University of Groningen Nutritional and metabolic aspects of the … · 2016. 3. 9. · Proefschrift ter verkrijging van de graad van doctor aan de Rijksuniversiteit Groningen op

University of Groningen

Nutritional and metabolic aspects of the hepatorenal axisDeetman, Petronella Elisabeth

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2015

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Deetman, P. E. (2015). Nutritional and metabolic aspects of the hepatorenal axis. University of Groningen.

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 18-08-2021

https://research.rug.nl/en/publications/nutritional-and-metabolic-aspects-of-the-hepatorenal-axis(cf6ec7bb-0982-4c59-aa84-2affe6fa968a).html

Nutritional and metabolic aspects of the hepatorenal axis

Petronella E. (Nicole) Deetman

Deetman, P.E.Nutritional and metabolic aspects of the hepatorenal axisDissertation University of Groningen, with summary in Dutch

ISBN: 978-90-367-7805-3 (printed version)ISBN: 978-90-367-7804-6 (digital version)

© P.E. (Nicole) Deetman 2015All rights reserved. No part of this publication may be reproduced, copied, modified or adapted, without the prior consent of the author.Financial support for the printing of this thesis was kindly provided by the University of Groningen, University Medical Center Groningen, Graduate School for Drug Exploration.

Cover: Douwe OppewalLayout, typesetting: Michal Slawinski, thesisprint.euPrinted in Poland

Nutritional and metabolic aspects of the hepatorenal axis

Proefschrift

ter verkrijging van de graad van doctor aan deRijksuniversiteit Groningen

op gezag van derector magnificus prof. dr. E. Sterken

en volgens besluit van het College voor Promoties.

De openbare verdediging zal plaatsvinden op

woensdag 10 juni 2015 om 16:15 uur

door

Petronella Elisabeth Deetmangeboren op 23 februari 1989

te Heemskerk

Promotores Prof. dr. S.J.L. Bakker Prof. dr. G. Navis

Beoordelingscommissie Prof. dr. J.A. Lisman Prof. dr. ir. E.J.M. Feskens Prof. dr. Y.M. Smulders

Paranimfen Drs. I.J. Riphagen Dr. J. Seggers

Contents

1 Introduction and aims of thesis 9

2 Bilirubin and progression of nephropathy in type 2 diabetes: A post-hoc analysis of RENAAL with independent replication in IDNT 21Diabetes 2014;63(8):2845–53

3 Plasma bilirubin and late graft failure in renal transplant recipients 45Transpl Int 2012;25(8):876–81

4 Diet, lifestyle, and total bilirubin: the Zutphen elderly study 57Submitted

5 Urinary urea excretion and long-term outcome after renal transplantation 71Transplantation 2014;12:(epub ahead of print)

6 Alanine aminotransferase and mortality in patients with type 2 diabetes (ZODIAC-38) 89Eur J Clin Invest, provisionally accepted

7 Uncovering of body mass index as a risk factor for poor long-term outcome after renal transplantation 105Transplantation 2015;99(1):e5–6, accepted in abbreviated form

8 General discussion and future perspectives 119

9 Nederlandse samenvatting 133

10 Dankwoord 147

Abbreviations 151

Author affiliations 152

List of Publications 154

1

Introduction

and aims of thesis1

11Introduction and aims of thesis

General introductionAging of the population along with the rising prevalence of obesity, type 2 diabetes, and hypertension have led to a worldwide increase in prevalence of chronic kidney disease (Figure 1) (CKD) (1). In the United States, the prevalence of patients with CKD was es-timated at 14 percent in the period from 2005 until 2010 (1). CKD may ultimately result in end stage renal disease (ESRD), necessitating dialysis or renal transplantation (2). In line with the rise in prevalence of CKD, the costs of ESRD are also rising in the United States, already costing approximately $34 billion in the year 2011, which is 6.3 percent of total Medicare costs (1). In the Netherlands, the number of patients with CKD has been estimated to be 1 million, of which 16 000 patients with ESRD (3).

CKD is defined by a state of kidney damage or decreased kidney function for three or more months (4). Kidney damage is identified by the presence of pathological abnor-malities or a rise in markers of kidney damage. Decreased kidney function is defined by a glomerular filtration rate (GFR) of less than 60 mL/min per 1.73 m2 (4). Once diagnosed with CKD, patients are classified into one of the five stages of CKD, which are determined by the level of GFR, albuminuria, or both (5). The highest stage of CKD (i.e. CKD stage 5) is defined by a GFR of less than 15 mL/min per 1.73m2 and is also known as ESRD.

The stages of CKD have been defined based on the associated risk of future CKD-related morbidity and mortality (4). This is especially important because patients with CKD are at a considerable increased risk of morbidity and mortality, and this risk in-creases with decreasing kidney function (6). The risk of mortality decreases after renal transplantation, but remains high compared to the general population (7). It is impor-

Figure 1. Prevalence of CKD, by age group based on the 2012 USRDS Annual data report. Adapted from JAMA 2007;298(17):2038–2047

12

Cha

pter

1

tant to realize that the high mortality rates among patients with CKD more often result from CKD-related comorbidities, than from kidney failure itself. A longitudinal study that investigated 27,998 patients with CKD stages 2 to 4 at baseline has reported that death was far more common as adverse outcome than dialysis at all stages of CKD (8,9). CKD patients are at a 5-fold increased risk of cardiovascular disease compared to the general population (10,11). Furthermore, in young patients receiving dialysis therapy, the risk of cardiovascular mortality has been estimated to be 1,000-fold higher compared to age-matched patients (12). The increased risk of cardiovascular disease in patients with CKD has been found to be independent of the classical cardiovascular risk factors (13). Specific, CKD-related, metabolic abnormalities, vascular calcification, and oxida-tive stress have been suggested to potentially contribute to this vascular injury (13,14).

Crosstalk

An integrative pathophysiological approach may open important new perspectives, as organ crosstalk can exert major effects on disease onset and progression. This has already been shown for cardiorenal interaction (15,16). There are data to suggest that interaction between the kidney and the liver, as the two major metabolic organs of the body, may be relevant to CKD as well. The hepatorenal axis, however, has mainly received attention in the context of acute liver failure. Acute liver failure is thought to result in an imbalance between vasoconstrictor factors and vasodilator factors that adversely affect the renal circulation, which may subsequently lead to renal dysfunction (17). The combination of renal dysfunction and liver dysfunction has been defined as the hepatorenal syndrome. The probability of developing the hepatorenal syndrome is approximately 20 percent in the first year and increases to 40 percent in the fifth year after a diagnosis of severe liver dysfunction (17,18). There are two types of the hepatorenal syndrome. Type 1 is characterized by a rapid progression in renal function, whereas type 2 is characterized by a relatively slow progression in renal function (17). The average prognosis of patients with the hepatorenal syndrome varies from 8 to 10 weeks to up to 6 months, depending on the type of hepatorenal syndrome (17).

The condition of hepatorenal syndrome demonstrates that the kidney can be profoundly affected by liver disease, but the possible role of the liver in chronic kidney disease has hardly been subject of investigation. In this thesis we will investigate several aspects of liver function, in the nonfailing liver, with respect to relevance for kidney disease and its complications.


Bilirubin

A commonly used marker of liver function is bilirubin. Bilirubin is the end product of heme catabolism. Heme is split into carbon monoxide, ferrous iron, and biliverdin by the enzyme heme oxygenase-1 (HO-1) (Figure 2) (19,20). Biliverdin is subsequently re-duced to unconjugated bilirubin by biliverdin reductase (19,20). In healthy subjects, un-conjugated bilirubin by far makes up the largest proportion of circulating bilirubin (21). Bilirubin is then conjugated by the liver and subsequently excreted in the bile (Figure 2). For decades, bilirubin has been thought to be a toxic waste product. Recent studies have, however, indicated that bilirubin has cytoprotective and antioxidant properties (22–24). In animal studies, exogenous administration of bilirubin protected against kidney isch-emia reperfusion injury and renal mesangial expansion, the latter being a typical feature of diabetic nephropathy (25). In addition, patients with both type 2 diabetes and diabetic nephropathy exhibited lower total bilirubin levels compared to patients without diabetic nephropathy (26). Therefore, we postulated that bilirubin is not merely a marker of liver function, but also a potential renoprotective agent if the bilirubin concentration is within the normal range. However, little is known about the association of bilirubin with re-nal function in humans. Therefore, in chapter 2, we prospectively investigated whether bilirubin was associated with progression of diabetic nephropathy in patients with type 2 diabetes. In chapter 3 we prospectively investigated whether bilirubin was associated with development of late graft failure and mortality after renal transplantation.

Figure 2. Schematic overview of bilirubin production. CO, carbon monoxide; UDP-GT, uridine diphosphate glucuronyltransferase

14

Cha

pter

1

The suggested protective effects of bilirubin raise questions about why its con-centrations vary between healthy individuals and how bilirubin concentrations can be elevated. Since HO-1 is the enzyme that catalyzes the rate limiting step in heme deg-radation (27,28), induction of HO-1 could result in higher levels of end products of heme degradation, including bilirubin. It has been suggested that bilirubin levels are largely utilized in the presence of increased oxidative stress (29). Several studies have demonstrated that smoking, a major contributor to oxidative stress (29–31), is inversely associated with bilirubin concentrations. Furthermore, a healthy dietary pattern like the Mediterranean diet was found to be associated with decreased plasma markers of oxi-dative stress (32–34). However, little is known about the effects of diet and lifestyle on bilirubin concentrations. In chapter 4 we aimed to investigate the associations of the Mediterranean diet, individual food groups, and lifestyle factors such as smoking with bilirubin concentrations in elderly men without major chronic diseases.

Urea production

Another kidney-related marker generated by the liver is urea. Urea is a waste product of protein metabolism, that is excreted by the kidneys. As such, urea can be used as a marker of protein intake (35). Lowering dietary protein intake is an important component of management of CKD patients (36–38). In these patients, moderate to high protein intake is supposed to aggravate proteinuria, a risk factor for progression of CKD (36,39–41). However, dietary protein restriction increases the risk of malnutrition (42–44). This may particularly be the case in patients with advanced CKD, because they often experience a loss of appetite with advancing kidney disease (42–44). Furthermore, patients with ESRD (CKD stage 5) experience increased catabolism of protein during treatment with dialysis. Studies in dialysis patients have consistently shown that low protein intake is associated with higher rates of morbidity and mortality (45,46). Accordingly, upon the start of dialysis, patients are recommended to consume relatively high amounts of protein rather than maintaining a low protein diet (38,47,48). However, dietary guidelines are lacking once patients have undergone transplantation. Therefore, in chapter 5, we inves-tigated the association of urinary urea excretion with graft failure and mortality in RTR.

Obesity, the liver, and the kidney

The liver may not only have distant effects on kidney function, the kidney and the liver may also be parallel victims of adverse factors to which they are both exposed. Obesity may be one of such factors, because it has been shown to increase both the risk of kidney


dysfunction and of liver dysfunction (49,50). The increased risk of kidney dysfunction associated with obesity is partly attributed to comorbidities such as hypertension and lipid disorders. However, obesity may also directly lead to CKD. Pathologic histological findings such as podocyte hypertrophy, mesangial cell expansion, and focal segmental glomerular sclerosis are more commonly present among obese individuals compared to healthy individuals, independent of blood pressure and serum glucose (51,52). In the liver, the major obesity-related abnormality is hepatic fat accumulation. In the absence of alcohol abuse (which is a different cause of fat accumulation in the liver that is in-dependent of obesity), the presence of hepatic fat accumulation is defined as nonalco-holic fatty liver disease (NAFLD) (53). NAFLD is common among patients with obesity. Compared to individuals without obesity, the prevalence of NAFLD is almost five times higher among obese individuals (54,55). At present, hepatic fat accumulation is one of the most common causes of liver injury (53), with NAFLD now becoming one of the most important indications for liver transplantation (53).

The definition of obesity is a controversial issue. A high BMI is often used to define obesity, but it is important to emphasize that a high BMI is not only determined by high fat mass, but also by other potentially beneficial constituents, like muscle mass. It may be because of this notion that several studies have reported absent or even in-verse associations of BMI with renal outcomes. A prospective study of 499 patients with CKD has reported that BMI was not associated with an increased rate of progression of existing CKD (56). Another study of 3,334 participants aged 65 years and older showed a surprising inverse association of BMI with CKD in older diabetic patients (57). In that study, BMI was not associated with CKD in patients without diabetes (57). Furthermore, the lack of an association of BMI with outcomes has been repeatedly observed in pa-tients with chronic diseases (58–61). Because the presence of NAFLD is an indicator of high fat mass, we hypothesized that NAFLD might be a better predictor of progression of renal dysfunction than BMI. To this end, we investigated whether ALT, as a marker of NAFLD, was associated with progression of renal function and mortality in patients with type 2 diabetes in chapter 6 (Figure 3).

Like mentioned, muscle mass is another constituent of BMI. A high muscle mass is an established marker of better outcomes, both in RTR (62) and in other populations, including the general population (63,64). It has been hypothesized that protein energy malnutrition and physical activity underlie the association of muscle mass with outcome (62). We hypothesized that BMI is a better marker of a high fat mass, if muscle mass is also taken into account. Accordingly, we investigated whether a high muscle mass, as determined by 24h urinary creatinine excretion, confounded associations of BMI with outcomes in chapter 7.

16

Cha

pter

1

General overview of thesisThe aim of this thesis is to investigate the possible role of the nonfailing liver in chron-ic kidney disease and its complications. In chapter 2, we will prospectively investigate whether bilirubin is associated with decline in renal function in patients with diabetic nephropathy. In chapter 3, we will prospectively investigate the associations of bilirubin with graft failure and mortality in RTR. In chapter 4, we will focus on possible ways to naturally elevate bilirubin concentrations, by investigating associations of a healthy dietary pattern, individual food groups, and lifestyle components such as smoking with bilirubin concentrations in elderly men without major chronic diseases. In chapter 5, we will investigate another kidney-related product that is generated by the liver; urea. In this chapter, the association of urinary urea excretion with graft failure and mortal-ity is studied in RTR. In chapter 6, we will investigate whether ALT, as a marker of he-patic fat accumulation, may be a predictor of progression of renal function and mortality in patients with type 2 diabetes. In chapter 7, we zoom in on the association of BMI with outcomes, by investigating whether BMI is a better marker of a high fat mass, and there-fore a better predictor of mortality in RTR, if muscle mass is also taken into account.

Figure 3. Simplified hypothetical associations of the hepatorenal axis. ALT, alanine amino-transferase


References 1. U.S. Renal Data System. USRDS 2013 Annual Data Report: Atlas of Chronic Kidney Disease

and End-Stage Renal Disease in the United States, National Institutes of Health, National Insti-tute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD

2. James MT, Hemmelgarn BR, Tonelli M. Early recognition and prevention of chronic kidney disease. Lancet 2010 Apr 10;375(9722):1296–1309.

3. Dutch Kidney Foundation. Available at: http://www.nierstichting.nl/nieren/onzenieren/feiten-en-cijfers, 2014.

4. National Kidney Foundation. K/DOQI clinical practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Am J Kidney Dis 2002 Feb;39(2 Suppl 1):S1–266.

5. CKD Work Group. KDIGO 2012 Clinical Practice Guideline for the Evaluation and Manage-ment of Chronic Kidney Disease. Kidney Int Suppl 2013;3:1–150.

6. Muntner P, He J, Hamm L, Loria C, Whelton PK. Renal insufficiency and subsequent death result-ing from cardiovascular disease in the United States. J Am Soc Nephrol 2002 Mar;13(3):745–753.

7. Wolfe RA, Ashby VB, Milford EL, Ojo AO, Ettenger RE, Agodoa LY, et al. Comparison of mor-tality in all patients on dialysis, patients on dialysis awaiting transplantation, and recipients of a first cadaveric transplant. N Engl J Med 1999 Dec 2;341(23):1725–1730.

8. Dennis VW. Coronary heart disease in patients with chronic kidney disease. J Am Soc Nephrol 2005 Nov;16 Suppl 2:S103–6.

9. Keith DS, Nichols GA, Gullion CM, Brown JB, Smith DH. Longitudinal follow-up and out-comes among a population with chronic kidney disease in a large managed care organization. Arch Intern Med 2004 Mar 22;164(6):659–663.

10. Aakhus S, Dahl K, Wideroe TE. Cardiovascular disease in stable renal transplant patients in Nor-way: morbidity and mortality during a 5-yr follow-up. Clin Transplant 2004 Oct;18(5):596–604.

11. Kasiske BL. Cardiovascular disease after renal transplantation. Semin Nephrol 2000 Mar;20(2):176–187.

12. Foley RN, Parfrey PS, Sarnak MJ. Clinical epidemiology of cardiovascular disease in chronic renal disease. Am J Kidney Dis 1998 Nov;32(5 Suppl 3):S112–9.

13. Stenvinkel P. Chronic kidney disease: a public health priority and harbinger of premature car-diovascular disease. J Intern Med 2010 Nov;268(5):456–467.

14. Henry RM, Kostense PJ, Bos G, Dekker JM, Nijpels G, Heine RJ, et al. Mild renal insufficien-cy is associated with increased cardiovascular mortality: The Hoorn Study. Kidney Int 2002 Oct;62(4):1402–1407.

15. Anand IS. Cardiorenal syndrome: a cardiologist’s perspective of pathophysiology. Clin J Am Soc Nephrol 2013 Oct;8(10):1800–1807.

16. Elsayed EF, Tighiouart H, Griffith J, Kurth T, Levey AS, Salem D, et al. Cardiovascular disease and subsequent kidney disease. Arch Intern Med 2007 Jun 11;167(11):1130–1136.

18

Cha

pter

1

17. Cardenas A. Hepatorenal syndrome: a dreaded complication of end-stage liver disease. Am J Gastroenterol 2005 Feb;100(2):460–467.

18. Gines A, Escorsell A, Gines P, Salo J, Jimenez W, Inglada L, et al. Incidence, predictive factors, and prognosis of the hepatorenal syndrome in cirrhosis with ascites. Gastroenterology 1993 Jul;105(1):229–236.

19. Vitek L. The role of bilirubin in diabetes, metabolic syndrome, and cardiovascular diseases. Front Pharmacol 2012;3:55.

20. Abraham NG, Asija A, Drummond G, Peterson S. Heme oxygenase -1 gene therapy: recent advances and therapeutic applications. Curr Gene Ther 2007 Apr;7(2):89–108.

21. Fevery J. Bilirubin in clinical practice: a review. Liver Int 2008 May;28(5):592–605. 22. Abraham NG, Kappas A. Heme oxygenase and the cardiovascular-renal system. Free Radic

Biol Med 2005 Jul 1;39(1):1–25. 23. Vitek L, Schwertner HA. The heme catabolic pathway and its protective effects on oxidative

stress-mediated diseases. Adv Clin Chem 2007;43:1–57. 24. Morita T. Heme oxygenase and atherosclerosis. Arterioscler Thromb Vasc Biol 2005

Sep;25(9):1786–1795. 25. Adin CA, Croker BP, Agarwal A. Protective effects of exogenous bilirubin on ischemia-

reperfusion injury in the isolated, perfused rat kidney. Am J Physiol Renal Physiol 2005 Apr;288(4):F778–84.

26. Zelle DM, Deetman N, Alkhalaf A, Navis G, Bakker SJ. Support for a protective effect of biliru-bin on diabetic nephropathy in humans. Kidney Int 2011 Mar;79(6):686; 686–7.

27. Riphagen IJ, Deetman PE, Bakker SJ, Navis G, Cooper ME, Lewis JB, et al. Bilirubin and Pro-gression of Nephropathy in Type 2 Diabetes: A Post-Hoc Analysis of RENAAL with Indepen-dent Replication in IDNT. Diabetes 2014 Mar 27.

28. Kirkby KA, Adin CA. Products of heme oxygenase and their potential therapeutic applications. Am J Physiol Renal Physiol 2006 Mar;290(3):F563–71.

29. Schwertner HA, Jackson WG, Tolan G. Association of low serum concentration of bilirubin with increased risk of coronary artery disease. Clin Chem 1994 Jan;40(1):18–23.

30. Pignatelli B, Li CQ, Boffetta P, Chen Q, Ahrens W, Nyberg F, et al. Nitrated and oxidized plasma proteins in smokers and lung cancer patients. Cancer Res 2001 Jan 15;61(2):778–784.

31. Schwertner HA. Association of smoking and low serum bilirubin antioxidant concentrations. Atherosclerosis 1998 Feb;136(2):383–387.

32. Dai J, Jones DP, Goldberg J, Ziegler TR, Bostick RM, Wilson PW, et al. Association between adher-ence to the Mediterranean diet and oxidative stress. Am J Clin Nutr 2008 Nov;88(5):1364–1370.

33. Fito M, Guxens M, Corella D, Saez G, Estruch R, de la Torre R, et al. Effect of a traditional Mediterranean diet on lipoprotein oxidation: a randomized controlled trial. Arch Intern Med 2007 Jun 11;167(11):1195–1203.

34. Yubero-Serrano EM, Garcia-Rios A, Delgado-Lista J, Delgado-Casado N, Perez-Martinez P, Rodriguez-Cantalejo F, et al. Postprandial effects of the Mediterranean diet on oxidant and antioxidant status in elderly men and women. J Am Geriatr Soc 2011 May;59(5):938–940.


35. Maroni BJ, Steinman TI, Mitch WE. A method for estimating nitrogen intake of patients with chronic renal failure. Kidney Int 1985 Jan;27(1):58–65.

36. Fouque D, Pelletier S, Mafra D, Chauveau P. Nutrition and chronic kidney disease. Kidney Int 2011 Aug;80(4):348–357.

37. KDOQI. KDOQI Clinical Practice Guidelines and Clinical Practice Recommendations for Dia-betes and Chronic Kidney Disease. Am J Kidney Dis 2007 Feb;49(2 Suppl 2):S12–154.

38. Clinical practice guidelines for nutrition in chronic renal failure. K/DOQI, National Kidney Foundation. Am J Kidney Dis 2000 Jun;35(6 Suppl 2):S1–140.

39. Brantsma AH, Atthobari J, Bakker SJ, de Zeeuw D, de Jong PE, Gansevoort RT. What predicts progression and regression of urinary albumin excretion in the nondiabetic population? J Am Soc Nephrol 2007 Feb;18(2):637–645.

40. Levey AS, Greene T, Beck GJ, Caggiula AW, Kusek JW, Hunsicker LG, et al. Dietary protein restriction and the progression of chronic renal disease: what have all of the results of the MDRD study shown? Modification of Diet in Renal Disease Study group. J Am Soc Nephrol 1999 Nov;10(11):2426–2439.

41. Chauveau P, Combe C, Rigalleau V, Vendrely B, Aparicio M. Restricted protein diet is associ-ated with decrease in proteinuria: consequences on the progression of renal failure. J Ren Nutr 2007 Jul;17(4):250–257.

42. Kopple JD, Greene T, Chumlea WC, Hollinger D, Maroni BJ, Merrill D, et al. Relationship between nutritional status and the glomerular filtration rate: results from the MDRD study. Kidney Int 2000 Apr;57(4):1688–1703.

43. Carrero JJ. Mechanisms of altered regulation of food intake in chronic kidney disease. J Ren Nutr 2011 Jan;21(1):7–11.

44. Kalantar-Zadeh K, Block G, McAllister CJ, Humphreys MH, Kopple JD. Appetite and inflam-mation, nutrition, anemia, and clinical outcome in hemodialysis patients. Am J Clin Nutr 2004 Aug;80(2):299–307.

45. Shinaberger CS, Greenland S, Kopple JD, Van Wyck D, Mehrotra R, Kovesdy CP, et al. Is con-trolling phosphorus by decreasing dietary protein intake beneficial or harmful in persons with chronic kidney disease? Am J Clin Nutr 2008 Dec;88(6):1511–1518.

46. Fouque D, Pelletier S, Guebre-Egziabher F. Have recommended protein and phosphate intake recently changed in maintenance hemodialysis? J Ren Nutr 2011 Jan;21(1):35–38.

47. Kopple JD, National Kidney Foundation K/DOQI Work Group. The National Kidney Founda-tion K/DOQI clinical practice guidelines for dietary protein intake for chronic dialysis pa-tients. Am J Kidney Dis 2001 Oct;38(4 Suppl 1):S68–73.

48. Fouque D, Vennegoor M, ter Wee P, Wanner C, Basci A, Canaud B, et al. EBPG guideline on nutrition. Nephrol Dial Transplant 2007 May;22 Suppl 2:ii45–87.

49. Kwakernaak AJ, Toering TJ, Navis G. Body mass index and body fat distribution as renal risk factors: a focus on the role of renal haemodynamics. Nephrol Dial Transplant 2013 Nov;28 Suppl 4:iv42–9.

20

Cha

pter

1

50. Pinto-Sietsma SJ, Navis G, Janssen WM, de Zeeuw D, Gans RO, de Jong PE, et al. A central body fat distribution is related to renal function impairment, even in lean subjects. Am J Kid-ney Dis 2003 Apr;41(4):733–741.

51. Ix JH, Sharma K. Mechanisms linking obesity, chronic kidney disease, and fatty liver disease: the roles of fetuin-A, adiponectin, and AMPK. J Am Soc Nephrol 2010 Mar;21(3):406–412.

52. Serra A, Romero R, Lopez D, Navarro M, Esteve A, Perez N, et al. Renal injury in the extremely obese patients with normal renal function. Kidney Int 2008 Apr;73(8):947–955.

53. Targher G, Byrne CD. Clinical Review: Nonalcoholic fatty liver disease: a novel cardiometa-bolic risk factor for type 2 diabetes and its complications. J Clin Endocrinol Metab 2013 Feb;98(2):483–495.

54. Angulo P. Nonalcoholic fatty liver disease. N Engl J Med 2002 Apr 18;346(16):1221–1231. 55. Hong HC, Hwang SY, Choi HY, Yoo HJ, Seo JA, Kim SG, et al. Relationship between sarcopenia

and nonalcoholic fatty liver disease: the Korean Sarcopenic Obesity Study. Hepatology 2014 May;59(5):1772–1778.

56. Brown RN, Mohsen A, Green D, Hoefield RA, Summers LK, Middleton RJ, et al. Body mass index has no effect on rate of progression of chronic kidney disease in non-diabetic subjects. Nephrol Dial Transplant 2012 Jul;27(7):2776–2780.

57. Kao YM, Chen JD. Inverse association between body mass index and chronic kidney disease in older diabetic adults. Ann Epidemiol 2013 May;23(5):255–259.

58. Kalantar-Zadeh K, Block G, Horwich T, Fonarow GC. Reverse epidemiology of conventional cardiovascular risk factors in patients with chronic heart failure. J Am Coll Cardiol 2004 Apr 21;43(8):1439–1444.

59. Marks WH, Florence LS, Chapman PH, Precht AF, Perkinson DT. Morbid obesity is not a con-traindication to kidney transplantation. Am J Surg 2004 May;187(5):635–638.

60. Nicoletto BB, Fonseca NK, Manfro RC, Goncalves LF, Leitao CB, Souza GC. Effects of obesity on kidney transplantation outcomes: a systematic review and meta-analysis. Transplantation 2014 Jul 27;98(2):167–176.

61. Kalantar-Zadeh K, Block G, Humphreys MH, Kopple JD. Reverse epidemiology of cardiovas-cular risk factors in maintenance dialysis patients. Kidney Int 2003 Mar;63(3):793–808.

62. Oterdoom LH, van Ree RM, de Vries AP, Gansevoort RT, Schouten JP, van Son WJ, et al. Uri-nary creatinine excretion reflecting muscle mass is a predictor of mortality and graft loss in re-nal transplant recipients. Transplantation 2008 Aug 15;86(3):391–398.

63. Oterdoom LH, Gansevoort RT, Schouten JP, de Jong PE, Gans RO, Bakker SJ. Urinary creatinine excretion, an indirect measure of muscle mass, is an independent predictor of cardiovascular disease and mortality in the general population. Atherosclerosis 2009 Dec;207(2):534–540.

64. Sinkeler SJ, Kwakernaak AJ, Bakker SJ, Shahinfar S, Esmatjes E, de Zeeuw D, et al. Creati-nine excretion rate and mortality in type 2 diabetes and nephropathy. Diabetes Care 2013 Jun;36(6):1489–1494.

2

Bilirubin and progression

of nephropathy

in type 2 diabetes:

A post‑hoc analysis of

RENAAL with independent

replication in IDNT

Ineke J. Riphagen1, Petronella E. Deetman1, Stephan J.L. Bakker1, Gerjan Navis1, Mark E. Cooper2, Julia B. Lewis3, Dick de Zeeuw4, and Hiddo J. Lambers Heerspink4.

1 Department of Medicine, Division of Nephrology, University of Groningen, University Medical Center Groningen, Groningen, the Netherlands

2 Baker IDI Heart & Diabetes Institute, Melbourne, Victoria, Australia

3 Division of Nephrology, VanderBilt University, Nashville, TN, USA

4 Department of Clinical Pharmacy and Pharmacology, University of Groningen, University Medical Center Groningen, Groningen, the Netherlands

Diabetes 2014;63(8):2845–53

2

22

Cha

pter

2

AbstractBackground. Bilirubin, a potent endogenous antioxidant, was found to protect against development of diabetic nephropathy (DN) in rodents. In humans, cross-sectional stud-ies found an inverse relation between bilirubin and DN. We prospectively investigated whether bilirubin is associated with progression of DN toward end-stage renal disease (ESRD). Methods. We performed a post hoc analysis in the Reduction of Endpoints in non-insu-lin dependent diabetes mellitus (NIDDM) with the Angiotensin II Antagonist Losartan (RENAAL) trial with independent replication in the Irbesartan Diabetic Nephropathy Trial (IDNT). Subjects with type 2 diabetes and nephropathy with alanine aminotrans-ferase, aspartate aminotransferase (AST), and bilirubin levels <1.5 times the upper limit of normal were included. The renal end point was defined as the composite of confirmed doubling of serum creatinine or ESRD. Results. Bilirubin was inversely associated with the renal end point in RENAAL inde-pendent of age, gender, race, BMI, smoking, total cholesterol, diastolic blood pressure, HbA1c, treatment, estimated glomerular filtration rate, albumin-to-creatinine ratio, and AST. These results were confirmed in IDNT. Conclusions. This study indicates an independent inverse association of bilirubin with progression of nephropathy in RENAAL and IDNT. These data suggest a protective ef-fect of bilirubin against progression of nephropathy in type 2 diabetes. The well-estab-lished role of bilirubin as an antioxidant is a potential explanation for our findings.

23Bilirubin and diabetic nephropathy

IntroductionThe incidence of type 2 diabetes and its complications are increasing worldwide. One of the major complications of type 2 diabetes is diabetic nephropathy (DN). Nephropathy develops in approximately 20–40% of patients with diabetes and is the single leading cause of end-stage renal disease (ESRD) around the world (1).

Bilirubin is a product of heme catabolism and is known to be a potent endoge-nous antioxidant (2). As such, bilirubin has consistently been associated with protection against development of cardiovascular disease (CVD) (3,4). A recent study in rodents suggested that bilirubin is also protective against progression of DN (5). This notion is supported by several cross-sectional studies in humans demonstrating that low levels of bilirubin are associated with DN (6–8).

To our knowledge, there are no prospective studies to date that investigated whether bilirubin levels are associated with progression of DN toward ESRD. Therefore, our primary objective was to prospectively investigate the association of bilirubin with progression of nephropathy in patients with type 2 diabetes. To this end, we performed a historical prospective study in the Reduction of Endpoints in non-insulin dependent diabetes with the Angiotensin II Antagonist Losartan (RENAAL) trial (9,10). Subse-quently, independent replication was sought in the Irbesartan Diabetic Nephropathy Trial (IDNT) (11,12). In the RENAAL and IDNT studies, patients were treated with an angiotensin receptor blocker (ARB) (losartan in RENAAL and irbesartan in IDNT). Several studies have shown that ARBs reduce hemoglobin levels (13–15). Because bili-rubin is a product of heme catabolism, the use of ARBs could consequently reduce bili-rubin levels. Therefore, our secondary aim was to investigate the effect of ARB treatment on serum concentrations of hemoglobin and bilirubin.

Patients and methods

Study Design and Population

The present study was conducted in patients with type 2 diabetes and nephropathy par-ticipating in the RENAAL and IDNT studies. The design, rationale, and study outcomes for these trials have been published elsewhere (9–12). Both trials investigated the ef-ficacy of an ARB (losartan in RENAAL and irbesartan in IDNT) on renal outcomes in patients with type 2 diabetes, nephropathy, and proteinuria. Inclusion criteria for both trials were similar, with minor differences in details. Patients with type 2 diabetes, hy-pertension, and nephropathy aged 30–70 years were eligible for inclusion in both trials. Serum creatinine levels ranged from 1.0 to 3.0 mg/dL. All subjects were required to have proteinuria defined as a urinary albumin-to-creatinine ratio (ACR) ≥300 mg/g or

24

Cha

pter

2

a 24-h urinary protein excretion >500 mg/day in RENAAL and ≥900 mg in IDNT. Major exclusion criteria for participation in both trials were type 1 diabetes, nondiabetic renal disease, and screening values of liver enzymes (alanine aminotransferase [ALT], aspar-tate aminotransferase [AST]) or total bilirubin >1.5 times the upper limit of normal. The inclusion and exclusion criteria of the RENAAL and IDNT studies are summarized in Supplementary Table S1 [adapted from Packham et al. (16)]. Subjects with missing data for baseline measurements of total bilirubin were excluded from the analyses (RE-NAAL n = 15 [1.0%], IDNT n = 8 [0.5%]).

Measurements and Clinical End Points

Laboratory and physical assessment data were collected every 6 months during follow-up for subjects participating in RENAAL and IDNT and included blood pressure measure-ments, glycated hemoglobin (HbA1c), lipid profile, hemoglobin, total bilirubin, serum albumin, ALT, AST, and serum creatinine. For both trials, all biochemical measurements were conducted in a central laboratory according to standardized conditions. Estimated glomerular filtration rate (GFR) was calculated using the Modification of Diet in Renal Disease (MDRD) equation (17). The primary end point for the current study was the composite of a confirmed doubling of serum creatinine (DSCR) or ESRD (defined as the need for long-term dialysis or renal transplantation). All end points were adjudicated by an independent committee using rigorous guidelines and definitions.

Statistical Analyses

Statistical analyses were performed using SPSS version 18.0 for Windows (IBM Cor-poration, Chicago, IL) and Stata 11 (StataCorp LP, College Station, TX). Results are presented as mean ± SD for variables with a normal distribution and as median (inter-quartile range) for variables with a nonnormal distribution. Nominal data are presented as the total number of patients with percentages. A two-sided P < 0.05 was considered statistically significant.

To assess which baseline variables were associated with baseline total bilirubin, the study populations were subdivided into tertiles of baseline total bilirubin concentra-tion, and patient characteristics were presented accordingly. P‑values for trend across tertiles of baseline total bilirubin were assessed using linear regression analyses. Vari-ables with a skewed distribution were log-transformed to fulfill criteria for linear regres-sion analyses. Multivariable linear regression analyses were used to investigate which clinical parameters at baseline were independently associated with bilirubin at baseline.


The course of clinical parameters over time are presented according to tertiles of baseline bilirubin. We investigated whether the change in total bilirubin concentra-tion over time differed among tertiles in subjects with complete bilirubin measurements at baseline and 12, 24, and 36 months using one-way ANOVA.

To investigate the association of total bilirubin with progression of nephropa-thy, we used Cox proportional hazard regression analyses with time-varying covari-ates, which takes the change of clinical parameters over time into account. Logarithmic transformation (base 2) of bilirubin levels was applied in order to present the hazard ratios (HRs) derived from Cox regression analyses per doubling of bilirubin levels. Mul-tivariable analyses were conducted using a Cox regression model, including the potential confounding factors of age, sex, baseline eGFR, baseline log ACR, race, smoking at base-line, history of CVD at baseline, baseline BMI, total cholesterol, diastolic blood pressure, HbA1c, treatment assignment, and log AST.

In sensitivity analyses, we repeated the Cox regression analyses in the subgroups receiving an ARB (i.e., losartan in RENAAL, irbesartan in IDNT) or placebo in both trials. In further sensitivity analyses, we investigated whether the change in bilirubin val-ues during the course of the trials was associated with renal progression. To investigate the effect of treatment with an ARB on serum concentrations of hemoglobin and total bilirubin, we used the independent sample t test to compare hemoglobin and bilirubin concentrations between treatment groups in both trials.

Results

Patient Characteristics

In RENAAL, bilirubin concentrations were available for 1,498 (99.0%) patients. Mean baseline bilirubin level was 0.57 ± 0.19 mg/dL. Baseline patient characteristics according to tertiles of baseline bilirubin levels are presented in Table 1A and 1B. Prevalence of male sex, age, history of CVD, hemoglobin, serum albumin, liver enzymes, and eGFR increased with increasing bilirubin levels, whereas the prevalence in use of diuretics, use of insulin, BMI, HbA1c, cholesterol, triglycerides, and urinary ACR decreased with increasing bilirubin levels.

26

Cha

pter

2

Table 1A. Baseline patient characteristics of the RENAAL study population presented as tertiles of bilirubin concentrations.

RENAALAll subjects Tertile 1 Tertile 2 Tertile 3 P‑value

N 1,498 374 776 348 ‑ Total bilirubin (mg/dL) 0.57 ± 0.19 0.1‑0.4 0.5‑0.6 0.7‑2.1 ‑ Age (years) 60.1 ± 7.4 57.5 ± 7.7 60.8 ± 7.2 61.4 ± 7.0 <0.001 Male gender (n, %) 946 (63.2) 187 (50.0) 490 (63.1) 269 (77.3) <0.001 History of CVD (n, %) 443 (29.2) 88 (23.5) 249 (32.1) 106 (30.5) 0.04Race White (n, %) 723 (48.3) 119 (31.8) 416 (53.6) 188 (54.0) 0.06 Black (n, %) 228 (15.2) 92 (24.6) 112 (14.4) 24 (6.9) Hispanic (n, %) 276 (18.4) 109 (29.1) 113 (14.6) 54 (15.5) Asian (n, %) 252 (16.8) 49 (13.1) 125 (16.1) 78 (22.4) Other (n, %) 19 (1.3) 5 (1.3) 10 (1.3) 4 (1.1)Smoking status Smoker (n, %) 270 (18.0) 80 (21.4) 134 (17.3) 56 (16.1) 0.06Body composition BMI (kg/m2) 29.7 ± 6.3 30.2 ± 7.1 29.9 ± 6.1 28.7 ± 5.6 0.002Blood pressure Systolic blood pressure (mmHg) 153 ± 19 152 ± 19 153 ± 20 151 ± 19 0.5 Diastolic blood pressure (mmHg) 82 ± 10 82 ± 10 82 ± 10 83 ± 11 0.1 Use of ACEi/ARB (n, %) 769 (51.3) 192 (51.3) 415 (53.5) 162 (46.6) 0.2 Use of diuretics (n, %) 870 (58.1) 251 (67.1) 440 (56.7) 179 (51.4) <0.001Glucose homeostasis Diabetes duration ≥ 5 years (n, %) 1,351 (90.2) 342 (91.4) 705 (90.9) 304 (87.4) 0.05 HbA1c (%) 8.4 ± 1.6 8.8 ± 1.6 8.5 ± 1.6 8.2 ± 1.6 <0.001 HbA1c (mmol/mol) 69 ± 18 73 ± 18 69 ± 18 66 ± 18 <0.001 Use of insulin (n, %) 901 (60.1) 252 (67.4) 471 (67.4) 178 (51.1) <0.001Laboratory measurements Hemoglobin (g/dL) 12.5 ± 1.8 11.6 ± 1.5 12.5 ± 1.7 13.5 ± 1.8 <0.001 Serum albumin (g/dL) 3.8 ± 0.4 3.5 ± 0.4 3.8 ± 0.4 4.0 ± 0.3 <0.001Lipids Total cholesterol (mg/dL) 228 ± 56 244 ± 61 226 ± 54 216 ± 48 <0.001 HDL cholesterol (mg/dL) 45 ± 15 48 ± 17 44 ± 14 44 ± 14 0.001 LDL cholesterol (mg/dL) 142 ± 46 152 ± 53 141 ± 44 134 ± 39 <0.001 Triglycerides (mg/dL) 172 [122‑245] 181 [133‑270] 172 [120‑244] 160 [111‑228] <0.001Liver function ALT (U/L) 15 [12‑21] 14 [11‑19] 15 [12‑21] 16 [13‑24] <0.001 AST (U/L) 16 [13‑20] 15 [12‑19] 16 [13‑20] 17 [14‑23] <0.001Renal function

ACR (mg/g)1247

[560‑2559]1917

[882‑3730]1193

[544‑2334]855

[433‑1749]<0.001

Serum creatinine (mg/dL) 1.9 ± 0.5 1.9 ± 0.5 1.9 ± 0.5 1.8 ± 0.4 <0.001 eGFR, MDRD (mL/min/1.73m2) 39.8 ± 12.4 38.2 ± 12.7 39.7 ± 12.5 41.8 ± 11.5 <0.001

ACEi, angio‑converting enzyme inhibitor; ACR, urinary albumin‑to‑creatinine ratio; ALT, alanine aminotransfer‑

ase; ARB, angiotensin receptor blocker; AST, aspartate aminotransferase; BMI, body mass index; CVD, cardiovascu‑

lar diseases; eGFR, estimated glomerular filtration rate; HDL, high‑density lipoprotein.


Table 1B. Baseline patient characteristics of the IDNT study population presented as tertiles of bilirubin concentrations.

IDNTAll subjects Tertile 1 Tertile 2 Tertile 3 P‑value

N 1,707 588 413 706 ‑ Total bilirubin (mg/dL) 0.54 ± 0.21 0.1–0.4 0.5 0.6–2.0 ‑ Age (years) 58.9 ± 7.8 57.4 ± 8.4 58.7 ± 7.7 60.3 ± 7.0 <0.001 Male gender (n, %) 1,134 (66.4) 297 (50.5) 270 (65.4) 567 (80.3) <0.001 History of CVD (n, %) 481 (28.2) 168 (28.6) 101 (24.5) 212 (30.0) 0.5Race White (n, %) 1,238 (72.5) 323 (54.9) 323 (78.2) 592 (83.9) <0.001 Black (n, %) 224 (13.1) 138 (23.5) 39 (9.4) 47 (6.7) Hispanic (n, %) 83 (4.9) 45 (7.7) 18 (4.4) 20 (2.8) Asian (n, %) 85 (5.0) 55 (9.4) 13 (3.1) 17 (2.4) Other (n, %) 77 (4.5) 27 (4.6) 20 (4.8) 30 (4.2)Smoking status Smoker (n, %) 299 (17.5) 119 (20.2) 76 (18.4) 104 (14.7) 0.009Body composition BMI (kg/m2) 30.8 ± 5.8 32.2 ± 6.8 30.3 ± 5.2 30.0 ± 4.9 <0.001Blood pressure Systolic blood pressure (mmHg) 159 ± 20 160 ± 21 159 ± 19 159 ± 19 0.3 Diastolic blood pressure (mmHg) 87 ± 11 86 ± 11 87 ± 10 88 ± 11 <0.001 Use of ACEi/ARB (n, %) 797 (46.7) 306 (52.0) 175 (42.4) 316 (44.8) 0.01 Use of diuretics (n, %) 802 (47.0) 320 (54.4) 198 (47.9) 284 (40.2) <0.001Glucose homeostasis Diabetes duration ≥ 5 years (n, %) 1,533 (89.8) 537 (91.3) 374 (90.6) 622 (88.1) 0.06 HbA1c (%) 8.1 ± 1.7 8.1 ± 1.8 8.3 ± 1.8 8.1 ± 1.7 0.9 HbA1c (mmol/mol) 65 ± 19 65 ± 20 67 ± 20 65 ± 19 0.9 Use of insulin (n, %) 985 (57.7) 367 (62.4) 232 (56.2) 386 (54.7) 0.006Laboratory measurements Hemoglobin (g/dL) 12.9 ± 1.9 12.0 ± 1.8 12.9 ± 1.8 13.8 ± 1.7 <0.001 Serum albumin (g/dL) 3.8 ± 0.4 3.6 ± 0.5 3.9 ± 0.4 4.0 ± 0.3 <0.001Lipids Total cholesterol (mg/dL) 228 ± 58 239 ± 64 229 ± 58 218 ± 51 <0.001 HDL cholesterol (mg/dL) 42 ± 14 43 ± 15 44 ± 15 41 ± 13 0.1 LDL cholesterol (mg/dL) 142 ± 46 150 ± 50 144 ± 48 136 ± 41 <0.001 Triglycerides (mg/dL) 177 [119–270] 185 [127–276] 178 [116–270] 169 [115–266] 0.06Liver function ALT (U/L) 18 [13–25] 17 [12–24] 18 [13–25] 19 [14–26] 0.001 AST (U/L) 18 [14–23] 17 [14–22] 18 [14–23] 18 [15–24] 0.001Renal function

ACR (mg/g)1500

[780‑2759]2130

[1163‑3692]1429

[781‑2609]1106

[604‑2015]<0.001

Serum creatinine (mg/dL) 1.7 ± 0.6 1.8 ± 0.6 1.7 ± 0.6 1.6 ± 0.5 <0.001 eGFR, MDRD (mL/min/1.73m2) 47.4 ± 17.5 43.3 ± 16.7 45.9 ± 18.2 51.6 ± 17.0 <0.001



lar diseases; eGFR, estimated glomerular filtration rate; HDL, high‑density lipoprotein.

28

Cha

pter

2

Multivariable linear regression analyses showed that baseline bilirubin levels were inde-pendently associated with age, smoking, BMI, HbA1c, hemoglobin, serum albumin, log AST, and total cholesterol (Table 2A and 2B).

In IDNT, bilirubin concentrations were available for 1,707 patients (99.5%). The mean baseline bilirubin level in IDNT was similar to that in RENAAL (0.54 ± 0.21 mg/dL). In general, associations and trends of bilirubin with baseline characteristics were similar to those observed in RENAAL. In multivariable linear regression analyses, all variables that were independently associated with bilirubin in RENAAL (except AST) were also independently associated with bilirubin in IDNT. Sex, race, diastolic blood pressure, duration of diabetes (≥5 years), and eGFR were also significantly associated with bilirubin levels in IDNT.

Clinical Parameters Over Time

The course of clinical parameters over time is shown in Supplementary Table S2. In RE-NAAL, the change in total bilirubin concentration only differed among tertiles of base-line bilirubin at 12 months. After 24 and 36 months, there were no significant differences in change in bilirubin concentrations among tertiles of bilirubin (Supplementary Table S3). These results were confirmed in IDNT (Supplementary Table S3).

Progression of Nephropathy

After a mean follow-up period of 3.4 years, 471 (31%) subjects had reached the renal end point of DSCR or ESRD in RENAAL. Univariable time-varying Cox regression analysis showed that total bilirubin was significantly associated with progression of nephropathy in RENAAL (HR 0.54 [95% CI 0.45–0.65], P < 0.001) (Table 3, model 1). These asso-ciations remained significant after adjustment for potential confounding factors, which were age, sex, race, eGFR, log ACR, BMI, smoking status, history of CVD, total choles-terol, diastolic blood pressure, HbA1c, treatment, and log AST (HR 0.67 [0.55–0.83], P < 0.001). The risk for the renal end point according to total bilirubin concentrations in the RENAAL trial is shown in Figure 1A.

In IDNT, 381 (22%) patients reached the renal end point after a mean follow-up period of 2.6 years. The results of time-varying Cox proportional hazard regression analy-ses were similar to those of RENAAL (HR 0.64 [0.55–0.76], P < 0.001) for the final multi-variable model (Table 3). The graph indicating the risk for the renal end point according to total bilirubin concentrations in IDNT is similar to that for RENAAL (Figure 1B).

In sensitivity analyses, we investigated whether total bilirubin was associated with progression of nephropathy irrespective of ARB or placebo assignment. In RENAAL, to-tal bilirubin was significantly and inversely associated with progression of nephropathy


Figure 1. Histogram of baseline bilirubin concentrations in the RENAAL (A) and IDNT (B) studies. The line in the graph represents the risk for DSCR or ESRD. The grey area represents the 95% CI of the HR.

for subjects receiving losartan (HR 0.66 [0.48–0.89], P = 0.008) and those receiving pla-cebo (HR 0.70 [0.52–0.94], P = 0.02). These results were confirmed in IDNT for subjects receiving irbesartan (HR 0.59 [0.43–0.81], P = 0.001) and those receiving placebo (HR 0.61 [0.46–0.80], P < 0.001).

In further sensitivity analyses, the change in total bilirubin during the course of the trial was not associated with the renal end point (HR 1.07 [0.97–1.19], P = 0.2, per 0.1 mg/dL), whereas total bilirubin remained significantly associated with the renal end point in RENAAL (HR 0.59 [0.44–0.79], P < 0.001). These results were confirmed in IDNT for change in total bilirubin (HR 1.05 [0.97–1.13], P = 0.2) and for total biliru-bin (HR 0.62 [0.52–0.74], P < 0.001). The results remained essentially unchanged when stratified for treatment.

30

Cha

pter

2

Table 2A. Univariable and multivariable associations of log2-transformed bilirubin con-centrations with clinical parameters in RENAAL.

RENAAL

Univariable Multivariable

Beta SE P‑value Beta SE P‑value Age (years) 0.01 0.001 <0.001 0.005 0.001 <0.001 Gender ‑0.17 0.02 <0.001 Race ‑0.02 0.008 0.04 History of CVD 0.06 0.03 0.02Smoking status Smoking ‑0.06 0.03 0.05 ‑0.05 0.03 0.05Body composition BMI (kg/m2) ‑0.006 0.002 0.002 ‑0.004 0.002 0.02Blood pressure Systolic blood pressure (10 mmHg) ‑0.005 0.006 0.4 Diastolic blood pressure (10 mmHg) 0.01 0.01 0.3 Use of ACEi/ARB ‑0.04 0.02 0.08 Use of diuretics ‑0.10 0.02 <0.001Glucose homeostasis Diabetes duration ≥ 5 years ‑0.07 0.04 0.09 HbA1c (%) ‑0.03 0.007 <0.001 ‑0.02 0.006 <0.001 Use of insulin ‑0.09 0.02 <0.001Laboratory measurements Hemoglobin (g/dL) 0.09 0.006 <0.001 0.06 0.006 <0.001 Serum albumin (g/dL) 0.45 0.03 <0.001 0.28 0.03 <0.001Lipids Total cholesterol (10 mg/dL) ‑0.02 0.002 <0.001 ‑0.008 0.002 <0.001 HDL cholesterol (10 mg/dL) ‑0.03 0.007 <0.001 LDL cholesterol (10 mg/dL) ‑0.02 0.003 <0.001 Log triglycerides (log mg/dL) ‑0.20 0.04 <0.001Liver function Log ALT (log U/L) 0.34 0.05 <0.001 Log AST (log U/L) 0.52 0.07 <0.001 0.37 0.06 <0.001Renal function Log ACR (log mg/g) ‑0.26 0.03 <0.001 Serum creatinine (mg/dL) ‑0.09 0.02 <0.001 eGFR, MDRD (mL/min/1.73m2) 0.004 0.001 <0.001



lar disease; eGFR, estimated glomerular filtration rate; F, females; HDL, high‑density lipoprotein; LDL, low‑density

lipoprotein; M, males.

Variables tested in multivariable analyses: age, gender, race, smoking, history of CVD, BMI, diastolic blood pressure,

use of ACEi/ARB, use of diuretics, hemoglobin, serum albumin, total cholesterol, duration diabetes, HbA1c, use of

insulin, log AST, log ACR, eGFR.


Table 2B. Univariable and multivariable associations of log2-transformed bilirubin con-centrations with clinical parameters in IDNT.

IDNT

Univariable Multivariable

Beta SE P‑value Beta SE P‑value Age (years) 0.01 0.002 <0.001 0.007 0.002 <0.001 Gender ‑0.33 0.03 <0.001 ‑0.10 0.03 0.001 Race ‑0.05 0.008 <0.001 ‑0.04 0.007 <0.001 History of CVD 0.01 0.03 0.6Smoking status Smoking ‑0.11 0.04 0.002 ‑0.17 0.04 <0.001Body composition BMI (kg/m2) ‑0.02 0.002 <0.001 ‑0.01 0.002 <0.001Blood pressure Systolic blood pressure (10 mmHg) ‑0.01 0.007 0.2 Diastolic blood pressure (10 mmHg) 0.06 0.01 <0.001 0.04 0.01 <0.001 Use of ACEi/ARB ‑0.06 0.03 0.03 Use of diuretics ‑0.12 0.03 <0.001Glucose homeostasis Diabetes duration ≥ 5 years ‑0.10 0.05 0.04 ‑0.08 0.04 0.08 HbA1c (%) 0.003 0.008 0.7 0.02 0.008 0.06 Use of insulin ‑0.09 0.03 0.002Laboratory measurements Hemoglobin (g/dL) 0.12 0.007 <0.001 0.08 0.009 <0.001 Serum albumin (g/dL) 0.50 0.03 <0.001 0.26 0.03 <0.001Lipids Total cholesterol (10 mg/dL) ‑0.02 0.002 <0.001 ‑0.01 0.002 <0.001 HDL cholesterol (10 mg/dL) ‑0.03 0.01 0.02 LDL cholesterol (10 mg/dL) ‑0.02 0.003 <0.001 Log triglycerides (log mg/dL) ‑0.09 0.06 0.09Liver function Log ALT (log U/L) 0.22 0.06 <0.001 Log AST (log U/L) 0.30 0.08 <0.001Renal function Log ACR (log mg/g) ‑0.43 0.04 <0.001 Serum creatinine (mg/dL) ‑0.17 0.02 <0.001 eGFR, MDRD (mL/min/1.73m2) 0.006 0.001 <0.001 0.002 0.001 0.6



lar disease; eGFR, estimated glomerular filtration rate; F, females; HDL, high‑density lipoprotein; LDL, low‑density

lipoprotein; M, males.

Variables tested in multivariable analyses: age, gender, race, smoking, history of CVD, BMI, diastolic blood pressure,

use of ACEi/ARB, use of diuretics, hemoglobin, serum albumin, total cholesterol, duration diabetes, HbA1c, use of

insulin, log AST, log ACR, eGFR.

32

Cha

pter

2

Table 3. Associations of log2-transformed bilirubin concentrations with the composite renal endpoint of DSCR or ESRD in univariable (model 1) and multivariable models adjusted for potential confounding factors.

ModelRENAAL IDNT RENAAL & IDNT

HR (95% CI) P HR (95% CI) P HR (95% CI) P

1. 0.54 (0.45–0.65) <0.001 0.48 (0.43–0.55) <0.001 0.50 (0.45–0.55) <0.001

2. 0.59 (0.49–0.72) <0.001 0.51 (0.45–0.58) <0.001 0.53 (0.48–0.59) <0.001

3. 0.60 (0.50–0.73) <0.001 0.55 (0.48–0.63) <0.001 0.55 (0.49–0.61) <0.001

4. 0.73 (0.60–0.89) 0.002 0.64 (0.55–0.74) <0.001 0.65 (0.58–0.73) <0.001

5. 0.67 (0.55–0.83) <0.001 0.64 (0.55–0.76) <0.001 0.67 (0.59–0.76) <0.001

Model 1: crude.

Model 2: adjusted for age and sex.

Model 3: adjusted for age, sex, and baseline eGFR.

Model 4: adjusted for age, sex, baseline eGFR, and baseline log ACR.

Model 5: adjusted for age, sex, baseline eGFR, baseline log ACR, race, smoking, history of CVD, baseline BMI, total

cholesterol, DBP, HbA1c, treatment assignment, and log AST.

ACR, urinary albumin‑to‑creatinine ratio; AST, aspartate aminotransferase; BMI, body mass index; CI, confidence

interval; CVD, cardiovascular disease; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate.

ARB Treatment Effect

Because treatment with ARBs influence serum concentrations of hemoglobin (13) and could consequently affect bilirubin levels, we investigated the effect of treatment with an ARB (losartan in RENAAL and irbesartan in IDNT) on serum concentrations of hemoglobin and bilirubin. Hemoglobin and bilirubin concentrations over time in the RENAAL trial are shown in Figure 2A and 2B. Hemoglobin levels slowly decreased over time in the placebo group, while an initial decrease followed by a stabilization of hemoglobin levels was seen in losartan-treated patients (Figure 2A). After 48 months of treatment, hemoglobin levels were no longer significantly different between treatment groups (Figure 2A).

Bilirubin levels were slightly, but not significantly, lower in the losartan group than in the placebo group (Figure 2B). Despite the initial fall in hemoglobin levels, there was no initial fall in bilirubin levels in the losartan group. Bilirubin values decreased over time in both treatment groups, and no significant differences in bilirubin concentrations were observed between treatment groups after 12, 36, and 48 months.

Hemoglobin and bilirubin values of subjects using placebo and irbesartan in the IDNT trial are shown in Figure 2C and 2D. In general, the pattern of changes in these


Figu

re

2.

Hem

oglo

bin

(A,

C)

and

tota

l bi

lirub

in (

B, D

) co

ncen

trat

ions

(m

ean

± SE

M)

of s

ubje

cts

usin

g lo

sart

an o

r pl

aceb

o in

the

REN

AA

L tr

ial

or ir

besa

rtan

or p

lace

bo in

the

tria

l. Th

e in

sets

sho

w a

mor

e-de

taile

d ve

rsio

n of

eac

h gr

aph.

Th

e ch

arts

und

er e

ach

grap

h in

dica

te t

he n

umbe

r of

sub

-je

cts

with

ava

ilabl

e m

easu

re-

men

ts.

34

Cha

pter

2

markers over time in IDNT was similar to the RENAAL trial. Although hemoglobin levels significantly decreased after initiation of treatment with irbesartan (Figure 2C), as in RENAAL, no significant differences in bilirubin concentrations were observed be-tween treatment groups (Figure 2D).

DiscussionIn this historical prospective analysis of the RENAAL trial, we found an independent inverse association of bilirubin levels with progression of nephropathy in patients with type 2 diabetes. This finding was independently replicated in IDNT. Furthermore, we showed that treatment with the ARBs losartan or irbesartan did not result in a decrease in bilirubin concentrations, despite an initial decrease in hemoglobin levels.

One of the major pathophysiologic mechanisms that has been identified in the development and progression of DN is oxidative stress, described as increased levels of reactive oxygen species (18–20). Bilirubin is known to be a potent endogenous antioxi-dant (2), and a recent study in rodents found a protective effect of bilirubin against DN through inhibition of oxidative stress by downregulation of renal NADPH oxidase (5).

In humans, several cross-sectional studies have provided additional evidence for a protective effect of bilirubin on DN. Inoguchi et al. (6) showed a lower prevalence of vascular complications as well as reduced markers of oxidative stress and inflamma-tion in patients with Gilbert syndrome (a congenital hyperbilirubinemia) and diabetes. Fukui et al. (8) reported a negative correlation of bilirubin with log ACR and a posi-tive correlation with eGFR. In addition, it was shown that bilirubin levels were higher in patients without DN than in those with DN (7). However, these studies were cross-sectional in design, precluding investigation of the prospective association of bilirubin with renal impairment. To our knowledge, the current prospective study is the first to indicate an inverse association of bilirubin and progression of nephropathy toward ESRD in type 2 diabetes.

In animal models, antioxidants have been shown to be effective in treating DN (21,22). In combination with the current human data showing an independent asso-ciation between bilirubin and renal outcome, we speculate that treatments intended to slightly raise bilirubin levels might have a beneficial effect on progression of nephropa-thy in patients with type 2 diabetes and low bilirubin levels.

A moderate increase in bilirubin levels could be attained through induction of heme oxygenase-1 (HO-1), the enzyme that catalyzes the rate-limiting step in heme deg-radation. HO-1 splits heme into carbon monoxide (CO) and biliverdin, which is subse-quently reduced to bilirubin (2,23). The HO-1 system and heme degradation products CO, biliverdin, and bilirubin have repeatedly been shown to have renoprotective prop-erties (2,23). Therefore, the renoprotective effects of bilirubin in this study are possibly, and at least partly, mediated by induction of HO-1 and by-products of heme degradation


(i.e., CO, biliverdin). A study in rats showed that induction of HO-1 reduces renal oxida-tive stress and protects against diabetes-related renal injury (24). HO-1 is a highly induc-ible enzyme, which can be induced by many drugs routinely used in clinical medicine (i.e., nonsteroidal anti-inflammatory drugs [NSAIDs], peroxisome proliferator-activat-ed receptor α agonists) (4). However, given the adverse effects of NSAIDs, long-term use of NSAIDs is not recommended in patients with advanced renal function impairment. Natural HO-1 inducers include curcuma and polyphenols (i.e., resveratrol) (4,25). Par-tial inhibition of conjugation of bilirubin by uridine diphosphate-glucuronyltransferase, an enzyme encoded by the UGT1A1 gene, reduces bilirubin excretion and is known to result in mild increases in bilirubin concentrations (4,23,26). Pharmaceuticals capable of a partial inhibition of UGT1A1 are probenecid and atazanavir (4).

A number of studies have reported that the use of ACE inhibitors and ARBs de-crease hemoglobin levels (13–15), which can be enhanced by the use of diuretics (15). Be-cause bilirubin is a product of heme catabolism, changes in hemoglobin levels could sub-sequently influence bilirubin concentrations. Although the use of losartan and irbesartan slightly, but significantly, decreased hemoglobin levels compared with placebo, it did not affect bilirubin levels in either trial. Several studies reported that the use of ACE inhibi-tors and ARBs reduce erythropoietin and, consequently, hemoglobin levels by blocking the effects of angiotensin II on erythropoiesis (27,28). Because the enzymatic degradation of hemoglobin by HO-1 is known to be the rate-limiting step in the formation of bilirubin (23,29) and not the synthesis of hemoglobin, it is conceivable that small changes in the synthesis of hemoglobin do not affect the formation and levels of bilirubin.

This study has several limitations. First, patients with liver enzymes (ALT, AST) and bilirubin levels >1.5 times the ULN were excluded from participation in both trials, which resulted in a relatively narrow range of bilirubin concentra-tions (i.e., 0.1–2.1 mg/dL, with a mean value of 0.57 mg/dL in RENAAL and 0.54 mg/dL in IDNT). In earlier cross-sectional studies on the association of bilirubin with DN, bilirubin levels were higher, with values of 1.4 (1.3–1.6) mg/dL in subjects with Gilbert syndrome in the study of Inoguchi et al. (6), and 0.71 ± 0.21 mg/dL in subjects with type 2 diabetes in the study of Fukui et al. (8). In a prospective study on development and progression of albuminuria in patients with type 2 diabetes of Mashitani et al. (30), mean bilirubin levels were 0.63 ± 0.28 mg/dL. Within the relatively small range of bilirubin lev-els in the current study, we could not identify a non-linear component in the association of bilirubin with the renal end point. A larger range of bilirubin concentrations in future studies might allow for identification of a cutoff value of bilirubin above which the as-sociation with progression of renal function might flatten, which could help to identify an optimal target concentration for bilirubin in intervention trials. Second, in both RE-NAAL and IDNT, only total bilirubin was measured. Direct (conjugated) bilirubin was not measured separately because serum bilirubin consists for >95% of indirect (uncon-jugated) bilirubin (26), and subjects with total bilirubin levels >1.5 times the ULN were excluded from participation in both trials. Therefore, examining differences between un-

36

Cha

pter

2

conjugated and conjugated bilirubin levels was not possible. Furthermore, given the ob-servational nature of this study and the inability to focus on the HO-1 system and its by-products in more detail, it is impossible to draw a definitive conclusion about the causal-ity of bilirubin and progression of DN. Mendelian randomization has been proposed as a method that enables estimation of causal relationships in observational studies (31,32). This method uses genotype to estimate causal relationships between a gene product and physiological outcomes (32). Because there is a strong relation between UGT1A1 (geno-type) and bilirubin levels (phenotype) (32), mendelian randomization can be used to establish a possible causal relation between bilirubin and DN. The strengths of this study are the large sample size, the large number of renal events, and the independent replica-tion of the current findings in another large cohort of >1,700 subjects.

In conclusion, the results show an independent inverse association of bilirubin levels with progression of nephropathy in patients with type 2 diabetes, suggesting that measurement of bilirubin may identify subjects at risk for renal disease progression. In addition, the study suggests a protective effect of bilirubin against progression of DN, thereby potentially implying its role as an antioxidant.

AcknowledgementsThis research was performed within the framework of CTMM, the Center for Trans-lational Molecular Medicine (www.ctmm.nl), project PREDICCt (grant 01C-104), and supported by the Dutch Heart Foundation, Dutch Diabetes Research Foundation and Dutch Kidney Foundation. The work leading to this paper has received funding from the European Community’s Seventh Framework Programme under grant agree-ment no. HEALTH–F2–2009–241544 (Syskid). IJR and SJLB received support from the Netherlands heart foundation, Dutch Diabetes Research Foundation and Dutch Kidney Foundation, together participating in the framework of the CTMM project PREDICCt. HJLH is supported by a VENI grant from the Netherlands Scientific Organisation. The RENAAL trial was funded by Merck & Co. The IDNT trial was sponsored by Bristol Myer Squibb Institute for Medical Research and Sanofi-Synthelabo. DdZ and MEC have received financial support from Merck for their participation in the RENAAL Steering Committee. No other potential conflicts of interest relevant to this article were reported. We acknowledge the supportive role of all RENAAL and IDNT investigators, support staff, and participating patients.


References 1. American Diabetes Association. Standards of medical care in diabetes 2012. Diabetes Care

2012 Jan;35 Suppl 1:S11–63. 2. Adin CA, Croker BP, Agarwal A. Protective effects of exogenous bilirubin on ischemia-


3. Lin JP, Vitek L, Schwertner HA. Serum bilirubin and genes controlling bilirubin concentrations as biomarkers for cardiovascular disease. Clin Chem 2010 Oct;56(10):1535–1543.


5. Fujii M, Inoguchi T, Sasaki S, Maeda Y, Zheng J, Kobayashi K, et al. Bilirubin and biliverdin protect rodents against diabetic nephropathy by downregulating NAD(P)H oxidase. Kidney Int 2010 Aug 4.

6. Inoguchi T, Sasaki S, Kobayashi K, Takayanagi R, Yamada T. Relationship between Gilbert syndrome and prevalence of vascular complications in patients with diabetes. JAMA 2007 Sep 26;298(12):1398–1400.

7. Zelle DM, Deetman N, Alkhalaf A, Navis G, Bakker SJ. Support for a protective effect of biliru-bin on diabetic nephropathy in humans. Kidney Int 2011 Mar;79(6):686; author reply 686–7.

8. Fukui M, Tanaka M, Shiraishi E, Harusato I, Hosoda H, Asano M, et al. Relationship be-tween serum bilirubin and albuminuria in patients with type 2 diabetes. Kidney Int 2008 Nov;74(9):1197–1201.

9. Brenner BM, Cooper ME, de Zeeuw D, Keane WF, Mitch WE, Parving HH, et al. Effects of losartan on renal and cardiovascular outcomes in patients with type 2 diabetes and nephropa-thy. N Engl J Med 2001 Sep 20;345(12):861–869.

10. Brenner BM, Cooper ME, de Zeeuw D, Grunfeld JP, Keane WF, Kurokawa K, et al. The losar-tan renal protection study--rationale, study design and baseline characteristics of RENAAL (Reduction of Endpoints in NIDDM with the Angiotensin II Antagonist Losartan). J Renin Angiotensin Aldosterone Syst 2000 Dec;1(4):328–335.

11. Lewis EJ, Hunsicker LG, Clarke WR, Berl T, Pohl MA, Lewis JB, et al. Renoprotective effect of the angiotensin-receptor antagonist irbesartan in patients with nephropathy due to type 2 diabetes. N Engl J Med 2001 Sep 20;345(12):851–860.

12. Rodby RA, Rohde RD, Clarke WR, Hunsicker LG, Anzalone DA, Atkins RC, et al. The Irbesar-tan type II diabetic nephropathy trial: study design and baseline patient characteristics. For the Collaborative Study Group. Nephrol Dial Transplant 2000 Apr;15(4):487–497.

13. Mohanram A, Zhang Z, Shahinfar S, Lyle PA, Toto RD. The effect of losartan on hemoglobin concentration and renal outcome in diabetic nephropathy of type 2 diabetes. Kidney Int 2008 Mar;73(5):630–636.

14. Robles NR, Angulo E, Grois J, Barquero A. Comparative effects of fosinopril and irbesartan on hematopoiesis in essential hypertensives. Ren Fail 2004 Jul;26(4):399–404.

38

Cha

pter

2

15. Slagman MC, Sinkeler SJ, Hemmelder MH, Waanders F, Vogt L, Kluin-Nelemans HC, et al. Erythropoietin is reduced by combination of diuretic therapy and RAAS blockade in proteinuric renal patients with preserved renal function. Nephrol Dial Transplant 2010 Oct;25(10):3256–3260.

16. Packham DK, Alves TP, Dwyer JP, Atkins R, de Zeeuw D, Cooper M, et al. Relative incidence of ESRD versus cardiovascular mortality in proteinuric type 2 diabetes and nephropathy: results from the DIAMETRIC (Diabetes Mellitus Treatment for Renal Insufficiency Consortium) da-tabase. Am J Kidney Dis 2012 Jan;59(1):75–83.

17. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 1999 Mar 16;130(6):461–470.

18. Baynes JW. Role of oxidative stress in development of complications in diabetes. Diabetes 1991 Apr;40(4):405–412.

19. Giugliano D, Ceriello A, Paolisso G. Oxidative stress and diabetic vascular complications. Dia-betes Care 1996 Mar;19(3):257–267.

20. Giacco F, Brownlee M. Oxidative stress and diabetic complications. Circ Res 2010 Oct 29;107(9):1058–1070.

21. Stanton RC. Oxidative stress and diabetic kidney disease. Curr Diab Rep 2011 Aug;11(4):330–336. 22. Koya D, Hayashi K, Kitada M, Kashiwagi A, Kikkawa R, Haneda M. Effects of antioxidants

in diabetes-induced oxidative stress in the glomeruli of diabetic rats. J Am Soc Nephrol 2003 Aug;14(8 Suppl 3):S250–3.

23. Abraham NG, Cao J, Sacerdoti D, Li X, Drummond G. Heme oxygenase: the key to renal func-tion regulation. Am J Physiol Renal Physiol 2009 Nov;297(5):F1137–52.

24. Elmarakby AA, Faulkner J, Baban B, Sullivan JC. Induction of hemeoxygenase-1 reduces renal oxidative stress and inflammation in diabetic spontaneously hypertensive rats. Int J Hypertens 2012;2012:957235.

25. Barbagallo I, Galvano F, Frigiola A, Cappello F, Riccioni G, Murabito P, et al. Potential thera-peutic effects of natural heme oxygenase-1 inducers in cardiovascular diseases. Antioxid Redox Signal 2013 Feb 10;18(5):507–521.

26. Fevery J. Bilirubin in clinical practice: a review. Liver Int 2008 May;28(5):592–605. 27. Mrug M, Stopka T, Julian BA, Prchal JF, Prchal JT. Angiotensin II stimulates proliferation of

normal early erythroid progenitors. J Clin Invest 1997 Nov 1;100(9):2310–2314. 28. Freudenthaler SM, Schreeb K, Korner T, Gleiter CH. Angiotensin II increases erythropoietin

production in healthy human volunteers. Eur J Clin Invest 1999 Oct;29(10):816–823. 29. Sassa S, Kappas A, Bernstein SE, Alvares AP. Heme biosynthesis and drug metabolism in mice

with hereditary hemolytic anemia. Heme oxygenase induction as an adaptive response for maintaining cytochrome P-450 in chronic hemolysis. J Biol Chem 1979 Feb 10;254(3):729–735.

30. Mashitani T, Hayashino Y, Okamura S, Tsujii S, Ishii H. Correlations between serum bilirubin levels and diabetic nephropathy progression among Japanese type 2 diabetic patients: a pro-spective cohort study (Diabetes Distress and Care Registry at Tenri [DDCRT 5]). Diabetes Care 2014 Jan;37(1):252–258.


31. Verduijn M, Siegerink B, Jager KJ, Zoccali C, Dekker FW. Mendelian randomization: use of genetics to enable causal inference in observational studies. Nephrol Dial Transplant 2010 May;25(5):1394–1398.

32. McArdle PF, Whitcomb BW, Tanner K, Mitchell BD, Shuldiner AR, Parsa A. Association be-tween bilirubin and cardiovascular disease risk factors: using Mendelian randomization to as-sess causal inference. BMC Cardiovasc Disord 2012 Mar 14;12:16.

40

Cha

pter

2

Supplementary Table S1. Inclusion and exclusion criteria for the RENAAL and IDNT trials.

RENAAL IDNT

Inclusion criteria

Age range (years) 31 – 70 30 – 70

Diagnosis of Type 2 Diabetes Yes Yes

Diabetic Nephropathy

Proteinuria/albuminuriaACR ≥ 300m/g or 24h urinary

protein excretion > 500 mg24h urinary protein excretion ≥ 900

mg

SCr Women: SCr 1.3 – 3.0 mg/dL

Men: SCr 1.5 – 3.0 mg/dLWomen: SCr 1.0–3.0 mg/dL

Men: SCr 1.2–3.0 mg/dL

HypertensionHypertension and seated SBP/DBP ≤ 200/110 mmHg or normotension

(SBP ≥ 100 mmHg)

Seated SBP >135 mmHg and/or seated DBP > 85 mmHg or receiving

antihypertensive medication

Exclusion criteria

Other diseasesHistory of non‑diabetic renal disease

or type 1 diabetesAge onset type 2 diabetes < 20 years

or diagnosis of type 1 diabetes

Concomitant therapiesPatients with absolute requirements

for ACEI or ARB

Patients with absolute requirements for ACEI, calcium antagonists or

ARB

Cardiovascular disease

• History of MI or CABG within past month of study entry

• CVA or PTCA within past 6 months

• TIA within past 12 months• History of HF

• Unstable AP, MI, CABG, PTCA, or CVA within past 3 months

of study entry• TIA within past 6 months• History of HF (class III or

IV NYHA)

Years of conduct of the trials 1996 – 2000 1996 – 2000

Abbreviations: ACR, urinary albumin‑to‑creatinine ratio; AP, Angina Pectoris; CABG, coronary artery bypass

grafting; CVA, cerebrovascular accident; DBP, diastolic blood pressure; HF, heart failure; MI, myocardial infarction;

NYHA, New York Heart Association; PTCA, percutaneous transluminal coronary angioplasty; SBP, systolic blood

pressure; SCr, serum creatinine; TIA, transient ischemic attack.


Supp

lem

enta

ry T

able

S2.

Tim

e co

urse

of m

ain

patie

nt ch

arac

teris

tics o

f all

subj

ects

in th

e RE

NA

AL

and

IDN

T tr

ials

and

stra

tified

by

tert

iles a

c-co

rdin

g to

bas

elin

e bi

lirub

in co

ncen

trat

ions

.

RENA

ALID

NT

Tim

eAl

l sub

ject

sTe

rtile

1Te

rtile

2Te

rtile

3Al

l sub

ject

sTe

rtile

1Te

rtile

2Te

rtile

3

Varia

ble

(Mon

th)

NM

ean

NM

ean

NM

ean

NM

ean

NM

ean

NM

ean

NM

ean

NM

ean

Bilir

ubin

Ba

selin

e14

980.

5737

40.

3877

60.

5434

80.

8417

070.

5458

80.

3441

30.

5070

60.

72(m

g/dL

)12

1165

0.53

280

0.39

609

0.51

276

0.71

1450

0.52

479

0.38

360

0.48

611

0.66

2495

20.

4920

70.

3649

80.

4724

70.

6411

570.

5035

00.

3628

20.

4752

50.

6136

599

0.48

106

0.37

314

0.44

179

0.62

613

0.47

163

0.30

151

0.43

299

0.58

4816

20.

4820

0.39

830.

4459

0.56

218

0.47

630.

3153

0.44

102

0.59

SBP

Base

line

1498

153

374

152

776

153

348

151

1707

159

588

160

413

159

706

159

(mm

Hg)

1213

1714

732

114

968

514

831

114

614

8614

349

114

337

014

262

514

424

1087

144

237

146

577

144

273

142

1149

142

340

142

282

141

527

142

3661

514

210

114

233

114

218

314

061

614

116

114

215

014

030

514

148

9614

211

141

4714

238

142

205

139

5314

251

137

101

139

DBP

Bas

eline

1498

8237

482

776

8234

883

1707

8758

886

413

8770

688

(mm

Hg)

1213

1779

321

8068

579

311

8014

8679

491

7737

079

625

8024

1087

7723

779

577

7727

377

1148

7833

977

282

7852

779

3661

575

101

7633

175

183

7661

678

161

7715

078

305

7848

9674

1173

4774

3874

204

7653

7651

7610

075

HbA

1C

Base

line

1490

8.5

371

8.8

773

8.5

346

8.2

1553

8.1

553

8.1

368

8.3

632

8.1

(%)

1211

608.

627

68.

960

88.

627

68.

314

538.

348

08.

336

08.

661

38.

224

936

8.6

202

9.0

490

8.6

244

8.2

1145

8.2

342

8.2

277

8.6

526

8.1

3658

48.

410

48.

830

58.

517

58.

261

98.

216

58.

115

28.

430

28.

148

159

8.4

198.

281

8.5

598.

221

48.

161

8.2

528.

010

18.

0

42

Cha

pter

2

RENA

ALID

NT

Tim

eAl

l sub

ject

sTe

rtile

1Te

rtile

2Te

rtile

3Al

l sub

ject

sTe

rtile

1Te

rtile

2Te

rtile

3

Varia

ble

(Mon

th)

NM

ean

NM

ean

NM

ean

NM

ean

NM

ean

NM

ean

NM

ean

NM

ean

HbA

1C

Base

line

1490

6937

173

773

6934

666

1553

6555

365

368

6863

265

(mm

ol/m

ol)

1211

6070

276

7360

870

276

6714

5367

480

6736

070

613

6624

936

7020

275

490

7124

466

1145

6734

266

277

7052

665

3658

469

104

7230

569

175

6661

966

165

6515

269

302

6548

159

6819

6681

6959

6721

465

6166

5264

101

64Ch

oles

tero

l Ba

selin

e14

9822

837

424

477

622

634

821

615

6422

855

823

937

122

963

521

8(m

g/dL

)12

1165

221

280

233

609

219

276

212

1445

220

479

225

359

223

607

214

2495

221

020

722

149

820

924

720

311

4921

034

820

928

220

951

921

036

600

202

107

215

314

202

179

194

613

203

163

206

150

206

300

200

4816

219

220

210

8319

559

183

216

197

6320

651

201

102

190

SCr

Base

line

1498

1.9

374

1.9

776

1.9

348

1.8

1698

1.7

584

1.8

412

1.7

702

1.6

(mg/

dL)

1212

672.

330

42.

666

72.

329

62.

114

532.

048

02.

436

12.

061

21.

824

1064

2.5

226

2.8

569

2.6

269

2.3

1160

2.2

351

2.5

283

2.2

526

2.0

3660

72.

799

3.1

330

2.7

178

2.6

614

2.3

163

2.7

150

2.3

301

2.1

4894

2.9

112.

847

3.2

382.

721

82.

463

2.8

532.

410

22.

3

DBP

, dia

stolic

blo

od p

ress

ure;

SBP,

systo

lic b

lood

pre

ssur

e; SC

r, se

rum

crea

tinin

e.

Supp

lem

enta

ry T

able

S2.

(con

tinue

d)


Supp

lem

enta

ry T

able

S3.

Mea

n an

nual

chan

ge in

tota

l bili

rubi

n co

ncen

trat

ions

in su

bjec

ts w

ith co

mpl

ete b

iliru

bin

mea

sure

men

ts at

bas

elin

e, 12

, 24

, and

36

mon

ths i

n RE

NA

AL

(n =

553)

and

IDN

T (n

= 5

87). RE

NAAL

IDN

T

Tim

e (M

onth

s)Te

rtile

1Te

rtile

2Te

rtile

3P‑

valu

eTe

rtile

1Te

rtile

2Te

rtile

3P‑

valu

e

Δ Bi

lirub

in (m

g/dL

)0–

120.

061

‑0.0

18‑0

.126

<0.0

010.

068

0.02

6‑0

.026

<0.0

01

12–2

4‑0

.067

‑0.0

55‑0

.071

0.5

‑0.0

39‑0

.042

‑0.0

630.

3

24–3

60.

002

‑0.0

35‑0

.026

0.2

‑0.0

70‑0

.049

‑0.0

620.

6

3

Plasma bilirubin and late

graft failure in renal

transplant recipients3Petronella E. Deetman, Dorien M. Zelle, Jaap J. Homan van der Heide, Gerjan J. Navis, Reinold O.B. Gans, Stephan J.L. Bakker

Department of Medicine, Division of Nephrology, University of Groningen, University Medical Center Groningen, Groningen, the Netherlands

Transpl Int 2012;25(8):876–81

46

Cha

pter

3

AbstractBackground. Exogenous bilirubin has been shown to protect against oxidative stress in ischemia-reperfusion injury. Oxidative stress has been implicated in the pathophysi-ology of chronic transplant dysfunction leading to late graft failure after renal transplan-tation. We prospectively investigated whether high endogenous bilirubin is protective against development of late graft failure in renal transplant recipients (RTR). Methods. Baseline data were collected between August 2001 and July 2003 in non-icteric outpatient RTR with a functioning graft for > 1 year. At baseline, bilirubin and liver en-zymes were measured by routine assays on a Merck Mega analyzer. Graft failure was prospectively recorded until May 19 2009. Results. During follow-up for 7.1 (6.2–7.2) years, 55 RTR developed graft failure. We found that circulating levels of bilirubin are inversely associated with late graft failure in RTR (HR = 0.29 [95% CI 0.16–0.52], P < 0.001), This association was independent of potential confounders, including creatinine clearance, urinary protein excretion, calci-neurin inhibitors and gender (HR = 0.31 [95% CI 0.15–0.62] P = 0.001). Conclusions. Our findings are consistent with a protective effect of increased endog-enous bilirubin against development of late graft failure in RTR. If our findings are con-firmed by other studies, intervention with endogenous or exogenous bilirubin may be of interest for long-term preservation of renal function in RTR.

47Plasma bilirubin and late graft failure

IntroductionShort-term outcome of renal transplantation has improved impressively during the last decades (1). However, beyond one year after transplantation, outcome is still poor be-cause many renal transplant recipients (RTR) experience a slow, but relentless decline in graft function (2–5). This will ultimately lead to graft failure, necessitating renewal of dialysis or re-transplantation (2–5). Frequently, the slow decline in graft function is not the consequence of immunological rejection, but an entity that is called chronic transplant dysfunction (3). The exact pathophysiology is not known, but oxidative stress has been implicated (6–9). Endogenous bilirubin is an established cytoprotectant and anti-oxidant (10–12). In experimental studies, it has been shown that exogenous bilirubin is protective against oxidative stress after ischemia-reperfusion injury in the kidney (13). It is, however, not known whether endogenous levels of bilirubin influ-ence long term renal outcome in renal transplant recipients. Because oxidative stress has been implicated a detrimental role in the process of chronic transplant dysfunction (6–9), we hypothesized that high circulating bilirubin concentrations protect against late graft failure in RTR. We prospectively investigated this hypothesis a large single center cohort of RTR.


Study population

To investigate whether high circulating bilirubin concentrations protect against late graft failure, we collected data in non-icteric outpatient RTR with a functioning graft for > 1 year. Baseline data were collected between August 2001 and July 2003 at a median of six years after transplantation. All 603 patients available for analysis signed written informed consent. Standard immunosuppressive therapy was given as described previ-ously (14). Approval for this study has been obtained by the Institutional Review Board (METc 2001/039). Detailed explanation of this study has been published before (14–16).

Laboratory analyses

Creatinine clearance was assessed by the Jaffé method on a MEGA AU 510 (Merck Diag-nostica, Darmstadt, Germany) and calculated using the Cockcroft-Gault formula. Con-centration of total endogenous bilirubin was measured using the Bilirubin DPD method on an MEGA (Merck Diagnostica, Darmstadt, Germany). In our laboratory, a bilirubin concentration lower than 17 μmol/L is considered normal.

48

Cha

pter

3

Delayed graft function was defined as oliguria for more than 6 days. Allograft rejections were biopsy proven. To assess change in renal function over time, difference between creatinine clearance at baseline and latest creatinine clearance up to 4 years beyond baseline was calculated. All participating subjects visited the out-patient clinic at least once a year and creatinine clearance was assessed at every visit. Graft failure was defined as the return to dialysis or retransplantation and was censored for death. Graft failure and mortality of all RTR were prospectively recorded until May 19, 2009. There was no loss to follow-up.

Statistical analyses

Statistical calculations were performed using SPSS (version 18.0, SPSS inc. Chicago, IL). Total bilirubin concentration was divided into sex-stratified tertiles. Normally dis-tributed data are given as mean ± SD, whereas skewed distributed data are expressed as median (interquartile range), categorical distributed variables are given as number (percentage). Skewed variables were normalized using logarithmic transformation. Cal-culations have been made using ANOVA for normally distributed variables, Chi-square test for categorical distributed variables and a Kruskal Wallis test in the case of a skewed distribution. Initial survival analyses were performed according to Kaplan-Meier with log-rank testing. To determine whether bilirubin concentrations were independently as-sociated with outcome, we subsequently performed cumulative adjustment for potential confounders in multivariate Cox-regression analyses. A probability value of < 0.05 was considered significant.

ResultsIn 603 RTR (age 51 ± 12 years, 55% men), median (IQR) baseline bilirubin concentra-tion was 17 (14–21) µmol/L. Bilirubin concentration was significantly higher in men than in women (P < 0.001). In women, median bilirubin concentration was 16 (13–20) µmol/L. In men, median bilirubin concentration was 18 (15–23) µmol/L. In 318 (53%) RTR bilirubin concentration was above the range considered normal for the gen-eral population. However, none of the included RTR was icteric or suffered from severe liver disease. Median time between transplantation and baseline measurements was 6.0 (2.6–11.4) years. In Table 1, recipient and donor characteristics are shown according to sex-stratified tertiles of bilirubin concentration. Bilirubin was inversely associated with use of ACE inhibitors or angiotensin II antagonists, cerebrovascular accidents, HbA1c, serum creatinine and urinary protein and sodium excretion. Bilirubin was positively as-sociated with ASAT, ALAT and use of calcineurin inhibitors.


Decline in serum creatinine clearance was least pronounced in RTR with highest bilirubin concentrations (Figure 1, P < 0.001). In line with this observation, graft failure was less frequent in RTR with high bilirubin concentrations than in RTR with lower bilirubin concentrations, with frequencies of 16, 10 and 2% in the respective tertiles after a median follow-up of 6.9 (6.1–7.4) years (log-rank P < 0.001). A Kaplan-Meier analysis of graft failure is shown in Figure 2A. Estimated 5-year graft survival was 88% in the first, 92% in the second and 98.5% in the third tertile of bilirubin distribution. Mortality rates were similar in all three tertiles (log-rank P = 0.70). A Kaplan-Meier analysis of mortal-ity is shown in Figure 2B. Estimated 5-year patient survival was 84% in the three tertiles of bilirubin distribution.

Table 1. Demographic characteristics according to sex-stratified tertiles of bilirubin.

Tertiles of serum bilirubin concentration

I II III P‑value

N 181 218 204

Men, n (%) 101 (56) 116 (53) 113 (55)

Serum bilirubin (μmol/L) 13 (6–15) 17 (16–20) 25 (21–62)

Women, n (%) 80 (44) 102 (47) 91 (45)

Serum bilirubin (μmol/L) 12 (7–13) 15 (14–18) 22 (19–47)

Recipient demographics

Age (yr) 49.8 ± 11.4 51.9 ± 12 52.3 ± 12.7 0.10

Body composition

BMI (kg/m2) 25.8 ± 4.1 25.8 ± 4.2 25.4 ± 4.6 0.51

Waist circumference (cm) 96.7 ± 13.5 97.6 ± 13.5 97.1 ± 14.2 0.78

Smoking status 0.08

Non‑smoker, n (%) 47 (26) 77 (36) 91 (45)

Current smoker, n (%) 63 (35) 44 (20) 25 (12)

Ex‑smoker, n (%) 10 (39) 97 (45) 87 (43)

Blood pressure

Systolic (mmHg) 151.0 ± 22.1 156 ± 25.9 151.5 ± 19.4 0.06

Diastolic (mmHg) 89.6 ± 10.2 89.7 ± 9.9 90.4 ± 9.6 0.72

Use of ACE‑I or ARB, n (%) 78 (43) 78 (43) 50 (25) 0.001

Number of antihypertensives 1.9 ± 1.1 2.0 ± 1.2 1.8 ± 1.1 0.08

Lipids

Total cholesterol (mmol/L) 5.6 ± 1.2 5.6 ± 1.1 5.6 ± 0.9 0.88

HDL (mmol/L) 1.1 ± 0.3 1.1 ± 0.3 1.1 ± 0.3 0.30

LDL (mmol/L) 3.5 ± 1.1 3.6 ± 1.0 3.5 ± 0.9 0.80

Triglycerides (mmol/L) 2.2 ± 1.3 2.2 ± 1.2 2.1 ± 1.2 0.74

Use of statins, n (%) 88 (49) 106 (49) 105 (52) 0.80

50

Cha

pter

3

Tertiles of serum bilirubin concentration

I II III P‑value

Diabetes Mellitus

Insulin (μU/ml) 13.6 (8.6–16.3) 10.9 (7.8–16.9) 10.8 (7.7–14.8) 0.26

Glucose (mmol/L) 4.7 (4.2–5.1) 4.5 (4.1–5) 4.5 (4–4.9) 0.30

Diabetes after Tx, n (%) 31 (17) 38 (17) 37 (18) 0.96

Use of antidiabetic drugs, n (%) 26 (14) 29 (13) 24 (12) 0.75

CRP (mg/L) 2.7 (0.8–6.5) 2.1 (0.9–4.6) 1.8 (0.7–4.0) 0.15

Liver function

ASAT (U/L) 21 (17–25) 22 (19–27) 24 (20–29) <0.001

ALAT (U/L) 17 (13–23) 19 (14–25) 18 (15–25) 0.01

Alkaline phosphatase (U/L) 72 (57–90) 75 (58–98) 71 (56–94) 0.34

Gamma glutamyl transferase (U/L) 23 (16–36) 25 (17–40) 24 (18–41) 0.27

CMV status

CMV seropositivity, n (%) 119 (66) 165 (76) 146 (72) 0.09

Donor demographics

Age (yr) 38 (16) 37 (15) 36 (15) 0.51

Men, n (%) 98 (54) 122 (56) 107 (53) 0.80

Number of HLA mismatches, 2 (1–2.5) 2 (0–3) 2 (1–3) 0.79

Time between Tx and baseline (yr) 6.1 (3–12) 5.6 (3–12) 5.9 (3–11) 0.38

Dialysis duration prior to Tx (mo) 26 (14–47) 27 (15–49) 28 (11–51) 0.40

Renal allograft function

Serum creatinine (μmol/L) 141 (112–178) 136 (113–165) 128 (112–149) 0.009

Creatinine clearance (ml/min) 60.1 ± 23.1 62.5 ± 22.9 63.4 ± 21.5 0.34

Urinary protein excretion (g/24h) 0.3 (0.1–0.6) 0.2 (0.0–0.5) 0.2 (0.0–0.4) <0.001

Transplantation type

Postmortem donor, n (%) 149 (82) 189 (87) 183 (90) 0.11

Living donor, n (%) 32 (18) 29 (13) 21 (10) 0.11

Acute rejection, n (%) 92 (51) 86 (39) 93 (46) 0.07

Delayed graft function, n (%) 28 (16) 25 (12) 22 (11) 0.36

Immunosuppression

Prednisolon dose (mg) 10 (7.5–10) 10 (7.5–10) 10 (7.5–10) 0.75

Use of calcineurin inhibitor, n (%) 122 (67) 177 (81) 174 (85) <0.001

Use of ciclosporin, n (%) 83 (46) 147 (67) 158 (77) <0.001

Use of tacrolimus, n (%) 39 (22) 30 (14) 16 (8) 0.001

Use of proliferation inhibitor, n (%) 143 (79) 160 (73) 142 (70) 0.11

Normally distributed data are given as mean ± SD, skewed distributed data as median (interquartile range) and cat‑

egorical distributed variables as number (percentage). BMI, body mass index; ACEi, ACE inhibitor; ARB, angioten‑

sin receptor blocker; CRP, C‑reactive protein; ASAT, aspartate aminotransferase; ALAT, alanine aminotransferase;

CMV, Cytomegalovirus; Tx, Transplantation.

Table 1. (continued)


Table 2. Cox regression analyses for late graft failure in RTR in sex-stratified tertiles of bilirubin concentration.

Graft failure

HR (95% CI) P‑value

Model 1 0.29 (0.16–0.52) <0.001

Model 2 0.28 (0.15–0.51) <0.001

Model 3 0.33 (0.18–0.61) <0.001

Model 4 0.32 (0.17–0.61) 0.001

Model 5 0.35 (0.18–0.67) 0.001

Model 6 0.31 (0.15–0.62) 0.001

Hazard ratio’s per 10 μmol/L change in bilirubin concentration

Model 1: crude association for bilirubin

Model 2: Model 1 + adjustments for age and sex

Model 3: Model 2 + adjustments for creatinine clearance and urinary protein excretion

Model 4: Model 3 + adjustments for use of calcineurin inhibitor, use of ARB or ACEi, number of antihypertensives

Model 5: Model 4 + adjustments for AST, ALT

Model 6: Model 5 + adjustments for CMV seropositivity, smoking status, systolic blood pressure, acute rejection

ARB, angiotensin receptor blocker; ACEi, ACE inhibitor; AST, aspartate aminotransferase; ALT, alanine amino‑

transferase; CMV, cytomegalovirus.

In Cox-regression analyses for bilirubin as a continuous variable, the inverse as-sociation of bilirubin concentration with late graft failure was confirmed, with a hazard ratio (HR) of 0.29 (95% CI 0.16–0.52), P < 0.001 for the unadjusted analysis (Table 2). After cumulative adjustment in multivariate analyses, this association appeared inde-pendent of potential confounders, including creatinine clearance, urinary protein excre-tion, use of calcineurin inhibitors and gender, with a HR of 0.31 (95% CI 0.15–0.62), P = 0.001 for the final model.

52

Cha

pter

3

Figure 1. Changes in creatinine clearance (mean ± SEM) between baseline and follow-up ac-cording to sex-stratified tertiles of serum bilirubin. Differences between groups were tested with Kruskal Wallis test.

Figure 2. Kaplan-Meier curves for graft failure (A; P < 0.001) and mortality (B; P = 0.7) accord-ing to sex-stratified tertiles of bilirubin (I-III).

A

B


DiscussionIn this prospective study we investigated whether endogenous bilirubin assessed in sta-ble, outpatient RTR at more than 1 year after transplantation is associated with subse-quent development of late graft failure. We found that relatively high circulating levels of bilirubin are independently associated with low risk of late graft failure in RTR.

Oxidative stress is important in the pathophysiology of chronic transplant dys-function (6–9). Chronic transplant dysfunction is the leading cause of a slow, but progres-sive decline in renal allograft function (3). It is thought that chronic transplant dysfunc-tion is usually not the result of immunological rejection, but of a constant ‘response to injury’ due to a combination of several immunological and non-immunological insults (3,17). In the first year after transplantation, transplant dysfunction is predominantly caused by immunological factors (14), whereas after one year chronic transplant dys-function is mainly attributed to non-immunological risk factors, with many aspects of the pathophysiology shared with diabetic nephropathy (14). Recent studies suggest a protec-tive effect of relatively high levels of bilirubin against risks imposed by such non-immu-nological risk factors. One is a study in which bilirubin was inversely associated with dys-lipidemia in patients with metabolic syndrome (18). The other study indicated that biliru-bin may protect against progression of diabetic nephropathy (19), showing that bilirubin significantly inhibited reactive oxygen species production in human renal mesangial cells (19). In line, diabetic hyperbilirubinemic rats did not develop renal mesangial expansion, a typical feature in diabetic nephropathy. Furthermore, Adin et al. showed that exogenous bilirubin protected against kidney ischemia reperfusion injury (13).

Accumulating evidence has shown that higher bilirubin levels are associated with low prevalence of death and disease as high serum bilirubin levels were suggested to be protective against coronary artery disease (20), cancer mortality (21) and colorectal can-cer (22). But to our knowledge we are the first to report that relatively high circulating levels of bilirubin are associated with low risk of late graft failure in RTR. In line with this association, we found that the decline in creatinine clearance was least pronounced in RTR with highest bilirubin concentrations.

Accumulating evidence implicates bilirubin as a powerful anti-oxidant (10,23,24). Bilirubin has been shown to protect against kidney ischemia/reperfusion injury in rats (13). It has also been shown that bilirubin might modulate immunological factors (23,25). High concentrations of biliverdin, a precursor of bilirubin, downregulate sev-eral inflammatory pathways, hereby inducing tolerance to cardiac allografts in mice (25). Protective effects in organ transplantation may include anti-inflammatory, anti-apoptot-ic and anti-proliferative properties (23). Thus, it may not only be non-immunological anti-oxidative effects of bilirubin that explain our findings, but also effects that involve modulation of the immune system, with less pronounced chronic rejection in subjects with higher circulating concentrations of bilirubin.

54

Cha

pter

3

This study has some limitations. Despite the fact that bilirubin has been impli-cated as protective against ischemia-reperfusion injury (13), we found no association of circulating bilirubin concentrations with occurrence of delayed graft function. It should, however, be realized that our study was designed to prospectively investigate the hy-pothesis that circulating endogenous bilirubin concentrations are associated with late graft failure and that assessment of bilirubin for our study was performed 1 year or more after transplantation. A prospective study of a potential association with occurrence of delayed graft function would have required assessment of pre- or peri-operative concen-trations of bilirubin. We found that bilirubin concentrations above what is considered normal for the general population are quite common late after transplantation. This may indicate preferential occurrence of delayed graft failure and early graft failure in RTR with constituently low bilirubin concentrations. In future studies it would therefore be interesting to investigate the potential association of bilirubin concentrations assessed in the pre- or peri-operative phase with occurrence of delayed graft function and early graft failure. We did not record data on extended criteria donors, but such donor data, known to influence occurrence of delayed graft function of recipient characteristics would be important to be recorded to allow for adjustment as potential confounders in multivariate analyses in such studies. Other limitations are that this study was set up as a single center study, so differences with other centers have not been assessed. Also, RTR patients were included at different time points after transplantation (2001–2003), which might have caused healthy survivor bias. Another limitation is that we have no repeated measurements of bilirubin concentrations. Most epidemiological studies use a single baseline measurement to predict outcomes, which adversely affects predictive properties of variables associated with outcomes. If intra-individual variability of predic-tive parameters is taken into account, this results in much stronger relations with out-comes (26,27). A particular strength of this study is that there was no loss to follow-up.

In conclusion, our study suggests a protective effect of endogenous bilirubin against development of graft failure in RTR, hereby implying its potential role as an anti-oxidant. If our findings are confirmed by other studies, and a mechanistic role sup-ported, intervention with exogenous bilirubin may be of interest for long-term preserva-tion of renal function in RTR. Given the results of our study, and particularly because we found relatively high bilirubin concentrations in RTR with preserved graft function until relatively late after transplantation, it would also be of interest to address in future studies whether pre- or peri-operative bilirubin concentrations or bilirubin concentra-tions shortly after transplantation would be associated with delayed graft function or early graft loss.


AcknowledgementsThis work was supported by the Dutch Kidney Foundation, Nierstichting Nederland (C00.1877).

References 1. Merville P. Combating chronic renal allograft dysfunction : optimal immunosuppressive regi-

mens. Drugs 2005;65(5):615–631. 2. Meier-Kriesche HU, Schold JD, Kaplan B. Long-term renal allograft survival: have we made

significant progress or is it time to rethink our analytic and therapeutic strategies? Am J Trans-plant 2004 Aug;4(8):1289–1295.

3. Paul LC. Chronic allograft nephropathy: An update. Kidney Int 1999 Sep;56(3):783–793. 4. Meier-Kriesche HU, Schold JD, Srinivas TR, Kaplan B. Lack of improvement in renal allograft

survival despite a marked decrease in acute rejection rates over the most recent era. Am J Transplant 2004 Mar;4(3):378–383.

5. Sayegh MH, Carpenter CB. Transplantation 50 years later--progress, challenges, and promises. N Engl J Med 2004 Dec 23;351(26):2761–2766.

6. Kim J, Seok YM, Jung KJ, Park KM. Reactive oxygen species/oxidative stress contributes to progression of kidney fibrosis following transient ischemic injury in mice. Am J Physiol Renal Physiol 2009 Aug;297(2):F461–70.

7. Djamali A, Sadowski EA, Muehrer RJ, Reese S, Smavatkul C, Vidyasagar A, et al. BOLD-MRI assessment of intrarenal oxygenation and oxidative stress in patients with chronic kidney al-lograft dysfunction. Am J Physiol Renal Physiol 2007 Feb;292(2):F513–22.

8. Raj DS, Lim G, Levi M, Qualls C, Jain SK. Advanced glycation end products and oxidative stress are increased in chronic allograft nephropathy. Am J Kidney Dis 2004 Jan;43(1):154–160.

9. Albrecht EW, Stegeman CA, Tiebosch AT, Tegzess AM, van Goor H. Expression of inducible and endothelial nitric oxide synthases, formation of peroxynitrite and reactive oxygen species in human chronic renal transplant failure. Am J Transplant 2002 May;2(5):448–453.


11. Duann P, Lianos EA. GEC-targeted HO-1 expression reduces proteinuria in glomerular im-mune injury. Am J Physiol Renal Physiol 2009 Sep;297(3):F629–38.

12. Baranano DE, Rao M, Ferris CD, Snyder SH. Biliverdin reductase: a major physiologic cytopro-tectant. Proc Natl Acad Sci U S A 2002 Dec 10;99(25):16093–16098.

13. Adin CA, Croker BP, Agarwal A. Protective effects of exogenous bilirubin on ischemia-reperfusion injury in the isolated, perfused rat kidney. Am J Physiol Renal Physiol 2005 Apr;288(4):F778–84.

56

Cha

pter

3

14. de Vries AP, Bakker SJ, van Son WJ, van der Heide JJ, Ploeg RJ, The HT, et al. Metabolic syn-drome is associated with impaired long-term renal allograft function; not all component crite-ria contribute equally. Am J Transplant 2004 Oct;4(10):1675–1683.

15. van Ree RM, de Vries AP, Oterdoom LH, Seelen MA, Gansevoort RT, Schouten JP, et al. Plasma procalcitonin is an independent predictor of graft failure late after renal transplantation. Trans-plantation 2009 Jul 27;88(2):279–287.

16. van Ree RM, Oterdoom LH, de Vries AP, Gansevoort RT, van der Heide JJ, van Son WJ, et al. Elevated levels of C-reactive protein independently predict accelerated deterioration of graft function in renal transplant recipients. Nephrol Dial Transplant 2007 Jan;22(1):246–253.

17. Massy ZA, Guijarro C, Wiederkehr MR, Ma JZ, Kasiske BL. Chronic renal allograft rejection: immunologic and nonimmunologic risk factors. Kidney Int 1996 Feb;49(2):518–524.

18. Giral P, Ratziu V, Couvert P, Carrie A, Kontush A, Girerd X, et al. Plasma bilirubin and gamma-glutamyltransferase activity are inversely related in dyslipidemic patients with metabolic syn-drome: relevance to oxidative stress. Atherosclerosis 2010 Jun;210(2):607–613.


20. Schwertner HA, Jackson WG, Tolan G. Association of low serum concentration of bilirubin with increased risk of coronary artery disease. Clin Chem 1994 Jan;40(1):18–23.

21. Temme EH, Zhang J, Schouten EG, Kesteloot H. Serum bilirubin and 10-year mortality risk in a Belgian population. Cancer Causes Control 2001 Dec;12(10):887–894.

22. Zucker SD, Horn PS, Sherman KE. Serum bilirubin levels in the U.S. population: gender effect and inverse correlation with colorectal cancer. Hepatology 2004 Oct;40(4):827–835.

23. Ollinger R, Wang H, Yamashita K, Wegiel B, Thomas M, Margreiter R, et al. Therapeu-tic applications of bilirubin and biliverdin in transplantation. Antioxid Redox Signal 2007 Dec;9(12):2175–2185.

24. Ryter SW, Tyrrell RM. The heme synthesis and degradation pathways: role in oxidant sensitiv-ity. Heme oxygenase has both pro- and antioxidant properties. Free Radic Biol Med 2000 Jan 15;28(2):289–309.

25. Yamashita K, McDaid J, Ollinger R, Tsui TY, Berberat PO, Usheva A, et al. Biliverdin, a nat-ural product of heme catabolism, induces tolerance to cardiac allografts. FASEB J 2004 Apr;18(6):765–767.

26. Danesh J, Wheeler JG, Hirschfield GM, Eda S, Eiriksdottir G, Rumley A, et al. C-reactive pro-tein and other circulating markers of inflammation in the prediction of coronary heart disease. N Engl J Med 2004 Apr 1;350(14):1387–1397.

27. Koenig W, Sund M, Frohlich M, Lowel H, Hutchinson WL, Pepys MB. Refinement of the as-sociation of serum C-reactive protein concentration and coronary heart disease risk by correc-tion for within-subject variation over time: the MONICA Augsburg studies, 1984 and 1987. Am J Epidemiol 2003 Aug 15;158(4):357–364.

4

Diet, lifestyle,

and total bilirubin:

the Zutphen elderly study4Petronella E. Deetman1*, Ineke J. Riphagen1,2*, Gerjan Navis1, Stephan J.L. Bakker 1,2, Daan Kromhout3

1 Department of Medicine, Division of Nephrology, University Medical Center Groningen and University of Groningen, the Netherlands;

2 Top Institute Food & Nutrition, Wageningen, the Netherlands;

3 Division of Human Nutrition, Wageningen University, the Netherlands

*Both authors contributed equally

Submitted

58

Cha

pter

4

AbstractBackground. Levels of serum bilirubin, a potent endogenous antioxidant, were found to be lower in chronic disorders that are associated with increased levels of oxidative stress including diabetes and cardiovascular disease. Levels of oxidative stress are also influenced by diet and lifestyle. Little is known about the effects of diet and lifestyle on bilirubin concentrations. Therefore, our aim was to investigate the associations of the Mediterranean diet, individual food groups, and lifestyle factors such as smoking with bilirubin concentrations in elderly men without major chronic diseases. Methods. A cross-sectional study was performed in elderly men participating in the Zutphen elderly study. Habitual alcohol and dietary consumption were assessed. Dietary patterns were studied with the Mediterranean Diet Score (MDS). Linear regression analyses were performed to determine associations of the MDS, components of the MDS, tea, wine consumption, and smoking status with bilirubin. Results. In total, 509 male subjects were included. Median age was 71 (IQR 67–75) years. Mean bilirubin level was 7.7 ± 3.5 µmol/L. Both adherence to the MDS with more than 4 points and wine consumption were significantly associated with higher serum bili-rubin concentrations (unstandardized B = 0.84, P = 0.009 and unstandardized B = 0.91, P = 0.013 respectively). Although not statistically significant, we found that tea consump-tion was positively associated and that smoking status was inversely associated with bili-rubin concentrations in univariable analyses. Conclusions. These findings suggest that a healthy dietary pattern like the Mediterra-nean diet and consumption of wine may increase bilirubin levels which could have ben-eficial effects on oxidative stress and oxidative stress-related diseases.

59Diet, lifestyle, and total bilirubin: the Zutphen elderly study

IntroductionFor decades, bilirubin was believed to be a potentially toxic waste product of heme ca-tabolism (1,2). During the last decades, however, it has been demonstrated that bilirubin is a potent endogenous cytoprotectant and antioxidant (2–4). Furthermore, decreased bilirubin concentrations might reflect increased levels of oxidative stress, which results in increased utilization of the endogenous antioxidant bilirubin (5,6). Moreover, it has been consistently shown that bilirubin concentrations were lower in patients with chron-ic disorders that are also associated with increased levels of oxidative stress including chronic kidney disease, diabetes, and cardiovascular disease (6).

Besides chronic diseases, lifestyle factors are also known to affect levels of oxida-tive stress. A major contributor to increased oxidative stress is smoking (7). Accordingly, bilirubin concentrations were found to be significantly lower in individuals who smoked, possibly as a consequence of increased utilization of bilirubin (5). In addition, diet was also found to be associated with oxidative stress. Adherence to the Mediterranean diet was found to be significantly associated with decreased markers of oxidative stress (8–10).

However, little is known about the effects of diet and lifestyle on bilirubin con-centrations. Therefore, our aim was to investigate the associations of diet patterns, indi-vidual foods, and lifestyle factors such as smoking and wine consumption with bilirubin concentrations in elderly men without major chronic diseases.


Study population

The population-based Zutphen Elderly Study was initiated in 1985 to collect data on diet and risk factors of cardiovascular disease in elderly men, living in a middle-sized city in the eastern part of The Netherlands. Details of the study were described previ-ously (11). In brief, of the 1,266 men who were invited, a total of 939 (74%) men, aged 64–85 years (response rate 74%), participated in the study. Subjects with missing data on the baseline measurement of bilirubin were excluded from the analyses (n = 128, 14%). Because coronary heart disease, cardiovascular disease, and cancer increase the risk of death and may induce changes in dietary habits, 261 subjects with a history of myocardial infarction, stroke, diabetes or cancer and with complete data on dietary intake at baseline were excluded at baseline. Furthermore, because underweight can be considered as a hallmark of underlying chronic disease, 9 subjects with underweight, defined as a BMI below 18.5 (12), (n = 8) or missing data on body mass index (BMI, n = 1) were excluded. Of the remaining 541 subjects, 32 were excluded because of miss-ing data on food consumption. Finally, 509 subjects were eligible for analyses. The

60

Cha

pter

4

study was approved by the Medical Ethics Committee of the University of Leiden, The Netherlands and adhered to the Declaration of Helsinki. Written informed consent was obtained from all participants.

Definitions and measurements

In 1985, information on patient characteristics was collected including height, weight, systolic and diastolic blood pressure, information on occupation, physical activity, use of medication, smoking status, history of myocardial infarction, stroke, diabetes mel-litus, and cancer. BMI was calculated as weight (kg) divided by height (m) squared. Systolic and diastolic blood pressure were measured twice by using a random-zero sphygmomanometer while participants were in the supine position. The mean values of the 2 blood pressure measurements are presented. Information on occupation was collected by a self-administered questionnaire. Two occupational groups were defined: professionals and men with administrative jobs (high socioeconomic status [SES]); small business owners and manual workers (low SES). Physical activity was assessed with a questionnaire and summed into the total physical activity in minutes per week. Medication use was reported to the examining physician by the participant and coded to the ATC system. Information on smoking habits was collected using a standard-ized questionnaire (13). For the current study, men were divided into current smoker or non-smoker. The estimated GFR (eGFR) was calculated using the Chronic Kidney Disease Epidemiology Collaboration equation (14).

Baseline serum total bilirubin, glucose, creatinine, aspartate aminotransferase (AST), alanine aminotransferase (ALT), and alkaline phosphatase (AP) were measured on a SMAC (Technicon, Tarrytown, NY) in the Central Clinical Chemical Labora-tory of the Academic Hospital in Leiden, The Netherlands. Non-fasting serum total cholesterol and high density lipoprotein (HDL) were determined enzymatically with the CHOD-PAP mono-testkit (Boehringer, Mannheim, Germany) in the Standardized Lipid Laboratory of Wageningen University, The Netherlands.

Assessment of food and alcohol consumption

The habitual alcohol and dietary consumption were assessed by dieticians, using the cross-check dietary history method (15), adapted to the Dutch situation. This meth-od provides information about the habitual food and alcohol consumption, based on consumption in the month preceding the interview. The interview consisted of three


steps. First, information about the usual food consumption pattern during weekdays or weeks was collected. This was checked by collecting data on the average consump-tion of foods during a day or a week. The second check consisted on the quantity of foods bought for the whole family during a week. Consumed foods were encoded according to the Uniform Food Encoding System (16) and foods were categorized into 22 food groups.

Dietary patterns were studied with the modified Mediterranean Diet Score (MDS). The MDS was calculated according to Trichoupoulou et al. (17). First, median dietary intake was calculated of the following components of the MDS: ratio of mono-unsaturated to saturated fat; legumes, nuts, and seeds; grains; fruit; vegetables; fish; meat; dairy products. For the components ratio of mono-unsaturated to saturated fat (M/S); legumes, nuts, and seeds; grains; fruit; vegetables; and fish, a 1 was assigned to subjects whose consumption was at least as high as the median. The remaining subjects were assigned a 0. For the components meat and dairy products, a 0 was assigned to subjects whose consumption was at least as high as the median. A 1 was assigned to the remaining subjects. This was done because the consumption of meat and dairy prod-ucts was low in the traditional Mediterranean Diet (17). Maximal adherence to the MDS renders a score of 8 points. Low adherence to the MDS was defined as 4 points or less and high adherence as more than 4 points.

Statistical analysis

Baseline characteristics of the study population were calculated according to tertiles of bilirubin. Normally distributed variables are presented as mean ± standard deviation (SD), skewed distributed variables are presented as median (interquartile range [IQR]), and categorical variables are given as number (percentage). Differences in baseline characteristics across tertiles of bilirubin were tested using the ANOVA test for normal distributed data, the Kruskal-Wallis test for nonnormal distributed data, and the Chi-square test for categorical variables. Differences in bilirubin between men with a high or a low MDS and components of the MDS were tested with a t-test. A t-test was also applied to determine differences in bilirubin for the 2 levels of groups of diet and life-style components. Correlations between the MDS, smoking, consumption of wine and tea were investigated with Fisher’s exact test. Finally, to assess the independent associa-tions of bilirubin and its correlates, multivariable linear regression analysis was used and stepwise backward elimination was applied. Statistical analyses were performed using SPSS (version 22.0, SPSS Inc. Chicago, IL, USA). A two-sided P‑value < 0.05 was considered statistically significant.

62

Cha

pter

4

ResultsA total of 509 male subjects were included in the present study. Median age was 71 (IQR 67–75) years. Mean total bilirubin level was 7.7 ± 3.5 µmol/L. Baseline characteristics are shown in Table 1, for the whole population and according to tertiles of bilirubin. HDL cholesterol and use of antihypertensive medication, diuretics in particular, and the enzymes ALT and AP were associated with bilirubin concentrations. Age, BMI, physical activity, total energy intake, blood pressure, glucose, total cholesterol, use of lipid lower-ing medication, indicators of renal function, and the enzyme AST were not related to bilirubin.

In the whole study population, the median M/S ratio was 0.6 (0.5–0.7), the me-dian consumption of legumes, nuts, and seeds was 8 (0–24) g/day, grains and bread 151 (116–196) g/day, meat 108 (85–135) g/day, and dairy products 346 (200–492) g/day. Furthermore, 193 (38%) men adhered to the MDS with more than 4 components. The consumption of components of the MDS, using the medians as cutoffs, and associations of the components of the MDS with total bilirubin levels are shown in Table 2. The eight individual components of the MDS were not significantly related to serum bilirubin lev-els. In elderly men who adhered to the MDS with more than 4 points, however, bilirubin concentrations were significantly higher (8.2 ± 3.6 versus 7.4 ± 3.4 µmol/L). These results remained unchanged after adjustment for age.

Table 1. Baseline participant characteristics of the study population and according to tertiles of bilirubin concentrations.

No. ofsubjects

Study Population

Tertiles of bilirubin

I II III P‑value

N 509 134 202 173

Bilirubin (µmol/L) 509 7.7 ± 3.5 4.1 ± 1.0 6.9 ± 0.8 11.5 ± 3.1

Age (years) 509 71 (67–75) 71 (67–75) 71 (67–75) 71 (68–75) 0.58

SES (n low, %) 492 227 (45) 67 (50) 93 (46) 67 (39) 0.088

BMI (kg/m2) 509 26 ± 3 26 ± 3 26 ± 3 26 ± 3 0.88

Physical Activity 495

Not active (n, %) 235 (48) 62 (48) 99 (51) 74 (44) 0.71

≤150 METS active (n, %) 124 (25) 31 (24) 45 (23) 48 (28)

>150 METS active (n, %) 136 (28) 36 (28) 52 (27) 48 (28)

Total energy intake (Kcal) 509 2276 ± 507 2274 ± 539 2311 ± 508 2237 ± 480 0.38

Systolic blood pressure (mmHg) 509 150 ± 21 151 ± 21 150 ± 19 151 ± 23 0.59

Diastolic blood pressure (mmHg) 509 86 ± 11 85 ± 11 85 ± 11 87 ± 12 0.29


No. ofsubjects

Study Population

Tertiles of bilirubin

I II III P‑value

Use of antihypertensives (n, %) 509 84 (17) 23 (17) 24 (12) 37 (21) 0.046

Beta Blockade (n, %) 45 (9) 12 (9) 14 (7) 19 (11) 0.39

Diuretics (n, %) 50 (10) 16 (12) 10 (5) 24 (14) 0.010

Glucose (mmol/L) 505 6.2 ± 1.4 6.3 ± 1.4 6.2 ± 1.3 6.3 ± 1.5 0.71

Total cholesterol (mmol/L) 509 6.1 (5.4–6.7) 6.2 ± 1.0 6.0 ± 1.0 6.0 ± 1.0 0.39

HDL cholesterol (mmol/L) 509 1.14 ± 0.27 1.07 ± 0.27 1.13 ± 0.27 1.19 ± 0.27 <0.001

Lipid lowering drugs (n, %) 509 2 (0.4) 0 (0) 1 (0.5) 1 (0.6) 0.69

Serum creatinine (µmol/L) 492 104 ± 22 106 ± 23 102 ± 25 103 ± 15 0.30

eGFR (ml/min/1.73m2) 492 65 ± 13 63 ± 14 66 ± 13 64 ± 11 0.18

Liver function

AST/SGOT (U/L) 492 8 (7–10) 8 (7–10) 8 (7–9) 8 (7–10) 0.17

ALT/SGPT (U/L) 492 6 (5–10) 8 (6–12) 6 (5–9) 6 (4–9) <0.001

AP (U/L) 402 36 (30–44) 38 (31–46) 38 (31–44) 34 (29–40) 0.002

SES, Socioeconomic status; BMI, body mass index; HDL, high density lipoprotein; eGFR, estimated glomerular filtra‑

tion rate; AST, aspartate aminotransferase; ALT alanine aminotransferase; AP, alkaline phosphatase.

Furthermore, the associations of different lifestyle components with total biliru-bin levels are shown in Table 3. Non-smokers and subjects who consumed > 2 cups of tea per day had higher concentrations of bilirubin, albeit these associations were not statisti-cally significant. Twenty-four percent of the elderly men used wine of which 67% was red wine. Subjects who consumed wine had higher concentrations of bilirubin compared to subjects that did not. Consumption of beer and liquor were not related to bilirubin levels. Results were essentially unchanged after the adjusting these associations for age. In multivariable linear regression analysis that included age, wine, beer, and liquor as determinants of bilirubin, the association of wine with bilirubin levels was confirmed, while beer and liquor were not associated with bilirubin.

Furthermore, we investigated whether the MDS, smoking, consumption of wine and tea were correlated. Wine consumption was positively associated with the MDS (χ2 = 15.12, P < 0.001), and with tea consumption (χ2 = 6.64, P = 0.010). Smoking sta-tus was inversely associated with wine consumption (χ2 = 9.7, P = 0.002) and tea con-sumption (χ2 = 17.02, P < 0.001). The MDS was neither associated with smoking status (χ2 = 0.37, P = 0.56), nor with tea consumption (χ2 = 1.76, P = 0.19).


64

Cha

pter

4

Table 2. Associations of components of the Mediterranean Diet Score (MDS) with total bilirubin.

No. ofsubjects

Median consumption (g/day)

Total Bilirubin

Bilirubin (µmol/L) P‑value

M/S Fat ratio

< median 258 0.54 (0.49–0.58) 7.7 ± 3.4 0.84

≥ median 251 0.71 (0.67–0.81) 7.7 ± 3.6

Legumes, Nuts, seeds

< median 244 0 (0–3) 7.7 ± 3.6 0.86

≥ median 265 23 (14–33) 7.7 ± 3.5

Grains, Bread

< median 247 115 (91–132) 7.7 ± 3.1 0.74

≥ median 262 194 (166–233) 7.8 ± 3.9

Fruit

< median 268 111 (49–136) 7.5 ± 3.5 0.11

≥ median 241 269 (225–341) 8.0 ± 3.6

Vegetables

< median 249 130 (107–147) 7.6 ± 3.5 0.38

≥ median 260 211 (183–251) 7.8 ± 3.5

Meat

≥ median 258 134 (119–158) 7.5 ± 3.3 0.11

< median 251 85 (71–98) 8.0 ± 3.7

Dairy products

≥ median 254 492 (413–644) 7.6 ± 3.5 0.36

< median 255 201 (119–262) 7.9 ± 3.5

Fish

< median 243 0 (0–6) 7.5 ± 3.8 0.18

≥ median 266 26 (17–38) 7.9 ± 3.2

MDS

≤4, low 316 7.4 ± 3.4 0.013

>4, high 193 8.2 ± 3.6

M/S fat ratio, monounsaturated/saturated fat ratio; MDS, Mediterranean diet score.


Table 3. Associations of lifestyle components with total bilirubin.

Total Bilirubin

No. of subjects Mean Bilirubin P‑value

Tea

≤ 2 cups 198 7.4 ± 3.1 0.082

>2 cups 311 7.9 ± 3.7

Wine

no 388 7.5 ± 3.5 0.012

yes 121 8.4 ± 3.7

Beer

no 371 7.7 ± 3.6 0.67

yes 138 7.8 ± 3.4

Liquor

no 216 7.6 ± 3.7 0.52

yes 293 7.8 ± 3.4

Smoking statusa

Non‑smoker 349 7.9 ± 3.7 0.079

Current smoker 159 7.3 ± 3.1

amissing data in one subject

Table 4. Multivariable associations of total bilirubin with its correlates.

Total Bilirubin

Model 1 Model 2 Model 3

Beta SE P‑value Beta SE P‑value Beta SE P‑value

MDS 0.84 0.32 0.009 0.70 0.32 0.031 0.68 0.32 0.036

Smoking ‑0.59 0.34 0.079 ‑0.41 0.34 0.23

Wine 0.91 0.36 0.013 0.66 0.37 0.079 0.79 0.37 0.033

Tea 0.58 0.32 0.072 0.40 0.32 0.22

Unstandardized beta’s are shown

Model 1: Crude analysis

Model 2: multivariable model: bilirubin = MDS + smoking + wine + tea

Model 3: multivariable model after stepwise backward elimination: bilirubin = MDS + wine

MDS, Mediterranean Diet Score; SE, standard error

66

Cha

pter

4

In univariable linear regression analyses (Table 4, model 1), the MDS and wine consumption were positively associated with bilirubin levels. Smoking status and tea consumption were not significantly associated with bilirubin. In a multivariable model with MDS, smoking status, wine, and tea consumption (Table 4, model 2), the MDS was significantly associated with bilirubin. After removing smoking status and tea consump-tion from the model, the MDS and wine consumption were significantly associated with bilirubin levels (Table 4, model 3). The men who adhered to the MDS and consumed wine had a 1.5 µmol/L (20%) higher bilirubin level compared to the men who did not adhere to the MDS and did not consume wine.

DiscussionIn this study we demonstrated that, even though the individual components of the Medi-terranean diet were not significantly associated with serum bilirubin concentrations, over-all adherence to the Mediterranean diet and wine consumption were significantly associ-ated with higher serum bilirubin concentrations. Although not statistically significant, we found that smoking status was inversely associated with serum bilirubin and that tea con-sumption was positively associated with bilirubin concentrations in univariable analyses.

The traditional Mediterranean diet incorporates high consumption of fruits, veg-etables, legumes, nuts, cereals, fish, unsaturated oils, and low intake of dairy products and red meat (18). Trichopoulou et al. (17) combined the consumption of these food groups into what is known as the MDS. In the present study, significant associations of the individual components with bilirubin were not found, whereas high adherence to the MDS was significantly associated with higher bilirubin levels. Similar positive associations with health have been described in previous studies (17,19). Apparently, small effects of the individual food components may enlarge when the individual food components are combined into a nutritionally adequate dietary pattern score like the MDS (17).

A number of studies have investigated the effects of the Mediterranean diet on oxidative stress (8–10). Dai et al. (8) studied 138 twin pairs and found a 7% decrease in the ratio of reduced to oxidized glutathione, a plasma marker of oxidative stress, per unit increase in the MDS. In addition, Urquiaga et al. (18) showed that adherence to the Mediterranean diet resulted in better antioxidant defenses and less oxidative dam-age compared to a Western or US diet. In line with these findings, we demonstrate that adherence to the Mediterranean diet is associated with higher concentrations of serum bilirubin, and with a lower level of oxidative stress.

Bilirubin is the end product of heme catabolism. Heme oxygenase-1 (HO-1) splits heme into carbon monoxide, ferrous iron, and biliverdin, which is subsequently reduced to bilirubin by biliverdin reductase (1,20). Since HO-1 is the enzyme that catalyzes the rate-limiting step in heme degradation (21,22), induction of HO-1 could result in higher


levels of end products of heme degradation, including bilirubin. HO-1 is a highly induc-ible enzyme, influenced by bioactive compounds such as resveratrol, which is commonly found in grapes and red wine (23). In the current study, we found that consumption of wine, which prevailingly consisted of red wine (i.e. 67%), was indeed independently as-sociated with higher bilirubin levels.

Several studies have demonstrated that smoking, a major contributor to oxida-tive stress (7), is inversely associated with HO-1 activity and bilirubin concentrations. In vitro, over-expression of HO-1 inhibited cigarette smoke-induced cell death (24). Furthermore, mice exposed to cigarette smoke showed elevated expression of HO-1 (24), which results in higher concentrations of bilirubin. Schwertner et al. have shown in humans that serum bilirubin concentrations were inversely related to cigarette smoking (5). In the present study, we found that current smokers had lower bilirubin levels com-pared to non-smokers, albeit this was not statistically significant. Given the correlations between smoking status, wine consumption, tea consumption and the MDS, it is pos-sible that the association of smoking with bilirubin was confounded by these variables.

This study has several limitations. First, the study population was relatively small and did not include women or middle-aged participants. It remains to be established whether the Mediterranean diet and wine consumption are also associated with biliru-bin in these groups. Second, given the observational nature of the current study, cause-effect relationships cannot be ascertained. Finally, because only total bilirubin was mea-sured we were not able to differentiate between direct and indirect bilirubin. A specific strength of this study was the collection of detailed information on the consumption of different food items and alcoholic beverages. In addition, the detailed information about smoking status and different types of alcohol consumption made it possible to study the independent associations of the MDS and wine consumption with bilirubin.

In conclusion, a healthy dietary pattern like the Mediterranean diet and con-sumption of wine may increase bilirubin levels. High bilirubin levels may have beneficial effects on oxidative stress and subsequent oxidative stress-related diseases. When our findings are confirmed by other studies, the MDS and wine consumption may enhance bilirubin concentrations to protect against oxidative stress.

Acknowledgements The Zutphen Elderly Study was supported by the Netherlands Prevention Foundation. The contribution of Daan Kromhout was supported by the Royal Netherlands Academy of Arts and Sciences.

68

Cha

pter

4

References 1. Vitek L. The role of bilirubin in diabetes, metabolic syndrome, and cardiovascular diseases.

Front Pharmacol 2012;3:55. 2. Morita T. Heme oxygenase and atherosclerosis. Arterioscler Thromb Vasc Biol 2005

Sep;25(9):1786–1795. 3. Abraham NG, Kappas A. Heme oxygenase and the cardiovascular-renal system. Free Radic

Biol Med 2005 Jul 1;39(1):1–25. 4. Vitek L, Schwertner HA. The heme catabolic pathway and its protective effects on oxidative

stress-mediated diseases. Adv Clin Chem 2007;43:1–57. 5. Schwertner HA. Association of smoking and low serum bilirubin antioxidant concentrations.

Atherosclerosis 1998 Feb;136(2):383–387. 6. Schwertner HA, Jackson WG, Tolan G. Association of low serum concentration of bilirubin

with increased risk of coronary artery disease. Clin Chem 1994 Jan;40(1):18–23. 7. Pignatelli B, Li CQ, Boffetta P, Chen Q, Ahrens W, Nyberg F, et al. Nitrated and oxidized plasma

proteins in smokers and lung cancer patients. Cancer Res 2001 Jan 15;61(2):778–784. 8. Dai J, Jones DP, Goldberg J, Ziegler TR, Bostick RM, Wilson PW, et al. Association between adher-

ence to the Mediterranean diet and oxidative stress. Am J Clin Nutr 2008 Nov;88(5):1364–1370. 9. Fito M, Guxens M, Corella D, Saez G, Estruch R, de la Torre R, et al. Effect of a traditional

Mediterranean diet on lipoprotein oxidation: a randomized controlled trial. Arch Intern Med 2007 Jun 11;167(11):1195–1203.

10. Yubero-Serrano EM, Garcia-Rios A, Delgado-Lista J, Delgado-Casado N, Perez-Martinez P, Rodriguez-Cantalejo F, et al. Postprandial effects of the Mediterranean diet on oxidant and antioxidant status in elderly men and women. J Am Geriatr Soc 2011 May;59(5):938–940.

11. Hertog MG, Feskens EJ, Hollman PC, Katan MB, Kromhout D. Dietary antioxidant fla-vonoids and risk of coronary heart disease: the Zutphen Elderly Study. Lancet 1993 Oct 23;342(8878):1007–1011.

12. World Health Organization. Global database on Body Mass Index. Available at: http://apps.who.int/bmi/index.jsp?introPage = intro.html.

13. Streppel MT, Boshuizen HC, Ocke MC, Kok FJ, Kromhout D. Mortality and life expectancy in relation to long-term cigarette, cigar and pipe smoking: the Zutphen Study. Tob Control 2007 Apr;16(2):107–113.

14. Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AF,3rd, Feldman HI, et al. A new equa-tion to estimate glomerular filtration rate. Ann Intern Med 2009 May 5;150(9):604–612.

15. Bloemberg BP, Kromhout D, Obermann-De Boer GL, Van Kampen-Donker M. The repro-ducibility of dietary intake data assessed with the cross-check dietary history method. Am J Epidemiol 1989 Nov;130(5):1047–1056.

16. Kommissie UCV. UCV Tabel: uitgebreide voedingsmiddelentabel 1985 (UVC Table: extended food composition table 1985). The Hague: Voorlichtingsbureau voor de Voeding (Nutrition Education Bureau), 1985 (in Dutch). 1985.


17. Trichopoulou A, Costacou T, Bamia C, Trichopoulos D. Adherence to a Mediterranean diet and survival in a Greek population. N Engl J Med 2003 Jun 26;348(26):2599–2608.

18. Urquiaga I, Strobel P, Perez D, Martinez C, Cuevas A, Castillo O, et al. Mediterranean diet and red wine protect against oxidative damage in young volunteers. Atherosclerosis 2010 Aug;211(2):694–699.

19. Knoops KT, de Groot LC, Kromhout D, Perrin AE, Moreiras-Varela O, Menotti A, et al. Medi-terranean diet, lifestyle factors, and 10-year mortality in elderly European men and women: the HALE project. JAMA 2004 Sep 22;292(12):1433–1439.


21. Riphagen IJ, Deetman PE, Bakker SJ, Navis G, Cooper ME, Lewis JB, et al. Bilirubin and Pro-gression of Nephropathy in Type 2 Diabetes: A Post-Hoc Analysis of RENAAL with Indepen-dent Replication in IDNT. Diabetes 2014 Mar 27.

22. Kirkby KA, Adin CA. Products of heme oxygenase and their potential therapeutic applications. Am J Physiol Renal Physiol 2006 Mar;290(3):F563–71.

23. Barbagallo I, Galvano F, Frigiola A, Cappello F, Riccioni G, Murabito P, et al. Potential thera-peutic effects of natural heme oxygenase-1 inducers in cardiovascular diseases. Antioxid Redox Signal 2013 Feb 10;18(5):507–521.

24. Slebos DJ, Ryter SW, van der Toorn M, Liu F, Guo F, Baty CJ, et al. Mitochondrial localization and function of heme oxygenase-1 in cigarette smoke-induced cell death. Am J Respir Cell Mol Biol 2007 Apr;36(4):409–417.

5

Urinary urea excretion

and long‑term outcome

after renal transplantation5Petronella E. Deetman1, M. Yusof Said1, Daan Kromhout2, Robin P.F. Dullaart1, Jenny E. Kootstra-Ros3, Jan-Stephan F. Sanders1, Marc A.J. Seelen1, Reinold O.B. Gans1, Gerjan Navis1, Michel M. Joosten1,4, and Stephan J.L. Bakker1,4.

1 Department of Medicine, Division of Nephrology, University of Groningen, University Medical Center Groningen, Groningen, the Netherlands;

2 Division of Human Nutrition, Wageningen University, the Netherlands;

3 Department of Laboratory Medicine, University Medical Center Groningen, the Netherlands;

4 Top Institute Food & Nutrition, Wageningen, the Netherlands

Transplantation 2014;12:(epub ahead of print)

72

Cha

pter

5

AbstractBackground. Little is known about optimal protein intake after transplantation. The aim of this study was to prospectively investigate associations of urinary urea excretion, a marker for protein intake, with graft failure and mortality in renal transplant recipients (RTR) and potential effect-modification by body mass index (BMI) and estimated glo-merular filtration rate (eGFR). Methods. Urinary urea excretion was measured in repeated 24-hr urine collections be-tween 6 and 18 months after transplantation. Results. In total, 940 RTR were included. During 4.4 (2.3–7.8) years of follow-up for graft failure and 4.8 (2.5–8.3) years for all-cause mortality, 78 RTR developed graft failure and 158 RTR died. Urinary urea excretion was not associated with graft failure in the overall population, but was inversely associated with graft failure in RTR with a BMI less than 25 kg/m2 (Hazard Ratio [HR], 0.64 [0.28–1.50] and 0.27 [0.09–0.83] for the second and third tertiles, respectively; P < 0.001), and in RTR with an eGFR of 45 mL per min per 1.73 m2 or higher (HR, 0.34 [0.15–0.79]; P = 0.015 and HR, 0.31 [0.11–0.86]; P = 0.025 for the second and third tertiles, respectively), both independent of potential confounders. Compared to the first tertile, RTR in the second and third tertiles of urinary urea excre-tion were at a lower risk of all-cause mortality (HR, 0.47 [0.32–0.69]; P < 0.001 and HR, 0.42 [0.26–0.68]; P < 0.001, respectively), independent of potential confounders. Body mass index and eGFR did not influence this association.Conclusions. Urinary urea excretion, a marker for protein intake, was inversely related to graft failure in RTR with a BMI less than 25 kg/m2 and in RTR with an eGFR of 45 mL per min per 1.73 m2 or higher. In addition, urinary urea excretion was inversely related to mortality.

73Urinary urea excretion and outcome after renal transplantation

IntroductionOptimizing protein intake is an important component of dietary management of chron-ic kidney disease (CKD) patients. Because high protein intake is supposed to aggravate proteinuria (1–4), scientific societies consider a low protein diet cornerstone of renopro-tective treatment (3,5,6). Guidelines recommend daily allowances of 0.6 to 0.8 g protein/kg per day in patients with CKD stages 1 to 4 (3,5,6).

In patients with advanced CKD (stage 5), the opposite is true. Mild anorexia and low protein intake are often already present at CKD stage 3, but are particularly pronounced once patients enter dialysis (7–9). Hence, dialysis patients usually start with relatively low protein intake (10,11), while dialysis induces protein catabolism and losses that must be compensated. Low protein intake has been associated with lower survival rates in hemodi-alysis patients (12,13). Therefore, current guidelines recommend a minimal protein intake of 1.1 g/kg per day and preferably 1.2 to 1.3 g/kg per day in dialysis patients (6,14,15).

Although recommended intake is unambiguous for patients with CKD stage 1 to 4 and CKD stage 5 treated with or without dialysis, advisable protein intake is unknown after transplantation. Therefore, the aim of the present study was to prospectively inves-tigate whether urinary urea excretion (UUE), a marker of protein intake, is associated with graft failure and all-cause mortality in renal transplant recipients (RTR).


Study design and population

RTR were eligible for this study if they received a renal transplant for the first time be-tween January 1993 and February 2008 in the University Medical Center Groningen in the Netherlands. Details of the cohort have been published previously (16). We made use of the regular patient care program for outpatients. For each visit, RTR are requested to collect a 24-hr urine sample in which several parameters including UUE, protein and sodium are routinely assessed. Data of total cholesterol, albumin, creatinine and 24-hr urinary excretion of urea, protein, sodium, and creatinine, assessed between 6 and 18 months, were extracted from our hospital laboratory system (Supplementary Figure S1). The median of these measurements was used for analyses. Patient characteristics were retrieved from medical records. To assess change in renal function over time, the dif-ference between serum creatinine at baseline and last serum creatinine available was calculated. In case of graft failure or mortality, the last measurement before the event was used. Body surface area (BSA) was calculated as: BSA = (W0.425 × H0.725) × 0.007184

74

Cha

pter

5

(17) and body mass index (BMI) as weight (kg) by height (m) squared. Delta BMI was calculated by subtracting baseline BMI from the BMI at the time of transplantation. Delayed graft function was defined as oliguria for more than 6 days after transplantation.

Shortly after transplantation, RTR are often not yet stable, and infections and rejections play a major role. Therefore, data of the first 6 months after transplantation were discarded. Furthermore, we only included patients who survived with a function-ing graft beyond 6 months after transplantation and beyond the period of baseline data collection. Finally, patients with missing data on UUE were excluded (n = 9, 0.84%), leav-ing 940 patients eligible for analyses.

The Institutional Review Board of the University Medical Center Groningen ap-proved the study protocol, which adhered to the declaration of Helsinki. According to Dutch law, general consent for organ donation and transplantation includes consent for research projects.

Laboratory measurements

Non-fasting blood and urine samples were directly analyzed after collection. Measure-ments were performed on a Merck Mega Analyzer before March 2006 (Merck, Darm-stadt, Germany). From March 2006 onward, measurements were performed on a Roche Modular (Roche Ltd., Mannheim, Germany). In the case of significant differences be-tween results obtained by the Merck Mega and Roche Modular, results were converted according to conversion equations provided by our laboratory (Supplementary Table S1). Urea was measured with a urease glutamate dehydrogenase reaction (Intra-assay coefficient of variation Merck Mega 3.1% and Roche Modular 2.4%; Merck Mega and Roche Modular). Urinary urea excretion was based on a median (interquartile range) of 6 (5–8) collections of 24-hr urine per patient.

Clinical end-points

The primary end-point of this study was graft failure and the secondary endpoint was all-cause mortality. Graft failure was defined as return to dialysis or the need for retransplan-tation. Follow-up was recorded until October 2009. Person-time of follow-up was com-puted for each participant from 1.5 years after transplantation until the incidence of graft failure or death, or censored after 10 years of follow-up. There was no loss to follow-up.



Baseline characteristics of the study population were calculated. P‑values for a linear trend across tertiles of UUE were determined with linear regression. The difference in decline of serum creatinine between tertiles of UUE was determined with a Mann Whitney U test. The estimated GFR (eGFR) was calculated using the CKD Epidemiology Collaboration equation (18).

Several patients had missing values for one or more variables. Bias is introduced if subjects with missing values are excluded. Therefore, we performed multiple imputa-tion for values with less than 30% missing data. Finally, the results were combined into pooled results which were only used in Cox regression models.

Univariable survival analyses were performed using log-rank tests. Because sex is an important confounder in many associations (19), we first determined whether asso-ciations of UUE with graft failure and all-cause mortality were modified by sex. Further-more, to discern whether there might be a parallel with the obesity paradox in dialysis patients (20), we explored potential effect modification by BMI. In addition, we inves-tigated potential effect modification by eGFR. Separate Cox regression analyses were performed for men, for women, for RTR with a BMI less than 25 kg/m2, and RTR with a BMI of 25 kg/m2 or higher, for RTR with an eGFR less than 45 mL per min per 1.73 m2 and for RTR with an eGFR of 45 mL per min per 1.73 m2 or higher. Because similar results were obtained in men and women, results were combined for both sexes. Cumu-lative adjustment was performed in Cox regression analyses for age, sex, donor age and sex, urinary sodium excretion, eGFR, and BMI.

Furthermore, we estimated protein intake from UUE. Therefore, we used data from a previous cohort of RTR for which UUE and data on protein intake estimated by a food frequency questionnaire was available (21). Multivariable linear regression pro-vided the following equations: protein intake in women (g/24 hr): 52.67 + 1.03 × UUE (g/24 hr) + 0.10 × age (years) - 3.30; protein intake in men (g/24 hr): 52.67 + 1.03 × UUE (g/24 hr) + 0.10 × age (years). Finally, Cox regression analyses were carried out for the association of estimated protein intake with graft failure and all-cause mortality.

To explore whether 24-hr UCE and serum albumin could lie in the causal pathway between UUE and outcome, we performed secondary analyses, in which we additionally adjusted for serum albumin and UCE. As further secondary analyses, we also repeated Cox regression analyses with adjustment for the change in BMI over the first posttransplant year rather than baseline BMI. Finally, to determine whether the use of high doses of corticoste-roids (i.e. during acute rejection), might have influenced the observations, we performed secondary analyses in which Cox regression analyses were repeated after exclusion of 32 sub-jects who experienced an acute rejection between 5 and 18 months after transplantation.

Statistical analyses were performed using SPSS (version 20.0, IBM, Chicago, IL, USA) and STATA (version 11.0, StataCorp LP, College Station, Texas, USA). A P‑value less than 0.05 was considered statistically significant.

76

Cha

pter

5

Results

Patient characteristics

Median age was 50 (interquartile range, 40–59) years and 59% was male. Mean eGFR was 54.1 ± 18.3 mL per min per 1.73 m2. Mean UUE was higher in men (24.3 ± 5.9 g/24 hr) compared to women (20.6 ± 5.4 g/24 hr; P < 0.001). Baseline characteristics are shown according to tertiles of UUE (Table 1). Prevalence of male sex, age, BSA, BMI, delta BMI, diastolic blood pressure, prevalence of diabetes, albumin, urinary sodium excre-tion, urinary creatinine excretion (UCE), delayed graft function, and use of proliferation inhibitors were positively associated with UUE. Prevalence of tubular interstitial disease as primary renal disease, use of mammalian target of rapamycin inhibitors and cold isch-emic time were inversely related to UUE. Renal transplant recipients with greater UUE received a kidney more frequently from a living donor and less frequently from a CMV seropositive donor. Urinary urea excretion was not significantly associated with eGFR, urinary protein excretion, and maintenance dose of corticosteroids.

Table 1. Baseline characteristics according to tertiles of urea excretion in 940 renal trans-plant recipients.

Tertiles of urinary urea excretion g/24 hrb

Na I II III Pc

Urinary urea, g/24 hr 940 16.6 ± 2.5 22.3 ± 1.4 29.4 ± 4.0Recipient demographics Age, yr 940 49 (37–60) 52 (39–59) 50 (42–59) 0.031 Men, n (%) 940 126 (40) 191 (61) 234 (75) < 0.001 BSA, m2 910 1.83 ± 0.16 1.93 ± 0.17 2.05 ± 0.20 < 0.001 BMI, kg/m2 910 25 ± 4 26 ± 4 28 ± 5 < 0.001 Delta BMId 696 1.5 ± 2.5 1.6 ± 2.6 1.8 ± 2.4 0.031Blood pressure Systolic, mm Hg 899 143 (130–155) 140 (132–153) 143 (133–153) 0.31 Diastolic, mm Hg 899 83 (77–90) 83 (79–90) 85 (79–91) 0.024Total cholesterol, mg/dL 866 223 ± 47 218 ± 41 214 ± 38 0.071Diabetes, n (%) 732 15 (5) 26 (8) 31 (10) 0.013Primary renal disease, n (%) 916 Primary glomerular disease 67 (21) 81 (26) 88 (28) 0.052 Glomerulonephritis 14 (5) 24 (8) 16 (5) 0.75 Tubular interstitial disease 42 (13) 32 (10) 24 (8) 0.004 Polycystic renal disease 57 (18) 53 (17) 63 (20) 0.18 Dysplasia and hypoplasia 9 (3) 5 (2) 3 (1) 0.091 Renovascular disease 22 (7) 23 (7) 32 (10) 0.42


Tertiles of urinary urea excretion g/24 hrb

Na I II III Pc

Diabetic nephropathy 11 (4) 14 (5) 14 (5) 0.92 Other or unknown cause 82 (26) 71 (23) 69 (22) 0.58Donor demographics Donor age, yr 940 46 (35–54) 47 (34–55) 47 (35–55) 0.73 Male donor, n (%) 940 161 (51) 162 (52) 152 (49) 0.65Renal allograft function Serum albumin, g/dL 940 4.2 (4.0–4.4) 4.2 (4.1–4.4) 4.2 (4.1–4.4) 0.007 Serum creatinine, mg/dL 940 1.35 (1.08–1.82) 1.37 (1.12–1.70) 1.41 (1.21–1.75) 0.39 eGFR, ml/min/1.73m2 940 52.7 ± 18.8 55.1 ± 18.6 54.4 ± 17.5 0.14 Urinary protein, g/24 hr 940 0.20 (0.11–0.32) 0.20 (0.11–0.30) 0.25 (0.14–0.39) 0.15 Urinary sodium, mEq/24 hr 940 129 ± 37 154 ± 42 191 ± 55 <0.001 Urinary creatinine, mg/24 hr 940 1149 ± 245 1373 ± 275 1633 ± 323 <0.001Transplantation details Living donor, n (%) 940 63 (20) 86 (27) 100 (32) <0.001 Acute rejection, n (%) 940 100 (32) 98 (31) 116 (37) 0.38 Delayed graft function, n (%) 940 77 (25) 71 (23) 99 (32) 0.040 HLA mismatches, n 746 2 (0–3) 2 (1–3) 2 (1–3) 0.51 Cold ischemic time, hr 939 18 (13–24) 17 (3–23) 16 (3–21) <0.001 Warm ischemia time, min 940 41 ± 12 40 ± 13 42 ± 13 0.43 Dialysis duration prior to Tx, mo 919 37 (18–55) 35 (17–57) 37 (18–58) 0.64CMV status CMV seropositivity recipient, n (%) 661 133 (43) 117 (37) 116 (37) 0.52 CMV seropositivity donor, n (%) 935 155 (50) 164 (52) 143 (46) 0.044Immunosuppressants Use of corticosteroids, n (%) 940 303 (97) 305 (97) 301 (96) 0.59 Maintenance dose, mg/day 776 10 (10–10) 10 (10–10) 10 (10–10) 0.26 Use of calcineurin inhibitor, n (%) 940 267 (88) 277 (88) 281 (90) 0.28 Use of proliferation inhibitor, n (%) 940 232 (74) 243 (77) 260 (83) 0.007 Use of mTOR, n (%) 940 15 (5) 12 (4) 8 (3) 0.027 Other immunosuppressants, n (%) 940 68 (22) 66 (21) 80 (26) 0.06

aNumber of patients with complete data. bRange of urinary urea excretion (g/24 hr) in tertiles: I: ≤20.1, II: 20.1–24.8,

III: ≥ 24.8. cP for linear trend is shown. dDelta BMI was calculated as the difference between baseline BMI and

BMI at transplantation. Dialysis duration prior to transplantation, urinary protein, and serum creatinine were log

transformed. Normally distributed data are given as mean ± standard deviation (SD), skewed data are presented as

median (interquartile range), and categorical distributed variables are given as number (percentage). SI conversion

factors: To convert urinary urea from g/24 hr to mmol/24 hr, multiply values by 16.6513; total cholesterol from mg/

dL to mmol/L, multiply values by 0.02586; serum creatinine from mg/dL to µmol/L , multiply values by 88.4; urinary

creatinine from mg/24 hr to mmol/24 hr, multiply values by 0.00884. BMI, body mass index; BSA, body surface area;

eGFR, estimated glomerular filtration rate; Tx, transplantation; CMV, cytomegalovirus; mTOR, mammalian target

of rapamycin; IQR, interquartile range.


78

Cha

pter

5

UUE and graft failure

Decline in serum creatinine was least pronounced in RTR in the second tertile of UUE (P = 0.048, compared to the first tertile). Serum creatinine was not different between the first and third tertile of UUE (P = 0.34). In line with this observation, graft failure rates were 36 (12%) in the first, 19 (6%) in the second, and 23 (7%) in the third tertile of UUE (log-rank; P = 0.041; Figure 1). In Cox regression analyses, UUE was not associated with graft failure, when all RTR were analyzed together (Table 2). However, there was an in-teraction with BMI. In RTR with a BMI less than 25 kg/m2, we observed a strong inverse association of UUE with graft failure (Supplementary Table S2). Urinary urea excretion was not associated with graft failure in RTR with a BMI of 25 kg/m2 or higher (Supple-mentary Table S3). Furthermore, UUE was not associated with graft failure in RTR with an eGFR less than 45 mL per min per 1.73 m2 (Supplementary Table S4). However, there was an independent inverse association of UUE and graft failure in RTR with an eGFR of 45 mL per min per 1.73 m2 or higher (Supplementary Table S5).

UUE and all-cause mortality

During 4.8 (2.5–8.3) years of follow-up, 158 RTR died. All-cause mortality rates were twice as low in the second and third tertiles of UUE (Table 3 and Figure 2). This associa-tion was independent of sex, BMI, eGFR and other potential confounders.

Because the first tertile of UUE was associated with a greater risk of mortality compared to the second and third tertiles, we estimated the protein intake for the upper border of the first tertile. This upper border was 20.1 g per 24 hr. According to our equa-

Figure 1. Kaplan Meier curves for graft failure according to tertiles of urinary urea excretion. Log-rank test for graft failure; P = 0.041. Range of urinary urea excretion (g/24 hr) in tertiles: I: ≤20.1, II: 20.1–24.8, III: ≥ 24.8.


tion (Materials and Methods), for a 50-year-old man with a weight of 84 kg, this translates into an estimated protein intake of 0.93 g/kg per day. In a woman of the same age, with a weight of 74 kg, this corresponds to an estimated protein intake of 1.01 g/kg per day.

Table 2. Cox regression analyses for graft failure according to tertiles of urinary urea excretion.

Tertiles of urinary urea excretiona

I II III

Reference category

HR (95% CI) P HR (95% CI) P

No. events 36 19 23

Model 1 1.0 0.50 (0.29–0.87) 0.014 0.71 (0.42–1.19) 0.19

Model 2 1.0 0.49 (0.28–0.86) 0.013 0.69 (0.40–1.20) 0.19

Model 3 1.0 0.49 (0.28–0.87) 0.015 0.70 (0.40–1.21) 0.20

Model 4 1.0 0.47 (0.26–0.84) 0.012 0.66 (0.34–1.26) 0.21

aRange of urinary urea excretion (g/24 hr) in tertiles: I: ≤20.1, II: 20.1–24.8, III: ≥ 24.8.

Model 1: Crude association

Model 2: Model 1 + age, sex

Model 3: Model 2 + donor age, and donor sex

Model 4: Model 3 + urinary sodium excretion, eGFR, and BMI

HR, hazard ratio; CI, confidence interval; eGFR, estimated glomerular filtration rate; BMI, body mass index

Figure 2. Kaplan Meier curves for mortality according to tertiles of urinary urea excretion. Log-rank test for mortality; P = 0.001. Range of urinary urea excretion (g/24 hr) in tertiles: I: ≤20.1, II: 20.1–24.8, III: ≥ 24.8.

80

Cha

pter

5

Table 3. Cox regression analyses for mortality according to tertiles of urinary urea excretion.

Tertiles of urinary urea excretiona

I II III

Reference category

HR (95% CI) P HR (95% CI) P

No. events 78 42 38

Model 1 1.0 0.54 (0.37–0.78) 0.001 0.59 (0.40–0.86) 0.007

Model 2 1.0 0.48 (0.33–0.70) < 0.001 0.52 (0.35–0.78) 0.002

Model 3 1.0 0.48 (0.33–0.71) < 0.001 0.53 (0.35–0.79) 0.002

Model 4 1.0 0.47 (0.32–0.69) < 0.001 0.42 (0.26–0.68) < 0.001

aRange of urinary urea excretion (g/24 hr) in tertiles: I: ≤20.1, II: 20.1–24.8, III: ≥ 24.8.




Model 4: Model 3 + urinary sodium excretion, eGFR, and BMI

HR, hazard ratio; CI, confidence interval; eGFR, estimated glomerular filtration rate; BMI, body mass index

Secondary analyses

In secondary analyses, we repeated Cox regression analyses with estimated protein in-take, based on our equation, instead of UUE. In these analyses, results remained essen-tially unchanged. Furthermore, results remained essentially unchanged after additional adjustment for serum albumin for graft failure and mortality. However, adjustment for UCE caused marked weakening of associations of UUE with graft failure (Hazard Ratio [HR], 0.59 [0.32–1.09]; P = 0.092 and HR, 1.02 [0.52–2.00]; P = 0.95 in the second and third tertiles, respectively) and mortality (HR, 0.61 [0.41–0.91]; P = 0.016 and HR, 0.62 [0.89–0.55]; P = 0.62 in the second and third tertiles, respectively). In addition, Cox re-gression analyses were repeated with adjustment for the change in BMI over the first post-transplant year rather than baseline BMI. Results were materially similar for graft failure and mortality. Furthermore, similar results were obtained for graft failure and mortality after exclusion of subjects who experienced rejection during the baseline as-sessment period (n = 32).


DiscussionUUE was strongly inversely associated with graft failure in RTR with a BMI less than 25 kg/m2 and in RTR with an eGFR of 45 mL per min per 1.73m2 or higher. Urinary urea excretion was not related to graft failure in RTR with a BMI of 25 kg/m2 or higher or in RTR with an eGFR less than 45 mL per min per 1.73m2. A more than 50% lower all-cause mortality risk was found in the second and third tertiles of UUE. This association was independent of BMI.

To our knowledge, this study is the first to prospectively investigate associations of protein intake, assessed by UUE, with late graft failure and mortality in a large cohort of RTR. Another method to assess protein intake is by means of a food frequency ques-tionnaire (FFQ). In one cross-sectional study in RTR using an FFQ, no association was found of protein intake with renal function (21).

A study of Dunkler et al. (17) showed that FFQ-derived protein intake was in-versely associated with CKD in patients with type 2 diabetes. In contrast, two reports of the modification of diet in renal disease (MDRD) intervention study (22,23) showed that neither a low (0.58 g/kg/d) nor a very low (0.28 g/kg/d) protein diet slowed renal function progression (22,23), but the assignment to a very low protein diet increased the risk of death in subjects with CKD stage 4 (23). In line, we also found no association of UUE with graft failure in the overall population or in RTR with an eGFR less than 45 mL per min per 1.73 m2, but RTR with a relatively preserved renal function showed a slightly beneficial effect of protein intake on graft failure.

Alongside an FFQ, protein intake can be estimated with the equation by Maroni et al. which is based on UUE. The equation of Maroni et al. was used in a prospective study investigating the association of estimated protein intake and renal function decline in 48 RTR (24). RTR were stratified into a low or high protein group. Renal function deteriorated most in the high protein group. Unfortunately, in that study associations of estimated protein intake with graft failure and mortality were not studied. Our results of an inverse association of protein intake with mortality are in accordance with a study by Halbesma et al. (25) which showed that healthy subjects with the lowest estimated protein intake had the highest all-cause mortality rates.

In our population of RTR, optimal protein intake was higher than what is cur-rently recommended in CKD patients (0.6–0.8 g/kg/day) (3,5,6), but comparable to rec-ommended protein intake in patients treated with dialysis (≥ 1.1 g/kg/day) (6,14,15). Possibly, corticosteroid treatment in RTR induced a tendency for protein catabolism, hereby shifting the optimal protein intake upward. We acknowledge that in the current transplant era, many RTR are transplanted steroid free. However, RTR not using steroids are at an increased risk of acute rejection and chronic allograft nephropathy long-term after transplantation (26,27). Hence, within the first year after transplantation, steroids

82

Cha

pter

5

are often added to a treatment regimen that was initially free of corticosteroids (26), resulting in steroids still being part of the long-term maintenance treatment regimen in approximately 70% of cases (26). We, therefore, believe that our data are applicable to many outpatient RTR long-term after transplantation.

Low protein intake may result in decreased muscle mass. Low UCE, as a marker of low muscle mass, has been consistently shown to be a predictor of mortality in the general population (28) and RTR (29–31). Hence, muscle mass hypothetically lies in the causal pathway between protein intake and outcome. Adjustment for variables in the causal pathway between exposure and outcome may induce bias and can cause signifi-cant associations to disappear (32). Indeed, adjustment for UCE caused material weak-ening of associations of UUE with graft failure and mortality, suggesting that UCE re-flecting muscle mass indeed lies in the causal pathway between UUE reflecting protein intake and outcome.

In dialysis patients, robust inverse associations were found of BMI with mortality rates (20,33). This phenomenon is called the ‘obesity paradox’ (20,33). To study whether the obesity paradox might also prevail among RTR, we explored effect modification by BMI on the associations of UUE with graft failure and mortality. No effect modification was found by BMI on the association of UUE with mortality. In contrast, the association of UUE with graft failure was most pronounced in RTR with a BMI less than 25 kg/m2. Hence, we speculate that at some point, the adverse effects of malnutrition counterbal-ance the supposed renoprotective effects of low protein intake in RTR.

It is plausible that high protein intake might be a marker of a healthy diet, low in saturated fat, low in sugar, or high in fibre. Since we did not collect information on diet, we cannot discern whether this was the case in our study. In addition, we did not have information on the type of protein ingested. We only collected data on sodium ex-cretion, a marker for sodium intake, which was positively associated with protein intake. The associations of UUE with graft failure and mortality remained after adjustment for sodium intake. Although we adjusted for potential confounders, we cannot exclude the possibility of residual confounding. Worth mentioning is that we found UUE positively associated with delayed graft function, while it was inversely associated with ischemia time. This is somewhat unexpected because long ischemia time is an acknowledged risk factor for delayed graft function. We have no good explanation for this observation.

Some limitations of this study need to be addressed. First, a unique asset of this study was that we obtained data from regular patient care. As a consequence, we did not have data on inflammatory markers, body composition and energy intake which are of con-siderable interest. Second, UUE was used as a marker of protein intake. Commonly, pro-tein intake is estimated with Maroni’s equation, taking also body weight into account (34). However, Maroni’s equation was validated for subjects with chronic renal disease only (34). Therefore, we estimated protein intake using a regression equation obtained from a differ-ent study of RTR. Another limitation is that we have no repeated measurements of UUE.


In most epidemiological studies like ours, a baseline classification for a parameter that may not be stable over time is used to study potential associations with outcome (35,36). It is generally acknowledged that this may adversely affect strength of associations with outcomes and even lead to false negation of otherwise truly existing associations (35,36). We, however, found a significant association despite these shortcomings of using tertiles of UUE. Finally, in observational studies, cause-effect relationships cannot be ascertained. The strength of the present study is that UUE was measured multiple times over a 6-month period. Using the median of multiple measurements reduces measurement errors. More-over, we included a large number of stable RTR, and the follow-up period exceeded that of many other studies. Another strength of our study is that there was no loss to follow-up.

In conclusion, UUE, a marker for protein intake, was not associated with graft failure, but was inversely associated with graft failure in RTR with a BMI less than 25 kg/m2 and in RTR with an eGFR of 45 mL per min per 1.73 m2 or higher. Furthermore, RTR with UUE below 20.1 g per 24 hr had the highest mortality rates. These findings need to be confirmed in other prospective cohorts and in intervention studies. In the meantime, it may be better to avoid low protein intake in RTR and be heedful of a habitual low protein intake in RTR.

AcknowledgementsThe contribution of Daan Kromhout was supported by the Royal Netherlands Academy of Arts and Sciences.

References 1. Brantsma AH, Atthobari J, Bakker SJ, de Zeeuw D, de Jong PE, Gansevoort RT. What predicts

progression and regression of urinary albumin excretion in the nondiabetic population? J Am Soc Nephrol 2007 Feb;18(2):637–645.




5. KDOQI Clinical Practice Guidelines and Clinical Practice Recommendations for Diabetes and Chronic Kidney Disease. Am J Kidney Dis 2007 Feb;49(2 Suppl 2):S12–154.

84

Cha

pter

5



8. Carrero JJ. Mechanisms of altered regulation of food intake in chronic kidney disease. J Ren Nutr 2011 Jan;21(1):7–11.

9. Kalantar-Zadeh K, Block G, McAllister CJ, Humphreys MH, Kopple JD. Appetite and inflam-mation, nutrition, anemia, and clinical outcome in hemodialysis patients. Am J Clin Nutr 2004 Aug;80(2):299–307.

10. Aparicio M, Cano N, Chauveau P, Azar R, Canaud B, Flory A, et al. Nutritional status of hae-modialysis patients: a French national cooperative study. French Study Group for Nutrition in Dialysis. Nephrol Dial Transplant 1999 Jul;14(7):1679–1686.

11. Burrowes JD, Larive B, Cockram DB, Dwyer J, Kusek JW, McLeroy S, et al. Effects of dietary intake, appetite, and eating habits on dialysis and non-dialysis treatment days in hemodialysis patients: cross-sectional results from the HEMO study. J Ren Nutr 2003 Jul;13(3):191–198.





16. Reznichenko A, Snieder H, van den Born J, de Borst MH, Damman J, van Dijk MC, et al. CUBN as a novel locus for end-stage renal disease: insights from renal transplantation. PLOS One 2012;7(5):e36512.

17. DuBois D, DuBois EF. A formula to estimate the approximate surface area if height and weight be known. Arch Intern Medicine 1916;17:863–17.


19. Ruiz-Cantero MT, Vives-Cases C, Artazcoz L, Delgado A, Garcia Calvente MM, Miqueo C, et al. A framework to analyse gender bias in epidemiological research. J Epidemiol Community Health 2007 Dec;61 Suppl 2:ii46–53.

20. Kalantar-Zadeh K, Abbott KC, Salahudeen AK, Kilpatrick RD, Horwich TB. Survival advan-tages of obesity in dialysis patients. Am J Clin Nutr 2005 Mar;81(3):543–554.

21. van den Berg E, Engberink MF, Brink EJ, van Baak MA, Gans RO, Navis G, et al. Dietary protein, blood pressure and renal function in renal transplant recipients. Br J Nutr 2013 Apr 28;109(8):1463–70.


22. Levey AS, Greene T, Sarnak MJ, Wang X, Beck GJ, Kusek JW, et al. Effect of dietary protein restriction on the progression of kidney disease: long-term follow-up of the Modification of Diet in Renal Disease (MDRD) Study. Am J Kidney Dis 2006 Dec;48(6):879–888.

23. Menon V, Kopple JD, Wang X, Beck GJ, Collins AJ, Kusek JW, et al. Effect of a very low-protein diet on outcomes: long-term follow-up of the Modification of Diet in Renal Disease (MDRD) Study. Am J Kidney Dis 2009 Feb;53(2):208–217.

24. Bernardi A, Biasia F, Pati T, Piva M, D’Angelo A, Bucciante G. Long-term protein intake control in kidney transplant recipients: effect in kidney graft function and in nutritional status. Am J Kidney Dis 2003 Mar;41(3 Suppl 1):S146–52.

25. Halbesma N, Bakker SJ, Jansen DF, Stolk RP, De Zeeuw D, De Jong PE, et al. High protein in-take associates with cardiovascular events but not with loss of renal function. J Am Soc Nephrol 2009 Aug;20(8):1797–1804.

26. Schold JD, Santos A, Rehman S, Magliocca J, Meier-Kriesche HU. The success of continued steroid avoidance after kidney transplantation in the US. Am J Transplant 2009 Dec;9(12):2768–2776.

27. Meier-Kriesche HU, Magee JC, Kaplan B. Trials and tribulations of steroid withdrawal after kidney transplantation. Am J Transplant 2008 Feb;8(2):265–266.



30. Streja E, Molnar MZ, Kovesdy CP, Bunnapradist S, Jing J, Nissenson AR, et al. Associations of pretransplant weight and muscle mass with mortality in renal transplant recipients. Clin J Am Soc Nephrol 2011 Jun;6(6):1463–1473.

31. Huang CX, Tighiouart H, Beddhu S, Cheung AK, Dwyer JT, Eknoyan G, et al. Both low muscle mass and low fat are associated with higher all-cause mortality in hemodialysis patients. Kid-ney Int 2010 Apr;77(7):624–629.

32. Hernan MA, Hernandez-Diaz S, Werler MM, Mitchell AA. Causal knowledge as a prerequisite for confounding evaluation: an application to birth defects epidemiology. Am J Epidemiol 2002 Jan 15;155(2):176–184.



35. Danesh J, Wheeler JG, Hirschfield GM, Eda S, Eiriksdottir G, Rumley A, et al. C-reactive pro-tein and other circulating markers of inflammation in the prediction of coronary heart disease. N Engl J Med 2004 Apr 1;350(14):1387–1397.

36. van Ree RM, de Vries AP, Oterdoom LH, Seelen MA, Gansevoort RT, Schouten JP, et al. Plasma procalcitonin is an independent predictor of graft failure late after renal transplantation. Trans-plantation 2009 Jul 27;88(2):279–287.

86

Cha

pter

5

Supplementary Table S1. Conversion equations for results of laboratory measurements performed on Merck Mega Analyzer or Roche Modular.

Measurement Conversion equation

Urinary urea Ya = (Xb ‑ 9) / 0.996

Total cholesterol No difference

Serum albumin Ya = (Xb + 6) / 1.132

Serum creatinine Ya = (Xb ‑ 8) / 1.07

Urinary protein Ya = (Xb + 0.05) / 1.403

Urinary sodium Ya = Xb / 0.918

Urinary creatinine No difference

aRoche Modular (Roche Ltd., Mannheim, Germany)bMerck Mega Analyzer (Merck, Darmstadt, Germany)

Conversion equations are presented in SI units.

Supplementary Table S2. Cox regression analyses for graft failure according to BMI-stratified tertiles of urinary urea excretion, in subjects with BMI <25 kg/m2.

BMI‑stratified tertiles of urinary urea excretiona

I II III

Reference category HR (95% CI) HR (95% CI) P‑valueb

No. of events 16 10 5

Model 1 1.0 0.67 (0.30–1.47) 0.37 (0.14–1.01) 0.001

Model 2 1.0 0.57 (0.25–1.28) 0.31 (0.11–0.88) <0.001

Model 3 1.0 0.60 (0.26–1.36) 0.29 (0.10–0.81) <0.001

Model 4 1.0 0.64 (0.28–1.50) 0.27 (0.09–0.83) <0.001

aRange of urinary urea excretion (g/24h) in tertiles: I, ≤18.4; II, 18.4–22.8; III, ≥ 22.8bP for trend is shown




Model 4: Model 3 + urinary sodium excretion and estimated GFR

Abbreviations: HR, hazard ratio; CI, confidence interval


Supplementary Table S3. Cox regression analyses for graft failure according to BMI-stratified tertiles of urinary urea excretion, in subjects with BMI ≥ 25 kg/m2.

BMI‑stratified tertiles of urinary urea excretiona

I II III

Reference category HR (95% CI) HR (95% CI) P‑valueb


Model 1 1.0 0.86 (0.42–1.74) 0.98 (0.48–1.98) 0.59

Model 2 1.0 0.80 (0.39–1.66) 0.95 (0.46–1.97) 0.58

Model 3 1.0 0.83 (0.40–1.71) 1.00 (0.48–2.09) 0.73

Model 4 1.0 1.10 (0.51–2.34) 1.31 (0.57–3.04) 0.78

aRange of urinary urea excretion (g/24h) in tertiles: I, ≤ 21.1; II, 21.1–25.7; III, ≥ 25.8bP for trend is shown




Model 4: Model 3 + urinary sodium excretion and estimated GFR


Supplementary Table S4. Cox regression analyses for graft failure according to eGFR-stratified tertiles of urinary urea excretion, in subjects with eGFR <45 ml/min/1.73m2.

eGFR‑stratified tertiles of urinary urea excretiona

I II III

Reference category HR (95% CI) P‑value HR (95% CI) P‑value


Model 1 1.0 0.39 (0.18–0.86) 0.020 0.85 (0.43–1.66) 0.63

Model 2 1.0 0.45 (0.20–1.02) 0.055 1.02 (0.48–2.13) 0.97

Model 3 1.0 0.45 (0.20–1.01) 0.054 0.97 (0.46–2.04) 0.94

Model 4 1.0 0.41 (0.18–0.94) 0.035 0.95 (0.40–2.27) 0.91

aRange of urinary urea excretion (g/24h) in tertiles: I, ≤20.1; II, 20.1–23.7; III, ≥ 23.7




Model 4: Model 3 + urinary sodium excretion and body mass index


88

Cha

pter

5

Supplementary Table S5. Cox regression analyses for graft failure according to eGFR-stratified tertiles of urinary urea excretion, in subjects with eGFR ≥ 45 ml/min/1.73m2.

eGFR‑stratified tertiles of urinary urea excretiona

I II III

Reference category HR (95% CI) P‑value HR (95% CI) P‑value


Model 1 1.0 0.61 (0.28–1.35) 0.23 0.60 (0.26–1.40) 0.24

Model 2 1.0 0.53 (0.24–1.19) 0.12 0.55 (0.23–1.32) 0.18

Model 3 1.0 0.47 (0.21–1.07) 0.071 0.55 (0.23–1.34) 0.19

Model 4 1.0 0.34 (0.15–0.79) 0.015 0.31 (0.11–0.86) 0.025

bRange of urinary urea excretion (g/24h) in tertiles: I, ≤22.3; II, 22.3–24.1; III, ≥ 24.1




Model 4: Model 3 + urinary sodium excretion and body mass index


Supplementary Figure S1. Study design. The median of laboratory measurements between 0.5 year (y) and 1.5 years after transplantation was used for analyses. From 1.5 years after trans-plantation, follow-up was recorded until the incidence of death, or after 10 years of follow-up.

6

Alanine aminotransferase

and mortality in patients

with type 2 diabetes

(ZODIAC‑38)6Petronella E. Deetman1, Alaa Alkhalaf2, Gijs W.D. Landman3,4, Klaas H. Groenier4,5, Jenny E. Kootstra-Ros6, Gerjan Navis1, Henk J.G. Bilo1,4, Nanne Kleefstra,4,7, Stephan J.L Bakker1

1 Department of Medicine, University of Groningen, University Medical Center Groningen, the Netherlands;

2 Department of Gastroenterology, Isala Clinics, Zwolle, the Netherlands;

3 Department of Medicine, Gelre Hospital, Apeldoorn, the Netherlands;

4Diabetes Centre, Isala Clinics, Zwolle, the Netherlands; 5 Department of General Practice, University of Groningen, University Medical Center Groningen, the Netherlands;

6 Department of Laboratory Medicine, University Medical Center Groningen, the Netherlands;

7 Langerhans Medical Research Group, Zwolle, the Netherlands.

Eur J Clin Invest, provisionally accepted

90

Cha

pter

6

Abstract Background. Combined data suggest a bimodal association of alanine aminotransferase (ALT) with mortality in the general population. Little is known about the association of ALT with mortality in patients with type 2 diabetes. We therefore investigated the as-sociation of ALT with all-cause, cardiovascular, and non-cardiovascular mortality in pa-tients with type 2 diabetes. Methods. A prospective study was performed in patients with type 2 diabetes, treated in primary care, participating in the Zwolle Outpatient Diabetes project Integrating Available Care (ZODIAC) study. Cox regression analyses were performed to determine associations of log2-transformed baseline ALT with all-cause, cardiovascular, and non-cardiovascular mortality. Results. In 1,187 patients with type 2 diabetes (67 ± 12 years, 45% female), ALT levels were 11 (8–16) U/l. During median follow-up for 11.1 (6.1–14.0) years, 553 (47%) pa-tients died, with 238 (20%) attributable to cardiovascular causes. Overall, ALT was in-versely associated with all-cause mortality (Hazard Ratio [HR] 0.81; 95% Confidence Interval [CI] 0.72–0.92), independently of potential confounders. This was less attribut-able to cardiovascular mortality (HR 0.87; 95% CI 0.72–1.05), than to non-cardiovascular mortality (HR 0.77; 95% CI 0.65–0.90). Despite the overall inverse association of ALT with mortality, it appeared that a bimodal association with all-cause mortality was pres-ent with increasing risk for levels of ALT above normal (P = 0.003). Conclusions. In patients with type 2 diabetes, low levels of ALT are associated with an in-creased risk of all-cause mortality, in particular non-cardiovascular mortality, compared to normal levels of ALT, while risk again starts to increase when levels are above normal.

91ALT and mortality in patients with type 2 diabetes

IntroductionThe prevalence of type 2 diabetes is on the rise, with already more than 382 million pa-tients affected worldwide (1). Type 2 diabetes and the metabolic syndrome are associated with hepatic fat accumulation (2,3). Hepatic fat accumulation, in the absence of alcohol abuse, is known as non-alcoholic fatty liver disease (NAFLD). NAFLD is a heteroge-neous disease that ranges from simple steatosis to steatohepatitis. Approximately 50–75% of the patients with type 2 diabetes will develop some form of NAFLD during their life (4,5). Because NAFLD is highly prevalent among subjects with the metabolic syn-drome, NAFLD is considered the hepatic component of the metabolic syndrome (2,6).

The presence of NAFLD has been associated with an increased risk of micro-vascular and macrovascular complications in patients with type 2 diabetes (5,7). One of those complications may be chronic kidney disease (CKD). It has been shown that the prevalence of CKD is higher among patients with NAFLD (8,9). Patients with type 2 dia-betes and NAFLD had a 2-fold increased risk of mortality compared to type 2 diabetes patients without NAFLD (7). In most cases of NAFLD, the enzyme alanine aminotrans-ferase (ALT) is elevated (10). Elevated levels of ALT are therefore regarded as a marker of NAFLD, and have been independently associated with an increased risk of coronary heart disease in population-based studies (11,12). However, there are now also indica-tions that low ALT is a risk factor for mortality, suggesting that there could be a bimodal, U-shaped association of ALT with mortality (13–15).

It is unknown whether ALT is associated with long-term outcome in patients with type 2 diabetes. Therefore, the aim of the present study was to investigate the as-sociation of ALT with progression of renal function, all-cause, cardiovascular, and non-cardiovascular mortality in a large prospective cohort of patients with type 2 diabetes, and to investigate whether these associations are U-shaped rather than linear.



The Zwolle Outpatient Diabetes Project Integrating Available Care (ZODIAC) study was initiated in 1998 in the Zwolle region of the Netherlands. The design and details of this study have been published elsewhere (16,17). In brief, this is a prospective study in pri-mary care patients with type 2 diabetes. In 1998, 1,143 patients with type 2 diabetes were included and in 2001, an additional 546 patients with type 2 diabetes were enrolled, re-sulting in a combined cohort of 1,689 patients (18). ALT was measured in 1,187 patients (70%). The ZODIAC study was approved by the local medical ethics committee, and all patients provided informed consent.

92

Cha

pter

6

Data collection and measurements

Baseline data, collected in 1998 and 2001, consisted of a full medical history, including a history of cardiovascular disease (CVD), use of medication, and tobacco consump-tion. Subjects were considered to have a positive history of CVD if they had a history of angina pectoris, myocardial infarction, percutaneous transluminal coronary angioplas-ty, coronary artery bypass grafting, stroke, or transient ischemic attack. The RAND-36 questionnaire was taken. According to the Fried frailty criteria (19), physical weakness is a component of the frailty phenotype. The physical functioning subscale of the RAND-36 questionnaire can serve as the physical weakness component of the Fried frailty cri-teria (20). Baseline laboratory and physical assessment data were collected and included gamma-glutamyltransferase (GGT), nonfasting lipid profile, HbA1c, serum creatinine, urinary albumin, and blood pressure. Every year, data on serum creatinine and urine albumin concentration were collected.

Samples were stored at -80° C until analyzed for ALT. In a previous study it was shown that the level of ALT was stable after three months of storage at -20° C (21). The stability of ALT was confirmed by data from the manufacturer of the assay, which indi-cate that ALT levels are stable when stored at -70° C (unpublished data). ALT was mea-sured with an UV test with pyridoxal phosphate activation according to IFCC recom-mendations (Roche Modular, Roche, Mannheim, Germany) (22). In our laboratory, the coefficients of variation (CVs) were determined for ALT levels of 51 U/L and 164 U/L. At these specific levels, the inter-assay CVs for the entire measurement period are 1.9% and 1.4%, respectively. The intra-assay CVs are 2.1% and 1.2%, respectively. In our labo-ratory, the upper limit of normal of ALT is < 45 U/L for men and < 34 U/L for women (22). Blood pressure was measured twice with a Welch Allyn sphygmomanometer in the supine position after at least 5 min of rest. The eGFR was calculated according to the Chronic Kidney Disease Epidemiology Collaboration equation (23).

Clinical end-points

In this study, associations were examined of baseline ALT with three clinical end points: progression of renal function, all-cause, cardiovascular, and non-cardiovascular mortal-ity. Progression of renal function was defined as a confirmed doubling of serum cre-atinine or incident micro- or macroalbuminuria. Microalbuminuria was defined as an albumin to creatinine ratio (ACR) of >2.5 mg/mmol for men and >3.5 mg/mmol for women. In 2012, vital status and cause of death were retrieved from medical records. Cause of death was coded according to the ICD-9. Cardiovascular death was defined as any death in which the principal cause of death was cardiovascular in nature (ICD-9 codes 390–459) (16,17).



Baseline characteristics of the study population were calculated according to sex-strati-fied tertiles of ALT. Results are expressed as mean ± SD for normally distributed data and median (interquartile range [IQR]) for skewed data. Categorical distributed data are pre-sented as the total number of subjects with percentage (n [%]). P‑values for a linear trend across sex-stratified tertiles of ALT were determined with linear regression analyses. Re-siduals were checked for normality and variables were log-transformed when appropriate.

Univariable survival analyses were performed using log-rank tests. Adjustment for potential confounders was performed in multivariable Cox regression analyses. Cox regression analyses were performed with ALT as a continuous variable. ALT was log2-transformed. Therefore, HRs and 95% CI are shown per each doubling of ALT. We also tested for potential interactions of ALT with age and sex for associations with mortality by means of product-terms of the respective variables. The final model included age, sex, BMI, smoking, systolic blood pressure, cholesterol/HDL ratio, diabetes duration, serum creatinine, HbA1c, cardiovascular history, use of ACE inhibitors, and ACR. The pro-portionality of hazards was tested by checking Schoenfeld’s residuals. Finally, we tested for potential non-linearity of the prospective associations of ALT with mortality (24). For the analyses in which we investigated the association of ALT with incident micro or macroalbuminuria, patients with albuminuria at baseline were excluded.

To investigate the discriminative and predictive capabilities of ALT, we deter-mined the Harrell’s C, the Net Reclassification Improvement (NRI) and Integrated Dis-crimination Improvement (IDI). The Harrell’s C is a measure of how well a model distin-guishes subjects who will get the outcome from subjects who will not get the outcome. The Harrell’s C was calculated for crude and multivariable-adjusted Cox regression models and for a Cox regression model without ALT. To estimate to which extent ALT improves risk classification, the NRI and IDI were calculated. With the NRI procedure, subjects are classified into low, intermediate and high risk categories. Cut off points for risk classification, as applied for NRI were 15, 30 and 40 for all-cause mortality and 10, 20, and 30 for non-cardiovascular mortality. IDI is a continuous version of the NRI (25). These analyses were done for all-cause and non-cardiovascular mortality only, because ALT was significantly associated with these outcomes. Finally, a goodness-of-fit test, as proposed by Grønnesby and Borgan, was performed for Cox regression analyses.

As sensitivity analyses, Cox regression analyses were performed after excluding subjects using statins, and after excluding subjects with ALT levels above the upper limit of normal. Data on the physical functioning subscale of the RAND-36 questionnaire were available in approximately two-thirds of patients (n = 735, 62%). As additional sen-sitivity analyses, we repeated the Cox regression analyses with additional adjustment for the physical functioning subscale of the RAND-36 questionnaire. We performed these analyses only in patients with complete data on this measure.

94

Cha

pter

6

Statistical analyses were performed with SPSS version 20.0 for Windows (IBM Corp., Armonk, NY), and STATA version 11 (StataCorp LP, College Station, TX, USA). A P < 0.05 was considered to indicate statistical significance.

Results

Patients characteristics

A total of 1,187 patients with type 2 diabetes (mean age 67 ± 12 years, 54% female) were included. At baseline, the median (IQR) duration of diabetes was 4 (2–9) years. The me-dian ALT level was 11 (8–16) U/l. A total of 23 patients had an ALT level above the upper limit of normal. Baseline characteristics are shown according to sex-stratified tertiles of ALT (Table 1). ALT was positively associated with GGT, HbA1c, BMI, physical func-tioning, systolic blood pressure, diastolic blood pressure, total cholesterol, triglycerides, lipid lowering drugs, eGFR, and albuminuria. There were inverse relationships of ALT with age, duration of diabetes, HDL, and serum creatinine. ALT was not associated with smoking status, use of ACE inhibition, and ACR.

ALT and outcomes

During a median follow-up of 11.1 (6.1–14.0) years, 553 (47%) patients died. Of those subjects, 238 (43%) subjects died of CVD. All-cause mortality rates were 221 (62%), 211 (46%), and 121 (32%) in the respective tertiles of ALT (P < 0.001, Figure 1). A total of 127 patients (16%) developed a 50% increase in serum creatinine. Of the patients with normoalbuminuria at baseline, 126 (32%) of the patients developed albuminuria.

In a univariable Cox regression analysis, ALT was not associated with 50% in-crease in serum creatinine (Hazard Ratio [HR], 1.24, 95% CI 0.88–1.73; P = 0.22) or in-cident albuminuria (HR, 1.03, 95% CI 0.79–1.34; P = 0.81). These associations remained nonsignificant after adjustment for potential confounders. ALT was inversely associated with all-cause mortality (Table 2), independently of potential confounders. No interac-tions were found of ALT with age and sex. ALT was not associated with cardiovascular mortality, but ALT was associated with non-cardiovascular mortality. In addition, de-spite the overall inverse association of low to normal levels of ALT with mortality, there was a U-shaped association of ALT and all-cause mortality with again increasing risk for levels of ALT above normal (P = 0.003, Figure 2).


Table 1. Baseline characteristics of the study population presented for the whole study population, and for sex stratified tertiles of alanine aminotransferase (ALT).

ALTa

Study

populationI II III P‑valueb

Liver Function

ALT (U/L) 11 (8–16) 7 (6–8) 11 (10–12) 19 (16–25)

GGT (U/L) 30 (21–47) 23 (16–34) 28 (21–41) 45 (30–75) <0.001

Demographics

Age (years) 67 ± 12 71 ± 11 67 ± 11 63 ± 11 <0.001

Male sex, n(%) 540 (46) 162 (46) 205 (45) 173 (46)

Current smoker, n(%) 225 (19) 72 (20) 89 (20) 64 (17) 0.73

Duration of diabetes (yr) 4 (2–9) 5 (2–10) 4 (2–9) 3 (2–8) 0.002

HbA1c, (%) 7.0 (6.2–8.0) 6.7 (6.1–7.7) 6.9 (6.2–7.9) 7.4 (6.5–8.5) <0.001

HbA1c, (mmol/mol) 53 (44–64) 50 (43–61) 52 (44–63) 57 (48–69) <0.001

History of CVD, n(%) 342 (29) 113 (32) 135 (30) 94 (25) 0.014

Body Composition

BMI (kg/m2) 29 ± 5 28 ± 5 29 ± 4 30 ± 5 <0.001

Physical functioningc 60 (30–85) 55 (25–80) 60 (35–85) 70 (35–90) 0.012

Blood pressure

Systolic BP (mmHg) 152 ± 24 150 ± 24 152 ± 23 154 ± 24 0.034

Diastolic BP (mmHg) 83 ± 11 80 ± 11 84 ± 10 86 ± 11 <0.001

Use of ACE inhibition, n(%) 320 (27) 97 (27) 125 (27) 98 (26) 0.73

Lipids

Total cholesterol (mmol/L) 5.51 ± 1.12 5.4 ± 1.1 5.5 ± 1.1 5.7 ± 1.1 0.002

HDL (mmol/L) 1.1 (1.0–1.4) 1.2 (1.0–1.5) 1.2 (1.0–1.4) 1.1 (0.9–1.3) <0.001

Triglycerides (mmol/L) 2.1 (1.5–3.0) 1.6 (1.1–2.2) 2.0 (1.4–2.7) 2.8 (1.9–4.1) <0.001

Lipid lowering drugs, n(%) 191 (16) 36 (10) 79 (17) 76 (20) 0.007

Renal function

Serum creatinine (µmol/L) 92 (82–103) 94 (84–107) 91 (81–103) 90 (81–100) <0.001

eGFR (ml/min/1.73m2) 69 ± 17 64 ± 16 69 ± 16 72 ± 17 <0.001

ACR (mg/mmol) 1.9 (0.9–6.5) 2.1 (1.0–6.8) 1.6 (0.8–6.2) 2.0 (0.9–6.7) 0.98

Albuminuria, n(%) 452 (38) 137 (39) 159 (35) 156 (42) 0.048

aRange of ALT (U/L) in men: I ≤9; II 10–14; III ≥ 15. Range of ALT (U/L) in women: I ≤8; II 9–13; III ≥ 14. bP

for linear trend is shown.cPhysical functioning was estimated using the physical functioning subscale of the RAND‑

36 questionnaire. ALT, GGT, duration of diabetes, HDL, triglycerides, serum creatinine, and ACR were log‑trans‑

formed. ALT, alanine aminotransferase; GGT, gamma‑glutamyltransferase; CVD, cardiovascular disease; BP, blood

pressure; ACR, albumin to creatinine ratio.

96

Cha

pter

6

Table 2. Cox regression analyses of log2-transformed alanine aminotransferase (ALT) with all-cause, cardiovascular and non-cardiovascular mortality.

All-cause mortality Cardiovascular mortalityNon-cardiovascular

mortality

HR (95%CI) P‑value HR (95%CI) P‑value HR (95%CI) P‑value

Number of events 515 219 296

Model 1 0.61 (0.54–0.68) <0.001 0.63 (0.53–0.75) <0.001 0.59 (0.51–0.68) <0.001

Model 2 0.81 (0.72–0.91) <0.001 0.84 (0.70–1.01) 0.062 0.78 (0.67–0.92) 0.002

Model 3 0.79 (0.70–0.88) <0.001 0.82 (0.69–0.98) 0.032 0.76 (0.65–0.89) 0.001

Model 4 0.81 (0.72–0.92) 0.001 0.87 (0.72–1.05) 0.15 0.77 (0.65–0.90) 0.001

HRs per each doubling in ALT.

Model 1; crude model

Model 2; model 1 + age, sex

Model 3; model 2 + BMI, cholesterol/HDL ratio

Model 4; model 3 + smoking, systolic blood pressure, diabetes duration, serum creatinine, glycated hemoglobin, car‑

diovascular history, ACE inhibtion, albumin to creatinine ratio

Figure 1. Kaplan Meier curve for sex-stratified tertiles of alanine aminotransferase (ALT) and cumulative survival (%). Range of ALT (U/L) in men: I ≤9; II 10–14; III ≥ 15. Range of ALT (U/L) in women: I ≤8; II 9–13; III ≥ 14. Log-rank test P < 0.001.


Predictive value of ALT and goodness of fit

The discriminative capabilities of ALT for all-cause and non-cardiovascular mortality, as determined by the Harrell’s C, are shown in Table 3. As expected, when more vari-ables were added to the model in addition to ALT, the discriminative ability increased. Discriminative abilities were similar in final multivariable-adjusted models with and without ALT. This indicates that the addition of ALT to Cox regression models does not materially improve the discriminative ability of saturated survival models. In addition, both the NRI and IDI were relatively low Table 3 indicating small additional predictive value of ALT on top of established predictors in Cox regression models. Furthermore, the goodness-of-fit tests for cardiovascular and non-cardiovascular mortality were not statistically significant.

Figure 2. Fractional polynomial of the hazard ratio for all-cause mortality, relative to the dis-tribution of log2-transformed alanine aminotransferase (ALT). In this figure, log2-transformed values of ALT are shown. Data of four patients with extreme values of log2-transformed ALT (≤1 and > 7) is not shown

98

Cha

pter

6

Table 3. Additional predictive value of log2-transformed alanine aminotransferase (ALT) on all-cause and non-cardiovascular mortality.

Harrell’s C (95%CI)

IDI (95%CI) NRI (95%CI)Grønnesby & Borgan test

All-cause mortality

Model 1 0.61 (0.58–0.63) NA NA 0.21

Model 2 0.77 (0.75–0.78) 0.51 (0.06–0.95) 4.28 (1.28–7.28) 0.22

Model 3 0.77 (0.75–0.79) 0.77 (0.24–1.29) 2.00 (‑1.51–5.52) 0.10

Model 4 0.79 (0.77–0.81) 0.52 (0.07–0.97) 3.03 (‑0.17–6.22) 0.63

Confounders only 0.79 (0.77–0.81) NA NA 0.34

Non-cardiovascular mortality

Model 1 0.61 (0.58–0.64) NA NA 0.70

Model 2 0.77 (0.75–0.80) 0.30 (‑0.08–0.68) 0.73 (‑3.14–4.61) 0.46

Model 3 0.77 (0.75–0.80) 0.35 (‑0.06–0.76) 0.60 (‑3.45–4.66) 0.14

Model 4 0.79 (0.76–0.81) 0.41 (‑0.06–0.88) 2.72 (‑0.90–6.34) 0.73

Confounders only 0.78 (0.76–0.81) NA NA 0.48

Model 1: crude model

Model 2: model 1 + age, sex

Model 3: model 2 + BMI, cholesterol/HDL ratio

Model 4: model 3 + smoking, systolic blood pressure, diabetes duration, serum creatinine, glycated hemoglobin, car‑

diovascular history, ACE inhibition, albumin to creatinine ratio

IDI, integrated discrimination improvement; NRI, net reclassification improvement; BMI, body mass index; HDL,

high density lipoprotein;ACE, angiotensin converting enzyme.

Sensitivity analyses

In a sensitivity analysis, we investigated whether ALT was associated with all-cause and non-cardiovascular mortality, irrespective of statin treatment. After the exclusion of subjects us-ing statins (n = 191), results remained essentially similar. Furthermore, after the exclusion of subjects with ALT levels above the ULN (≥ 45 U/l for men and ≥ 34 U/l for women, n = 23), we found similar results for the association with all-cause and non-cardiovascular mortality. The U-shape of the association of ALT with all-cause mortality disappeared after exclusion of these subjects (P for non-linearity = 0.32). Finally, results remained materially unchanged after additional adjustment for an indicator of physical functioning.


DiscussionIn this prospective cohort of patients with type 2 diabetes, we found an inverse associa-tion of serum levels of ALT with all-cause mortality. Furthermore, ALT was not associ-ated with cardiovascular mortality, but particularly with non-cardiovascular mortality. These associations were independent of potential confounders. ALT was not associated with progression of renal function. Furthermore, there was a bimodal, U-shaped, asso-ciation of ALT with all-cause mortality.

To our knowledge, the association of ALT with progression of renal function has not been investigated to date. In previous studies it has been shown that the prevalence of CKD is significantly higher among patients with both diabetes and NAFLD compared to those without NAFLD (8,9). In those studies, the majority of patients had ALT levels within the normal range (8,9). Thus, ALT may be an imperfect marker for NAFLD. This may explain why we did not find an association of ALT with progression of renal function.

It is established that elevated levels of ALT are associated with NAFLD. Besides other components of the metabolic syndrome, NAFLD is associated with increased car-diovascular mortality (26,27). In contrast with our findings of an inverse association of ALT with mortality, three population-based cohort studies have reported positive as-sociations of ALT with mortality among subjects with levels of ALT above the ULN (12,28,29). Like mentioned, our study differs from these studies in that we investigated the association of ALT with mortality comparing subjects within the normal range. Sev-eral population-based studies that mainly compared subjects within the normal range of ALT also found an inverse association of ALT with mortality (13–15,30–32).

Apparently, studies that focused on high versus normal ALT found positive as-sociations of ALT with mortality, whereas studies that focused on ALT within the normal range found inverse associations of ALT with mortality. It is tempting to speculate that both high and low levels of ALT are associated with mortality. Such a trend would be reflected in a U-shaped relationship or bimodal association. Indeed, other studies found bimodal associations of ALT with all-cause mortality (13,28). In line, the results of the present study also showed a significant bimodal association of ALT with all-cause mortal-ity. It might also be speculated that ALT above the normal range heralds NAFLD (10) and might therefore be a risk factor of cardiovascular mortality, whereas low ALT increases the risk of non-cardiovascular mortality. Because few subjects in our study had ALT levels above the normal range, this might explain why we did not find an association of ALT with cardiovascular mortality whereas previous studies that focused on elevated levels of ALT indeed found a positive association of ALT with cardiovascular mortality (29).

There are several potential explanations for the inverse association of ALT with all-cause mortality in the present study. Firstly, ALT levels are associated with age (33–

100

Cha

pter

6

35). Elinav et al. reported the highest ALT levels at 50 years, with decreasing ALT levels with advancing age (35). In the present study, ALT and age were also correlated. Interest-ingly, in studies that are in disagreement with our findings, mean age was approximately 13 years lower compared to our study population (12,36). However, we did not find evi-dence for an interaction of ALT with age and adjustment for age did not alter the associa-tion of ALT with mortality. It has been suggested that ALT levels decrease with advanc-ing age because of hepatic aging (14). During aging, hepatic blood flow decreases, which subsequentially reduces the number of functional hepatocytes (14). Because hepatocytes are the main site for production of ALT, a decrease in functional hepatocytes is reflected by a decrease in ALT (37,38). A decrease in functional hepatocytes has detrimental ef-fects, since hepatocytes are important mediators in detoxification, lipid metabolism, and glucose regulation (39,40).

Frailty may be another potential explanation. In our analyses, we used the physi-cal functioning subscale of the RAND-36 questionnaire, which has been used to serve as the physical weakness component of the Fried frailty criteria (19,20), rather than the complete set of Fried frailty criteria. This means that, although we found the association of ALT with mortality to be independent of physical functioning, we cannot exclude the possibility that frailty actually mediates the association of ALT with mortality, as has been found in a previous study on this subject (15). Furthermore, although ALT is generally thought of in relation to the liver, ALT is also found in other tissues including skeletal muscle, albeit in smaller quantities (41). It may be possible that increased mor-tality found with low ALT could result from smaller muscle mass. Because we do not have data available about muscle mass we, unfortunately, could not take this possibility into consideration.

It should also be noted that the level of ALT was lower than expected. Because ALT is stable when stored at -20° C (21), and because in the present study samples were stored at -80° C, we believe that it is unlikely that frozen storage has affected the results concerning the measurement of ALT. In addition, pyridoxal phosphate (vitamin B6) serves as a co-enzyme for the activation of ALT. Therefore, a deficiency of vitamin B6 may result in lower activity of circulating ALT in vivo (14,42). Because we used an assay with activation by pyridoxal phosphate, (vitamin B6), and because pyridoxal phosphate activation causes activation of all available ALT, it is unlikely that vitamin B6 deficiency is involved in the associations of ALT with outcomes that we observed. It may be hypoth-esized that the relatively low level of ALT that we observed is a consequence of healthy survivor bias.

This study has some limitations. First, no data of alcohol use were available. ALT levels can be elevated in the presence of alcoholic liver disease, which might have caused unmeasured confounding. However, if associations of ALT and mortality are as we report in patients with type 2 diabetes, this would have caused underestimation rather than overestimation of our results. Second, no ultrasonography or biopsies were


performed to determine whether subjects had NAFLD. Thirdly, since ALT was only measured at baseline, variation in ALT over time cannot be taken into account. The strengths of this study are the prospective design, the large number of events and the length of follow-up period.

In conclusion, our results show an inverse association of ALT with all-cause mortality and particularly with non-cardiovascular mortality in patients with type 2 dia-betes. Furthermore, there appears to be a U-shaped, bimodal relationship of ALT with all-cause mortality. The association of ALT and mortality seems different when ALT is mainly studied within the normal range rather than with inclusion of a large number of subjects that have levels of ALT above the normal range. ALT was not associated with progression of renal function in this population of patients with type 2 diabetes.

AcknowledgementsThis research was supported by GlaxoSmithKline, the Netherlands. The funding source had no further involvement in study design, data collection, or interpretation of the data. No other potential conflicts of interest were reported.

References 1. International Diabetes Federation. Available at: http://www.idf.org/worlddiabetesday/toolkit/

gp/facts-figures. Accessed April, 2013. 2. Hamaguchi M, Kojima T, Takeda N, Nakagawa T, Taniguchi H, Fujii K, et al. The metabol-

ic syndrome as a predictor of nonalcoholic fatty liver disease. Ann Intern Med 2005 Nov 15;143(10):722–728.

3. Marceau P, Biron S, Hould FS, Marceau S, Simard S, Thung SN, et al. Liver pathology and the metabolic syndrome X in severe obesity. J Clin Endocrinol Metab 1999 May;84(5):1513–1517.

4. Targher G, Bertolini L, Rodella S, Tessari R, Zenari L, Lippi G, et al. Nonalcoholic fatty liver dis-ease is independently associated with an increased incidence of cardiovascular events in type 2 diabetic patients. Diabetes Care 2007 Aug;30(8):2119–2121.

5. Targher G, Byrne CD. Clinical Review: Nonalcoholic fatty liver disease: a novel cardiometa-bolic risk factor for type 2 diabetes and its complications. J Clin Endocrinol Metab 2013 Feb;98(2):483–495.

6. Kotronen A, Yki-Jarvinen H. Fatty liver: a novel component of the metabolic syndrome. Arte-rioscler Thromb Vasc Biol 2008 Jan;28(1):27–38.

7. Adams LA, Harmsen S, St Sauver JL, Charatcharoenwitthaya P, Enders FB, Therneau T, et al. Nonalcoholic fatty liver disease increases risk of death among patients with diabetes: a commu-nity-based cohort study. Am J Gastroenterol 2010 Jul;105(7):1567–1573.

102

Cha

pter

6

8. Targher G, Pichiri I, Zoppini G, Trombetta M, Bonora E. Increased prevalence of chronic kid-ney disease in patients with Type 1 diabetes and non-alcoholic fatty liver. Diabet Med 2012 Feb;29(2):220–226.

9. Targher G, Bertolini L, Rodella S, Zoppini G, Lippi G, Day C, et al. Non-alcoholic fatty liver disease is independently associated with an increased prevalence of chronic kidney disease and proliferative/laser-treated retinopathy in type 2 diabetic patients. Diabetologia 2008 Mar;51(3):444–450.

10. Pratt DS, Kaplan MM. Evaluation of abnormal liver-enzyme results in asymptomatic patients. N Engl J Med 2000 Apr 27;342(17):1266–1271.

11. Ioannou GN, Weiss NS, Boyko EJ, Mozaffarian D, Lee SP. Elevated serum alanine aminotrans-ferase activity and calculated risk of coronary heart disease in the United States. Hepatology 2006 May;43(5):1145–1151.

12. Lee TH, Kim WR, Benson JT, Therneau TM, Melton LJ,3rd. Serum aminotransferase activity and mortality risk in a United States community. Hepatology 2008 Mar;47(3):880–887.

13. Ruhl CE, Everhart JE. The Association of Low Serum Alanine Aminotransferase Activity With Mortality in the US Population. Am J Epidemiol 2013 Sep 26;178(12):1702–11.

14. Elinav E, Ackerman Z, Maaravi Y, Ben-Dov IZ, Ein-Mor E, Stessman J. Low alanine amino-transferase activity in older people is associated with greater long-term mortality. J Am Geriatr Soc 2006 Nov;54(11):1719–1724.

15. Le Couteur DG, Blyth FM, Creasey HM, Handelsman DJ, Naganathan V, Sambrook PN, et al. The association of alanine transaminase with aging, frailty, and mortality. J Gerontol A Biol Sci Med Sci 2010 Jul;65(7):712–717.

16. Drion I, Kleefstra N, Landman GW, Alkhalaf A, Struck J, Groenier KH, et al. Plasma COOH-terminal proendothelin-1: a marker of fatal cardiovascular events, all-cause mortality, and new-onset albuminuria in type 2 diabetes? (ZODIAC-29). Diabetes Care 2012 Nov;35(11):2354–2358.

17. Ubink-Veltmaat LJ, Bilo HJ, Groenier KH, Rischen RO, Meyboom-de Jong B. Shared care with task delegation to nurses for type 2 diabetes: prospective observational study. Neth J Med 2005 Mar;63(3):103–110.

18. Lutgers HL, Gerrits EG, Graaff R, Links TP, Sluiter WJ, Gans RO, et al. Skin autofluorescence provides additional information to the UK Prospective Diabetes Study (UKPDS) risk score for the estimation of cardiovascular prognosis in type 2 diabetes mellitus. Diabetologia 2009 May;52(5):789–797.

19. Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001 Mar;56(3):M146–56.

20. Sirola J, Pitkala KH, Tilvis RS, Miettinen TA, Strandberg TE. Definition of frailty in older men according to questionnaire data (RAND-36/SF-36): The Helsinki Businessmen Study. J Nutr Health Aging 2011 Nov;15(9):783–787.

21. Cuhadar S, Koseoglu M, Atay A, Dirican A. The effect of storage time and freeze-thaw cycles on the stability of serum samples. Biochem Med (Zagreb) 2013;23(1):70–77.


22. Schumann G, Bonora R, Ceriotti F, Ferard G, Ferrero CA, Franck PF, et al. IFCC primary ref-erence procedures for the measurement of catalytic activity concentrations of enzymes at 37 degrees C. International Federation of Clinical Chemistry and Laboratory Medicine. Part 4. Reference procedure for the measurement of catalytic concentration of alanine aminotransfer-ase. Clin Chem Lab Med 2002 Jul;40(7):718–724.


24. Hosmer DW, Lemeshow S, May S. Applied Survival Analysis: Regression Modeling of Time-to-Event Data, Chapter 5. 2nd ed.: John Wiley & Sons; 2008.

25. Pencina MJ, D’Agostino RB S, D’Agostino RB,Jr, Vasan RS. Evaluating the added predictive abil-ity of a new marker: from area under the ROC curve to reclassification and beyond. Stat Med 2008 Jan 30;27(2):157–72; discussion 207–12.

26. Calori G, Lattuada G, Ragogna F, Garancini MP, Crosignani P, Villa M, et al. Fatty liver index and mortality: the Cremona study in the 15th year of follow-up. Hepatology 2011 Jul;54(1):145–152.

27. Ekstedt M, Franzen LE, Mathiesen UL, Thorelius L, Holmqvist M, Bodemar G, et al. Long-term follow-up of patients with NAFLD and elevated liver enzymes. Hepatology 2006 Oct;44(4):865–873.

28. Koehler EM, Sanna D, Hansen BE, van Rooij FJ, Heeringa J, Hofman A, et al. Serum liver enzymes are associated with all-cause mortality in an elderly population. Liver Int 2014 Feb;34(2):296–304.

29. Yun KE, Shin CY, Yoon YS, Park HS. Elevated alanine aminotransferase levels predict mortality from cardiovascular disease and diabetes in Koreans. Atherosclerosis 2009 Aug;205(2):533–537.

30. Schooling CM, Kelvin EA, Jones HE. Alanine transaminase has opposite associations with death from diabetes and ischemic heart disease in NHANES III. Ann Epidemiol 2012 Nov;22(11):789–798.

31. Hovinen SM, Pitkala KH, Tilvis RS, Strandberg TE. Alanine aminotransferase activity and mortality in older people. J Am Geriatr Soc 2010 Jul;58(7):1399–1401.

32. Ford I, Mooijaart SP, Lloyd S, Murray HM, Westendorp RG, de Craen AJ, et al. The inverse relationship between alanine aminotransferase in the normal range and adverse cardiovascular and non-cardiovascular outcomes. Int J Epidemiol 2011 Dec;40(6):1530–1538.

33. Dong MH, Bettencourt R, Brenner DA, Barrett-Connor E, Loomba R. Serum levels of alanine aminotransferase decrease with age in longitudinal analysis. Clin Gastroenterol Hepatol 2012 Mar;10(3):285–90.e1.

34. Dong MH, Bettencourt R, Barrett-Connor E, Loomba R. Alanine aminotransferase decreases with age: the Rancho Bernardo Study. PLoS One 2010 Dec 8;5(12):e14254.

35. Elinav E, Ben-Dov IZ, Ackerman E, Kiderman A, Glikberg F, Shapira Y, et al. Correlation between serum alanine aminotransferase activity and age: an inverted U curve pattern. Am J Gastroenterol 2005 Oct;100(10):2201–2204.

104

Cha

pter

6

36. Ruhl CE, Everhart JE. Elevated serum alanine aminotransferase and gamma-glutamyltransfer-ase and mortality in the United States population. Gastroenterology 2009 Feb;136(2):477–85.e11.

37. Ruhl CE, Everhart JE. Upper limits of normal for alanine aminotransferase activity in the Unit-ed States population. Hepatology 2012 Feb;55(2):447–454.

38. Matloff DS, Selinger MJ, Kaplan MM. Hepatic transaminase activity in alcoholic liver disease. Gastroenterology 1980 Jun;78(6):1389–1392.

39. Martin R, Fitzl G, Mozet C, Martin H, Welt K, Wieland E. Effect of age and hypoxia/reoxygen-ation on mRNA expression of antioxidative enzymes in rat liver and kidneys. Exp Gerontol 2002 Dec;37(12):1481–1487.

40. Castro Mdel R, Suarez E, Kraiselburd E, Isidro A, Paz J, Ferder L, et al. Aging increases mito-chondrial DNA damage and oxidative stress in liver of rhesus monkeys. Exp Gerontol 2012 Jan;47(1):29–37.

41. Kim WR, Flamm SL, Di Bisceglie AM, Bodenheimer HC, Public Policy Committee of the American Association for the Study of Liver Disease. Serum activity of alanine aminotransfer-ase (ALT) as an indicator of health and disease. Hepatology 2008 Apr;47(4):1363–1370.

42. Ono K, Ono T, Matsumata T. The pathogenesis of decreased aspartate aminotransferase and alanine aminotransferase activity in the plasma of hemodialysis patients: the role of vitamin B6 deficiency. Clin Nephrol 1995 Jun;43(6):405–408.

7

Uncovering of body mass

index as a risk factor

for poor long‑term outcome

after renal transplantation7Petronella E. Deetman1, Jan-Stephan F. Sanders1, Marc A.J. Seelen1, Reinold O.B. Gans1, Gerjan Navis1, and Stephan J.L. Bakker1,2

1 Department of Medicine, Division of Nephrology, University Medical Center Groningen and University of Groningen, the Netherlands;

2 Top Institute Food & Nutrition, Wageningen, the Netherlands

Transplantation 2015;99(1):e5–6, accepted in abbreviated form

106

Cha

pter

7

AbstractBackground. A high BMI is a risk factor for neither graft failure nor mortality in renal transplant recipients (RTR). BMI is not only determined by fat mass, but also by muscle mass. It is unknown whether adjustment for muscle mass could uncover associations of BMI with mortality, by making BMI a better indicator of fat mass. The aim of this study was to investigate whether muscle mass, determined by 24hr urinary creatinine excre-tion (UCE), confounds associations of BMI with graft failure and mortality in RTR.Methods. RTR were included who were transplanted between 1993 and 2008. Baseline BMI was determined at one year after transplantation. UCE was measured in repeated 24hr urine collections gathered between 6 and 18 months after transplantation. Results. In total, 916 RTR were included (mean age 50 [39–59] and 58% was male). Mean BMI was 26 ± 4 kg/m2 and UCE 12.3 ± 3.0 mmol/24h. During 4.6 (2.5–7.9) years of follow-up for graft failure and 4.9 (2.7–8.4) years for mortality, 77 (8%) RTR developed graft failure and 153 (17%) RTR died. The age- and sex-adjusted association of BMI with graft failure was nonsignificant, (Hazard Ratio [HR], 1.20 [0.96–1.50]; P = 0.10), but became uncovered by adjustment for UCE (HR, 1.44 [1.14–1.82]; P = 0.003). Simi-larly, the age- and sex-adjusted association of BMI with mortality was nonsignifi-cant (HR, 1.06 [0.90–1.25]; P = 0.50), but became uncovered by adjustment for UCE (HR, .35 [1.13–1.62]; P = 0.001). Conclusions. Adjustment for UCE uncovers an adverse association of high fat mass, for which BMI became a better measure after adjustment for UCE, with both graft failure and mortality. These findings provide an additional potential explanation for the absence of increased risk observed in previous studies.

107Body mass index as a risk factor after renal transplantation

IntroductionThe short-term prognosis of renal transplant recipients (RTR) has improved significant-ly (1). However, long-term outcome beyond one year after transplantation is disappoint-ing, with success rates of only 50% at approximately 10 years after transplantation (2). One of the reasons for poor outcomes in RTR is the high prevalence of obesity, predis-posing patients to dyslipidemia, hypertension, and insulin resistance (3–5).

However, in RTR, obesity (i.e. body mass index [BMI] ≥ 30 kg/m2) is a risk fac-tor for neither graft failure nor mortality (6,7,8). It has been hypothesized that advances in immunosuppressive therapy, better control of obesity-related comorbidities and in-creasing experience in kidney transplantation underlie absence of increased risk associ-ated with obesity (6). However, it is important to realize that BMI is not only determined by fat mass, but also by other constituents, particularly muscle mass. Muscle mass could confound the association of BMI with long-term outcomes, because it has associations with long-term outcome opposite of those hypothesized for BMI. This is not only true for renal transplant recipients (9), but also for other populations (10,11).

To our knowledge, it is unknown whether adjustment for muscle mass could uncover associations of BMI with graft failure and mortality, by making BMI a better indicator of fat mass in RTR. To this end, we investigated whether muscle mass, deter-mined by 24 hr urinary creatinine excretion (UCE), confounds associations of BMI with graft failure and mortality in RTR. Because low muscle mass may be the consequence of low protein intake and because low protein intake has been associated with a high risk of graft failure and mortality in RTR (12), we also questioned whether protein intake might also confound associations of BMI with graft failure and mortality. Therefore, our sec-ondary aim was to investigate the effect of urinary urea excretion, as a marker of protein intake, on the association of BMI with graft failure and mortality.



RTR were eligible for this study if they received a renal transplant for the first time be-tween January 1993 and February 2008 in the University Medical Center Groningen in the Netherlands. Details of the cohort have been published previously (12,13). We made use of the regular patient care program for outpatients. For each visit, RTR are re-quested to collect a 24 hr urine sample in which several parameters are routinely assessed. Data of total cholesterol, serum creatinine and 24 hr urinary excretion of urea, creatinine

108

Cha

pter

7

and protein, assessed between 6 and 18 months, were extracted from our hospital labora-tory system (12). The median of these measurements was used for analyses. Patient char-acteristics were retrieved from medical records. BMI was defined as weight (kg) by height (m) squared. Delayed graft function (DGF) was defined as oliguria for more than 6 days after transplantation. The estimated glomerular filtration rate (eGFR) was calculated us-ing the Chronic Kidney Disease Epidemiology Collaboration equation (14).

Shortly after transplantation, RTR are often not yet stable and infections and re-jections play a major role. Therefore, data of the first 6 months after transplantation were discarded. Furthermore, we only included patients who survived with a function-ing graft beyond 6 months after transplantation and beyond the period of baseline data collection. Finally, patients with missing data on baseline BMI were excluded (n = 33, 3.5%), leaving 916 patients eligible for analyses. The Institutional Review Board of the University Medical Center Groningen approved the study protocol, which adhered to the declaration of Helsinki. According to Dutch law, general consent for organ donation and transplantation includes consent for research projects.

Laboratory measurements

Nonfasting blood and urine samples were directly analyzed after collection. Measure-ments were performed on a Merck Mega Analyzer before March 2006 (Merck, Darm-stadt, Germany). From March 2006 onward, measurements were performed on a Roche Modular (Roche Ltd., Mannheim, Germany).

Because there was no substantial difference in UCE and total cholesterol levels, these results were not converted. There was a substantial difference in serum creatinine and urinary protein; therefore, results were converted according to conversion equations provided by our laboratory (Supplementary Table S1). Before March 2006, UCE was measured with the Jaffé method, from March 2006 onwards, UCE was measured with the enzymatic colorimetric assay. UCE was based on a median (interquartile range) of 6 (5–8) collections of 24 hr urine per patient.

Definitions and clinical end points

The primary endpoint of this study was all-cause mortality. The secondary endpoint was graft failure. Graft failure was defined as the need for dialysis or retransplantation. Follow-up was recorded until October 2009. Person-time of follow-up was computed for each participant from 1.5 years after transplantation until the incidence of death or graft failure or after 10 years of follow-up. There was no loss to follow-up.



Baseline characteristics of the study population were calculated according to categories of BMI. According to WHO definitions (15), BMI is classified into the following categories; less than 18.5 kg/m2, 18.5 to 24.9 kg/m2, 25 to 30 kg/m2, and 30 kg/m2 or higher. Because only 14 (1.9%) RTR had a BMI less than 18.5 kg/m2, we subdivided RTR into the follow-ing BMI categories; less than 25 kg/m2, 25 to 30 kg/m2, and 30 kg/m2 or higher. Normally distributed variables are presented as mean ± standard deviation (SD), skewed distributed variables are presented as median (interquartile range), and categorical variables are given as number (percentage). Differences in baseline characteristics across categories of BMI were determined with linear regression. Variables were log-transformed when appropriate.

The effect of potential confounders on the associations of graft failure and mor-tality was assessed in multivariable Cox regression models. Adjustment for UCE was performed in these models to investigate a potential uncovering effect of UCE on asso-ciations of BMI with graft failure and mortality. The final multivariable-adjusted model included age, sex, UCE, diabetes, systolic blood pressure, use of antihypertensive drugs, cholesterol, use of lipid lowering drugs, smoking status, and eGFR. Because several sub-jects had missing values for one or more variables, multiple imputation was performed for missing data with five imputation cycles. Imputed data was used for Cox regression analyses only. The assumptions of proportionality of hazards and linearity were met.

As a secondary analysis, we tested whether adjustment for urinary urea excretion (UUE), as a marker of protein intake, uncovered associations of BMI with graft failure and mortality. Furthermore, to study whether the uncovering effect of UCE on outcomes was independent of UUE, UUE was added to the multivariable Cox regression models with UCE. As a further secondary analysis, Cox regression analyses were repeated with body weight rather than BMI. In addition, because a BMI lower than 18.5 kg/m2 may be a sign of underlying chronic disease, which may increase the risk of mortality (16), Cox regression analyses were repeated after exclusion of RTR with a BMI less than 18.5 kg/m2. To investigate whether transplant era might have influenced our results, Cox re-gression analyses were repeated after exclusion of RTR who were transplanted before 2000. Statistical analyses were performed using SPSS (version 20.0, IBM Inc. Chicago, IL, USA) and STATA (version 11.0, StataCorp LP, College Station, Texas, USA). A P‑value of less than 0.05 was considered statistically significant.

110

Cha

pter

7

Results

Patient characteristics

A total of 916 RTR were included in the present study. Median age was 50 (interquartile range 39–59) years, 58% was male. Mean BMI was 26 ± 4 kg/m2, eGFR was 54 ± 18 mL per min per 1.73 m2 and UCE was 12.3 ± 3.0 mmol/24h. Baseline characteristics are shown in Table 1, for all RTR and according to categories of BMI. BMI was positively associated with age, the use of antihypertensive medication, cholesterol, UCE, UUE, and warm ischemia time. BMI was inversely associated with the eGFR. No associations were found of BMI with sex, diabetes, systolic blood pressure, use of lipid lowering drugs, uri-nary protein excretion, dialysis vintage, living donor, cold ischemia time, delayed graft function, acute rejection, use of proliferation inhibitors, calcineurin inhibitors, and the use of mTOR inhibitors.

UCE and graft failure

A total of 77 (8%) RTR experienced graft failure during 4.6 (2.5–7.9) years of follow-up. Hazard ratios are shown for the associations of BMI with graft failure in Table 2. In a univariable analysis, BMI was not associated with graft failure. The association of BMI with graft failure remained nonsignificant after adjustment for age and sex, but became uncovered by further adjustment for UCE (Figure 1). This uncovering effect remained after further adjustment for systolic blood pressure, use of antihypertensive drugs, cholesterol, use of lipid lowering drugs, diabetes, smoking status, and eGFR.

UCE and all-cause mortality

A total of 153 (17%) RTR died during 4.9 (2.7–8.4) years of follow-up. Hazard ratio’s are shown in table 3. In a univariable Cox regression analysis, BMI tended to be positively associated with mortality. This trend disappeared after adjustment for age and sex, but became uncovered by adjustment for UCE (Figure 1). This uncovering effect remained after further adjustment for potential confounders.


Table 1. Baseline characteristics of the study population and according to categories of BMI.

Study population BMI (kg/m2)

< 25 25–30 > 30 P

n = 916 n = 362 n = 382 n = 172

Men, n 531 (58) 210 (58) 221 (58) 100 (58) 0.96

Age, yr 50 (39–59) 46 (35–57) 51 (41–60) 52 (44–60) < 0.001

Weight, kg 80 ± 15 70 ± 10 82 ± 9 99 ± 14 <0.001

Diabetes, n (%) 70 (10) 21 (8) 36 (9) 13 (8) 0.28

SBP, mm Hg 144 ± 17 142 ± 17 143 ± 18 144 ± 17 0.058

Use ofantihypertensives, n (%)

572 (87) 225 (83) 230 (88) 117 (94) 0.002

Cholesterol, mmol/L 5.7 ± 1.1 5.6 ± 1.1 5.7 ± 1.1 5.9 ± 1.1 < 0.001

Lipid lowering drugs, n (%)

237 (36) 89 (33) 96 (37) 52 (43) 0.14

eGFR, mL/min/1.73 m2 54 ± 18 56 ± 19 54 ± 17 52 ± 18 0.010

Urinary protein, g/24h 0.3 (0.1–0.3) 0.2 (0.1–0.3) 0.3 (0.1–0.3) 0.4 (0.1–0.4) 0.10

UCE, mmol/24h 12.3 ± 3.0 11.5 ± 2.8 12.5 ± 3.0 13.2 ± 3.1 < 0.001

UUE, mmol/24h 380 ± 98 351 ± 93 385 ± 93 428 ± 98 < 0.001

Dialysis vintage, mo 40 (18–57) 37 (20–57) 37 (17–60) 37 (18–53) 0.26

Living donor, n (%) 239 (26) 103 (29) 93 (24) 43 (25) 0.41

Warm ischemia time, min

40 (32–49) 38 (32–46) 40 (32–49) 41 (35–53) 0.002

Cold ischemia time, hr 17 (3–23) 16 (3–23) 18 (9–23) 17 (3–22) 0.86

Delayed graft function, n (%)

242 (26) 91 (25) 97 (25) 54 (31) 0.050

Acute rejection, n (%) 308 (34) 127 (35) 123 (32) 58 (34) 0.55

Use of PI, n (%) 726 (79) 287 (79) 292 (76) 147 (86) 0.29

Use of CI, n (%) 823 (90) 316 (87) 347 (91) 160 (93) 0.14

Use of mTOR, n (%) 36 (4) 16 (4) 16 (4) 4 (2) 0.15

PI, proliferation inhibitor; CI, calcineurin inhibitor; mTOR, mammalian target of rapamycin

112

Cha

pter

7

Table 2. Associations of body mass index with graft failure and mortality.

Graft Failure Mortality

HR (95% CI) P HR (95% CI) P

No. events 77 153

Model 1 1.09 (0.87–1.36) 0.45 1.16 (0.99–1.35) 0.067

Model 2 1.20 (0.96–1.50) 0.10 1.06 (0.90–1.25) 0.50

Model 3 1.44 (1.14–1.82) 0.003 1.35 (1.13–1.62) 0.001

Model 4 1.44 (1.09–1.91) 0.012 1.35 (1.10–1.66) 0.004

HRs are shown per each SD increase in BMI (4.3 kg/m2)

Model 1: Crude analysis with body mass index

Model 2: Model 1+age and sex

Model 3: Model 2+UCE

Model 4: Model 3+ systolic blood pressure, use of antihypertensive drugs, cholesterol, use of lipid lowering drugs,

diabetes, smoking status, and eGFR

HRs, hazard ratios; BMI, body mass index; UCE, urinary creatinine excretion; UUE, urinary urea excretion

In a secondary analysis, we tested whether UUE, similar to UCE, uncovered the as-sociations of BMI with graft failure and mortality. Indeed, there was an uncovering effect of UUE, albeit less pronounced than UCE (Table 3, model 1 and 2). When the Cox regression analyses with UCE were additionally adjusted for UUE, the uncovering effect of UCE remained essentially similar (Table 3, model 3 and model 4). In a further secondary analysis with body weight rather than BMI, we found that adjustment for UCE also uncovered the associations of body weight with graft failure and mortality (Table 4). In addition, results of Cox regression analyses remained essentially similar after exclusion of 14 RTR with a BMI lower than 18.5 kg/m2 as secondary analyses and after exclusion of RTR who were transplanted before 2000.


Table 3. Associations of body mass index with graft failure and mortality (uncovering effect of creatinine excretion).


HR (95% CI) P HR (95% CI) P

No. events 77 153

Model 1 BMI 1.44 (1.14–1.82) 0.003 1.35 (1.13–1.62) 0.001

UCE 0.55 (0.40–0.77) <0.001 0.49 (0.38–0.63) <0.001

Model 2 BMI 1.42 (1.11–1.82) 0.005 1.26 (1.05–1.51) 0.015

UUE 0.62 (0.46–0.84) 0.002 0.63 (0.51–0.78) <0.001

Model 3 BMI 1.51 (1.18–1.93) 0.001 1.38 (1.15–1.66) 0.001

UCE 0.65 (0.44–0.97) 0.035 0.55 (0.40–0.74) <0.001

UUE 0.77 (0.54–1.09) 0.14 0.85 (0.66–1.10) 0.21

Model 4 BMI 1.44 (1.04–2.00) 0.029 1.37 (1.11–1.68) 0.003

UCE 0.60 (0.40–0.90) 0.013 0.56 (0.39–0.80) 0.002

UUE 1.00 (0.62–1.60) 0.98 0.93 (0.67–1.30) 0.67

HRs are shown per each SD increase in BMI (4.3 kg/m2), UCE (3.0 mmol/24h) or UUE (97.8 mmol/24h)

Model 1: BMI, age, sex and UCE

Model 2: BMI, age, sex and UUE

Model 3: BMI, age, sex, UUE and UCE

Model 4: BMI, age, sex, UUE, UCE, systolic blood pressure, use of antihypertensive drugs, cholesterol, use of lipid

lowering drugs, diabetes, smoking status, and eGFR

BMI, body mass index; UCE, urinary creatinine excretion; UUE, urinary urea excretion

Secondary analyses

114

Cha

pter

7

Table 4. Associations of body weight with graft failure and mortality, and the uncovering effect of urinary creatinine excretion.


HR (95% CI) P HR (95% CI) P

No. events 77 153

Model 1 1.11 (0.89–1.39) 0.37 1.07 (0.91–1.25) 0.45

Model 2 1.25 (0.97–1.60) 0.084 1.06 (0.87–1.29) 0.57

Model 3 1.58 (1.21–2.07) 0.001 1.46 (1.18–1.81) 0.001

Model 4 1.53 (1.12–2.09) 0.008 1.45 (1.14–1.84) 0.002

HRs are shown per each SD increase in body weight (15 kg)

Model 1: Body weight

Model 2: Model 1+age and sex, and body length

Model 3: Model 2+UCE

Model 4: Model 3+ systolic blood pressure, use of antihypertensive drugs, cholesterol, use of lipid lowering drugs,

diabetes, smoking status, and eGFR

HRs, hazard ratios; BMI, body mass index; UCE, urinary creatinine excretion; UUE, urinary urea excretion

Figure 1. Uncovering of the association of BMI with mortality and graft failure by adjustment for urinary creatinine excretion. A: the age-and sex-adjusted association of BMI with graft failure; B: additional adjustment for urinary creatinine excretion; C: the age-and sex-adjusted association of BMI with mortality; D: additional adjustment for urinary creatinine excretion.


DiscussionIn this prospective analysis of a large cohort of RTR, we found a nonsignificant associa-tion of BMI with graft failure and mortality in age- and sex-adjusted analyses, which became uncovered after adjustment for muscle mass, which was determined by UCE. Furthermore, we showed a similar uncovering effect of UCE on the association of body weight with graft failure and mortality. The uncovering effect of UCE appeared indepen-dent of other potential confounders.

A high BMI (i.e. above 30 kg/m2) has been associated with dyslipidemia, hyperten-sion, and insulin resistance in the general population (3–5). A high BMI was, therefore, suggested as a risk factor for poor outcomes in various chronic diseases. However, in pa-tients with end-stage renal disease, studies consequently reported paradoxical inverse as-sociations of BMI with mortality (17,18). Similar associations were found in patients with coronary heart disease (19). In renal transplant recipients, both significant and nonsig-nificant associations of BMI with graft failure and mortality have been reported. A recent systematic review and meta-analysis by Nicoletto et al. concluded that BMI was not sig-nificantly associated with graft failure and mortality (6). Nicoletto et al. also observed that the direction of associations of BMI with graft failure and mortality was highly dependent on the transplant era (6). An inverse association was found if studies included RTR were transplanted before 2000. If RTR were transplanted after 2000, nonsignificant associations were found. Our findings are in line with the findings of Nicoletto et al. In the present study that mostly included RTR who were transplanted from 2000 and later, we also found no significant associations of BMI with graft failure and mortality. After exclusion of sub-jects who were transplanted before 2000, results were essentially similar.

To our knowledge, we are the first who reported an uncovering effect by UCE on associations of BMI with graft failure and mortality. Somewhat analogous to our find-ings, Kovesdy et al. have found that a nonsignificant association of BMI with mortality became protective after adjustment for waist circumference in RTR (20). It was postulat-ed that this uncovering effect could be explained by the beneficial effects of high muscle mass for which BMI became a better measure after adjustment for waist circumference as a measure of body fat distribution (20). Again analogous to our result, Beddhu et al. investigated patients with end-stage renal disease (ESRD) and reported that the protec-tive effect of a high BMI, as commonly observed in patients with ESRD, was absent in patients with low muscle mass (21).

To date, there was no explanation for the lack of associations of BMI with graft failure and mortality in RTR. It has been suggested that advances in therapeutic regi-mens in RTR have led to better control of obesity-related comorbidities (6). We speculate that because both high muscle and fat mass contribute to BMI, with opposite associa-tions with outcome (9,20,22), BMI alone is an imperfect estimate of adiposity. It may be possible that BMI becomes a better measure for adiposity after adjusting for UCE as a marker of muscle mass. Another finding of the present study was that UUE uncovered

116

Cha

pter

7

the associations of BMI with outcomes, which was slightly less pronounced than the un-covering effect of UCE. Hence, we conclude that protein intake may also be an important confounder in the associations of BMI with outcomes, but not as strong as muscle mass.

A limitation of this study is that UCE was measured in 24 hr collections of urine, as 24 hr urine collection is prone to collection errors. However, UCE was measured mul-tiple times over a 6-month period in RTR, which reduces the possibility of collection or measurement errors. Other methods to investigate muscle mass include computerized to-mography (CT), magnetic resonance imaging (MRI), and bioelectrical impedance analysis (23). These methods have been proven expensive and in some cases inaccurate, especially in patients with dehydration or edema (23), but also because fat infiltration in muscle may cause overestimation of muscle mass (23,24). A strength of this study is that a large number of stable RTR were included. Another strength of our study is that there was no loss to follow-up.

In conclusion, adjustment for muscle mass, as determined by UCE, seems to un-cover an adverse association of high fat mass, for which BMI became a better measure after adjustment for UCE, with both graft failure and mortality. These findings may provide new insight in what we now know as the obesity paradox and may provide an additional potential explanation for the absence of increased risk associated with obesity as previously observed.

References 1. Merville P. Combating chronic renal allograft dysfunction : optimal immunosuppressive regi-

mens. Drugs 2005;65(5):615–631. 2. Lamb KE, Lodhi S, Meier-Kriesche HU. Long-term renal allograft survival in the United States:

a critical reappraisal. Am J Transplant 2011 Mar;11(3):450–462. 3. Wilson PW, D’Agostino RB, Sullivan L, Parise H, Kannel WB. Overweight and obesity as de-

terminants of cardiovascular risk: the Framingham experience. Arch Intern Med 2002 Sep 9;162(16):1867–1872.

4. Steiger U, Lippuner K, Jensen EX, Montandon A, Jaeger P, Horber FF. Body composition and fuel metabolism after kidney grafting. Eur J Clin Invest 1995 Nov;25(11):809–816.

5. van den Ham EC, Kooman JP, Christiaans MH, Nieman FH, van Hooff JP. Weight changes after renal transplantation: a comparison between patients on 5-mg maintenance steroid therapy and those on steroid-free immunosuppressive therapy. Transpl Int 2003 May;16(5):300–306.


7. Johnson DW, Isbel NM, Brown AM, Kay TD, Franzen K, Hawley CM, et al. The effect of obesity on renal transplant outcomes. Transplantation 2002 Sep 15;74(5):675–681.

8. Massarweh NN, Clayton JL, Mangum CA, Florman SS, Slakey DP. High body mass in-dex and short- and long-term renal allograft survival in adults. Transplantation 2005 Nov 27;80(10):1430–1434.





12. Deetman PE, Said MY, Kromhout D, Dullaart RP, Kootstra-Ros JE, Sanders JS, et al. Urinary Urea Excretion and Long-Term Outcome After Renal Transplantation. Transplantation 2014 Nov 12.

13. Reznichenko A, Snieder H, van den Born J, de Borst MH, Damman J, van Dijk MC, et al. CUBN as a novel locus for end-stage renal disease: insights from renal transplantation. PLoS One 2012;7(5):e36512.


15. World Health Organization. Global database on Body Mass Index. Available at: http://apps.who.int/bmi/index.jsp?introPage = intro.html.

16. Physical status: the use and interpretation of anthropometry. Report of a WHO Expert Com-mittee. World Health Organ Tech Rep Ser 1995;854:1–452.

17. Kalantar-Zadeh K, Abbott KC, Salahudeen AK, Kilpatrick RD, Horwich TB. Survival advan-tages of obesity in dialysis patients. Am J Clin Nutr 2005 Mar;81(3):543–554.



20. Kovesdy CP, Czira ME, Rudas A, Ujszaszi A, Rosivall L, Novak M, et al. Body mass index, waist circumference and mortality in kidney transplant recipients. Am J Transplant 2010 Dec;10(12):2644–2651.

21. Beddhu S, Pappas LM, Ramkumar N, Samore M. Effects of body size and body composition on survival in hemodialysis patients. J Am Soc Nephrol 2003 Sep;14(9):2366–2372.

22. Alam A, Molnar MZ, Czira ME, Rudas A, Ujszaszi A, Kalantar-Zadeh K, et al. Serum adi-ponectin levels and mortality after kidney transplantation. Clin J Am Soc Nephrol 2013 Mar;8(3):460–467.

23. Al-Gindan YY, Hankey CR, Leslie W, Govan L, Lean ME. Predicting muscle mass from anthro-pometry using magnetic resonance imaging as reference: a systematic review. Nutr Rev 2014 Feb;72(2):113–126.

24. Heymsfield SB, Olafson RP, Kutner MH, Nixon DW. A radiographic method of quantifying protein-calorie undernutrition. Am J Clin Nutr 1979 Mar;32(3):693–702.

118

Cha

pter

7

Supplementary Table S1. Conversion equations for results of laboratory measurements performed on Merck Mega Analyzer or Roche Modular.

Measurement Conversion equation

Urinary creatinine No difference

Serum creatinine Ya = (Xb‑8)/1.07

Total cholesterol No difference

Urinary protein Ya = (Xb+0.05)/1.403

Urinary urea Ya = (Xb‑9)/0.996

aRoche Modular (Roche Ltd., Mannheim, Germany)bMerck Mega Analyzer (Merck, Darmstadt, Germany)

8

General discussion

and future perspectives8

121General discussion and future perspectives

PreludeAn interaction between the kidney and the liver has mainly received attention in the context of acute liver failure, leading to renal failure in which kidneys fail subsequential-ly. A possible role of the liver in chronic kidney disease (CKD) has hardly been subject of investigation. There are, however, data that suggest that an interaction between the liver and the kidney, being the two major metabolic organs of the body, may be relevant to CKD as well. Therefore, the aim of the present thesis was to investigate aspects of liver function, in the non-failing liver, for their relevance in CKD and its complications.

Several aspects of liver function were studied. There are several traditional mark-ers of liver function which might be interesting from the perspective of CKD. One of those is bilirubin, of which it has been suggested that it has a renoprotective effect. In this thesis we studied whether bilirubin protects against progression of CKD and whether the concentration of bilirubin can be elevated by lifestyle. Another aspect of liver func-tion that we have studied is protein metabolism. The liver and the kidney are intrinsically linked in protein metabolism. One of the important roles of the liver in protein metabo-lism is breakdown of amino acids to urea. The kidney subsequently facilitates the excre-tion of this breakdown product in urine. The amount of urea excreted in urine provides an estimate of protein intake from food. We explored whether protein intake assessed from 24 hr urine excretion influences long-term outcomes after renal transplantation. Another aspect where liver function and kidney function come together is where they are both adversely influenced by obesity. In this field, it is important to realize that the measure of body mass index (BMI), which is of the used to define obesity is composed of both fat mass and other bodily masses, including muscle mass. Use of BMI as a marker of high fat mass therefore has its limitations. We sought to investigate the utility of a liver-derived marker of excess fat mass and a breakdown product of muscles which is excreted in urine as measures to better relate obesity to long-term outcome, including CKD.

Bilirubin and protection against chronic kidney diseaseDiabetic nephropathy (DN) develops in approximately 20–40% of patients with diabe-tes and is the single leading cause of end-stage renal disease (ESRD) around the world (1). An important pathophysiological mechanism that has been identified in the devel-opment and progression of diabetic nephropathy is oxidative stress (2–4). Bilirubin is known to be a potent endogenous antioxidant (5) and accumulating evidence has shown that bilirubin has a protective effect against kidney disease (5–7).

Based on its anti-oxidant properties, we hypothesized that bilirubin is inversely associated with progression of diabetic nephropathy. To investigate this hypothesis, we tested the association of bilirubin with progression of diabetic nephropathy in a co-hort of 1,498 patients with type 2 diabetes, hypertension, and nephropathy (chapter 2).

122

Cha

pter

8

In this cohort, bilirubin was inversely and independently associated with progression of diabetic nephropathy. This supports a role for the anti-oxidant effect of bilirubin, exert-ing a mild protective effect on the kidney.

Besides in DN, oxidative stress has also been implicated in the progression of chronic allograft damage after renal transplantation (8–11). We therefore hypothesized that bilirubin protects against decline of renal function and late graft failure in renal transplant recipients (RTR) (chapter 3). To this end, we prospectively investigated 603 RTR with a functioning graft for more than one year. We found that the decline in renal function, determined by a decline in creatinine clearance, was least pronounced in RTR with highest bilirubin concentrations. We also found an independent inverse association of bilirubin with graft failure. Bilirubin was not associated with mortality.

Bilirubin: a causal association with chronic kidney disease?Given the observational design of these studies, it remains unknown whether bilirubin truly protects against renal disease or whether bilirubin is merely an indicator of a fa-vourable renal risk profile. Johansen et al. already pointed out that such a distinction is important: if bilirubin is a correlated biomarker, it might serve as a prognostic or diag-nostic indicator for renal outcomes such as CKD, diabetic nephropathy or long-term graft failure (12). However, if bilirubin is causally related, it could serve as a potential therapeutic target against these diseases (12).

A method to estimate causal relationships is Mendelian randomization (13,14). Mendelian randomization is a statistical method that is based on the fact that genetic variability is less likely to be influenced by lifestyle or exposure to environmental influ-ences compared to the phenotype (14). Mendelian randomization is increasingly used to limit biases such as (unmeasured) confounding and reverse causality that are often present in observational studies (14). Therefore, Mendelian randomization can be used as a tool to explore the possible causality of observed associations.

As much as 18% of the total variability in bilirubin is explained by the genetic variability in the gene UGT1A1. UGT1A1 encodes for the enzyme UDP-GT which is responsible for the regulation of bilirubin levels. Using Mendelian randomization, genetically elevated levels of bilirubin were associated with a lower incidence of type 2 diabetes in the general population (15). In contrast, Mendelian randomization stud-ies rendered disappointing results for the association of bilirubin with cardiovascular disease. A genetically elevated level of bilirubin was neither associated with risk of ischemic heart disease in a prospective study of 10,264 individuals nor in a case control study of 5,133 cases and 5,133 healthy individuals (16). In line, a Mendelian random-ization study of 868 healthy individuals found no association of genetically elevated levels of bilirubin with cardiovascular risk factors (17). These findings are in contrast to a number of observational studies which have demonstrated an inverse association


of circulating bilirubin with cardiovascular disease (18–20). To our knowledge, to date, no Mendelian randomization study has been performed to investigate the effect of bilirubin on renal outcomes.

Elevation of bilirubinAssuming that variation in the circulating concentration of bilirubin is relevant to renal outcomes, we investigated the determinants of variability in the normal range (chapter 4), with emphasis on dietary factors and lifestyle factors such as drinking wine and smoking. To this end, a cross-sectional study was performed in 509 male subjects without major chronic diseases from the Zutphen elderly study aged 64–85 years. We investigated associ-ations of bilirubin with individual food groups, a healthy dietary pattern, and lifestyle fac-tors such as smoking. The dietary pattern was studied with the Mediterranean Diet Score (MDS). Both adherence to the MDS with more than 4 points and wine consumption were significantly associated with higher serum bilirubin concentrations in uni- and multivari-able analyses. No data were available on oxidative stress in this population, it therefore remains therefore unknown whether this could have beneficial effects on oxidative stress.

Bilirubin: a correlated biomarker?It is possible that there is no causal association of bilirubin with renal outcomes and that bilirubin is no more than a correlated biomarker. If so, the association of bilirubin with chronic kidney disease could be mediated by other factors, e.g. by metabolites that are gen-erated during bilirubin production. These metabolites include carbon monoxide and bili-verdin (5,21). Like bilirubin, carbon monoxide and biliverdin have been suggested to have a beneficial effect on the kidney. Administration of both carbon monoxide and biliverdin has been shown to attenuate ischemia reperfusion injury in rats (22). Administration of carbon monoxide alone has also been shown to inhibit chronic transplant dysfunction in rats (23) and protect against renal fibrosis in mice (24). Furthermore, administration of biliverdin has been shown to protect against diabetic nephropathy in mice (6).

A limitation of our studies is that only total bilirubin was measured. Total biliru-bin is composed of two forms, conjugated (direct) bilirubin and unconjugated (indirect) bilirubin. It is important to distinguish between conjugated and unconjugated bilirubin. This is important because the conjugation of bilirubin may significantly alter its charac-teristics and thereby its influence on biological processes. It is also important because the amount of conjugated bilirubin may reflect the functionality of the liver to conjugate other metabolites. Such other metabolites include products from cellular catabolism, drugs, and vitamin D (25,26). Increased conjugation by the liver may cause unwant-ed excretion of substances with implied beneficial effects. In other words, if bilirubin

124

Cha

pter

8

conjugation is impaired, the concentration of bilirubin may be higher than normal, but also the concentration of other metabolites may be higher. A logical next step, therefore, would be to investigate whether it is conjugated or unconjugated bilirubin that is associ-ated with a more favorable renal risk profile.

UreaThe liver and the kidney are intrinsically linked in protein metabolism. Urea is synthe-sized by the liver from the breakdown of amino acids, as present in protein. Because the kidney facilitates the excretion of urea, circulating urea is used as a marker of kidney function. The urinary excretion of urea can be used a marker of protein intake.

Optimizing protein intake is an important component of dietary management of chronic kidney disease (CKD) patients. Because high protein intake is supposed to aggravate proteinuria (1–4), guidelines recommend daily allowances of 0.6 to 0.8 g pro-tein/kg per day in patients with CKD stages 1 to 4 (3,5,6). Once patients are on dialy-sis, protein catabolism and protein losses must be compensated. Guidelines therefore recommend a minimal protein intake of 1.1 g/kg per day and preferably 1.2 to 1.3 g/kg per day in patients receiving dialysis (27–29). Unfortunately, advisable protein intake is unknown after transplantation. Therefore, we studied whether 24-h urinary urea excre-tion (UUE), as a marker of protein intake, was associated with graft failure and mortality in a cohort of 940 RTR with a functioning graft for more than one year (chapter 5). We made use of the regular patient care program for outpatients. Data on repeated 24 hr urine collections was collected from our hospital laboratory system. In this cohort of RTR, optimal protein intake was higher than what is currently recommended in CKD patients (27,30,31), but comparable to recommended protein intake in patients treated with dialysis (≥ 1.1 g/kg/day) (27–29). UUE was not associated with graft failure in the overall population, but was inversely associated with graft failure in RTR with a BMI less than 25 kg/m2, and in RTR with an eGFR of 45 mL per min per 1.73 m2 or higher, independent of potential confounders. UUE was inversely associated with mortality, in-dependent of BMI and eGFR. From these data, we conclude that a relatively high protein intake may be beneficial for long-term outcomes in RTR.

After renal transplantation, effects of the renal allograft, rejection episodes, and immunosuppressive drugs contribute to high protein and energy expenditure in RTR (32), possibly shifting optimal protein intake upward. Furthermore, a major side effect of chronic renal disease is anorexia with concordant low protein intake (33,34). When nutritional demands are not balanced by intake, protein and energy stores are utilized. In some patients, this utilization of energy stores can progress into the ‘protein energy wasting syndrome’. The protein energy wasting syndrome is a generic term for malnu-trition with inadequate protein intake, and is a risk factor for inflammation, oxidative stress, cardiovascular disease, and mortality (35–38). We hypothesized that, at some


point, the adverse effects of a high protein intake (i.e. renal function deterioration) may counterbalance the beneficial effects of high protein intake (i.e. tissue repair, more mus-cle mass, better nutritional status) in RTR.

Given the observational nature of our study, it remains to be established whether high consumption of protein indeed protects against graft failure and mortality in RTR. In addition, we did not discern between sources of protein (i.e. animal or plant protein), which may have opposite associations with outcome. Markers to determine the intake of meat protein are urinary carnosine, 1-methylhistidine and 3-methylhistidine (39). Therefore, it would be of interest to investigate associations of the source of protein (as measured by carnosine, 1-methylhistidine, and 3-methylhistidine) with renal outcomes in future studies.

Obesity, the liver, and the kidneyThe kidney and the liver may also be parallel victims of adverse factors to which they are both exposed. Obesity is a common cause of both liver and kidney damage. However, the definition of obesity is a controversial issue. Several studies have surprisingly report-ed robust inverse associations of BMI with mortality in chronically ill individuals. This phenomenon has been referred to as the obesity paradox (40–44). Several explanations for this paradoxical association have been proposed. Survival bias has been suggested. Another potential explanation is that patients with a low BMI are at high risk for the protein energy wasting syndrome. As previously mentioned, the protein energy wasting syndrome is disadvantageous, and the long term effects of adiposity (i.e. cardiovascular morbidity and mortality) may not be relevant to patients with a short life expectancy. We propose that a high BMI can indeed reflect a high fat mass, but it can also reflect a high muscle mass, or a combination of both. It may be because of this notion that studies have reported inverse associations of BMI with mortality, and absent or even inverse associa-tions of BMI with renal outcomes (40–44).

Another method to investigate the effects of a high fat mass is by studying hepatic fat. A healthy liver contains little or no fat, but approximately 70% of obese individuals have some degree of liver fat accumulation (NAFLD) (45–47). We hypothesized that alanine aminotransferase (ALT), as a marker of NAFLD might be a better predictor of progression of incident CKD than BMI. We tested this hypothesis in 1,187 patients with type 2 diabetes (chapter 6). ALT was elevated in few subjects. ALT was not significantly associated with incident CKD. To our knowledge, to date, the association of ALT with CKD has not yet been investigated, but it has been shown that the prevalence of CKD is significantly higher among patients with both diabetes and NAFLD compared to those without NAFLD (48,49). In those studies, the majority of patients had ALT levels within the normal range (48,49). Because NAFLD may be present even if ALT levels are normal, this may explain why we did not find an association of ALT with incident CKD.

126

Cha

pter

8

Furthermore, ALT was inversely associated with all-cause mortality, particularly with non-cardiovascular mortality. These results are in line with several population-based studies that mainly included subjects with normal levels of ALT (50–55). It is tempting to speculate that both high and low levels of ALT are associated with mortality. Such a trend would be reflected in a U-shaped relationship or bimodal association. In-deed, other studies found bimodal associations of ALT with all-cause mortality (50,56). In line, the results of the present study also showed a significant bimodal association of ALT with all-cause mortality. A potential explanation for the association of ALT with noncardiovascular mortality is that lower levels of ALT reflect a low number of func-tional hepatocytes. This low amount of hepatocytes may reflect hepatic aging, which un-derlies extra-hepatic pathways that increase the risk of mortality (57,58). Because ALT is also produced in muscle cells (59), another explanation for this finding may be that low levels of ALT reflect a low muscle mass or frailty.

Another way to bring into view the harmful effects of a high fat mass is by taking both BMI and muscle into account. A recent meta-analysis by Nicoletto et al. concluded that obesity (i.e. body mass index [BMI] ≥ 30 kg/m2) is a risk factor for neither graft failure nor mortality in renal transplant recipients (42). As mentioned, muscle mass is an important component of BMI. A high muscle mass is an established marker of better outcomes, both in RTR (60) and in other populations, including the general population (61,62). Accordingly, we investigated whether muscle mass, deter-mined by 24 hr urinary creatinine excretion (UCE), confounds associations of BMI with graft failure and mortality in 916 RTR with a functioning graft for more than one year (chapter 7). In a univariable analysis, BMI was not associated with graft failure. The association of BMI with graft failure remained non-significant after adjustment for age and sex, but became uncovered by further adjustment for UCE. In a univari-able analysis, BMI tended to be positively associated with mortality. This trend disap-peared after adjustment for age and sex, but became uncovered by adjustment for UCE. In conclusion, adjustment for UCE seems to uncover an adverse association of high fat mass, for which BMI became a better measure after adjustment for UCE, with both graft failure and mortality. These findings provide an additional potential explanation for the absence of increased risk of BMI with graft failure and mortality, as previously observed in RTR (63,64). Based on these findings we infer that clinicians should not only monitor BMI, but also muscle mass, because preservation of muscle mass could improve outcome after transplantation.

Conclusion and future perspectivesBased on the findings of this thesis, we conclude that the non-failing liver may be in-volved in the pathophysiology of CKD. One aspect of liver function that we studied


was bilirubin. We found that bilirubin was significantly associated with CKD. However, because of the observational design of the studies, causality could not be ascertained. Causality of the association of bilirubin with CKD can be investigated using the Mende-lian randomization approach in future studies. If bilirubin is no more than a correlated biomarker, it would be interesting to investigate the predictive capabilities of bilirubin and to compare its predictive capabilities to already existing biomarkers. It is also pos-sible that bilirubin is a biomarker for a different causal association with CKD (e.g. for an association of carbon monoxide, or biliverdin with CKD). In that case, the associa-tion of bilirubin with CKD may disappear after controlling for the metabolites that are generated during the heme degradation cycle such as carbon monoxide and biliverdin. In addition, it is important that future studies distinguish between conjugated and un-conjugated bilirubin, because, as pointed out earlier, the effects of conjugated and un-conjugated bilirubin may be different.

Another aspect of liver function that we have studied is protein metabolism. From our research we learned that high protein intake may be beneficial for long-term outcomes in RTR. If these findings are confirmed by other studies, there is need for an intervention study that investigates the effect of high and low protein intake on out-comes after transplantation. It would also be interesting to determine the optimal range of protein intake for RTR. This information could aid clinicians in the treatment of RTR in the future. Furthermore, the most favorable source of protein intake (i.e. animal or plant protein) is unclear at present. It is also unknown why we found that the optimal protein intake of RTR was similar to the optimal protein intake of dialysis patients. We hypothesized that this was due to increased protein catabolism and/or the protein en-ergy wasting syndrome in both RTR and dialysis patients, but we were not able to inves-tigate this. A logical next question would be to determine whether a low protein intake is associated with increased markers of inflammation such as CRP, which are elevated as a consequence of the protein energy wasting syndrome. Furthermore, because a low pro-tein intake could reflect an overall low dietary intake, it should be investigated whether the association of protein intake with outcomes in RTR is independent of caloric intake.

‘Personalized medicine’ is gaining increasing attention with medical decisions becoming tailored to a specific patient rather than a group of patients. Overall, renal transplantation induces protein losses that must be compensated for and muscle mass often decreases after transplantation. However, this may not be similar in every patient. The optimal moment at which action is warranted with respect to protein intake and muscle mass after renal transplantation may be an important area of investigation. Fur-thermore, it would be interesting to investigate whether preservation of muscle mass after transplantation is sufficient, or whether muscle mass should be increased before transplantation is initiated. Knowledge about this optimal moment of intervention would aid clinicians in adequate treatment of RTR (nutritional intervention and exer-cise), subsequently increasing the long-term prognosis after transplantation.

128

Cha

pter

8

References 1. American Diabetes Association. Standards of medical care in diabetes--2012. Diabetes Care

2012 Jan;35 Suppl 1:S11–63. 2. Baynes JW. Role of oxidative stress in development of complications in diabetes. Diabetes 1991

Apr;40(4):405–412. 3. Giugliano D, Ceriello A, Paolisso G. Oxidative stress and diabetic vascular complications. Dia-

betes Care 1996 Mar;19(3):257–267. 4. Giacco F, Brownlee M. Oxidative stress and diabetic complications. Circ Res 2010 Oct

29;107(9):1058–1070. 5. Adin CA, Croker BP, Agarwal A. Protective effects of exogenous bilirubin on ischemia-



7. Hull TD, Agarwal A. Bilirubin: a potential biomarker and therapeutic target for diabetic ne-phropathy. Diabetes 2014 Aug;63(8):2613–2616.





12. Johansen CT, Hegele RA. Using Mendelian randomization to determine causative factors in cardiovascular disease. J Intern Med 2013 Jan;273(1):44–47.

13. Eckel RH, Alberti KG, Grundy SM, Zimmet PZ. The metabolic syndrome. Lancet 2010 Jan 16;375(9710):181–183.

14. Reaven GM. Role of insulin resistance in human disease (syndrome X): an expanded defini-tion. Annu Rev Med 1993;44:121–131.

15. Abbasi A, Deetman PE, Corpeleijn E, Gansevoort RT, Gans RO, Hillege HL, et al. Bilirubin as a po-tential causal factor in type 2 diabetes risk: a Mendelian randomization study. Diabetes 2014 Nov 3.

16. Stender S, Frikke-Schmidt R, Nordestgaard BG, Grande P, Tybjaerg-Hansen A. Genetically elevated bilirubin and risk of ischaemic heart disease: three Mendelian randomization studies and a meta-analysis. J Intern Med 2013 Jan;273(1):59–68.



18. Kang SJ, Kim D, Park HE, Chung GE, Choi SH, Choi SY, et al. Elevated serum bilirubin levels are inversely associated with coronary artery atherosclerosis. Atherosclerosis 2013 Oct;230(2):242–248.


20. Breimer LH, Wannamethee G, Ebrahim S, Shaper AG. Serum bilirubin and risk of ischemic heart disease in middle-aged British men. Clin Chem 1995 Oct;41(10):1504–1508.


22. Nakao A, Neto JS, Kanno S, Stolz DB, Kimizuka K, Liu F, et al. Protection against ischemia/reperfusion injury in cardiac and renal transplantation with carbon monoxide, biliverdin and both. Am J Transplant 2005 Feb;5(2):282–291.

23. Nakao A, Faleo G, Nalesnik MA, Seda-Neto J, Kohmoto J, Murase N. Low-dose carbon mon-oxide inhibits progressive chronic allograft nephropathy and restores renal allograft function. Am J Physiol Renal Physiol 2009 Jul;297(1):F19–26.

24. Wang L, Lee JY, Kwak JH, He Y, Kim SI, Choi ME. Protective effects of low-dose carbon mon-oxide against renal fibrosis induced by unilateral ureteral obstruction. Am J Physiol Renal Physiol 2008 Mar;294(3):F508–17.

25. Tukey RH, Strassburg CP. Human UDP-glucuronosyltransferases: metabolism, expression, and disease. Annu Rev Pharmacol Toxicol 2000;40:581–616.

26. Wang Z, Wong T, Hashizume T, Dickmann LZ, Scian M, Koszewski NJ, et al. Human UGT1A4 and UGT1A3 conjugate 25-hydroxyvitamin D3: metabolite structure, kinetics, inducibility, and interindividual variability. Endocrinology 2014 Jun;155(6):2052–2063.




30. KDOQI. KDOQI Clinical Practice Guidelines and Clinical Practice Recommendations for Dia-betes and Chronic Kidney Disease. Am J Kidney Dis 2007 Feb;49(2 Suppl 2):S12–154.


32. Molnar MZ, Czira ME, Rudas A, Ujszaszi A, Lindner A, Fornadi K, et al. Association of the malnutrition-inflammation score with clinical outcomes in kidney transplant recipients. Am J Kidney Dis 2011 Jul;58(1):101–108.

130

Cha

pter

8


34. Kopple JD, Levey AS, Greene T, Chumlea WC, Gassman JJ, Hollinger DL, et al. Effect of dietary protein restriction on nutritional status in the Modification of Diet in Renal Disease Study. Kidney Int 1997 Sep;52(3):778–791.

35. Keane WF, Collins AJ. Influence of co-morbidity on mortality and morbidity in patients treated with hemodialysis. Am J Kidney Dis 1994 Dec;24(6):1010–1018.

36. Stenvinkel P, Heimburger O, Paultre F, Diczfalusy U, Wang T, Berglund L, et al. Strong associa-tion between malnutrition, inflammation, and atherosclerosis in chronic renal failure. Kidney Int 1999 May;55(5):1899–1911.

37. Stenvinkel P, Holmberg I, Heimburger O, Diczfalusy U. A study of plasmalogen as an index of oxidative stress in patients with chronic renal failure. Evidence of increased oxidative stress in malnourished patients. Nephrol Dial Transplant 1998 Oct;13(10):2594–2600.

38. Owen WF,Jr, Lew NL, Liu Y, Lowrie EG, Lazarus JM. The urea reduction ratio and serum al-bumin concentration as predictors of mortality in patients undergoing hemodialysis. N Engl J Med 1993 Sep 30;329(14):1001–1006.

39. Altorf-van der Kuil W, Brink EJ, Boetje M, Siebelink E, Bijlsma S, Engberink MF, et al. Identi-fication of biomarkers for intake of protein from meat, dairy products and grains: a controlled dietary intervention study. Br J Nutr 2013 Sep 14;110(5):810–822.






45. Kotronen A, Westerbacka J, Bergholm R, Pietilainen KH, Yki-Jarvinen H. Liver fat in the meta-bolic syndrome. J Clin Endocrinol Metab 2007 Sep;92(9):3490–3497.

46. Hamaguchi M, Kojima T, Takeda N, Nakagawa T, Taniguchi H, Fujii K, et al. The metabol-ic syndrome as a predictor of nonalcoholic fatty liver disease. Ann Intern Med 2005 Nov 15;143(10):722–728.



48. Targher G, Pichiri I, Zoppini G, Trombetta M, Bonora E. Increased prevalence of chronic kid-ney disease in patients with Type 1 diabetes and non-alcoholic fatty liver. Diabet Med 2012 Feb;29(2):220–226.

49. Targher G, Bertolini L, Rodella S, Zoppini G, Lippi G, Day C, et al. Non-alcoholic fatty liver disease is independently associated with an increased prevalence of chronic kidney disease and proliferative/laser-treated retinopathy in type 2 diabetic patients. Diabetologia 2008 Mar;51(3):444–450.







56. Koehler EM, Sanna D, Hansen BE, van Rooij FJ, Heeringa J, Hofman A, et al. Serum liver enzymes are associated with all-cause mortality in an elderly population. Liver Int 2014 Feb;34(2):296–304.



59. Kim WR, Flamm SL, Di Bisceglie AM, Bodenheimer HC, Public Policy Committee of the American Association for the Study of Liver Disease. Serum activity of alanine aminotransfer-ase (ALT) as an indicator of health and disease. Hepatology 2008 Apr;47(4):1363–1370.



132

Cha

pter

8


63. Kovesdy CP, Czira ME, Rudas A, Ujszaszi A, Rosivall L, Novak M, et al. Body mass index, waist circumference and mortality in kidney transplant recipients. Am J Transplant 2010 Dec;10(12):2644–2651.

64. Postorino M, Marino C, Tripepi G, Zoccali C, CREDIT (Calabria Registry of Dialysis and Transplantation) Working Group. Abdominal obesity and all-cause and cardiovascular mor-tality in end-stage renal disease. J Am Coll Cardiol 2009 Apr 14;53(15):1265–1272.

9

Nederlandse samenvatting9

135Nederlandse samenvatting

InleidingEr is een wereldwijde toename van het aantal patiënten met chronische nierschade (1). Dit is grotendeels te wijten aan het stijgende aantal patiënten met obesitas, hoge bloed-druk en diabetes, welke risicofactoren zijn voor chronische nierschade (1). Bij sommige patiënten met chronische nierschade verslechtert de nierfunctie dusdanig dat gif- en af-valstoffen op een bepaald moment onvoldoende afgevoerd kunnen worden en deze zich ophopen. In dat stadium zijn patiënten afhankelijk van dialyse en/of niertransplantatie. In Nederland wordt het aantal patiënten met chronische nierschade geschat op 1 mil-joen, waarvan ongeveer 16.000 patiënten nierfalen hebben (2). Stadiëring van nierfunc-tie wordt gedaan aan de hand van de filtercapaciteit van de nier en/of het verlies van het eiwit albumine in de urine (3). Eindstadium nierfalen betekent een filtercapaciteit van minder dan 15 mL/min per 1.73m2 (3).

De stadiëring van chronische nierschade is belangrijk om het risico op comorbide aandoeningen zoals hart- en vaatziekten en voortijdige sterfte in te schatten. Dit is be-langrijk, omdat patiënten met chronische nierschade hier in vergelijking tot de algemene populatie een sterk verhoogde kans op hebben (4). Deze kans is dusdanig vergroot dat pa-tiënten met chronische nierschade er in de meeste gevallen niet aan toe komen daadwer-kelijk nierfalen te ontwikkelen, maar voortijdig overlijden aan bijvoorbeeld cardiovascu-laire ziekte. In gunstige gevallen hebben patiënten met chronische nierschade een 5 maal grotere kans op cardiovasculaire ziekten dan de algemene populatie (5,6). Bij jongere patiënten die dialyseren is deze kans zelfs 1000 keer groter dan bij leeftijdsgenoten (7). Gezien het grote aantal patiënten met nierschade en de relatief slechte prognose van deze patiëntengroep is het van groot belang dat er meer kennis wordt opgedaan over mogelijke factoren die de achteruitgang van nierschade kunnen vertragen of zelfs voorkomen.

Wij hypothetiseren dat zogenaamde ‘crosstalk’ tussen de nier en andere organen een belangrijke invloed kan hebben op het ontstaan en progressie van chronische nier-ziekte. Zoals eerder aangegeven heeft een slechte nierfunctie blijkbaar een slechte in-vloed op het cardiovasculaire systeem. Een ander orgaan waar de nier nauw mee in ver-binding staat is de lever. Een slechte leverfunctie kan de nierfunctie soms dusdanig doen verslechteren dat er nierfalen optreedt (het hepatorenaal syndroom) (8). Echter, het is nagenoeg onbekend of de leverfunctie en de nierfunctie elkaar beïnvloeden wanneer er geen sprake is van een slechte leverfunctie. Daarom was de doelstelling van dit proef-schrift het onderzoeken van de verschillende aspecten van de leverfunctie, hun effecten op chronische nierschade en complicaties van chronische nierschade.

BilirubineEen van de functies van de lever is het omzetten van bilirubine, een afbraakproduct van rode bloedcellen (9,10). De lever conjugeert bilirubine zodat het wateroplosbaar

136

Cha

pter

9

wordt en kan worden uitgescheiden. Bilirubine is een veelgebruikte maat voor de le-verfunctie. Tot voor kort beschouwde men bilirubine slechts als een giftig afvalpro-duct dat geassocieerd was met nierschade wanneer de concentraties te hoog worden in het lichaam (9,10). Echter, nieuwe inzichten suggereren dat een hoognormale con-centratie van bilirubine juist bescherming zou kunnen bieden tegen nierschade. Dit paradoxale beschermende effect wordt toegeschreven aan de antioxidatieve effecten van bilirubine indien de concentraties de normaalwaarde niet overschrijden (11–13). Dierstudies hebben een gunstig effect laten zien van hoognormaal bilirubine op chro-nische nierschade (14–16), in het bijzonder op diabetische nefropathie (14). Het ef-fect van bilirubine op chronische nierschade was tot op heden nog niet onderzocht in mensen. Daarom onderzochten wij de associatie tussen bilirubine en progressie van diabetische nefropathie (hoofdstuk 2). Daartoe includeerden wij 1,498 patiënten met type 2 diabetes, hypertensie, en diabetische nefropathie die reeds deelnamen aan de prospectieve RENAAL en IDNT studies. In dit onderzoek vonden wij een onafhan-kelijke inverse associatie van bilirubine met progressie van diabetische nefropathie, zowel in IDNT als RENAAL.

Een andere vorm van chronische nierschade waarbij oxidatieve stress een rol lijkt te spelen is het achteruitgaan van de nierfunctie na niertransplantatie (17–20). Als de transplantaatfunctie dusdanig vermindert dat de nierfunctie niet meer toereikend is wordt de patiënt weer afhankelijk van dialyse of moet opnieuw getransplanteerd wor-den. Dit wordt ook wel transplantaatfalen genoemd. Het komt vaak voor dat de trans-plantaatfunctie achteruitgaat, maar er is geen goede behandeling om dit te voorkomen. Omdat wordt verondersteld dat oxidatieve stress een belangrijke rol speelt in dit proces, en bilirubine een belangrijke antioxidant lijkt te zijn, hypothetiseren wij dat bilirubine mogelijk een beschermende rol zou kunnen spelen tegen transplantaatfalen. Daartoe onderzochten wij de associatie tussen bilirubine en transplantaatfalen (hoofdstuk 3). In totaal werden er 603 niertransplantatiepatiënten geïncludeerd om dit te onderzoeken. Transplantaatfalen werd gedefinieerd als het weer afhankelijk worden van dialyse of een re-transplantatie. Patiënten met een hogere bilirubineconcentratie hadden een minder uitgesproken achteruitgang van nierfunctie (gemeten als creatinine klaring). Daarnaast kwam transplantaatfalen significant minder vaak voor bij patiënten met een relatief hoge (maar steeds nog normale) bilirubineconcentratie. Bilirubine was niet geassocieerd met overleving. Op basis van deze data leiden wij af dat bilirubine mogelijk zou kunnen be-schermen tegen transplantaatfalen.

CausaliteitDe meeste studies die de associatie van bilirubine met renale uitkomsten onderzoch-ten, inclusief de onze, zijn observationele studies. Het is daarom nog altijd onbekend of een hogere concentratie van bilirubine werkelijk beschermt tegen nierschade, of dat


bilirubine slechts een maat is voor een betere nierfunctie. Resultaten van observationele studies kunnen dit onderscheid niet goed maken. Echter, als uit vervolgonderzoek blijkt dat bilirubine daadwerkelijk beschermt tegen nierschade zou bilirubine in de toekomst een rol kunnen spelen in de behandeling van chronische nierschade (21).

Een manier om het mogelijk bestaan van een causale associatie te onderzoeken is door middel van een ‘Mendelian randomization’ (22,23). ‘Mendelian randomization’ is een statistische methode die is gebaseerd op het feit dat genetische variabiliteit niet onderhevig is aan levensstijl of omgevingsfactoren (23). Maar liefst 18% van de variatie in bilirubine wordt verklaard door genetische variabiliteit in het gen UGT1A1. Het gen UGT1A1 codeert voor een enzym (UDP-GT) dat verantwoordelijk is voor de regulatie van de bilirubine concentratie. Enkele studies die ‘Mendelian randomization’ hebben toegepast om causaliteit van de associatie tussen bilirubine en cardiovasculaire ziekte te onderzoeken leverden teleurstellende resultaten. Zowel in een prospectieve studie als in een case cohort studie was een genetisch verhoogde concentratie bilirubine niet ge-associeerd met ischemische hartziekte (24,25). Deze bevindingen staan in contrast tot een aantal observationele studies waarbij er wel een verband werd gevonden tussen bi-lirubine en cardiovasculaire ziekte (9,26,27). Tot op heden is er nog geen studie geweest die heeft onderzocht of er werkelijk een causale associatie bestaat tussen bilirubine en chronische nierschade.

Verhoging van bilirubineErvan uitgaande dat er werkelijk een causale associatie bestaat tussen bilirubine en lange termijn uitkomsten bij chronische nierschade, is het interessant om te weten of biliru-bine ook op een natuurlijke manier verhoogd kan worden. Dit onderzochten wij in een cross-sectionele studie van mannen zonder chronische ziekten die reeds deelnamen aan de Zutphen elderly study (hoofdstuk 4). Associaties van bilirubine met individu-ele voedselgroepen, een gezond voedingspatroon, en levensstijlfactoren zoals roken en wijnconsumptie werden onderzocht. Kwalificering van het voedingspatroon werd toege-past door middel van de Mediterranean Diet Score (MDS). Een score van hoger dan 4 uit 8 betekende dat het voedingspatroon goed overeenkwam met het mediterrane dieet. Zo-wel een Mediterraans voedingspatroon, als wijnconsumptie was significant geassocieerd met een hogere concentratie van bilirubine. Of dit ook een effect zou kunnen hebben op oxidatieve stress en daaraan gerelateerde ziekten is nog onbekend.

Bilirubine, een marker of beschermende factor?Het is ook mogelijk dat de gesuggereerde beschermende effecten van bilirubine op nier-functie (voor een deel) te verklaren zijn door bijproducten die worden geproduceerd

138

Cha

pter

9

tijdens de degradatie van hemoglobine (en bilirubine productie), te weten koolstofmo-noxide en biliverdine (15,28). Experimentele studies hebben mogelijk beschermende ef-fecten gevonden van koolstofmonoxide en biliverdine op de nier (14,29–31).

Een limitatie van onze studies is dat alleen totaal bilirubine is gemeten. Bili-rubine bestaat in twee vormen; ongeconjugeerd (indirect) en geconjugeerd (direct) bilirubine. Het is van belang om onderscheid te maken tussen deze twee vormen van bilirubine. Enerzijds omdat de conjugatie van bilirubine de eigenschappen en effecten van bilirubine zou kunnen veranderen en daarmee de invloed op biologische proces-sen. Anderzijds omdat de hoeveelheid geconjugeerd bilirubine een maat zou kunnen zijn voor de functionaliteit van de lever om andere metabolieten te conjugeren. Der-gelijke metabolieten zijn afkomstig van cellulair catabolisme, medicatie, en vetoplos-bare vitaminen zoals vitamine D (32,33). Toegenomen conjugatie door de lever zou gepaard kunnen gaan met een ongewenste afbraak en excretie van stoffen en meta-bolieten met beschermende effecten, zoals bijvoorbeeld vitamine D. In andere woor-den, als de conjugatie van bilirubine is afgenomen, zal niet alleen de conjugatie van bilirubine hoger zijn dan normaal, maar ook de concentratie van goede metabolieten zoals vitamine D, die daarmee mogelijk het beschermende effect van een verhoogd bilirubine zouden kunnen verklaren.

UreumDe lever en de nier zijn ook nauw verbonden als het gaat om eiwit metabolisme. Een van de functies van de lever is het omzetten van aminozuren, die afkomstig zijn uit eiwitten, naar ureum. Omdat de nieren de uitscheiding van ureum faciliteren, wordt ureum ook gebruikt als maat voor de nierfunctie. De 24h ureum excretie kan gebruikt worden als maat van de eiwitinname (34).

De optimalisatie van eiwitinname speelt een belangrijke rol in de behandeling van chronische nierschade. Van een hoge eiwitintake wordt gedacht dat het een risi-cofactor is voor een snellere achteruitgang van nierfunctie (35–38). Daarom adviseren huidige richtlijnen een lage eiwitinname (0.6 tot 0.8 g eiwit/kg per dag) aan patiënten met chronische nierschade (37,39,40). Omdat patiënten die dialyseren een hoog eiwit catabolisme hebben adviseren de richtlijnen aan deze patiënten juist een hoge eiwitin-name (1.1 tot 1.3 g/eiwit/kg/dag) (41,42). Het is onbekend of de eiwitinname van nier-transplantatiepatiënten nu juist hoog of laag moet zijn. Daarom onderzochten wij of de 24-uurs ureum excretie, als maat voor eiwitinname, geassocieerd was met transplan-taatfalen en mortaliteit in een cohort van niertransplantatiepatiënten met een functione-rend transplantaat gedurende meer dan een jaar (hoofdstuk 5). Wij gebruikten hiervoor data van herhaalde 24-uurs urinecollecties die werden verzameld in het kader van de reguliere patiëntenzorg. In totaal werden er 940 niertransplantatiepatiënten onderzocht. In deze populatie was de optimale eiwitintake hoger dan wat wordt aanbevolen voor


patiënten met chronische nierschade (37,39,40), maar was vergelijkbaar met de hoe-veelheid die wordt aanbevolen voor patiënten die dialyseren (40,43,44). Ureum excretie was niet geassocieerd met transplantaatfalen in de hele populatie, maar er was wel een inverse associatie met transplantaatfalen in niertransplantatiepatiënten met een body mass index (BMI) van minder dan 25 kg/m2 en in niertransplantatiepatiënten met een geschatte glomerulaire filtratiesnelheid van 45 mL per min per 1.73 m2 of hoger. Ureum excretie was omgekeerd geassocieerd met mortaliteit, onafhankelijk van BMI en geschat-te glomerulaire filtratiesnelheid. Op basis van deze resultaten concluderen wij dat een relatief hogere eiwitinname voordelig zou kunnen zijn voor niertransplantatiepatiënten.

Na niertransplantatie dragen het gebruik van immunosuppressiva en bijvoor-beeld episoden van rejecties en infecties mogelijk bij aan een toegenomen eiwitverbruik bij niertransplantatiepatiënten (45). Wanneer de eiwitbehoefte niet voldoende wordt ge-compenseerd met intake, kan dit leiden tot het ‘protein energy wasting syndrome’. Dit syndroom is een risicofactor voor toegenomen inflammatie, oxidatieve stress, cardiovas-culaire ziekte, en mortaliteit (46–49). Mogelijk wegen de onwenselijke effecten van een hoge eiwitinname (verslechtering van nierfunctie) niet op tegen de wenselijke effecten van een hoge eiwitinname (weefselherstel, spiermassa en betere voedingsstatus) in nier-transplantatiepatiënten. Een kritische kanttekening bij onze studie is dat het wederom een observationele studie betreft. Causaliteit kan daarom wederom niet worden aange-toond. Tevens maakten wij geen onderscheid in de inname van plantaardige of dierlijke eiwitten. Het is goed mogelijk dat een hoge inname van plantaardige eiwitten beter is dan een hoge inname van dierlijke eiwitten. In toekomstige studies zou het daarom in-teressant zijn om te onderzoeken of er een verschil is in deze twee vormen van eiwitin-name en de associaties met uitkomsten.

Obesitas, de lever, en de nier De lever heeft niet alleen een effect op de nier, maar de nier en de lever kunnen ook een gezamenlijk slachtoffer zijn van factoren waaraan zij beiden worden blootgesteld. Obesi-tas zou zo’n factor kunnen zijn, omdat obesitas zowel het risico op een nierfunctiestoor-nis als op een leverfunctiestoornis vergroot (50,51). De definitie van obesitas is contro-versieel. Enkele studies hebben verrassende omgekeerde associaties gerapporteerd van BMI met mortaliteit in chronisch zieke patiënten (52–56). Verschillende verklaringen voor deze paradoxale associatie zijn gesuggereerd. Een van deze verklaringen was dat de studies onderhevig zouden kunnen zijn aan zogenaamde overlevingsbias. Een andere gesuggereerde verklaring is dat patiënten met een lage BMI een hoog risico hebben op het ‘protein energy wasting’ syndroom. Wij hypothetiseren dat een andere mogelijke verklaring zou kunnen zijn dat BMI een slechte maat is voor een hoge vetmassa. Theore-tisch zou een hoog BMI ook veroorzaakt kunnen worden door een hoge spiermassa, wat vaak geassocieerd is met betere gezondheidsuitkomsten (57–59).

140

Cha

pter

9

Een andere methode om de effecten van een hoge vetmassa te onderzoeken is door middel van het meten van accumulatie van vet in de lever. Een gezonde lever bevat nagenoeg geen vet, maar ongeveer 70% van de patiënten met obesitas heeft leververvet-ting (60–62). Wij hypothetiseren dat leververvetting sterker geassocieerd zou kunnen zijn met chronische nierschade dan BMI. Daarom onderzochten wij of alanine amino-transferase (ALT), een marker die veel gebruikt wordt om de mate van leververvetting te schatten, was geassocieerd met nierfunctie en mortaliteit in patiënten met type 2 dia-betes (hoofdstuk 6).

In dit hoofdstuk onderzochten wij patiënten met type 2 diabetes die deelnamen aan de prospectieve ZODIAC studie. In totaal zijn er 1.187 patiënten geïncludeerd. De overgrote meerderheid van de patiënten had een ALT waarde binnen de normaalwaarde. Er was geen associatie van ALT met nierfunctie. Verder vonden we een inverse associatie van ALT met sterfte, in het bijzonder non-cardiovasculaire sterfte. Deze bevinding is in overeenstemming met meerdere studies in de algemene populatie die patiënten heb-ben geïncludeerd met een normale ALT concentratie (63–68). Net zoals in deze studies in de algemene populatie had het merendeel van de patiënten in de ZODIAC studie een ALT concentratie binnen de normale range. Onze bevindingen stonden in tegenstelling tot andere studies die met name ALT boven de normale range bestudeerden. Deze studies vonden juist bij een verhoogde ALT concentratie een hoger risico op (cardiovasculaire) mortaliteit (69,70). Mogelijk is de associatie tussen ALT en mortaliteit afhankelijk van de range waarin ALT bestudeerd wordt; wij speculeren daarom dat een lage ALT con-centratie mogelijk meer een risicofactor is voor non-cardiovasculaire sterfte, terwijl een ALT boven de normale range mogelijk een risicofactor is voor cardiovasculaire sterfte.

Een verklaring voor het hogere risico op non-cardiovasculaire sterfte bij patiënten met een laag ALT zou kunnen zijn dat een lagere ALT een weerspiegeling is van minder functionele hepatocyten. Dit lagere aantal hepatocyten kan een gevolg zijn van ‘hepatic ageing’ (71,72). Daarnaast zou de bevinding in deze studie verklaard kunnen worden door een lagere spiermassa omdat ALT ook voor een deel in spierweefsel wordt geproduceerd.

Een andere manier om de slechte effecten van vetmassa beter in kaart te bren-gen is door niet alleen rekening te houden met BMI, maar ook met spiermassa. In een recente meta-analyse van Nicoletto et al. (54) werd geconcludeerd dat obesitas geen ri-sicofactor is voor transplantaatfalen of mortaliteit na niertransplantatie. Spiermassa is een belangrijk bestanddeel van BMI. Een hoge spiermassa is geassocieerd met betere gezondheidsuitkomsten (57,58). Wij onderzochten of spiermassa (gemeten als 24h ex-cretie van kreatinine) een ‘confounder’ is in de associatie van BMI met transplantaatfa-len en mortaliteit in niertransplantatiepatiënten (hoofdstuk 7). In totaal werden er 916 patiënten geïncludeerd. In de univariable analyse was BMI niet geassocieerd met trans-plantaatfalen. Deze associatie werd echter heel duidelijk zichtbaar na corrigeren voor spiermassa. Er was een vergelijkbaar effect van spiermassa op de associatie tussen BMI en mortaliteit. Gebaseerd op deze data leiden wij af dat BMI een betere maat wordt


voor obesitas als er rekening wordt gehouden met spiermassa en dat een te hoge vet-massa wel degelijk een risicofactor is voor transplantaatfalen en voortijdige sterfte na niertransplantatie. Deze resultaten benadrukken dat het erg belangrijk is om rekening te houden met de spiermassa na transplantatie. Behoud van spiermassa zou de uitkomsten na transplantatie voordelig kunnen beïnvloeden.

Conclusie en toekomstig onderzoekGebaseerd op de bevindingen van het huidige proefschrift concluderen wij dat de le-verfunctie binnen de normale range geassocieerd lijkt te zijn met het ontstaan en de prognose van chronische nierschade. In dit proefschrift beschrijven wij dat met name bilirubine mogelijk een beschermend effect heeft tegen chronische nierschade. Echter, het blijft de vraag of deze associatie causaal is. Zoals eerder genoemd zou een Mendelian Randomization studieopzet uitkomst kunnen bieden om te onderzoeken of de associatie tussen bilirubine en chronische nierschade causaal is. Als causaliteit niet kan worden aangetoond zou het interessant zijn om te onderzoeken of bilirubine een goede voor-speller is van chronische nierschade, en zo ja hoe goed. Het zou ook een interessante vraagstelling kunnen zijn of het beschermende effect van bilirubine te vinden is in het geconjugeerde of het ongeconjugeerde bilirubine.

Een ander aspect van de leverfunctie dat wij bestudeerd hebben is het eiwitmeta-bolisme. Wij vonden dat een hogere eiwitinname mogelijk beter zou zijn voor niertrans-plantatiepatiënten. Het zou interessant zijn om onze bevindingen te repliceren in een interventiestudie die de effecten van een lage en hoge eiwitinname na transplantatie zou bestuderen. Dan zou er ook meer duidelijkheid kunnen ontstaan over de optimale range en de beste bron van eiwitintake (plantaardig of dierlijk). Tot slot zou het interessant zijn om te onderzoeken wanneer het kritieke moment is dat er een interventie noodzakelijk is in het voedingspatroon of levensstijl om eiwitinname en spiermassa te optimaliseren na transplantatie.

Referenties 1. U.S. Renal Data System. USRDS 2013 Annual Data Report: Atlas of Chronic Kidney Disease

and End-Stage Renal Disease in the United States, National Institutes of Health, National Insti-tute of Diabetes and Digestive and Kidney Diseases 2013.

2. Dutch Kidney Foundation. Available at: http://www.nierstichting.nl/nieren/onzenieren/feiten-en-cijfers, 2014.

3. CKD Work Group. Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease. Kidney Int Suppl 2013;3:1–150.

142

Cha

pter

9

4. Wolfe RA, Ashby VB, Milford EL, Ojo AO, Ettenger RE, Agodoa LY, et al. Comparison of mor-tality in all patients on dialysis, patients on dialysis awaiting transplantation, and recipients of a first cadaveric transplant. N Engl J Med 1999 Dec 2;341(23):1725–1730.

5. Aakhus S, Dahl K, Wideroe TE. Cardiovascular disease in stable renal transplant patients in Nor-way: morbidity and mortality during a 5-yr follow-up. Clin Transplant 2004 Oct;18(5):596–604.

6. Kasiske BL. Cardiovascular disease after renal transplantation. Semin Nephrol 2000 Mar;20(2):176–187.

7. Foley RN, Parfrey PS, Sarnak MJ. Clinical epidemiology of cardiovascular disease in chronic renal disease. Am J Kidney Dis 1998 Nov;32:S112–9.

8. Cardenas A. Hepatorenal syndrome: a dreaded complication of end-stage liver disease. Am J Gastroenterol 2005 Feb;100(2):460–467.



11. Abraham NG, Kappas A. Heme oxygenase and the cardiovascular-renal system. Free Radic Biol Med 2005 Jul 1;39(1):1–25.

12. Vitek L, Schwertner HA. The heme catabolic pathway and its protective effects on oxidative stress-mediated diseases. Adv Clin Chem 2007;43:1–57.

13. Morita T. Heme oxygenase and atherosclerosis. Arterioscler Thromb Vasc Biol 2005 Sep;25(9):1786–1795.


15. Adin CA, Croker BP, Agarwal A. Protective effects of exogenous bilirubin on ischemia-reperfusion injury in the isolated, perfused rat kidney. Am J Physiol Renal Physiol 2005 Apr;288(4):F778–84.

16. Hull TD, Agarwal A. Bilirubin: a potential biomarker and therapeutic target for diabetic ne-phropathy. Diabetes 2014 Aug;63(8):2613–2616.






21. Johansen CT, Hegele RA. Using Mendelian randomization to determine causative factors in cardiovascular disease. J Intern Med 2013 Jan;273(1):44–47.

22. Eckel RH, Alberti KG, Grundy SM, Zimmet PZ. The metabolic syndrome. Lancet 2010 Jan 16;375(9710):181–183.

23. Reaven GM. Role of insulin resistance in human disease (syndrome X): an expanded defini-tion. Annu Rev Med 1993;44:121–131.


25. Stender S, Frikke-Schmidt R, Nordestgaard BG, Grande P, Tybjaerg-Hansen A. Genetically elevated bilirubin and risk of ischaemic heart disease: three Mendelian randomization studies and a meta-analysis. J Intern Med 2013 Jan;273(1):59–68.

26. Kang SJ, Kim D, Park HE, Chung GE, Choi SH, Choi SY, et al. Elevated serum bilirubin levels are inversely associated with coronary artery atherosclerosis. Atherosclerosis 2013 Oct;230(2):242–248.

27. Breimer LH, Wannamethee G, Ebrahim S, Shaper AG. Serum bilirubin and risk of ischemic heart disease in middle-aged British men. Clin Chem 1995 Oct;41(10):1504–1508.


29. Wang L, Lee JY, Kwak JH, He Y, Kim SI, Choi ME. Protective effects of low-dose carbon mon-oxide against renal fibrosis induced by unilateral ureteral obstruction. Am J Physiol Renal Physiol 2008 Mar;294(3):F508–17.

30. Nakao A, Faleo G, Nalesnik MA, Seda-Neto J, Kohmoto J, Murase N. Low-dose carbon mon-oxide inhibits progressive chronic allograft nephropathy and restores renal allograft function. Am J Physiol Renal Physiol 2009 Jul;297(1):F19–26.

31. Nakao A, Neto JS, Kanno S, Stolz DB, Kimizuka K, Liu F, et al. Protection against ischemia/reperfusion injury in cardiac and renal transplantation with carbon monoxide, biliverdin and both. Am J Transplant 2005 Feb;5(2):282–291.

32. Tukey RH, Strassburg CP. Human UDP-glucuronosyltransferases: metabolism, expression, and disease. Annu Rev Pharmacol Toxicol 2000;40:581–616.

33. Wang Z, Wong T, Hashizume T, Dickmann LZ, Scian M, Koszewski NJ, et al. Human UGT1A4 and UGT1A3 conjugate 25-hydroxyvitamin D3: metabolite structure, kinetics, inducibility, and interindividual variability. Endocrinology 2014 Jun;155(6):2052–2063.


35. Brantsma AH, Atthobari J, Bakker SJ, de Zeeuw D, de Jong PE, Gansevoort RT. What predicts progression and regression of urinary albumin excretion in the nondiabetic population? J Am Soc Nephrol 2007 Feb;18(2):637–645.

144

Cha

pter

9




39. KDOQI. Clinical Practice Guidelines and Clinical Practice Recommendations for Diabetes and Chronic Kidney Disease. Am J Kidney Dis 2007 Feb;49(2 Suppl 2):S12–154.






45. Molnar MZ, Czira ME, Rudas A, Ujszaszi A, Lindner A, Fornadi K, et al. Association of the malnutrition-inflammation score with clinical outcomes in kidney transplant recipients. Am J Kidney Dis 2011 Jul;58(1):101–108.

46. Keane WF, Collins AJ. Influence of co-morbidity on mortality and morbidity in patients treated with hemodialysis. Am J Kidney Dis 1994 Dec;24(6):1010–1018.

47. Stenvinkel P, Heimburger O, Paultre F, Diczfalusy U, Wang T, Berglund L, et al. Strong associa-tion between malnutrition, inflammation, and atherosclerosis in chronic renal failure. Kidney Int 1999 May;55(5):1899–1911.

48. Stenvinkel P, Holmberg I, Heimburger O, Diczfalusy U. A study of plasmalogen as an index of oxidative stress in patients with chronic renal failure. Evidence of increased oxidative stress in malnourished patients. Nephrol Dial Transplant 1998 Oct;13(10):2594–2600.

49. Owen WF,Jr, Lew NL, Liu Y, Lowrie EG, Lazarus JM. The urea reduction ratio and serum al-bumin concentration as predictors of mortality in patients undergoing hemodialysis. N Engl J Med 1993 Sep 30;329(14):1001–1006.

50. Kwakernaak AJ, Toering TJ, Navis G. Body mass index and body fat distribution as renal risk factors: a focus on the role of renal haemodynamics. Nephrol Dial Transplant 2013 Nov;28 Suppl 4:iv42–9.


51. Pinto-Sietsma SJ, Navis G, Janssen WM, de Zeeuw D, Gans RO, de Jong PE, et al. A central body fat distribution is related to renal function impairment, even in lean subjects. Am J Kid-ney Dis 2003 Apr;41(4):733–741.









60. Kotronen A, Westerbacka J, Bergholm R, Pietilainen KH, Yki-Jarvinen H. Liver fat in the meta-bolic syndrome. J Clin Endocrinol Metab 2007 Sep;92(9):3490–3497.

61. Hamaguchi M, Kojima T, Takeda N, Nakagawa T, Taniguchi H, Fujii K, et al. The metabol-ic syndrome as a predictor of nonalcoholic fatty liver disease. Ann Intern Med 2005 Nov 15;143(10):722–728.





146

Cha

pter

9




69. Ioannou GN, Weiss NS, Boyko EJ, Mozaffarian D, Lee SP. Elevated serum alanine aminotrans-ferase activity and calculated risk of coronary heart disease in the United States. Hepatology 2006 May;43(5):1145–1151.

70. Lee TH, Kim WR, Benson JT, Therneau TM, Melton LJ,3rd. Serum aminotransferase activity and mortality risk in a United States community. Hepatology 2008 Mar;47(3):880–887.



10

Dankwoord10

149Dankwoord

Dat dit proefschrift gereed is gekomen is te danken aan de inspanning van veel mensen binnen en buiten het UMCG. Mijn promotoren, collega’s, familie en vrienden hebben mij op een geweldige manier over de hindernissen geholpen die ik onderweg ben tegen-gekomen. Ik wil een aantal mensen in het bijzonder noemen;

Prof. dr. S.J.L. Bakker, beste Stephan, ik ben ontzettend dankbaar dat je mij de kans hebt geboden (en slechts een klein beetje hebt aangedrongen) om dit MD/PhD traject te doen. Mijn promotietijd was niet alleen ontzettend leerzaam, maar ook gewoon heel erg leuk. Ik heb veel geleerd van je taalgevoel, oog voor detail en kennis van statistiek. Ik heb altijd blindelings kunnen vertrouwen op je wetenschappelijke inzicht. Daarnaast vind ik het bewonderenswaardig dat je ondanks je drukke agenda belangeloos tijd hebt gestoken in mijn persoonlijke ontwikkeling.

Prof. dr. G. Navis, beste Gerjan, ik respecteer je wetenschappelijke inzicht enorm en het was fijn om op jouw expertise te kunnen vertrouwen. Ik wil je hartelijk danken voor je waardevolle adviezen.

Prof. dr. J.A. Lisman, Prof. dr. ir E.J.M. Feskens, en Prof. dr. Y.M. Smulders, hartelijk dank dat jullie de tijd hebben genomen om mijn proefschrift kritisch door te lezen en te beoordelen.

Alle patiënten die belangeloos hun gegevens beschikbaar hebben gesteld voor research; zonder jullie was dit proefschrift er niet geweest. Bedankt!

Alle co-auteurs van mijn manuscripten, bedankt voor de prettige samenwerking.

J.E. Kootstra, beste Jenny, bedankt dat ik altijd bij jou kon aankloppen voor hulp en in-formatie. Dr. R.P.F. Dullaart, beste Robin, bedankt voor de fijne samenwerking.

De collega’s van het Diabetes Centre, met name Nanno Kleefstra en Prof. dr. H.J.G. Bilo, bedankt voor de fijne samenwerking en jullie gastvrijheid op de EASD. Klaas Groenier, hartelijk dank voor je uitstekende statistische adviezen en hulp.

Prof. dr. D. Kromhout, hartelijk dank voor de prettige samenwerking.

Mijn paranimfen: Jorien, ik vind het bijzonder waardevol om zo’n goede en lieve vrien-din te hebben. Ineke, mijn partner in crime als het om STATA en statistiek gaat; ik heb niet alleen ontzettend veel van je geleerd, maar ook heel veel plezier met je gehad. Dank-julliewel dat jullie aan mijn zij willen staan tijdens de verdediging van mijn proefschrift.

150

Cha

pter

9

JC Mantis, bedankt voor het begrip wanneer de geneeskunde-mantissen het weer over leervragen hadden of er voor de zoveelste keer een woensdag werd overgeslagen voor een tentamen, maar ook als er werd gesproken over opleidingsplekken, sollicitaties en manus-cripten. We hebben samen een geweldige studententijd gehad en ik hoop dat we elkaar nog heel lang zullen blijven zien. In het bijzonder Jara, je filosofische blik op het leven heeft me geleerd dat niets vanzelfsprekend is. Anne-Marie, ik neem een voorbeeld aan je aangezien je het presteert om 120 uur per week te werken en dan nog steeds vrolijk blijft lachen!

Florine en Anna, jullie zijn lieve vriendinnen! Bedankt voor de gezellige tijd, etentjes en welkome afleiding samen in Groningen.

Collega’s op de brug en in het Triade gebouw: onder andere Solmaz, Edwin, Lyanne, Niek, Michel, Elise, Debbie, Esmée, Marieke, Gerald, Laura de Vries, Laura Harskamp, Marco, Annet, Harmke, Michèle, Isidor, Ilse, Jelmer, Charlotte en Dorien, bedankt dat jullie altijd open stonden om samen te werken en om vragen te beantwoorden. Onze meetings hebben bijgedragen aan mijn bredere wetenschappelijke ontwikkeling, be-dankt voor al jullie bijdragen daarin. Niels, het was heel leerzaam om je te begeleiden, dank daarvoor. Winie, je bent de allerbeste! Bedankt voor je geduld en al je hulp. In het bijzonder Tsjitske, jouw relativerende kijk op de wereld en onderzoek heeft mij op de moeilijkere momenten opgebeurd. Janneke, het was een genoegen om deelgenoot te zijn van jouw Groningse nuchterheid. Arjan, jouw kennis is niet te evenaren en ik mis nog altijd jouw updates over de nieuwste studies en weetjes uit de literatuur.

Chris, Lies, Corine, Khaossou, Tineke en Joris, hartelijk dank dat ik zo welkom ben in jullie familie. Bedankt voor alle gezellige familie bijeenkomsten en jullie belangstelling in mijn promotie onderzoek.

Paul, jij hebt mij de handvatten gegeven om “bottlenecks” te elimineren en om “out of the box” te denken. Met jouw hulp heb ik zo veel mogelijk “key issues” weten te “tacke-len”. Wat ik voor onmogelijk houd, maak jij mogelijk. Wanneer ik problemen zie, schiet jij ze onderuit. Met niemand kan ik zo goed de wereld doornemen als met jou. Ik ben enorm dankbaar voor zo’n lieve, opgewekte vriend en steun!

Papa en mama, bedankt dat jullie mij alle kansen hebben gegeven om mij zowel binnen als buiten het medische vlak te ontwikkelen. Papa, in de tweede klas had ik al discussies met jou over inverse verbanden. Wie had dat gedacht dat ik me daar nu nog steeds mee bezig zou houden. Zonder al jullie hulp had ik hier zeker niet gestaan. Het is een waardevol goed om twee liefdevolle ouders te hebben, waar je onvoorwaardelijk op terug kunt vallen.

Nicole

151

95% CI 95% Confidence interval ACE Angiotensin

converting enzyme ACR Albumin/creatinine ratio ALT Alanine aminotransferase ANOVA Analysis of variance AP Alkaline phosphatase ARB Angiotensin II

receptor blocker AST Aspartate aminotransferase B Unstandardized beta BMI Body mass index BSA Body surface area CI Calcineurin inhibitor CKD Chronic kidney disease CKD-EPI Chronic kidney disease

epidemiology collaboration CMV Cytomegalovirus CO Carbon monoxide CRP C-reactive protein CVD Cardiovascular disease DGF Delayed graft function DN Diabetic nephropathy DSCR Doubling of serum creatinine eGFR Estimated glomerular

filtration rate ESRD End-stage renal disease FFQ Food frequency

questionnaire GFR Glomerular filtration rate HbA1c Glycated hemoglobin HDL High density lipoprotein HLA Human leucocyte antigen HO-1 Heme oxygenase -1 HR Hazard ratio HRs Hazard Ratio’s

IDI Integrated discrimination improvement

IDNT Irbesartan Diabetic Nephropathy Trial

IQR Interquartile range MDRD Modification of diet

in renal disease MDS Mediterranean diet score METc Medical ethical committee mTOR Mammalian target

of rapamycin NADPH Nicotinamide adenine

dinucleotide phosphate NAFLD Non-alcoholic fatty

liver disease NIDDM Non-insulin dependent

diabetes mellitus NRI Net reclassification

improvement NSAID’S Nonsteroidal anti-

inflammatory drugs PI Proliferation inhibitor RENAAL Reduction of endpoints

in NIDDM with the angiotensin II antagonist losartan

RTR Renal transplant recipients SD Standard deviation SES Socioeconomic status Tx Transplantation UCE Urinary creatinine excretion UUE Urinary urea excretion ZODIAC Zwolle Outpatient Diabetes

project Integrating Available Care

Abbreviations

152

Cha

pter

9

Alaa Alkhalaf Dept. of Gastroenterology, Isala Clinics, Zwolle, the Netherlands

Stephan J.L. BakkerDiv. of Nephrology, Dept. of Medicine, University of Groningen, University Med-ical Center Groningen, Groningen, the Netherlands; Top Institute Food and Nu-trition, Wageningen, the Netherlands

Henk J.G. BiloDiabetes Centre, Isala clinics, Zwolle, the Netherlands; Dept. of Medicine, Univer-sity of Groningen, University Medical Center Groningen, Groningen, the Neth-erlands

Mark E. CooperBaker IDI Heart & Diabetes Institute, Mel-bourne, Australia

Petronella E. DeetmanDiv. of Nephrology, Dept. of Medicine, University of Groningen, University Med-ical Center Groningen, Groningen, the Netherlands

Robin P.F. DullaartDiv. of Endocrinology, Dept. of Medicine, University of Groningen, University Medi-cal Center Groningen, Groningen, the Netherlands

Reinold O.B. GansDept. of Medicine, University of Gronin-gen, University Medical Center Gronin-gen, Groningen, the Netherlands

Klaas H. GroenierDiabetes Centre, Isala clinics, Zwolle, the Netherlands; Dept. of General PracticeUniversity of Groningen, University Medi-cal Center Groningen, Groningen, the Netherlands

Jaap J. Homan van der HeideDiv. of Nephrology, Dept. of Medicine, University of Groningen, University Med-ical Center Groningen, Groningen, the Netherlands

Michel M. JoostenDiv. of Nephrology, Dept. of Medicine, University of Groningen, University Med-ical Center Groningen, Groningen, the Netherlands; Top Institute Food and Nu-trition, Wageningen, the Netherlands

Nanne KleefstraDiabetes Centre, Isala clinics, Zwolle, the Netherlands

Jenny E. Kootstra-RosDept. of Laboratory Medicine, University of Groningen, University Medical Center Groningen. Groningen, the Netherlands

Daan KromhoutDiv. of Human Nutrition, Wageningen University, Wageningen, the Netherlands

Hiddo J. Lambers HeerspinkDept. of Clinical Pharmacy and Pharma-cology, University of Groningen, Universi-ty Medical Center Groningen, Groningen, the Netherlands

Author affiliations

153Author affiliations

Gijs W.D. LandmanDiabetes Centre, Isala clinics, Zwolle, the Netherlands; Dept. of Medicine, Gelre Hospital Apeldoorn, the Netherlands

Julia B. LewisDiv. of Nephrology, Dept. of Medicine, VanderBilt University, Nashville, TN, USA

Gerjan NavisDiv. of Nephrology, Dept. of Medicine, University of Groningen, University Med-ical Center Groningen, Groningen, the Netherlands

Ineke J. RiphagenDiv. of Nephrology, Dept. of Medicine, University of Groningen, University Med-ical Center Groningen, Groningen, the Netherlands

M. Yusof SaidDiv. of Nephrology, Dept. of Medicine, University of Groningen, University Med-ical Center Groningen, Groningen, the Netherlands

Jan-Stephan F. SandersDiv. of Nephrology, Dept. of Medicine, University of Groningen, University Med-ical Center Groningen, Groningen, the Netherlands

Marc A.J. SeelenDiv. of Nephrology, Dept. of Medicine, University of Groningen, University Med-ical Center Groningen, Groningen, the Netherlands

Dick de ZeeuwDept. of Clinical Pharmacy and Pharma-cology, University of Groningen, Universi-ty Medical Center Groningen, Groningen, the Netherlands

Dorien M. ZelleDiv. of Nephrology, Dept. of Medicine, University of Groningen, University Med-ical Center Groningen, Groningen, the Netherlands

154

Cha

pter

9

List of Publications

Plasma procalcitonin and long‑term outcomes after renal transplantation. Deetman PE, Zelle DM, Navis G, Bakker SJL. In preparation.

Diet, lifestyle, and total bilirubin: the Zutphen elderly study. Deetman PE*, Riphagen IJ*, Navis G, Bakker SJL, Kromhout D. Submitted. *both authors contributed equally

Alanine aminotransferase and mortality in patients with type 2 diabetes (ZODIAC‑38). Deetman PE, Alkhalaf A, Landman GWD, Groenier KH, Kootstra-Ros JE, Navis G, Bilo HJG, Kleefstra N, Bakker SJL. Eur J Clin Invest, provisionally accepted.

Uncovering of body mass index as a risk factor for poor long‑term outcome after renal transplantation. Deetman PE, Sanders JS, Seelen MA, Gans RO, Navis G, Bakker SJ. Transplantation 2015;99(1):e5–6

Urinary urea excretion and long‑term outcome after renal transplantation. Deetman PE, Said MY, Kromhout D, Dullaart RP, Kootstra-Ros JE, Sanders JS, Seelen MA, Gans RO, Navis G, Joosten MM, Bakker SJ. Transplantation 2014;12:(epub ahead of print).

Bilirubin as a potential causal factor in type 2 diabetes risk: a Mendelian randomization study. Abbasi A, Deetman PE, Corpeleijn E, Gansevoort RT, Gans RO, Hillege HL, van der Harst P, Stolk RP, Navis G, Alizadeh BZ, Bakker SJ. Diabetes 2014;3:(epub ahead of print).

Bilirubin and progression of nephropathy in type 2 diabetes: a post hoc analysis of RENAAL with independent replication in IDNT. Riphagen IJ, Deetman PE, Bakker SJ, Navis G, Coo-per ME, Lewis JB, de Zeeuw D, Lambers Heerspink HJ. Diabetes 2014;63(8):2845–53.

The relationship of the anti‑oxidant bilirubin with free thyroxine is modified by insulin resistance in euthyroid subjects. Deetman PE, Bakker SJ, Kwakernaak AJ, Navis G, Dullaart RP; PREVEND Study Group. PLoS One 2014;9(3):e90886.

155List of Publications

High sensitive C‑reactive protein and serum amyloid A are inversely related to serum bili‑rubin: effect‑modification by metabolic syndrome. Deetman PE, Bakker SJ, Dullaart RP. Cardiovasc Diabetol 2013;12:166.

Low‑normal free thyroxine confers decreased serum bilirubin in type 2 diabetes mellitus. Deetman PE, Kwakernaak AJ, Bakker SJ, Dullaart RP. Thyroid 2013;23(11):1367–73.

Support for a protective effect of bilirubin on diabetic nephropathy in humans. Zelle DM, Deetman N, Alkhalaf A, Navis G, Bakker SJ. Kidney Int. 2011 Mar;79(6):686.

Plasma bilirubin and late graft failure in renal transplant recipients. Deetman PE, Zelle DM, Homan van der Heide JJ, Navis GJ, Gans RO, Bakker SJ. Transpl Int 2012;5(8):876–81

Documents

University of Groningen Nutritional and metabolic aspects of the … · 2016. 3. 9. · Proefschrift ter verkrijging van de graad van doctor aan de Rijksuniversiteit Groningen op