Kidney International, Vol. 56, Suppl. 72 (1999), pp. S-37–S-40

Acid-base balance and substitution fluid duringcontinuous hemofiltration

PETER HEERING, KATRIN IVENS, OLIVER THUMER, MATTHIAS BRAUSE, and BERND GRABENSEE

Department of Nephrology, Heinrich-Heine-University of Dusseldorf, Dusseldorf, Germany

Acid-base balance during continuous hemofiltration in patients CRRT. Acetate- and lactate-based physiological solu-with acute renal failure. Critically ill patients with acute renal tions have the advantage of being commercially avail-failure usually present with an unstable acid-base balance, often able. Under normal conditions, acetate and lactate areleading to cardiovascular complications and multi-organ fail-

metabolized rapidly to bicarbonate. In addition, the ad-ure. Therefore, to prevent metabolic acidosis, acid-base bal-ministration of large amounts of lactate may result inance must be normalized and maintained; these patients are

primarily treated with continuous hemofiltration techniques increased urea generation (abstract; Olbricht et al, Intusing different replacement fluids to influence the acid-base Soc Blood Purif 10:48, 1992). Bicarbonate has the majorvalues. Dialysate solutions can be an acetate-based, lactate-

advantage of being the most physiologic anion. Unfortu-based, citrate-based or bicarbonate-based buffer. This articlenately, the production of a commercially available bicar-discusses the strengths and weaknesses of each type of hemofil-

tration replacement fluid. bonate-based physiologic solution is difficult because ofthe loss of bicarbonate from the solution or the formationof calcium and magnesium salts. Persistent acidosis has

Acute renal failure (ARF) complicating the course of been demonstrated to be an indicator of poor prognosiscritical illness continues to be associated with a high [3]. Acetate, as well as lactate, must be metabolized inmortality, despite many advances made over the last the organism in order to compensate for the loss ofdecades in the care of the critically ill patients. The ma- endogenous bicarbonate. Acetate is converted on a 1:1jority of these patients suffer from multiple- rather than basis to bicarbonate by both the liver and skeletal muscle.single-organ failure. Continuous venovenous hemofil- Lactate is converted in the liver on a 1:1 basis to bicar-tration (CVVH) has become common as a renal replace- bonate, and each type of anion is capable of providingment therapy, which is often better tolerated than inter- an alkali load sufficient to correct the acidemia of ARF.mittent hemodialysis at facilitating extracellular fluid Previous studies have suggested that the metabolism ofvolume control and reducing cardiovascular disturbances acetate and lactate was impaired in critical illness and[1, 2]. During continuous renal replacement therapies might be able to produce acidosis [2, 4]. Furthermore,(CRRTs), alkali administration, either in the replace- it was suggested that lactate and acetate might act as ament fluid or by diffusive uptake from the dialysate, must cardiovascular depressant [4]. The correction of meta-replace not only the bicarbonate lost in the buffering of bolic acidosis is a primary aim for renal replacementendogenous acid production, but also the bicarbonate therapy. Depending on the type of dialytic modality em-lost across the hemodiafilter. In most patients with ARF, ployed, the buffer balance is influenced by several fac-a buffer concentration of approximately 30 to 35 mmol/ tors; among those, the type of buffer employed influencesliter and an ultrafiltration rate (or dialysate flow rate) not only acid-base correction, but possibly also clinicalapproaching or exceeding 1 liter/hr can be expected to outcomes. In this brief review, the importance of differ-normalize the acid-base status within one or two days. ent buffers applied is reviewed under the topic of ARFIn individuals with ARF and lactic acidemia, the buffer in intensive care medicine.load may need to approach $50 mmol/liter to correctthe associated metabolic acidemia. Acetate, bicarbonate,citrate, and lactate are the possible choices for bicarbon- DIALYTIC TECHNIQUESate or bicarbonate-equivalent administration during The buffer balance depends on buffer losses with ultra-

filtrate and buffer gain with replacement fluid. Whencontinuous venous-venous hemofiltration is applied toKey words: critical care, multi-organ failure, metabolic acidemia, acido-

sis, dialysate buffer. reduce the patient’s fluid overload, huge volumes of bi-carbonate are lost and this bicarbonate loss has to be 1999 by the International Society of Nephrology

S-37

Heering et al: Acid-base balance during CVVHS-38

replaced. When metabolic acidosis needs to be corrected,the amount of buffer in the replacement fluid must ex-ceed the amount lost in ultrafiltrate, thus providing apositive balance of buffer. Several observations havereported that the loss of bicarbonate in the ultrafiltrateexceeds that of plasma so that a sieving coefficient ofhigher than 1 can be calculated (between 1.1 and 1.25)[5]. Because of their greater stability compared with bi-carbonate, acetate or lactate is used in the buffer ofcommercial fluids. The natural buffer, bicarbonate, pre-cipitates as an insoluble salt when sterilized togetherwith divalent cations, calcium, and magnesium. Recent

Fig. 1. Plasma bicarbonate levels in patients treated with lactate ( ),technological advances have enabled this problem to beacetate (h) or bicarbonate (j) based buffer in continuous venous-

overcome. A double chamber bag with two separate venous hemofiltration (CVVH). (Data are from: abstract; Heering etal, J Am Soc Nephrol 8:159, 1997).compartments, one containing bicarbonate and the other

calcium and magnesium, allows the bicarbonate solutionto be sterilized and stored. Bicarbonate is the most physi-ologic of the possible anions available for use in CRRT.

and had a presumably higher protein catabolic rate (ab-Based on our experience and that of other investigators,stract; Olbricht et al, Int Soc Blood Purif 10:48, 1992).normalization of the acid-base balance in patients with

The use of lactate-based hemofiltration/hemodiafiltra-metabolic acidemia resulting only from ARF can be ac-tion solutions is generally well tolerated in most patientscomplished within 24 to 48 hours using these rates ofundergoing continuous therapies. Most commerciallybicarbonate gain.available solutions are buffered with approximately 40 to45 mmol/liter of lactate. The resulting exogenous lactate

CLINICAL FINDINGS load is capable of being cleared by the liver even in thesetting of fulminant hepatic failure. However, there areFew clinical studies have been performed in criticallycomplications that may arise from the use of lactate-ill patients treated with CVVH in order to assess thebased replacement fluids in critically ill patients. Theefficacy of acidosis correction and to compare the differ-administration of large amounts of lactate without itsently buffered replacement solutions. However, a greatredox partner pyruvate can possibly result in an in-deal of information could be derived from the treatmentcreased protein catabolic rate and myocardial depres-of chronic patients or from intermittent treatments ofsion. The administration of lactate in patients with severecritically ill patients. The effect on acid-base status de-hypoxia, liver impairment, or pre-existing lactic acidemiapends on the buffer concentration in the fluid and oncan result in a worsening of lactic acidemia. In criticallythe type of buffer used. In general, each type of anionill patients with shock and liver failure, the use of lactate-is capable of providing an alkali load sufficient to correctbased fluids in CRRTs may be associated with a paradox-the acidemia of ARF.ical worsening of acid-base balance. If lactate-based flu-However, there are clinical conditions that might in-ids are to be used in these individuals, close attention tofluence the clinician’s decision to chose a particularthe patient’s catabolic rate, acid-base status, and arterialbuffer. Hyperacetatemia has a peripheral vasodilatinglactate levels may be necessary. If hemodynamic instabil-effect and a myocardial depressant effect and is ableity worsens, an assessment of myocardial function and—ifto increase oxygen consumption as a result of acetateindicated—inotropic support may be appropriate. In ad-metabolism [4]. In a recent study on critically ill patientsdition, serial measurements of the serum anion gap maytreated with CVVH and a 35 mmol/liter acetate replace-be necessary to detect the accumulation of d-lactate,ment solution, no substantial increase in blood bicarbon-which is not measured by standard laboratory methodol-ate concentration was recorded (Fig. 1) (abstract; Heer-ogy. Continuous lactate-based buffer is still the treat-ing et al, J Am Soc Nephrol 8:159, 1997). In this study,ment of choice. The exceptions are impairment of liverthe correction of metabolic acidosis occurred much morefunction and lactic acidosis [2].rapidly and completely in patients who were treated with

When oxidizable anions are used in the replacementCVVH and lactate- or bicarbonate-based replacementfluids, the anion (acetate or lactate) must be completelytherapy. Lactate is a powerful peripheral vasodilator,oxidized to CO2 and H2O in order to generate an equimo-affects myocardial contractility, and reduces blood pres-lar amount of bicarbonate. If the metabolic conversion ofsure. It has been demonstrated that the use of lactate-nonbicarbonate anions proceeds without accumulation,buffered replacement solution was associated with a

higher serum urea concentration and urea generation the buffering capacity of them is equal to that of bicar-

Heering et al: Acid-base balance during CVVH S-39

One goal of CRRT is to maintain normal acid-basebalance in patients with ARF, thereby preventing thedetrimental effects of acidemia on cardiovascular perfor-mance. Studies on acetate-based dialysates have shownundesirable metabolic side-effects such as hypotensiveepisodes resulting from the conversion of sodium acetateto sodium bicarbonate [7]. Meanwhile, some studies haveshown that nonsurvivors remain more acidotic (abstract;Heering et al, J Am Soc Nephrol 8:159, 1997) [8, 9]. Ithas been suggested that nonsurvivors have a decreasedability to correct metabolic acidosis [10]. The fact thatbicarbonate did not significantly improve these acid-based disturbances supports the evidence that this mightFig. 2. Plasma bicarbonate levels in patients treated with lactate (h)be due to processes other than a defect in lactate metabo-or bicarbonate ( ) based buffer in CVVH. (Data are from: abstract;

Riegel et al, J Am Soc Nephrol 8:170A, 1997). lism.Lactate-based replacement fluids are also associated

with untoward effects [6], but are reported to cause fewerdialysis related symptoms, including hypotension. Lac-

bonate. Thus, the effect on acid-base status depends on tate and acetate buffers, in particular, have been shownthe buffer concentration in the fluid rather than on the to exert a negative influence on the mean arterial bloodkind of buffer used. When the metabolic conversion is pressure and cardiac function [4]. The problems of wors-impaired (for example, a hyperlactatemia could occur in ening lactic acidemia in patients with severe circulatorypatients with an associated liver disease who were failure and hepatic dysfunction who are hemofilteredtreated with a lactate-buffered replacement solution) [6], using the commercially available lactate-buffered replace-the increase blood concentration of the anion leads to ment fluids are now well recognized [11]. An increase inincrease losses of this anion across the dialytic mem- carbon dioxide production is an inevitable consequencebrane, thus reducing the buffer gain. of the correction of lactic acidosis by bicarbonate-buffered

Similar results were reported after the use of bicarbon- hemofiltration, and particular attention should be paidate (abstracts; Heering et al, J Am Soc Nephrol 8:159, to the ventilation of the patient to ensure that the pCO21997; Riegel et al, J Am Soc Nephrol 8:170A, 1997). level does not rise [12]. Hyperlactatemia occurred inBicarbonate is the physiological buffer for the body sys- all groups, with the lactate group having higher levelstem. Because of its solubility characteristics, no commer- throughout. Although the lactate levels exceeded normalcial products have been available until now. Recently, values in most cases, they were not generally associatedthe results of a large multicenter study on patients with a negative effect on arterial pH. Previous studiestreated with CVVH and bicarbonate-buffered fluid have of lactate-based hemofiltration and dialysis fluids havebeen published. The comparison of a 33.75 mmol/liter reported an accumulation of lactate during treatment [13].lactate-buffered replacement solution with a new 35 The use of bicarbonate-buffered hemofiltration solu-mmol/ml bicarbonate fluid demonstrated the efficacy of tions has only been studied scarcely in the literaturethe bicarbonate solution with the regards to the acidosis (abstract; Olbricht et al, Int Soc Blood Purif 10:48, 1992)correction and in the lowered hyperlactatemia (Fig. 2) [8, 9]. This is probably because of the fact that bicarbon-(abstract; Riegel et al, J Am Soc Nephrol 8:170A, 1997). ate-buffered hemofiltration solution was found to be un-The actual bicarbonate levels were improved in patients stable and, therefore, is not officially approved, at leasttreated with bicarbonate-based buffer as compared with in Germany. An improvement in the prognosis of thislactate-based buffer. Future studies need to focus on extremely sick group of patients has been described iffurther indications. It is expected that the relevance of bicarbonate-buffered replacement fluid is used, althoughbicarbonate-based buffer will increase in the future and controversial results have been reported [11, 14].that the use of this buffer will be able to improve cardio-vascular hemodynamics and prognosis in these patients.

CONCLUSIONS

There is established evidence that the correction ofDISCUSSION metabolic acidosis is mandatory to the survival of acute

Continuous hemofiltration procedures have been uremic patients and that this correction could be ade-found to be the method of choice in the treatment of quately accomplished by means of continuous renal re-critically ill patients with ARF (abstract; Marangoni et placement therapys (CRRTs). Because of continuous

and gentle application, no clinically untoward effectsal, Artif Organs 19:490, 1995).

Heering et al: Acid-base balance during CVVHS-40

DJ: Continuous high, venous-venous hemofiltration in acute renalseem to occur during the treatments that are linked to thefailure. Intensive Care Med 15:358–363, 1989

use of different buffers. The choice of the hemofiltration 4. Jacob AD, Elkins N, Reiss O, Chan L, Shapiro JI: Effects ofacetate on energy metabolism and function in the isolated perfusedreplacement fluid is essential for influencing acid-baserat heart. Kidney Int 52:755–760, 1997values. Although lactate- and bicarbonate-based re-

5. Feriani M, Dell’Aquila R: Acid-base balance and replacementplacement fluids lead to a remarkable elevation in serum solutions in continuous renal replacement therapies. Kidney Int

66(Suppl):156–159, 1998bicarbonate and arterial pH, acetate-based replacement6. Macias W: Choice of replacement fluid/dialysate anion in continu-fluid has been reported to compensate metabolic acidosis ous renal replacement therapy. Am J Kidney Dis 28(Suppl 3):S5–

insufficiently. An absence of an increase in serum bicar- S20, 19967. Hakim RM, Pontzer MA, Tilton D, Lazarus JM, Gottlieb M:bonate and arterial pH can be seen as a poor prognosis

Effects of acetate and bicarbonate dialysate in stable chronic dial-for the outcome in lactate- and bicarbonate-based but ysis patients. Kidney Int 28:535–540, 1985not in acetate-based hemofiltration. However, in patients 8. Heering P, Ivens K, Thumer O, Morgera S, Heintzen M, Pas-

slick-Reetjen J, Willers R, Strauber BE, Grabensee B: Thewith multiple-organ failure, the physiological buffer bi-use of different buffers during continuous filtration in critically illcarbonate is preferable because of its lack of metabolic patients with acute renal failure. Intensive Care Med (in press)

9. Thomas AN, Guy JM, Kishen R, Geraghty IF, Bowles BMJ,interference. Future studies are needed to prove that theVadgama P: Comparison of lactate and bicarbonate buffered he-application of bicarbonate-buffered hemofiltration fluidmofiltration fluids: Use in critically ill patients. Nephrol Dial Trans-

does not only improve acid-base metabolism, but also plant 12:1212–1217, 199710. Nimmo GR, Mackenzie SJ, Walker S, Grant IS: Acid base re-the overall prognosis of patients treated with CVVH.

sponse to high, hemofiltration in the critically ill patient. NephrolDial Transplant 8:854–857, 1993Reprint requests to Peter Heering, M.D., Department of Nephrology

11. Kierdorf H, Leue C, Heintz B, Riehl J, Melzer H, Sieberthand Rheumatology, Heinrich-Heine-University, 40001 Dusseldorf, Ger-HG: Continuous venovenous hemofiltration in acute renal failure:many.Is a bicarbonate or lactate buffered substitution better? in Continu-ous Extracorporal Treatment in Multiple Organ Dysfunction Syn-

REFERENCES drome. Contrib Nephrol, vol. 116, edited by Sieberth HG, Stumm-voll HK, Kierdorf H, Karger, 1995, pp 38–47

1. Agazia B, Ferrari G, Santoro A: Acid-base balance during hemo- 12. Santoro A, Ferrari G, Bolzani R, Spongano M, Zucceli P:filtration with bicarbonate substitution fluid: A prospective multi- Regulation of base balance in bicarbonate hemofiltration. Int Jcenter trial, in Istituto Italiano Per Gli Studi Filosofici Acid Base Artif Organs 17:27–36, 1994and Electrolyte Balance, edited by De Santo NG, Capasso G, 13. Locatelli F, Pontoriero G, Di Filippo S: Electrolyte disordersEditoriale Bios, Cosenza, 1995, pp 341–346 and substitution fluid in continuous renal replacement therapy.

2. Davenport A, Will EJ, Davidson AM: Improved cardiovascular Kidney Int 66(Suppl):151–155, 1998stability during continuous modes of renal replacement therapy in 14. Wakabayashi Y, Ohwada T, Kikawada R: Hemodialysis/hemo-critically ill patients with acute hepatic and renal failure. Crit Care filtration using sodium bicarbonate depresses cardiac function inMed 21:328–338, 1993 critically ill patients with acute renal failure. Jpn Circ J 58:81–86,

3. Wendon J, Smithies M, Sheppard M, Bullen K, Tinker J, Bihari 1994