Changes in Urinary Uric Acid Excretionin Obstructive Sleep Apnea Before andAfter Therapy With Nasal ContinuousPositive Airway Pressure*Hamid Sahebjami, MD, FCCP

Study objective: To assess the utility of urinary uric acid excretion as a marker of nocturnalhypoxia in patients with obstructive sleep apnea-hypopnea syndrome (OSAHS) before and afterthe institution of nasal continuous positive airway pressure (CPAP).Design: Prospective, open.Setting: Sleep Disorders Laboratory, Veterans Affairs Medical Center.Participants: Thirty consecutive male subjects, 20 with OSAHS and 10 without OSAHS.Measurements and methods: Spot morning urine and venous blood samples were obtained in allsubjects; samples were also obtained after the application of CPAP in those with OSAHS. Uricacid excretion, normalized to creatinine clearance, was calculated as the product of urinary uricacid and serum creatinine concentrations divided by urine creatinine concentration. In patientswith OSAHS, uric acid excretion was 0.55±0.1 mg/dL before CPAP therapy and decreased to0.30±0.01 mg/dL after CPAP therapy (p<0.001). The latter value did not differ significantly fromthe mean value (0.32 ±0.03 mg/dL) in the control group. Uric acid excretion in OSAHS patientscorrelated significantly with the apnea-hypopnea index (r=0.42; p< 0.0003).Conclusion: Uric acid excretion is increased in OSAHS patients and normalizes after CPAPtreatment, most likely reflecting differences in tissue oxygenation between the two conditions.Further studies in large number of patients may confirm the usefulness of this simple test fordiagnosis and follow-up of patients with OSAHS. (CHEST 1998; 113:1604-08)

Key words: anaerobic metabolism; sleep disorders; uric acid excretion

Abbreviations: ADP= adenosine diphosphate; AMP= adenosine monophosphate; ATP=adenosine triphosphate;CI=confidence interval; CPAP=continuous positive airway pressure; OSAHS=obstructive sleep apnea-hypopneasyndrome; Sa02=arterial saturation of oxygen

TP he integrity of cellular metabolic processes de--*- pends on the adequate supply of oxygen.oxy¬gen delivery.for the preservation and maintenanceof aerobic metabolism. Oxygen consumption by tis¬sues represents the aerobic production of adenosinetriphosphate (ATP), a crucial compound for main¬

taining cellular homeostasis. Under hypoxic condi¬tions, when oxygen supplies are inadequate to meet

oxygen demands of the cells, formation of ATP fromadenosine diphosphate (ADP) is impaired and a net

degradation of ATP to ADP and adenosine mono¬

phosphate (AMP) occurs. This leads to the release ofpurine nucleotide intermediates (adenosine, inosine,

*From the Pulmonary and Critical Care Section, Veterans AffairsMedical Center, and the Department of Medicine, University ofCincinnati College of Medicine, Cincinnati, Ohio.Manuscript received July 22, 1997; revision accepted November5, 1997.Reprint requests: Hamid Sahebjami, MD, FCCP, VA MedicalCenter, 3200 Vine St, Cincinnati, OH 45220

hypoxanthine, and xanthine) and the purine catabolicend product, uric acid.1 Elevated levels of ATPdegradation products in bodily fluids, therefore,represent a marker for cell energy crisis as a result ofcellular hypoxia.2 Increased levels of ATP degrada¬tion products, as a marker of ischemia and hypoxia,have been reported in isolated organs of animals andin human studies.3-6 Furthermore, clinical studieshave shown increased amounts of these products inneonates with infant respiratory distress syndrome,78during strenuous exercise,9 in patients with hypoten¬sion10 or acute respiratoiy failure,11 and in criticallyill patients.12Three studies have examined the utility of an

overnight change in urinary uric acid:creatinine ratioas a marker of tissue hypoxia in patients with ob¬structive sleep apnea-hypopnea syndrome (OSAHS)compared with control subjects.1315 Results of thesestudies have been conflicting and have revealed that

1604 Clinical Investigations

Downloaded From: http://journal.publications.chestnet.org/ on 09/13/2013

the overnight change in the urinary uric acid:creati-nine ratio has poor sensitivity in detecting nocturnalhypoxemia and does not correlate with disorderedbreathing events or associated desaturation. Further¬more, two sets of blood and urine samples.bothevening and morning.are necessary for the calcu¬lation of this ratio.A simple, convenient method for assessing urinary

uric acid excretion has been described and validatedin normal adult men and gouty patients.16 Accordingto this method, uric acid excretion is measured in

milligrams of urinary uric acid per deciliter of glo¬merular filtrate using spot urine and venous bloodsamples.16 In addition to its simplicity, this test ismore physiologic than the urinaiy uric acid:creati-nine ratio because uric acid excretion is normalizedto creatinine clearance.16The goal of the present preliminary study was to

assess the utility of urinary uric acid excretion as a

marker of hypoxia in patients with OSAHS beforeand after the institution of therapy with nasal con¬

tinuous positive airway pressure (CPAP) and to

compare it with the utility of the urinary uric acid:creatinine ratio obtained under similar conditions.

Materials and Methods

Patients

Thirty male subjects referred to the Sleep Disorders Labora¬tory at the Cincinnati Veterans Affairs Medical Center forpolysomnography were studied. Twenty men were found to haveOSAHS; the remaining 10 had no evidence of disordered breath¬ing and served as controls.

Measurements

Flow rates, lung volumes, and single-breath carbon monoxidediffusing capacity were determined using an automated system(Collins model GS/Plus; Warren E. Collins Inc; Rraintree, Mass).Recommendations for standardized procedure for lung functiontesting were followed.17 Predicted equations for pulmonaryfunction tests were from Knudson et al,18 Goldman and Reek-lake,19 and Gaensler and Smith.20

Arterial blood samples were drawn from the radial artery withthe patient in a sitting position while breathing room air. Arterialblood gas analysis was performed using the ARG-520 system(Radiometer American Inc; Westlake, Ohio).

Details of technique and scoring polysomnography in our sleeplaboratory have been published previously.21 For staging sleep,EEGs (two channels), chin electromyogram (one channel), andelectro-oculograms (two channels) were recorded. Thoracoab-dominal excursions and airflow were measured qualitatively.Arterial oxyhemoglobin saturation was recorded using an ear

oximeter (Riox IIA; RT, Inc; Roulder, Colo). These variableswere recorded on a multichannel polygraph (Model 78D; GrassInstrument; Quincy, Mass). Apnea was defined by the cessationof inspiratory airflow for at least 10 s in the presence of thoracicand abdominal excursions. A hypopnea was defined by a reduc¬tion of airflow for at least 10 s that was associated with at least a

4% reduction in overnight oxyhemoglobin desaturation or arous¬

al.21 The time spent below saturation of 88% was measuredmanually.A urine specimen and a venous blood sample were collected in

the morning after the completion of polysomnography. To some

ofthe urine specimens, 5% sodium hydroxide was added and pHwas adjusted to greater than 8. The urine was stored at 4°C.Urinary uric acid concentration was measured enzymaticallyaccording to the method described by Tietz.22 Sodium, creati¬nine, and uric acid concentrations were measured in the bloodspecimen and the remaining urine specimen, using standardtechniques. Uric acid excretion was calculated as the product ofurine uric acid and serum creatinine concentrations divided byurine creatinine concentration, all in mg/dL.16

Patients were told to avoid drinking caffeinated beveragesduring the hospital stay as a routine instruction for conductingsleep studies. They continued taking medications as prescribedby their physicians. Urine passed during polysomnography was

discarded. After the diagnosis of OSAHS was established duringthe first night of study, patients undenvent a second night ofpolysomnography with the application of adequate levels ofCPAP to eliminate apnea-hypopnea and maintain an arterialoxyhemoglobin saturation above 90%.

Statistical AnalysisFor each parameter measured or calculated, values for the

individual patients in each group were averaged, and the SEMwas calculated. Recause the common variance assumption re¬

quired by the t test was not appropriate for some of themeasurements, the Kruskal-Wallis nonparametric analysis ofvariance test was used to assess the significance of differencesamong the three studies. A p value of less than 0.05 was

considered significant with the significance adjusted by theRonferroni-Dunn method. We calculated 95% confidence inter¬vals (CIs) using the approximate degrees of freedom for the tstatistics. Kendall's nonparametric rank correlation coefficient ofvarious parameters of uric acid excretion, as the dependentvariable, was computed.23

Results

Patients and the control group did not differsignificantly in age and in body mass index (Table 1).Among pulmonary function tests and arterial bloodgases, only total lung capacity was significantly dif¬ferent in the two groups (Table 1).

Polysomnographic parameters (Table 2) showedthat in patients with OSAHS the apnea-hypopneaindex was 53±6 episodes per h (CI, 39.2 to 66.9episodes per h), which was associated with severe

arterial oxyhemoglobin desaturation. During the sec¬

ond study night with CPAP, all polysomnographicparameters improved significantly in patients withOSAHS and were similar to those in the controlgroup (Table 2).

Results of blood and urine tests are shown in Table3. In patients with OSAHS, fractional excretion ofsodium and uric acid excretion were significantlyhigher after the first night of study compared withthe values obtained after the second night withCPAP and compared with the control group's values.

CHEST/113/6/JUNE, 1S 1605

Downloaded From: http://journal.publications.chestnet.org/ on 09/13/2013

Table 1.Age, Body Mass Index, Pulmonary FunctionTests, and Arterial Blood Gases*

VariablesControls(n=10)

Patients(n=20)

Age, yr 57.7±3.3 55.1+2.8Body mass index, kg/m2 33.9±2.9 39.6±1.9FVC, % predicted 84.8±5.0 77.1+3.1FEVT, % predicted 78.1±6.5 75.2±3.5FEV/FVC, % 73.9±3.5 78.6±1.5Total lung capacity, % predicted 96.8±2.8f 85.7±3.2Residual volume, % predicted 111.0+9.0 100.8±8.5Residual volume/total lung capacity, 109.6±7.7 109.1 ±5.5% predicted

Diffusion of carbon monoxide, 87.9±8.7 88.4±5.7% predicted

Pa02, mm Hg 78.7±1.9 72.4±2.0PaC02, mmHg 39.1±0.5 42.5±1.1

pH 7.41±0.005 7.41±0.006*Values are means± SEM.

fp<0.05.

Mean uric acid excretion was 0.32±0.03 mg/dL (CI,0.24 to 0.40 mg/dL) in the control group, 0.55±0.1mg/dL (CI, 0.27 to 0.82 mg/dL) in patients beforeCPAP, and 0.30±0.01 mg/dL (CI, 0.26 to 0.34mg/dL) in patients after CPAP therapy. Figure 1shows individual values for uric acid excretion in

patients and in the control group. One patient had a

very high value before CPAP therapy, resulting in a

large variation in this group; the inclusion or exclu¬sion of this patient did not alter the statistical profileofthe results. A significant correlation existed for allpatients between uric acid excretion as the depen¬dent variable and the apnea-hypopnea index

Table 2.Polysomnography Data in Controls and inPatients With OSHAS Before and After Treatment

With Nasal CPAP*

Patients (n=20)

VariablesControls(n=10)

ReforeCPAP

AfterCPAP

Total dark time, minTotal sleep time, min

Sleep efficiency, %Baseline Sa02,f %Lowest SaOa during REM

sleep, %Lowest Sa02 during non-

REM sleep, %Apnea-hypopnea index,

episodes/hSa02<88%, min

Sa02<88%, % total sleeptime

38631183:95:91:

90

3+ 15:3:0.5:0.5

0.6

:3

380±3285±1076±293 ±0.465±4+

66±5*

53±6j

108±24*38±8*

390 ±6329±1182±393 ±0.490±0.3

90±0.2

1±0.6

4±41±1

*Data presented as mean±SEM.fSa02=arterial saturation of oxygen.*p<0.001 vs other groups.

(r=0.42; p<0.0003) and percent total sleep timewith saturation less than 88% (r=0.32; p<0.01). Theurinary uric acid:creatinine ratio was 27.2±2.8% (CI,20.8 to 33.6%) in the control group, 50.1±13% (CI,21.3 to 78.9%) in patients before CPAP, and26.9±2.2% (CI, 22.3 to 31.5%) in patients afterCPAP therapy. The urinary uric acid:creatinine ratio,as the dependent variable, correlated significantlywith apnea-hypopnea index (r=0.34; p<0.003) andpercent total sleep time with saturation less than88% (r=0.28; p<0.02).

Discussion

The results of this study revealed that in patientswith OSAHS uric acid excretion was significantlyhigher before an overnight application of CPAP thanafter CPAP. Although there was a considerableoverlap of individual values of uric acid excretion incontrol subjects and in OSAHS patients beforeCPAP therapy, in all but one subject the uric acidexcretion fell following the institution of CPAP (Fig1). Since the major difference between the two

nights of study was the correction of hypoxemia byCPAP, it can be concluded that the change in uricacid excretion was most likely the result of differ¬ences in oxygenation during the two nights. How¬ever, numerous hormonal and sympathetic nervous

system changes occur during the night in untreatedOSAHS patients, which might have influenced uri¬nary uric acid excretion independent of and inaddition to tissue hypoxia.2427Three published reports have examined overnight

changes in the uric acid:creatinine ratio in OSAHS.In two of these, a significant increase in the uricacid:creatinine ratio was observed in association withnocturnal hypoxemia; CPAP treatment led to a

significant reduction in this ratio.1314 The otherstudy failed to confirm this observation.15 Collec¬tively, these studies showed a substantial overlap andhigh variability of urinary uric acid:creatinine ratiosamong patients with OSAHS. This was reflected in a

substantial number of false-negative findings of over¬night change in this ratio and in the absence of anysignificant correlation between this ratio and variousparameters of desaturation.

Thus, we chose to test the utility of urinary uricacid excretion, normalized to creatinine clearance, inpatients with OSAHS before and after CPAP ther¬apy. This is a more physiologic expression than theurinary uric acid:creatinine ratio because it correctsfor the functional renal mass of the individual.16Furthermore, it obviates the need for an additionalevening collection necessary for the calculation ofovernight change in urinary uric acid:creatinine ra-

1606 Clinical Investigations

Downloaded From: http://journal.publications.chestnet.org/ on 09/13/2013

Table 3.Blood and Urine Tests in Controls and in Patients With OSAHS Before and After Treatment With NasalCPAP*

VariablesControl(n=10)

Patients (n=20)Before CPAP After CPAP

Serum creatinine, mg/dLSerum sodium, mg/dLSerum uric acid, mg/dLUrinary creatinine, mg/dLUrinary sodium, mg/dLUrinary uric acid, mg/dLFractional excretion of sodium,Uric acid excretion, mg/dLUrinary uric acid:creatinine, %

1.2±0.6140.0±1.3

7.2±0.6145.2±1895.7±1840.4±70.53±0.090.32±0.0327.2±2.8

1.2±0.07140.2±0.7

6.6±0.2101.7±13125.3±1038.7±51.4±0.1f

0.55±0.1§so.^m1

1.1±0.07140.2±0.4

6.8±0.3154.9±2098.3±1042.7±60.73±0.10.30±0.0126.9±2.2

*Data presented as mean±SEM.

fp<0.01 vs control and p<0.05 vs after CPAP.Ip<0.05 vs control and p<0.01 vs after CPAP.

tio.1315 It is more convenient, less costly, and more

physiologic, and therefore may prove to be more

useful.In 19 of our 20 patients (95%), uric acid excretion

was higher before the application of CPAP than itwas after CPAP. False-negative results were found in

only 1 of 20 patients (5%) with OSAHS, comparedwith previous reports of 17% and 30% false negativ¬ity.1314 Importantly, the mean value of uric acidexcretion in our study was similar in the control

3.0

0.6

0.5 h

5E,co

£ 0.4oXLU2 0.3o<o"5 0.2

0.1

0.0

P <0.05 P <0.01

_L _LControl Before CPAP After CPAP

Figure 1. Uric acid excretion in 10 control subjects and 20patients with OSAHS before and after therapy with nasal CPAP.Shaded areas represent mean±SEM.

group (0.32 ±0.03 mg/dL) and in patients(0.30±0.01 mg/dL) after CPAP therapy (Table 3),suggesting that CPAP normalized uric acid excre¬

tion. We also compared changes in the urinary uricacid:creatinine ratio in the control group and in

patients before and after CPAP therapy. This ratiowas significantly higher before than after the institu¬tion of CPAP therapy (Table 3), confirming theresults of previous studies.1314

If the increase in uric acid excretion is a marker forcellular hypoxia, arterial oxyhemoglobin desaturationalone may not reflect its magnitude because arterialoxyhemoglobin saturation represents only one com¬

ponent of the oxygen delivery system. Of the othertwo major components, cardiac output and hemoglo¬bin concentration, the latter was similar during thetwo study nights; the status of cardiac output duringthe two nights of study, however, could have beenvariable. Episodes of apnea-hypopnea and resultanthypoxemia and hypercapnia during the first studynight, as well as application of CPAP during thesecond study night, could influence cardiac outputleading to variable changes in oxygen delivery.2830 Itis interesting that in the present study, uric acidexcretion best correlated with the apnea-hypopneaindex (r=0.42; p<0.0003) rather than desaturation(r-0.32; p<0.01).The results of this study also confirm increased

nocturnal natriuresis in OSAHS patients and itsnormalization after CPAP therapy (Table 3).31 Na¬triuresis in these patients is in part due to a nocturnalincrease in plasma levels of atrial natriuretic factors,which decrease after treatment with CPAP.3132 In¬creased negative intrathoracic pressure and pulmo¬nary vasoconstriction during sleep, with a resultantincrease in right atrial pressure, are likely causes ofenhanced secretion of atrial natriuretic factor inOSAHS.3334

CHEST/113/6/JUNE, 1£ 1607

Downloaded From: http://journal.publications.chestnet.org/ on 09/13/2013

The usefulness of measuring uric acid excretion inpatients with OSAHS lies more in outpatient man¬

agement than in diagnosis. With respect to diagnosis,future studies of large numbers of subjects are

needed to perhaps define a threshold. However,once the level of uric acid excretion has beendetermined in a given patient with OSAHS afterCPAP treatment, the measurement can be repeatedperiodically to determine the success of long-termCPAP therapy or to assess patient compliance. Thiswould be useful particularly for the management ofpatients who do not have easy access to health-carefacilities and sleep laboratories. However, additionalresearch is needed before any guidelines can beestablished.

ACKNOWLEDGMENT: The author thanks Dr. Shahrokh Java-heri for his assistance in the performance and interpretation ofsleep studies.

References1 Fox IH. Metabolic basis for disorders of purine nucleotide

degradation. Metabolism 1981; 30:616-342 Fox IH, Pallela TD, Kelly WN. Hyperuricemia: a marker for

cell energy crisis. New Engl J Med 1987; 317:111-123 Van Rilsen M, van der Vusse GJ, Coumans WA, et al.

Degradation of adenine nucleotides in ischemic and reper-fused rat heart. Am J Physiol 1989; 257:H47-54

4 Siems W, Schmidt H, Muller M, et al. H202 formation duringnucleotide degradation in the hypoxic rat liver: a quantitativeapproach. Free Radic Res Commun 1986; 1:289-95

5 Fox AC, Reed GE, Glassman E, et al. Release of adenosinefrom human hearts during angina induced by rapid atrialpacing. J Clin Invest 1974; 53:1447-57

6 Fox AC, Reed GE, Meilmam H, et al. Release of nucleosidesfrom canine and human hearts as an index of prior ischemia.Am J Cardiol 1979; 43:52-58

7 Jensen MH, Rrinklov MM, Lillquist K. Urinary loss ofoxypurines in hypoxic premature neonates. Riol Neonate1980; 38:40-48

8 Jensen MH. The catabolism of purine nucleotides in thehuman organism [thesis]. Institute of Anaesthesiology,Odense University. Odense, Denmark: CAVI Rogtrykkeri,1986

9 Ketai LH, Simon RH, Kreit JW, et al. Plasma hypoxanthineand exercise. Am Rev Respir Dis 1987; 136:98-101

10 Woolliscroft JO, Fox IH. Increased body fluid purine levelsduring hypotensive events: evidence for ATP degradation.Am J Med 1986; 81:472-78

11 Christensen EF, Jacobsen J, Anker-Moller E, et al. Increasedurinary loss of uric acid in adults with acute respiratory failurerequiring mechanical ventilation. Chest 1992; 102:556-59

12 Grum CM, Simon RH, Dantzker DR, et al. Evidence foradenosine triphosphate degradation in critically-ill patients.Chest 1985; 88:763-67

13 Hasday JD, Grum CM. Nocturnal increase of urinary uricacid:creatinine ratio: a biochemical correlate of sleep-associ¬ated hypoxemia. Am Rev Respir Dis 1987; 135:534-38

14 Rraghiroli A, Sacco C, Erbetta M, et al. Overnight urinaryuric acid:creatinine ratio for detection of sleep hypoxemia:

validation study in chronic obstructive pulmonary disease andobstructive sleep apnea before and after treatment with nasalcontinuous positive airway pressure. Am Rev Respir Dis1993; 148:173-78

15 McKeon JL, Saunders NA, Murree-Allen K, et al. Urinaryuric acid:creatinine ratio, serum erythropoietin, and blood2,3-diphosphoglycerate in patients with obstructive sleepapnea. Am Rev Respir Dis 1990; 142:8-13

16 Simkin PA, Hoover PL, Paxson CS, et al. Uric acid excretion:quantitative assessment from spot, midmornmg serum andurine samples. Ann Intern Med 1979; 91:44-47

17 Ferris BG. Recommended standardized procedures for pul¬monary function testing. Am Rev Respir Dis 1978; 118:55-88

18 Knudson RJ, Lebowitz MD, Holberg CJ, et al. Changes in thenormal maximal expiratory flow volume curve with growthand aging. Am Rev Respir Dis 1983; 127:725-34

19 Goldman HI, Recklake MR. Respiratoiy function tests: nor¬

mal values at median altitudes and the prediction of normalresults. Am Rev Tuberculosis 1959; 79:457-67

20 Gaensler ISA, Smith AA. Attachment for automated singlebreath diffusing capacity measurement. Chest 1973; 63:136-45

21 Javaheri S, Parker TJ, Wexler L, et al. Occult sleep disorderedbreathing in stable congestive heart failure. Ann Intern Med1995; 122:487-92

22 Tietz NW. Clinical guide to laboratory tests. 2nd ed. Phila¬delphia: W.B. Saunders, 1990; 568

23 Snedecor GW7, Cochran WC. Statistical methods. 7th ed.Ames, Iowa: Iowa State University Press, 1980; 193-95

24 Hedner J, Darpo R, Ejnel H, et al. Reduction in sympatheticactivity after long-term CPAP treatment in sleep apnea:cardiovascular implications. Eur Respir J 1995; 8:222-29

25 Somers VK, Dyken ME, Clary MP, et al. Sympathetic neuralmechanisms in obstructive sleep apnea. J Clin Invest 1995;96:1897-904

26 Smith ML, Niedermaier ON, Hardy SM, et al. Role ofhypoxemia in sleep apnea-induced sympathoexcitation. J Au-ton Nerv Syst 1996; 56:184-90

27 Dimsdale JE, Coy T, Anconi-Israel S, et al. Sympatheticnervous system alterations in sleep apnea: the relative impor¬tance of respiratory disturbances, hypoxia, and sleep quality.Chest 1997; 111:639-42

28 Montner PK, Greene R, Murata GH, et al. Hemodynamiceffects of nasal and mask continuous positive airway pressure.Am J Respir Crit Care Med 1994; 149:1614-18

29 Schroeder JS, Motta J, Guilleminault C. Hemodynamic stud¬ies in sleep apnea. In: Guilleminault C, Dement WC, eds.Sleep apnea syndrome. New York: Alan R. Liss, 1978; 177-90

30 Guilleminault C, Motta J, Mihm F, et al. Obstructive sleepapnea and cardiac index. Chest 1986; 89:331-34

31 Krieger J, Laks L, Grunstein RR, et al. Atrial natriureticpeptide release during sleep in patients with obstructive sleepapnea before and during treatment with nasal continuouspositive airway pressure. Clin Sci 1989; 77:407-11

32 Lin CC, Tsan KW, Lin CY. Plasma levels of atrial natriureticfactor in moderate to severe obstructive sleep apnea syn¬drome. Sleep 1993; 16:37-39

33 Edwards BS, Zimmerman RS, Schwab TR, et al. Atrialstretch, not pressure, is the principal determinant controllingthe acute release of atrial natriuretic factor. Circ Res 1988;62:191-95

34 Lew RA, Raertschi A. Mechanisms of hypoxia-induced atrialnatriuretic factor release from rat heart. Am J Physiol 1989;257.H147-H156

1608 Clinical Investigations

Downloaded From: http://journal.publications.chestnet.org/ on 09/13/2013