Urinary Lactate Dehydrogenase Isoenzym e Analysis in Adult Population*

ANNALS O F CLINICAL A N D LABORATORY SC IE N C E, Vol. 15, No. 1 Copyright 1985, Institute for Clinical Science, Inc. Urinary Lactate Dehydrogenase Isoenzym e Analysis in Adult Population* TSIEH SUN, M.D.,t CHRISTO PHER CHOW,* M ELINDA MCVICAR, M.D., and LIO N EL MAILLOUX, M.D. fdepartment of Laboratories, Children s Kidney Center, Department of Pediatrics, Division of Nephrology, Department. North Shore University Hospital, Manhasset, NY 11030 and fdepartment of Pathology, Department o f Pediatrics, and Department of Medicine, Cornell University Medical College, New York, NY 10021 and fmassachusetts Institute of Technology, Boston, MA 02139 ABSTRACT This investigation was a systemic study on an adult population of urinary lactate d ehydro g enase (LD H ) isoenzym e analysis for th e distinction betw een upper and lower urinary tract infections. The study included 160 urine sam ples from patients and healthy individuals. On the basis of clinical symptoms, urinary bacterial colony counts, renal function tests and radiologic findings, the adults w ere divided into (1) pyelonephritis group, (2) cystitis group, (3) pelvic lesion group, and (4) control group. This technique correctly identified 23 of 26 patients with pyelonephritis by the presence of elevated LDH-V (over 10 percent) and all of 12 patients with cystitis by the presence of elevated LD H -I (over 60 relative units) but low LDH-V (below 10 percent or lower than LDH-I). In the pelvic group, the results of eight patients were consistent with cystitis and four w ith pyelonephritis. O ur study confirms the sensitivity and specificity of the LD H isoenzyme technique for the differential diagnosis of urinary tract infection on adult patients and is consistent w ith previous studies on pediatric patients. However, one should be cautious to interpret the results of LD H isoenzymogram before extra-urinary tract lesions are excluded. Introduction clinical problem which needs laboratory assistance. T hese two clinical e n titie s The differentiation betw een upper and may not have d istin g u ish ed clinical lower urinary tract infections has been a symptoms and signs or they may even be I,,,,,,,, asym ptom atic. How ever, th eir distinc- * Reprint requests should be addressed to Tsieh,.. i,, * Sun, M.D., D epartm ent o f Laboratories, North Shore University Hospital, Manhasset, NY 11030. tlon ls m andatory as th e m anagem ent and prognosis of th ese two groups are 32 0091-7370/85/0100-0032 $01.20 Institute for Clinical Science, Inc.

URINARY LDH ISOENZYMES 33 obviously different. For instance, cases of cystitis (lower urinary tract infection) respond well to single-dose antimicrobial therapy, w hereas pyelonephritis (upper urinary tract infection) is a more serious clinical p ro b lem req u irin g in ten siv e treatm ent.8 To help solve this problem, m any laboratory procedures have been devised to locate th e site of in fe c tio n.17 Am ong direct m ethods, th ere are culture of kidney tis s u e,416 c u ltu re of u re th e ra l u rin e,23 and F airley s bladder-w ashout technique.10 These m ethods, while generally reliab le, are tim e consum ing, potentially hazardous, and costly.17 Indirect m ethods include the assessment of maximal urinary concentrating ability,6,22 urinary enzym e assays,12 20 seru m a n tibody against th e infecting o rg an ism,2 leukocyte excretion r a te,9 and d e te c tion of antibody coating b a cteria in u rin e.13'14,24 These tests, while sim ple, non-invasive and cost-effective, have been questioned as to th eir accuracy.117 A relatively recent developm ent is the analysis of urinary lactate dehydrogenase (LD H ) isoenzym es for th e differential diagnosis betw een upper and lower urinary tract infections.5 This inexpensive technique is generally available in clinical laboratories, and its sensitivity and specificity have b e e n w ell estab lish ed in c h ild ren by several clinical s tu d ies. 1 3 7 17 18 How ever, the usefulness of this technique has rarely been studied in the ad u lt p o p u la tio n.1718 T he p re se n t study is to evaluate w hether or not the conclusion based on studies of pediatric cases is also valid in adult patients. Materials and Methods The present study included 160 urine samples from patients and healthy individuals consisting of 81 m ales and 79 females betw een 18 and 89 years of age. The grouping of patients was essentially based on the criteria used by Devaskar and M ontgom ery7 with the exception of the pelvic lesion group which is an additional group that has not been included in any previous studies of th e sam e nature. Pyelonephritis G roup There were 26 patients in this group. All patients had urinary bacterial colony counts exceeding 100,000 per ml plus at least one of the following diagnostic criteria: (1) related clinical symptoms: fever, chills, vom iting, costovertebral te n d e r ness, or loin pain; (2) radiologic abnormalities: staghorn calculi, vesicoureteral reflux, cortical scarring, calyceal dilatation or distortion; (3) inability to concentrate urine above 700 Osm per L after prolo n g ed w ater dep riv atio n ; and (4) im pairm ent of renal function: elevated creatinine and blood urea nitrogen levels or abnorm al creatinine clearance test. Cystitis G roup This group was com posed of 12 patients, all of whom had urinary bacterial colony counts exceeding 100,000 per ml but who lacked o th er criteria of kidney infection, as listed in th e pyelonephritis group. Pelvic Lesion G roup This group consisted of 37 p atien ts who had pelvic lesions (prostatic carcinoma, benign prostatic hypertropy, ovarian carcinom a, cervical carcinom a, p reg nancy, and pelvic inflammatory disease) but who lacked evidence of urinary tract infection. Control G roup This group com prised 85 healthy persons. All had no history or p rese n t evidence of urinary tract infection or p re existing renal abnormality.

34 SUN, CHOW, MCVICAR, AND MAILLOUX Random urine samples were collected in m orning hours and LD H isoenzyme assays were perform ed within two hours after collection. D u rin g th e in terim, u rin e sam ples w ere c o n c en tra te d 50 times by ultrafiltration in a Minicon B 15 concentrator.* E lectrophoresis was carried out on agarose films for 30 m inutes in 0.065 M barbital buffer at a ph of 8.6. After electrophoresis, a m ixture of substrate and coloring reagentf was applied onto the agarose film, and L D H isoenzymes w ere then dem onstrated by the stained bands based on the principle of the form azan m ethod. Q uantitation of LD H isoenzymes was accomplished by scanning the p a tte rn with a densitom e ter:]: utilizing a 520 nm filter. The fraction that m igrated fastest to the anode was d esig n ated isoenzym e I (L D H -I); the slowest, isoenzym e V (LDH-V). Comparison of the scans was based on two d ifferen t m easu rem en ts. As the p resen ce of L D H isoenzym e V was always accom panied by the other four fractions, th e com parisons of LD H -V w ere based on th e p e rc en tag e of this isoenzyme in the total activity. On the other hand, isoenzyme I was frequently present alone; calculation of its percentage was found to be deceiving. Since all samples w ere concentrated to the same magnitude and all slides were scanned at th e sam e sensitivity, fraction I curves were, therefore, compared on the basis of the area under the curve. M easurements of dimensions of the curves were taken u n d e r a dissectin g m icroscope. Though the curves were obviously not triangles, th e form ula for area of a triangle (V2 height x width of the base) was found to be adequate for relative com parisons. Area units will be referred to as relative units (RU s). Results Pyelonephritis Group In this group, total urinary LD H activity was generally high as judged by gross inspection of the patterns. All but three of the 26 cases had LDH-V higher than 10 percent, ranging from 11 percent to 39 percent of the total activity (table I and figure 1). T he rem aining th re e showed 0 percent, 3 percent, and 4 percent of LDH-V, respectively. LD H -I was also m ore prom inant in this group than in the control group with 17 cases showing a peak greater than 60 RU, a cutoff point derived from the study of th e cystitis group (table I). Cystitis G roup In this group, total urinary LD H activity was also high. All of the 12 cases show ed a p red o m in an t L D H -I w ith activity over 60 RU, ranging from 65.5 to 248 RU (table I and figure 2). The m edian was 141.7 RU. LD H -II was frequently present but usually at a lower level than LD H -I. LDH-V was dem onstrated in six cases, but only two were over 10 percent (13 percent and 23 per- TABLE I C o m p a riso n o f L a c ta t e D e h y d ro g e n a se Isoenzym e A n aly ses in Four S tu dy Groups P y e lo n e - p h r i t i s G roup C y s t i t i s Group P e l v i c L e s io n Group C o n tr o l Group LDH-V* P o s i t i v e 23 2 4 0 N e g a tiv e 3 10 33 85 T o t a l 26 12 37 85 L D H -It P o s i t i v e 17 12 8 0 N e g a tiv e 9 0 29 85 T o t a l 26 12 37 85 * Amicon. t Corning, t Helena. A c t i v i t y g r e a t e r t h a n 10 p e r c e n t i s c o n s id e r e d p o s i t i v e. t A c t i v i t y g r e a t e r t h a n 60 r e l a t i v e u n i t s (R U 's) i s c o n s id e r e d p o s i t i v e.

35 URINARY LDH ISOENZYMES sitivity of using L D H -I for the diagnosis of cystitis is 100 percent, and the speci ficity 94 percen t w hen the 17 cases of pyelonephritis are excluded as false-pos itive. These 17 cases w ere diagnosed as pyelonephritis and not cystitis because of the presence of five L D H fractions with a prom inent LDH-V. 13 Discussion F ig u r e 1: A case m a r k e d in c r e a s e in o f p y e lo n e p h r itis LD H -V s h o w in g a n d t h e p r e s e n c e o f a ll fiv e f r a c tio n s in is o e n z y m o g r a m a n d tra c in g. cent). Even in these two cases, L D H -I was far m ore prom inant than LDH-V. Pelvic Lesion Group In th is g ro u p, 20 o f th e 37 cases show ed no distin g u ish ab le peaks. T he re m a in in g 17 cases re v e a le d L D H -I a c tiv ity w ith e ig h t cases o v e r 60 RU. F our cases dem onstrated a LD H -V level over 10 percent (11 percent, 12 percent, 14 percent, and 24 percent) (table I). Urinary L D H isoenzym e analysis was first used by Carvajal et al5 for the local ization of the site of urinary tract infec tions.5 T heir study revealed that the ele vated LD H -V correctly identified 15 out of 16 c h ild re n w ith p y e lo n e p h ritis, w h e re as L D H -V was a b s e n t in all 22 patients with cystitis. Although this con c lusion was c h a lle n g e d by a few re p o rts,115 m ost of the subsequent stud ies c o n firm e d th e re lia b ility of L D H a n a ly sis.3,717 18 B ro u h a rd a n d C u n ningham,3 for instance, rep o rted that 97 to 100 percen t of the u p p e r urinary tract infections w ere accurately localized by this analysis. C a rv a ja l s orig in al stu d y in c lu d e d q u an titatio n of total L D H.5 H ow ever, I I Control Group In th is g ro u p, 61 of th e 85 cases show ed no distin g u ish ab le peaks. T he rem aining 24 cases revealed low L D H -I and L D H -II activities; the highest being 37.5 RU. These cases also dem onstrated insignificant levels of LD H -V (table I and figure 1). The X2 test, including correction for continuity, shows statistical significance of th e association b e tw e en te st results and the clinical classification of subjects. The sensitivity of using LD H -V for the diagnosis of pyelonephritis is 88 percent and the specificity 96 percent. The sen F ig u r e 2 : L DH -I and A case LD H -II o f c y s titis s h o w in g p r o m in e n t w ith th e p re s e n c e o f in is o e n z y m o g ra m a n d tra c in g. LDH -III

36 SUN, CHOW, MCVICAR, AND MAILLOUX L orentz and R esnick18 co n sid ered it unnecessary, as did D evaskar and M ontgomery.7 The form er authors m aintained th at calculation of th e p e rc en tag e of LDH-V was sufficient for the purpose of differential diagnosis, and 10 percent was used as th e lev el w hich d istin g u ish ed pyelonephritis from cystitis.18 By utilizing the cutoff point of 10 percent for LDH-V, 23 of our 26 patients w ith p y e lo n e p h ritis w ere accurately identified. At the same time, when 60 RU was used as a cutoff point for LD H- I in combination with the quantitation of LDH-V, all of 12 patients with cystitis were correctly localized. As a m atter of fact, a gross inspection of the LD H isoenzymogram is usually adequate to distinguish u p p e r and low er u rinary tract infections u n d er m ost circum stances: a predom inant L D H -I is consistent with cystitis, w hile a p ro m in e n t L D H -V is indicative of pyelonephritis. The calculation of RU for LD H -I is only helpful when a borderline case betw een normal and cystitis is encountered. The kidney tissu e contains all five LD H isoenzymes. The cortex is high in LD H -I-III, while the m edulla LDH-V.25 As p y e lo n e p h ritis involves m ainly the m edulla, th a t explains why LD H -V is p red o m in an t in cases of this disease. LDH-V is released from necrotic renal tu b u le s.1,5,21 The m echanism of elevations of L D H -I in cystitis is not clear. Normal urinary bladder tissue contains only five percent of L D H -I.11 However, in normal urine, LD H -I is the predom inant fraction.517 T herefore, the elevation of LD H -I in cystitis may m erely represent exaggeration of the normal pattern owing to the in crease of total L D H. W hen cystitis subsides, urinary L D H returns to norm al w ithin two or three w eek s.5 O n th e o th e r hand, the inflam m ed bladder may release disproportionally higher L D H -I than norm al tissue. A com parable exam ple can be seen in the elevation of CPK-II isoenzyme in myositis while normal skeletal muscle does not contain CPK-II at all.11 H ow ever, one should keep in m ind that lysis of leukocytes in urine may also cause elevation of LD H -V,1,19 and when urinary ph is below 5.5, LDH-V dim inishes rapidly.1 These factors may account for the two cases of cystitis with raised LD H-V and the th re e cases of pyelonephritis with normal LDH-V. W hen fresh urine sample is examined, the chance of lysis of leukocytes and change in urinary ph is greatly minimized. Several studies showed no clear-cut correlation betw een p y u ria and e n zy m u ria.5,7 In th e study conducted by D evaskar and M ontgom ery,7 patients from both cystitis and pyelonephritis groups had significant pyuria, but the urinary LD H patterns were distinctly different in these two groups of patients. Leukocytes contain 5,000 mu of LD H per 108 cells, while erythrocytes contain 240 m U p e r 108 c e lls.21 Since norm al urine contains 13 mu LD H per ml, it requires the com plete lysis of 2.5 X 105 leukocytes p e r ml (about 50 cells p e r high pow er field) or lysis of 5 X 10 erythrocytes per ml (about 1,000 erythrocytes per high power field) to double the L D H activity in urine.21 From this calculation, it seems that the influence of leukocytes and erythrocytes to urinary LD H is very lim ited, except in unfresh specim ens of p y u ria and w h ere large num bers of cells are lysed. O ur inclusion of a pelvic lesion group in our study reveals some new problems. The basic pathology in this group is nonurinary, but urine samples are frequently received from patients of this group and the results of LD H isoenzyme analysis are confused with those of the urinary groups. In our study, for instance, the results of eight cases of the pelvic lesion group w ere consistent with cystitis while four w ere consistent with pyelonephritis. However, all of the 12 patients had no clinical sym ptom s of urinary tract infec

URINARY LDH ISOENZYMES 37 tion, and their urinary bacterial colony counts w ere below 100,000 p er ml. It seem s likely th at pelvic lesions m ight have caused obstruction of the urinary tract or disturbances of the blood circulation of the urinary system. In the case of carcinom a, tu m o r m etastasis m ight have involved the tissues of kidney or u rin ary b lad d e r directly. As a resu lt, L D H isoenzym es from th ese organs w ere released into urine. False positive results could thus be obtained when a pelvic lesion is present. Therefore, when the results of LD H isoenzymogram are not consistent w ith clinical and m icrobiological examinations, clinicians should look for pelvic lesions to rule out the possibility of false positive resu lts. O bviously, the release of LD H is not lim ited to inflam m atory condition b u t also occurs in o th er destructive processes, such as tumors or obstruction of urinary flow. No m atter what the basic mechanism is, our study indicates that L D H isoenzyme analysis is a reliable test to distinguish u p p e r and low er u rin ary tract infection when infection is confirmed by clinical and microbiological observations. Acknowledgments Thanks are extended to Robert Brody, statistician of North Shore University Hospital, for help in the statistical work, and York Lien, Ph.D. for technical guidance. References 1. A p p e l m e l k, B. J. and M a c L a r e n, D. M. : Localization of urinary tract infection with urinary lactic dehydrogenase isoenzyme 5. Lancet i: 1417-1418, 1981. 2. B r e n n e n, D. A., F a i r l e y, K. F., O K e e f e, C., et al: The serum antibody response in renal and bladder infections. Med. J. Aust. i:1069-1076, 1969. 3. B r o u h a r d, B. H. and C u n n i n g h a m, R. J. : Single-dose antibiotics for urinary infections. Lancet t : 3 3 1, 1 9 8 1. 4. B r u n, C., R a s c h o n, F., and E r i k s o n, K. R. : Simultaneous bacteriologic studies of renal biopsies and urine. Progress in Pyelonephritis. Kass, E., ed. Philadelphia, F. A. Davis Co., 1965, pp. 461-465. 5. C a r v a j a l, H. F., P a s s e y, R. U., B e r g e r, M., et al: Urinary lactic dehydrogenase isoenzym e 5 in the differential diagnosis of kidney and bladder infections. Kidney Int. 8: ] 76-184, 1975. 6. C l a r k, H., R o n a l d, A. R., and C u t l e r, R. E., et al: The correlation between the site of infection and maximal concentrating ability in bacteriuria. J. Infect. Dis. 120:47-53, 1969. 7. D e v a s k a r, U. and M o n t g o m e r y, W.: Urinary lactic dehydrogenase isoenzym e IV and V in the differential diagnosis of cystitis and pyelonephritis. J. Pediatr. 93:789-791, 1978. 8. E d i t o r i a l : Single-dose treatment of urinary tract infections. Lancet i:26, 1981. 9. F a i r l e y, K. F : Leukocyte excretion rate as a screening test for bacteriuria. Lancet :420, 1967. 10. F a i r l e y, K. F., B o n d, A. G., and B r o w n, R. B. et al: Simple test to determ ine the site of urinary tract infection. Lancet ii:427, 1965. 11. G a l e n, R. S. : Isoenzym es and myocardial infarction. Diag. Med. i:4 0-5 2, 1977. 12. G o l d b e r g, W. M., C h a k r a b a r t i, S., and F i l i p i c h, R. : Urinary lactic dehydrogenase, alkaline phosphatase and lysozyme studies in renal disease. Canad. Med. Assoc. J. 94:1265-1268, 1966. 13. H e l l e r s t e i n, S., K e n n e d y, E., N u s s b a u m, L., et al: Localization of the site o f urinary tract infections by means of antibody-coated bacteria in the urinary sediments. J. Pediatr. 92:188-193, 1978. 14. J o n e s, S. R., S m i t h, J. W., and S a n f o r d, J. P.: Localization of urinary tract infections by detection of antibody-coated bacteria in urine sediment. N. Engl. J. Med. 290:591-593, 1974. 15. K e n n e d y, E. D., S h a r m a, P., G r o s s m a n, H. M., et al: Determinants of urinary lactic dehydrogenase (LDH) activity in girls with urinary tract infections. Pediatr. Res. J3:515, 1979. 16. K i p n i s, G. P., J a c k s o n, G. G., D a l l e n b a c h, F. D., et al: Renal biopsy in pyelonephritis: Correlative study of kidney morphology, bacteriology and function in patients with chronic urinary tract infections. Arch. Intern. Med. 95:445-459, 1965. 17. L o r e n t z, W. B. : Localization o f urinary tract infection. Urol. Clin. North Amer. 6:519-527, 1979. 18. L o r e n t z, W. B. and R e s n i c k, M. I.: Comparison of urinary lactic dehydrogenase with antibodycoated bacteria in the urine sedim ent as means of localizing the site of urinary tract infection. Pediatrics 64:672-677, 1979. 19. M a l i k, G. M., C a n a w a t i, H. N., K e y s e r, A. J., et al: Correlation of urinary lactic dehydrogenase with polym orphonuclear leukocytes in urinary tract infections in patients with spinal cord injuries. J. Infect. Dis. 147:161, 1983. 20. M a t t e n h e i m e r, H. : Enzymes in renal diseases. Ann. Clin. Lab. Sci. 7:422-431, 1977.

38 SUN, CHOW, MCVICAR, AND MAILLOUX 21. M a t t e n h e i m e r, H.: Enzymes in the urine. Med. Clin. North Amer. 55:1493-1508, 1971. 22. R o n a l d, A. R., C u t l e r, R. E., and T u r c k, M.: Effect of bacteriuria on the renal concentrating mechanism. Ann. Intern. M ed. 70:723 733, 1969. 23. S t a m e y, T. A., G o v a n, D. E., and P a l m e r, J. M.: The localization and treatment o f urinary tract infections: The role of bactericidal urine levels as opposed to serum levels. M edicine 44:1-36, 1965. 24. T h o m a s, V., S h e l o k o v, A., and F o r l a n d, M.: Antibody-coated bacteria in the urine and the site of urinary-tract infection. N. Engl. J. Med. 290:588-590, 1974. 25. P a p a d o p o u l o s, N. M.: Clinical applications of lactate dehydrogenase isoenzym es. Ann. Clin. Lab. Sci. 7:506-510, 1977.