Ranitidine-Associated Interstitial Nephritis and Fanconi Syndrome

CASE REPORTS

Ranitidine-Associated Interstitial Nephritis and Fanconi Syndrome Kotresha Neelakantappa, MD, Gloria R. Gallo, MD, and Jerome Lowenstein, MD • Ranitidine is an H2 blocking agent widely used in the treatment of peptic ulcer disease. Although acute interstitial nephritis has been described in association with a related drug, cimetidine, this complication secondary to ranitidine has been reported only once before. We describe a patient who presented with acute renal failure and features of Fanconi syndrome in association with the use of ranitidine. Renal biopsy showed acute interstitial nephritis. The patient had renal tubular acidosis with alkaline urine, aminoaciduria, renal glycosuria, uricosuria, and phosphaturia. There was a marked disparity between the degree of elevation of urea nitrogen and creatinine concentrations in the serum that was attributable to diminished urea generation, since simultaneous measurement of inulin, creatinine, and urea clearances showed that they were comparably reduced. Urinary nitrogen loss in the form of aminoaciduria contributed significantly to decreased urea generation. Both acute renal failure and Fanconi syndrome resolved after treatment with prednisone. © 1993 by the National Kidney Foundation, Inc. INDEX WORDS: Ranitidine; interstitial nephritis; Fanconi syndrome; aminoaciduria; urea generation.

R

ANITIDINE (Zantac; Glaxo, Research Triangle Park, NC) is a selective H2 receptor blocker widely used in the treatment of peptic ulcer disease. Although a related drug, cimetidine, has been known to cause acute interstitial nephritis, there is only one reported case of this secondary to ranitidine. 1 In fact, patients who have had acute interstitial nephritis secondary to cimetidine have been reported to tolerate ranitidine without adverse effects. 2 We report a case of acute interstitial nephritis and Fanconi syndrome secondary to ranitidine. CASE REPORT A 39-year-old Hispanic woman was admitted to the hospital January 5, 1991, with complaints of epigastric discomfort and nausea for 4 weeks' duration. She had taken 150 mg ranitidine orally twice a day for 3.5 weeks and I g sucralfate (Carafate; Marion Laboratories, Kansas City, MO) orally four times a day for 9 days prior to admission. She denied taking any other medication except for 0.1 mg levothyroxine sodium USP (Synthroid, Lincolnshire, IL). Specifically, the patient denied the use of antibiotics and nonsteroidal anti-inflammatory agents, including over-the-counter medication. Physical examination revealed a blood pressure of 110/80 mm Hg and absence of edema. There were no skin eruptions, alopecia, or lymphadenopathy. The remainder of the examination was unremarkable. Initial urinalysis by dipstick (Multistix, Ames Reagent Strips; Miles, Inc., Elkhart, IN) showed pH of 5.0, specific gravity of 1.015, protein of 30 mgjdL, and no glucose. Microscopic examination of the urine revealed two to three red blood cells, 20 to 25 white blood cells, and 2+ bacteria per high-power field. There were no crystals or casts. Blood urea nitrogen (BUN) was 22 mgjdL and creatinine was 3.9 mgjdL on admission. The serum sodium was 134 mEq/L, potassium was 4.4 mEq/ L, chloride was 100 mEq/L, and total venous carbon dioxide

content was 18.7 mEq/L. Serum uric acid concentration reached a nadir of 1.9 mgjdL. The hematocrit was 35%, the white blood cell count was 8,100, and platelets were 354,000/mm 3• Because of pyuria and bacteriuria, the patient was treated with ampicillin for the first 5 days until the urine culture tested negative. She remained afebrile and without symptoms of urinary infection. The patient was also treated with 50 mg ranitidine intravenously every 8 hours for the gastrointestinal symptoms. Eight days after admission the serum creatinine was 4.7 mg/dL and BUN was 13 mg/dL; ranitidine was discontinued. Because of the striking disparity between the concentrations of creatinine and urea, measurement of inulin, creatinine, and urea clearances was performed on the 10th hospital day. A 2-hour collection of urine with bladder catheterization was used for the clearance study. Inulin clearance was performed as per standard technique. 3 Creatinine was measured by an autoanalyzer technique that uses alkaline picrate and measures noncreatinine chromogens in addition to true creatinine.' Simultaneous clearance determinations showed the following: inulin clearance of 17.7 mL/min, creatinine clearance of 15.4 mL/min, and urea clearance of l7.1 mL/min (Table I). A percutaneous renal biopsy was performed 13 days after hospitalization. The biopsy specimen (Fig I) contained 14 normal-appearing glomeruli. There was severe dilfuse mononuclear and polymorphonuclear cell infiltration of the interstitium and focal invasion and necrosis of epithelium of proximal and distal tubules. The vessels appeared normal. Immunofluorescence

From the Departments of Medicine and Pathology, New York University Medical Center, New York, NY; and the Department of Medicine, The Methodist Hospital of Brooklyn, Brooklyn, NY. Received October 5,1992; accepted in revisedform February 23,1993. Address reprint requests to Kotresha Neelakantappa, MD, New York University Medical Center, 550 First Ave, Room TH 440, New York, NY 10016. © 1993 by the National Kidney Foundation, Inc. 0272-6386/93/2202-0015$3.00/0

American Journal of Kidney Diseases, Vol 22, No 2 (August), 1993: pp 333-336

333

NEELAKANTAPPA, GALLO, AND LOWENSTEIN

334 Table 1.

Cinulin

Patient Normal values

17.7 mL/min 120 ± 14 mL/min

Ce,

Cu"",

FEuncackl

FEphosphorua

24-hr Urinary Aminoacids

24-hr Urinary Glucose

15.4 mL/min Ca/C"''''n = 0.88 Ca/C"'"n = 1-1 .15

17.1 mL/min C....}Croln = 0.97 C....}ClrMJ 1mL/min

48.9%

28.8%

9,359 mg

11,505 mg

6%-20%

5%-20%

< 390 mg

None

NOTE. Clearances were calculated for 1.73 m2 body surface area. Twenty-four-hour urinary aminoacid and glucose as well as FE of uric acid and phosphorus were determined when Ce, was 25.6 mL/min.

examination showed no immunoglobulin deposits. A diagnosis of acute interstitial nephritis was made. Urine output ranged from 1.5 to 2.5 Lid. Twenty-fourhour urine protein excretion was 720 and 1, 129 mg on two occasions. The arterial blood gas analysis showed a pH of 7.28, PC02 of 29 mm Hg, and He0 3 of 13.2 mEq/L. The

urine pH was consistently above 7.0 on multiple occasions, except at the time of admission. It was 7.5 on the day blood pH was 7.28. Twenty-four-hour urinary ammonia excretion was 41 mmo!. The serum creatinine peaked at 6.8 mg/dL with a blood urea of 45 mgjdL 6 days after stopping ranitidine. Two weeks after stopping ranitidine, the serum creatinine was

Fig 1. (A) The renal biopsy demonstrates severe interstitial inflammatory cell infiltration around both proximal (large arrows) and distal (small arrows) tubules. (Hematoxylin-eosin stain; magnification X360.) (B) In this field there is leukocytic cell infiltration of tubules (small arrows) and disruption of tubular basement membranes in a necrotic tubule (large arrow). (Periodic acid silvermethenamine, counterstained with hematoxylin-eosin; magnification x 660.)

RANITIDINE-ASSOCIATED INTERSTITIAL NEPHRITIS

still 5.0 mg/dL. In light of the acute interstitial nephritis and failure to improve, the patient was treated with oral prednisone 60 mg/d along with I g sucralfate four times a day. Six weeks after initiation of prednisone therapy, serum creatinine was 2.1 mg/dL and creatinine clearance had increased to 25.6 mL/min. Other evidences of Fanconi syndrome were recognized. Twenty-four-hour urine collection showed the following (Table I): glycosuria of 11.5 g (while the daily fasting blood sugars never exceeded 160 mgfdL and most of the blood sugar levels were < 110 mgfdL), generalized aminoaciduria of 9.4 g, uric acid excretion of 268 mg with FEuric ac;d of 48.9%, and urinary phosphorus excretion of 221 mg with a FE""osphorus of 28.8%. The serum C-terminal parathyroid hormone level was 0.6 ngfmL, which is within the normal range for a serum calcium of9.1 mg/dL and indicates that the increase in fractional excretion of phosphorus was not due to excess parathyroid hormone. Serum creatinine fell to 1.4 mg/dL after 8 weeks of prednisone therapy. The urinalysis, however, still showed a pH of 7, glycosuria (500 mg/dL), and proteinuria (30 mg/dL) by dipstick examination. The serum CO 2 content was 21 mEq/ L. Prednisone was discontinued after 9 weeks. On a followup visit 10 months later, serum creatinine was 1.3 mgfdL, BUN was 17 mgfdL, uric acid was 3.5 mgfdL, and venous CO 2 content was 24.4 mEq/ L. Urinal ysis showed a pH of6.5 and a trace of protein . There was no glycosuria.

DISCUSSION

The patient presented with acute interstitial nephritis and evidence ofFanconi syndrome. She had renal tubular acidosis, renal glycosuria, amino aciduria, phosphaturia, and uricosuria. These are all features of proximal tubular dysfunction. The finding of a urinary pH of 7.5 in the face of metabolic acidosis is consistent with either a severe defect in proximal tubular bicarbonate reabsorption with a threshold lower than the plasma level (13.2 mEq/L) and continued bicarbonaturia or with a distal acidification defect in addition to the proximal defect. A striking feature of the patient's presentation was the disparity between the degree of elevation of BUN and serum creatinine concentrations. In the second week of hospitalization, BUN was 10 to 14 mg/dL and creatinine was 4.0 to 4.6 mg/ dL. The possibility that this reflected inhibition of tubular secretion of creatinine by ranitidine was considered. Ranitidine, like cimetidine, is an organic cation and, at physiologic plasma pH, acts as a weak base. A fivefold increase in clearances of these cations over creatinine clearance (Ccimetidine/Ccr and Cranitidine/Ccr >5) has been found in sheep, suggesting that they are excreted largely by tubular secretion. s Elevation of plasma creatinine and reduction in creatinine clearance without a

335

reduction in inulin clearance has been described with cimetidine administration in healthy volunteers. 6 This phenomenon has been attributed to competitive inhibition of tubular secretion of creatinine by cimetidine. Ranitidine, on the other hand, has not caused significant elevation in serum creatinine.? In a group of renal transplant patients treated with H2 blockers, Pachon et al found no change in the ratio Ccr to Cinulin after 1 week of treatment with ranitidine as opposed to a decrease in this ratio from 1.8 to 1.4 with cimetidine. 8 Ranitidine is a much more efficient H2 blocker than cimetidine and, hence, is used in smaller doses. The difference in dosage might explain the relatively insignificant interference with creatinine secretion by ranitidine. Only 5% to 20% of the excreted creatinine is accounted for by tubular secretion when the glomerular filtration rate (GFR) is normal. Therefore, inhibition of tubular secretion alone should not cause more than 20% elevation in plasma creatinine concentration as long as the GFR is normal. The striking elevation in serum creatinine in this patient suggested a marked reduction in GFR, which was confirmed by measurement of inulin clearance of 17.7 mL/min/m2. Creatinine clearance usually exceeds inulin clearance because creatinine, unlike inulin, is excreted both by glomerular filtration (GFR) and tubular secretion. The Ccr to Cinulin ratio increases as renal function declines, despite accumulation of the noncreatinine chromogens in the plasma. These chromogens are measured as creatinine by the autoanalyzer and are not excreted in the urine, hence accounting for an underestimation of true creatinine clearance. The underestimation of GFR by the measurement of creatinine clearance attributable to accumulation of noncreatinine chromogens in the plasma is usually offset by an even greater overestimation due to tubular secretion of creatinine as creatinine excretion by glomerular filtration declines. In our patient the clearance of creatinine was lower than that of inulin (Ccr/Cinulin = 0.88); this may be attributed to impairment of creatinine secretion as part of a more generalized picture of tubular dysfunction in a patient with severe interstitial nephritis and Fanconi syndrome as well as the expected accumulation of noncreatinine chromogens when GFR is reduced. Since simultaneous measurements of creatinine, urea, and inulin clearances gave clear evi-

336

dence that glomerular filtration was markedly reduced, the disparity in BUN and creatinine concentrations could be ascribed to an unusually low blood urea concentration. The urea clearance relative to creatinine clearance was only modestly increased and could not explain the disparity in the plasma concentrations of these substances. Therefore, reduced blood urea concentration was attributable to reduced urea generation. The 24hour urinary urea nitrogen excretion was only 1,403 mg. On a diet providing 1 g proteinjkg daily, one would expect the 24-hour urinary urea nitrogen to be approximately 6,500 mg with approximately 4,000 mg of non-urinary urea nitrogen loss per day. A reasonable explanation for low urea generation in our patient lies in urinary nitrogen loss as amino acids. Aminoaciduria of 9,400 mg/ d represents approximately 1,500 mg of nitrogen that would have been used for urea synthesis. The urinary nitrogen loss due to aminoaciduria was as great as the urinary urea nitrogen. Urea generation and, hence, BUN concentration would have been twice that observed at the same urea clearance in the absence of aminoaciduria. Enhanced ammonia excretion can represent another source of urinary non-urea nitrogen loss. Carlisle et al found that increased urinary ammonia excretion reduced the nitrogen pool available for urea synthesis in hyperchloremic metabolic acidosis resulting from toluene toxicity associated with glue sniffing. 9 One would expect increased urinary ammonium in chronic metabolic acidosis. Urinary NH4 excretion has been described to reach 350 mmol/d in humans during chronic metabolic acidosis. lo Under these circumstances, the urine is maximally acidified. Based on the difference between urinary and plasma pH as well as urine and peri tubular flow rates, it can be calculated that another 150 mmol/ day of NH4 enter the renal vein. Unlike the patient described by Carlisle et ai, urinary ammonia in our patient was not elevated, measuring only 41 mmol/24 hr. In our patient the urine pH was 7.5 and the plasma pH was 7.28. Based on pH difference alone the quantity ofNH4 exiting into the plasma would be 1.66 times greater than that entering the urine. We did not measure renal plasma flow rate in our patient and can only estimate peritubular plasma flow rate. Allowing for a markedly reduced peri tubular flow, if a non-ionic diffusion

NEELAKANTAPPA, GALLO, AND LOWENSTEIN

equilibrium of NH3/NH4 were attained, the finding of urinary ammonium excretion of 41 mmolj24 hr would lead us to estimate that a considerably greater amount of ammonia entered the peritubular plasma and that renal ammoniagenesis was increased in response to metabolic acidosis. Unlike chronic metabolic acidosis with unimpaired urinary acidification in which the bulk of renal ammoniagenesis results in urinary ammonium excretion, and thereby represents a loss ofnitrogen,9 in our patient, impaired urinary acidification prevented trapping of ammonia in the tubular lumen. Virtually all the NH4 was returned via the renal venous blood, ultimately to be used (together with a-ketoglutarate) in hepatic urea synthesis. The present case represents a rare complication of ranitidine and illustrates the unique features of creatinine handling and nitrogen balance encountered with severe interstitial nephritis and the Fanconi syndrome. ACKNOWLEDGMENT The authors are indebted to Charles E. Langs, MD, and Sui Leong for their assistance in clearance studies.

REFERENCES I. Freeman HJ: Ranitidine associated interstitial nephritis in

a patient with celiac sprue. Can J Gastroenterol 2:35-36, 1988 2. Potter PH, Westby RG: Interstitial nephritis after cimetidine but not ranitidine. JAMA 249:351, 1983 3. Smith HW: Principles of Renal Physiology. New York, NY, Oxford University, 1956 4. Doolan PD, Alpern EL, Theil GB: Aclinical appraisal of the plasma concentration and endogenous clearance of creatinine. Am J Med 32:65-79, 1962 5. Czuba MA, Morgan DJ, Ching MS, Mihaly GW, Hardy KJ, Smallwood RJ: Ontogeny of fetal renal organic cation excretion: A study with cimetidine and ranitidine during the latter half of gestation in pregnant ewe. J Pharmacol Exp Ther 255:1177-1182, 1990 6. Dubb JW, Stote RM, Familiar RG, Lee K, Alexander F: Effect of cimetidine on renal function in normal man. Clin Pharmacol Ther 24:76-83, 1978 7. Rocci ML, Jr, Vlasses PH, Ferguson RK: Creatinine serum concentrations and H2-receptor antagonists. Clin Nephrol 22:214-215, 1984 8. Pachon J, Lorber MI, Bia M: Effects ofH2-receptor antagonists on renal function in cyclosporin-treated renal transplant patients. Transplantation 47:254-259, 1989 9. Carlisle EJ, Donnely SM, Vasuvattakul S, Kamel KS, Tobe S, Halperin ML: Glue-sniffing and distal renal tubular acidosis: Sticking to the facts. J Am Soc Nephrol 1:1019-1027, 1991 10. Halperin ML, Jungas RL, Pichette C, Goldstein MB: A quantitative analysis of renal ammoniagenesis and energy balance: Atheoretical approach. Can J Physiol Pharmacol60: 1431-1435, 1982