A Case of Cocaine-Induced Acute Interstitial Nephritis

Discussion 

When evaluating a patient with AKI, one must always be mindful that the cause may be multifactorial. Also, given the inherently complex nature of AKI, a systematic approach must be used. We attempted to work through the cause of this patient's AKI and will provide what we believe to be the most likely diagnosis.

The patient's recent cocaine use seemed a likely cause of AKI. The notion of AKI secondary to cocaine abuse is not a novel concept; 2 common causes are rhabdomyolysis and arterial hypertension induced by intense vasoconstriction.3, 4, 5 There are 3 substantiated mechanisms of renal failure caused by rhabdomyolysis: vasoconstriction with resultant ischemia, direct toxic effects of the myoglobin heme-protein on renal tubular cells, and intraluminal cast formation leading to mechanical obstruction.6 The patient clearly showed biochemical evidence of rhabdomyolysis. However, did the rhabdomyolysis completely account for the severity of the patient's AKI? The renal biopsy specimen and urine microscopy did not show casts to suggest intratubular obstruction. Also, unlike this patient, patients with renal failure caused by severe rhabdomyolysis are often oliguric. Additionally, changes seen in tubules were primarily inflammatory in nature, including infiltrating eosinophils. There were some mild tubular ischemic changes on the biopsy specimen, but no evidence of tubular necrosis to explain the AKI. Lastly, when factors predictive of AKI in patients with rhabdomyolysis were evaluated, the majority of patients who developed AKI were older (mean age, 53.7 years) and had a peak CK level greater than 16,000 U/L.7 The patient in this case was younger and had a lower peak CK level. CK levels decrease approximately 39% per day in healthy persons, but are prolonged in patients with renal failure.8 CK levels typically peak in 24 to 36 hours after the onset of rhabdomyolysis.8 It is unlikely that a significantly greater CK level was missed because the patient was actively using cocaine 2 days before presentation.

Acute cocaine toxicity also may lead to vasoconstriction of the renal vasculature with resultant medullary hypoxia and tubular ischemia.4 Two such examples have been reported in the literature.4, 5 In these cases, renal function returned to normal within 10 days or less, suggesting a more self-limited process.4, 5 Our patient showed evidence of mild vasoconstriction on the biopsy specimen, which, in addition to rhabdomyolysis, accounts for the tubular ischemic changes. However, this patient's Cr level was still increased at day 15, suggesting more than transient vasoconstriction and ischemia as the sole cause of AKI.

Given the recent drug exposure, nonoliguria, and eosinophiluria, AIN also was considered in the differential diagnosis. It is important to remember that the classic triad of fever, rash, and eosinophilia is present in only one third of patients with drug-induced AIN, and the absence of these findings should not be used to exclude the diagnosis.9 The biopsy specimen was consistent with the diagnosis of AIN, showing tubulointerstitial nephritis with admixed eosinophils. But what was the offending agent? The patient admitted to ibuprofen use; however, the overall presentation was not typical for nonsteroidal anti-inflammatory drug (NSAID)-induced AIN. In a retrospective study by Schwarz et al,10 it was found that NSAID-induced AIN was more likely to result in nephrotic-range proteinuria compared with other causes of AIN (38% versus 14%, respectively). Additionally, NSAID exposure occurred within 24 hours of the initial presentation to the hospital. In most cases of NSAID-induced AIN, the exposure to NSAIDS is present for a much longer period, often several weeks to months, before developing AKI.10, 11 Lastly, NSAID-induced AIN typically presents with a 1:2 man-woman ratio, with a mean age of 60 years.11 The patient did not fulfill any of the typical characteristics for NSAID-induced AIN.

We reconsidered the patient's heavy cocaine abuse as a cause of AKI. Two reports were identified in the literature of AIN caused by cocaine use.12, 13 In both cases, the AKI was preceded by heavy cocaine abuse, which required a short course of hemodialysis therapy. The 2 prior reported cases also resulted in complete renal recovery. In the report by Decelle et al,13 prednisone therapy was also initiated with a serum Cr level of 1.7 mg/dL on day 9. In the report by Alvarez et al,12 the patient declined prednisone therapy because of concerns of weight gain, and serum Cr level was still greater than 10 mg/dL at week 4. The role of steroids in AIN is not completely clear. There have been no true prospective randomized trials assessing the efficacy of steroids. In a study by Galpin et al,14 the addition of prednisone, 60 mg/d, hastened the recovery of renal function compared with patients who did not receive prednisone (9.3 versus 54 days, respectively). Given the lack of prospective data for AIN, it is difficult to make a determination on the role of steroid therapy in renal recovery.

Lastly, given the newly diagnosed HIV status, we also wanted to exclude any possibility of concomitant HIV-associated nephropathy (HIVAN). HIVAN typically is associated with both proliferative and apoptotic changes on the renal biopsy specimen, both of which were absent in this patient.15 Additionally, no viral inclusions were noted and no tubuloreticular structures were identified. HIVAN typically presents with nephrotic-range proteinuria, which was absent in our patient.15 However, the patient was noted to have enlarged kidneys on ultrasound. It traditionally has been believed that patients with HIVAN and those with AIN may present with enlarged kidneys. However, a recent cross-sectional study by Atta et al16 showed in an HIV-positive patient population that underwent kidney biopsy for evaluation of renal disease that kidney size on ultrasound was not associated with HIVAN status. Conversely, a study by Pusey et al17 showed that 78% of patients with drug-associated AIN had radiological evidence of enlarged kidneys.

When faced with the diagnosis of AKI, a broad differential diagnosis must be considered because multiple potential causes often are present. In this case, we believe that although the cause of AKI was multifactorial, drug-induced AIN accounted for the majority of the renal injury. Although there is no definitive way to establish causality between exposure to cocaine and the development of AIN, we believe that cocaine was the most likely culprit. When considering a diagnosis of AIN, a thorough drug history is always a prudent endeavor. A urine toxicology screen should be performed, and if cocaine-induced AIN is suspected, some consideration should be given to the use of corticosteroid therapy because it appears to hasten the recovery of renal function.