Purpose: Human immunodeficiency virus (HIV)-associated nephropathy is a common and serious cause of progressive renal insufficiency in patients with HIV, frequently presenting with nephrotic range proteinuria. The purpose of this study is to document the histopathologic diagnoses seen in HIV-positive patients with and without nephrotic range proteinuria and to evaluate the predictive value of both nephrotic range proteinuria and CD4 count in diagnosing HIV-associated nephropathy.
Methods: We performed a cross-sectional, single-center study of all 107 HIV-positive patients who had both a renal biopsy and urine protein measurement between 1995 and 2002. Nephrotic range proteinuria was defined as a urine protein-to-creatinine ratio > 3 or a 24-hour urine protein > 3 g. Clinical and laboratory characteristics of those patients with and without HIV-associated nephropathy were compared. Sensitivity, specificity, and positive and negative predictive values of nephrotic range proteinuria in the diagnosis of HIV-associated nephropathy were determined.
Results: Fifty-five biopsied patients had nephrotic range proteinuria, among whom 29 (53%) were diagnosed with HIV-associated nephropathy. Among the remaining patients, 12 had non-HIV-associated nephropathy focal segmental glomeruloscerlosis, 3 had membranoproliferative glomerulonephritis, 2 had AA Amyloid, 2 had diabetic nephropathy, and 7 had other diagnoses. Sensitivity, specificity, and positive and negative predictive values of nephrotic proteinuria in the diagnosis of HIV-associated nephropathy were 73%, 61%, 53%, and 79%, respectively. The patients with HIV-associated nephropathy had a significantly higher creatinine (8.2 mg/dL vs 2.5 mg/dL, P < .001) and a lower CD4 count (158 count/mm3 vs 349 count/mm3, P < .01) at the time of biopsy. Although significantly more patients with HIV-associated nephropathy had a CD4 count below 200 (P = .03), among those with a CD4 count below 200, 10 of 30 patients (33%) had diagnoses other than HIV-associated nephropathy. Injection drug use, presence of hepatitis C, and hypertension were not associated with HIV-associated nephropathy.
Conclusion: Our results suggest that HIV patients with nephrotic range proteinuria warrant a kidney biopsy because the presence of nephrotic range proteinuria, even in the presence a low CD4 count, does not establish the diagnosis of HIV-associated nephropathy.