Nonalbumin proteinuria predominates in biopsy-proven tenofovir nephrotoxicity.

OBJECTIVE

Tenofovir disoproxil fumarate (TDF) nephrotoxicity is characterized by proximal renal tubular injury and dysmorphic mitochondria resulting in proteinuria, orthoglycemic glycosuria, and other markers of proximal tubular dysfunction. The objective of this study was to determine the pattern of proteinuria in patients with biopsy-proven TDF nephrotoxicity.

DESIGN

Retrospective chart review.

METHODS

Patients with biopsy-proven TDF nephrotoxicity were identified and their medical charts and biopsy reports were reviewed. Comparison was made with HIV-infected patients not on TDF who underwent kidney biopsy.

RESULTS

We identified 43 biopsy-proven cases of TDF nephrotoxicity; mean age 54.7 ± 0.4 years, 53% men, 42% whites. Thirty-seven cases reported proteinuria by dipstick of which only 60% had at least 2+ proteinuria. Twenty-seven patients had urine protein quantified by either 24-h collection or spot urine protein-to-creatinine ratio; median proteinuria was 1742 mg/day [interquartile range (IQR) 1200-2000 mg] and 1667 mg/g creatinine (IQR 851-1967 mg/g), respectively. Ten patients had concurrent urinary albumin measured, with a median 236 mg/g creatinine (IQR 137-343 mg/g). The mean urine albumin-to-urine protein ratio (uAPR) was 0.17 (IQR 0.14-0.19), confirming that TDF nephrotoxicity is primarily associated with nonalbumin proteinuria. Control cases had a uAPR of 0.65 (IQR 0.55-0.79) P < 0.001. Histopathology showed the predominance of proximal tubular injury with characteristic mitochondrial abnormalities.

CONCLUSION

In the largest published cohort of patients with biopsy-proven TDF nephrotoxicity, we show that low uAPR is a reliable feature of this disease. Because of the predominance of nonalbumin proteinuria, dipstick urinalysis may be unreliable in TDF nephrotoxicity.