Chronic kidney disease (CKD) is associated with increased mortality and cardiovascular disease in patients with HIV [1,2]. Cumulative exposure to tenofovir and some protease inhibitors has been associated with CKD defined according to reductions in estimates of glomerular filtration rate (GFR) . Proteinuria can precede reductions in GFR and represents significant renal disease, including proximal tubular dysfunction. Proteinuria has been shown to predict all-cause mortality in individuals with untreated and treated HIV . Proteinuria has been associated with tenofovir therapy, although the prevalence, predictors, and outcomes of persons who develop proteinuria in association with tenofovir therapy are not well described .
All patients undergo annual proteinuria screening in our clinic. This is performed by calculating the protein:creatinine ratio on a random urine specimen. The medical records of patients who received tenofovir for more than 1 year were retrospectively reviewed. Age, sex, HIV viral load, current CD4 cell count, antiretroviral status at time of initiation of tenofovir (naive or experienced), concomitant antiretroviral therapy, duration of tenofovir therapy, estimated glomerular filtration rate (eGFR) calculated by the Modification of Diet in Renal Disease equation (values over 90 ml/min/1.73m2 reported as >90), nonfasting serum phosphate (PO4), and urine protein:creatinine ratio (UPCR) were collated. Proteinuria was defined as a UPCR more than 15 g/mol on repeat testing having excluded urinary tract infection. Hypertension, diabetes, and BMI were not well recorded and have not been included in this analysis. Factors associated with proteinuria were examined using univariate (Student's t-test for continuous variables and χ2 test for categorical variables) and multivariate analyses. Changes in UPCR were compared in patients who ceased tenofovir because of proteinuria with those of patients who continued tenofovir despite proteinuria.
One hundred and fifty-three patients were identified and all were included in this analysis. The characteristics of the patients are shown in Table 1. All patients had a viral load below the limit of detection of 40 copies/ml. Forty-two (27%) patients had proteinuria (mean UPCR 38, SD 34). Multivariate analyses indicated that longer duration of tenofovir use [P = 0.006, odds ratio (OR), 1.29 per year, 95% confidence interval (CI), 1.08–1.54] and concomitant protease inhibitor therapy (P = 0.004, OR, 7.36, 95% CI, 1.88–28.9) were associated with proteinuria. Age, sex, current CD4 cell count, or treatment status at the time of initiation of tenofovir did not predict proteinuria in multivariate analyses. Proteinuria was associated with lower eGFR (median 80.5 versus >90 ml/min/1.73m2, approximate OR, 0.91 per ml/min/1.73m2, approximate 95% CI, 0.87–0.95) and lower nonfasting serum PO4 (mean 0.99 versus 1.06 mmol/l P = 0.009, OR, 0.70 per 0.1 mmol/l increment, 95% CI, 0.53–0.91). However, the differences in these parameters between patients with and without proteinuria were small and may not be considered to be significant by the clinician.
Twelve patients ceased tenofovir and commenced an alternative nucleoside reverse transcpritase inhibitor because of proteinuria. These patients did not alter other medication. In particular, they did not commence angiotensin-converting enzyme inhibitor therapy, other antihypertensive agents, diabetic therapy, or alter protease inhibitor therapy. Patients who ceased tenofovir because of proteinuria had higher mean UPCR prior to ceasing tenofovir than patients with proteinuria who continued tenofovir (64 versus 28; P = 0.03). Proteinuria significantly reduced over a 6-month period in 11 of 12 patients ceasing tenofovir (64–14; P = 0.001), whereas UPCR did not significantly change over a similar period in 30 patients continuing tenofovir despite proteinuria. UPCR increased from 71 to 123 in one patient despite cessation of tenofovir. This patient was also taking ritonavir-boosted darunavir. He was normotensive and had a normal fasting blood glucose level and eGFR more than 90. His urinary albumin creatinine ratio was also increased (87, cf normal less than 1). No cause has yet been identified for this patient's proteinuria.
The prevalence of proteinuria among those patients taking tenofovir in our cohort was 27%. We found an association between duration of tenofovir therapy and proteinuria and estimated that the odds of proteinuria increased by 2% for every month of tenofovir therapy (29% per year). These findings are consistent with other studies. The risk of proteinuria increased by 30% per year of exposure to tenofovir in one study . Another study reported that the incidence of proteinuria in patients taking tenofovir was 19% over 2 years . The co-administration of protease inhibitor therapy with tenofovir increased the odds of proteinuria by seven times in our study. This is consistent with another study . Possible causes of this association include ritonavir inhibition of enzymes involved in tenofovir elimination from the kidney.
Tenofovir-associated proteinuria was reversible in the majority (11/12) of patients who ceased the drug in our cohort. Other studies have not demonstrated such reversibility in renal abnormalities in patients who cease tenofovir [5,6]. Renal parameters did not improve following tenofovir cessation in 42% of patients in one study . The mean eGFR of the patients who ceased tenofovir in that study was 51 ml/min/1.73m2 compared with 76.3 ml/min/1.73m2 in our study. It is interesting to speculate that proteinuria may be an early and reversible effect of tenofovir-associated renal damage. As hypertension and diabetes were not well documented in this cohort, we cannot exclude the possibility that the management of these conditions impacted upon the improvements of proteinuria in patients ceasing tenofovir because of proteinuria; however, this seems unlikely. This retrospective study was also limited by its inability to provide information regarding the timing of the onset of proteinuria in patients taking tenofovir.
Proteinuria was commonly detected in patients taking tenofovir for more than 1 year in this cohort. The risk of proteinuria increased with a longer duration of tenofovir therapy and was greater in persons also taking protease inhibitor therapy. Proteinuria was generally observed in the absence of marked decreases in either serum phosphate or eGFR. Tenofovir-associated proteinuria generally resolved upon cessation of the drug. Larger studies are warranted to confirm these initial findings; to describe the natural history of tenofovir-associated proteinuria; to determine the value of regular proteinuria screening in persons taking tenofovir; and to address the impact of tenofovir-associated nephrotoxicity on bone disease.
The authors acknowledge Dr Andrew Redmond's careful review of the manuscript.
Conflicts of interest
There are no conflicts of interest.
1. Choi A, Scherzer R, Bacchetti P, Tien PC, Saag MS, Gibert CL, et al
. Cystatin C, albuminuria, and 5-year all-cause mortality in HIV-infected persons.Am J Kidney Dis
2. Choi AI, Li Y, Deeks SG, Grunfeld C, Volberding PA, Shlipak MG. Association between kidney function and albuminuria with cardiovascular events in HIV-infected persons.Circulation
3. Mocroft A, Kirk O, Reiss P, De Wit S, Sedlacek D, Beniowski M, et al
.; and EuroSIDA Study Group. Estimated glomerular filtration rate, chronic kidney disease and antiretroviral drug use in HIV-positive patients.AIDS
4. Wyatt CM, Hoover DR, Shi Q, Seaberg E, Wei C, Tien PC, et al
. Microalbuminuria is associated with all-cause and AIDS mortality in women with HIV infection.J Acquir Immune Defic Syndr
5. Calza L, Trapani F, Tedeschi S, Piergentili B, Manfredi R, Colangeli V, Viale P. Tenofovir-induced renal toxicity in 324 HIV-infected, antiretroviral-naïve patients.Scand J Infect Dis
6. Scherzer R, Estrella M, Li Y, Choi AI, Deeks SG, Grunfeld C, Shlipak MG. Association of tenofovir exposure with kidney disease risk in HIV infection.AIDS