Background:  Tenofovir (TDF) is a nucleotide analogue active against HIV and hepatitis B virus. Although TDF does not seem to affect kidney glomerular function, several studies have found significant impairment in kidney tubular function after prolonged exposure to TNF, as compared with other nucleoside analogs. Possible influence of TDF exposure on kidney tubular function has not been explored so far.

Methods:  All consecutive HIV patients receiving TDF for longer than 6 months at a referral HIV clinic in Madrid during year 2008 were identified. Kidney tubular function was assessed in all patients in plasma and 24-hours urine. Altered kidney tubular function was defined when at least 2 of the following abnormalities were present:  non-diabetic glucosuria, urine phosphate wasting, hyperaminoaciduria, b₂-microglobulinuria, and increased fractional excretion of uric acid. Mid-dose plasma levels of TDF were measured using a validated high-performance liquid chromatography (HPLC) -mass method.

Results:  A total of 92 HIV patients, with a median exposure to TDF of 33 months (IQR 10 to 46), were included in the study:  18 patients had altered kidney tubular function, and 74 controls had normal kidney tubular function. Creatinine clearance (24 h) was normal and comparable between groups (94 vs 118 mL/minute, respectively). Both groups were well matched in gender, age, weight, hepatitis C virus (HCV) co-infection, months under TDF, exposure to nephrotoxic drugs, concomitant PI use, and history of hypertension. History of diabetes was more frequent in cases (44%) than controls (22%) (p = 0.05). Plasma levels of TDF were higher in cases than controls (182 [IQR 105 to 220] vs 104 [IQR 75 to 139] ng/mL, respectively; p = 0.001). The best TDF plasma concentration threshold associated with altered kidney tubular function was 160 ng/mL (AUROC 0.75 [95%CI 0.63 to 0.87], p = 0.001) (61% sensitivity and 80% specificity). Multivariate analysis including all relevant variables showed that only TDF plasma levels >160 ng/mL were associated to abnormal kidney tubular function (OR, 4.8 [95%CI 1.5 to 16], p = 0.008).

Conclusions:  Plasma levels of TDF are increased in patients experiencing abnormal kidney tubular function on treatment. These results confirm the potential specific toxicity of TDF on the kidney tubular epithelium, as other nucleoside monophosphates (adefovir and cidofovir). Studies assessing the efficacy and safety of lower doses of TDF are warranted, and this information could be helpful for hepatitis B mono-infected patients as well.