Background: Fanconi syndrome (FS) is a well recognised complication of use of the antiretroviral agent of tenofovir disoproxyl fumarate (TDF).
Despite millions of patient years of TDF use, the incidence and predictors of FS are not known. The aim of this study was to determine the incidence and predictors of FS in a clinic cohort of patients taking TDF.
Methods: Clinical records and laboratory investigations from patients receiving ART between January 2002 and March 2016 at the Melbourne Sexual Health Centre, Australia, were extracted. FS was defined as new onset normoglycaemic glycosuria and proteinuria and at least one other marker of renal dysfunction. Kaplan-Meier survival curves were performed using duration of exposure to ART: not containing TDF, or containing TDF, with or without ritonavir co-administration. Cox regression analysis was performed on TDF exposures with using the covariates ritonavir co-administration, age, sex, co-morbidities (hypertension, hyperlipidemia, diabetes, viral hepatitis), CD4 cell count nadir and baseline eGFR.
Results: 1537 patients received ART, including 1260 who received TDF, of whom 398 patients received TDF co-administered ritonavir, representing 10401 patient years (PY) of ART, 5327 PY of TDF and 1641 PY of TDF ritonavir co-administration. 13 cases of FS were identified. All cases were taking TDF and the mean duration of exposure was 55 months (12-98) before developing FS. The incidence of FS was 1.09/1000PYs (95% CI 0.54-1.63) of TDF exposure (without ritonavir) and 5.48/1000PYs (3.66-7.33) of TDF-ritonavir co-administration (p=0.0057). The adjusted hazards ratio for ritonavir co-administration was 4.71 (1.37-16.14, p=0.014). Known risk factors for chronic kidney disease were not associated with an increased risk of FS.
Conclusions: In this first published study of the incidence of Fanconi syndrome, we find that ritonavir co-administration but not other factors is associated with a greater risk of FS. FS developed late. Out study supports ongoing renal monitoring in long term supressed patients with twice yearly urinalysis, particularly if serum laboratory monitoring is not available.

Download the e-Poster