Objective: To investigate the association between tenofovir disoproxil fumarate (TDF) use and renal abnormality in a large cohort of HIV-1-infected children on antiretroviral therapy (ART).
Design: Nested case-control study.
Methods: Patients were from the Collaborative HIV Paediatric Study, a cohort of approximately 95% of HIV-1-infected children in the UK/Ireland. Serum (but not urine) biochemistry results for 2002-2008 were obtained for 456 ART-exposed children (2-18 years) seen at seven hospitals. Cases had either confirmed hypophosphataemia DAIDS grade at least 2 or estimated glomerular filtration rate (eGFR) less than 60 ml/min per 1.73 m; three controls per case were matched by hospital. Conditional logistic regression identified risk factors for renal abnormality.
Results: Twenty of 456 (4.4%) had hypophosphataemia, and one had eGFR less than 60 ml/min per 1.73 m. Ten of 20 (50%) cases versus 11 of 60 (18%) controls had taken TDF-containing ART for a median [interquartile range (IQR)] of 18 [17-20] months, as part of second-line or salvage therapy. The hypophosphataemia incidence rate was 4.3/100 person-years in the TDF group versus 0.9/100 person-years in those not exposed to TDF. In multivariable analysis, only TDF exposure in the previous 6 months was associated with hypophosphataemia [odds ratio (OR) = 4.81, 95% confidence interval (CI) 1.45-16.0, P = 0.01]. In six of 10 children with hypophosphataemia and at least four subsequent phosphate measurements, phosphate values returned to normal when TDF was stopped; in four with three measures or less, values rose but remained subnormal.
Conclusions: Hypophosphataemia was uncommon (4%), but was associated with prolonged TDF use, and was generally reversible following TDF withdrawal. Findings highlight the importance of continuing to monitor longer-term renal function, in particular tubular function, especially in those taking TDF. Further studies assessing urine biochemistry measures which more accurately indicate renal tubular damage are required.