975 
Effect of Tenofovir on Renal Function in Patients Using HAART
Michael Horberg*1, B Tang2, W Towner2, S Bersoff-Matcha3, D Klein4, M Silverberg1, W Chen2, L Hurley4, J Chang2, and J Blank3
1Kaiser Permanente, Oakland, CA, US; 2Kaiser Permanente Southern California, Pasadena, US; 3Kaiser Permanente Mid-Atlantic, Washington, DC, US; and 4Kaiser Permanente Northern California, Oakland, US
Background: Tenofovir (TDF) is widely used in HAART, yet
its effect on renal function remains unclear. We studied long-term TDF effects
on renal function during therapy in a real world cohort.
Methods: We performed a retrospective cohort analysis comparing
Kaiser Permanente patients starting a TDF-containing HAART regimen to patients
starting a non-TDF-containing one in years 2002 to 2005 with serum creatinine
<1.3mg/dL and HIV RNA (viral load) >75/mL at baseline (<90 days prior
to regimen start). Baseline measures included demographics, co-morbidities,
HAART data, laboratory values for serum creatinine, phosphorus, HCO3,
urinalysis, CD4 count, and viral load. We recorded all follow-up values of the
same measures through December, 2006. We calculated glomerular filtration rate
(GFR) by modification of diet in renal disease (MDRD) equation and defined
Fanconi syndrome as having ≥3 of the following criteria in a 30-day period:
increased serum creatinine by 50% or by ≥0.5 mg/dL or serum creatinine >1.5
mg/dL, phosphorus <2.7mg/dL, proteinuria, glucosuria with normal serum
glucose, HCO3 <23 mEq/L. We report crude rate ratios (RR) and
adjusted hazard ratios (HR) by Cox models with predictors (including diabetes
mellitus) retained in multivariable models if p <0.10.
Results: We analyzed 1742 TDF and 623 non-TDF patients (n
= 2365); 90% were men and 19% were black. Median follow-up was 2.7 years (IQR
1.8 to 3.8). At baseline, TDF patients were less often ARV-naive than non-TDF,
but had higher CD4 counts and higher prevalence of diabetes (all p <0.05);
otherwise the groups were similar. The groups had similar rates of achieving viral
load <75/mL (80% vs 84%; p = 0.14). TDF patients had greater risk of
≥50% decrease in GFR (RR = 1.8; HR = 1.76, 95%CI, 1.10 to 2.82; p =
0.02), a higher risk of serum creatinine rising >2.0 mg/dL (RR = 1.9; HR = 1.81,
1.00 to 3.29; p = 0.05), and more frequent development of Fanconi syndrome
(174 vs 21 events, RR = 3.1; HR = 2.91, 1.80 to 4.71; p <0.001), with
proteinuria, elevated serum creatinine, and HCO3 being the criteria
most frequently met. We measured <3 Fanconi criteria for 16% of TDF and 28% of
non-TDF patients (p <0.001). Regimen discontinuation was more frequent
among TDF patients (39% vs 32%, HR = 1.37, 1.16 to 1.62; p <0.001);
TDF patients were more likely to be viral load <75/mL at time of
discontinuation (p = 0.002).
Conclusions: While highly efficacious, TDF is associated
with statistically significant decrease in renal function including Fanconi syndrome
as we define it over extended follow-up in this large cohort. Close monitoring
of renal function and the components of Fanconi syndrome among patients taking
TDF is warranted.
|
