780
Didanosine and Lower Baseline Body Weight Are Associated with Declining Renal Function among Patients Receiving Tenofovir
Heidi Crane*, R Harrington, S Van Rompaey, and M Kitahata
Univ of Washington, Seattle, US
Background: Factors associated with renal impairment among
patients receiving tenofovir DF (TDF) have not been determined.
Methods: Between November 2001 and September, 2005, we
conducted this observational cohort study of all HIV-infected patients receiving
TDF in an urban HIV clinic population. Development of abnormal renal function
was defined as an absolute glomerular filtration rate (GFR) ≥90 before
initiation of TDF, and <90 at the last creatinine measurement on TDF. We
used logistic regression analysis to examine factors associated with abnormal
renal function, including ART medications, weight at initiation of TDF, weight change by last measurement on TDF,
race, age, gender, CD4 count nadir, HIV-1 RNA level, hepatitis C virus (HCV)
antibody, diabetes, and hypertension. We used the Cockcroft-Gault (CG)
equation that incorporates body weight to calculate GFR because weight changes
are common among HIV-infected patients. We also examined GFR using the modification
of diet in renal disease (MDRD) equation.
Results: Among the 497 patients initiating TDF, 87
patients developed abnormal renal function, which was associated with
didanosine (ddI) use (p = 0.009), lower baseline weight (p = 0.001),
and older age (p = 0.009) compared with patients with normal renal
function. 79 of 87 patients had a moderate decline in GFR (≤60). Of 87
patients, 8 had a severe decline in GFR (>80) that was associated with
ritonavir-boosted regimens (p=0.03)
containing either amprenavir (p=0.04) or lopinavir (p=0.009), and a trend toward ddI use (38% vs 15%, p = 0.09).
After adjusting for other factors, ddI (OR 2.5, p = 0.02), lower
baseline weight (OR 0.95/kg, p = 0.001), and increasing age (OR 1.07/year,
p <0.002) were significantly associated with abnormal renal function.
The results were unchanged when only patients with moderate abnormalities were
examined; there were too few patients with severe abnormalities for separate
analysis. When GFR was calculated using MDRD, fewer patients had abnormal renal
function than when the CG equation was used.
Conclusions: Increased age, ddI use, and lower baseline
weight are significantly associated with risk for renal impairment among
patients receiving TDF. GFR results using the MDRD were inconsistent with those
calculated using CG which points to the importance of weight in the measurement
of GFR among HIV-infected patients. Our findings suggest there may be a
moderate and severe category of renal abnormalities associated with TDF and
that ritonavir may contribute to severe abnormalities.
|