| |
Tenofovir and Changes in Renal Function
|
| |
| |
Clinical Infectious Diseases August 2005
Letters To the Editor & Response by Gallant
Samir K. Gupta
Division of Infectious Diseases, Department of Medicine, Indiana University School of Medicine, Indianapolis, Indiana
SIR—I read with interest the recent article by Gallant et al. [1] detailing the changes in renal function in HIV-infected patients treated with a tenofovir-containing antiretroviral regimen, compared with patients treated with alternative nucleoside analogues. The authors should be applauded for providing comparative longitudinal data on the impact of tenofovir on renal function. However, one major issue requires clarification and explanation.
Not only did the serum creatinine level increase by a median of 0.15 mg/dL in the tenofovir group, but, interestingly, there was also a median increase of 0.10 mg/dL in the non-tenofovir group. This translated to a median decline in the creatinine clearance rate of 〜16 mL/min in the non-tenofovir group over the course of 1 year. One might presume that renal function would improve with HAART in a predominantly African-American population that is more at risk for HIV-related renal diseases [2]. Therefore, it would be important to determine whether these within-group changes were significant and whether there was a significant change in renal function in the overall cohort.
If there was indeed a significant decline in the creatinine clearance rate in the nucleoside reverse-transcriptase inhibitor group, what was driving it? The changes in serum creatinine levels presented in Gallant et al. [1] appear to explain the observed changes in the estimated creatinine clearance rate (a 1-year increase in age only reduces the creatinine clearance rate by 〜1 mL/min). However, this assumes that total body weight (the third time-varying parameter, along with age and serum creatinine level, in the Cockcroft-Gault equation [3]) was stable during the 1-year study period. However, data on weight at baseline in the 2 groups and on changes in weight are not provided by Gallant et al. [1] to confirm this; nor does it appear that these variables were included in the multivariable analyses as age at baseline was. If weights were indeed stable, a potential explanation for at least a portion of the decline in the estimated creatinine clearance rate that was observed in both groups may lie in changes in body composition during initial antiretroviral treatment.
Muscle is the source of serum creatinine. Muscle mass could potentially increase during the initial stages of HAART without an appreciable increase in total body weight. Therefore, an increase in the serum creatinine level would just be the normal result of the effects of HAART on body composition, as opposed to being the result of true impairment in renal function. This phenomenon would be most likely to occur in subjects with wasting and lower CD4 cell counts [4]. Notably, the authors found that having a CD4 cell count of <50 cells/μL was an independent predictor of a decline in the creatinine clearance rate. Were dual-energy x-ray absorptiometry scans or bioelectric impedance measurements available to test this hypothesis? If not, it would still be helpful to know the weight data for this cohort. If weight or muscle mass does not at least partially account for the observed renal function decline, then further research will be needed to determine the effects of all antiretrovirals on renal function. Furthermore, it would be important to study and derive an HIV/AIDS-specific renal function estimating equation that would account for the rapid changes in body composition seen with the use of HAART.
Acknowledgments
Potential conflicts of interest. Since this letter was accepted for publication, S.K.G. has received an honorarium from Gilead Sciences, the manufacturer of tenofovir. Gilead is also currently considering whether to fund a research study recently proposed by S.K.G.
References
1. Gallant JE, Parish MA, Keruly JC, Moore RD. Changes in renal function associated with tenofovir disoproxil fumarate treatment, compared with nucleoside reverse-transcriptase inhibitor treatment. Clin Infect Dis 2005; 40:1194–8. First citation in article | Full Text | PubMed
2. Ross MJ, Klotman PE. HIV-associated nephropathy. AIDS 2004; 18:1089–99. First citation in article | PubMed | CrossRef
3. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron 1976; 16:31–41. First citation in article | PubMed
4. Shikuma CM, Zackin R, Sattler F, et al. Changes in weight and lean body mass during highly active antiretroviral therapy. Clin Infect Dis 2004; 39:1223–30. First citation in article | Full Text | PubMed
Reply to Gupta
Joel E. Gallant, Michelle A. Parish, Jeanne C. Keruly, and Richard D. Moore
Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland
Sir—We agree with Dr. Gupta [1] that our initial results could be at least partially explained by changes in body weight, which affects the creatinine clearance rate calculated using the Cockcroft-Gault equation. Specifically, one might expect initial increases in body weight and muscle mass in patients starting initial therapy, of whom more were in the alternative nucleoside reverse-transcriptase inhibitor (NRTI) arm. One might also expect to see decreases in muscle mass associated with mitochondrial toxicity in patients receiving NRTIs (i.e., stavudine, zidovudine, and didanosine). In our cohort, significantly more patients in the NRTI arm used those 3 drugs. Thus, there were differences between the 2 arms that could have differentially affected the calculated creatinine clearance value in either direction.
Because ours was an observational study, dual-energy x-ray absorptiometry data were not available. However, in an attempt to address this issue, we have reanalyzed our data with 1.5 years of follow-up time, calculating the glomerular filtration rate using the modification of diet in renal disease (MDRD) equation, which is not dependent on body weight values. The results were essentially the same as those of the published 1-year analysis that used the Cockcroft-Gault equation. Tenofovir DF recipients had greater increases in serum creatinine level and greater absolute and percentage declines in the glomerular filtration rate, compared with NRTI recipients, although there was again no difference in the rate of discontinuation coincident with the maximum decline in the glomerular filtration rate. The median changes in glomerular filtration rate in the tenofovir DF arm and in the NRTI arm at month 6 after the start of therapy were -14 and -2 mL/min per 1.73 m2, respectively (P = .01); at month 12, they were -18 and -10 mL/min per 1.73 m2, respectively (P = .005); and at month 18, they were -19 and -11 mL/min per 1.73 m2, respectively (P = .005). The greatest difference in the slope of the curves for the 2 arms occurred through month 6, with the difference between therapies remaining relatively stable thereafter. The percentage change in glomerular filtration rate was also associated with longer duration of treatment with tenofovir DF or NRTI, with a CD4 cell count of <50 cells/mm3 and an HIV RNA level of >100,000 copies/mL (P < .05).
In a multivariate analysis, only tenofovir DF use, duration of therapy, and lower baseline CD4 cell count were associated with a decrease in the glomerular filtration rate (P < .05). In this analysis, diabetes, hypertension, use of lopinavir-ritonavir or any other concomitant antiretroviral agents, use of cotrimoxazole, previous use of adefovir, age, sex, race, and HIV transmission risk group were not associated with a decrease in the glomerular filtration rate.
We believe that this reanalysis of data from our cohort study supports our earlier published findings and suggests that the decline in the creatinine clearance rate is real and not an artifact of body composition changes induced by antiretroviral therapy. However, with longer follow-up, it appears that the changes in renal function are most pronounced shortly after the initiation of therapy and begin to level off with time. If confirmed, this observation suggests that the loss of renal function seen in some patients taking tenofovir DF may not be progressive. Moreover, as we stated in the original article, the decline in renal function is modest and of unclear clinical significance.
Acknowledgments
Potential conflicts of interest. J.E.G. has been a consultant and received honoraria and research support from Gilead Sciences. R.D.M. has received research funding with Bristol-Myers Squibb and Gilead Sciences. M.A.P. and J.C.K.: no conflicts.
|
|
| |
| |
|
|
|
