Jim, I work with MS not HIV, we have only used the creatinine but kidney disease per say is rare in MS. For the studies we do that monitor kidney function they use creatinine.

Two articles wil help you to answer to this question. However, cystatin C may be a good marker for risk prediction, cardiovascular events, and  long term decrease in GFR in HIV-infected  patients, but data are lacking.

Gagneux-Brunon A, Delanaye P, Maillard N, Fresard A, Basset T, Alamartine E, et al. Performance of creatinine and cystatin C-based glomerular filtration rate estimating equations in a European HIV-positive cohort. AIDS. 2013 Jun 19;27(10):1573–81.

Inker LA, Wyatt C, Creamer R, Hellinger J, Hotta M, Leppo M, et al. Performance of Creatinine and Cystatin C GFR Estimating Equations in an HIV-positive population on Antiretrovirals. Journal of acquired immune deficiency syndromes (1999) [Internet]. 2012 Jul 26 [cited 2012 Aug 3]; Available from: http://www.ncbi.nlm.nih.gov/pubmed/22842844

I believe that the problem of patients who have AIDS is the same as that of other patients: marker and the formula used performances depend on the patient's weight, muscle mass and the value of the glomerular filtration rate.(The age also of course but I do not think there is currently elderly patients with AIDS). The CKD-EPI combined cystatin / creatinine formula looks best in various populations. Articles sent by A. Gagneux-Brunon also confirms this for these patients.

Generally when using drugs that target metabolic pathways and alter them, such as cancer chemotherapy, or ART, there can be errors induced in the usual endogenous metabolic markers used to roughly estimate GFR. Even PrEP will bump creatinine by 2%, and that is very possibly not renal at all. It is likely due to metabolic change as evidenced by increased creatine kinase levels. Cystatin C is also a metabolite, so the same comments apply, although differently since the metabolic pathways are different. I urge caution interpreting estimated GFR results from heuristic formulas of either creatinine or cystatin C. These formulas were generally calibrated using radiometric methods that are themselves not the best possible ones, and are often scaled to body size using methods that are themselves questionable. For example, we scale to body surface area knowing full well that skin surface does not produce urine. Cr clearance scales to body size differently than GFR, which latter scales roughly as Kleiber's law (weight to the 3/4 power).

Endogenous markers are especially problematic for sick patients, e.g., with fluid disturbances, renal insufficiency, muscle wasting. Consider for example, that there is no significant  meaning of a creatinine level in a patient who is post albumin infusion, post dialysis for renal insufficiency in the course of hepatorenal failure with ascites. The take home message is that the worse the patient condition, the less meaningful assaying an endogenous metabolite is. In those cases, the Tk-GV radiometric assay may yield a reasonable GFR measurement, where other radiometric methods are more sensitive to fluid disturbances. Tk-GV assay would require administering an exogenous radioactive GFR marker and assaying concentration in 4 blood samples over a four hour period, see file.

For both, plasma level is related inversely to GFR, exponentially. Thus they are both relatively insensitive to small decreases in GFR.

Thanks all, very helpful.