Background
The recognition of kidney tubular dysfunction (KTD) in subjects treated with tenofovir (TDF) has prompted to include tubular and glomerular function as part of current HIV monitoring in most patients. Although CCR in 24 h urine is the most reliable method to determine GF, practical issues make CG or MDRD estimates more adequate in a daily basis. The correlation between all these methods used to assess GF, and the possible interference of KTD, has not been examined in the HIV population.
Methods
All consecutive HIV+ patients attending a reference HIV outclinic were assessed for CCR (mL/min) in 24 h urine. Subjects collecting an urine volume <450 mL were excluded. Estimates of CCR by CG and MDRD were calculated for each patient. Intraclass correlation (IC) and greatest survival-agreement plots for 75% of the population (SAP-75) were applied to test CG vs MDRD differences with respect to CCR. Correlation analyses were also performed according to the presence of glomerular impairment (CCR <60 mL/min) or KTD (at least two of the following: glucosuria, hyperaminoaciduria, hyperphosphaturia, hyperuricosuria or beta2-microglobulinuria, being present at least one of the first three).
Results
A total of 417 patients were examined (22% ARV-naïve, mean age 39 years-old, 87% males, mean CD4 count 445 cells/μL), Table 1.
Table 1.
| All | CCR >60 mL/min |
CCR <60 mL/min |
No KTDa | KTDb | a vs b [p] | |
|---|---|---|---|---|---|---|
| No. of patients (%) | 417 | 393 (94) | 24 (6) | 337 (86) | 54 (14) | |
| Mean CCR (mL/min) | 113 | 117 | 47 | 116 | 103 | 0.01 |
| Mean CG (mL/min) | 100 | 101 | 89 | 103 | 90 | 0.003 |
| Mean MDRD (mL/min) | 91 | 91 | 76 | 92 | 81 | <0.001 |
| IC [p] | 0.64 [<0.001] | 0.66 [<0.001] | 0.63 [0.02] | 0.59 [<0.001] | 0.77 [<0.001] | |
| SAP-75* [CG minus CCR] (mL/min) |
20.3 | 19.9 | 24.6 | 21.1 | 18.8 | |
| SAP-75** [MDRD minus CCR] (mL/min) |
21.3 | 21.3 | 20.8 | 24.1 | 14.8 | |
| Log-Rank [SAP-75* vs SAP-75**] |
0.05 | 0.04 | 0.3 | 0.04 | 0.38 |
Discussion
In HIV+ patients, indirect methods to assess GF show high correlation with calculated CCR. In patients with normal CCR, CG may be more accurate than MDRD to estimate GF. Glomerular dysfunction or KTD do not seem to alter the correlation between CG or MDRD and CCR. Significant reductions in GF, as assessed by CCR, CG or MDRD, are more frequent in patients with than without KTD.