More than a decade after the introduction of highly active antiretroviral therapy (HAART) in 1996, concern has been raised regarding the renal effects of long-term HIV infection and its treatment. Although HAART has improved survival considerably, patients are at increased risk of chronic kidney disease due to a number of factors, including ageing and HAART-related metabolic complications—hypertension,1 diabetes mellitus,2 dyslipidemia3—and the adverse effects of some antiretroviral drugs, especially tenofovir.4,5 Indeed, kidney function is abnormal in up to 30% of HIV-infected patients, and kidney disease may be associated with progression to AIDS and death.6–8 In addition, an independent association between decreased kidney function and the risk of cardiovascular events has been demonstrated in HIV-infected patients.9 Most studies have investigated the association between the presence of cardiovascular disease (CVD) and impaired kidney function using the values for estimated glomerular filtration rate (eGFR) established for chronic kidney disease in the US National Kidney Foundation Guidelines (<60 mL·min−1·1.73 m−2)10; however, most patients experience only mild renal function abnormalities whose clinical significance has yet to be defined.11 Importantly, in the general population, the presence of low-grade albuminuria, a low eGFR, and a rapid decline in kidney function have been associated with all-cause and cardiovascular mortality.6,12–14
The Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) recently developed a formula to calculate the eGFR–EPI equation, which has proved to be more accurate than the routinely used modification of diet in renal disease (MDRD) formula, especially in patients with normal kidney function.15–17
In this context, the early and accelerated atherosclerotic process described in HIV-infected patients18 may be an additional contributing factor in kidney disease. Prompt diagnosis of subclinical atherosclerosis may enable aggressive management of contributing factors before the onset of clinical CVD. Several noninvasive approaches have been suggested to detect patients at increased risk of CVD, including measurement of carotid intima-media thickness (cIMT), which has proved to be an important predictor of cardiovascular events in both the general population19,20 and in HIV-infected patients.21
Given that mild renal abnormalities have been associated with an increased incidence of cardiovascular events in community-based populations12–14 and that indirect measurement of subclinical atherosclerosis such as cIMT has demonstrated a relationship between these abnormalities and cardiovascular events,19,20 we assessed the hypothesis that the presence of incipient alterations in renal function could independently predict the presence of subclinical atherosclerosis in HIV-infected patients.
Study Design, Participants, Setting, and Eligibility
We conducted an observational cross-sectional study of 145 consecutive HIV-infected patients who attended our university based HIV clinics in a tertiary hospital in Madrid between 2009 and 2010. The exclusion criteria were known CVD (previous stroke, myocardial infarction, or intermittent claudication) and/or known chronic kidney disease. The local ethics committee approved the study, and all patients gave their written informed consent to participate.
Clinical and Laboratory Measurements
Medical records were carefully reviewed at interview, a questionnaire was completed, and a thorough physical examination was performed. Gender, age, body mass index, smoking status, family history of early CVD, and treatment with antiretroviral drugs were recorded. The presence of hypertension, hypercholesterolemia, hypertriglyceridemia, and metabolic syndrome was defined according to the Adult Treatment Panel III criteria.22 Lipodystrophy was defined as the presence of body fat changes that could be clearly recognized by the patient and confirmed by the doctor and diagnosed following the lipodystrophy severity grading scale of Lichtenstein et al.23
A sample of fasting venous blood was obtained to determine concentrations of glucose, insulin, interleukin-6, total cholesterol, high-density lipoprotein cholesterol, and triglycerides using standard enzymatic methods. Insulin resistance was calculated with the following equation: HOMA (homeostasis model assessment) = insulin (μU/mL) x glucose (mg/dL)/22.5. Low-density lipoprotein cholesterol concentrations were calculated using the Friedewald equation.24
Plasma viral load was measured using the Cobas TaqMan HIV-1 assay (Roche Diagnostics Systems, Inc, Branchburg, NJ), and CD4 lymphocyte count was determined by flow cytometry (Beckman-Coulter, Inc, Münster, Germany). Hepatitis C virus coinfection was diagnosed by a positive serology result using a standard enzyme-linked immunosorbent assay.
Plasma levels of high-sensitivity C-reactive protein (CRP) were measured using nephelometry (VISTA System, Siemens Healthcare Diagnostics Inc, Deerfield, IL). D-dimers were measured using turbidimetry (Beckman-Coulter, Inc, Münster, Germany) and fibrinogen concentration using a prothrombin time–derived method. N-terminal pro-B–type natriuretic peptide was measured using a luminescent oxygen-channeling assay (VISTA System, Siemens Healthcare Diagnostics Inc).
Renal Function Assessment
One first morning urine sample was collected from each participant and urinary albumin and creatinine concentrations were determined by turbidimetry (Olympus Diagnostics AU2700 autoanalyzer). Creatinine was measured using the kinetic Jaffe method (Olympus Diagnostics AU2700 autoanalyzer), and C-cystatin was analyzed by nephelometry (BN-Prospect System, Siemens Diagnostics). Albuminuria was calculated using the albumin/creatinine urine ratio (ACR) measured in 1 urine sample.
eGFR was calculated using the CKD-EPI formula.16 The decrease in eGFR was determined by calculating the difference in eGFR value 3 years before the onset of the study and at the onset of the study. Given the absence of a standardized definition of IRI, several definitions were analyzed using data from studies that had previously explored the association between IRI and CVD.12–14 The IRI definitions assessed were as follows: “A” = stage 2 chronic kidney disease (eGFR between 60 and 90 mL/min)10 ± rapid decline in eGFR (>3% per year); “B” = “A” ± ACR >5 mg/g (median); “C” = “A” ± ACR >10 mg/g; “D” = “A” ± ACR >30 mg/g.12,13,21,22
Measurement of cIMT
cIMT measurements were obtained by ultrasonography (HD7 model, US Philips) according to the Mannheim Criteria.25 Measurements were made at the common carotid artery (1 cm proximal to the bifurcation). Far wall cIMT images were obtained and digitalized for each patient. cIMT detection software was previously calibrated using QLab (Advanced Quantification Software).26 More than 400 measurements were performed for each patient, and the median value was used for the statistical analyses. A plaque was defined as a thickness >1.5 mm or a focal structure that encroached into the arterial lumen by at least 0.5 mm or at least 50% of the surrounding cIMT value. Participants were categorized in 2 groups: those who presented a cIMT ≥ 75th percentile or in whom a plaque was demonstrated were considered to be at increased CVD risk and were classified as having subclinical atherosclerosis; the remaining patients were included in the comparison group. Several factors were taken into consideration before cIMT was classified as being above the 75th percentile. First, cIMT correlates with the incidence of cardiovascular events, and no threshold effect is observed.20 Second, cIMT depends largely on age; however, reference values of cIMT adjusted for age are lacking in our population. Third, in our study, the dispersion in mean age was slight. Finally, the 75th percentile is indicative of increased CVD risk according to the American Society of Echocardiography cIMT Task Force.27 Measurements were performed by 2 trained technicians who had previously participated in a pilot study (repeated and blinded measurements performed in 29 patients). The intraclass correlation coefficient was >0.90.
Qualitative variables were summarized as a frequency distribution and normally distributed quantitative variables as mean ± standard deviation. The continuous nonnormally distributed variables were summarized as median and interquartile range. Means for variables with a normal distribution were compared using the t test. Nonparametric variables were studied using the Mann–Whitney test. Regression analysis and the Pearson correlation coefficient were used to evaluate the relationship between cIMT and eGFR. The Spearman rank correlation was used to evaluate the relationship between cIMT and ACR in first morning urine samples. A logarithmic transformation of the raw ACR data was applied to meet the requirements of regression analysis. Patients with and without IRI were compared using the χ2 test or Fisher exact test when more than 25% of the expected values were less than 5. Logistic regression analysis was used to study the variables associated with the presence of subclinical atherosclerosis (as a binary variable).
Of the 4 IRI definitions proposed, we selected the one which best classified patients according to their degree of atherosclerosis by calculating the sensitivity, specificity, and area under the curve using a logistic regression model.
A logistic regression model was built to evaluate the effect of IRI on the degree of subclinical atherosclerosis. The variables included in the model were those that had a P value <0.10 and/or were clinically relevant between patients with and without IRI. The adjustment strategy involved calculation of the crude odds ratio of the association between the presence of IRI and the degree of subclinical atherosclerosis followed by individual adjustment of the effect of each potential confounding variable using a series of bivariate models. If any of these variables led to a change in the initial odds ratio for the presence of IRI >10%, they were included in a multivariate model (full model). Finally, the adjusted odds ratio (OR) and its 95% confidence interval (95% CI) were calculated. The null hypothesis was rejected by a type I error <0.05 (α < 0.05). Statistical analyses were performed using SPSS 15.0.
Table 1 summarizes the characteristics of the 145 HIV-infected participants included in the study. The study sample was representative of a middle-aged (41 ± 10 years) HIV-positive population, and the most frequent cardiovascular risk factor was smoking (46.5%), followed by hypertriglyceridemia (43.4%), hypercholesterolemia (38.6%), hypertension (16.6%), and diabetes mellitus (9.0%). Metabolic syndrome was detected in 12.1% of patients. Most patients were on HAART (91.4%) and had an undetectable viral load (77.1%).
Mean cIMT was 0.59 (0.13) mm [median: 0.56 (0.50–0.65) mm]. The association between cIMT as a continuous variable with eGFR and ACR is shown in Figures 1, 2. cIMT correlated inversely with eGFR (r = −0.381, P < 0.001, Pearson correlation) and positively with log ACR (r = 0.35, P < 0.001, Spearman correlation). The presence of plaque was demonstrated in 11 (6.4%) cases, and 44 individuals (31.0%) were classed as having subclinical atherosclerosis. Although patients with subclinical atherosclerosis had a lower eGFR and a higher ACR than patients without subclinical atherosclerosis, both groups showed normal eGFR and ACR below the threshold of microalbuminuria (30 mg/g) (Table 2). No differences were found for creatinine and C-cystatin.
Table 3 summarizes the sensitivity, specificity, and area under the curve of the 4 different definitions of IRI for the diagnosis of subclinical atherosclerosis. The definition that best classified patients was “B”.
Association Between IRI and Subclinical Atherosclerosis
A total of 95 patients (64.1%) presented IRI according to definition “B”. Table 1 summarizes the characteristics of these patients.
HIV patients with IRI were older and had a higher waist circumference than patients without IRI. Hypertriglyceridemia, diabetes, and metabolic syndrome were more frequent in patients with IRI. Regarding HIV-related variables, patients with IRI more frequently had lipodystrophy, a lower CD4 lymphocyte count, lower CD4 lymphocyte nadir, undetectable viral load, and longer exposure to antiretroviral therapy. However, IRI was not associated with accumulated exposure to tenofovir or with ongoing treatment with tenofovir.
In the univariate analysis, the presence of IRI was statistically associated with the presence of subclinical atherosclerosis (OR: 4.3; 95% CI: 1.7 to 10.6; P = 0.001). Subsequently, a logistic regression model was built to analyze the presence of IRI as an independent predictor of subclinical atherosclerosis. We examined the impact of a number of potential confounding variables (age, diabetes mellitus, hypertension, hypertriglyceridemia, time to HIV diagnosis, accumulated exposure to antiretroviral drugs, CD4 lymphocyte count, detectable viral load, lipodystrophy, and CRP levels) by constructing a series of bivariate models according to the strategy previously described. Finally, we built the “full model” including the following confounding variables: age, diabetes mellitus, use of angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, and accumulated exposure to nonnucleoside reverse transcriptase inhibitors and protease inhibitors. Multivariate analysis showed the presence of IRI to be independently associated with subclinical atherosclerosis (OR: 3.8; 95% CI: 1.3 to 11.0; P = 0.013).
Cardiovascular Biomarkers and IRI
HIV-infected patients with IRI tended to show higher levels of circulating cardiovascular biomarkers (CRP, fibrinogen, D-dimers, N-terminal pro-B–type natriuretic peptide) than patients without IRI; however, these increments were not statistically significant (Table 4).
Principal Findings and Comparisons With the Literature
Although several studies have demonstrated that HIV-infected patients have thicker cIMT than noninfected individuals, the factors associated with subclinical atherosclerosis in these patients remain unclear. In addition, the presence of IRI is common in these individuals and has proved to be an independent marker of cardiovascular events in the general population7,12,13; however, to our knowledge, its association with subclinical atherosclerosis has not yet been investigated in HIV-infected patients. In our study, individuals with IRI, as determined by slight decreases in eGFR, a rapid decline in kidney function and/or low-grade albuminuria, had a 4-fold higher risk of subclinical atherosclerosis. Importantly, this increased risk was present at levels below the current threshold for microalbuminuria.
Current Infectious Diseases Society of America guidelines advocate evaluation of both eGFR and proteinuria when assessing renal function in HIV-infected patients.28 The Infectious Diseases Society of America recommends using eGFR measured by serum creatinine–based equations in routine clinical practice, mainly the MDRD equation. However, serum creatinine among HIV-infected persons can vary considerably according to body mass, associated metabolic abnormalities, and exposure to drugs that affect renal tubular creatinine secretion.29,30 Furthermore, the MDRD equation tend to underestimate actual GFR in subjects with normal or only mildly impaired kidney function,31,32 as is the case in most HIV-infected patients.11 More recently, the CKD-EPI equation has proved to provide GFR estimates with better accuracy and less bias than the MDRD equation16,33 and, in contrast to the MDRD equation, eGFR >60 ml/min/1.73 m2 can be calculated accurately using the CKD-EPI equation.34 Although a growing body of evidence supports the use of this new equation, results should be confirmed in larger groups; however, for our study, it seemed to be the most appropriate estimator of eGFR, as we aimed to detect patients in the earliest stages of kidney disease.
Previous reports on patients with proteinuria clearly show that the presence of microalbuminuria (ACR >30 mg/g or a positive urine dipstick result) is associated with an increased risk of death.35,36 In our study, the median ACR value was low (5 mg/g). As mentioned above, the presence of low-grade albuminuria, a low eGFR, and a rapid decline in kidney function have also been associated with all-cause and cardiovascular mortality in community-based populations.6,12–14 In this study, we demonstrated a relationship between the presence of IRI and increased cIMT, a validated surrogate marker of cardiovascular events in HIV-infected patients.
Possible Mechanisms for the Observed Association
We suggest that the link observed between subclinical atherosclerosis and the presence of IRI reflects different consequences of a common phenomenon: the accelerated atherosclerotic process that is present in HIV-infected persons.
Both the proatherogenic effect of HIV itself through HIV-associated systemic inflammation and antiretroviral-induced metabolic disorders have been suggested to predispose to premature atherosclerosis.37,38 Patients with IRI also showed a higher prevalence of traditional cardiovascular risk factors, in particular hypertrygliceridemia and diabetes.
Several studies have suggested a proatherogenic effect of HIV through HIV-associated systemic inflammation. HIV-infected patients with IRI tended to show higher levels of circulating cardiovascular biomarkers supporting the presence of a low-grade systemic inflammation than patients without IRI. In non–HIV-infected patients, a low-grade systemic inflammation has been associated with premature atherosclerosis39 and the incidence of CVD events.40
The possible influence of antiretroviral treatment on IRI cannot be completely ruled out, given the relationship observed between cumulative use of antiretroviral drugs and renal impairment in the univariate analysis. Although we detected a relationship between cumulative use of antiretroviral therapy and the presence of IRI, the use of tenofovir, a generally well-tolerated drug that may cause tubular toxicity, was not statistically related. The limited number of patients in our series precludes us from drawing conclusions on the effect of individual drugs, but we consider that this hypothesis should be explored in greater depth.
Management of HIV-infected patients to date has been based on optimization of 2 following surrogate parameters: viral load and the CD4 lymphocyte count. However, mortality data show that although HIV-infected persons are clearly living longer as a consequence of effective HAART, they may be dying earlier than the general population from conditions not traditionally associated with HIV infection, such as kidney disease and CVD.41 Nowadays, one of the clinical challenges in the management of these individuals is the noninvasive detection of patients at increased risk of CVD to prevent the progression of atherosclerosis. We demonstrated that mildly decreased eGFR associated with a rapid decline in kidney function and/or albuminuria levels within the normal range was an important prognostic factor for increased cIMT. Although measurement of cIMT is considered the gold standard for the noninvasive detection of subclinical atherosclerosis, this approach cannot be generalized, as it is expensive and not universally available. Therefore, periodic monitoring of eGFR and ACR might help to better identify subjects at increased risk of CVD to initiate aggressive management of risk factors. Our findings provide support for the hypothesis that mild abnormalities of renal function can independently predict an increased atherosclerotic burden and behave as a useful surrogate marker of subclinical atherosclerosis.
Our study is subject to a series of limitations. First, cross-sectional studies can neither prove causality nor distinguish between risk and prognostic factors for a disease, in this case, subclinical atherosclerosis. Second, we used a surrogate marker of cardiovascular events, namely, cIMT, instead of studying the incidence of coronary events. Nevertheless, cIMT is a validated surrogate marker of cardiovascular mortality and is the only test recommended by the American Heart Association for the assessment of the burden of atherosclerotic plaque.18 Third, although we identified the presence of subclinical atherosclerosis with degrees of albumin excretion as low as 5 mg/g, further studies are needed to investigate the generalizability of our results to other HIV patient populations. In addition, ACR was measured only once in our study. Although it is unknown whether a confirmatory ACR determination should be performed when evaluating the presence of low-grade albuminuria, there is a potential risk of misclassification bias in our analysis, as urinary albumin excretion may provide false results under specific circumstances.42 Fourth, given the sample size and the number of events of the dependent variable, we could not simultaneously adjust the effect for all the possible confounders, only for those variables producing a significant change in the crude odds ratio (>10%). However, we believe that the effect of the presence of IRI is properly adjusted, given the weakness of any possible residual confounders. Finally, most subjects in this cohort had IRI (64.1%), thus reflecting the poor specificity of definition “B”. Consequently, our data must be interpreted carefully and should be reproduced in further studies in HIV-infected patients. It is also necessary to evaluate strategies to improve the diagnostic values of IRI before considering it a reliable marker of subclinical atherosclerosis when screening for CVD. In our opinion, these limitations are overwhelmed by the proof of concept that mild abnormalities of kidney function within the normal range may reflect an increased risk for CVD.
IRI is an independent predictor of increased cIMT (OR: 3.8; 95% CI: 1.3 to 11) and could be an easy and accessible way to identify patients with greater subclinical atherosclerosis and, therefore, an increased risk of CVD.
The authors would like to thank Thomas O'Boyle for his critical review and suggestions on the article and Maria Rodrigo for her work in data collection.
1. Seaberg EC, Munoz A, Lu M, et al.. Association between highly active antiretroviral therapy and hypertension in a large cohort of men followed from 1984 to 2003. AIDS. 2005;19:953–960.
2. Brown TT, Cole SR, Li X, et al.. Antiretroviral therapy and the prevalence and incidence of diabetes mellitus in the multicenter AIDS cohort study. Arch Intern Med. 2005;165:1179–1184.
3. Riddler SA, Smit E, Cole SR, et al.. Impact of HIV infection and HAART on serum lipids in men. JAMA. 2003;289:2978–2982.
4. Calza L, Trapani F, Tedeschi S, et al.. Tenofovir-induced renal toxicity in 324 HIV-infected, antiretroviral-naive patients. Scand J Infect Dis. 2011;43:656–660.
5. Horberg M, Tang B, Towner W, et al.. Impact of tenofovir on renal function in HIV-infected, antiretroviral-naive patients. J Acquir Immune Defic Syndr. 2010;53:62–69.
6. Gardner LI, Holmberg SD, Williamson JM, et al.. Development of proteinuria or elevated serum creatinine and mortality in HIV-infected women. J Acquir Immune Defic Syndr. 2003;32:203–209.
7. Gupta SK, Mamlin BW, Johnson CS, et al.. Prevalence of proteinuria and the development of chronic kidney disease in HIV-infected patients. Clin Nephrol. 2004;61:1–6.
8. Szczech LA, Hoover DR, Feldman JG, et al.. Association between renal disease and outcomes among HIV-infected women receiving or not receiving antiretroviral therapy. Clin Infect Dis. 2004;39:1199–1206.
9. George E, Lucas GM, Nadkarni GN, et al.. Kidney function and the risk of cardiovascular events in HIV-1-infected patients. AIDS. 2010;24:387–394.
10. Levey AS, Coresh J, Balk E, et al.. National Kidney Foundation practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Ann Intern Med. 2003;139:137–147.
11. Deti EK, Thiebaut R, Bonnet F, et al.. Prevalence and factors associated with renal impairment in HIV-infected patients, ANRS C03 Aquitaine Cohort, France. HIV Med. 2010;11:308–317.
12. Arnlov J, Evans JC, Meigs JB, et al.. Low-grade albuminuria and incidence of cardiovascular disease events in nonhypertensive and nondiabetic individuals: the Framingham heart study. Circulation. 2005;112:969–975.
13. Matsushita K, van d V, Astor BC, et al.. Association of estimated glomerular filtration rate and albuminuria with all-cause and cardiovascular mortality in general population cohorts: a collaborative meta-analysis. Lancet. 2010;375:2073–2081.
14. Shlipak MG, Katz R, Kestenbaum B, et al.. Rapid decline of kidney function increases cardiovascular risk in the elderly. J Am Soc Nephrol. 2009;20:2625–2630.
15. Levey AS, Coresh J, Greene T, et al.. Using standardized serum creatinine values in the modification of diet in renal disease study equation for estimating glomerular filtration rate. Ann Intern Med. 2006;145:247–254.
16. Levey AS, Stevens LA, Schmid CH, et al.. A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009;150:604–612.
17. Levey AS, Stevens LA. Estimating GFR using the CKD Epidemiology Collaboration (CKD-EPI) creatinine equation: more accurate GFR estimates, lower CKD prevalence estimates, and better risk predictions. Am J Kidney Dis. 2010;55:622–627.
18. Hsue PY, Lo JC, Franklin A, et al.. Progression of atherosclerosis as assessed by carotid intima-media thickness in patients with HIV infection. Circulation. 2004;109:1603–1608.
19. Lorenz MW, von Kegler S, Steinmetz H, et al.. Carotid intima-media thickening indicates a higher vascular risk across a wide age range: prospective data from the Carotid Atherosclerosis Progression Study (CAPS). Stroke. 2006;37:87–92.
20. Lorenz MW, Markus HS, Bots ML, et al.. Prediction of clinical cardiovascular events with carotid intima-media thickness: a systematic review and meta-analysis. Circulation. 2007;115:459–467.
21. Mangili A, Polak JF, Quach LA, et al.. Markers of atherosclerosis and inflammation and mortality in patients with HIV infection. Atherosclerosis. 2011;214:468–473.
22. NCEP Expert Panel. Adult Treatment Panel III. Third report of the National Cholesterol Education Program (NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (adult treatment panel III) final report. Circulation. 2002;106:3143–3421.
23. Lichtenstein KA, Ward DJ, Moorman AC, et al.. Clinical assessment of HIV-associated lipodystrophy in an ambulatory population. AIDS. 2001;15:1389–1398.
24. Gomez F, Camps J, Simo JM, et al.. Agreement study of methods based on the elimination principle for the measurement of LDL- and HDL-cholesterol compared with ultracentrifugation in patients with liver cirrhosis. Clin Chem. 2000;46:1188–1191.
25. Touboul PJ, Hennerici MG, Meairs S, et al.. Mannheim carotid intima-media thickness consensus (2004–2006). An update on behalf of the advisory board of the 3rd and 4th watching the risk symposium, 13th and 15th European stroke conferences, Mannheim, Germany, 2004, and Brussels, Belgium, 2006. Cerebrovasc Dis. 2007;23:75–80.
26. Secil M, Altay C, Gulcu A, et al.. Automated measurement of intima-media thickness of carotid arteries in ultrasonography by computer software. Diagn Interv Radiol. 2005;11:105–108.
27. Stein JH, Korcarz CE, Hurst RT, et al.. Use of carotid ultrasound to identify subclinical vascular disease and evaluate cardiovascular disease risk: a consensus statement from the American Society of Echocardiography Carotid Intima-Media Thickness Task Force. Endorsed by the society for vascular medicine. J Am Soc Echocardiogr. 2008;21:93–111.
28. Gupta SK, Eustace JA, Winston JA, et al.. Guidelines for the management of chronic kidney disease in HIV-infected patients: recommendations of the HIV medicine association of the infectious diseases society of America. Clin Infect Dis. 2005;40:1559–1585.
29. Maia BS, Engelson ES, Wang J, et al.. Antiretroviral therapy affects the composition of weight loss in HIV infection: implications for clinical nutrition. Clin Nutr. 2005;24:971–978.
30. Roy MT, First MR, Myre SA, et al.. Effect of co-trimoxazole and sulfamethoxazole on serum creatinine in normal subjects. Ther Drug Monit. 1982;4:77–79.
31. Bostom AG, Kronenberg F, Ritz E. Predictive performance of renal function equations for patients with chronic kidney disease and normal serum creatinine levels. J Am Soc Nephrol. 2002;13:2140–2144.
32. Rule AD, Larson TS, Bergstralh EJ, et al.. Using serum creatinine to estimate glomerular filtration rate: accuracy in good health and in chronic kidney disease. Ann Intern Med. 2004;141:929–937.
33. Bonjoch A, Bayes B, Riba J, et al.. Validation of estimated renal function measurements compared with the isotopic glomerular filtration rate in an HIV-infected cohort. Antiviral Res. 2010;88:347–354.
34. Stevens LA, Schmid CH, Greene T, et al.. Comparative performance of the CKD Epidemiology Collaboration (CKD-EPI) and the Modification of Diet in Renal Disease (MDRD) Study equations for estimating GFR levels above 60 mL/min/1.73 m2. Am J Kidney Dis. 2010;56:486–495.
35. Choi A, Scherzer R, Bacchetti P, et al.. Cystatin C, albuminuria, and 5-year all-cause mortality in HIV-infected persons. Am J Kidney Dis. 2010;56:872–882.
36. Wyatt CM, Hoover DR, Shi Q, et al.. Microalbuminuria is associated with all-cause and AIDS mortality in women with HIV infection. J Acquir Immune Defic Syndr. 2010;55:73–77.
37. Hsue PY, Hunt PW, Schnell A, et al.. Role of viral replication, antiretroviral therapy, and immunodeficiency in HIV-associated atherosclerosis. AIDS. 2009;23:1059–1067.
38. Friis-Moller N, Reiss P, Sabin CA, et al.. Class of antiretroviral drugs and the risk of myocardial infarction. N Engl J Med. 2007;356:1723–1735.
39. Amer MS, Elawam AE, Khater MS, et al.. Association of high-sensitivity C-reactive protein with carotid artery intima-media thickness in hypertensive older adults. J Am Soc Hypertens. 2011;5:395–400.
40. Ridker PM. Clinical application of C-reactive protein for cardiovascular disease detection and prevention. Circulation. 2003;107:363–369.
41. Palella FJ Jr, Baker RK, Moorman AC, et al.. Mortality in the highly active antiretroviral therapy era: changing causes of death and disease in the HIV outpatient study. J Acquir Immune Defic Syndr. 2006;43:27–34.
42. Nguyen MT, Maynard SE, Kimmel PL. Misapplications of commonly used kidney equations: renal physiology in practice. Clin J Am Soc Nephrol. 2009;4:528–534.