Chronic Kidney Disease in the Aging Human Immunodeficiency Virus–Infected Population
Year: 2017
Type: article
Abstract: (See the major article by Kooij et al, on pages 622–31.) In this issue of The Journal of Infectious Diseases, Kooij and colleagues report on the prevalence of chronic kidney disease (CKD) in a prospective, longitudinal cohort of 598 human immunodeficiency virus–infected (HIV+) individuals, with a median age of 45 years [1]. This carefully performed study has deepened our understanding of kidney disease among middle-aged HIV+ subjects with long-term exposure to antiviral therapy, having a median duration of therapy of 10 years. The principal renal findings in the HIV+ cohort, compared to an age-matched human immunodeficiency virus–uninfected (HIV–) cohort, were increased rates of tubular dysfunction (odds ratio [OR], 7); albuminuria, defined as urine albumin/creatinine ratio >3 mg/mmol (equivalent to 27 mg/g) (OR, 6); and reduced estimated glomerular filtration rate (eGFR)–creatinine (OR, 2). The study had 3 principal renal outcomes: CKD, albuminuria, and proximal tubular dysfunction. CKD was defined as eGFR <60 mL/minute/1.73 m2, based on serum creatinine. Serum creatinine is chiefly influenced by muscle mass and by glomerular filtration rate; to the extent that some HIV+ individuals may have reduced muscle mass, this would tend to underestimate the prevalence of CKD. In multivariable analysis, variables significantly associated with baseline CKD included only age. Associations with baseline albuminuria were adjusted for diabetes, hypertension, dyslipidemia, and smoking; variables that remained significant included age, current smoking, use of hypertensive drugs, systolic blood pressure, diabetes, nadir body weight, and soluble CD14 (a lipopolysaccharide receptor expressed mainly on macrophages that is a marker for macrophage activation). The authors point out that chronic inflammation, present even when HIV replication is suppressed, is present and is a likely driver of CKD in this population. Nevertheless, at baseline, markers of inflammation were not different between those who had reduced eGFR and those who had normal eGFR. Associations with baseline proximal tubular dysfunction (defined as increased urine retinol binding protein and increased fractional excretion of phosphate) included age, smoking, cardiovascular disease, nadir body weight, protease inhibitor use, and, most strongly of all, cumulative tenofovir disoproxil fumarate (TDF) use. The study offers a longitudinal perspective regarding the effects of HIV infection on the aging kidney. Over a median of 4 years of follow-up, the HIV+ group (n = 377) manifested faster GFR decline (a difference of –0.56 mL/minute/year) and worsening albuminuria compared to the HIV– group (n = 479). Rapid GFR decline, defined as >5 mL/minute/1.73 m2/year, was more common among HIV+ individuals (5.8%) vs controls (2.3%). Interestingly, baseline albuminuria predicted faster eGFR decline in the controls, but not the HIV+ group. One hypothesis might be that albuminuria in HIV disease involves glomerular disease in some subjects and tubular disease in others (as well as metabolic syndrome in still others, as discussed below), and it is glomerular disease that is most likely to progress, resulting in falling eGFR. Current atazanavir use (but not TDF use) and duration of prolonged T-cell lymphopenia also correlated with GFR decline. The study has several important strengths. The cohorts were of sufficient size (596 HIV+ and 544 HIV– subjects) to provide reasonable power. The longitudinal design provides insight into progression of CKD. The study also has certain limitations. Although the disease and control cohorts were generally well matched, there was an excess of individuals of African descent (14% vs 6%), current smokers (32% vs 25%), and hypertensive subjects (50% vs 37%) in the HIV+ group. All 3 factors are associated with an increased risk for CKD. By contrast, the rate of diabetes mellitus was similar (7% vs 5%) between the groups. The authors do not present data on the fraction of each subgroup that had albuminuria and/or reduced GFR at the end of the study. Studies from several large cohorts have documented increased rates of CKD in the setting of HIV infection; Table 1 presents several cohorts that have included a comparison group, either HIV– controls or population databases [1–3]. The cohorts differ substantially in the fractions that are of African descent (an important risk for kidney disease in HIV+ individuals, in large part due to genetic variation in APOL1 [4]), and are composed of men having sex with men (most cases in the Dutch study, and 46% of cases in the Danish study). Recent Studies Comparing Chronic Kidney Disease Incidence or Prevalence in HIV-Infected and HIV-Uninfected Groups African Americans were present only in the 2 North American cohorts. Men having sex with men predominated in the Dutch cohort and comprised 46% of the Danish cohort, and were likely a smaller subset of the North American cohorts. Intravenous drug users were present in all cohorts. Abbreviations: eGFR, estimated glomerular filtration rate; ESKD, end-stage kidney disease; HIV, human immunodeficiency virus; OR, odds ratio; USRDS, United States Renal Data Systems. Recent Studies Comparing Chronic Kidney Disease Incidence or Prevalence in HIV-Infected and HIV-Uninfected Groups African Americans were present only in the 2 North American cohorts. Men having sex with men predominated in the Dutch cohort and comprised 46% of the Danish cohort, and were likely a smaller subset of the North American cohorts. Intravenous drug users were present in all cohorts. Abbreviations: eGFR, estimated glomerular filtration rate; ESKD, end-stage kidney disease; HIV, human immunodeficiency virus; OR, odds ratio; USRDS, United States Renal Data Systems. An important consideration is whether to define eGFR based on serum creatinine (which may be reduced in HIV disease due to reduced muscle mass or increased by antiretroviral medications that interfere with tubular secretion of creatinine, eg, ritonavir, cobicstat, or dolutegravir) or serum cystatin C (which may be elevated in HIV disease due to increased inflammation) [6]. More precise assessment of GFR by plasma clearance or plasma decay of iothalamate or iohexol is not practical for large cohort studies. What are the likely causes of CKD in this and similar HIV+ cohorts? Without kidney biopsies, it cannot be known with certainty what is causing albuminuria and impaired eGFR, but several possibilities can be considered with regard to the Dutch cohort studied by Kooij et al. First, these subjects were likely exposed to several nephrotoxins, particularly TDF, which was taken by 85% of study subjects at some time, for a median of 4 years. TDF causes tubular injury, manifesting as tubular proteinuria; the newer therapeutic, tenofovir alafenamide, is much less likely to cause tubular injury. Tubular proteinuria is predominantly composed of low molecular weight proteins; these proteins are normally filtered to some extent through the glomerulus and are taken up by the proximal tubular epithelium. TDF that was taken by individuals in the present study causes chronic (and less commonly acute) tubular injury, leading to tubular proteinuria. What is the cause of albuminuria in this cohort? The urine albumin/creatinine ratio correlated with the retinol-binding protein/creatinine ratio (r = 0.44). This suggests that some portion of the albuminuria may have been due to tubular dysfunction, as the glomerular ultrafiltrate includes both retinol binding protein (molecular weight ~23 kDa) and, to a much lesser extent, albumin (molecular weight 67 kDa). Microalbuminuria (defined as urine albumin/creatinine ratio [ACR] 3–30 mg/mmol) was more common in HIV+ subjects (22% vs 5%); this may well represent a diverse array of glomerular endothelial or tubular dysfunction, and in some cases, glomerulosclerosis [7]. On the other hand, macroalbuminuria (defined as urine ACR > 3 mg/mmol) was much more common among HIV+ subjects (although the numbers were small, 26 cases and 1 case, respectively); this most certainly represents structural glomerular disease, chiefly glomerulosclerosis. What mechanisms might contribute to CKD in this study? HIV infection, even with suppressed viral replication to the extent that retroviral RNA is undetectable in serum, is a state of heightened systemic inflammation. Thus, various cohort studies have shown elevated serum markers of inflammation, including high-sensitivity C-reactive protein, interleukin 6, and cystatin C (the latter also a marker of glomerular filtration), markers of thrombosis (D-dimer), and markers of endothelial damage, including soluble intercellular cell adhesion marker 1 and soluble vascular cell adhesion marker 1 [8] Few studies have characterized the renal histologies that characterize the aging kidney in HIV+ individuals. Wyatt and colleagues examined autopsies in 89 individuals with long-term antiretroviral exposure, with an average age of 47; 50% were African American [9]. Of these, 75 had renal pathology, including arterionephrosclerosis (commonly misattributed to hypertension), HIV-associated nephropathy, and glomerulonephritis. HIV-associated chronic inflammation appears to be the common thread that weaves together metabolic syndrome, hypertension [10], neurocognitive disorders [11], cardiovascular diseases [12], and kidney diseases. There are conflicting data, such as in a study from Denmark, suggesting that only kidney disease, among these other chronic diseases, is more common in the HIV+ aging population [3]. Similarly, data from a cohort of US veterans found that end-stage kidney disease, but not myocardial infarction, occurred at a younger age among HIV+ individuals, compared with HIV– controls with similar age distribution [5]. Taken together, the data suggest that HIV+ individuals on antiretroviral therapy with long-term viral suppression have a proinflammatory state and that the kidney, and especially the glomerular endothelium, is a common and perhaps uniquely sensitive target of systemic inflammatory processes. Thus, the kidney, via assessment of albuminuria, may serve as the proverbial canary in the coal mine, signaling widely distributed danger in vascular endothelia throughout the body. Acknowledgments. This work was supported by the Intramural Research Program of the National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, Maryland. We are pleased to acknowledge helpful comments from Dr Christina Wyatt, Mount Sinai School of Medicine. Potential conflicts of interest. Author certifies no potential conflicts of interest. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.
Author: Jeffrey B Kopp
Language: English
Subfield: Nephrology
Field: Medicine
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W2626024086