Investigating the Cumulative Incidence of Chronic Kidney Disease Following HIV Diagnosis and the Effects of Comorbidities and COVID-19 on Kidney Function Over Time in a Cohort of People With HIV
Open Access Deposited0.52–1.00). Black race was associated with a 67% greater risk of CKD compared with White race, and IVDU was associated with a nearly twofold increased risk of CKD. However, age, health insurance status, and alcohol abuse were not significantly associated with the risk of CKD. Conclusions Multimorbidity among people with CKD centers on a small set of highly connected cardiometabolic and behavioral conditions. Both hypertension and dyslipidemia were independently and strongly associated with incident CKD, consistent with the joint cumulative incidence results showing early and frequent co-occurrence. These findings emphasize the importance of aggressive blood pressure and lipid management to mitigate kidney disease progression in aging populations. Analysis 3 Objective The COVID-19 pandemic has posed new challenges to kidney health, particularly among PWH, who may have heightened vulnerability due to chronic inflammation, comorbidities, and long-term immune dysregulation. Although acute kidney injury related to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection is well documented, the long-term impact of COVID-19 on kidney function in PWH remains poorly understood. To address this gap, we assessed post–COVID-19 changes in kidney function, quantified the risk of a ≥25% reduction in eGFR, and evaluated the independent effect of COVID-19 infection compared with propensity-score-matched unexposed controls. Using longitudinal data, we employed repeated-measures linear mixed-effects (LME) models and logistic regression to estimate the magnitude and predictors of kidney function decline, clarifying how COVID-19 influences longitudinal kidney function change beyond traditional and HIV-related risk factors. Methods Adult participants were eligible if they had ≥1 pre- and post-COVID eGFR during the study period. COVID-19 cases were identified based on laboratory test results or International Statistical Classification of Diseases and Related Health Problems, Tenth Edition (ICD-10) codes, and required ≥1 month of follow-up. Participants with stage III–V CKD or eGFR <60 mL/min/1.73 m2 at diagnosis were excluded. PWH with COVID-19 were matched with unexposed PWH without COVID-19 at a 2
6.71–14.82, p < 0.0001), and dyslipidemia was associated with a nearly threefold increase risk of stage III–V CKD (HR = 2.73, p < 0.0001) and a nearly fivefold increased risk of stage IV/V CKD (HR = 4.80, p < 0.0001). The history of opportunistic infections increased the risk of CKD by 26%–43%, whereas longer tenofovir exposure decreased the risk by 7%–9% per year. A lower CD4 molecule (CD4)+ T cell count was associated with an increased risk of CKD, and viral suppression was associated with a lower risk of stage IV/V CKD (HR = 0.72, 95% CI
“Low and Declining” (18.39%, low baseline eGFR followed by rapid loss), “Intermediate and Mild Decline” (45.74%, moderate baseline eGFR followed by gradual decline), and “High and Declining” (35.87%, high baseline eGFR followed by modest decline). In adjusted models, Black race was significantly associated with a lower odds of belonging to the Intermediate and Mild Decline group compared with the High and Declining group (aOR = 0.49, 95% confidence interval [CI]
ever having an eGFR <60 or <30 mL/min/1.73 m2. We then constructed multimorbidity networks using co-occurrence matrices of chronic conditions, restricted to those present in >5% of participants and to participants with ≥2 comorbidities. We used mixed graphical models and the Louvain community detection algorithm to identify highly connected clusters. Finally, we estimated both the marginal and joint cumulative incidence of stage III–V and IV/V CKD with hypertension and dyslipidemia, and calculated conditional probabilities of CKD given each comorbidity. Results Of 8,335 eligible participants, 1,646 met the criteria for stage III–V CKD and 634 for stage IV/V CKD. The cohort was predominantly Black (76.8%). Hypertension was common (51.0% overall) and was more prevalent among those with worse kidney function, present in 72.8% of those with stage III–V CKD and 80.1% of those with stage IV/V CKD. Dyslipidemia and DM were also common in participants with a lower eGFR. Multimorbidity networks revealed two dense clusters
1.01–3.94), and higher serum creatinine (OR = 1.43, 95% CI
0.07–0.68). In the matched analysis (493 with and 913 without COVID-19), COVID-19 was associated with a decline in eGFR (β = −2.14 per 6 months, p < 0.001). At 24 months, the predicted cumulative change in eGFR was greater among PWH with COVID-19 (−5.91 mL/min/1.73 m2, 95% CI
0.99–5.33), while having public insurance was associated with lower odds of belonging to the Low and Declining group (aOR = 0.51, 95% CI
4.25–6.90, p < 0.0001) and a nearly tenfold increased risk of stage IV/V CKD (HR = 9.97, 95% CI
a cardiometabolic cluster (hypertension, dyslipidemia, and DM) and a liver/behavioral cluster (chronic liver disease and alcohol abuse). Hypertension exhibited the highest overall connectivity, linking to most other conditions. The marginal cumulative incidence was 25.7% for stage III–V CKD, 9.2% for stage IV/V CKD, 66.3% for hypertension, and 54.9% for dyslipidemia. Joint cumulative incidence curves showed earlier and more rapidly increasing risk of CKD and hypertension than CKD and dyslipidemia. In fully adjusted time-varying Cox models, hypertension was associated with a fivefold increased risk of stage III–V CKD (HR = 5.41, 95% CI
0.63–1.41). Conclusions COVID-19 was independently associated with kidney function decline in PWH but not with a ≥25% reduction in eGFR. Female sex, DM, and HBV coinfection increased the risk of declining kidney function, while higher CD4+ T cell counts were protective. The findings demonstrate that COVID-19 independently impacts kidney function decline beyond traditional and HIV-related risk factors.
[−7.59, −4.23]) compared with PWH without COVID-19 (−0.46 mL/min/1.73 m2, 95% CI
[−1.61, 0.69]). The proportion with a ≥25% reduction in eGFR did not differ significantly between PWH with and without COVID-19 (8.4% vs. 8.3%
1.80–10.09), DM (OR = 2.00, 95% CI
1 ratio using propensity scores based on sociodemographic, HIV-related, and comorbidity-related factors. The outcomes were the absolute eGFR change and a ≥25% reduction in eGFR over 24 months following COVID-19 diagnosis. We estimated longitudinal change using repeated-measures LME models (β, mL/min/1.73 m2 per 6 months) a random intercept for matched sets and participant-level random intercepts and slopes for time. We used logistic regression models to assess the binary outcome of a ≥25% reduction in eGFR, reported as ORs with 95% CIs. Results Among 517 participants with COVID-19 (median age 54 years and 71% Black), 41 (7.4%) developed stage III–V CKD. Lower serum ALB (β = 4.89, p = 0.0130), higher serum creatinine (β = 0.99, p = 0.0013), and hepatitis B virus (HBV) infection (β = −9.32, p = 0.0003) were significantly associated with a decline in eGFR. Female sex (OR = 4.26, 95% CI
OR = 0.95, 95% CI
0.31–0.85). Conclusions These findings emphasize the importance of age-specific kidney risk assessment and early detection of DM-associated CKD in this population. HIV diagnosis during the HAART era was associated with an increased risk of CKD, likely reflecting older age and longer cumulative exposure to HIV and earlier ART regimens. Integrating metabolic and kidney health management into HIV care, along with close monitoring of ALB levels, may help identify and protect those most vulnerable to CKD progression. Analysis 2 Objectives Despite improvements in survival and sustained engagement in care, PWH increasingly experience CKD alongside other comorbid conditions. Hypertension, dyslipidemia, DM, and liver disease frequently contribute to an increasing multimorbidity burden that may accelerate CKD progression and complicate its management. To better characterize these comorbidities, we applied network-based analytic methods to describe comorbidity profiles across kidney function thresholds, identify “communities” of highly connected conditions, and quantify the joint occurrence of CKD with major cardiometabolic conditions. By integrating network metrics with joint cumulative incidence estimates for stage III–V and IV/V CKD, we aimed to clarify how cardiometabolic multimorbidity contributes to CKD risk. Methods We analyzed data from adult participants with≥2 eGFR measurements and ≥12 months of follow-up between 2011 and 2023. We summarized comorbidity distributions across two thresholds of kidney dysfunction
1.20–1.70) were associated with an increased risk of a ≥25% reduction in eGFR, whereas a CD4+ T cell count ≥500 cells/µL was associated with a decreased risk (OR = 0.22, 95% CI
0.27–0.87). Current or former intravenous drug use (IVDU) was associated with greater odds of belonging to the Low and Declining group (aOR = 2.29, 95% CI
Abstract of DissertationInvestigating the Cumulative Incidence of Chronic Kidney Disease Following HIV Diagnosis and the Effects of Comorbidities and COVID-19 on Kidney Function Over Time in a Cohort of People With HIV Background Chronic kidney disease (CKD) remains a leading non-AIDS comorbidity among people with HIV (PWH), driven by overlapping effects of aging, systemic inflammation, and cumulative metabolic burden. In contemporary US cohorts, the incidence of CKD among PWH ranges from 5.0 to 12.4 per 1,000 person-years, and the incidence of end-stage kidney disease is approximately 3.0 per 1,000 person-years, with higher risks observed among older and Black PWH. However, less is known about how kidney disease emerges after HIV diagnosis in the modern antiretroviral therapy (ART) era, when improved survival has shifted attention toward the long-term preservation of organ function. As PWH age, CKD increasingly co-occurs with cardiometabolic and behavioral comorbidities, forming tightly connected multimorbidity clusters that may accelerate disease progression and complicate care. The lasting effects of the coronavirus disease 2019 (COVID-19) pandemic have introduced an additional layer of kidney risk through both direct viral injury and indirect impacts on inflammation and care continuity. Together, these stages, from early post-diagnosis vulnerability to the accumulation of chronic conditions and the impact of emerging infections, define a continuum of kidney health challenges. This dissertation examines each of these stages to characterize the incidence, trajectories, and determinants of CKD within the DC Cohort in Washington, DC. Study Design We used data from the DC Cohort, a large multicenter, prospective observational study of PWH receiving care at 14 clinical sites across Washington, District of Columbia (DC). The analytic period spanned 2011 to 2024, encompassing more than a decade of clinical, laboratory, and demographic data extracted from electronic health records. Analyses were restricted to adults (≥18 years) with longitudinal estimated glomerular filtration rate (eGFR) measurements and relevant covariates. Study-specific inclusion criteria and analytic methods were defined for each analysis as described below. Analysis 1 Objectives CKD remains a key comorbidity among PWH, and early identification of kidney function decline is essential to prevent progression and related complications. Among newly diagnosed participants entering HIV care, the early years after HIV diagnosis represent a critical period to understand patterns of kidney function change and identify those at greatest risk of decline. To characterize the onset and early progression of CKD following HIV diagnosis, we evaluated the cumulative incidence of stage III–V and IV/V CKD during follow-up, identified demographic, clinical, and HIV-related factors associated with CKD onset, and classified distinct longitudinal trajectories of kidney function using group-based trajectory modeling (GBTM) to clarify early patterns of kidney function decline. Methods We analyzed data from newly diagnosed adult participants to estimate the cumulative incidence of stage III–V (eGFR <60 mL/min/1.73 m2) and IV/V (eGFR <30 mL/min/1.73 m2) CKD using Fine and Gray’s competing risks models. We employed GBTM to identify distinct patterns of eGFR over time and used multinomial logistic regression models to examine predictors of trajectory group membership. Results Among 844 participants (median age 36 years, 70% male, and 73% Black), 77 (9.12%) developed stage III–V CKD, among whom 19 (2.25% of the total, 24.68% of the cases) progressed to stage IV/V CKD, yielding a cumulative incidence rate of 12.16 and 2.88 per 1,000 person-years, respectively. In adjusted models, older age was associated with an increased risk of stage III–V (aHR = 2.19 per 10 years) and IV/V (aHR = 1.46 per 10 years) CKD. Diabetes mellitus (DM, encompassing both types 1 and 2) was significantly associated with stage III–V CKD (aHR = 1.64, p = 0.0430) and showed a non-significant trend toward stage IV/V CKD (aHR = 2.06, p = 0.1100). Lower serum albumin (ALB) levels were associated with stage IV/V CKD (aHR = 0.39 per g/dL). HIV diagnosis during the modern ART era (2015–2023) was associated with a reduced risk of both stage III–V and IV/V CKD compared with HIV diagnosis during the highly active ART (HAART) era (p < 0.01). While not statistically significant, Black race was associated with a 37% greater risk of stage III–V CKD and 94% greater risk of stage IV/V CKD. GBTM identified three eGFR patterns
- All rights reserved
Notice to Authors
If you are the author of this work and you have any questions about the information on this page, please use the Contact form to get in touch with us.