|Home | About | Journals | Submit | Contact Us | Français|
Nephropathy in HIV-infected patients has been associated with progression to AIDS and death. The virus, several co-morbid conditions and certain medications may contribute to the development and progression of kidney disease.
This study analyzed data collected from HIV-infected persons enrolled in a HIV registry in Puerto Rico during January 1998 through September 2006. Demographic factors, clinical manifestations, laboratory findings at enrollment, and antiretroviral therapy (ART) prescriptions were compared between patients with and without kidney disease. Death status and cause of death by December 2006 were also evaluated and compared.
The study included 1,283 subjects, 69.0% male, 39.7% injecting drug users, 19.5% hepatitis C infected, 6.5% with diabetes mellitus (DM-II), 11.6% had hypertension (HTN) and 9.0% had kidney disease. Patients with kidney disease had significantly higher (p<0.05) HIV viral load mean (273,499 vs. 202,858 copies/mL), CD4 T-cell count < 200 (57.0% vs. 44.4%), underweight (22.9% vs. 10.9%), DM-II (13.9% vs. 5.8%), HTN (27.8% vs. 10.0%) and mortality (15.9 vs. 5.7 deaths per 100 years of follow-up) than those without it. Cox proportional hazard analysis showed that patients with kidney disease had higher mortality risk (2.1) after controlling for age, gender, HIV risk factor, ART prescript in the last year and HIV disease duration.
This study demonstrated a substantial disparity in mortality for Puerto Ricans HIV-infected patients with nephropathy. Kidney disease preventive strategies that include aggressive control of HIV-infection and chronic medical conditions such as hypertension and diabetes are recommend as an approach to reduce this health disparity.
The patterns of mortality among HIV-infected persons have been changing with the introduction of highly active antiretroviral therapy (HAART). More specifically, the presence of kidney disease has been increasing as an important cause of death amongst HIV-infected patients. National trends derived from U.S. death certificates among persons with HIV demonstrate that from 1987 through 1999, the proportion of deaths associated with kidney disease accounted for 10.4%. 1 Similarly, mortality trends of a HIV-infected cohort followed in Puerto Rico since 1992 revealed an increase of renal disease as a clinical condition reported as a cause of death. With the introduction of HAART in the island, kidney disease was associated as a cause of death in 7.3% of patients as compared to 6.3% before HAART. 2 While an important decrease in AIDS-related opportunistic conditions have been documented an increase in the presence of chronic conditions such as diabetes mellitus type II (DM-II), cardiovascular diseases (CVDs) and renal diseases have become more relevant in defining the prognosis among HIV-infected subjects.3,4 The presence of kidney disease has been associated with the severity of HIV-infection, rate of progression to AIDS, as cause of hospital admissions and as an adverse finding associated to increased mortality. 4,5 Survival is adversely affected by the presence of end-stage renal disease particularly in the absence of antiretroviral therapy.6,7
The pathogenesis of kidney disease in HIV-infected patient is frequently multifactorial with a number of specific entities well defined. These include HIV-associated nephropathy (HIVAN), and diabetic nephropathy amongst others.8 While HIVAN has been described as occurring overwhelmingly in persons of African or Haitian descent, there is limited data describing the epidemiology of kidney disease in a Hispanic cohort and no published data from Puerto Rico. Puerto Rico has a high prevalence of HIV and has the largest concentration of Hispanics within the United States and its territories.9,10 National data on abnormal kidney function and chronic kidney disease are not available for Hispanic HIV-infected patients. The availability of this data is relevant for the planning of evidence-based prevention activities that may prolong the lives of the affected population. The present study examines the epidemiology of HIV-infected patients with and without kidney disease and the risk for death associated to kidney disease.
The study evaluated the prevalence of kidney disease among a cohort of adult HIV-infected persons that entered the Retrovirus Research Center (RRC) at Bayamon, Puerto Rico between January 1998 through September 2006 and that had at least one serum creatinine measure. Demographic factors, clinical manifestations, laboratory findings, and antiretroviral therapy (ART) prescriptions at enrollment or in the previous 12 months were compared between patients with and without kidney disease. Study participants were followed at the Ramón Ruiz Arnau University Hospital or in the HIV ambulatory clinic of our institution up to the death or the study end by December 2006. The study was conducted with the approval of the Institutional Review Board (IRB) of the Universidad Central del Caribe.
Once an informed consent was obtained, a baseline questionnaire was completed and laboratory testing is performed including CD4+ T-lymphocyte count, HIV viral load, HCV viral load (if co-infected), white and red blood cell count and a comprehensive panel of tests that included creatinine. Enrollment included a retrospective abstraction of clinical data and laboratory findings from the last 12 months. HIV risk behaviors and socio demographic data were gathered via personal face to face 30 minutes interview. The information was complemented with medical record abstraction in which diseases, laboratory data, therapy related information, and medical complications were abstracted and recorded. The AIDS-defining illnesses included Pneumocystis jirovecii pneumonia (PJP), cerebral toxoplasmosis (CT), recurrent bacterial pneumonia (RP), pulmonary tuberculosis (PTb) and Kaposi’s sarcoma (KS) were evaluated. HIV disease duration was estimated based on the initial positive HIV test reported. Kidney disease was defined as a glomerular filtration rate (GFR) below 60 ml/min. The GFR was estimated by the Modification of Diet in Renal Disease (MDRD) equation that included standardized serum creatinine level, age and gender.11 GFR was expressed in millimeters per minute per 1.73 m2. The participants were considered to have a history of DM-II or hypertension (HTN) if this was documented in the medical record. Body mass index (BMI) was calculated using weight divided by height square (Kg/m2). Participants were classified as underweight if their BMI was below 18.5 kg/m2. Prescription of antiretroviral therapies (ART), including nucleoside reverse transcriptase inhibitors (NRTI), non nucleoside reverse transcriptase inhibitors (NNRTI) and protease inhibitors (PI) were also collected. The use of HAART was defined as the use of three or more antiretroviral drugs in the same regimen at or before enrollment.
Mortality analysis used death status available through December 2006. Mortality data was obtained from a review of the institutional medical records and the Puerto Rican AIDS surveillance system. The Puerto Rico Demographic mortality registry was also used to establish date and certificate based causes of death. The reported causes of death were tabulated and organized into systems or organ failure.
The Statistical Package of Social Sciences (SPSS) program was used to perform univariate, and bivariate analyses. Differences between patients with and without kidney disease including enrollment findings, mortality rates and causes of death were analyzed with the Chi-square or Fisher exact test, and the student t test. Odds ratios and 95% confidence intervals (CI) were presented. In addition, Cox proportional hazard analysis with relevant covariates, including variables with P values less or equal 0.05 in the bivariated analyses, were use to evaluate mortality risk in the cohort. The data were presented as mortality hazard ratios (HR) with their 95% CI. The P value used to determine statistical significance was ≤ 0.05.
Of the total of 1,518 HIV-infected patients who visited the RRC during the study period, 1,440 entered the HIV cohort. Of them, 1,283 persons had at least one measure of serum creatinine level at enrollment. Among those with creatinine data, 69.0% were male, the mean educational level was below 9th grade, and the HIV transmission risk factor was 39.7% injecting drug use (IDU) and 27.3% men who have sex with men (MSM). The mean age was 40.8 ± 9.6 year. Approximately 2.3% had a history of PJP, 2.1% CT, 0.5% RP, 0.5% KS, 19.5% HCV, 11.6% HTN and 6.5% DM-II at study entry (data not shown). A CD4+T cell count < 200 cells/μl was observed in 45.3 %, 9.0% had moderate to severe kidney disease with a GFR estimated by the MDRD study equation below 60 ml × min, 11.9% were underweight and 57.3% had received HAART at enrollment or in the previous 12 months.
Patients with kidney disease were more likely to have a CD4+ T cell count <200 cells/μL, higher HIV mean viral load, and higher prevalence of underweight, CT, DM-II and HTN at enrollment than patients without kidney disease (table 1). Patients without kidney disease were more likely to be prescribed with HAART than those with kidney disease (Table 1).
Among the 1,283 HIV-infected patients, 323 (25.2%) had died at the end of the observation period, with a significantly higher mortality in patients with kidney disease (15.9 vs. 5.7 deaths per 100 years of follow up). Bivariate analysis demonstrated that patients with kidney disease were more likely to die during follow-up than those without kidney disease (OR = 5.2, 95% CI 2.18–12.48). Death was significantly more likely to be attributed to kidney disease and less likely to be attributed to liver disease among the patients with kidney disease (table 2). Cox proportional hazard analysis confirmed that patients with kidney disease had a higher mortality than those without it (HR= 2.12, 95% CI1.50–3.00), after controlling for age, sex, estimated HIV disease duration, underweight, history of IDU, CD4+ T cell count, HTN, DM-II, HCV seropositive status, prevalent AIDS-defining conditions and history of HAART prescription (table 3). Additionally, mortality was associated with IDU, DM-II, underweight, PJP, lower CD4+Tcell count, and the absence of HAART.
Most of the reported data about renal disease in HIV-infected patients is derived from studies in which there is a substantial majority of African Americans and/or few Hispanics. African American experiences the highest rate of HIV-infection and chronic kidney disease in the US.12,13 Our data provides an estimate of the prevalence of kidney disease in Hispanic HIV-infected patients and more specifically in Puerto Ricans. The 9% prevalence of kidney disease in this cohort is slightly higher than that reported by previous studies. Gardner et al in their study of 885 US HIV-infected women, found a prevalence of 7.2% at baseline and Choi et al in their 15,135 HIV-infected cases study found that 7.3% of the white and 6.9% of the African American patients had a renal dysfunction.14,15 Choi ‘s publication and our study use GFR levels to define kidney dysfunction. Other studies use the presence of proteinuria or an elevated serum creatinine as index of renal dysfunction.14 GFR is accepted as the best overall measure of kidney function since it controls the variation in the serum creatinine levels causes by sex, age and BMI. 11 These variables are often quite critical in the evaluation of target organ. We believe that along with Choi’s study we have utilized a more accurate measure of kidney function.
Chronic kidney disease is a serious condition associated with premature mortality, decreased quality of life and increased health-care expenditures.13 Although our study did not examine end stage renal disease, it demonstrated a clinically and statistically significant increased risk for death among patients with kidney disease, even after controlling for other factors including prescription of HAART. Our findings highlight the importance of including this condition among the priority considerations when designing any intervention to prolong the lives of HIV-infected Puerto Rican patients. Our study suggests a higher degree of immune dysfunction in patients with kidney disease. Higher level of HIV RNA viral load, lower CD4+ cell count and higher prevalence of AIDS related conditions were significantly seen in patients with kidney disease. In addition low BMI, the presence of chronic conditions including DM-II and HTN, were more common in these patients. Consistent with previous work these findings supported an association between higher AIDS conditions and higher chronic medical conditions in HIV infection patients with kidney dysfunction. 4, 14
With regards to disease outcome, our results demonstrate a substantial difference in mortality among HIV-infected patients with kidney disease. They have twice the risk to death than patients without renal disease (MR= 2.12), after controlling for other important cofactors.
Our study demonstrates the applicability to the Hispanic population of the recommendation of the Infectious Diseases Society of America that all patients at the time of HIV diagnosis should be assessed for existing kidney disease with a screening urine analysis and a calculated estimate of renal function.16 High risk populations for developing kidney disease, including African American persons, those with CD4+ T cell count < 200 cell/μL or HIV viral load > 4000 copies/mL, or those with HTN, DM-II or HCV should undergo annual screening for renal dysfunction. Patients with proteinuria or reduced renal function should be promptly referred to nephrologists and undergo additional evaluations. These recommendations highlight the importance of aggressively managing Puerto Rican HIV-infected population, as one with high risk for developing kidney disease.
Our study had several limitations: 1) The study group was selected from a passive surveillance cohort, where the patients had come to the center to have the data collected. This type of surveillance could increase the probability of lost of followed of the participants, which would generate missing data that could affect the study findings. 2) Many patients of the cohort had only one creatinine measure at the study time and it is highly recommended the use of at least two consecutive creatinine measures to evaluate the renal function.4 Consecutive measures would minimize the transient creatinine variations related to nonintrinsic renal processes, increasing so the accuracy of the kidney disease diagnosis.
The present study was performed in a Hispanic HIV infected population and demonstrates an increase prevalence of kidney disease. Improvements of prevention strategies are being included as of the study recommendation to decrease health disparity in this population.
The present study was sponsored by RCMI grant G12RR03035 and 1U54RR01950701. We would like to thanks the Puerto Rico Demographic Registry and the Puerto Rico Department of Health for their help and collaboration.
Angel M. Mayor, Retrovirus Research Center, Internal Medicine Department, Universidad Central del Caribe, School of Medicine.
Mark Dworkin, Epidemiology and Biostatistics Division of Epidemiology and Biostatistics, School of Public Health, University of Illinois at Chicago.
Luis Quesada, Retrovirus Research Center, Internal Medicine Department, Universidad Central del Caribe, School of Medicine.
Eddy Rios-Olivares, Department of Microbiology, Universidad Central del Caribe, School of Medicine.
Robert F. Hunter-Mellado, Retrovirus Research Center, Internal Medicine Department, Universidad Central del Caribe, School of Medicine.