A prospective cohort study in a Rwandan referral hospital of the incidence of rapid GFR decline and its risk factors in patients with hypertension.

Abstract

Introduction: Hypertension, behind diabetes mellitus, is second most common cause of ESRD. It is estimated that one-third of Rwandans between the age 55 and 64 years have hypertension. In a country where access to renal replacement therapy is very limited, understanding the role of traditional risk factors for rapid GFR decline, and other modifiable factors related to access to medications and adherence to treatment may help prevent loss of kidney function among hypertensive patients. Methods: We enrolled 137 outpatients – having hypertension as sole diagnosis – followed-up at Centre Universitaire Hospitalier de Kigali. Measurements at Month 0: demographics, BMI, socioeconomic status, level of education, baseline BP, baseline creatinine (baseline GFR calculated using CKD-EPI equation), proteinuria, HbA1c, BP control over 12 months, questionnaire administration for adherence to treatment and access to medications. Month 12: GFR. The primary outcome was the incidence of rapid GFR decline – defined as a decline of >= 4ml/min/1.73m2over one year. Secondary outcomes were traditional factors associated with rapid GFR decline; and exploratory outcomes were measures of adherence to treatment and access to medications and their association with rapid GFR decline. We used binary logistic regression (univariate and multivariate analysis) to identify independent factors. Results: 24 participants were lost to follow-up. The mean age was 56 years and the median age was 58 years; women made the majority (65%) of the study population; 56% were either overweight or obese; 83.7% had at least finished primary school; and two-third were in the middle socioeconomic status category. The incidence of rapid GFR decline was 28.1%; 18.5% of participants with rapid GFR decline progressed to incident CKD. Fourteen participants (10.8%) were newly diagnosed with diabetes mellitus and 18 participants (13.3%) were found to have proteinuria. We found that 43% of the study population had no full access to medications; 63% had good adherence; and 52% had uncontrolled blood pressure. The association between rapid GFR decline and age, socioeconomic status, baseline GFR or access to medications was statistically significant in univariate analysis, however, in multivariate analysis, there was no statistically significant association. Limitations: Short duration of follow-up and low statistical power. Measurements of creatinine are influenced by a number of factors. Conclusion and recommendations: We found a high incidence of rapid GFR decline; more studies with larger sample size and longer follow-up periods are needed to understand the problem. We emphasize the need for regular routine screening of diabetes and proteinuria among hypertensive patients. In-depth analysis of root causes of shortage of antihypertensive medications is recommended.