Background: Studies have shown that hypertensive retinal changes (HRC) have a moderate accuracy in predicting microalbuminuria (MA) in elderly hypertensive patients (age >65 years). This study is an effort to identify a similar relationship in hypertensive patients aged <65 years.
Methods: Eight hundred and seventy consecutive hypertensive patients (males, 460; females, 410) aged 18-65 years were assessed for their demographic characteristics and other laboratory variables. Patients with diabetes mellitus, metabolic syndrome and overt nephropathy were excluded. Optic fundi were assessed for HRC after pupillary dilatation, which were photographed. MA (albumin-creatinine ratio) was measured as an average of two non-consecutive overnight spot urine samples.
Results: Mean age was 45 +/- 13.4 years. Prevalence of MA and HRC was 36.7 and 38%, respectively. MA showed a strong association with HRC (P < 0.0001). Logistic regression identified the association between MA, duration of hypertension (HTN) (P = 0.016), smoking (P = 0.012) and elevated high-sensitivity C-reactive protein (HsCRP) (P = 0.032). HRC were associated with duration of HTN (P = 0.021) and smoking (P < 0.0001). Tests of accuracy for HRC as a predictor of MA showed sensitivity, specificity, positive predictive value, negative predictive value, likelihood ratio of a positive test and likelihood ratio of a negative test of 78%, 86%, 76%, 87%, 5.2 and 0.26, respectively. Area under the receiver operating characteristic curve was 81%. Similarly, the individual grades of HRC had a moderate predictive accuracy. Higher grades had higher predictive accuracy. Inter- and intra-observer correlation in interpreting HRC was acceptable.
Conclusions: HRC of any grade have moderate accuracy in predicting MA and hence can be used as a cost-effective screening tool to predict MA especially in a resource-poor setting.