Direct ophthalmoscopy versus detection of hypertensive retinopathy: a comparative study.

BACKGROUND Detection of hypertensive retinopathy (HR) with direct ophthalmoscopy is part of the assessment of hypertensive patients. Its use has been questioned because of its subjectivity and high interobserver variability. OBJECTIVE To determine the prevalence of HR in hypertensive patients under outpatient monitoring, the correlation between diagnosis and ophthalmoscopy and angiography, and to correlate it with other target organ damages. METHODS An observational, analytical and cross-sectional evaluation of 99 patients. Direct ophthalmoscopy and angiography performed by two investigators independently. Classification of RH, as described by Keith, Wagener and Barker. RESULTS The prevalence of HR of any grade was higher than 90.0% by both methods. On ophthalmoscopy, we observed grade I abnormalities in 51.0%, grade II in 43.0%, with one patient with grade III HR. On angiography, we observed grade I abnormalities in 42.0% and grade II in 52.0%. We detected three patients with grade III HR, two of which were not detected by ophthalmoscopy. Observers' agreement for the presence and severity of HR was poor with direct ophthalmoscopy and good with angiography. Renal dysfunction, ECG abnormalities (ventricular hypertrophy, pathological Q wave, repolarization abnormalities), and history of stroke were observed in 70.0%, 27.0% and 10.0% of patients, respectively. There was no relationship between the severity of HR and other target organ damages. CONCLUSION We observed a high prevalence of HR using both methods. Observers' agreement for the diagnosis and determination of the severity of HR was better with angiography. In our sample, there was no association of the severity of HR with other target organ damages.