Microvascular narrowing and BP monitoring: A single centre observational study

Abstract

Introduction: Half of all hypertensive individuals have inadequately-controlled BP because monitoring methods are ineffective. This single centre study examined consecutive subjects undergoing 24 hour BP measurements for clinic and ambulatory BP levels, and for end-organ damage (retinal microvascular abnormalities and left ventricular hypertrophy, LVH, > 1.1 cm). Retinal images were graded for microvascular retinopathy (Wong and Mitchell classification), and vessel calibre using a semiautomated method. Features were compared using chi-squared, Fisher's exact or the student's t test.

Methods: One hundred and thirty-one individuals (59 male, 45.0%, mean age 61.7 ± 14.5 years) were studied. Ninety-nine (76.2%) had a clinic BP ≥ 140/90 mm Hg, 84 (64.6%) had a mean awake systolic BP ≥ 135 mm Hg, 100 (76.9%) had a mean sleeping systolic BP ≥ 120 mm Hg, and 100 (76.2%) had abnormal nocturnal BP dipping patterns. Sixty-nine individuals had undergone echocardiography and 23 (33.3%) had LVH.

Results: All participants had a mild (88.5%) or moderate (11.5%) microvascular retinopathy. Moderate microvascular retinopathy was found in 86.7% of those with a mean awake systolic BP ≥135 mm Hg (p = 0.058) but was not associated with other abnormal BP measurements, abnormal dipping patterns or LVH. However retinal arteriole calibre was reduced in subjects with a mean 24 hour awake systolic BP ≥ 135 mm Hg (p = 0.05). Retinal arteriole calibre was smaller in subjects with LVH (128.1 ± 13.5 μm compared with 137.6 ± 14.1 μm in normals, p = 0.014). Venular calibre was also less in subjects with LVH (185.4 ± 24.6 μm compared with 203.0 ± 27.2 μm in normals, p = 0.016). Arteriole narrowing predicted an increased risk of LVH (AUC 0.69, 95%CI 0.55 to 0.83) that was comparable with 24 hour systolic BP ≥130 mm Hg (AUC 0.68, 95%CI 0.53 to 0.82) and mean awake systolic BP ≥135 mm Hg (AUC 0.68, 95%CI 0.54 to 0.83).

Conclusions: This study suggests that retinal arteriole narrowing may be equally accurate in predicting LVH as any clinic or ambulatory BP measurement. The convenience and accuracy of microvascular calibre measurement mean that it should be investigated further for a role in routine hypertension assessment and monitoring.