Blood pressure measurement protocol determines hypertension phenotypes in a Middle Eastern population

Clinic blood pressure (BP) measurement remains a crucial step in managing hypertension. While the number of measures recorded in different settings varies, with typically 1-3 measures, there has been no prior justification for the actual number of measures required. We investigated the pattern of BP variability over 5 consecutive automated readings (R1-R5) and the influence of patient characteristics on this pattern to identify the phenotype of hypertension in a Middle Eastern population. There were 1389 outpatients (51% men, 49% women), age range (18-87 y) who had 5 unattended automated consecutive BP measurements with one-minute intervals using the validated Datascope Mindray Passport V Monitor with the patient blinded from the results. Mean (±SEM) SBP for R1 (136.0 ± 2 mm Hg) was similar to R2 (136.2 ± 2 mm Hg). Thereafter SBP progressively declined till R5 by total of 5.5 mm Hg. The SBP decline was less (4.2 mm Hg) in older (>50 years) vs younger participants (8.1 mm Hg; P <.001) and was blunted in diabetic and hypertensive participants. Overall, 43% of participants had R2 > R1, and 24% additionally had R5 > R1. Age was a strong independent predictor of having both R2 > R1 and R5 > R1, as well as diabetes. Diastolic blood pressure (DBP) decreased by average 2.8 mm Hg from R1 to R5. Females had a 5-fold greater total decline in DBP vs males (P <.001). Using the mean of 5 BP measures resulted in fewer participants being classified as hypertensive (36% of the population) compared to using one measurement (46%), or established BP guidelines which use different combinations of R1-R3 (37%-42%). Our findings in a Middle Eastern population highlight the importance of the BP measurement protocol in combination with patient characteristics in determining whether a patient is diagnosed with hypertension. Protocols that rely on different combinations of only 3 measures (R1-3) will classify more participants as hypertensive, compared to using 5 measures or disregarding a high R2.