ABSTRACT
Background Office blood pressure (BP) measurement is affected by the white-coat phenomenon and shows a weaker correlation with the gold standard ambulatory blood pressure monitoring (ABPM). To overcome this limitation, 24-hour ABPM is recommended by the guidelines for the diagnosis of hypertension. However, 24-hour ABPM is expensive and cumbersome, which limits its use in low to middle-income countries like Pakistan. We aimed to assess if an abbreviated ABPM interval can be utilized to diagnose hypertension effectively in our population. Methods A cross-sectional study, involving 150 participants as part of the Post Clinic Ambulatory Blood Pressure (PC-ABP) study, was conducted in the cardiology clinics. Participants ≥18 years of age, who were either hypertensive or referred for assessment of hypertension, were included. Blood pressure (BP) readings were taken with an ambulatory BP monitor over a 24-hour period. After excluding the first hour called the 'white-coat window,' the mean of the first six systolic readings taken every half hour during the daytime was calculated and was called systolic three-hour ABPM. Pearson correlation coefficients were calculated and Bland-Altman plots were constructed to determine the correlation and limits of agreement between mean systolic three-hour ABPM and daytime-ABPM. Receiver operating characteristic (ROC) curve for systolic and diastolic three-hour daytime ABPM and area under the curve (AUC) were analyzed for the level of accuracy in predicting hypertension. Results Of the 150 participants, 49% were male, and 76% of all were hypertensive. The mean age of participants was 60.3 ± 11.9 years. The mean systolic three-hour ABPM was 135.0 ± 16 mmHg. The mean systolic daytime ABPM was 134.7 ± 15 mmHg. Pearson correlation coefficient between mean systolic three-hour ABPM and mean systolic daytime ABPM was 0.85 (p-value <0.001). The limits of agreement were 18 mmHg to -17 mmHg between the two readings on Bland-Altman plots and the area under the curve of the receiver operating characteristic (ROC) was 0.96, suggesting that three-hour systolic ABPM is a good predictor of hypertension. Conclusion Three-hour ABPM correlates well with 24-hour ABPM in the Pakistani population. We recommend considering the use of this abbreviated ABPM to screen hypertension where a full-length ABPM cannot be used. Further studies can be conducted on a larger sample size to determine the prognostic implications of this shortened ABPM.
