In-Clinic Blood Pressure Prediction of Normal Ambulatory Blood Pressure Monitoring in Pediatric Hypertension Referrals.

OBJECTIVE: Since younger patients have low pretest probability of hypertension and are susceptible to reactive and masked hypertension, ambulatory blood pressure monitoring (ABPM) can be useful. To better target use in referred patients, we sought to define in-clinic systolic blood pressure (SBP) measures that predicted normal ABPM and target end organ damage. DESIGN, SETTING, PATIENTS, OUTCOME MEASURES: Data were collected on consecutive patients referred for high BP undergoing an ambulatory BP monitor from 2010 to 2013 (n = 248, 33.9% female, mean age 15.5 ± 3.6 years). Candidate in-clinic predictors were systolic maximum, minimum, or average BPs obtained by auscultative, oscillometric, or both. Multivariable logistic regression models were used to determine the prediction of normal ABPM by in-clinic BP predictors. Separate models considered predicting left ventricular hypertrophy (LVH) by in-clinic SBP vs. ABPM-defined hypertension. Identified predictor utility was tested with receiver operator characteristic curves. RESULTS: Maximum (OR 0.97 [95% CI 0.94-0.99]; P = .047), minimum (0.96 [0.94-0.99]; P = .002), and average (0.97 [0.95-1.00]; P = .04) in-clinic auscultative SBP predicted normal ABPM. Each had a c-statistic of 0.58. LVH was associated with in-clinic auscultative minimum SBP treated continuously (1.05, [1.01-1.10], P = .01) or dichotomized at the 90th percentile (8.23, [1.48-45.80], P = .02), as well as ABPM-defined hypertension (3.31, [1.23-8.91], P = .02). Both predictors had poor sensitivity and specificity. CONCLUSION: In youth, normal auscultative in-clinic systolic blood pressure indices weakly predicted normal ambulatory blood pressure and target end organ damage.