Comparison of Invasive and Oscillometric Blood Pressure Measurement in Obese and Nonobese Children

Abstract

Abstract BACKGROUND Obesity and hypertension are public health priorities, with obesity considered to be a potential cause of hypertension. Accurate blood pressure (BP) determination is required and often obtained by automated oscillometric cuff devices. We sought to determine the correlation of oscillometric measurement in children, and if obesity was associated with worse correlation between methods than nonobese children. METHODS Retrospective matched case-controlled study of 100 obese (97–99th percentile) and 100 nonobese (25–70th percentile) children after cardiac surgery with simultaneous systolic, diastolic, and mean invasive and oscillometric measurements. Matching was 1:1 for age, sex, race, and Risk Adjustment for Congenital Heart Surgery-1 score. Intraclass correlation coefficients and Bland–Altman plots were used to determine agreement with 0.75 as threshold. RESULTS Median age was 13 years (10–15). Agreement was low for systolic (0.65 and 0.61), diastolic (0.68 and 0.61), and mean measurements (0.73 and 0.69) (obese/nonobese). Bland–Altman plots demonstrated oscillometric BP measurements underestimated systolic hypertension (oscillometric readings lower than intra-arterial). Oscillometric measurements underestimated hypotension (systolic oscillometric measurements were higher than intra-arterial). This occurred in obese and nonobese patients. Correlation of oscillometric measurements was similar for nonobese and obese patients. CONCLUSIONS In this first ever study of simultaneous BP measurement by oscillometric vs. intra-arterial in obese and nonobese children, correlation is below accepted norms. The correlation of oscillometric cuff measurements is not affected by habitus in children. There is less correlation between oscillometric measurements and intra-arterial measurements during hypertension or hypotension. Healthcare providers should be aware of the limitations of oscillometric measurements.