Abstract 2759: Approved 24-hr Ambulatory Blood Pressure Monitors Inaccurate and Unreliable? Approval Protocols Should be Modified and Devices Should Adjust Sensors for Posture and Activity
Given concerns about 24-hr ambulatory blood pressure monitor (ABPM) accuracy & reliability & environmentally-isolated approval criteria established by the Association for the Advancement of Medical Instruments (AAMI) & European Society of Hypertension (ESH), we embarked upon a series of rigorous lab & field studies to test ABPM validity & reproducibility. We used a novel dual-monitor protocol (DMP) enabling simultaneous, same arm testing by two clinicians (C1,C2) with mercury column and one or two ABPMs (A1,A2) to test 47 subjects: normotensives (n=29, age=27 yr, 16 – 60 yr), nonmedicated & medicated hypertensives (n=12, age=51 yr, 17– 68 yr) & primary alcohol-dependent patients (n=6, age=39 yr, 30–59 yr). Transitional measurements just after a change in posture were most problematic. During lab testing A1 & A2 significantly (P<0.001) underestimated C1 & C2 diastolic blood pressure (DBP) by over 7 mm Hg in standing normotensives. A1 & A2 measuring same arm field pressures simultaneously in a labile hypertensive male had a low level of DBP agreement (R2=0.52). In hypertensives A1 & A2 significantly (P<0.01) underestimated lab relaxed arm DBP by 6 mm Hg. C1 & C2 common variance was 97% for systolic BP (SBP) & 95% for DBP, while A1 & A2 common variance was 88% for SBP & 77% for DBP. BP postural differences that may help distinguish hypertensives from normotensives were detected by C1 & C2 but not A1 & A2. ABPMs underestimated 24-hr DBP up to 10 mm Hg (avg=5.4) & blurred distinctions among normal, high normal, Stage I & white-coat hypertensives. In patients post-detoxification, A1 significantly underestimated C1 & C2 lab DBP by 6 mm Hg. We conclude in normotensives, hypertensives & alcohol-dependent patients that:
despite having achieved AAMI & ESH approval, popular ABPMs may be inaccurate & unreliable in assessing DBP,
approval protocols should require simultaneous same arm measurements as provided by our DMP,
unless posture & activity are accounted for, auscultatory & oscillometric ABPMs are susceptible to clinically significant & unpredictable errors &
because Korotkoff sound intensity varies with posture & exercise, new ABPMs should assess these variables proactively to adjust sensor sensitivity before each reading to improve validity & reliability.