Automated Blood Pressure Recording in Pregnancy

Abstract

The use of an automated blood pressure (BP) recorder (the Takeda UA-751) was evaluated in 50 pregnant and 50 age-matched non-pregnant women. Four alternate BP recordings were made using the Takeda UA-751 and mercury sphygmomanometry. The order of recordings was allocated randomly and the observer blinded to the automated recordings. Diastolic BP was recorded as both phase IV and phase V Korotkoff sounds in pregnant women and as the phase V sound in non-pregnant women. Systolic blood pressure (SBP) was overestimated by the Takeda UA-751 by a mean absolute value of 4mmHg (95% C.I: +2 to +6mmHg) in pregnant women and by 2mmHg (9556 CI: 0 to +4mmHg) in non-pregnant women. Diastolic blood pressure (DBP) phase IV was underestimated by the Takeda UA-751 by a mean of 3mmHg (95% C.I: -5 to -1mmHg) and DBP phase V overestimated by a mean of 2mmHg (95% C.I: 0 to +5mmHg) in pregnant women. DBP did not differ (95% C.I: -1 to +2mmHg) in non-pregnant women. Limits of agreement, i.e. the mean ± two standard deviations of the differences between blood pressures recorded by these two methods, were wide: the automated recorder differed from auscultatory DBP by up to 17mmHg in pregnant women and 10mmHg in non-pregnant women, and from auscultatory SBP by up to 16mmHg and 10mmHg respectively in these groups. Mean SBP and DBPs for both groups were similar with the two methods of BP recording. If mercury sphygmomanometry is considered the gold standard then the Takeda UA-751 is suitable for measuring blood pressures during pregnancy when group values are being assessed. However, clinicians must be aware that automated BP recording may differ considerably from mercury sphygmomanometry in an individual subject, particularly during pregnancy.