Objective: To test the validity of ambulatory heart rate (HR) assessment with a cuff ambulatory blood pressure (ABP) monitor.
Design: Cross-instrument comparison of HR measured intermittently by a cuff ABP monitor (SpaceLabs, Redmond, Washington, USA), with HR derived from continuous electrocardiogram (ECG) recordings (1) in a controlled laboratory experiment and (2) during long-term recording in a true naturalistic setting.
Participants: Six normotensive subjects participated in the laboratory study. A total of 109 male white-collar workers underwent ambulatory monitoring, of which 30 were mildly hypertensive.
Methods: Four different laboratory conditions (postures: lying, sitting, standing, walking), repeated twice, were used to assess the short-term effects of cuff inflation on the HR. To test the actual ambulatory validity, participants simultaneously wore a continuous HR recorder and the ABP monitor from early morning to late evening on 2 workdays and one non-workday. Diary and vertical accelerometery information was used to obtain periods of fixed posture and (physical) activity across which HR from both devices was compared.
Results: Laboratory results showed that the ABP device reliably detected HR during blood pressure measurement, but that this HR was systematically lower than the HR directly before and after the blood pressure measurement. The ambulatory study confirmed this systematic underestimation of the ongoing HR, but additionally showed that its amount increased when subjects went from sitting to standing to light physical activity (2.9; 4.3 and 9.1 bpm (beats/min), respectively). In spite of this activity-dependent underestimation of HR, the correlation of continuous ECG and intermittent ABP-derived HR was high (median r= 0.81). Also, underestimation was not different for normotensives and mild hypertensives.
Conclusions: A direct effect of cuff inflation leads to the underestimation of ongoing HR during cuff-based ABP measurement. Additional underestimation of HR occurs during periods with physical activity, probably due to behavioural freezing during blood pressure measurements. HR underestimation was not affected by hypertensive state. When its limitations are taken into account, ABP-derived ambulatory HR can be considered a reliable and valid measure.