Actimeter-derived sleep and wake data and nocturnal ambulatory blood pressure estimation in subjects with stroke and transient ischaemic attack.

BACKGROUND/AIMS: Abnormalities in nocturnal blood pressure control identified using ambulatory blood pressure monitoring are associated with adverse cardiovascular outcomes. Sleep and wake episodes during such studies are usually identified by means of sleep diaries but these may be inaccurate in stroke patients. We performed a study to determine whether sleep-wake data obtained using wrist-mounted actimeters would significantly influence the results of routinely performed nocturnal ambulatory blood pressure monitoring when compared with diary-based sleep-wake recording and fixed time-period data.

METHODS

Actimetry was performed using a wrist-mounted device during routine ambulatory blood pressure monitoring in subjects who had suffered a transient ischaemic attack or stroke. The mean nocturnal blood pressure readings were calculated using sleep data derived from actimetry and diaries and compared for a fixed time period from 11:00 pm to 8:00 am.

RESULTS

Twenty subjects (mean age 68 years, and 13 female) were studied. Patients were found to have slept for a median of six-hours (one- to eight-hours) by diary and five-hours (zero- to eight-hours) by actimeter data. Diary and actimeter data agreed in 69% of recordings. The mean sleeping systolic blood pressure was lower when calculated by actimeter data than by diary data (119·6 mmHg vs. 123·2 mmHg, P=0·049, paired t-test) but there was no significant difference in diastolic blood pressure. The mean nocturnal blood pressure calculated from 11:00 pm to 7:00 am was higher than sleeping blood pressure calculated from diary data. (mean systolic blood pressure: 127·6 mmHg vs. 123·6 mmHg, P=0·065; mean diastolic blood pressure 69·0 vs. 64·0, P=0·028).

CONCLUSION

Calculation of nocturnal and sleeping blood pressure is lower in subjects with stroke and transient ischaemic attack when objective actimeter-derived sleep/wake data are used.