| Summary: | Background: Oscillometric mean arterial pressure (MAP) agrees closely with invasive MAP, [1] but most devices do not report MAP and it is usually estimated by a form factor (FF). However, blood pressure (BP) measurement errors will affect FF, its correlations with exposures, and introduce errors into MAP estimated from the BP waveform.
Methods: Brachial BP was measured using a Pulsecor device in 1,112 participants in the Southall and Brent Revisited study (68.8 ± 6.1 y; 78.2% male; 47.4% White-European; 38.3% South-Asian; 14.3% African-Caribbean). Form factors (FFosc and FFwave) were calculated as (MAP-diastolic BP)/(systolic BP-diastolic BP) by oscillometry (MAPosc) or from the BP waveform (MAPwave).
Results: FFosc and FFwave differed (0.28 (SD = 0.02) vs. 0.36 (SD = 0.04); p <0.001) and were negligibly correlated (r = 0.07). Neither FFosc nor FFwave were associated with ethnicity, prevalent cardiovascular disease or current smoking status, and neither showed significant correlations with age, total- or HDL-cholesterol, or physical activity. Both FFosc and FFwave were lower in men (difference (Δ) = −0.005(95% CI = −0.007, −0.002) vs −0.015(95% CI = −0.020, 0.009) respectively) and were negatively correlated with height (r = −0.14 both), but only FFwave correlated with body mass index (r = 0.02 vs r = 0.10) and heart rate (r = −0.06 vs r = 0.20). ΔMAPosc-MAPwave correlated with age (r = 0.10), height (r=0.15) and heart rate (0.17) and was greater in women (0.9(95% CI = 0.5, 1.3) mmHg).
Conclusions: FFwave agrees poorly with FFosc probably due to measurement errors. This creates spurious associations between exposures and FF and causes systematic errors in estimated MAPwave. These errors have the potential to confound associations in epidemiological studies.
|
|---|