Bias Associated with Peripheral Non-Invasive Compared to Invasive Arterial Blood Pressure Monitoring in Healthy Anaesthetised and Standing Horses Using the Bionet BM7Vet.

Abstract

To compare arterial blood pressure (ABP) measured invasively (IBP) to ABP measured non-invasively (NIBP) via oscillometry in healthy anaesthetised and standing horses using the Bionet BM7Vet. Fourteen horses were anaesthetised for elective procedures (anaesthetised group) and 10 horses were enrolled for standing blood pressure manipulation (standing group). In both groups, IBP and NIBP-corrected to heart level were measured every 3 min using the Bionet BM7Vet. The overall mean difference (bias), standard deviation and limits of agreement (LOA) were calculated for paired IBP and NIBP systolic (SAP), mean (MAP) and diastolic (DAP) blood pressure measurements. In anaesthetised horses, the NIBP cuff was placed at either the proximal tail base or the metacarpus. Invasive MAP was used to retrospectively characterise measurements into hypotensive (≤70 mm Hg), normotensive (71–110 mm Hg) or hypertensive (≥111 mm Hg) subgroups. In standing horses, the NIBP cuff was placed at the tail base only and invasive MAP was manipulated to achieve hypertension (≥126 mm Hg) and hypotension (≤90 mm Hg) using phenylephrine and acepromazine, respectively. When measuring NIBP at the tail in anaesthetised horses, the Bionet BM7Vet failed on 8/185 occasions and overestimated SAP, MAP and DAP during hypotension and normotension. The biases (lower, upper LOA) for MAP were −11.4 (−33.3, 10.5) and −6.0 (−25.8, 13.8) mm Hg, respectively. Hypertension could not be evaluated. When measuring NIBP at the metacarpus in anaesthetised horses, the Bionet BM7Vet failed on 24/65 occasions and underestimated SAP, MAP and DAP when all ABP subgroups were combined. The bias (lower, upper LOA) for pooled MAP was 3.6 (−44.3, 51.6) mm Hg. When measuring NIBP at the tail in standing horses, the Bionet BM7Vet failed on 64/268 occasions and underestimated SAP, MAP and DAP during hypotension, normotension and hypertension. The biases (lower, upper LOA) for MAP were 16.3 (−10.5, 43.1), 16.6 (−19.5, 52.7) and 30.0 (−8.1, 68.0) mm Hg, respectively. Monitoring NIBP on the Bionet BM7Vet in anaesthetised horses overestimated ABP at the tail and underestimated ABP at the metacarpus. The device inaccurately detected hypotension and should be used cautiously. In standing horses, the Bionet BM7Vet underestimated ABP at the tail, especially during pharmacologically induced hypertension.