Validity of a protocol to estimate patients' pre-morbid basal blood pressure

Abstract

Purpose: The pre-illness basal mean arterial BP (MAP) is an important reference point to gauge the degree of relative hypotension among unwell patients. We aimed to assess mean bias, correlation, and agreement between basal MAP measured during nighttime ambulatory BP monitoring (ABPM) and basal MAP estimated using a standardized protocol.

Materials and methods: For a cohort of 137 consecutive patients, aged ≥40 years, who recently underwent ABPM, a blinded investigator estimated basal MAP from up to five most recent clinic BP measurements. Both basal MAP values, measured and estimated, were compared pairwise for each participant.

Results: We traced a median of 4 [interquartile range 3-5] previous BP measurements per patient over a median period of 132 [interquartile range 55-277] days up until the ABPM test. The estimated basal MAP (mean 88 ± 8 mmHg) was linearly related (Pearson's r = 0.41, p = 0.0001) to the measured basal MAP (mean 88 ± 12 mmHg). Bland-Altman plot revealed a mean bias of 0.3 mmHg with agreement limits of ±22 mmHg.

Conclusions: The mean bias between estimated and measured values for basal MAP was insignificant and modest. When a recent nighttime ABPM is unavailable, a protocol based on recent clinic BP readings can be used to estimate patient's basal MAP.

Study registration: Australian New Zealand Clinical Trials Registry ACTRN12613001382763.

Keywords: Blood pressure; baseline; critical illness; protocol; relative hypotension; targets.