Effect of overcuffing on the accuracy of oscillometric blood pressure measurements

Abstract

Overcuffing (using too large a blood pressure [BP] cuff), is known to artificially lower auscultatory BP measurements; however, its effect on oscillometric readings is unclear. The possibility that overcuffing biases oscillometric readings is currently widely disregarded. We sought to confirm that overcuffing lowers auscultatory readings and to assess its effect on oscillometric measurements. Community–dwelling adults (aged ≥18 years) with arm circumferences within the standard range of 25–32 cm were recruited. Using primarily the International Standards Organization 2009 protocol, we compared the standard Baum adult (25–35 cm) to the large adult (33–47 cm) cuff. The standard cuff was considered the “reference standard.” In Phase I, auscultatory measurements were performed by two trained observers. In Phase II, oscillometric measurements were performed. Each phase was analyzed independently using paired t–tests and by generating Bland–Altman plots. Of 108 participants, 87 contributed data for Phase I, 85 for Phase II, and 69 were common to both phases. Phase I mean age was 38.0 ± 18.5 years, mean arm circumference was 28.0 ± 1.9 cm, and 21% had a past hypertension. The Phase I results confirmed that overcuffing reduced auscultatory BP measurements by 3.6 ± 5.1/2.8 ± 4.0 mm Hg (P–values <.0001 for both). For Phase II, mean age was 39.3 ± 18.3 years, mean arm circumference was 28.0 ± 1.9 cm, and 22% had past hypertension. Mean BPs were 112.2 ± 13.1/67.8 ± 7.3 mm Hg for the large cuff and 117.8 ± 13.3/71.2 ± 7.1 mm Hg for the standard cuff (5.5 ± 5.9/3.4 ± 5.2 mm Hg lower with the large cuff; P–values <.0001).Overcuffing leads to a clinically important downward bias in oscillometric measurements. An upper size limits for oscillometric cuffs should be specified.