According to the published literature, blood pressure (BP) measurements performed in the outpatient clinical setting are often inaccurate. The white coat effect and improper technique are the main causes of this imprecision. Construction of a set of readings without them could improve the accuracy of BP measurement.
To evaluate the accuracy and agreement of successive office BP measurements using the awake blood pressure average (ABPa) as the gold standard.
BP was measured in 852 patients using three techniques: in office (OBPa); seven successive measurements performed by a nurse using an automatic device; and 24 h of ambulatory BP monitoring. BP averages (BPa) were obtained from the nurse’s measurements: 1–2BPa (first and second), 3–7BPa (third to seventh), and 1–7BPa (first to seventh). OBPa and successive measurements were tested against ABPa by calculating the following: average difference in BP of 1–2BPa, 3–7BPa and OBPa, and the area under the curve.
Among the 834 patients eligible, 374 (43.9%) were considered to be hypertensive on the basis of the ABPa (≥135/85 mmHg). 3–7BPa showed the lowest average difference (4/3 mmHg). By contrast, OBPa showed the highest result (21/11 mmHg). The mean difference with 1–2BPa was 8/4 mmHg. The areas under the curve were better with 3–7BPa (0.82–0.85) and 1–2BPa (0.82–0.83) than OBPa (0.67–0.71) for both systolic and diastolic BP.
All means from successive measurements showed a better precision than OBPa, even the two first readings. However, more research needs to be carried out before recommendation of the use of this technique in routine practice.
aCardiac Diagnostic Center
bTriângulo Mineiro Federal University, Uberaba
cHeart Institute of the University of São Paulo Faculty of Medicine, (INCOR-HCFMUSP)
dHeart Institute of the University of São Paulo Faculty of Medicine, (INCOR-HCFMUSP) & Sírio-Libanês Hospital, São Paulo, Brazil
Correspondence to Marco A. Vieira da Silva, MD, Cardiac Diagnostic Center (CDC), Dr Thomaz Ulhoa, 544 Square, 38025-050 Uberaba, Brazil Tel: +55 343 331 5252; fax: +55 343 331 5224; e-mail: firstname.lastname@example.org
Received March 12, 2015
Received in revised form September 1, 2015
Accepted September 3, 2015