Self-measurement of blood pressure in arterial hypertension--preliminary results from the AMPA study

The clinical usefulness of home blood pressure monitoring (HBPM) is still uncertain, and is currently a major topic of scientific debate. Some studies have stressed its potential role in the clinical decision-making process, but there have been few prospective studies addressing this subject. The AMPA study is intended to contribute to this debate, exploring the potential usefulness of this methodology in the clinical setting of arterial hypertension using a prospective, observational and multicenter design. The study included 685 hypertensive patients (346 female), with a mean age of 54.2 +/- 11.1 years (range: 17-86 years). All patients were being followed in primary care centers by their family doctors, and were being treated for arterial hypertension and other comorbidities. Forty-seven patients were smokers (6%), 90 (13%) had a personal history of cardiovascular disease, 42 (6%) were diabetic, 255 (37%) had dyslipidemia, and 31 (5%) were both diabetic and dyslipidemic. Blood pressure (BP) was measured in the brachial artery with a validated automatic blood pressure measurement device (Colson MAM BP 3AA1-2; Colson, Paris). This device has solid state memory (sufficient for 60 measurements) and an adaptable printer. A cuff appropriate for the arm size of each patient was used. All patients were instructed on how to operate the device correctly and how to perform the measurements in compliance with the study protocol. BP was always measured after a 5-minute resting period in a seated position. The protocol consisted of an HBP program over a period of five working days. Each day the patient performed six BP measurements in two different periods: three in the morning (between 6 and 10 am) and three in the evening (between 6 and 10 pm). Other clinical and anthropometric data were also collected. The HBP reference values adopted were 135 mmHg for systolic and 85 mmHg for diastolic BP. Analysis of BP behavior over time demonstrated a significant white-coat effect, with regression to the mean of BP levels after the first day of the HBP program. As a consequence, the first day values were excluded in determining mean HBP. This behavior was independent of gender, and was more pronounced in diabetic patients. Analysis of diagnostic concordance between office BP and HBP showed discrepancies in 27.4% of the patients. This prompted a change in diagnosis based on HBP values, with 133 patients (19.4%) presenting uncontrolled office BP levels but normal HBP values, whi