¹Section of Hypertension and Clinical Pharmacology, University of Connecticut Health Center, Farmington, Connecticut

Correspondence: Dr. George Mansoor, Section of Hypertension and Clinical Pharmacology, 263 Farmington Avenue, Farmington, CT 06030-3940 E-mail: Mansoor@NSO1.uchc.edu

^*This study was supported in part by General Clinical Research Center Grants (MO1-RR-06192 and K23 RR 15545) from the National Institutes of Health (Bethesda, MD), and by grants from the Catherine and Patrick Donaghue Medical Research Foundation (Hartford, CT).

Received 1 April 2004; Revised 1 June 2004; Accepted 9 June 2004.

Abstract

Background: Physicians are commonly uncertain whether a person with office blood pressure (BP) around 140/90 mm Hg actually has hypertension. This is primarily because of BP variability. One approach is to perform self-measured home BP and determine if home BP is elevated. There is a general agreement that if home BP is 135/85 mm Hg, then antihypertensive therapy may be commenced. However, some persons with home BP below this cut-off will have ambulatory hypertension. We therefore prospectively study the role of home BP in predicting ambulatory hypertension in persons with stage 1 and borderline hypertension.

Methods: We studied in a cross-sectional way home and ambulatory BP in a group of 48 patients with at least two elevated office BP readings. The group was free of antihypertensive drug therapy for at least 4 weeks and performed 7 days of standardized self-BP measurements at home. We examined the relationships of the three BP methods and also defined a threshold (using receiver operating curves) for home BP that captures 80% of ambulatory hypertensives (awake BP 135/85 mm Hg).

Results: Office systolic BP (145 13 mm Hg) was significantly higher than awake (139 12 mm Hg, P = .013) and home (132 11 mm Hg, P < .001) BP. Office diastolic BP (88 4 mm Hg) was higher than home diastolic BP (80 8 mm Hg, P < .001) but not different from awake diastolic BP (88 8 mm Hg, P = .10). Home BP had a higher correlation (compared with office BP) with ambulatory BP. The home BP-based white coat effect correlated with ambulatory BP-based white coat effect (r = 0.83, P = .001 for systolic BP; r = 0.68, P = .001 for diastolic BP). The threshold for home BP of 80% sensitivity in capturing ambulatory hypertension was 125/76 mm Hg.

Conclusions: Our preliminary data suggest that a lower self-monitored home BP threshold should be used (to exclude ambulatory hypertension) in patients with borderline office hypertension.