BackgroundResults of several studies have shown that subjects with white-coat hypertension (WCH) have more target-organ damage than do normotensive controls with similar ambulatory blood pressures.ObjectiveTo investigate whether this is due to a selection bias.SettingSeventeen hypertension clinics in northeast Italy.Main outcome measuresEchocardiographic data in relation to WCH status.Patients and methodsMild hypertensive subjects from the HARVEST (n = 565) who underwent two ambulatory blood pressure monitorings 3 months apart and M-mode echocardiography, and 95 normotensive control subjects.ResultsFrom first ambulatory monitoring, 90 hypertensive subjects were classified as having WCH (mean daytime blood pressure < 130/80 mmHg). Their 24 h blood pressure was similar to that of the normotensive subjects, but their left ventricular mass index was greater. From second ambulatory monitoring, only 38 of the 90 subjects still had WCH, whereas 24 h blood pressure in the other 52 had risen beyond the limit of WCH. Left ventricular mass index (89.2 ± 2.4 g/m2), wall thickness (18.1 ± 0.3 mm), and relative wall thickness (0.359 ± 0.006%) of the 38 subjects with WCH at both recordings were still greater than those of the normotensive subjects (82.4 ± 1.5 g/m2,P= 0.02; 17.2 ± 0.2 mm,P= 0.002; and 0.337 ± 0.004%,P= 0.025) and similar to those of the 52 subjects who no longer had WCH (88.5 ± 2.0 g/m2, 18.7 ± 0.2 mm, and 0.375 ± 0.005%, all NS).ConclusionsOwing to regression toward the mean, over 50% of the subjects with WCH could no longer be classified as such from repeated ambulatory monitoring, indicating that the current diagnosis of WCH is subject to selection bias. Cardiac remodeling was present also in the subjects confirmed to have WCH by repeated blood pressure recording, suggesting that the effect of WCH has an actual impact on target organs.