Assessing the white-coat effect: which blood pressure measurement should be considered?

The alerting reaction and blood pressure increase triggered in a patient by consultation may reduce the accuracy of conventional office blood pressure measurements and prevent them from accurately reflecting the ‘true’ blood pressure levels in a given subject. The occurrence of a transient pressor response when blood pressure is measured in a clinic or office environment was first described in 1897 by Scipione Riva-Rocci [1]. Nearly 50 years later, Ayman and Goldshine [2] observed that blood pressure values measured by the patient at home were invariably lower than blood pressure values recorded by physicians in their office, and that such a difference persisted over a relatively long follow-up period. The quantitative assessment of the magnitude and the time course of this phenomenon was first provided in 1983 by Mancia et al. [3,4] through the use of continuous intra-arterial ambulatory blood pressure recordings carried out before and during a doctor's visit (Fig. 1) [3,4].Fig. 1These original observations were subsequently confirmed by several other studies, which: (i) termed the blood pressure rise associated with the doctor's visit the ‘white-coat effect’ (WCE) [5]; (ii) provided clear evidence that clinic blood pressure is, in most cases, higher than ambulatory or home blood pressure; and (iii) offered an indirect quantification of the WCE through calculation of the difference between blood pressure values measured in the presence of a physician and blood pressure values obtained in daily life conditions [6,7]. The quantification of this phenomenon is believed to be of clinical importance because it may reduce the possibility of an overestimation of patient's blood pressure levels in the absence of treatment, and/or of an underestimation of the effect of antihypertensive drugs. How to measure white-coat effect and to assess isolated office hypertension The direct and precise quantitative assessment of the blood pressure rise associated with the doctor's visit, obtained by intra-arterial recording [3,4], is obviously unsuitable for use either in a clinical setting or in large-scale human research. An alternative method could be the use of devices for non-invasive continuous finger blood pressure-monitoring at the time of consultation [8,9,11–13]. Although useful in a research setting, this approach is hardly applicable in clinical practice due to the high cost of finger blood pressure-monitoring systems and their overall complexity. This has stimulated the development of ‘surrogate’ solutions to quantify the WCE that might be more easily used in a clinical environment. Such a development was made easier by the introduction of non-invasive automated blood pressure-monitoring devices that are able to provide blood pressure values out of the doctor's office [10]. The most common approach is to assess the WCE as the difference between office and average daytime ambulatory or home blood pressure values [5], based on the assumption that the condition characterized by a persistently high blood pressure in the physician's office and a persistently normal blood pressure away from the clinical environment, commonly referred to as ‘white-coat hypertension’, could reflect the persistence over time of an alerting reaction to the physician's visit. However, it has been shown that the difference between office and ambulatory blood pressure values does not specifically reflect the acute blood pressure rise induced in the patient by the physician's visit, as assessed directly by continuous blood pressure monitoring before and during consultation [8,11–13], and therefore does not have much to do with the true WCE. This is not surprising because the difference between office and ambulatory blood pressure values obviously depends on factors affecting not only blood pressure values in the doctor's office, but also blood pressure values in daily life. Indeed, several observations have raised the possibility that behavioural factors affecting daytime blood pressure might be even more important determinants of the clinic–daytime blood pressure difference than the pressor effect of the physician's visit. Thus, the term ‘white-coat hypertension’ is inappropriate when referring to a clinic–daytime blood pressure difference characterized by blood pressure values that are elevated in the doctor's office and normal in ambulatory conditions. Similar problems exist for the difference between clinic and home blood pressure [14]. Against this background, the 1999 WHO/ISH Guidelines recommended that the condition characterized by a persistently high clinic blood pressure and normal daytime ambulatory or home blood pressure values should be named ‘isolated office hypertension’ instead of using the appealing, but misleading, term ‘white-coat hypertension’ [15,16]. A brief glossary of the various terms used in this setting is provided in Table 1[17].Table 1: White-coat-related phenomenaRecently, another method for indirectly quantifying the WCE has been proposed, namely the adoption of blood pressure measurements repeatedly taken in the clinic or in the physician's office, when waiting for consultation, either by technicians, nurses or automated devices [18]. In this issue of the journal, Gerin et al. [19] report on the results of a study aimed at exploring whether this approach may represent a valuable solution to the problem. Resting blood pressure taken in a non-medical setting (a laboratory room) by a research assistant was compared with the blood pressure values (obtained at rest and by the same research assistant on a different day) in the physician's office. Both sets of resting values were also compared with ambulatory blood pressure values obtained during daytime activities. Resting blood pressure levels taken in a non-medical setting were lower than those taken in the clinic before the patient saw the physician, with this being the case both in normotensive and in hypertensive subjects. As a result, in hypertensive patients, the surrogate measure of the WCE based on the difference between blood pressure values taken by the physician and those taken in a non-medical environment was significantly higher than the assessment of WCE based on the difference between blood pressure values measured in the former situation and those taken by the research assistant in the clinic when the patient was waiting to see the doctor. At variance from the results of previous studies based on continuous blood pressure recording before and during the physician's visit [3,4,8], in normotensive individuals, blood pressure values measured by the physician were slightly lower than blood pressure values assessed in a non-medical setting, leading to the conclusion of an absence, or even an inversion, of the WCE in subjects with normal blood pressure when using this method. The conclusion drawn by Gerin et al. [19] is that the WCE may therefore characterize not only the period when the patient actually sees the physician, but also the entire period spent in a clinic setting. This means that any improvement of the methods for measuring blood pressure in the clinic or in the physician's office is unlikely to remove the confounding influence of the WCE on conventional blood pressure measurements. Thus, to properly account for this phenomenon, out-of-office blood pressure measurements appear to be necessary. This has a practical relevance, because it undermines the suggestion of using automated blood pressure-measuring devices in the clinic or office waiting room, to obtain blood pressure measurements prior to the appearance of the physician, with the aim of removing the diagnostic error associated with a WCE. In this regard, the data by Gerin et al. [19] are in line with the results obtained by Stergiou et al. [20] in their study exploring the difference between self-blood pressure measurements obtained at home and in a clinic environment. Self-blood pressure measured in the clinical setting was higher than self-measured blood pressure taken at home, suggesting that the setting where blood pressure is taken may be a more important contributor to blood pressure alteration than the subject who is in charge of the measurement. Another methodological implication of the data by Gerin et al. [19] is that the WHO–ISH Hypertension Guidelines [15] proposal, to use the term ‘isolated office hypertension’ rather than the term ‘white-coat hypertension’ when referring to conditions characterized by persistently elevated office blood pressure accompanied by normal out-of-office blood pressure values, might also apply to the description of the acute blood pressure rise occurring at the time of consultation. On the basis of their results, which show that, in hypertensive patients, blood pressure is elevated in a medical setting even before the patient sees the physician, Gerin et al. [19] suggest that the term WCE might be more properly changed into ‘isolated office effect’ [19]. However, this suggestion disregards the data obtained by continuous blood pressure-monitoring during the physician's visit, which showed that the appearance of the physician next to the patient does induce a marked blood pressure rise on top of the blood pressure level existing prior to the visit in the medical environment [3,4,8,11–13]. Furthermore, the conclusion that normotensive subjects are characterized by an inverted difference between blood pressure values measured by the physician and ‘resting’ or ambulatory blood pressure values measured in his/her absence needs to be critically re-assessed. For example, in the study by Gerin et al. [19], hypertensive patients were older than normotensive subjects. Thus, the probability exists that this somewhat unexpected finding was due to an age-related increase in cardiovascular reactivity to the stress associated with the physician's visit, as well as to a reduced daytime physical activity in the older hypertensive group [21]. In conclusion, the paper by Gerin et al. [19] clearly confirm that WCE is still a difficult phenomenon to diagnose and quantify. It also indicates that out-of-office blood pressure-monitoring appears to be the best method for identifying the ‘real’ blood pressure level of a patient, and thus is a more sensitive prognostic marker than isolated clinic or office readings, as shown in several studies [22–25].