Masked hypertension: an independent predictor of organ damage

Over the past 20 years, the increasing use of ambulatory blood pressure monitoring (ABPM) and home blood pressure monitoring (HBPM), in addition to conventional blood pressure measurement, has allowed the identification of different blood pressure conditions, respectively defined as sustained normotension (normal office and out-of-office blood pressure), sustained hypertension (elevated office and out-of-office blood pressure), white-coat hypertension or isolated clinic hypertension (ICH, elevated office and normal out-of-office blood pressure) and its reverse phenomenon (i.e. masked hypertension; MHT, normal office and elevated out-of-office blood pressure). Classification of patients according to these patterns is currently regarded as a useful means for more precise risk stratification and therapeutic decisions. This is because several population-based studies and prospective clinical trials have provided clear evidence on the superiority of out-of-office blood pressure measurement (ABPM or HBPM) versus office or clinic blood pressure measurement to predict the risk of subsequent fatal and nonfatal events [1–3]. Furthermore, several studies have shown that the target organ damage of hypertension (TOD), namely left ventricular hypertrophy (LVH), proteinuria and carotid artery thickening or plaques, correlates more closely with average blood pressure values assessed by ABPM or recorded at home, than with isolated clinic or office blood pressure measurements [4–6]. Discrepancies between the clinical value of different combinations between office and ambulatory or home blood pressure elevation have been reported frequently. Most studies investigating the association of ICH with target organ damage or cardiovascular events conclude that ICH is a condition at relatively low risk of cardiovascular morbidity [7–10], although this statement has been recently questioned [11,12]. Until recently, less attention has been paid to the clinical and prognostic value of the reverse phenomenon (i.e. MHT), despite the potential therapeutic relevance of the finding of elevated ambulatory or home but normal office blood pressure [13]. In this issue of the journal, Hara et al. [14] provide a further contribution in the latter field, by reporting data on the association between MHT and carotid atherosclerosis in a sample of subjects aged older than 55 years taken from the general population of Ohasama, a small town in Northern Japan. The authors show that the atherosclerotic involvement in individuals with MHT is greater than in subjects with ICH or sustained normotension and similar to that of sustained hypertensives. Despite some limitations, this study, carried out in a large population sample, extends and refines previous evidence, as provided by studies often performed in smaller groups of subjects, on the prevalence of MHT and on the impact of this condition on vascular structure. Masked hypertension: prevalence and clinical correlates A growing body of evidence indicates that MHT is a more powerful predictor of cardiovascular events than the isolated elevation in office blood pressure. However, a number of issues related to this condition are still a matter of debate, including the best method to identify those subjects at risk of having MHT, the question of whether the definition of MHT can be applied both to treated and untreated subjects, and the actual prevalence of this phenomenon. Although no definitive data are available, MHT has been estimated to occur in approximately 10–30% of individuals, with the variability in these figures depending on the diagnostic criteria used to identify this condition and on the characteristics of the population examined. In 319 clinically normotensive volunteers who underwent five clinic blood pressure measurements and a 12-h daytime ABPM, MHT, defined as a daytime blood pressure > 135/85 mmHg and an office blood pressure < 140/90 mmHg, was present in 23% of the study sample [15]. Male gender, older age and smoking were more prevalent in subjects with MHT than in their sustained normotensive counterparts. Liu et al. [16], reported that, in 61 out of 295 clinically normotensive subjects (20%), awake ambulatory systolic or diastolic blood pressure exceeded 134/90 mmHg. Obesity, cholesterol levels and nicotine use tended to be higher in subjects with MHT compared to sustained normotensives. In a retrospective analysis of 1494 ambulatory blood pressure recordings performed for clinical indications in untreated (33%) and treated subjects in a community hospital, Ben-Dov et al. [17] found a 11% prevalence of MHT, according to a day-time ambulatory blood pressure cut-off level of 135/85 mmHg. The masked hypertension phenomenon was correlated with male sex, young age and higher awake heart rate. In a subset of 710 untreated older patients included in the Second Australian National Blood Pressure (ANBP 2) study, a substantial proportion had higher day-time (≥ 5 mmHg) systolic and diastolic ambulatory blood pressure than clinic blood pressure (13 and 24%, respectively), with smoking status, previous treatment and lower clinical blood pressure levels being the main predictors of this condition [18]. In 136 clinically normotensive children and young adults, MHT, diagnosed according to systolic or diastolic day-time blood pressure ≥ 95th ambulatory blood pressure percentile for sex- and height-adjusted reference values, was present in 11% of the cases, with its prevalence being higher in boys (19%) than in girls (5%) [19]. Two studies also investigated the magnitude of the MHT phenomenon in a general population. In the Ohasama population (i.e. in the same population in which the present study was performed), 10% of subjects with a normal screening blood pressure displayed average 24-h ambulatory blood pressure values ranging between 134/79 and 144/85 mmHg; in 3% of the subjects examined, 24-h ambulatory blood pressure was equal or even higher than 145/86 mmHg [20]. In the PAMELA study, the only published study providing data on MHT based on either home self-measured blood pressure or ABPM, this condition involved 9–12% of a population sample of untreated adult and elderly individuals, depending on the use of either ambulatory or home measurement, and diastolic or systolic blood pressure values [21]. The degree of agreement between ABPM and HBPM in the diagnosis of MHT was prospectively investigated by Stergiou et al. [22] by performing a direct comparison of the two methods in 438 untreated and treated hypertensive patients. The authors found that a similar, but not identical, percentage of subjects with MHT were detected by ABPM (14%) and HBPM (11%). A disagreement in the diagnosis of MHT between ABPM and HBPM was observed in 23% of subjects for systolic and in 30% for diastolic blood pressure. The reproducibility of the MHT pattern is essential for identifying this condition as a clinical entity. In a pediatric population of 535 subjects, MHT persisted in approximately 50% of them over repeated assessments whereas, in 10% of subjects, a progression from MHT to sustained HT was observed during a median follow-up period of 37 months [23]. In a small group of 82 patients studied by Ben-Dov et al. [17], 71% of subjects with MHT according to the first ABPM session confirmed this pattern at the time of the second ABPM, whereas 80% of those with MHT at the second ABPM session had masked or sustained hypertension at the time of the first session [17]. Masked hypertension: target organ damage and prognosis A pioneering study looking at the impact of MHT on preclinical TOD was published by Devereux's group in 1999 [16] (Table 1). For the first time, these investigators showed that MHT patients had a significantly higher left ventricular (LV) wall thickness and mass compared to subjects with sustained normotension. LV mass index (LVMI) was similar, on average, in patients with MHT (86 g/m2) and in those with sustained hypertension (90 g/m2), despite an average difference in clinic blood pressure of 35/16 mmHg between these groups. Compared to patients with sustained normotension, patients with MHT had a greater carotid wall thickness and cross-sectional area, as well as a higher prevalence of atherosclerotic plaques, although these differences were no more significant in a multivariate analysis. In the PAMELA population, average LVMI, LV wall thickness and the prevalence of LVH (14%) in subjects with MHT were similar to those of subjects with ICH (15%) and lower compared to subjects with both office and ambulatory or home hypertension (26%), but greater than in subjects with normal blood pressure in and outside the physician's office (4%) [21]. In a 30-month longitudinal study, we ourselves reported that in treated patients with MHT the decrease of LVMI and the normalization rates of altered LV patterns (LVH or LV concentric remodelling) and of microalbuminuria were significantly lower than in subjects with both office and ambulatory blood pressure reaching the normal range [24]. More recently, Tomiyama et al. [25] offered additional data on this topic by assessing office and 24-h ambulatory blood pressure as well as cardiac and extracardiac TOD in 332 treated hypertensive patients. Patients with MHT had a greater LVMI, maximal intima–media thickness and urinary albumin level than hypertensive patients with both office and ambulatory blood pressure controlled or with ICH. The degree of TOD in MHT, conversely, was similar to that of patients with sustained hypertension.Table 1: Masked hypertension: prevalence and association with target organ damageMHT has been found to entail an increased incidence of cardiovascular events in a number of longitudinal studies (Table 2). Bjorklund et al. [26] investigated the prognostic significance of MHT for cardiovascular morbidity in a population-based cohort of 578 untreated 70-year-old men [26]. They found that both MHT (relative risk = 2.77) and sustained hypertension (relative risk = 2.94) were independent predictors of cardiovascular events. Bobrie et al. [27] evaluated the prognostic impact of clinic and home blood pressure in 4939 elderly treated hypertensives followed-up for a mean of 3.2 years. According to a multivariate model, patients with MHT and those with uncontrolled sustained hypertension had a two-fold increase in the risk of developing events compared to patients with controlled hypertension. Pierdomenico et al. [28] showed that, among 446 treated hypertensive patients with controlled office hypertension, those with MHT were at higher cardiovascular risk (relative risk = 2.28) compared to those with sustained control of both office and out-of-office blood pressure. Among the 1332 participants in the Ohasama study followed for 10 years, the composite risk of cardiovascular mortality and stroke morbidity was two-fold greater in patients with MHT or sustained hypertension than in those with sustained normotension or ICH [29]. Finally, the risk of mortality associated with selected and combined elevations in office, home, and ambulatory blood pressure was recently examined in the PAMELA population by Mancia et al. [30]. Compared to subjects with normal office and 24-h blood pressure, the hazard ratio for cardiovascular death showed a progressive increase from those with selective office elevation (ICH), to those with selective 24-h blood pressure elevation (MHT), up to those with elevation in both office and 24-h blood pressure. This was also the case when the above conditions were identified by considering office and home rather than ambulatory blood pressure values.Table 2: Masked hypertension: prevalence and association with cardiovascular eventsTaken together, these data seem to convey an important general message, namely that out-of-office blood pressure, either monitored at home or in ambulatory conditions over the 24 h, has a greater prognostic value than office or clinical blood pressure readings. Moreover, the progressive increase in out-of-office blood pressure, which occurs going from sustained normotension to ICH, to MHT and to sustained hypertension, indicates that the prognostic relevance of home or ambulatory blood pressure may stand independently from the classification of patients into the above blood pressure categories, given the closer continuous relationship of the cardiovascular consequences of hypertension with out-of-office than with office blood pressure. New insights and limitations regarding the study by Hara et al. The report by Hara et al. [14] in this issue of the Journal confirms and expands previous evidence on the prognostic relevance of MHT by showing that an elevated blood pressure at home, in the presence of a normal office blood pressure, may affect the carotid structure in both untreated and treated subjects. The same study also indicates that the extent of organ damage at the carotid level (i.e. carotid thickness or plaque) in ICH patients is similar to that of sustained normotensives and lower than that of sustained hypertensives. These findings therefore suggest that: (i) clinic blood pressure measurements alone are insufficient to discriminate those subjects who are at higher risk from those at lower risk of subclinical vascular alterations; and (ii) self-measured blood pressure at home is a reliable predictor of carotid atherosclerosis. The results of the study by Hara et al. [14] deserve several comments, but we will limit our considerations to the prevalence of MHT and its impact on organ damage. In the general population of adults older than 55 years studied by Hara et al. [14], the prevalence of MHT based on office and home blood pressure readings was approximately five-fold lower (6.6%) than that of ICH (32.2%), and was generally lower than the prevalence of MHT found in the same population by combining office measurements and ABPM in a previous study [29], in agreement with the observations made by Stergiou et al. [22]. An important criticism of the study by Hara et al. [14] is that clinic blood pressure was measured only on one occasion (at the time of the carotid ultrasound examination after the subject had been resting in the seated position for only 2 min): this probably led to an overestimation of the ICH phenotype. Given the dynamic pattern of the blood pressure changes during a physician's visit [31], and because of the widely documented decrease in office blood pressure after repeated visits, a reliable diagnosis of ICH should be based on several rather than on a single visit in the doctor's office. On the other hand, the demographic characteristics of the population (i.e. old age and a 68% prevalence of women) might have led to an underestimation of MHT. According to the available evidence, young age and male gender are important factors affecting the prevalence of MHT [32]. It is conceivable that the analysis of a more representative population, including a similar proportion of men and women with a broader range of age, would had provided more reliable information on the characteristic prevalence of both blood pressure patterns. Another probably more relevant finding of the study by Hara et al. [14] is that carotid intima–media thickness, a powerful predictor of coronary and cerebrovascular events, was significantly higher in subjects with MHT than in those with ICH or sustained normotension and similar to that of sustained hypertensives. This finding represents a novel observation in the comparison among subjects taken from a general population sample and classified in different blood pressure categories based on clinic blood pressure and self-measured home blood pressure. The data suggest that HBPM is a robust witness of organ damage, independently of clinic blood pressure values. Of note, the stronger association between carotid atherosclerosis and home blood pressure, as compared to clinic blood pressure, was not apparently related to the larger number of home BP measurements because it persisted even when taking into account the same number of readings (two measurements) taken in both settings. This confirms that out-of-office self-measured blood pressure has a greater predictive power than clinic blood pressure, even when a few measurements are used, as previously shown in the frame of the Ohasama study [33]. Some additional limitations of the study by Hara et al. [14] should be acknowledged. First, their conclusions are based on data collected from a small group of subjects with MHT (n = 54), and need to be confirmed in larger population sample. Moreover, having been collected in an Asian population, they should be generalized with caution to populations with different ethnic, demographic and clinical characteristics. Finally, the MHT group was not a homogeneous entity because two-thirds of patients classified as MHT received antihypertensive treatment. Thus, only a minority of MHT were untreated individuals without previous evidence of high blood pressure in the medical setting (‘true MHT’). It is conceivable (although data on this issue are not available) that these two conditions may recognize distinct pathophysiological mechanisms, clinical profiles and association with TOD. Notwithstanding these limitations, the study by Hara et al. [14] further supports the relevance of out-of-office blood pressure values, in particular home blood pressure, in the prognostic assessment of hypertensive patients because conventional blood pressure measurements may fail to identify a notable fraction of individuals at low (ICH) or high risk (MHT). An inaccurate assessment of blood pressure-related risk resulting from a single or a few office blood pressure measurements may lead to over-treating subjects at lower risk (i.e. ICH) and, more importantly, to under-treating subjects at higher risk (i.e. MHT) of developing TOD and cardiovascular events. The practical implication of all the above considerations is that out-of-office blood pressure should be more frequently obtained in the clinical management of hypertension.