Hypertension has a strong, continuous relationship with cardiovascular risk. Traditionally, OBP measurements have been used for cardiovascular risk stratification and for defining targets of therapy. The classification of BP categories (i.e., optimal BP <120/80; pre-hypertension 120–139/80–89; and hypertension ≥140/90 mmHg) as well as definition of BP targets to be achieved by treatment, has been based on epidemiological studies using OBP measurements [3, 25]. These values cannot be directly extrapolated to HBPM, because meta-analyses of several studies on unselected populations or hypertensive patients [26, 27], comparing HBP and OBP distribution curves, have demonstrated HBP values to be lower than corresponding OBP values. Longitudinal studies in general populations [28–37] and in hypertensive subjects [38–40] as well as clinical trials on the use of HBPM have confirmed that the cut-off limit to define hypertension based on HBP should be lower than that used for OBP [41, 42]. Although the relationship between BP values self-measured at home and the incidence of CV morbidity and mortality should be further clarified by prospective studies, there is an agreement to diagnose hypertension when HBP is >135/85 mmHg (corresponding to an OBP of >140/90 mmHg). Prospective data are still needed, however, to formally recommend the proposed thresholds of <120/80 mmHg and <130/85 mmHg to define optimal and normal HBP, respectively. A couple of studies suggested that HBP thresholds for hypertension in high-risk patients might be lower than 135/85 mmHg [30, 39]. Although the target HBP to be achieved with treatment should logically be below the threshold used to diagnose hypertension (i.e., <135/85 mmHg), these target HBP levels are still unknown, being currently explored by the ongoing HBPM studies [43]. Although attaining therapeutic goals may be difficult in some patients, it should be remembered that, even if BP is not fully controlled, each mmHg of reduction in HBP is important, as it contributes to the prevention of CV complications.

As a general remark, it has to be acknowledged that the evidence available to support the prognostic value of HBPM is less than for ABPM, also because of the smaller number of outcome studies available so far [8, 9]. When averaged over a period of a few days, home BP measures have been shown to significantly predict the development of major nonfatal cardiovascular events (myocardial infarction, stroke) [28–35, 38, 44–52] as well as cardiovascular (fatal cardiovascular events) and all-cause mortality [28, 34, 35, 39, 47, 48, 53]. In most available studies, the prognostic value of HBPM has been found superior to that of OBP measurements with one exception, where a similar predictive value was observed for both techniques [38] (see Fig. 2.1 and Table 2.2).

Longitudinal and cross-sectional studies have reported that target organ involvement, including left ventricular mass index (LVMI), carotid intima-media thickness, and microalbuminuria, is more strongly correlated with HBP measurements than with OBP measurements in patients with hypertension [51, 54–60] as well as in patients with chronic kidney disease (CKD) on hemodialysis (HD) [61], in elderly people, in women with pre-eclampsia, and in hypertensive patients with diabetes [8]. In the case of patients with CKD, HBP has been shown to be a better predictor of progression of CKD (as assessed with eGFR) [40, 62], including its progression to end stage renal disease (ESRD) [39] and of cardiovascular events and mortality [63] than OBP. In particular, in ESRD, HBPM may be more informative than pre- and post-dialysis OBP readings as it provides BP measurements that are more representative of the BP load over the interdialytic period. Indeed, several studies in ESRD have found HBP to be prognostically superior than OBP also in predicting subclinical organ damage (i.e., LVH) [61] and cardiovascular events (i.e., all-cause and CV mortality) [64, 65].

As discussed above, HBPM and ABPM provide out-of-office BP measurements detecting BP changes in real life conditions and preventing the alarm reaction associated with OBP [66]. It is thus not surprising that BP levels measured in the clinic setting are in general higher than ambulatory BP measurements performed out of the clinic environment [67]. This is considered a major explanation for the frequently observed disagreements between OBP and out-of-office BP measurements when classifying hypertensive subjects [30]. Indeed, when considering the threshold values to define hypertension using OBP (≥140/90 mmHg) and HBP or 24h ABP (≥130/80 mmHg), a given individual may fall into one of four BP categories: sustained normotension (normal office BP and normal home or 24h ABP), sustained hypertension (high OBP and high home or 24h ABP), white coat hypertension (high office BP and normal home or 24h ABP), or masked hypertension (normal office BP and high home or 24h ABP) (Fig. 2.2).

Evidence on the ability of HBPM to identify WCH and MH [56, 68] was provided in a report of the Pressioni Arteriose Monitorate e Loro Associazioni (PAMELA) study in which the initial diagnosis of WCH (i.e., identified as office BP >140/90 mmHg and 24-h BP mean <125/79 mmHg or home BP <132/82 mmHg) was reassessed 10 years later. Overall, the study showed similar results in the ability of HBPM and ABPM for identifying WCH, sustained hypertension, true normotension, and masked hypertension, even if a substantial percentage of subjects changed from one category to another, including progression to sustained hypertension (Fig. 2.3).

In the light of the available evidence supporting the prognostic and clinical advantages offered by HBPM, current international and national guidelines recommend the use of HBPM as part of the routine diagnostic and therapeutic approach to hypertension, particularly in treated patients [8, 69–74]. By providing accurate and frequent BP measures at regular time intervals over several days, weeks, or months, in a setting of typical daily living, HBPM is able to accurately track changes in BP levels induced by antihypertensive treatment and becomes a better indicator of BP control than OBP measurements alone [8]. HBPM may be an excellent tool to assess and improve the achievement of BP control, particularly in patients with apparent resistant hypertension in whom BP cannot be easily controlled even with several classes of antihypertensive medications. In support of this concept, several studies exploring the benefits of HBPM for the long-term management of patients on antihypertensive therapy have shown that when properly implemented, HBPM may significantly increase achievement of BP control when compared to conventional OBP [75, 76], while reducing the need of follow-up medical visits [77]. The benefits of HBPM in this regard may be derived from several factors. First, the use of HBPM improves adherence to prescribed treatment (see below). Secondly, in subjects who receive antihypertensive treatment, OBP measurements alone may be inaccurate in assessing true BP control. For instance, the alerting reaction to the medical visit may continue to be present in anyone treated for hypertension, regardless of the number of drugs being taken [78]. It is not uncommon to find patients with mild hypertension based on HBPM or ABPM who yet appear to have severe hypertension in the clinic, due to a white coat effect in this condition [79], or treated subjects who, despite achieving adequate out-of-office BP control with antihypertensive drugs, continue to present elevation in office BP levels because of a persistent emotional reaction to the medical visit. This phenomenon, which is equivalent to WCH in untreated patients, has been addressed as “white coat resistant hypertension” (WCRH) or false resistant hypertension in order to emphasize its occurrence in subjects receiving antihypertensive treatment [14].

Observational and interventional studies in treated hypertensives implementing OBP measures along with ambulatory or home BP monitoring have shown overwhelmingly that up to one-third of treated hypertensives may be mistakenly classified as having resistant hypertension, when they actually have “false resistant hypertension” due to a persisting white coat effect [80]. A condition of greater clinical concern is masked resistant hypertension (MRH) or false BP control (i.e., BP appear to be controlled based on OBP, but is elevated when out-of-office BP levels are recorded)—this condition has been also reported to occur in about 30 % of treated subjects [81, 82].

The high prevalence of MRH and WCRH among treated hypertensive individuals further reinforces the clinical relevance of identifying these conditions. On the one hand, identification of WCRH would prevent undesirable modifications of antihypertensive treatment, i.e., an unnecessary increase in dose or number of antihypertensive drugs, and reduction of the chance of adverse effects associated with improperly prescribed multidrug therapy that often interferes with patients’ quality of life, leading in the end to poor compliance with treatment. At the same time, it would reduce the expenditures associated with unnecessary additional pharmacological treatment and/or unnecessary interventional device-based strategies (i.e., carotid baroreceptor activation [83] and renal denervation [84]) for the management of resistant hypertension. Indeed, given the elevated costs and the invasive nature of these approaches, as well as their potential adverse effects when improperly indicated, discarding WCRH based on out-of-office BP measures is currently considered among the eligibility criteria before proceeding with interventional treatment of resistant hypertension [85]. In contrast, identification of MRH would indicate the need to implement early modifications on antihypertensive treatment in order to prevent development/progression of subclinical organ damage and cardiovascular events associated with this condition.

Regarding the ability to identify masked hypertension and white coat hypertension, several studies have comparatively explored the performance of HBPM against the reference standard for out-of-office BPM represented by ABPM . Although MH was first studied with ABPM [86], it has been demonstrated that HBP can be as reliable as ABPM in identifying this phenomenon as well as the associated target-organ damage associated with MH [87]. Evidence has also been provided that HBPM is as reliable as ABPM in identifying WCH [87] and useful in identifying “truly” hypertensive patients likely to benefit from implementation of antihypertensive therapy from those with WCH in whom antihypertensive treatment is probably not needed [41]. In a recent study conducted in a group of subjects on stable treatment with ≥3 antihypertensive drugs using ABPM as reference method [88] in which resistant hypertension was defined as elevated OBP (≥140/90 mmHg) and true resistant hypertension as concomitant elevation in-office and out-of-office BP (SBP and/or DBP ≥ 135/85 mmHg for HBP or awake ABP), there was agreement between ABP and HBP in diagnosing clinic or “white coat” resistant hypertension in 82 % of the cases (59 % with and 23 % without clinic resistant hypertension; kappa 0.59). Regarding the diagnosis of true resistant hypertension, there was agreement between ABP and HBP in 74 % of the cases (49 % with and 25 % without true resistant hypertension; kappa 0.46). The sensitivity, specificity, and positive and negative predictive values for HBP in detecting white coat resistant hypertension were 93 %, 63 %, 81 %, and 83 %, respectively. The respective values for HBP in detecting true resistant hypertension were 90 %, 55 %, 71 %, and 82 %, indicating that HBP may be a useful tool in the evaluation of false and true resistant hypertension [88].

Based on the above data, it may be concluded that a proper assessment of BP control and classification of treated hypertensive patients with the combined use of office, ambulatory, and ideally home BP measurements are essential for defining the need of performing additional diagnostic procedures (i.e., screening tests for secondary causes of resistant hypertension) and/or implementing more aggressive pharmacological or interventional strategies (Fig. 2.4) [89].

While emphasizing the above advantages of HBPM in assessing BP control by treatment, we have also to acknowledge that HBPM may not provide information on BP levels during night-time sleep, which have shown to be of major clinical relevance because of their demonstrated prognostic value [35, 44, 90–93]. However, in recent years validated, memory-equipped devices have been designed that can be programmed to provide nocturnal BP readings comparable to those obtained with 24-h ABPM [22–24].

HBPM is admittedly less effective than ABPM in assessing the time distribution of BP control by treatment over 24 h. However, HBPM performed in the morning (before drug intake) and in the evening over different days may provide useful information about the efficacy of therapeutic coverage over 24 h and in the long term and may identify cases of morning hypertension attributable to insufficient duration of action of prescribed antihypertensive medications.

HBPM allows patients to perform repeated and regular BP measurements over extended periods of time and may be particularly advantageous in the case of treated hypertensive subjects with CKD, and particularly in those with ESRD. In hemodialysis (HD) patients, BP control poses unique challenges because of the marked reductions in intravascular volume immediately after HD and its progressive increases throughout the inter-dialytic period, which induce an extremely variable behavior of BP [94]. In this context, HBPM provides potential advantages such as the possibility of sampling BP at various times throughout the inter-dialytic period to aid in tracking daytime and day-to-day BP variations and providing BP measurements that are more representative of subject’s actual BP burden.

In view of the limitations characterizing OBP measurements, it becomes clear that an adequate assessment of BP control and a proper diagnosis of resistant hypertension cannot be based on just isolated OBP readings. Indeed, a recent position paper on ABPM of the European Society of Hypertension [14] recommends performing 24-h ABPM and/or HBPM for detecting the presence of WCH and identifying the presence of true hypertension and masked hypertension in all patients with uncomplicated, stage 1 and 2 hypertension before starting antihypertensive drug therapy. Based on the evidence from several studies supporting the clinical value of ABPM either for selecting patients for treatment or for assessing the effects of antihypertensive drug therapy, ABPM is currently considered the standard method for confirming the diagnosis of hypertension in clinical practice [12, 14] and for assessing BP control in treated hypertensive patients [3, 6, 14, 10]. However, ABPM is not always available everywhere and requires trained clinic staff and specialized equipment and software for its analysis [9].

When performed on a regular basis and following standardized protocols [9], repeated BP measures obtained by patients at home result in accurate and frequent out-of-office BP measurements not only during a single day, but also over several days, weeks, or months in the non-medical setting, hence providing more reliable measures not only on the degree, but also on the consistency of BP control over time [9].

In view of the available evidence supporting the superior prognostic value of home vs. office BP levels, as well as the clinical advantages of HBPM, current hypertension guidelines recommend more extensive use of HBPM not only for the initial diagnostic approach to hypertension (i.e., to identify “truly” hypertensive patients, likely to benefit from implementation of antihypertensive therapy [41]), but also for the long-term follow-up of treated hypertensive patients, even if they have controlled OBP, in order to better define the actual BP normalization rate achieved by various drug regimens [8, 9, 3, 25, 95]. Although HBPM shares many of the advantages of ABPM, including a cost-effective approach to the diagnosis of hypertension, it should not be considered as a substitute but rather as a complement to ABPM, since these methods are likely to pick up different types of BP behavior in a person’s activities of daily living.

Poor adherence to therapy has been recognized as one of the most important factors contributing to uncontrolled hypertension. By encouraging patients to become actively involved in their care, and by positively affecting their perceptions about the management of hypertension, HBPM offers the possibility to improve patient’s compliance and adherence to lifestyle changes and/or medical treatment [96]. Recent meta-analyses of randomized controlled trials have shown that compared to usual care based on OBP measurements, HBPM-guided antihypertensive treatment may significantly increase rates of achievement of BP control [75, 97] probably as a consequence of better compliance to treatment. In fact, HBPM is being increasingly implemented in clinical settings not only to guide antihypertensive therapy and to assess long-term BP control, but also as a means to improve patient’s compliance and adherence to antihypertensive treatment [98]. Another important advantage of HBPM in clinical practice is that it may help to overcome therapeutic inertia, since more information is provided to practitioners that allow more appropriate clinical decisions.

HBPM may offer clinically relevant information when considering BPV over long periods of time. Although most studies on the prognostic relevance of BPV have focused on short-term BP changes assessed from 24-h ABPM, evidence from recent studies and clinical trials has suggested that an increased BPV in the midterm (day-to-day) and in the long-term (i.e., between weekly, monthly, or yearly visits) relates to adverse implications for CV prognosis [99–103]. Although an extensive assessment of BPV for intermediate periods could theoretically be obtained by performing ABPM over consecutive days (i.e., during 48 h or more), this approach is neither well-accepted by patients nor available in all clinical settings. An alternative method for assessment of day-by-day BPV consists of its calculation from BP measurements performed by patients at home over several days. Although HBPM cannot provide extensive information on nocturnal BP and BP profiles as ABPM does, a major advantage of this technique is that it provides information on the consistency of BP control over time earlier than when considering long-term visit-to-visit BPV, thus allowing early adjustment of antihypertensive treatment (and thus timely preventing development/progression of organ damage associated with inconsistent BP control). Besides, HBP monitors are widely available and are well-accepted by patients and are a more feasible approach for the evaluation of day-by-day BPV by applying different metrics: (1) Blood pressure standard deviation [104, 105], but with accounting for its dependence on mean BP levels, i.e., by calculating the coefficient of variation (SD × 100/BP mean) [104]; (2) morning maximum and minimum blood pressure (MMD); (3) “average real variability” (ARV), computed as the average of the absolute differences between consecutive BP measurements, focusing on the sequence of BP readings, thus reflecting reading-to-reading, within-subject variability in BP levels [106]; (4) variance independent of the mean (VIM), a method proposed to exclude the effect of mean BP from BPV by applying nonlinear regression analysis (i.e., plotting SD against mean) [99]. These indices of day-by-day BPV have been shown to be of prognostic value as indicated by a series of studies in which an increased day-by-day BPV independently of average home BP levels was predictive of development, establishment, and evolution of cardiac, vascular, and renal organ damage [107]. In a cross-sectional analysis of a population of never-treated participants with hypertension, an increased day-by-day BPV in home systolic BP (assessed as the maximum mean triplicate in home systolic BP over 14 consecutive days) was positively correlated with left ventricular mass index, increased carotid intima-media thickness, and urinary albumin/creatinine ratio over and above mean home SBP levels [107]. In a population of type 2 diabetes patients from Japan, increasing values of day-by-day variability (assessed as CV of morning and evening HBP measured over 14 consecutive days) were significantly higher in subjects presenting with macroalbuminuria (i.e., urinary albumin excretion ≥300 mg/g creatinine). Additionally, the CVs of morning systolic and diastolic BP and evening systolic BP were significantly correlated with urinary albumin excretion independently of other confounders [108]. A further report, also in type 2 diabetic patients, found higher values of SD of morning systolic HBP to be associated with increased arterial stiffness (i.e., higher pulse wave velocity) independent of other known risk factors [109]. Another study, conducted in a cohort of hypertensive patients, found home systolic BPV and max systolic BP to be associated with urinary albumin excretion [110]. In the frame of the Home Blood Pressure for Diabetic Nephropathy (HBP-DN) study , a prospective study in type 2 diabetic patients with microalbuminuria, higher values of SD of home systolic blood pressure (HSBP) were observed among subjects with the lowest values of estimated GFR (eGFR) [111]. Regarding the predictive value for cardiovascular events and mortality, the two main population studies exploring the prognostic value of mid-term BPV have found increasing values of day-by-day BPV to be associated with an increased risk of fatal and nonfatal cardiovascular events [101–103]. In the Ohasama study from Japan, increasing values of variability in systolic HBP were associated with a higher risk of the composite end point of cardiac and stroke mortality, but only with a significant risk of stroke mortality, when the outcomes were independently considered [101]. In another report from the Ohasama study, increasing values of variability in systolic HBP were associated with a higher risk of cerebral infarction in ever smokers, but not in never smokers [102]. When the prognostic value of novel indices of BPV derived from self-measured HBP was evaluated in the population of the Ohasama study, increasing values of VIM and ARV, but not of morning maximum and minimum blood pressure (MMD) determined on a median of 26 readings, were associated with an increased risk of cardiovascular and total mortality. However, when adjustment was performed by accounting for average BP and common confounders, the incremental predictive value of VIM, MMD, and ARV over and beyond HBP level was only marginal (i.e., from <0.01 to 0.88 %) [102]. In the Finn–Home study in a cohort of adults from the general population [103], increasing variability in systolic and diastolic HBP measures performed over seven consecutive days was associated with a higher risk of cardiovascular events after 7.8 years of follow-up, which remained significant even after adjusting for age and average HBP levels, thus supporting the additive value of HBP variability in predicting CV prognosis [103]. Contrasting results were reported after 12-years of follow-up in a Belgium population in which no predictive value for HBP variability was observed for either cardiovascular mortality or morbidity after accounting for average BP levels [112]. In relation to the effects of antihypertensive treatment, despite the wide availability of monitors for HBP monitoring, only few interventional studies in hypertension have implemented routine assessment of HBPV in order to address whether reducing day-by-day BP variability with antihypertensive treatment in addition to reducing average home BP levels is associated with improvements in cardiovascular protection.