At present, clinic blood pressure (BP) evaluation is being increasingly complemented by ambulatory BP measurements for the evaluation of haemodynamic patterns during daily activities and sleep. Nondipping pattern, a measure of decreased attenuation of nighttime over daytime BP, has been correlated with enhanced target organ damage and adverse cardiovascular (CV) outcomes in different clinical settings beyond pure hypertensive cohorts. As the nondipping pattern is a derivative extract of both daytime and nighttime BP, it is yet questionable whether the crude estimate of nocturnal BP is superior to daytime BP and nondipping pattern in the prediction of subclinical damage and CV events. In this review, we aimed at comparing the CV predictive value of the nondipping pattern with that of nocturnal BP using cross-sectional and longitudinal data obtained from different cohort studies within the past 10 years. Our findings suggest that nocturnal BP including the phenotype of isolated nocturnal hypertension is better associated with CV target organ damage and ‘hard end points’ as compared with the nondipping pattern.
Ambulatory blood pressure (BP) monitoring (ABPM) is a widely used tool in hypertension management,1 and compared with isolated office BP, it provides first, a more accurate estimation of the overall cardiovascular (CV) risk burden and second, demonstrates higher reproducibility in the prediction of target organ damage and adverse CV outcomes.2
Along with 24-h systolic and diastolic BP, both reflecting the cardinal clinical components of the haemodynamic load magnitude, the prognostic impact of diverse time-dependent ABPM components (namely daytime and nighttime BP) and circadian BP variability patterns (such as dipping status, night-to-day BP ratio, early morning surge) have been previously investigated.3 Although both averaged nocturnal BP values4, 5, 6 and blunted nocturnal decrease with respect to daytime BP7, 8, 9 were found to be associated with subclinical target organ damage and to heighten CV morbidity and mortality, their relative importance is still controversial.
In this review, we aimed at addressing whether either averaged nocturnal BP or estimates of nighttime-focused circadian BP variability, namely dipping and nondipping patterns, are better associated with intermediate (that is, target organ damage) and hard CV outcomes (that is, CV morbidity and mortality), acknowledging that these ABPM-derived clinical qualifications may coexist. In this perspective, we reviewed 144 clinical studies conducted between 2000 and 2009 identified in the PubMed and Cochrane Database by implementing the search terms: ‘nocturnal hypertension’, ‘nocturnal BP’, ‘nighttime BP’, ‘dipping pattern’, ‘nondipping pattern’, ‘BP nondipping’, ‘BP dipping’ and each one of the aforementioned quoted phrases plus ‘cardiovascular disease’, ‘cardiovascular morbidity/mortality’, ‘all-cause mortality’, ‘cardiovascular risk’, ‘myocardial infarction’, ‘heart failure’, ‘stroke’, ‘cerebral dysfunction’, ‘left ventricular hypertrophy’, ‘left ventricular mass’, ‘relative wall thickness’, ‘left atrial enlargement’, ‘diastolic dysfunction’, ‘arterial stiffness’, ‘pulse wave velocity’, ‘increased intima-media thickness’ and ‘cardiac/vascular damage’; among them, those that recruited >100 participants were considered eligible for analysis and divided as per study design; in these lines, 43 cross-sectional and 33 longitudinal studies, as well as 3 meta-analyses were finally analyzed (Figure 1).
ABPM estimates: at the crossroads of definitions
Dipping status is an at least 10% reduction in both systolic and diastolic BP during nighttime compared with daytime, and BP fall <10% reflects the nondipping status. Dippers and nondippers exhibiting a BP fall >20% and increased averaged BP during nighttime versus daytime, respectively, comprise two extreme subgroups, namely extreme dippers and reverse dippers.10 Another way for defining the dipping/nondipping status is by night-to-day BP ratio, and accordingly, dippers lie between 0.90 and 0.80, nondippers between 0.91 and 1.00, extreme dippers below 0.80 and reverse dippers above 1.00. However, the lack of unanimity in the above definitions and selective implementation of nocturnal systolic or diastolic BP fall alone, further complicate the clinical importance of the dipping/nondipping pattern.
According to the American Heart Association Council on High Blood Pressure Research and the European Society of Hypertension, nocturnal BP is considered within the normal range if the average of nighttime values is lower than 120/70 mm Hg,1, 10 whereas values higher than 125/75 mm Hg are deemed abnormal.10 Nighttime is usually designated through arbitrarily defined fixed time intervals that are either wide (from 2200 hours to 0600 hours) or narrow (from midnight to 0600 hours). The use of narrow fixed intervals in the absence of sleeping diaries is considered a more accurate estimate of sleeping BP;11 consequently, not accounting for nighttime awakenings and daytime naps might contribute to misclassification of nocturnal hypertension.
In the majority of studies so far, the proportion of successful readings during ABPM is not reported as the measure of testing accuracy, and additional testing is usually performed only once. With respect to the latter, there is an ongoing debate for the reduced reproducibility of ABPM measurements possibly resulting in misclassification of the dipping/nondipping status.12 Furthermore, shift and nighttime work constitute a further issue not addressed by the design of the majority of ABPM-oriented studies, and finally, variations in nocturnal body position and diurnal physical activity may affect nighttime and daytime BP, respectively.13
The nondipping profile is frequently accompanied by enhanced nocturnal BP; thus, the discrimination between them might be an eternal unfulfilled target to address. Furthermore, both are not always concomitantly present and the pathophysiological and clinical significance of each may differ. In a study of repeated ABPM, agreement in diagnosing nocturnal hypertension on the basis of two measurements was greater using an absolute nocturnal systolic BP threshold value rather than the percentage of nocturnal systolic BP fall.14 In addition, both the nondipping pattern and the elevated nocturnal BP are associated with adverse cardiometabolic profile with pathophysiological mechanisms not yet fully elucidated. However, it could be suggested that sympathovagal imbalance,15 disturbed breathing during sleep,16 increased salt sensitivity,3 leptin and insulin resistance17 and endothelial dysfunction18 constitute a unified potential pathophysiological substrate for impaired nighttime haemodynamics.
Nocturnal BP phenotypes and CV organ damage
Although there could be a sex-guided difference in the relationship between the dipping profile and the left ventricular (LV) mass,19 the latter represents a powerful predictor of nonfatal cardiac and cerebrovascular events, as well as of that CV and total mortality in various settings, including hypertension (Tables 1a and b). Twenty-four-hour BP has been extensively associated with increased ventricular mass,20 whereas there are no prospective studies addressing the comparative prognostic value of either nocturnal BP values or night-to-day BP fall on the same outcome. On the contrary, a large number of cross-sectional studies in hypertensive subjects have established an almost constant association between nocturnal BP phenotypes with cardiac adaptations including abnormal LV mass and geometry.7, 8, 12, 21, 22 In addition, nondippers and those with hypertensive nocturnal BP values demonstrated impaired LV diastolic function, abnormal aortic root diameters and left atrium structural properties compared with dippers and those with normotensive nocturnal BP levels, respectively.23, 24 However, other studies have reported no difference in cardiac adaptations among dippers and nondippers,25, 26, 27, 28 whereas nocturnal hypertension was constantly related to LV mass, an issue not addressed by absolute nocturnal systolic and diastolic BP.25 It is worth observing that the lack of accordance among studies in the association between the nondipping profile and CV adaptations could be at least partly attributed to the different definitions of dipping pattern.
By accounting for divergences in the definitions of nocturnal BP phenotypes, the evidence so far qualifies the categorically commissioned nocturnal hypertension as the main determinant of cardiac adaptations. Indeed, it might be an absolute nocturnal BP threshold below of which impaired cardiac adaptations are less likely to occur, and additionally, the nondipping status in the context of normal daytime BP levels might not be clinically relevant to intermediate cardiac outcomes.
Vascular and cerebral adaptations
Carotid intima–media thickness and pulse wave velocity are established surrogates of subclinical progressive atherosclerotic vascular damage and independently predict the development of hard CV outcomes (Tables 1a and b).29, 30 Nocturnal hypertension with respect to the nondipping pattern more frequently determines impaired vascular function underlining the importance of the magnitude of BP levels on the progression of atherosclerotic damage.24, 31, 32 In the same lines, nocturnal BP contributes to the progression of vascular damage even below the threshold of nocturnal hypertension (that is, <125/75 mm Hg); by contrast, the nondipping pattern is less frequently associated with vascular damage. These findings from cross-sectional observational studies highlight the detrimental role of the magnitude of nighttime haemodynamics on vascular function, suggesting a lower threshold of nighttime BP that triggers vascular impairment. Nonetheless, the progression of vascular damage below that hypothetical threshold might be still deleterious with mechanisms independent of increased haemodynamic load, such as sympathetic overactivity, sodium-mediated endothelial dysfunction and increased systemic inflammatory load. Furthermore, in sleep-disordered breathing and sleep disruption, BP surges are intermittent and may not be accompanied by either nocturnal hypertension or nondipping status; indeed, periodical increase in negative intrathoracic pressure and paroxysmal arousals or frank awakenings could promote CV damage independently of the registered nocturnal crude BP levels.
Finally, brain microbleeds reflecting uncovered cerebrovascular disease have been independently associated with nocturnal hypertension, whereas nondipping failed to determine the same outcome.33 In addition, among hypertensive patients, those with a positive history of stroke demonstrated higher levels of systolic and diastolic BP than did those without a stroke history.34, 35 Nonetheless, white matter lesions and leukoaraiosis were associated with altered circadian BP pattern, suggesting that chronic ischaemic lesions are more likely associated with sustained 24-h BP elevation.36, 37
Nocturnal BP phenotypes and ‘hard CV outcomes’
General population and hypertensive cohorts
Among 7458 enrolled patients in prospective population studies, both nighttime BP and night-to-day ratio significantly predicted all mortality outcomes (such as total, CV and non-CV) in adjusted models (Tables 2a and b). However, the mortality hazard ratios for systolic and diastolic nighttime BP were always greater than those of night-to-day ratios. In the total population, although nighttime BP was a significant predictor for all CV events irrespective of the treatment status, the night-to-day ratio omitted its prognostic significance for all combined fatal and nonfatal outcomes.38 In another prospective study that recruited 3957 patients, a superior prognostic significance for all-cause mortality was attributed to nocturnal hypertension in comparison with the nondipping status defined according to either systolic BP or diastolic BP fall.6
A meta-analysis of 3468 hypertensive patients’ nighttime ambulatory BP significantly predicted all-cause CV and non-CV mortality, coronary heart disease and stroke, independently of confounders. In addition, nighttime ambulatory BP adds to the prognostic significance of daytime ambulatory BP for all of the end points. By contrast, although the night-to-day BP ratio predicted all end points, after additional adjustment for 24-h ambulatory BP, it limited its prediction on all-cause mortality.39 However, a further meta-analysis performed by the same authors analyzing data of 302 patients with hypertension and CV disease, showed that the higher the night-to-day ratio, the higher the mortality and the incidence of CV events.40
The IDACO (International Database on ambulatory BP in relation to Cardiovascular Outcomes)41 highlighted the observation that isolated nocturnal hypertension was associated with a significantly higher risk of total mortality and all of the remaining ‘hard CV outcomes’ in >8700 patients from 10 different populations. In the same study, a 1 s.d. increase in either nighttime systolic or nighttime diastolic BP was accompanied by almost 20% higher CV risk. Although the recent real-world meta-analysis is not reported in Tables 2a and b—because of the implemented exclusion criteria (Figure 1)—reflects the importance of crude nighttime haemodynamics on hard CV outcomes, in line with our findings.
In a population of 872 elderly men (aged >70 years), nighttime systolic BP constituted a significant predictor of CV morbidity in both treated and untreated subjects (Tables 2a and b). On the contrary, the night-to-day ratio of systolic and diastolic BP was not associated with CV morbidity in this cohort, irrespective of the cutoff value used for the definition of nondipping.42 These results diverge greatly from the findings of the Systolic Hypertension in Europe (Syst-Eur) study43 and the PIUMA (Progetto Ipertensione Umbria Monitoraggio Ambulatoriale) study,44 in which the nondipping BP pattern was a predictor of CV morbidity.
In a further prospective study of elderly men free of heart failure, valvular disease and LV hypertrophy showed that the presence of the nondipping pattern was better associated with increased risk of heart failure than with nighttime ambulatory BP.45 Interestingly, when office systolic BP was accounted for as a confounder, nondipping status qualified as the predictor of heart failure, whereas ambulatory nighttime BP lost its significance.
In a longitudinal study, elevated nighttime systolic BP predicted increased risk of CV disease in both diabetic and nondiabetic patients, whereas the nondipping pattern—defined as the conjoint subset of risers plus true nondippers—was not associated with increased risk of CV disease in both groups (Tables 2a and b).46 Similarly, a prospective study of 300 untreated patients with uncomplicated type II diabetes mellitus showed that nondipping and rising patterns were not associated with incident CV disease.47 Finally, a longitudinal study of 392 subjects with type II diabetes mellitus demonstrated that the mean nighttime systolic BP predicted nonfatal vascular events more strongly than nocturnal fall in systolic BP.48
Chronotherapy and other interventions for nighttime BP
Although there is mounting evidence supporting the association between nocturnal hypertension and the nondipping profile with increased mortality and CV morbidity, it is yet unknown whether restoring the normal circadian rhythm or reversing nocturnal hypertension would ameliorate the prognosis. However, two options for circadian haemodynamic pattern restoration are available so far: first, implementing chronotherapy issues and second, targeting the pathophysiological mechanisms underlying the altered circadian pattern, including nighttime hypertension.
Several studies support a potentially favourable effect of afternoon doses of antihypertensive drugs on 24-h BP control, highlighting the purposeful timing of medication in relation to the normal circadian BP variation and drug pharmacokinetics. A relatively recent prospective parallel group trial on 250 resistant hypertensive patients evaluated the impact of the modification of time of treatment on BP.49 Patients assigned to receive one of their antihypertensive medicines at bedtime showed a further ambulatory BP reduction than did those receiving all drugs in the morning, whereas dippers increased from 16 to 57% in the former group. In another parallel study,50 on 200 nondipper hypertensive patients, 3 months of monotherapy with an angiotensin receptor blocker in the evening led to a significantly greater mean decrease in nocturnal BP, whereas 59% of patients in this treatment group managed to control their nighttime BP compared with only 33% of patients receiving the drug in the morning.
Till date, two large prospective randomized control trials have implemented nighttime administration of drugs, namely the HOPE (Heart Outcomes Prevention Evaluation) and the CONVINCE (Controlled Onset Verapamil Investigation of Cardiovascular End Points) trial. In the HOPE trial,51 ramipril was administered once daily at bedtime and resulted in an outstanding benefit on CV morbidity and mortality in high-vascular-risk subjects. In a small substudy of the HOPE trial,52 in patients with peripheral arterial disease, 24-h systolic and diastolic BP levels were significantly reduced by 10 and 4 mm Hg, respectively, mainly because of a more pronounced systolic and diastolic BP-lowering effect during nighttime (that is, 17 and 8 mm Hg, respectively). The night-to-day ratio was also significantly decreased in the ramipril group. Ambulatory BP showed greater falls, especially at night, than did office BP during treatment with ramipril administered once daily at bedtime. By contrast, in the CONVINCE trial,53 nighttime administration of verapamil was not equivalent to the morning doses of atenolol or hydrochlorothiazide in preventing CV disease-related events, and consequently, the result of this trial do not support the concept of chronotherapy in hypertension.
Targeting the responsible pathophysiological mechanisms for limiting the perpetuation of altered BP circadian pattern might be a promising auxiliary antihypertensive strategy. Indeed, dietary sodium restriction and potassium supplementation in tandem with anti-volume-expansion treatments with diuretics proved efficacious for restoring the dipping BP pattern in salt-sensitive hypertensive patients.54 In the context of obstructive sleep apnoea syndrome, continuous positive airway pressure (CPAP) therapy itself exerts a favourable effect on the disturbed nighttime BP variation.16 Nonetheless, correct timing of medication could further improve BP control in patients with sleep apnoea syndrome, while it may especially be considered as an alternative therapeutic approach in cases of low compliance to CPAP.
The evidence so far suggests that nocturnal BP as compared with the nondipping profile is superior in the evaluation and prediction of CV risk. Although the assessment of circadian BP variability is better refined by the nondipping pattern over nocturnal BP alone, there are methodological and definition problems that carry out potential barriers to its predictive value. The common denominator of adverse nocturnal haemodynamics including altered BP variability is increased body adiposity in tandem with the rest of metabolic syndrome components and sleep-quality issues. From a clinical point of view, crude estimates of nighttime BP—independent from definition thresholds—adversely impact the CV system, thus increasing the risk for developing events. It might be that isolated nocturnal hypertension associated with the nondipping profile further heightens the risk beyond nighttime BP per se.
Chronotherapy and targeting the potential pathophysiological mechanisms responsible for elevated nocturnal BP are the two available therapeutic options to restore blunted circadian BP variability. Whether these interventions are translated into ameliorated CV prognosis is yet unknown.
About this article
Nighttime vs. daytime blood pressure as a predictor of changes in left ventricular mass in hypertensive subjects
Hypertension Research (2013)