Background
The circadian rhythm plays a critical role in multiple neurohormonal processes, thus modulating the cardiovascular system [
1,
2]. The circadian changes in blood pressure have received increasing interest. Additionally, with the introduction of ambulatory blood pressure monitoring (ABPM) into regular hypertension management, day-night patterns of blood pressure can be observed [
3]. Early morning is the time of the highest incidence of cardiovascular events during the day [
4]. Accumulating evidence demonstrated that an exaggerated morning blood pressure surge (MBPS) is closely related to the increased risk of cardiovascular diseases and all-cause mortality [
5‐
8]. Besides, systematic review also indicated that when using a continuous variable to test correlations, a 10 mm Hg increase in MBPS was related to high risk of stroke [
9]. Thus, the control of MBPS is clinically relevant and important. In addition, night blood pressure dipping, an index of blood pressure decline, has also been reported as an important prognostic marker for cardiovascular events and morbidity in hypertensives [
10,
11].
Chronotherapy is the adaption of medication to biological rhythm, to achieve maximal effectiveness by altering the time of drug administration [
12], which could influence the pharmacokinetic properties of antihypertensive medications. Multiple works have evaluated the administration-time effects in hypertension therapy. Hermida et al. revealed that administration of at least one prescribed antihypertensive medications in hypertensives at bedtime, compared to upon wakening, significantly improved blood pressure control (especially bedtime blood pressure) and remarkably diminished the occurrence of major cardiovascular events, such as stroke, myocardial infarction, death, etc. [
13]. A Cochrane systemic review showed that bedtime ingestion of antihypertensive drugs was more effective to decrease 24-h blood pressure without additional adverse effects than morning regimen [
14].
Some studies documented that bedtime ingestion of blood pressure-lowering agents had a more efficient antihypertensive effect during night-time and at the early morning period [
15‐
17]. Most studies working on the administration time of antihypertensive medications were small-scaled and single-centered. And the bedtime administration has not been recommended in guidelines. Whether the bedtime administration of antihypertensive medications has a more significant effect on lowering MBPS than conventional morning administration has not been reported in any systematic review and meta-analysis. Therefore, this systematic review was conducted to investigate the efficacy of chronotherapy of once-daily antihypertensive drugs on reducing MBPS, systolic and diastolic blood pressure, and adverse effects in hypertensive patients, thus providing more evidence to the bedtime regimen.
Methods
The protocol for this systematic review and meta-analysis was registered in PROSPERO (CRD42020180166).
Search strategy
The searching databases included: Ovid MEDLINE, EMBASE, Cochrane (Wiley) CENTRAL Register of Controlled Trials and Cochrane Database of Systematic Reviews, the Chinese Biomedical literature database up to April 2020. Besides, we also tracked completed clinical trials that met our inclusion criteria on ClinicalTrials.gov. The major search terms included: “hypertension”, ‘‘morning surge”, “morning blood pressure surge”, “MBPS”, “night* decline”, “Chronotherapy”, “morning OR day OR am OR diurnal OR daytime OR awake”, “evening OR bedtime OR night OR nocturnal OR pm” (shown in Additional file
1: Table 1). In addition to databases, studies in references lists of relevant articles that met our criteria were also hand-searched and screened as a significant supplement. No language and date restrictions were applied to avoid missing any related investigations.
Table 1
Characteristics of included studies
Hermida et al. | 2009 | Spain | 238 | 8 | 53.3 ± 11.4 | 130 (55%) | Grade 1 or 2 essential hypertensiona | Nifedipine-GITS (CCB) | Tablet counts and interviews | ①②③④⑤⑥⑦⑨ |
Hoshino et al. | 2010 | Japan | 31 | 32 | 69 ± 11 | 19 (61%) | Essential hypertensionb | Olmesartan + Amlodipine (CCB + ARB) | Not mention | ①②③④⑤⑥⑦ |
Acelajado et al. | 2012 | USA | 38 | 8 | 51.7 ± 11.6 | 17 (46%) | Grade 1 or 2 essential hypertensiona | Nebivolol (β-blocker) | Professional instruction | ① |
Peng et al. | 2013 | China | 54 | 8 | 58.3 ± 10.7 | 26 (48%) | Grade 1 or 2 essential hypertensiona and 24 h mean ambulatory blood pressure more than more than 130 / 80 mm Hg | Telmisartan + Amlodipine (CCB + ARB) | Not mention | ②③④⑤⑥⑦ |
Zhang et al. | 2014 | China | 156 | 8 | 56.3 ± 6.1 | 92 (59%) | Essential hypertensionb | Amlodipine + Losartan (CCB + ARB) | Not mention | ①②③④⑤⑥⑦ |
Dion et al. | 2015 | Germany, Spain, France, Italy and the Netherlands | 639 | 12 | 61.6 ± 10.6 | 281 (44%) | Grade 1 or 2 essential hypertensiona and 24 h mean ambulatory BP (maBP) more than 130/80 mmHg | Valsartan (ARB) | Professional instruction | ①②③④⑤⑥⑦⑧⑨⑩ |
Lai et al. | 2015 | China | 120 | 2 | 60.6 ± 5.3 | 55 (46%) | Essential hypertensionb | Losartan (ARB) | Not mention | ① |
Qiao et al. | 2015 | China | 108 | 4 | 64.7 ± 8.3 | 62 (57%) | Essential hypertensionb | Candesartan (ARB) | Not mention | ①④⑤⑥⑦ |
Zhao et al. | 2015 | China | 244 | 48 | 74.5 ± 9.1 | 104 (43%) | Essential hypertensionb | Nifedipine-GITS (CCB) | Interviews every two weeks | ①②③④⑤⑥⑦⑧ |
Li et al. | 2016 | China | 96 | 12 | 65.1 ± 9.4 | 49 (51%) | Essential hypertensionb | Enalapril (ACEI) | Not mention | ①②③④⑤⑥⑦ |
Inclusion and exclusion criteria
Study selection was performed independently by two reviewers (Z.X. and J.Z.) by viewing the titles and the abstracts of the search strategies. The disagreements were judged by a third reviewer who was blinded to the first two reviewers’ decisions. The selection process was repeated twice by each reviewer. Studies meeting the inclusion criteria were checked in detail (full text).
Only randomized control trials (RCT) were included to assess the effects of chronotherapy. Randomized cross-over trials that only had two treatment periods (two interventions) were also considered. Adults (more than 18) with primary hypertension, which is defined as systolic and/or diastolic blood pressure levels more than 140/90 mmHg were included. Secondary hypertension, alternating shift workers, and severe cardiac insufficiency (NYHA III-IV) were the exclusion criteria. Studies reporting one-drug therapy or combined therapy (two or more drugs) with the antihypertensive drug(s) administered once daily at bedtime when the control group was the same drug(s) at the same dose(s) once a day upon awakening were included. Antihypertensive drugs comprised diuretics, adrenergic beta-antagonists (β-blockers), alpha-antagonists, calcium channel blockers (CCB),vasodilator agents and renin-angiotensin system inhibitors (RASI, including angiotensin II receptor blockers and angiotensin-converting enzyme inhibitors). Administration time in the evening was from 18:00 to 24:00, while in the morning was from 6:00 to 12:00.
Included studies must clearly define and measure MBPS. Both continuous and categorical variables were included. Definitions of MBPS must belong to any one of the following: (1) the sleep-trough surge, calculated by the mean value of morning blood pressure within 2 h after waking minus the lowest night-time blood pressure; (2) the prewaking surge, the mean blood pressure within 2 h after waking to subtract the mean blood pressure within 2 h before waking; (3) the rising blood pressure surge, morning blood pressure reading upon rising minus blood pressure reading in the lying position 30 min before rising. The measurement of night dipping is the percentage of reduction in mean nighttime blood pressure relative to mean daytime blood pressure.
Outcomes
The primary outcome was the change in MBPS from baseline to the end of treatment (or the value of MBPS at the end of treatment if the baseline is comparable) when it was calculated as a continuous variable; or the ratio of patients whose MBPS exceeded the settled threshold after treatment when MBPS was defined as a categorical variable.
The secondary outcomes contained night blood pressure dipping, 24-h mean systolic blood pressure (SBP) and diastolic blood pressure (DBP), daytime mean SBP and DBP (measured by ABPM from the time patients wake up in the morning to the time they fall asleep in the evening or from 6–8:00 to 22–24:00), night-time SBP and DBP (measured by ABMP from the time patients fall asleep in the evening to the time they wake up in the morning or from 22–24:00 to 6–8:00). Safety outcomes include overall adverse effects and withdrawals due to adverse effects during treatment. Adverse effects can be any unplanned and unfavorable symptom, or disease temporally associated with the use of medicine. Withdrawals due to adverse effects are reported as events leading to permanent trial discontinuation.
Data extraction and management
Data were extracted by two independent reviewers (Z.X. and J.Z.) based on a standard form. Necessary information of studies was extracted, such as patients’ demographics (age, sex, region, race, hypertensive status, medication history), study methods (patients’ recruitment, randomization, crossover, treatment duration), interventions (drugs, dose, administration time), outcomes (definition and measurement of MBPS, night blood pressure dipping, 24-h mean SBP/DBP, daytime mean SBP/DBP, night-time SBP/DBP) and safety endpoints (headache, nasopharyngitis, edema, bronchitis, pain and withdrawals due to adverse effects). The numeric data (e.g., blood pressure) were collected from text and tables. Data in graphs were not extracted due to a possible measurement error. All the studies were double coded by the two reviewers. If there were missing data, we attempted to contact the authors to provide the missing information. Regarding missing data for the standard deviation of the change in MBPS, imputation was conducted based on other similar trials.
Risk of bias assessment
Evaluation of the risk of bias in all included trials was executed by two independent reviewers (Z.X. and J.Z.) based on the guide of the Cochrane risk of bias tool. Characteristics of assessment included random sequence generation, allocation concealment, blinding, incomplete outcome data, selective reporting, and other bias (e.g., conflict of interest). The assessment was done at the study level.
Data analysis
All data analysis and synthesis in the meta-analysis was conducted by RevMan 5.4 (RevMan 5.4; Cochrane Collaboration, Oxford, UK). Comparisons of continuous MBPS changes, 24-h, daytime and night-time SBP and DBP between groups were presented as mean ± standard deviation (SD) with corresponding 95% confidence intervals. These data were entered using a generic inverse variance. Categorical MBPS, overall adverse effects, and withdrawals were presented as relative risk ratios (RR). The outcomes were combined using a fixed- effect model (I2 ≤ 25% ) and a random-effect model (I2 > 25%).
Subgroup analysis
Subgroup analysis were conducted by classifying the trials into those using different MBPS definitions, and those using antihypertensive drugs from different classes. Subgroup analysis can help locate the sources of heterogeneity and assess the effectiveness of different kinds of medicines.
Discussion
This systematic review and meta-analysis investigated the effects of evening and morning dosage of antihypertensive medications on MBPS and blood pressure control in patients with essential hypertension.
Morning administration of medication is commonly used in antihypertensive treatment in clinical practice. However, some recent clinical trials reported evening administration of the antihypertensive drug to be efficient in lowering blood pressure, improving organ functions, and preventing cardiovascular events [
27‐
30]. This study indicated that the evening regimen of antihypertensive drugs exerted better effects on decreasing night-time blood pressure than morning therapy. Since studies showed that night-time blood pressure was significantly better than daytime blood pressure in predicting all-cause, cardiovascular and non-cardiovascular mortality, stroke and cardiovascular events [
31,
32]. Besides, this analysis also demonstrated that bedtime administration significantly better enhanced night blood pressure dipping, relative to morning regimen, which was consistent with several previous studies. Increased nocturnal decline drives blood pressure towards more of a dipper pattern, thus improving blood pressure variability and lowering cardiovascular risk [
33,
34]. Therefore, chronotherapy targeting night-time blood pressure control may favorably affect cardiovascular morbidity and mortality in hypertensive patients.
This study found that evening administration of blood pressure-lowering medication significantly improved MBPS, which has not been systematically reviewed before. The increase of MBPS was first associated with the incidence of stroke in 2003 [
8]. A clinical trial with 519 hypertensive participants found that a 10 mm Hg raised in MBPS caused a 25% increased stroke incidence. Several but not all subsequent studies indicated that exaggerated MBPS had been related to increased risk for cardiovascular events and all-cause mortality. However, in recent years, the prognostic role of MBPS is controversial.
One possible reason is the distinct definitions and cut-off points of MBPS. There are four definitions commonly used (the sleep-trough surge, the prewaking surge, the rising blood pressure surge, and the morning-evening difference), and the thresholds are various. The thresholds of MBPS could be defined as top decile, quartiles of MBPS in participants, or determined according to hypertensive guidelines, which have not reached a consensus [
35]. Among included studies, two articles analyzed MBPS as categorical variable with different predetermined threshold, 23.58 mmHg and 35 mmHg, respectively. Due to the limited number and different thresholds of studies, the heterogeneity between studies of categorical MBPS is very high. And a previous meta-analysis found that there was no significant association between MBPS and all-cause mortality or cardiovascular events when MBPS was analyzed as categorical variable [
9]. Therefore, a single threshold dichotomizing MBPS as normal and exaggerated may not be powerful to define the significance of MBPS. Besides, in patients with non-dipper hypertension and nocturnal hypertension, which are described as a sleep-to-awake SBP ratio of less than 10%, and night-time SBP more than 120 mmHg and DBP more than 70 mmHg respectively [
36], the MBPS may be low and not suitable for the evaluation of cardiovascular prognosis. This was also supported by the Jackson Heart Study, that no significant association between MBPS and the incidence of cardiovascular events was found in the black population with non-dipper and nocturnal hypertension [
37]. Consequently, more research is needed to analyze MBPS’s definitions and reach a consensus on the target value of MBPS control. In patients with nocturnal dipping hypertension, a prognostic indicator needs to be developed to predict cardiovascular events and serve as a target of hypertension management.
On the other hand, ethnicity is a critical factor affecting morning blood pressure surge. The Ohasama Study [
7] and JMS-ABPM Study [
8] found significant relationship between exaggerated MBPS and stoke incidence in Japanese patients. Conversely, Bombelli et al. [
38] reported that high MBPS was not associated with increased mortality and cardiovascular events in a white population. Furthermore, systematic review showed that the degree of sleep-trough surge was higher in Japanese than in European patients with hypertension [
39]. The Jackson Heart Study also revealed that there was no clear evidence for the associations of sleep-trough MBPS, prewaking MBPS, and rising blood pressure surge with the incidence of cardiovascular events and all-cause mortality in black adults [
37]. Discrepancies in the pathogenesis and manifestations of hypertension and related cardiovascular outcomes have been consistently reported among different races. Asians have more active sympathetic nerve activity during the morning period and higher incidence of stroke, which may account for the high degree and the prognostic role of MBPS [
4]. Therefore, therapeutic strategies targeting MBPS control are of considerable significance in Asian population.
This meta-analysis discovered a prominent role of evening CCB in lowering blood pressure. The CCBs in included studies were nifedipine gastrointestinal therapeutic system (GITS) and amlodipine, which both belongs to the dihydropyridine calcium blocker family with long duration of action [
40]. Amlodipine has been shown to be more effective for the MBPS control compared with valsartan in the VALUE trial [
41]. Besides, previous evidence observed that the pharmacokinetic pharmacodynamics relationship of CCB is circadian rhythm-dependent that a night dosing of amlodipine displayed a longer half-life and higher peak plasma concentration compared to the morning dosage [
42,
43]. This may also explain the remarkable effects of CCB in lowering night-time and MBPS when given at bedtime.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.