Chinese Herbal Medicine Combined with Conventional Therapy for Blood Pressure Variability in Hypertension Patients: A Systematic Review of Randomized Controlled Trials

Objective. The aim of this systematic review is to evaluate effect of Chinese medicine combined with conventional therapy on blood pressure variability (BPV) in hypertension patients. Methods. All randomized clinical trials (RCTs) comparing Chinese medicine with no intervention or placebo on the basis of conventional therapy were included. Data extraction, analyses, and quality assessment were performed according to the Cochrane standards. Results. We included 13 RCTs and assessed risk of bias for all the trials. Chinese medicine has a significant effect in lowering blood pressure (BP), reducing BPV in the form of standard deviation (SD) or coefficient of variability (CV), improving nighttime BP decreased rate, and reversing abnormal rhythm of BP. Conclusions. Chinese medicine was safe and showed beneficial effects on BPV in hypertension patients. However, more rigorous trials with high quality are warranted to give high level of evidence before recommending Chinese medicine as an alternative or complementary medicine to improve BPV in hypertension patients.


Background
It is estimated that there are nearly one billion people suffering from hypertension worldwide, and the number of patients will increase to 1.5 billion by 2050 [1]. Blood pressure variability (BPV) means the degree of blood pressure (BP) fluctuations in a certain period of time. BPV is regarded as a separate index which is different from the BP reflecting cardiovascular activities. Many studies have confirmed that BPV could impact on hypertensive target organ damage and overall prognosis of patients [1][2][3][4][5]. An important factor to improve the prognosis of hypertensive patients is to reduce the BPV effectively. Clinicians pay close attention to lowering pressure steadily nowadays. There have been many studies that elucidated the relationship between western medicine and BPV. Although the results are inconclusive, most studies have shown that calcium channel blockers (CCB) are the most effective to reduce the BPV, especially amlodipine. Amlodipine in combination with other drugs may be more effective [6][7][8].
Short-term BPV within the 24-hour period is easier to obtain measured results. Ambulatory blood pressure monitoring (ABPM) is a way to assess the short-term BPV at present. By ABPM, a large amount of cross-sectional studies confirmed the increased BPV indicating the aggravated target organ damage [9]. A quantitative analysis of 155 Chinese people's ABPM results showed that BPV of hypertensive patients was higher than that of healthy people [10].
Western medicine puts emphasis on quickly and effectively lowering BP and helps patients reach target BP as soon as possible. But even effective long-term control of BP by western medicine may not fully achieve goals of 2 Evidence-Based Complementary and Alternative Medicine protecting target organs against damage, because reversing the target organ damage is a long process. Chinese medicine highlights the overall concept, self-regulation mechanism, and multitargets action and often plays an important role in the protection of target organ of hypertensive patients [11]. However, whether Chinese medicine is effective on BPV remains unclear. So, we design this systematic review to evaluate effect of Chinese medicine on BPV by ABPM in hypertension patients.

Search Strategy.
Two reviewers (Z. Chen and L. Q. Wang) searched the following databases from their inception to April 15, 2014, for the identification of randomized clinical trials (RCTs) assessing Chinese herbal medicine for BPV: The Cochrane Library, Pubmed, China National Knowledge Infrastructure (CNKI), Chinese Scientific Journal Database (VIP), Wanfang Databases, and Sino-Med Database. There was no limitation in languages. We also searched references of included reviews, just in case eligible trials were missed. We searched using the following searching terms: "blood pressure variability, " "BPV, " "essential hypertension, " "standard deviation, " "coefficient of variability, " "traditional Chinese medicine, " "Chinese herbal medicine, " and "randomized controlled trial. " Chinese pinyin of these terms were "xue ya bian yi, " "xue ya bo dong, " "xue ya bian yi xing, " "xue ya bo dong xing, " "zhong yao, " "zhong chengyao, " "zhong cao yao, " "zhong xi yi, " "ke li, " "zhong yi yao, " "zhong yi, " and "sui ji. " Based on different characteristics of literature databases, we adapted the search strategies appropriately which are shown in Table 1.

Study Design.
RCTs were included regardless of blind method and language.

Participants.
Patients with BPV regardless of age and race were included, who were diagnosed by one of the following diagnostic criteria: WHO/ISH Hypertension Prevention Guide (1999 [12]/2003 [13]) or Chinese Hypertension Prevention Guide (2005 [14]/2010 [15]). Patients with severe liver and kidney dysfunction and serious complications were excluded.

Interventions.
Chinese herbal medicine was used as intervention in at least one group of the study, form of which can be decoction, granule, and Chinese patent medicine (capsule/tablet). The method of application was restricted to orally taken; therefore, injections were excluded. Control groups were no treatment, placebo, or conventional medicines. Trials of Chinese herbal medicine in combination of conventional medicine compared to conventional medicine alone were also included, if the conventional medicine applied in both groups was the same. Trials with a treatment course of less than four weeks were excluded. Trials with more than one kind of Chinese herbal medicine as intervention were also excluded.

Outcome Measures.
Primary outcome measures were BPV (measured by standard deviation (SD) and coefficient of variability (CV), at three different time points: 24 hours, day, and night), BP (24 hours, day, and night), and symptom improvement rate which had uniform criteria, that is, clinical guideline of new drugs for traditional Chinese medicine. Among all these indicators, BPV and ambulatory BP are both indispensable. Secondary outcome measures were adverse events, quality of life (QOL), the abnormal rhythm of BP reverse rate, and nighttime BP decrease rate.

Data Extraction and Quality Assessment.
Two authors (Z. Chen and L. Q. Wang) performed data extraction ( Figure 1) independently according to a predesigned form. Disagreements during cross-checking of the data extraction form were resolved by consensus or consultation from a third author (J. P. Liu). We assessed the methodological quality of these trials by risk of bias tool which was recommended by Cochrane Handbook [16]. Specific items of the risk of bias tool were as follows: selection bias (random sequence generation, allocation concealment), performance bias (blinding of participants and personnel), detection bias (blinding of outcome assessment), attrition bias (incomplete outcome data), reporting bias (selective outcome reporting), and other biases. Each item for all included trials was judged into "high risk, " "unclear, " and "low risk. " 2.4. Data Analysis and Synthesis. Revman 5.2 software provided by the Cochrane Collaboration was used for data analyses. We expressed dichotomous data as risk ratio (RR) and its 95% confidence intervals (CI) and continuous outcome as mean difference (MD) and its 95% CI. Since all included trials applied different interventions, we did not pool the data due to the clinical heterogeneity. We performed qualitative description of the data synthesis.
We also failed to conduct a funnel plot to explore publication bias, because the number of included studies was less than nine for each outcome.  [17][18][19][20][27][28][29] were academic dissertations, five [21][22][23][24][25] were journal articles, and one [26] was conference paper. All trials were carried Evidence-Based Complementary and Alternative Medicine 3  (4) CNKI (SU = "blood pressure variability" OR SU = "fluctuation of blood pressure" OR SU = "BPV") AND (FT = "traditional Chinese medicine" OR FT = "Chinese patent medicine" OR FT = "integrated Chinese and western medicine" OR FT = "granule" OR FT = "Chinese herbal medicine" OR FT = "Chinese medicine") AND (FT = "randomized") (5) VIP (Title or key word = blood pressure variability + fluctuation of blood pressure + BPV) and (Any Field = traditional Chinese medicine + integrated Chinese and western medicine + granule + Chinese patent medicine + Chinese herbal medicine + Chinese medicine) and (Any Field = randomized) and (Professional = pharmaceutical and health care) and (Range = all journals) (6) WanFang database (title or key word: (blood pressure variability) + title or key word: (fluctuation of blood pressure) + title or key word: (BPV)) * (Chinese medicine + Chinese patent medicine + integrated Chinese and western medicine + granule + Chinese herbal medicine + traditional Chinese medicine) * randomized out in mainland China. One article [21] was funded by Shandong province science and technology development project. Additional records identified through other sources (n = 0)

Description of Included
Records after duplicates removed   [17,23,24,26,29] did not report male/female ratio in different groups, three trials [24,26,28] did not report distribution of age, and five [21,23,24,26,29] papers did not report disease duration between the groups. The remaining trials clearly reported these items. High BP classification was limited to first and second degree, except two trials (one [24] did not report high BP classification and one [25] was not limited classification). One trial [17] explicitly mentioned complications, but there was no significant difference in the distribution and two groups can be comparable. Remaining trials did not report complications.

Comparisons.
There were two types of comparisons: (1) Chinese medicine combined with conventional therapy versus conventional therapy alone (twelve trials) and (2) Chinese medicine combined with conventional therapy versus placebo and conventional therapy (one trial). Conventional therapy of six trials [17, 20-22, 26, 29] was amlodipine besylate tablet 5 mg per day. One trial was levamlodipine [28]. One trial was nifedipine controlled released tablet [27]. Two trials [18,24] were CCB combined with ACEI. Two trials were ACEI [19,23]. One trial did not report the ingredients of Chinese medicine [25]. There were four dosage forms: decoction, capsule, granule, and tablet. (Herbal medicines and adverse effects in the included trials are presented in Table 4.)

Outcome Measures.
Nine trials [17,19,22,[24][25][26][27][28][29] reported adverse events. Two trials [20,23] clearly reported there was no adverse drug reaction. Two trials [23,29] reported QOL. No trial reported health-economic indicators or follow-up visit. All trials reported BPV in the form of SD or CV, symptom improvement rate (antihypertensive effect) of which diagnostic criterion was Clinical Research Guideline of New Drugs for Traditional Chinese Medicine. Antihypertensive effect is divided into three levels according to the BP value: markedly effective, effective, and invalid. [17,19,26,27,29] used random number table to generate the random sequence. Two trials [17,29] referred to opaque sealed envelopes. Blinding of participants and personnel was mentioned in only one trial [29] which was double-blinding. No trial blinded the outcome assessors. There was not sufficient information to judge whether outcome assessors were blinded or not. Eleven trials did not miss outcome data, among which two trials [17,26] clearly claimed no drop-out patient. All trials reported their prespecified primary outcomes except one trial. All trials declared baseline characteristics were comparable. One trial [21] was lack of the inclusion and exclusion criteria. Unfortunately, no trial reported sample size calculation. (Risk of bias summaries are presented in Figure 2.) Evidence-Based Complementary and Alternative Medicine 5

Effects of Interventions.
Since every trial had different Chinese herbal medicines as treatment, none of trials could be analyzed by meta-analysis, because of clinical heterogeneity. We presented the effects of interventions by qualitative description, according to the two types of comparisons: Chinese medicine combined with conventional therapy versus the same conventional therapy and Chinese medicine combined with conventional therapy versus placebo combined with the same conventional therapy. All BPV were measured by ambulatory BP meter. BPV mainly was expressed in two forms: SD and CV. (Effect estimate of outcomes is presented in Table 3 = 80). Xiang tian ma decoction combined with benazepril also reduced 24 h systolic SD (MD −5.20; 95% CI −6.37 to −4.03; = 78). Xuezhikang capsule was integrated with angiotensin converting enzyme inhibitors (ACEI) and calcium channel blockers (CCB) (MD −1.79; 95% CI −2.79 to −0.79; = 110). There were no significant differences in the other four trials; they were Tianma Shuxin granule combined with amlodipine besylate tablet, Qianyang Yuyin granule combined with felodipine and enalapril, Tiao ping kang tablet combined with nifedipine controlled released tablets, and Gouteng Siwu decoction combined with levamlodipine besylate tablet.      Not reported: the article did not report any information about adverse effects. Not observed: the article reported that no adverse effects were observed in the studies.
Day systolic SD was reported by nine trials. Five trials [17-23, 25, 27]  The other four trials did not find significant difference between groups, one trial about Qianyang Yuyin granule plus felodipine with enalapril and three trials compared amlodipine besylate tablets, that were Tianma Shuxin granule, Sangji Wendan decoction, and Songling Xuemaikang capsule.
(2) CV after Treatment. Only one trial (Tianma Shuxin granule combined with amlodipine besylate tablet) reported CV. There were statistically significant differences in two outcomes. There was a reduction in day systolic CV in experimental group superior to that in control group (MD −0.01; 95% CI −0.02 to −0.00; = 60). There was a reduction in 24 h diastolic CV in experimental group inferior to that in control group (MD 0.03; 95% CI 0.02 to 0.04; = 60). Nevertheless, there was a reduction in day systolic CV in control group inferior to that in experimental group (MD −0.01; 95% CI −0.02 to −0.00; = 60). There was no statistically significant advantage in reducing CV in experimental group in terms of 24 h systolic CV, day diastolic CV, night systolic CV and night diastolic CV.
(4) Antihypertensive Effect. Antihypertensive effect was measured taking clinical guideline of new drugs for TCM as standard. Only one trial [26] showed treatment group superior to control group in improving antihypertensive effect.
(5) QOL. One trial [23] referred to QOL but did not report which specific test scale it used.
(6) Nighttime BP Decreased Rate. Three trials [19,26,28] reported this outcome. Only one trial [26] presented that combination therapy is superior to conventional treatment in nighttime BP decreased rate of both systolic BP and diastolic BP.
(7) Frequency of Reversed Abnormal Rhythm of BP. Two trials [25,27] showed treatment group superior to control group in increasing frequency of reversed abnormal rhythm of BP.

Chinese Medicine Combined with Conventional Therapy versus Placebo
Combined with the Same Conventional Therapy. There was only one trial [29] under this category, that is, Qing xuan granule combined with amlodipine besylate tablet versus placebo combined with amlodipine besylate tablet.
(1) SD after Treatment. All the outcomes about SD of this trial were of significant statistical heterogeneity. Qing xuan granule combined with amlodipine besylate tablet had a better efficacy than the control in terms of lowering (2) CV after Treatment. No noteworthy statistical differences of this endpoint in any of the trials were noted.
(3) BP after Treatment. There was no significant difference between the experimental and the control groups regarding this outcome which contained 24 h systolic BP, 24 h diastolic BP, day systolic BP, day diastolic BP, night systolic BP, and night diastolic BP.
(4) Antihypertensive Effect. There was no significant difference between the experimental and the control groups regarding this outcome.
(5) QOL. There was also no significant difference between the experimental and the control groups regarding this outcome which was measured by SF-36 health related QOL scale.  3.4. Safety. There were 9 trials reporting 50 cases of adverse events. The experimental groups had 18 patients with side reactions, and the control groups had 29 patients with side reactions. Three patients had adverse events but with no specific group information. No significant difference about adverse events was found between two groups. The most commonly reported adverse events in the 9 trials were intestinal disturbance (abdominal distension, nausea, and constipation) [17,22,25,28], ankle edema [17,27], dizziness [22,27,29], palpitation [22], facial flushing [26], and dry cough [19,24].

Discussion
There are thirteen RCTs including 1103 participants that were included in this systematic review. From this review, we may be able to speculate that both the TCM and combination therapy have a significant effect in lowering BP, lowering BPV, and adjusting the circadian rhythm. As we known, efficacy of amlodipine besylate tablets in lowering BP variation is better than other western medicines. Most of trials in this paper selected amlodipine besylate tablets as conventional therapy in the compare group of Chinese medicine combined with conventional therapy versus the same conventional therapy. One trial took placebo combined with conventional therapy as control group. According to TCM theories, the treatment is based on syndrome differentiation, and even the same disease could have a variety of syndromes; therefore, the treatments could be modified with different Chinese herbal medicines. Although the differences in TCM prescription, forms, course of treatment, and control drugs prevent us from performing meta-analysis which could provide precise effect estimate of intervention, from Table 4 we gladly found that there were two kinds of TCM (Ramulus Uncariae Rhynchophyllae cum Uncis and Radix Achyranthis Bidentatae) that appeared most. The most TCM syndromes were yin deficiency and yang excess. This is also consistent with the theory of TCM. It was confirmed that the integrative medicine based on syndrome differentiation, a person-centered and balanced medicine [30], was safer and more effective than western medicine alone in the treatment of hypertension [31]. Adverse reaction related to Chinese medicine to reduce BPV is relatively rare. Only a few patients had ankle edema, dizziness, and nausea during the treatment.
It is too early to recommend this conclusion to clinical practitioners considering the limitations of this review. Firstly, the methodological quality of included trials in the review generally needs to be improved. Studies have shown that the degree of rigorous design and quality of methodology of the study have a direct impact on the effectiveness of intervention [32,33]. The randomization was unclear in most of trials. Only five RCTs described random allocation in detail, and two RCTs mentioned random hidden method. Blinding is an effective way to control measurement bias; however, only one trial mentioned double-blind; no trial reported estimation of sample size. Moreover, this review focuses on short-term BPV and all of the included trials did not have follow-up research, so we cannot evaluate longterm effect of Chinese herbal medicine for BPV. In addition, all the trials did not publish their protocols, so we can only judge their reporting bias by a compromise method, which is comparing whether the outcomes mentioned in method and results are consistent. Two trials reported QOL, but only one trial evaluated QOL by certain scale. Fifth, due to the clinical features of hypertension, hypertensive patients commonly also suffered from other diseases, such as coronary heart disease or hyperlipidemia, but only a small number of clinical trials on the baseline data reported these accompanied diseases. Most of the other tests did not report this information, but we cannot rule out the possibility of patients with accompanied disease. Sixth, we only included Evidence-Based Complementary and Alternative Medicine 15 13 trials with a relative small sample size in this review, and we failed to perform meta-analysis. Therefore, further rigorously designed RCTs are needed before recommending TCM to patients with hypertension. Moreover, we suggest that the design and reporting of RCTs on TCM strictly comply with both CONSORT statement [34] and that for herbal interventions [35].

Conclusion
TCM showed potentially short-term beneficial effects on BPV. However, because of small sample size and potential bias of most trials, this result should be interpreted with caution. More high quality trials, safety evidence, and longterm effects are warranted before TCM is recommended as an alternative or complementary medicine for BPV in hypertension patients. Information of QOL and economic effectiveness should also be paid more attention in future clinical trials.