Evaluation of HuoXueHuaYu therapy for nonalcoholic fatty liver disease: a systematic review and meta-analysis of randomized controlled trial

Background To evaluate the effectiveness and safety of HuoXueHuaYu (HXHY) therapy in treating nonalcoholic fatty liver disease (NAFLD) through a systematic review and meta-analysis. Methods We performed comprehensive searches on Embase, Pubmed, Cochrane Library, CNKI, VIP and Wanfang databases up to June 2017 for randomized controlled trials using HXHY in the treatment of NAFLD compared with conventional treatment. Results This meta-analysis included 13 studies involving 1429 patients which 775 patients belonged to HXHY group and 654 patients belonged to conventional treatment group. The results of meta-analysis showed that HXHY can significantly improve B ultrasonic level (OR = 2.33; 95% CI:1.60, 3.40; P < 0.00001) of NAFLD compared with conventional treatment. As to lipids, HXHY was tested to be better on reduction of total cholesterol (TC) (MD = -0.38, 95% CI: − 0.48, − 0.29; P < 0.00001) and triglyceride (TG) (MD = -0.31; 95% CI: − 0.37, − 0.24; P < 0.00001) than conventional treatment. HXHY also had a greater beneficial effect on liver function in reducing alanine transaminase (ALT) (MD = -1.69; 95% CI: − 2.24, − 1.14; P < 0.00001) and aspartate transaminase (AST) (MD = -22.53; 95% CI: − 33.16, − 11.90; P < 0.00001) compared with conventional treatment. HXHY can also significantly improve the effective rate (OR = 3.55; 95% CI:2.65, 4.76; P < 0.00001) compared with conventional treatment. No serious adverse reactions were reported. Conclusions HXHY seems to be an effective and safe therapy for NAFLD. It is suggested that further study of HXHY in the treatment of NAFLD requires trials with rigorous design, multicenter, large-scale and high-quality worldwide.


Background
Nonalcoholic fatty liver disease (NAFLD) is a common chronic liver disease, with prevalence between 14 and 45% in the world [1,2]. One clinical study suggested that about 1/3 of NAFLD patients could develop into nonalcoholic steatohepatitis (NASH), and once they developed into NASH, the risks of liver cirrhosis, liver cancer, and liver failure might increase significantly [3].
The main therapies of NAFLD in conventional treatment are lifestyle intervention and drug therapy. Lifestyle intervention is hard to be applied due to lack of compliance. Therefore, drug therapy plays an important role in treating patients with NAFLD. Vitamin E and pioglitazone showed positive effects on liver function and lipid deposition. However, in spite of some beneficial effects, vitamin E does not have therapeutic effect on liver fibrosis and pioglitazone causes weight gain [4]. Other drugs such as metformin, orlistat and statins were of limited benefit [5]. Therefore, development of an effective therapy is of significant importance for NAFLD.
In Traditional Chinese medicine (TCM), chronic liver diseases are usually considered to be accompanied by blood stasis [6,7]. Promoting blood circulation (Chinese name in pinyin "Huo Xue HuaYu" (HXHY)) is an important therapy in the treatment of NAFLD [8]. A previous study compared different TCM therapies for NAFLD and indicated that HXHY therapy was superior to other therapies in treating patients with NAFLD [9]. More and more traditional Chinese herbs with the function of activating blood circulation have been proved to be effective in treating NAFLD [10]. Though there are several clinical trials suggested that HXHY therapy has therapeutic potential in treating NAFLD, the effectiveness of HXHY has not been assessed in system. Therefore, the present meta-analysis aimed to evaluate the effectiveness and safety of HXHY in treating NAFLD by a systematic review and meta-analysis of randomized controlled trial (RCTs) to provide evidence for clinical practice.

Search strategy
The study was performed following the PRISMA guidelines [11]. The literature search was conducted using Cochrane Library (1993 to June 2017), the PubMed database (2000 to June 2017), the Embase database (1974 to June 2017), the China National Knowledge Infrastructure database (1979 to June 2017), the Wanfang database (1982 to June 2017), the VIP database (1989 to June 2017). Search terms were (NAFLD OR nonalcoholic fatty liver disease OR fatty liver disease) AND (HuoXueHuaYu OR activating blood circulation OR Chinese medicine OR herbs OR herbal medicine).
Exclusion criteria were as following: (a) duplicated or redundant study; (b) nonhuman studies; (c) nonrandomized controlled trials.

Outcome indicators
The primary outcome was the level of type-B ultrasonic of liver, and the secondary outcomes were levels of total cholesterol (TC), triglyceride (TG), alanine transaminase (ALT) and aspartate transaminase (AST) and the effective rate.

Data extraction and analysis
Data Extraction of the included studies were performed by two researchers independently. They discussed and recorded any disagreement. The third researcher resolved the disagreement that could not be resolved through discussion. Cochrane Risk of Bias Tool was used to evaluate the quality of RCTs. Mean difference (MD) was reported for TC, TG, AST and ALT. Odds ratio (OR) was reported for B ultrasonic level and effective rate using Review Manager software (RevMan 5.3). 95% confidence interval (CI) will be used as an effective size for the combined analysis. I 2 statistics is used to estimate heterogeneities. If there is no heterogeneity (I 2 < 50% and P > 0.1), a fixed-effect model is used to synthesize the data; Otherwise, if there is heterogeneity (50% < I 2 < 75%), a random-effect model was applied. When I 2 > 75%, subgroup analysis or sensitivity analysis was performed to identify the causes of the heterogeneity. A funnel plot was selected to assess the publication bias.

Study selection and quality evaluation
Firstly, we searched out 381 studies completely and then keep 280 studies after deleting repeated records. Eliminating case reports, reviews and animal researches, we achieved 13 studies [12][13][14][15][16][17][18][19][20][21][22][23][24] (Fig. 1). The 13 studies included 1429 cases in total. Of which, 775 belonged to the HXHY therapy group and 654 belonged to the conventional treatment group. The patients included in each study were all classified as NASH. Table 1 listed the characteristics of the studies. Table 2 listed the compositions of the herbal formulae. Figure 2 showed the quality evaluation of the included studies. In terms of random sequence generation, only 1 study used a table of random number [12], while 12 studies mentioned "random", but there was no detail of the randomization method. In the aspect of blinding, no studies mentioned blinding of the patients and personnel. In addition, no study reported allocation concealment. Selective reporting, incomplete outcome data and blinding of outcome assessment were evaluated as low risk of bias.

The effect of HuoXueHuaYu therapy on B ultrasonic level in patients with NAFLD
Seven studies reported type-B ultrasonic. These studies involved 590 patients including 327 patients in HXHY therapy group and 263 patients in the control group. We found no significant heterogeneity in these studies ( I 2 = 0%, P= 0.85). A fixed effects model analysis showed that HXHY was more beneficial to change type-B ultrasonic level in NAFLD Patients when compared to the conventional treatment group (OR = 2.33; 95% CI: 1.60, 3.40; P < 0.0001) (Fig. 3).

The effect of HuoXueHuaYu therapy on blood lipids in patients with NAFLD
Five studies reported total cholesterol (TC). These studies including 358 patients which 202 patients belonged to the HXHY therapy group and 156 patients belonged to the conventional treatment group. As the results showed, the heterogeneity of TC was high (I 2 = 94%, P < 0.00001) among trials when comparing HXHY therapy with conventional treatment group. The random-effect model analysis showed that patients with NAFLD who received HXHY had significantly lower TC levels than those who received conventional therapy. (MD = − 0.38; 95% CI: − 0.48, − 0.29; P < 0.00001) (Fig. 4a). Sensitivity analysis results suggested that the study carried out by Guangjun Tian 2009 made a great contribution to the high heterogeneity. There was no heterogeneity existed when the study was removed (I 2 = 0%, P = 0.73) (Fig. 4b). Meanwhile, we found that the duration of the study was 24 weeks which was obviously longer than other studies, indicating that the duration maybe a source of heterogeneity. Six studies reported triglyceride (TG). These studies including 418 patients which 232 patients belong to the HXHY therapy group and 186 patients belong to the conventional treatment group. We found high significant heterogeneity in TG (I 2 = 95%, P < 0.00001) among trials when comparing HXHY therapy with conventional treatment group. A random-effect model analysis showed that HXHY therapy significantly decrease the level of TG than conventional treatment (MD = − 0.31; 95% CI − 0.37, − 0.24; P < 0.00001) (Fig. 5a). Sensitivity analysis results suggested that the study carried out by Guangjun Tian. 2009 made a great contribution to the high heterogeneity. The heterogeneity was much smaller when this study was removed. (I 2 = 49%, P = 0.10) (Fig. 5b).

The effect of HuoxueHuayu therapy on liver function in patients with NAFLD
Six studies reported alanine transaminase (ALT). These studies including 418 patients which 232 patients belong Fig. 4 a The effect of HXHY on blood lipid total cholesterol in NAFLD patients. b Sensitivity analysis was performed by omitting one study to the HXHY therapy group and 186 patients belong to the conventional treatment group. As the results showed, we found high significant heterogeneity in ALT (I 2 = 82%, P < 0.0001) among trials when comparing HXHY therapy with conventional treatment group. A random-effect model analysis showed that HXHY therapy significantly reduce the level of ALT than conventional treatment in the NAFLD patients (MD = − 1.69; 95% CI: − 2.24, − 1.14; P < 0.00001) (Fig. 6). Although we conducted sensitivity analysis and subgroup analysis, there was still a high heterogeneity. Five studies reported aspartate transaminase (AST). We found high significant heterogeneity in AST (I 2 = 97%, P < 0.00001) among trials when comparing HXHY with conventional treatment group. A random-effect model analysis showed that HXHY significantly reduce the level of AST than conventional treatment in the NAFLD patients (MD = − 22.53; 95% CI:− 33.16, − 11.90; P < 0.0001) (Fig. 7). Although we conducted sensitivity analysis and subgroup analysis, there was still a high heterogeneity.
The effect of HuoXueHuaYu therapy on the effective rate in patients with NAFLD Twelve studies reported effective rate of Huoxue-Huayu therapy in patients with NAFLD. These studies including 1369 patients which 745 patients belong to the therapy group and 624 patients belong to the conventional treatment group. The effective rate in seven studies [13,18,[20][21][22][23] refers to the proportion of participants with improvement of clinical symptoms and level of type-B ultrasonic of liver. The effective rate in the other five studies [12,14,17,19,24]   refers to the proportion of participants with improvement of clinical symptoms and level of type-B ultrasonic of liver as well as ≥30% reduction in level of liver function and blood lipids. There was no significant heterogeneity with I 2 = 0%, P = 0.99. A fixed effects model analysis showed that HXHY was more beneficial for the effective rate in NAFLD Patients when compared with the conventional treatment group (OR = 3.55; 95% CI:2.65, 4.76; P < 0.00001) (Fig. 8). Funnel plot was selected to assess the publication bias for the effective rate, which showed that the distribution is generally almost symmetrical (Fig. 9).

The safety evaluation of HuoxueHuayu therapy on NAFLD patients
There were no adverse reactions reported in all the 13 articles. Therefore, we may need to assess the safety of HXHY therapy on NAFLD in facilitate further researches.

Discussion
Based on the meta-analysis of 13 RCTs, it can be documented that HXHY can significantly improve B ultrasound in NAFLD patients when compared with the conventional treatment group. Meanwhile, HXHY therapy also can improve the blood lipid, liver function and the effective rate. Furthermore, there was no obvious adverse reaction reported in treating NAFLD. Our results suggested that HXHY is effective and safe in treating NAFLD.
The mechanism of NAFLD is not fully understood. Recently, researches show that insulin resistance, free radicals and oxidative stress, endoplasmic reticulum stress, and inflammation may involve in the mechanism of NAFLD [25][26][27][28]. The general philosophical underlining of Chinese medicine is holistic medicine [29,30]. TCM has anti-inflammatory effect and high safety in treating chronic liver diseases [31,32]. Previous studies suggested that HXHY therapy can protect hepatic cells, improve liver function and control the development of hepatic fibrosis. For example, salvianolic acid B extracted from Radix Salvia miltiorrhiza were demonstrated to attenuate liver damage, hepatic steatosis, and reduce the levels of pro-inflammatory cytokines [33]. Hawthorn leaf flavonoids significantly lowered liver/body weight ratio, improved serum parameters and liver dysfunction and alleviated hepatic lipid accumulation [34].  Our study had some limitations. First, the quality of the included trials was generally not high. None of the studies provided the methods of blinding and allocation concealment [35]. Clinical trials should be reported in accordance with the Consolidated Standards of Reporting Trials (CONSORT) standards [36,37]. Second, the treatment duration of most studies was short. Because NAFLD is chronic disease, longer duration should be taken to assess the safety and effectiveness of HXHY in the treatment of NAFLD. The sensitivity analysis suggested that treatment duration may be the main source of heterogeneity. Third, the sample size of some of the included studies is small. It is necessary to demonstrate whether the effects of HXHY will not be changed in future large-scale trials. Fourth, a wide variety of drugs were used in the control group across the studies, which may be another important source of the heterogeneity. Despite the limitations of this study, to the best of our knowledge, this is the first study to evaluate HXHY therapy for NAFLD.

Conclusions
In conclusion, this study indicates that HXHY therapy is more effective compared with conventional treatment for patients with NAFLD, suggesting that HXHY may be a new option for treating NAFLD. Due to the pool quality of the included studies, it is necessary to validate the conclusions by more rigorously designed, multi-centered RCTs with high quality.