Evidence-based review of oral traditional Chinese medicine compound recipe administration for treating weight drop-induced experimental traumatic brain injury

Background Recently, a number of studies conducted and published in China have suggested that traditional Chinese medicine compound recipe (TCMCR) may be beneficial in the treatment of experimental traumatic brain injury (TBI). In this study, we conducted a systematic review and meta-analysis of the efficacy of TCMCR in TBI model with weight drop method to provide robust evidence on the effects of TCMCR and to determine whether TCMCR can be recommended for routine treatment or considered as a standard treatment for TBI. Methods We identified eligible studies by searching five electronic databases on April 1, 2014, and pooled the data using the random-effects model. Results were reported in terms of standardized mean difference (SMD). We also calculated statistical heterogeneity, evaluated the studies’ methodological quality and investigated the presence of publication bias. Results Totally, 187 relevant publications were searched from databases, 25 of which met our inclusion criteria. The overall methodological quality of the most studies was poor, and there was evidence of statistical heterogeneity among studies along with small-study effects. Meta-analysis showed statistically significant effects indicating that TCMCR has a beneficial effect on TBI. Conclusions Despite the limitations, we concluded that TCMCR may reduce brain water content, improve BBB permeability, and decrease TNF-α/NO expression after experimental TBI in terms of overall efficacy. However, our review also indicates that more well-designed and well-reported animal studies are needed.


Background
Traumatic brain injury (TBI) remains the leading cause of long-term disability in individuals under 35 years worldwide [1], severely affects the quality of life of surviving patients and brings a significant social and economic burden. However, there is currently no effective pharmacological interventions options for TBI [2].
Because research aimed at therapy development has focused almost exclusively on single therapies, all of which have failed in multicenter clinical trials [3]. Fortunately, the focus of research has recently shifted to modify multiple targets, either through combination therapies or through the use of single agents that modulate multiple key secondary events following TBI [4,5].
For thousands of years, traditional Chinese medicine (TCM) has been widely practice in China, and holds a key role in maintaining the health of the Chinese population, merits far greater attention from researchers because of its diverse pharmacological functions and targets, which can provide improved treatment of complex diseases by its ability to aim at several targets simultaneously [5]. The traditional Chinese medicine compound recipe (TCMCR), the main form of TCM drug treatment, may represent an ideal source for developing safe and effective agents for TBI treatment because it contains ≥2 Chinese herbs, and therefore more closely conforms to TCM theories and more accurately reflects the characteristics of TCM than does the administration of a single herb [6]. Recently, a number of studies conducted and published in China have suggested that TCMCR may be beneficial in TBI treatment and rehabilitation [7][8][9][10][11]. However, it is uncertain whether robust evidence exists on the effects of TCMCR or whether TCMCR can be recommended either for routine treatment or considered as a standard treatment for TBI. Moreover, a systematic review and meta-analysis of the efficacy of TCMCR in treating TBI has not yet been investigated in experimental animal studies.
Therefore, the primary aim of the present study was to conduct a systematic review and meta-analysis to investigate in an unbias manner whether the evidence from experimental studies indicated a beneficial effect of TCMCR in the treatment of TBI in animal models.

Literature search
All studies reporting the efficacy of TCMCR in animal TBI models prior to April 1, 2014 were included. Articles were searched in the following databases: PubMed, ScienceDirect, CNKI, Wan-Fang Data, and Vip. The key search terms are summarized in Table 1 and were kept broad to capture all potentially relevant articles. In addition, the reference lists of all relevant articles were searched for further relevant publications.

Study selection
Three investigators assessed the records to assess for eligibility based on title, abstracts. The copies of all relevant articles were obtained and further assessed whether each met the prespecified inclusion criteria, the details of which are presented in Table 2. Disagreements among investigators were resolved by consensus after discussion.

Data extraction
Two investigators independently extracted details of each included studies including the animal species used, type of TBI model, treatment groups, time/dose of drug administration, anesthetic used, and the main outcomes. Information on sample sizes and substances used as experimental and control treatments was also extracted. Disagreements were resolved through consultation with a third party author.
When sufficient data were not available, authors were contacted and requested to provide missing data. The digital ruler software was used to estimate numerical values from the graphs, if no reply was received. And the study was excluded from the meta-analysis, when the required data were not obtainable.

Study quality
The methodological quality of each included studies was assessed based on a 10-point quality checklist modified from the CAMARADES study as previously described, with minor modifications [12,13], comprising (1) publication in a peer-reviewed journal; (2) random group allocation; (3) blinded induction of brain injury; (4) blinded assessment of outcome; (5) monitoring of physiological parameters including temperature; (6) sample size calculation; (7) compliance with animal welfare regulations; (8) avoidance of anesthetics with marked intrinsic neuroprotective properties (ketamine); (9) statement of potential conflicts of interest; (10) use of accurate/suitable/adequate animal models.
One point was given for written evidence of the quality criteria.

Statistical analysis
Data were processed as described previously [14]. Briefly, for the meta-analysis, results were calculated as standardized mean difference (SMD), and 95 % confidence intervals (CI) with random-effects model to avoid heterogeneity were used to assay differences of the global estimate effect [13]. The Cochran's Q-statistic was used to assess within-and between-study variation or heterogeneity [15,16]. Heterogeneity was quantified with the I 2 metric, with higher values denoting a greater degree of heterogeneity. I 2 values ≤ 50 % indicate acceptable heterogeneity among studies [17]. For studies comparing different doses and/or timing of drug administration with a single control group, we pooled data from all experimental groups for comparison with the control group. The possible publication bias was assessed using funnel plots and Egger's tests. [18]. All statistical analyses were performed using Stata software (version 12.0).

Study selection
On the basis of predefined standards, we identified 187 potentially relevant articles. After removing duplicate articles, 95 articles remained. Through screening titles and abstracts, 68 were excluded because they were not published in peer-reviewed journals, TCMCR was not administered orally, or traditional Chinese medicine injection/a single Chinese herb medicine was administered in the treatment group. After full-text evaluation of the remaining 26 articles, one article was excluded due to unobtainable data. Thus, 25 studies [7,8, were included for systematic review. Moreover, 14 studies [8, 19, 22-29, 32, 34, 35, 38] were ultimately included in the meta-analysis. Figure 1 shows a flow chart of the study selection.

Methodological quality of included studies
Overall, the median quality of the 25 included studies was poor (3, interquartile range, 3-4) with scores ranging from 2 to 6. No studies scored 0 or had a high quality rating (7-10 points). We found one study [8] with a quality score of 6, three [30,34,40] with a score of 5, and four [21,29,37,39] with a score of 4. Animals were allocated treatment by randomization in all included articles except one [27]. Only two studies [8,40] that were included failed to report the monitoring of physiological parameters (although the majority of these only monitored body temperature). All of the studies failed to report potential conflicts of interest, blinded outcome assessment, and blinded induction of TBI.

Meta-analysis Brain water content
In 12 studies [8, 22-29, 32, 38], there were 25 comparisons (involving 432 animals) of brain water content after TBI, which was determined by the wet and dry weight method [42]. Pooled analysis indicated that animals in the treatment groups had significantly reduced brain water content compared to animals in the control groups (SMD = -1.421, 95 % CI: -1.704 to -1.1379, P < 0.0001).

NO levels
In two studies [19,34], there were six comparisons (involving 108 animals) of NO after TBI. The pooled analysis indicated that there was a significant difference in NO levels between the treatment and control groups (SMD = −1.550; 95 % CI, −1.987 to −1.112; P < 0.0001).

Possible drug protection mechanism analysis
All of the studies selected during initial screening assessed the biological mechanisms of TCMCR activity. A wide variety of possible neuroprotective mechanisms were proposed in these studies. The neuroprotective effect of TCMCR was attributed primarily to preservation of blood-brain barrier integrity, amelioration of cerebral edema, and inhibition of inflammatory response. In addition, it was found that TCMCR may regulate cerebral blood flow. (Table 4)

Discussion
To date, numerous clinical trials [46][47][48] that have sought new therapeutic agents for treating TBI have proven unsuccessful. However, there is increasing evidence that traditional Chinese medicine, including   TCMCR, extracts, and acupuncture, have clinical benefit in the treatment of TBI patients [49][50][51]. Because TCMCR is the main form of TCM drug treatment, robust evidence of its effects on TBI must be provided. Therefore, we have conducted the first systematic review and meta-analysis of the effects of TCMCR in animal models of TBI. Because systematic review and metaanalysis of animal experiments could provide strong evidence in an unbiased manner. Although small-study effects and statistical heterogeneity among studies were present, our results indicated that TCMCR potentially exerts neuroprotective effects in terms of reduction of brain water content, amelioration of BBB permeability, and deduction of TNF-α/NO after TBI. Similar work [52] was performed in the context of experimental stroke that demonstrated the neuroprotective effects of Buyang Huanwu decoction, a well-known TCMCR, on animal stroke models. Though they are different diseases, many aspects of their respective pathologies are similar, and these investigations provide further evidence of the neuroprotective efficacy of TCMCR, supporting its potential use for human TBI therapy.
Concerning study quality, we found that the methodological quality of most included studies was generally poor, as many failed to report blinded outcome assessment, blinded induction of TBI, sample size calculation, compliance with animal welfare regulations, and potential conflicts of interest. However, because we sought to report on overall quality, we did not arbitrarily exclude them solely on the basis of these defects [53].
The current study has some limitations that have also been observed in previous studies [18,54,55]. First, although we made an effort to identify all relevant studies, our analysis could only be based on articles published in English or Chinese and did not take the unpublished data and the relevant articles published in other languages into account. Moreover, most of them published in local journal particularly, some publications were showed in journal from university. So there was evidence of small-study effects and publication bias should be considered. Second, no studies specifying the degree of severity (e.g., mild, moderate, or severe). The results of different studies could have been more accurately compared if injury severity had been reported consistently. Third, as in previous studies [56][57][58], the methodological quality of the included studies was generally poor. Due to the poor quality of the studies, the results of this review are likely to be influenced by many factors. Of course, it should be noted that negative judgment did not necessarily indicate that the experiment itself was performed inadequately; it indicated that there was inadequate information for assessing its quality. Fourth, although the findings indicate that TCMCR treatment benefits can be found in the TBI model with weight-drop mothed. For the weight-drop model, this model is limited in the production of primary lesions that are macroscopically meaningful once it is not capable of creating cranial fractures with levels of energy compatible with life after impact. It presented discrete focal lesions in a small number of animals and only at elevated levels of energy, but in accordance with what was described in similar studies [59]. Moreover, any single animal model may not be able to fully mimic the highly heterogeneous nature of human TBI [60]. Lastly,  we limited our analysis to the alternation of BBB integrity, brain water content, and TNF-α/NO levels following TBI, largely due to insufficient data regarding histopathology, such as lesion volume and neurobehavioral outcomes such as cognitive performance and motor function.
Additionally, heterogeneity must be considered for any meta-analysis. The main reasons for heterogeneity were the limited number of trials and small cohorts [61]; therefore, additional large-scale clinical trials are required. Another important reason for the existence of heterogeneity was the low quality and potential bias of the trials selected for analysis. The surprisingly low heterogeneity of NO levels in the meta-analysis requires further consideration. It is also associated with the poor methodological quality of the selected trials, which require additional investigation.
To improve the clinical translation, our recommendations for the conduct of future animal studies of TCMCR or other TCM drugs are as follows: (1) More studies have shown that TCMCR is a whole greater than the sum of its parts in terms of its composition and pharmacodynamic action. In order to make full use of the advantages offered by TCMCR, it is essential to address the difficulties in studying TCMCR, including the role played by the material basis and physical basis in TCM's therapeutic effects, and the rules of compatibility, pharmacology, and action mechanism of TCMCR [62]; (2) Other TBI models are needed to investigate the effects of TCM/TCMCR on TBI; (3) Further researchers are strongly recommended to consult and follow the ARRIVE (Animal Research: Reporting In Vivo Experiments) guidelines to report their animal experimental results [63,64]; (4) Other long-term neuropsychological outcomes should also be focused on, such as the cognitive performance, the motor function.