The Efficacy and Safety of a Qiliqiangxin Capsule Combined with Sacubitril/Valsartan in the Treatment of Chronic Heart Failure: A Systematic Review and Meta-Analysis

Background Qiliqiangxin (QLQX) capsules are a commonly used proprietary Chinese medicine for the adjuvant treatment of chronic heart failure (CHF) in China. In recent years, several randomized controlled trials (RCTs) have reported on the efficacy and safety of QLQX combined with sacubitril/valsartan for CHF. Objective The purpose of this study was to systematically analyze the clinical efficacy and safety of QLQX combined with sacubitril/valsartan in the management of CHF and to provide clinicians as well as scientists with optimal evidence-based medical evidence. Methods We searched RCTs to evaluate the efficacy and safety of QLQX combined with sacubitril/valsartan in the treatment of CHF in the Wanfang Database, China National Knowledge Infrastructure, China Science and Technology Journal Database, PubMed, Embase, and Cochrane Library databases from their inception until January 8, 2022. RCTs on QLQX in combination with sacubitril/valsartan for CHF were included. The outcome measures considered were total effective rate, left ventricular ejection fraction (LVEF), left ventricular end-diastolic dimension (LVEDD), 6-minute walking distance (6-MWD), and adverse events. The quality of the included RCTs was assessed thereafter using the Cochrane risk of bias tool. RevMan 5.3 software was used to conduct the meta-analysis. Results The meta-analysis included 17 trials involving 1427 CHF patients. The results indicated that with sacubitril/valsartan administration combined with QLQX treatment, the total effective rate (relative risk (RR) = 1.24; 95% confidence interval (CI) (1.17, 1.31); p < 0.01), LVEF (mean difference (MD) = 6.20; 95% CI (5.36, 7.05; p < 0.01)), and 6-MWD (MD = 55.87; 95% CI (40.66, 71.09); p < 0.01) of CHF patients were significantly increased, and the LVEDD value of CHF patients was noted to be significantly reduced (MD = −3.98; 95% CI (−4.47, −3.48); p < 0.01). Moreover, there was no increase in the number of adverse events during treatment (RR = 0.67; 95% CI (0.33, 1.34); p < 0.01). Conclusions This study indicated that in CHF patients, on the basis of sacubitril/valsartan treatment, combination with QLQX can potentially enhance the total effective rate, improve LVEF and 6-MWD, and reduce LVEDD values, with good safety. However, considering the poor quality of the included studies, a multicenter, randomized, double-blind controlled study is needed for further confirmation.


Introduction
Chronic heart failure (CHF), which is the end-stage manifestation of cardiovascular disease and the leading cause of mortality, is one of the two major challenges in the cardiovascular feld in the 21st century [1]. Globally, approximately 2% of the population sufers from CHF, and the prevalence is continuously increasing [2]. Te absolute number of patients with CHF has also been increasing due to an aging population, global population growth, and improved survival rates after the initial diagnosis [3]. Te magnitude of this public health issue can be refected in the heavy economic burden of a CHF-related health expenditure of approximately $31 billion per annum, which is expected to increase to $70 billion by 2030 [4]. In recent years, several new drugs have been used to treat CHF. Sacubitril/valsartan is a novel orally active angiotensin receptor-enkephalinase inhibitor. A single cocrystal composed of batroxobin and valsartan at a molar ratio of 1 : 1 has been approved as the frst-line drug for the treatment of heart failure [5]. Several lines of evidence demonstrate that sacubitril/valsartan can signifcantly reduce the risk of cardiovascular death and hospitalization in heart failure patients [6]. Despite advances in sacubitril/valsartan therapy, however, only 20-40% of heart failure patients with reduced ejection fraction have been found to be eligible for initiation therapy with sacubitril/valsartan [7]. CHF still displays a high mortality rate and a poor quality of life; therefore, the development of new drugs with which to treat CHF remains an urgent issue [8]. Chinese herbal medicines have been found to serve as important sources of potential drugs for the treatment of heart failure [9]. In China, to improve the clinical efcacy and safety of sacubitril/valsartan and expand its application, doctors often combine sacubitril/valsartan with Chinese herbal medicines to efectively treat CHF [10].
Te Qiliqiangxin (QLQX) capsule, a well-known proprietary Chinese medicine, has been approved by the National Medical Products Administration and is widely used as an adjunctive therapy in the treatment of CHF. In addition, various preclinical studies have shown that QLQX can signifcantly inhibit pathological cardiac remodeling through a multitargeted, multipathway mechanism of action. For example, it has been previously shown that QLQX can exert its antiapoptotic efects through the activation of PPAR gamma, thereby reducing pathological hypertrophy and improving myocardial fbrosis in isoproterenol-induced CHF mice [11]. Interestingly, in a left anterior descending coronary artery ligation-induced CHF mouse model, QLQX was also observed to markedly improve cardiac function, myocardial fbrosis, and infammatory infltration in rats with heart failure, which may be related to the optimization of glucose metabolism in both the distal and border zone myocardium by QLQX [12]. In addition, in a clinical study involving a multicenter, randomized, double-blind controlled study [13], QLQX was found to substantially improve cardiac function, increase the left ventricular ejection fraction (LVEF), increase the 6-minute walking distance (6-MWD), and thereby reduce composite endpoint events in patients with CHF with a favorable safety profle. All of these studies have suggested that QLQX can serve as a promising potential therapeutic agent for the treatment of CHF. In the last 5 years, clinical studies conducted on QLQX combined with sacubitril/valsartan for the treatment of CHF have been increasing, and these clinical studies have indicated that the combination of QLQX and sacubitril/valsartan may not only exhibit improved clinical efcacy in treating CHF but also reduce the incidence of various adverse events [14,15]. However, individual studies have been inconsistent, which may be due to their limitations. Terefore, we conducted a meta-analysis to evaluate the potential efcacy and safety of QLQX combined with sacubitril/valsartan in the treatment of CHF.

Protocol and Registration.
Before carrying out this study, we registered our research proposal on the website of Open Science Framework at the following registered address: https://osf.io/mkswr. Tis study followed a PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) statement [16] as shown in Supplementary Material S1.

Search Strategies.
Two reviewers independently searched three Chinese electronic databases, namely, the Wanfang Database, China National Knowledge Infrastructure, and China Science and Technology Journal Database, and three English electronic databases, namely, PubMed, Embase, and the Cochrane Library. In addition, to fnd various unpublished clinical studies, we also searched clinicaltrials.gov and the Chinese Clinical Trial Register (ChiCTR) website to further explore the diferent ongoing clinical trials. Our search keywords were "heart failure," "Qiliqiangxin," and "sacubitril/valsartan," and the search time was from inception to January 8, 2022. We also manually searched the reference lists of all full-text papers for other relevant reports.

Type of Studies.
Randomized controlled trials (RCTs) that investigated the efectiveness and safety of QLQX for CHF were included.

Type of Participants.
Te participants met the recognized diagnostic criteria, such as the "2016 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure" [17] and the "Guidelines for diagnosis and treatment of heart failure in China 2018" [18]. Race, sex, and source of the case were not limited.

Types of Interventions.
Te control group should have been treated with a conventional treatment regimen (e.g., diuretics, beta-blockers, and aldosterone antagonist) in combination with sacubitril/valsartan, and the intervention group should have been treated with QLQX capsule on the basis of the control group, without limiting the dose, frequency, duration of treatment, or manufacturer of QLQX.

Types of Outcome Measures. Te outcome indicators
were the LVEF, left ventricular end-diastolic dimension (LVEDD), 6-MWD, and total efective rate, and treatments resulting in remission of the various clinical signs and symptoms of CHF and with cardiac function improvement at grade 1 or above after treatment were considered to be "efective" according to the NYHA scale; otherwise, they were considered to be "inefective." Te ratio of the number of efective cases to the total number of cases was considered to be the efective rate.

Safety
Outcome. All adverse events that occur after a subject has received a test dose may be manifested by the presence of symptoms, signs, disease, or abnormal laboratory tests.

Exclusion Criteria.
Exclusion criteria include the following: (i) Case reports, literature reviews, meta-analyses, conference papers, and so on (ii) Inability of the study outcomes to be extracted (iii) Animal experiments 2.6. Study Selection and Data Extraction. Te literature selection and data extraction were carried out by two reviewers independently. If there was any disagreement during this process, it was decided primarily by means of discussion or seeking a third reviewer. Te reviewers frst read the title and abstract of the article, screened out the literature that could be included according to the inclusion criteria, further obtained the full text, and then read the full-text information in detail to screen the literature that accurately met the criteria. Te data extraction was mainly performed to extract basic data from the included studies, such as the frst author, year of publication, participation, mean age, sex, intervention measures with specifc drugs (such as QLQX dose, frequency, and manufacturer), treatment time, followup time, funding for information, posttreatment follow-up, clinical efectiveness of the CHF patients' LVEF before and after the treatment, potential changes in LVEDD and 6-MWD, and associated adverse events. If missing information was found during data extraction, the authors of the original study were contacted via e-mail to obtain it.

Risk of Bias Assessment.
In this study, two researchers independently assessed the quality of the included trials according to the Cochrane bias risk assessment tool [19]. Te assessment was carried out by means of evaluation of the following seven diferent items: (1) random sequence generation, (2) allocation concealment, (3) blinding of participants and personnel, (4) blinding of outcome assessment, (5) incomplete outcome data, (6) selective reporting, and (7) other bias. Each item was potentially assessed as having a high risk, low risk, or unclear risk.

Statistical Synthesis and Analysis.
In this study, Rev-Man v.5.3 software was used for meta-analysis of the relevant outcome indicators, and the efective values of dichotomous variables were expressed as the relative risk (RR) and the 95% confdence interval (CI). Continuous variables were represented by the mean diference (MD) and the 95% CI, and p < 0.05 indicated a signifcant diference. In this study, the chi-square test was also used to test the heterogeneity of the included literature, and appropriate statistical methods were selected to analyze the results according to the heterogeneity [20]. If I 2 was 0, this meant that there was relatively little possibility of heterogeneity among the multiple similar studies and that the homogeneity was good. If I 2 was <50%, this indicated small interstudy heterogeneity and that a fxedefects model could be used for statistical analysis; if I 2 was >50%, this indicated large interstudy heterogeneity, and the possible source of heterogeneity was analyzed frst; if it was determined that the various clinical factors caused large heterogeneity, a subgroup analysis was also performed. However, if after the subgroup analysis the heterogeneity was still large, a random-efects model was selected to calculate the statistics.

Subgroup Analysis.
Considering that the age of the patients and the duration of the treatment will afect the efcacy of QLQX in the treatment of CHF, we conducted a subgroup analysis based on the aforementioned two factors.

Sensitivity
Analysis. Sensitivity analysis is a common analytical method employed to quantitatively describe the infuence of the corresponding input variables of a specifc model upon its corresponding output variables. According to the scope of its various efects, it can be divided into global and local sensitivity analyses. Tis study mainly used local sensitivity analysis (conducted by the package "meta" (version 5.2-0) in R (version 4.1.2)). Te main method comprised removing each included RCT from the metaanalysis and then again conducting a comprehensive statistical analysis, which involved redrawing the forest map of the meta-analysis, followed by comparing the results with those of previous studies to evaluate the stability of the results [21].

Publication
Bias. If the observed outcome indicators were more than 10 included studies, a funnel plot was used to evaluate publication bias. However, if the funnel plot was asymmetric, Egger's test (conducted by the package "meta" (version 5.2-0) in R (version 4.1.2)) was employed to further evaluate the degree of publication bias, and p < 0.05 was considered to indicate the existence of publication bias.

Search Results.
A total of 68 studies were retrieved from the aforementioned database, leaving a total of 40 studies after removing the various duplicates. After reading the title and abstract, 14 studies were excluded because 12 of them included patients sufering from other diseases (e.g., atrial fbrillation and coronary heart disease), and two of them were non-RCT studies. Te full text of the 26 studies was further obtained. After reading the full text, nine studies were excluded, including two studies whose subjects were defned as having cardiac insufciency and did not meet the diagnostic criteria for CHF and one study that was also excluded due to the use of a multiarm clinical trial. In addition, since the duration of the treatment was not reported in one study, we also contacted the corresponding author of the article via e-mail to obtain more detailed information but failed to do so. Tus, considering the incompleteness of the study, this paper was excluded from the meta-analysis. Te remaining fve studies were excluded from inclusion because the interventions were based on the QLQX only. Finally, only 17 studies [14,15,[22][23][24][25][26][27][28][29][30][31][32][33][34][35][36] were included for systematic evaluation and meta-analysis. A fowchart of the study selection is shown in Figure 1.

Quality Evaluation.
Te Cochrane bias risk assessment tool was used to assess the quality of each study. We tried to e-mail the corresponding author of the article to collect more information on these selected studies; unfortunately, we did not receive any response. Fourteen trials [14, 15, 22-27, 29-32, 35, 36] were found to report a random allocation of participants, but only seven trials [14, 24-27, 31, 32] mentioned the use of a random number table. One study [33] used the Excel random sampling method for the grouping, which was considered to have a "low risk," while the remaining six trials [15,22,23,29,35,36] did not mention the method used for randomization in the sequence generation, which were therefore selected as having a "high risk." In addition, one trial [34] performed the grouping according to the therapeutic agent administered, which was considered to have a "high risk." Te remaining two trials [28,33] did not mention randomized grouping; therefore, we were not sure of the exact grouping method used and, thus, these were judged to have an "high risk." In terms of the allocation concealment, none of the studies mentioned specifc information on allocation concealment and, therefore, were judged to have an "unclear risk." Regarding the implementation of blinding, none of the studies were blinded to the patients, study personnel, and outcome measures; therefore, these features were judged to have a "high risk." In addition, all of the included RCTs contained no incomplete data or selective reporting and were labeled as having a "low risk." Although all of the included studies reported consistent fndings regarding between-group baselines, potential sources of bias (such as intention to treat and other adherence diferences) may still exist.
Terefore, other biases were assessed as having an "unclear risk" after attempts made to contact the authors via e-mail to clarify the various unreported information but did not receive a response. Te results of the risk of bias assessment are shown in Table 2.

Total Efective
Rate. Tirteen studies [15,22,[24][25][26][27][28][29][30][31][32][33]35] involving 1057 participants reported the total efective rate. Te heterogeneity test results indicated that the heterogeneity of the studies among each group was small (I 2 � 0%), and the fxed-efects model was used for the meta-analysis. Te results showed that therapy involving sacubitril/valsartan combined with QLQX signifcantly improved the total efective rate of CHF patients (RR � 1.24; 95% CI (1.17, 1.31); p < 0.01; Figure 2). Sensitivity analysis showed that the results were robust (Supplementary Material S2.1). In addition, subgroup analysis was performed to consider diferences in the treatment duration and the age of included patients between the study groups. Te results of the subgroup analysis showed that QLQX combined with sacubitril/valsartan was superior to sacubitril/valsartan alone in improving the total efective rate, regardless of the age of CHF patients being ≥65 years or <65 years and regardless of the duration of treatment being >30 days or ≤30 days. Moreover, subgroup analysis according to diferent ages and treatment durations showed no signifcant difference between the groups (p for interaction � 0.66 and 0.73, respectively; Supplementary Materials S2.2 and S2.3).

6-MWD.
Ten studies [15,23,24,28,30,31,[33][34][35][36] reported the 6-MWD outcome. Te results of the heterogeneity test showed that the heterogeneity of the studies between the various groups was signifcantly high (I 2 � 99%). Sensitivity analysis was performed by excluding studies one by one. After removing the studies reported by Zhang [36], the heterogeneity between studies was signifcantly reduced (I 2 � 82%; Supplementary Material S2.10). However, the heterogeneity was still >50%, and a random-efects model was used for the meta-analysis. Te results indicated that treatment involving sacubitril/valsartan combined with QLQX signifcantly increased the 6-MWD of CHF patients (MD � 55.87; 95% CI (40.66, 71.09); p < 0.01; Figure 5). Subgroup analysis according to diferent ages and treatment durations showed no signifcant diference between the groups (p for interaction � 0.38 and 0.48, respectively; Supplementary Materials S2.11 and S2.12), thereby suggesting that age and treatment duration may not be the primary sources of the heterogeneity.

Adverse Events.
Of the 17 included studies, eight trials [14,15,23,28,31,33,34,36] did not report adverse events. Adverse events were reported in nine studies [22, 24-27, 29, 30, 32, 35], among which no adverse reactions were observed during treatment in fve studies [24,26,29,30,35], and some adverse reactions (such as nausea, vomiting, headache, and dizziness) were observed in both the intervention group and the control group during treatment in four studies [22,25,27,32]; there were 3.4% (12/358) adverse reactions in the intervention group and 4.7% (17/358) adverse reactions in the control group, as shown in Table 3. Te results of the heterogeneity test showed that the heterogeneity of the various studies between the groups was relatively small (I 2 � 0%), and a fxed-efects model was used for the meta-analysis. Te results indicated that for sacubitril/valsartan combined with QLQX, there was no signifcant increase in adverse events during the treatment period (RR � 0.67; 95% CI (0.33, 1.34); p < 0.01; Figure 6).

Publication Bias.
Te RCTs of the total efective rate, LVEF, and 6-MWD were found to be 13, 16, and 10, respectively, which were signifcantly greater than 10, and funnel plots were constructed to assess the potential publication bias. A funnel plot of data relating to the total effective rate and 6-MWD showed asymmetry (Figures 7(a) and 7(b)), and Egger's test (Figures 7(d) and 7(e)) showed statistical signifcance (p � 0.0312 and 0.0002). A funnel plot of the LVEF data showed symmetry (Figure 7(c)), and the

Main Results of Tis Research.
In this study, we conducted a comprehensive search of the various relevant clinical randomized studies of QLQX combined with sacubitril/valsartan for the treatment of CHF by searching relevant databases in both English and Chinese, which resulted in a total of 17 studies involving 1427 participants for meta-analysis. Our study showed that QLQX combined with sacubitril/valsartan performed signifcantly better in terms of improved overall efcacy, LVEF, and 6-MWD and reduced LVEDD than did sacubitril/valsartan alone. Te results did not change in the subgroup analysis based on age and treatment duration. Regarding the safety of therapy with sacubitril/valsartan combined with QLQX, there was no signifcant increase in adverse events during the treatment period, and no serious adverse events were observed in   either group, which suggested that QLQX was generally efective and safe for patients with CHF.
We analyzed the quality of all the included studies through the Cochrane bias risk assessment tool, and our results indicated that the quality of all the included studies was relatively low. First, in terms of the randomization method, most of the included original literature only stated that "randomized" was used. Moreover, in terms of the randomization method, most of the included original literature only stated the word "random" but did not accurately describe the specifc randomization method used, or the randomization method employed was not reasonable. Second, in terms of allocation concealment, none of the included studies described the allocation concealment. In addition, in terms of blinding implementation, none of the included studies were blinded to the patients, investigators, or statisticians. Finally, none of the studies registered their study protocols in clinical research registries (e.g., the Chinese Clinical Trial Registry) at the time at which they were conducted. Tese results suggested that the poor quality of our included studies could further signifcantly reduce the credibility of the results of our meta-analysis. Based on the shortcomings of the currently available clinical studies related to the use of QLQX in combination with sacubitril/valsartan for the treatment of CHF, we make the following recommendations for conducting future clinical studies. First, researchers should refer to the contents of the CONSORT statement [37] in detail, develop a proper study protocol, and register it on the relevant websites before conducting the clinical studies. Second, an appropriate random assignment method (e.g., random number table) should be selected to assign the subjects to diferent groups, with detailed descriptions of the hidden contents of the assignment. In addition, if possible, the studies should be conducted in a blind manner, and subjects as well as the control group can be placebocontrolled, which would potentially exclude the placebo efect of QLQX.
Te total efective rate is a comprehensive index for assessing signifcant improvement in patients with heart failure and is widely used as a review indicator for the various drugs used to treat heart failure [26]. Our Study   meta-analysis showed that QLQX combined with sacubitril/ valsartan was substantially better than sacubitril/valsartan alone in improving the treatment efciency in CHF patients. Moreover, subgroup analysis showed no signifcant efect of the treatment duration and the mean age of the participants on this outcome, with the good heterogeneity and the sensitivity analysis suggesting robust results. We also observed signifcant improvements in LVEF, LVEDD, and 6-MWD values following the use of QLQX combined with sacubitril/valsartan for CHF. LVEF and LVEDD are considered to be important indicators of cardiac function and play an important role in the prognosis of heart failure patients [38,39]. Te 6-min walk test [40] is a simple, safe, and convenient test that has been widely used to assess the exercise tolerance of patients with CHF and evaluate the severity of heart failure. Our results suggested that QLQX combined with sacubitril/valsartan was superior to sacubitril/ valsartan alone in improving LVEF and 6-MWD and, thus, reducing LVEDD in patients with CHF. However, we also observed greater heterogeneity in the two outcome indicators (LVEF and 6-MWD). Terefore, we performed a metaanalysis by means of a random-efects model; in addition, we conducted a subgroup analysis based on the duration of the treatment and the mean age of the participants included in the study. We observed that none of the meta-analysis results were reversed, but the heterogeneity was also not reduced, and the sensitivity analysis suggested robust results. Furthermore, upon reviewing the characteristics of the included studies, we found that the source of heterogeneity may potentially arise from the fact that various aspects of the routine treatment of the participants difered between studies. In addition, all included clinical studies were conducted primarily within Chinese hospitals, most of which are at the county level in China, with varying levels of care and signifcant variation in the standardization of medication for CHF, further increasing the clinical heterogeneity between studies.
It has been established that the treatment goals for CHF include improving the quality of life, improving long-term prognoses, and reducing morbidity, mortality, and hospitalization rates [41]. However, only a few included studies focused on the various outcome indicators to improve quality of life, such as the Kansas City Cardiomyopathy Quality of Life Questionnaire [42] and the Minnesota Heart Failure Quality of Life Questionnaire [43]. Moreover, none of the studies have undertaken long-term follow-ups with patients. Reductions in both mortality and hospitalization rates are also considered to be key indicators with regard to the evaluation of heart failure medications [41], which, unfortunately, were not observed in any of the reported studies. In conclusion, it is still unknown whether QLQX combined with sacubitril/valsartan can improve the quality of life of patients with heart failure, improve long-term prognoses, and reduce mortality and hospitalization rates; therefore, future clinical studies should focus on these important outcome indicators.
In addition to clinical efcacy, safety is also regarded as the most important concern in herbal formulations. In the present study, regarding safety, we observed that, in comparison with sacubitril/valsartan alone, QLQX combined with sacubitril/valsartan did not show signifcant adverse efects during the treatment of CHF, which was consistent with the fndings of various previous studies. Tis fnding appeared to indicate that QLQX is relatively safe in the treatment of CHF, but we should also clearly recognize that there were eight studies [14,15,23,28,31,33,34,36] that did not report their safety results; therefore, the safety profle of QLQX was unclear in these studies. In addition, the longterm clinical safety of QLQX is unknown due to the lack of long-term follow-up data. In future studies, we should also observe the long-term safety profle of QLQX during the treatment of CHF.
Te pharmacokinetic study of compound traditional Chinese medicine is the core of the modernization of traditional Chinese medicine, which can provide an experimental basis for rational clinical drug use and safety evaluation [44]. A previous pharmacokinetic study used high-performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS) technology to screen the main active ingredients of QLQX [45]. Te research results showed that there were mainly 29 active ingredients of QLQX: 10 of these were saponins (astragaloside, ginsenoside Rg1, Rg3, Rb1, Rb2, Rc, Rd, Re, Rf, and F2), six were phenolic acids (salvianolic acid A, salvianolic acid B, lithospermic acid, danshensu, rosmarinic acid, and protocatechuic acid), six were favonoids (calycosin-7-glucoside, formononetin, hesperidin, hydroxysafor yellow A, rutin, and quercetin), and seven were alkaloids (aconitine, hypaconitine, mesaconitine, benzoylaconine, benzoylhypaconine, enzoylmesaconine, and sinapine bisulfate) [45]. Some active ingredients have been shown to have protective efects on the heart. For example, astragaloside can inhibit ventricular remodeling and improve cardiac function by reducing oxidative stress in cardiomyocytes and reducing mitochondrial damage [46], and salvianolic acid may exert cardioprotective efects by promoting angiogenesis [47].

Publication Bias.
Our study also showed publication bias for the outcome measures of the total efective rate and 6-MWD, which was inevitable because all of the studies were based in China and reported positive results. Although we searched the current major clinical research registries in an attempt to include various unpublished studies, we did not fnd any relevant studies. Tus, considering that none of the included original studies were registered prior to trial initiation, we suggest that future investigators follow the clinical trial registration system to bring their clinical trials into the public domain from the very beginning, increase the transparency of the clinical trials, and improve the overall standard of the clinical trials, thereby efectively reducing the publication bias.

Comparison with Other Meta-Analyses.
In this study, we compared the fndings of various previously reported metaanalyses. For instance, Xu's et al. [10] research team systematically evaluated the clinical efcacy and safety of QLQX combined with conventional drugs in the treatment of CHF, and the fndings indicated that the combination of QLQX with conventional drugs was not limited to QLQX combined with sacubitril/valsartan. Teir subgroup analysis showed that QLQX combined with sacubitril/valsartan was signifcantly better than sacubitril/valsartan alone in improving clinical efciency and LVEF, which was also a fnding that was consistent with the present study, but the previous report only included four clinical trials, totaling 358 patients with a search deadline of August 2020. In contrast, the present study included 17 clinical trials with a total of 1427 patients with CHF, and our current meta-analysis is the most recent and comprehensive, which further strengthens the results of the previous meta-analysis. In addition, our study is more reproducible because we detail the screening details of the literature in the appendix.

Limitations.
Tere are some limitations associated with this study that are worth discussing. First, we found large heterogeneity in the assessment of the efect of QLQX combined with sacubitril/valsartan in improving LVEF and 6-MWD in patients with CHF, as indicated by the I 2 values. Tis is predictable because of the interstudy diferences in the quality of studies evaluated, the populations enrolled, and the specifc dosing regimens of treatment (e.g., conventional drug therapy). Despite our use of random-efects models and subgroup analysis, the study's heterogeneity persisted. Second, we conducted only a quantitative metaanalysis based primarily on the secondary data, which might have led to inaccurate results due to insufcient raw individual patient data. Furthermore, considering that Chinese herbal medicines are also popularly used in Asian countries such as South Korea and Japan and given that only Chinese and English databases were searched in this study, it is possible, albeit highly unlikely, that other relevant literature was missed during analysis.

Conclusion
Overall, our study demonstrated a signifcantly better efcacy of QLQX combined with sacubitril/valsartan in comparison with sacubitril/valsartan alone in terms of an improved total efective rate, increased LVEF and 6-MWD, and reduced LVEDD with a favorable safety profle in the CHF treatment. However, because of the poor quality of the original study, more high-quality clinical studies are needed to verify the clinical efcacy and safety of QLQX.