Chronic heart failure (CHF), an end-stage state of cardiovascular disease, is characterized by stagnation of the pulmonary or somatic circulation and insufficient systemic blood perfusion (McDonagh et al., 2023). Epidemiologic data indicate that 64.3 million people worldwide suffer from heart failure (HF) (Savarese et al., 2023). Coronary artery disease, hypertension, diabetes mellitus, and obesity are the main causes of CHF(Heidenreich et al., 2022; Barghash, 2023); consequently, the average age of patients with HF is decreasing because the prevalence of these diseases continues to rise among young people (Lecoeur et al., 2023). HF also imposes an enormous economic burden on global public health. A study across 197 countries showed that the overall global economic cost of HF in 2012 was approximately $108 billion (Kapelios et al., 2023).

Most importantly, CHF is the leading cause of death from cardiovascular disease (Bozkurt et al., 2023). Thus, HF treatment is aimed at delaying its development, improving cardiac function, relieving clinical symptoms, and reducing mortality. Typically, diuretics, angiotensin-converting enzyme inhibitors (ACEIs), angiotensin AT-1 receptor blockers (ARBs), and beta-adrenergic blocking agents are effective CHF treatments (Heidenreich et al., 2022). Yet despite the maturity of current CHF treatment options, the mortality rate from this condition remains high. A longitudinal analysis of 86,000 patients with HF showed a one-year mortality rate of up to 32% after an HF event (Conrad et al., 2019). The search for safer, more effective CHF therapeutic options thus remains a major challenge. Traditional Chinese medicine (TCM) has recently been found to have significant advantages in the management of chronic diseases, including CHF.

Sen Gui capsules (SGCP) are a TCM preparation approved by the National Medical Products Administration of China in 2019 (Approval number: Z20000060, Specification: 0.3 g/capsule, Shanghai Yudan Pharmaceutical Co.). Animal experiments (Zongduo and Runtang, 2003) have shown that SGCP can dilate the rat coronary artery to increase blood flow, lower blood pressure, increase cardiac output, and protect cardiomyocytes. Clinical trials (Songlin et al., 2022) have also demonstrated that SGCP improves cardiac function, delays HF progression, and is efficacious in the adjuvant treatment of CHF. Although its adjunctive efficacy has been demonstrated, the quality of evidence from these clinical studies has not yet been assessed. To date, there have been no systematic reviews or meta-analyses of the efficacy and safety of SGCP in CHF treatment. Therefore, this study assessed the efficacy and safety of SGCP in the treatment of CHF through a systematic review and meta-analysis, to provide high-quality evidence for evidence-based medicine.

This systematic review and meta-analysis was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) (Page et al., 2021) and Cochrane Handbook 2019 (Cumpston et al., 2019) for Systematic Reviews. The PRISMA2020 checklist is detailed in the Supplementary Material S5. The study protocol was registered in the International Prospective Register of Systematic Reviews (https://www.crd.york.ac.uk/PROSPERO/logout.php, PROSPERO, No. CRD42023390409) and had no amendments to the information provided at registration.

To ensure study accuracy, these analyses adopted the consensus statement on the Phytochemical Characterisation of Medicinal Plant extract (ConPhyMP) as a reference when reporting SGCP. We also followed the guidelines for standardizing the scientific nomenclature of botanical drug components. Moreover, we validated these names by cross-referencing them with the websites “Plant of the World Online” (http://www.plantsoftheworldonline.org) and “The World Flora Online” (http://www.worldfloraonline.org) (see compositions of included SGCP in the Supplementary Material S1–S4).

This meta-analysis of data from nine RCTs, representing a cumulative 888 patients with CHF, found that conventional therapy combined with SGCP was more conducive to improving LVEF, increasing treatment efficiency, and decreasing BNP, LVEDD, and hs-CRP compared with conventional therapy alone; however, there was not a significant difference in reduced TNF-α or LVESD.

The quality of the included studies and evidence varied, with the quality of outcomes rated as intermediate-to-very-low, and no high-quality evidence studies. The main reasons for downgrading were inconsistency, limitations, and imprecision. Heterogeneity of LVEF, TNF-α, and BNP were all significant. Heterogeneity reflects widely varying estimates of treatment effects across individual studies, indicating real differences in potential treatment effects (Yang et al., 2023); thus, higher heterogeneity in outcomes across studies is downgraded due to inconsistency across results. In terms of limitations, no study described specific information on allocation concealment and blinding, thereby reducing the quality of evidence assessment based on the study design, which may lead to biased estimates of treatment effects (Wang et al., 2024). In terms of imprecision, when a study sample size is small, CIs for outcomes tend to be wider, contributing to uncertainty of the results (Guyatt et al., 2023). Consequently, the quality of the evidence for LVESD and TNF-α was downgraded, and that for effective rate, hs-CRP, and LVEDD, which were intermediate-quality evidence, had high confidence in the outcome estimation. However, further high-quality studies are required to confirm the effect of conventional treatment combined with SGCP on LVEF, TNF-α, BNP, and LVESD.

Regarding the quality of the literature, in terms of randomization, unbiased interventional studies need to ensure that similar participants receive respective interventions according to the principle of random assignment. Herein, five RCTs (Xiushuang et al., 2008; Fengmei, 2010; Chunjuaan et al., 2013; Qun, 2015; Xiaoyan and Wenping, 2016) did not describe sequence-generated information to determine the risk of bias, which raised the possibility of bias in the allocation of interventions. Intervention assignment in one RCT (Wentao et al., 2013) during sequence generation resulted in a high risk of selective bias. All these factors affect the overall literature quality. In terms of allocation sequence concealment, three RCTs (Ying, 2018; Rui et al., 2021; Songlin et al., 2022) used randomized number tables which may lead to selective recruitment based on prognostic factors. This open random assignment method did not accomplish adequate concealment of participant sequences and was prone to a high risk of selective bias. In addition, none of the other RCTs described the exact method of assigning sequence hiding. A study (Wood et al., 2008) of 146 meta-analyses found that in trials with subjective endpoints where allocation concealment was inadequate or unclear, there might be a component of amplification of the effect of their interventions on outcomes, thereby increasing the risk of bias. In addition, incomplete information made it impossible to determine whether there is a risk of selective reporting of results and the possibility of causing other biases, which would lead to a potential risk of bias in the reported results, which in turn would affect the study results.

Furthermore, none of the RCTs herein reported blinding of patients or outcome assessors. It has been shown that in randomized trials with ambiguous blinding or lack of blinding, the value of their intervention effect on outcomes was exaggerated from the estimated value (Tack, 2023). In general, the more subjective the trial outcome, the greater the bias it creates (Wang et al., 2023). However, the risk of measurement bias was low herein because the outcomes were objective and did not influence the measurement and assessment of the outcome. The data from only one RCT (Chunjuaan et al., 2013) was suspected to be incorrect, which would increase the likelihood of bias in the observed effect estimates. Therefore, these data were excluded to prevent such bias. All RCTs reported all outcomes and there was no risk due to selective reporting. Nonetheless, the studies might also be at risk of bias because of factors such as specific design, baseline imbalance, block randomization of unblinded trials, and varying diagnostic activity; however, no other sources of bias were identified in any of the included RCTs, partly maintaining their quality.

Sensitivity analyses revealed that all the results were robust. The main heterogeneity sources are clinical and methodological (Yalla and Lambrechts, 2023). To assess clinical heterogeneity, we performed subgroup analyses based on mean patient age (<60 or ≥60 years) and SGCP dose (9 capsules/day or 12 capsules/day), finding that neither was a source of heterogeneity. This was explored by excluding single studies on a one-by-one basis to re-run the meta-analysis, showing that patient NYHA classification may have been a primary source of heterogeneity for LVEF. Heterogeneity was reduced by excluding that study. In addition, most included studies did not mention the specific drugs and methods of administration of conventional therapy, and factors such as varying levels of medical care across study centers may have contributed to clinical heterogeneity, which was unavoidable due to the limitations of the objective factors. Methodological heterogeneity might have resulted from including evidence quality ranging from intermediate to very low, and risk of bias in the quality of the literature. Therefore, more studies with higher-quality evidence and higher-quality literature are needed to reduce these heterogeneities.

Characteristics of CHF include dyspnea or limitations in exercise due to impaired ventricular filling or ejection (Metra et al., 2023). Therefore, LVEF measured by echocardiography and serum BNP are important indicators in CHF diagnosis. HF with reduced ejection fraction occurs when the LVEF is ≤ 40%, and is accompanied by progressive left ventricular dilatation and ventricular remodeling (Murphy et al., 2020); this context increases LVEDD and LVESD. BNP is mainly secreted by ventricular myocytes, and promotes excretion, urination, and vasodilatation. When the ventricular load and ventricular wall tension are altered, BNP secretion is promoted to reduce cardiac load and protect cardiac function (Kuwahara, 2021). Therefore, when BNP is elevated above age- and underlying disease-specific thresholds, the potential for HF diagnosis increases accordingly. Moreover, because BNP is an effective predictor of death and acute cardiovascular events at 2–3 months (Yazdani et al., 2023), it can be used to assess cardiac function and long-term prognosis among patients with CHF. As an inflammatory biomarker, hs-CRP is associated with worsening CHF pathogenesis (Magnussen and Blankenberg, 2018). In a prospective cohort study (Burger et al., 2023) of patients with cardiovascular disease, hs-CRP was an independent risk marker for HF development. Therefore, hs-CRP levels may reflect severity and prognosis among patients with CHF. Clinical studies (Hanna and Frangogiannis, 2020) have reported significant TNF-α levels in patients with CHF, and mice with cardiac-specific TNF-α overexpression were more likely to develop dilated cardiomyopathy (Zhang and Dhalla, 2024). This mechanism may be related to TNF-α stimulating the synthesis of other pro-inflammatory cytokines with pro-apoptotic and negative inotropic properties, which leads to ventricular remodeling and HF progression (Galeone et al., 2023). Therefore, TNF-α also reflects disease severity in patients with CHF. These outcomes were chosen to assess treatment effectiveness herein to provide a reference basis for the efficacy of SGCP in combination with conventional HF therapy.

SGCP contains Panax ginseng C.A.Mey [ Araliaceae; ginseng radix et rhizoma], Oreocome striata (DC.) Pimenov and Kljuykov [ Apiaceae; chuanxiong rhizoma], and Neolitsea cassia (L.) Kosterm [Lauraceae; cinnamomi ramulus] with the main active metabolites of ginsenoside, ligustrazine, and cinnamaldehyde. Modern pharmacology (Yating et al., 2018) has demonstrated that ginsenoside Rb1 in P. ginseng C.A.Mey. Improves ventricular remodeling in CHF rats by affecting the expression of periostin proteins in myocardial tissues and inhibiting the TGF-β signaling pathway to slow CHF progression. Ligustrazine in O. striata (DC.) Pimenov and Kljuykov. Exerts anti-apoptotic effects and protects cardiomyocytes by regulating phosphatidylinositol 3-kinase/protein kinase B (Li et al., 2023), and inhibits the JAK kinase/signal transducer activator of transcription signaling pathway of hypertrophic cardiomyocytes to improve ventricular remodeling (Huajuan et al., 2023). Cinnamaldehyde in N. cassia (L.) Kosterm. Inhibits the release of histamine and prostaglandin E, and scavenges excessive oxygen free radicals to protect myocardial cell membranes and prevent myocardial damage (Chang et al., 2023). Substances extracted from N. cassia (L.) Kosterm. can reduce the expression of RhoA and ROCK2, and inhibit the phosphorylation of target molecules downstream of ROCK, thereby dilating central and peripheral blood vessels and reducing cardiac load (Liu et al., 2020). Different classes and sources of effective metabolites in SGCP can achieve multi-component, multi-pathway, and multi-target therapeutics through synergistic or complementary effects (Ning, 2019). Pharmacological experiments (Qian et al., 2018) have shown that SGCP can significantly improve the energy metabolism of the myocardium in rats with cardiac insufficiency after acute myocardial infarction, protect the myocardial mitochondria from lipid peroxidation, and reduce the content of plasma ET and Ang Ⅱ, improving the systolic and diastolic functions of the myocardium. SGCP may improve myocardial energy metabolism by reducing methylmalonic acid accumulation (Wang et al., 2022). In summary, SGCP can improve cardiac function through multiple pathways and has certain advantages in CHF treatment.

Four of the nine RCTs included herein reported adverse events in the SGCP and control groups, without a significant between-groups difference in these incidences. Therefore, SGCP can be considered safe for CHF treatment, with few side effects. At the same time, other clinical studies (Guo and Zhong, 2006; Zhou, 2016; Li, 2017) on SGCP have also proved that its liver and kidney functions before and after treatment are at normal levels. However, these studies were not conducted in recent years, and the side effects of many traditional Chinese medicines are still unclear due to the complex metabolites, resulted it difficult to explore. In addition, the short period of time in which the study was conducted may have contributed to the fact that no significant side effects were reported. Therefore, these reports of adverse events also indicate that the safety of SGCP needs further evaluation, in more high-quality studies.