Overall, a total of 812 studies were identified in the initial search (Fig. 1). After removing duplicates, 569 studies were retained. After screening titles and abstracts, 435 studies were excluded because of irrelevant study design. Afterward, 134 studies were eligible and examined, of which 58 were further excluded due to the following reasons: (1) the study type missed eligibility criteria (n = 26), (2) studies with insufficient data (n = 22), (3) studies without eligible outcomes (n = 5), (4) the study’s intervention missed eligibility criteria (n = 3), and (5) repetitive studies (n = 2). Furthermore, 8 studies were identified through expert advice and personal communication and 6 studies were excluded because they were irrelevant and animal experiment reports.

Flow chart of the search for eligible studies

Finally, 77 eligible RCTs that evaluated the use of 8 CPMs combined with CAM against DCM-HF were included for the NMA, including QLQX (47 RCTs), WX (11 RCTs), TXL (8 RCTs), QSYQ (2 RCTs), SXBX (3 RCTs), YXST (2 RCTs), YXSC (2 RCTs), and GTTL (2 RCTs); all of them were carried out in China between 2006 and 2023.

Seventy-seven RCTs with 6980 patients accorded with the eligibility criteria, including 3523 patients in the experimental groups and 3457 patients in the control groups [18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94]. Among the participants, more than half were men and the majority were middle-aged and elderly people. The intervention in the control groups was current CAM treatments, including angiotensin II receptor antagonists, angiotensin-converting enzyme inhibitors, angiotensin receptor-neprilysin inhibitors, mineralocorticoid receptor antagonists, cardiac glycosides, nitrates, beta-blockers, and so on. The experimental groups received one of the CPMs identified based on the control groups. The duration of RCTs ranged from 1 to 48 weeks; it was 12 weeks in 33.8% of RCTs and 4 weeks in 27.3% of them, respectively. The details of the study characteristics are presented in Table 2.

The compared connections among interventions for each outcome are shown in Fig. 2. Each node represents a different intervention and the size of nodes is positively correlated with the number of patients. The thickness of the line segment corresponds to the number of included studies for that intervention. The thicker the line segment, the larger the number of included studies for that intervention is. There were no closed loops formed between the studies, thus, the assumption of consistency between direct and indirect evidence was not utilized in this NMA.

Network graph of the outcomes. A Clinical effectiveness rate (CER). B Left ventricular ejection fraction (LVEF). C Left ventricular end-diastolic dimension (LVEDD). D Six-minute walk test (6MWT). E Brain natriuretic peptide (BNP). F Cardiac output (CO). Abbreviations: QLQX, Qili Qiangxin capsule; WX, Wenxin granule; TXL, Tongxinluo capsule; QSYQ, Qishen Yiqi dropping pill; SXBX, Shexiang Baoxin pill; YXST, Yangxinshi tablet; YXSC, Yixinshu capsule; GTTL, Getong Tongluo capsule; CAM, complementary and alternative medicine

The Cochrane risk-of-bias assessment tool was used to perform a quality evaluation. Thirty-one studies generated randomization via random number table [24, 25, 33, 34, 37, 39, 41, 42, 44, 46,47,48,49, 52,53,54,55, 57, 60, 62, 63, 66, 72, 75, 78,79,80, 82,83,84, 94], and two studies generated randomization via draw method [38, 51]. Besides, block randomization, stratified randomization, touch ball method, and double chromosphere method were also used for randomization [35, 59, 61, 87]. Seven studies used randomization methods prone to high risk of bias including grouping by the time of admission, odd and even outpatient numbers, and visiting sequence [40, 68, 70, 73, 86, 90, 91]. The remaining RCTs referred to only random grouping. One study described the information on allocation concealment which was grouped using the sealed envelope method [58] and one study was designed as a double-blind study [35]. Additionally, the rest of the studies did not provide detailed information on allocation concealment and blinding. The detection bias was evaluated as “low risk” because the measurement of related results of the included RCTs was not affected by the blinding toward the outcome assessors. All the study outcome reports were complete; therefore, it was considered that there was no risk of incomplete outcome data. Considering that the complete implementation scheme could not be acquired, the reporting bias was evaluated as “unclear risk.” No other obvious bias was observed in all included studies, so this review assumed that there were no other bias risks. The quality assessment of the included RCTs is shown in Fig. 3.

Fifty-eight RCTs reported the CER of eight types of CPMs. Figure 4A and Table 3 showed that in GTTL+CAM vs CAM (OR = 5.95, 95% CI 2.04–17.40), QSYQ+CAM vs CAM (OR = 5.82, 95% CI 2.08–16.28), YXST+CAM vs CAM (OR = 4.13, 95% CI 1.58–10.75), TXL+CAM vs CAM (OR = 3.73, 95% CI 2.34–5.92), QLQX+CAM vs CAM (OR = 3.52, 95% CI 2.78–4.46), WX+CAM vs CAM (OR = 3.45, 95% CI 2.27–5.23), YXSC+CAM vs CAM (OR = 1.94, 95% CI 0.91–4.13), and SXBX+CAM vs CAM (OR =1.89, 95% CI 0.59–6.07), it was observed that the CPMs (GTTL, QSYQ, YXST, TXL, QLQX, and WX) combined with CAM had a better clinical effectiveness rate compared with CAM alone.

Forest plot of the outcomes. A Clinical effectiveness rate (CER). B Left ventricular ejection fraction (LVEF). C Left ventricular end-diastolic dimension (LVEDD). D Six-minute walk test (6MWT). E Brain natriuretic peptide (BNP). F Cardiac output (CO). Abbreviations: QLQX, Qili Qiangxin capsule; WX, Wenxin granule; TXL, Tongxinluo capsule; QSYQ, Qishen Yiqi dropping pill; SXBX, Shexiang Baoxin pill; YXST, Yangxinshi tablet; YXSC, Yixinshu capsule; GTTL, Getong Tongluo capsule; CAM, complementary and alternative medicine

The results of SUCRA suggested that GTTL + CAM was the optimal combination, followed by QSYQ + CAM, YXST + CAM, and TXL + CAM (Table 4 and Fig. 5A).

Plot of the surface under the cumulative ranking curves for outcomes. A Clinical effectiveness rate (CER). B Left ventricular ejection fraction (LVEF). C Left ventricular end-diastolic dimension (LVEDD). D Six-minute walk test (6MWT). E Brain natriuretic peptide (BNP). F Cardiac output (CO). Abbreviations: QLQX, Qili Qiangxin capsule; WX, Wenxin granule; TXL, Tongxinluo capsule; QSYQ, Qishen Yiqi dropping pill; SXBX, Shexiang Baoxin pill; YXST, Yangxinshi tablet; YXSC, Yixinshu capsule; GTTL, Getong Tongluo capsule; CAM, complementary and alternative medicine

Sixty-six RCTs reported the LVEF of eight types of CPMs. As shown in Fig. 4B and Table