Efficacy of Jianpi formulas (健脾剂) in reducing the recurrence of colorectal adenoma after polypectomy: a systematic review and Meta-analysis

Abstract

OBJECTIVES:

To systematically review the efficacy and safety of Jianpi formulas (健脾剂) in reducing the recurrence of colorectal adenoma (CRA) after poly-pectomy.

METHODS:

Randomized controlled trials (RCTs) investigating Jianpi formulas for CRA post-polypectomy were systematically retrieved from eight electronic databases. The quality of the methodology was assessed using the Cochrane collaboration tool. The Grades of Recommendations Assessment Development and Evaluation (GRADE) approach was employed for evidence assessment. Statistical analyses were conducted using Statistics and Data Analysis (STATA) 17 (StataCorp (College Station, TX, USA) and Review Manager (RevMan) 5.4 (The Cochrane Collaboration (London, UK).

RESULTS:

The Meta-analysis, encompassing 18 RCTs with 1838 patients, revealed that Jianpi formulas significantly outperformed postoperative routine treatment. It demonstrated a reduction in the half-year recurrence rate [relative risk (RR) = 0.41, 95% confidence interval (CI) = 0.33-0.49, Z = 9.08, P < 0.000 01], the one-year recurrence rate [RR = 0.58, 95% CI= 0.49-0.69, Z= 6.12, P < 0.000 01], and an enhancement in the clinical effective rate [RR = 1.27, 95% CI = 1.19-1.36, Z= 7.06, P < 0.000 01]. The half-year recurrence rate and the clinical effective rate were medium-quality evidence. The one-year recurrence rate was low-quality evidence. Additionally, Jianpi formulas appear to be safe and do not increase adverse reactions compared to postoperative routine treatment alone.

CONCLUSION:

Jianpi formulas exhibit efficacy in reducing postoperative half-year and one-year recurrence rates while improving the clinical effective rate after polypectomy for CRA.

1. INTRODUCTION

Colorectal adenoma (CRA) is characterized by abnormal epithelial growth in the colorectal mucosa^1,2 and serves as the primary precancerous lesion for colorectal cancer (CRC), accounting for 85%-90% of all such lesions.³ The widely acknowledged 'adenoma-carcinoma' sequence in CRC progression indicates a 10-15 years transition from adenoma to carcinoma in situ, providing an opportunity for effective tumor incidence reduction through screening and diagnostic tools.

Despite the high removal rate and relative safety of polypectomy, long-term studies reveal a significant CRA recurrence rate: 37% at 1 year, 47% at 2 years, and 60% at 5 years.⁴ This underscores the urgent need for effective strategies to prevent CRA recurrence and reduce precancerous lesions.

In addition to polypectomy, the prevention and treatment of CRA by Western Medicine also included colonoscopy screening, drug therapy, regular follow-up, and lifestyle modifications.⁵ Consensus among relevant experts explicitly suggests that appropriate physical exercise, smoking cessation, and reducing red meat intake can reduce the recurrence rate of CRA.^6,⇓-8 The 2020 UK guidelines recommend surveillance at 3 years for high-risk patients.⁹ Conventional pharmaceutical treatments for CRA involve calcium, vitamin D, aspirin, and celecoxib, among others. However, there is considerable controversy regarding their efficacy in preventing adenoma recurrence.^10,11 For example, a Meta-analysis included 77 000 participants found that taking low-dose aspirin for 2-4 years reduced the risk of CRA recurrence, but high-dose aspirin did not reduce the recurrence rate of adenoma.¹² As one of the common drugs for the treatment of CRA recurrence, aspirin is associated with an increased risk of gastrointestinal diseases, cardiovascular diseases, and coagulation disorders with long-term use.^13,14 Given the limitations of current treatment, there is an urgent need for safe and effective complementary or alternative medicines.

According to Traditional Chinese Medicine (TCM), spleen deficiency is considered the root cause of CRA. Studies have shown that the application of Chinese herbal Jianpi formulas (健脾剂) to intervene and treat postoperative CRA patients has a positive effect on reducing the recurrence and canceration rates of CRA.¹⁵ Jianpi therapy can intervene in CRA by affecting the tumor microenvironment, enhancing the intestinal mucosal barrier function, blocking the canceration of mucosal epithelial cells, and slowing down or even reversing the progression of CRA to CRC.^16,17 Existing Meta-analyses has demonstrated that TCM significantly prevents and treats CRA. Chen et al ¹⁸ suggested that TCM effectively reduces the recurrence rate of colorectal polyps after polypectomy. Lin et al ¹⁹ reported that oral administration of Chinese herbs is effective in reducing the recurrence rate of CRA after polypectomy and improving symptoms of abdominal pain and diarrhea. Jianpi formulas refer to prescriptions aimed at strengthening the spleen. Some Meta-analyses have shown that these formulas have a notable curative effect on CRC and colorectal polyps after polypectomy. However, the available evidence for Jianpi formulas primarily concerns colorectal polyps or CRC rather than CRA.^20,21 The current state of evidence of Jianpi formulas in preventing CRA recurrence remains unclear.

Therefore, this study aims to construct a comprehensive review of polypectomy and postoperative treatment to confirm the efficacy and safety of Jianpi formulas in preventing CRA recurrence.

2. MATERIALS AND METHODS

2.1. Registration

The registration number for this study is No. CRD42023484240, which can be found on the International Prospective Register of Systematic Reviews (PROSPERO) website (https://www.crd.york.ac.uk/PROSPERO/).

2.2. Database search

The materials for this study were retrieved from the following Chinese and English databases up until August 2023: Public Medline (PubMed), Excerpta Medica Database (Embase), Web of Science (WOS), Cochrane Library, China National Knowledge Infrastructure (CNKI), Wanfang Data, Chinese Biomedical Literature Database (SinoMed), Chinese Scientific Journals Database (VIP). The search aimed to identify relevant randomized controlled trials (RCTs). To include unpublished studies, we also searched the website of the international clinical trial registry provided by the United States National Institutes of Health, the Chinese Clinical Trial Registry (ChiCTR) and International Clinical Trials Registry Platform (ICTRP). Ultimately, a total of 1015 clinical registration projects were identified, but all were excluded after screening. The search was conducted independently by two researchers (XIAO Jing and SONG Danlei).

2.3. Retrieval method

The search terms included two main categories combined with the Boolean operator "AND": Chinese medicine-related terms (Chinese herbal medicine, Traditional Chinese medicine, Classical Chinese herbal formulas, Chinese herb, Chinese herb therapy, Herbal medicine, Herb therapy, Herbal remedy) and adenoma-related terms (Adenoma, Adenomas, Polyps, Polyp, Adenomatous Polyp, Colonic Polyps, Intestinal Polyp, Intestinal Polyps, Colorectal adenoma, Adenomatous polyps, Colonic adenoma, Rectal adenoma, Colorectal neoplasms). These terms were translated into Chinese for searches in Chinese databases. The PubMed search strategy was as follows: (Adenoma [Title/Abstract] OR Adenomas [Title/Abstract] OR Polyps [Title/Abstract] OR Polyp [Title/Abstract] OR Adenomatous Polyp [Title/Abstract] OR Colonic Polyps [Title/Abstract] OR Intestinal Polyp [Title/Abstract] OR Intestinal Polyps [Title/Abstract] OR Adenomatous Polyp [Title/Abstract] OR Colorectal Adenoma [Title/Abstract] OR Adenomatous Polyps [Title/Abstract] OR Colonic Adenoma [Title/Abstract] OR Rectal Adenoma [Title/ Abstract] OR Adenomatous Polyp [Title/Abstract] OR Colorectal Neoplasms [Title/Abstract]) AND (Chinese Herbal Medicine [Title/Abstract] OR Traditional Chinese Medicine [Title/Abstract] OR Classical Chinese Herbal Formulas [Title/Abstract] OR Chinese Herb [Title/Abstract] OR Chinese Herb Therapy [Title/Abstract] OR Herbal Medicine [Title/Abstract] OR Herb Therapy [Title/Abstract] OR Herbal Remedy [Title/Abstract]).

2.4. Inclusion criteria

The inclusion criteria were as follows: (a) type of study: RCTs; (b) patients: individuals meeting the diagnostic criteria for CRA and receiving a pathological diagnosis of adenomas or adenomatous polyps following polypectomy, with no limitations related to age, ethnicity, gender; (c) intervention: oral Jianpi formulas combined with postoperative routine treatment (including conventional medicine, rehydration therapy, nutritional support, postoperative fasting, etc.); comparator: postoperative routine treatment alone; (d) primary outcomes: the recurrence rate of CRA, including the half-year recurrence rate and the one-year recurrence rate; Secondary outcomes: the clinical effective rate and adverse events. The recurrence rate of CRA was defined as the proportion of patients who underwent a colonoscopic examination at follow-up and had CRA detected again. The clinical effective rate was based on the standards of the Guiding Principles for the Clinical Research of New TCM.²² The categories were defined as follows: clinical cure-symptoms and signs disappeared; Marked effective-symptoms and signs improved; Effective-symptoms and signs improved; Ineffective-symptoms and signs did not improve or worsened. The clinical effective rate refers to the proportion or percentage of individuals who exhibit improvement or positive outcomes in clinical parameters or symptoms following a specific intervention or treatment. It is calculated as follows:

$$\begin{array}{l}\text{ Clinical effective rate }=\\ \left(\frac{\text{ Clinical cure }+\text{ Marked effective }+\text{ Effective }}{\text{ Total number of patients }}\right)\times 100\mathrm{\%}\end{array}$$

2.5. Exclusion criteria

The exclusion criteria were as follows: (a) non-RCTs and dissertation; (b) studies where both the treatment and control groups were treated with TCM; (c) study subjects who did not meet the diagnostic criteria for CRA; (d) duplicate publications and studies with incomplete data.

2.6. Data extraction and quality assessment

Data extraction and verification were independently carried out by two researchers. Any discrepancies were resolved through mutual consensus. In cases of persistent disagreement, a third researcher made the final decision regarding the inclusion or exclusion of articles with differing viewpoints. Detailed information, including the first author, publication date, total number of cases, intervention measures, treatment duration, primary outcome measures, and adverse events, was extracted. We used the Cochrane Collaboration’s risk of bias tool²³ to assess the risk of bias in each of the included studies. The Grading of Recommendations Assessment, Development and Evaluation (GRADE)²⁴ approach was employed to evaluate the methodological quality of the evidence. Each outcome was evaluated separately, and limitations were assessed in five categories: risk of bias, inconsistency, indirectness, imprecision, and publication bias. Any disagreements were resolved through discussion or by consulting another researcher's opinion.

2.7. Statistical analysis

Statistics and Data Analysis (STATA) 17 (StataCorp, College Station, TX, USA) and Review Manager (RevMan) 5.4 (The Cochrane Collaboration, London, UK) were used for the statistical analysis. Dichotomous outcomes were presented as relative risk (RR), while continuous outcomes were expressed as mean differences (MD) or standardized mean difference (SMD), both with corresponding 95% confidence intervals (CI). Heterogeneity was assessed using both the Q-test and I ² statistic. The I ² statistic ranges from 0% to 100% and was used to evaluate the degree of heterogeneity among multiple study results. A higher value indicated stronger heterogeneity, while a lower value indicated weaker heterogeneity. If I ² < 50%, indicating homogeneity among the studies, a fixed-effects model was employed to calculate the pooled effect size. If I ² ≥ 50%, signifying significant hetero-geneity, a random-effects model was used for the pooled analysis. In cases of substantial heterogeneity, sensitivity analysis or subgroup analysis was performed. If hetero-geneity could not be explained or data sources were unavailable, a descriptive systematic review was conducted. Publication bias was assessed using funnel plots.

3 RESULT

3.1. Search results and study characteristics

Based on our database search, a total of 2022 articles were identified. After excluding 815 duplicate articles, 17 irrelevant articles were further excluded through title and abstract screening. The remaining 50 articles were subjected to full-text reading. A total of 1015 was identified as clinical registration projects, but all were excluded after screening. Ultimately, 18 articles were included. The literature screening process was presented in Figure 1. The basic characteristics of the included studies are detailed in Table 1, and the classification of Chinese herbs is also provided in Table 2.

Figure 1. Flow chart of literature screening.

WOS: Web of science; CNKI: China National Knowledge Infrastructure Database; VIP: China Science and Technology Journal Database; SinoMed: Chinese biomedical literature database. ICTRP: International clinical trials registry platform. ChiCTR: Chinese clinical trial registry. CRA: Colorectal adenoma. RCTs: Randomized controlled trials.

Table 1.

Main characteristics of 18 included trials

Study ID	Participants		RCT design (randomization)	Intervention		Duration (months)	Outcome assessment
	T	C		T	C
Xiong XJ et al 202025	60	60	Random number table method	Jianpi Qushi Fang+PRT	PRT	3	a
Li L et al 202026	32	32	Random number table method	Jianpi Xingqi Tang+PRT	PRT	3	a, c
Niu XL et al 201527	30	30	Randomized	Jianpi Xiaotan Tang+PRT	PRT	12	a, b
Chen QT et al 202028	48	49	Random number table method	Jianpi Lichang Tang+PRT	PRT	2	a, b, c
Zhang R et al 201929	29	29	Random number table method	Jiawei Wuyao Wan+PRT	PRT	3	a, b, c
Wang HM et al 202130	40	40	Admission number randomized	Qingchang Quxi Tang+PRT	PRT	3	a, b
Pan ZM 202131	53	53	Random number table method	Liqi Liujunzi Tang+PRT	PRT	2	a, c
Meng K et al 201632	40	40	Random number table method	Fuzi Lizhong Tang and Xiangsha Liujunzi Tang+PRT	PRT	2	c
Xie JJ 202133	43	43	Random number table method	Jiedu Huoxue Jianpi Fang+PRT	PRT	2	a, c
Zhang BP et al 202034	176	183	Randomized	Tiaochang Xiaoliu Tang+PRT	PRT	3	b
Zhao L et al 202035	53	53	Random number table method	Jianpi Xiaoji Jiedu Tang+PRT	PRT	2	a
Guo HJ et al 202036	43	43	Randomized	Jianpi Xiaoji Qingre Tang+PRT	PRT	2	a, c
Fang MH et al 201837	30	30	Randomized	Jianpi Xiaoji Qingchang Tang+PRT	PRT	2	a, b, c
Xu YP et al 202138	48	48	Random number table method	Jianpi Huazhuo Tang+PRT	PRT	3	a, b, c
Yang JW et al 202239	30	30	Random number table method	Yuchang Tang+PRT	PRT	1.5	c
Yan XJ et al 201840	90	90	Randomized	Zhushao Keli+PRT	PRT	1.5	a
Li Y et al 202141	41	40	Random number table method	Tiaochang Xiaoliu Tang+PRT	PRT	3	b
Xiao J 201842	29	30	Randomized	Wumei Tang+PRT	PRT	3	a

Notes: PRT: postoperative routine treatment; T: treatment group; C: control group; Outcome: a: half a year recurrence rate; b: one-year recurrence rate; c: clinical effective rate; RCT: randomized controlled trial.

Table 2.

Frequencies of usage and distribution in TCM

TCM Category	Chinese herbs	Frequency	Rate (%)
Invigorating spleen and reinforcing Qi	Baizhu (Rhizoma Atractylodis Macrocephalae)	17	7.7
	Gancao (Radix Glycyrrhizae)	16	7.3
	Dangshen (Radix Codonopsis)	14	6.4
	Huangqi (Radix Astragali Mongolici)	8	3.6
Resolving dampness	Fuling (Poria)	12	5.5
	Yiyiren (Semen Coicis)	8	3.6
	Banxia (Rhizoma Pinelliae)	5	2.3
Regulating Qi	Chenpi (Pericarpium Citri Reticulatae)	9	4.1
	Muxiang (Radix Aucklandiae)	5	2.3
Blood activiatingand stasis dissolving	Ezhu (Rhizoma Curcumae Phaeocaulis)	5	2.3
	Sanqi (Radix Notoginseng)	5	2.3
Clearing away heat	Baihuasheshecao (Herba Hedyotdis)	5	2.3
Promoting digestion and relieving stasis	Shanzha (Fructus Crataegus Pinnatifidae)	6	2.7
Astringency of contraction	Wumei (Fructus Mume)	5	2.3

Notes: Chinese herbs mentioned fewer than five times are not displayed in the table. TCM: Traditional Chinese Medicine.

3.2. Risk of bias

Among the 18 included studies, 11 utilized random number tables for grouping,^{25,26,28,29,31,⇓-33,35,38,39,41} 6 mentioned randomization,^{27,34,36,37,40,42} and one employed hospital number randomization.³⁰ Regarding blinding implementation, one study mentioned the use of single-blinding,²⁵ and another mentioned double-blinding.³⁴ One study employed allocation concealment.³⁴ None of the studies had missing or selective reporting. One study had complete data and dropout criteria,²⁴ while 11 studies had complete data but no dropout criteria.^{26,29,30,32,⇓-34,36,⇓,⇓-39,42} The remaining studies mentioned the number of dropouts and exclusions, with one study supplemented dropout cases due to a small sample size,²⁷ primarily attributed to loss to follow-up, low medication compliance, and cases excluded due to incomplete colonoscopy. None of the studies reported other biases. All included studies had comparable baseline characteristics. The summary of the risk was shown in supplementary Figure 1, and the summary of the risk of bias was presented in supplementary Figure 2.

3.3. Quality assessment

Guideline Development Tool (GRADEpro GDT) was used to evaluate the quality of evidence for key outcome indicators. The half-year recurrence rate and the clinical effective rate were rated as medium-quality evidence, while the one-year recurrence rate was rated as low-quality evidence. The main reasons include publication bias and varying degrees of methodological defects in the included literature (Table 3).

Table 3.

GRADE summary of finding

Outcome	No. of studies	Risk of biasa	Inconsistency	Indirectness	Imprecision	Other considerations	RR	95% CI	Quality of the evidence
Half-year recurrence rate	14	Serious	Not serious	Not serious	Not serious	None	0.41	0.33-0.49	Moderate
One-year recurrence rate	8	Serious	Not serious	Not serious	Not serious	Publication bias	0.58	0.49-0.69	Low
Clinical effective rate	10	Serious	Not serious	Not serious	Not serious	None	1.27	1.19-1.36	Moderate

Notes: ^aThere are some shortcomings in the methodology of randomization, blind method and allocation concealment. GRADE: Grading of recommendations assessment development and evaluation. RR: relative risk; CI: confidence interval.

3.4. Meta-analysis results

3.4.1 The recurrence rate of CRA

Half-year recurrence rate: 14 articles reported the half-year recurrence rate as an outcome measure. No heterogeneity was observed among the studies (P = 0.87, I ² = 0%), and a fixed-effects model was used for the pooled analysis (Figure 2). The results indicated a lower recurrence rate in the treatment group compared to the control group [RR = 0.41, 95% CI (0.33, 0.49), Z = 9.08, P < 0.000 01].

Figure 2. Forest plot of half-year recurrence rate.

One-year recurrence rate: 8 articles reported the one-year recurrence rate as an outcome measure. Good homogeneity was observed among the studies (P = 0.14, I ² = 36%), and a fixed-effects model was employed to combine the statistical data (Figure 3). The results indicated that the recurrence rate was lower in the treatment group than in the control group [RR = 0.58, 95% CI = 0.49-0.69, Z = 6.12, P < 0.000 01].

Figure 3. Forest plot of one-year recurrence rate.

3.4.2 Clinical effective rate

The clinical effective rate was reported in ten articles. There was no heterogeneity observed among these studies (P = 0.90, I² = 0%). Due to the homogeneity of the studies, RR was determined using a fixed-effects model (Figure 4). The results indicated that the clinical treatment effectiveness was greater in the treatment group than in control group [RR = 1.27, 95% CI (1.19, 1.36), Z = 7.06, P < 0.000 01].

Figure 4. Forest plot of clinical effective rate.

3.5. Sensitivity analysis

Sensitivity analysis indicated that one study³⁴ of one-year recurrence rates was a potential source of heterogeneity. After excluding this study and re-analyzing the data, the results were [RR = 0.41, 95% CI = 0.30-0.56), Z = 5.44, P < 0.000 01]. Minimal changes were observed compared to the original results, suggesting the robustness of the initial conclusion (supplementary Figures 3A-3C, supplementary Figure 4).

3.6. Publication bias

A funnel plot analysis and Egger's test were conducted to evaluate the clinical effective rate, as well as the half-year and one-year recurrence rates of Jianpi formulas in the postoperative treatment of CRA. If the funnel plot indicates that most studies are located in the upper part of the "inverted funnel" with few studies in the bottom part, and if left-right symmetry is approximately maintained, then there is no apparent publication bias. Conversely, an asymmetrical distribution suggests significant publication bias. The funnel plot for the one-year recurrence rate displayed asymmetry, and Egger’s test (P = 0.001) indicated potential publication bias (supplementary Figures 5C, 5D). However, the funnel plots for the overall the half-year recurrence rate and clinical treatment efficacy exhibited a symmetrical distribution and Egger’s test showed no publication bias (half-year recurrence rate: P = 0.691; clinical treatment efficacy: P = 0.063) (supplementary Figures 5A, 5B, 5E, 5F). This bias may be attributed to the inclusion of studies with low methodological quality and the under reporting of negative trial results.

3.7. Adverse event reporting

Three studies provided information on adverse events. Xu et al ²² reported 2 cases of nausea, one cases of vomiting, and 3 cases of rash in the control group, and one case of nausea, 2 cases of vomiting, and 2 cases of rash in the treatment group. Yang et al ³⁹ reported one case of postoperative bleeding, and 3 cases of abdominal discomfort in the control group. Zhao et al ³⁵ reported 9 cases of abdominal pain, 8 cases of loose stool, 6 cases of a burning sensation of the anus, 12 cases of lassitude and 10 cases of poor appetite in the control group, and 3 cases of abdominal pain, 2 cases of loose stool, 3 cases of a burning sensation of the anus, 3 cases of lassitude, and 4 cases of poor appetite in the treatment group. Based on the adverse events reported in the selected studies, Jianpi formulas appear to be safe and do not demonstrate an increase in adverse reactions compared to the control group. The shedding and withdrawal mentioned in the study were all due to the patient's personal wishes, making follow-up impossible.

4. DISCUSSION

This study aimed to investigate the impact of Jianpi formulas on the recurrence of CRA. All studies were conducted or published before August 31, 2023. A total of 18 studies was included in the systematic review. The main findings of the Meta-analysis were as follows: (a) The efficacy of Jianpi formulas combined with postoperative routine treatment (PRT) was superior to PRT alone; (b) Compared with PRT alone, Jianpi formulas combined with PRT had a better effect on reducing the recurrence rate of CRA; (c) Jianpi formulas did not increase adverse reactions compared to conventional treatment. These findings suggest that Jianpi formulas have beneficial effects in preventing the recurrence of CRA.

Spleen deficiency, damp phlegm and blood stasis are considered the pathogenesis of tumors in TCM. CRC is widely recognized in TCM as having its basis in asthenia in origin and excess in superficiality. Phlegm dampness and stasis belong to excess in superficiality, while spleen deficiency is considered the root cause. Shenlin Baizhu decoction (参苓白术散, SLBZD), known for invigorating the spleen and reinforcing healthy Qi, is often used in combination with chemotherapy for CRC treatment.⁴³ Among the included studies, 6 articles^{25,28,34,36,⇓-38} contained the main Chinese herbs of SLBZD. Study have shown that Shenqinfuzheng injection could alleviate the development of CRC by improving the balance of Th1/Th2 cytokine network and reducing the apoptosis of effector T cells to regulate the immune escape of tumor microenvironment.⁴⁴ The prescriptions adopted by Wang et al ³⁰ and Fang et al ³⁷ were based on the classic and important Jianpi formulas, SLBZD. It was a well-known Chinese herbal medicine prescription for treating cancer during the past decade.⁴⁵ Xu et al ⁴⁶ demonstrated that SLBZD is a valid prescription to control the formation of intestinal adenomas in ApcMin/+ mice and can reduced the number and size of intestinal adenomas. Transforming growth factor beta (TGF-β) signal transduction pathway could induce tumor fibrosis, angiogenesis and immunosuppression. Modified SLBZD could significantly reduce the size of colorectal cancer tumors and the serum content of TGF-β, inhibited CRC cell migration and invasion by limiting TGF-β signaling.⁴⁷ The above studies have confirmed that invigorating spleen can affect colorectal cancer by regulating tumor microenvironment. Jianpi formulas have been considered the fundamental approach to reduce the recurrence rate of CRA and improved patient symptoms.

To analyze the frequency of occurrence of Chinese herbs monomers in the included studies, we found that the most frequently used herbs were Baizhu (Rhizoma Atractylodis Macrocephalae) (17 times), Gancao (Radix Glycyrrhizae) (16 times), Yiyiren (Semen Coicis) (8 times), Dangshen (Radix Codonopsis) (14 times), Fuling (Poria) (12 times), Chenpi (Pericarpium Citri Reticulatae) (9 times), and Huangqi (Radix Astragali Mongolici) (8 times). According to TCM theory, these herbs all could strengthen the spleen or assist in strengthening the spleen. Qin et al ⁴⁸ found that atractylenolide could inhibit the activation of Dynamin-related protein 1 (Drp1) and NOD-like receptor pyrin domain-containing protein 3 (NLRP3) inflammasomes and thus inhibited the formation of colon tumors. Licochalcone A could inhibit the proliferation of colon cancer human colorectal tumor 116 (HCT116) cells, regulate B-cell lymphoma 2 (Bcl2) / Bcl-2-associated X protein (Bax) / Cleavd-caspase3 signaling pathway and promote colon cancer cell apoptosis.⁴⁹ Polysaccharides isolated from Dangshen (Radix Codonopsis) could promote macrophage infiltration, inhibit cancer cell migration, and block and apoptosis of cancer cells in S phase.⁵⁰ Additionally, Fuling (Poria), Chenpi (Pericarpium Citri Reticulatae), Huangqi (Radix Astragali Mongolici), Yiyiren (Semen Coicis) and so on also had antitumor effect.^51,⇓-53 Therefore, a combination of these Chinese herbs with spleen-strengthening effect could theoretically treat CRC.

Most Chinese herbs with spleen-strengthening effect were sweet flavour, but Huanglian (Rhizoma Coptidis) were bitter cold. Huanglian (Rhizoma Coptidis) classified the meridians of stomach, spleen, large intestine, heart, gallbladder and liver in TCM theory. Although Huanglian (Rhizoma Coptidis) presented only 3 times in our statistics, a double-blind, randomized, placebo-controlled clinical trial found that berberine derived from rhizome of Huanglian (Rhizoma Coptidis) reduced the risk of CRA and recurrence of polypoid lesions after polypectomy.⁵⁴ At the same time, it could significantly inhibit the levels of tumor necrosis factor-alpha (TNF-α), interleukin-1 beta (IL-1β) and Interleukin-6 (IL-6) in colon tissue and reduce the pathological damage of colon tissue.⁵⁵ Excessive use of Huanglian (Rhizoma Coptidis) could be easy to damage the spleen and stomach. Therefore, in clinical practice, Huanglian (Rhizoma Coptidis) is often combined with spleen-strengthening TCM as an auxiliary drug of spleen-strengthening decoction. The ultimate goal is to better invigorate the spleen.

Inevitably, our systematic review and Meta-analysis have several limitations. Firstly, all included studies were conducted in mainland China and reported in Chinese. This geographically limited distribution may lead to sampling bias and is challenging to generalize to different populations worldwide. Secondly, the included studies contained a range of methodological flaws that could have introduced bias into the outcomes and affected the results' credibility. Thirdly, because most of the included studies were small studies with small sample sizes, there may be potential reporting bias.

In conclusion, this study indicates that Jianpi formulas has a comparable therapeutic effect on preventing CRA recurrence, significantly reducing the recurrence rate and improving the clinical effective rate after polypectomy. However, due to poor methodological quality in the included trials, further standardized research with multicenter, large-scale, and rigorous design is required.

5. SUPPORTING INFORMATION

Supporting data to this article can be found online at http://journaltcm.com.

Funding Statement

Supported by Jinhua Traditional Chinese Medicine Science and Technology Research Project: Gut Microbiota Distribution Characteristics in Spleen-Deficiency Colorectal Adenoma Patients (No. 2023KR04), Jinhua Traditional Chinese Medicine Science and Technology Major Research Project: Clinical Efficacy and Mechanism of Shugan Jianpi Tiaochang Pills in Preventing Postoperative Recurrence of Colorectal Adenomas based on Metabolomics Technology (No. 2024ZD03)