Effects of curcumin and Curcuma longa extract on inflammatory biomarkers in patients with rheumatoid arthritis (RA) and systemic lupus erythematosus (SLE): a systematic review and meta-analysis of randomized controlled trials

Abstract

Background

Curcumin and Curcuma longa extract have been proposed as adjunct therapies for autoimmune diseases such as rheumatoid arthritis (RA) and systemic lupus erythematosus (SLE) due to their anti-inflammatory and immunomodulatory properties. The purpose of this study was to determine the effects of curcumin and Curcuma longa extract on inflammatory biomarkers in patients with RA and SLE.

Methods

PubMed, EMBASE, and Cochrane CENTRAL databases were searched to May 14, 2025. Randomized controlled trials (RCTs) comparing curcumin or Curcuma longa extract with placebo or standard therapy in patients with RA or SLE were included. The primary outcomes were changes in disease activity score (DAS-28), erythrocyte sedimentation rate (ESR), and C-reactive protein (CRP) level analyzed using standardized mean differences (SMDs) with a random-effects model. Heterogeneity was assessed using the I² statistic, and sensitivity analyses were conducted using a leave-one-out approach.

Results

Seven RCTs involving 314 patients were included. The meta-analysis revealed no significant effect of curcumin or Curcuma longa extract on DAS-28 (pooled SMD = − 2.34, 95% confidence level [CI]: − 5.13 to 0.44; I² = 96.0%), ESR (pooled SMD = − 1.49, 95% CI − 3.71 to 0.72; I² = 91.7%), or CRP (pooled SMD = − 1.65, 95% CI − 3.88 to 0.58; I² = 91.6%). Substantial heterogeneity was observed across all outcomes.

Conclusions

This meta-analysis suggests that curcumin and Curcuma longa extract have limited and inconsistent effects on inflammatory biomarkers in patients with RA and SLE. Larger, well-designed RCTs are needed to clarify their clinical utility as adjunct therapies.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00011-025-02135-z.

Introduction

Autoimmune diseases encompass a heterogeneous group of chronic disorders characterized by dysregulation of immune tolerance, resulting in an aberrant immune response directed against self-antigens. Among these, rheumatoid arthritis (RA) and systemic lupus erythematosus (SLE) are among the most prevalent and clinically significant [1]. RA is a systemic autoimmune condition primarily targeting synovial joints, leading to chronic inflammation, cartilage degradation, and bone erosion. Globally, RA affects approximately 0.5% to 1% of the population, with a higher prevalence among females and an onset typically occurring between the third and fifth decades of life [2]. SLE, in contrast, is a prototypic systemic autoimmune disease that affects multiple organ systems, including the skin, kidneys, cardiovascular system, and central nervous system [3]. The global prevalence of SLE varies between 20 and 150 cases per 100,000 individuals, and similar to RA has a marked female predominance [4]. Both RA and SLE contribute significantly to long-term disability, decreased quality of life, and increased healthcare utilization [5, 6].

The pathogenesis of RA and SLE involves complex interactions between genetic susceptibility, environmental triggers, and immune dysregulation [7, 8]. These factors culminate in the production of autoantibodies and pro-inflammatory cytokines such as tumor necrosis factor-alpha (TNF-α), interleukin-6 (IL-6), and interleukin-1β (IL-1β) [9]. These mediators play central roles in sustaining the inflammatory milieu and driving tissue damage. Although conventional pharmacologic therapies, including corticosteroids, disease-modifying anti-rheumatic drugs (DMARDs), and biologic agents can effectively suppress disease activity, their long-term use is often limited by adverse effects, high cost, and the risk of immunosuppression [10, 11]. Consequently, there is an increasing interest in identifying safer, adjunctive therapies with immunomodulatory and anti-inflammatory properties.

Natural products have been utilized across human history for a wide range of purposes, and clinical studies have progressively demonstrated their therapeutic potential. Among them, Curcuma longa (turmeric) represents a notable example, comprising multiple active constituents, including curcuminoids and sesquiterpenes [12]. Curcumin, the principal curcuminoid derived from the rhizome of Curcuma longa, has been extensively studied for its therapeutic potential in chronic inflammatory and autoimmune conditions [13]. Owing to its poor aqueous solubility and chemical instability, curcumin is commonly extracted with organic solvents, while its stability and bioavailability are often improved through oil-based carriers or nanoparticle formulations [14]. Numerous in vitro and in vivo studies have demonstrated that curcumin exerts anti-inflammatory, antioxidant, and immunomodulatory effects through the inhibition of key signaling pathways such as nuclear factor-kappa B (NF-κB), mitogen-activated protein kinases (MAPKs), and Janus kinase/signal transducers and activators of transcription (JAK/STAT) [15]. Curcuma longa extract, which contains curcumin and other bioactive compounds have similarly demonstrated efficacy in reducing inflammatory responses [16].

Given the immunopathological overlap between the mechanisms of action of curcumin and the inflammatory pathways implicated in RA and SLE, there is a biologically plausible rationale for its therapeutic application in these diseases. Over the past decade, several clinical trials have investigated the impact of curcumin and Curcuma longa extract on inflammatory biomarkers and disease activity in RA and SLE patients. However, the findings have been inconsistent, and comprehensive synthesis of the available data is lacking.

This systematic review and meta-analysis aimed to evaluate the effects of curcumin and Curcuma longa extract on inflammatory biomarkers in patients with RA and SLE, with the goal of determining their usefulness as adjunctive therapies in the management of autoimmune inflammatory diseases.

Methods

Search strategy

This systematic review and meta-analysis adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [17]. A comprehensive literature search of PubMed, EMBASE, and Cochrane CENTRAL was conducted using specific keywords combined with Boolean operators, along with appropriate Medical Subject Headings (MeSH) for studies published up to May 14, 2025.

An example search string was: (“Lupus Erythematosus, Systemic”[Mesh] OR “Systemic Lupus Erythematosus” OR “SLE” OR “Arthritis, Rheumatoid”[Mesh] OR “Rheumatoid Arthritis” OR “RA”) AND (“Curcumin”[Mesh] OR Curcumin OR Curcumas OR Tumeric OR Tumerics OR Turmeric OR Turmerics OR “Curcuma zedoaria” OR “zedoaria, Curcuma” OR “zedoaria, Zedoary” OR “Curcuma longa” OR “longa, Curcuma” OR “Curcuma Longa”).

Additionally, the reference lists of the included studies were searched to identify any other potentially relevant research.

Ethics statement

This systematic review and meta-analysis was conducted using published data and the process did not require raw patient data utilization, nor direct patient involvement. Consequently, the study did require approval from the hospital's institutional review board (IRB) or explicit informed consent from the study subjects.

Selection criteria

This review was conducted following the PICOS criteria, which encompass participants (P), intervention (I), control (C), outcomes (O), and study design (S). Eligible studies included patients with RA or SLE. The intervention was taking curcumin in any form, such as curcumin or Curcuma longa extract, with any dosage and any route of administration. The control criteria were taking placebo, receiving conventional therapy, curcumin-free therapy, or no active treatment. Outcomes of interest were changes in the levels of specific inflammatory markers. Only randomized controlled trials (RCTs) were eligible for inclusion.

However, RCTs which provided no quantitative outcomes of interest and focused on autoimmune disease other than RA or SLE were excluded. In addition, non-human studies and studies published in a language other than English or Chinese were excluded. Eligibility was confirmed by 2 independent reviewers, with a third reviewer consulted if there was uncertainty for inclusion.

Data extraction

Data extracted from the included studies were the name of the first author, year of publication, country the study was performed in, type of autoimmune disease, age, sex, number of participants in the intervention and control groups, duration of the trial, and the type and dosage of the interventions administered to each group.

Outcomes

The primary outcomes were changes in the Disease Activity Score in 28 joints (DAS-28), erythrocyte sedimentation rate (ESR), and C-reactive protein (CRP) level. These outcomes were chosen because each was reported in at least e studies, allowing for meaningful synthesis. Changes from baseline were evaluated using the standardized mean difference (SMD).

Risk of bias assessment

The quality of the included studies was assessed using the Cochrane Collaboration tool for literature review and meta-analyses. It assesses the risk of 7 types of bias: selection bias (random sequence generation and allocation concealment), performance bias (blinding of participants and personnel), detection bias (blinding of outcome assessment), attrition bias (incomplete outcome), reporting bias (selective outcome reporting), and inclusion of intention-to-treat analysis. Quality assessment was performed by 2 independent reviewers, and a third reviewer was consulted if any discrepancies arose.

Statistical analysis

Heterogeneity among the studies was evaluated with the I² statistic as follows: 0 to 24%, low heterogeneity; 25 to 49%, moderate heterogeneity; 50 to 74%, substantial heterogeneity; and 75 to 100%, high heterogeneity. If I² was > 50% a random effects model was used; otherwise, a fixed effects model was used for analysis. Sensitivity analyses were conducted to identify the robustness of the outcomes by using the leave-one-out approach. The significance level was set at 0.05. All statistical analyses were performed using R Studio 4.3.2 with the packages “meta”, “dmetar”, and “metafor”.

Results

Study selection

A total of 15 full-text articles were assessed for eligibility, and 8 were excluded. Thus, 7 studies with a total of 314 patients were included in the meta-analysis [18–24]. A flow diagram of the study selection process is shown in Fig. 1.

Fig. 1.

Flow diagram of the study selection process

Characteristics of the included studies

The study characteristics are summarized in Table 1. Among the included studies, 1 focused on SLE, while the others focused on RA. The mean patient age ranged from 39 to 55 years, and the proportion of male patients ranged from 0.0% to 54%. The trial durations ranged from 8 to 12 weeks. Regarding dosage and treatment duration, curcumin administration across the included trials showed considerable variability. Daily doses ranged from 40 mg to 1,000 mg, and treatment durations spanned from 8 to 12 weeks, corresponding to total curcumin exposures between approximately 3,360 mg (Javadi et al., 2019) and 70,000 mg (Sedighi et al., 2025). Specifically, two studies administered 500 mg/day for 8 weeks (≈ 28,000 mg total), one study used 500 mg/day for 12 weeks (≈ 42,000 mg total), one study used 80 mg/day for 12 weeks (≈ 6,720 mg total), and one study applied 1,000 mg/day for 10 weeks (≈ 70,000 mg total). The wide range of cumulative doses (3.4 g to 70 g) underscores substantial heterogeneity in intervention intensity among trials, which may partly explain the variability in effect estimates across studies. Additional details regarding the solubility, extraction methods, and major active components of Curcuma longa are summarized in Supplementary Table 1 to provide readers with a concise overview of its physicochemical and pharmacological characteristics.

Table 1.

Summary of the included studies

First author (publication year)	Country	Autoimmune disease	Number of patients in each group	Mean patient age, years	Male, %	Trial duration, weeks	Intervention group treatment	Control group treatment
Sedighi [23]	Iran	SLE	Intervention: 30 Control: 32	39.4	6.4	10	Curcumin 1000 mg/day	Placebo
Pourhabibi-Zarandi [21]	Iran	RA	Intervention: 22 Control: 22	50.5	0.0	8	Curcumin 500 mg/day	Placebo
Rezaieyazdi [22]	Iran	RA	Intervention: 32 Control: 29	55.0	9.4	12	Curcumin (nano-micelles) 80 mg/day	Placebo
Pourhabibi-Zarandi [24]	Iran	RA	Intervention: 22 Control: 22	50.5	0.0	8	Curcumin 500 mg/day	Placebo
Javadi [20]	Iran	RA	Intervention: 24 Control: 25	55.0	10.2	12	Curcumin nanomicelle 40 mg/day	Placebo
Amalraj [18]	India	RA	Intervention: 12 Control: 12	38.9	54.2	12	Curcumin 500 mg/day	Placebo
Chandran [19]	India	RA	Intervention: 15 Control: 15	47.9	16.7	8	Curcumin 500 mg/day + diclofenac sodium 50 mg/day	Diclofenac sodium 50 mg/day

Meta-analysis

DAS-28

Five studies were included in the analysis of DAS-28 score [18–22]. Substantial heterogeneity was present (I² = 96%); thus, a random effects model was used. The pooled SMD showed no significant effect (pooled SMD = − 2.34, 95% confidence level [CI] − 5.13 to 0.44) (Fig. 2A).

Fig. 2.

Forest plots of DAS-28, ESR, and CRP level. A Disease activity score. B Erythrocyte sedimentation rate. C C-reactive protein. DAS-28, disease activity score in 28 joints; ESR, erythrocyte sedimentation rate; CRP, C-reactive protein

ESR

Six studies were included in the analysis of the ESR [18–20, 22–24]. Large heterogeneity was present (I² = 91.7%); thus, a random effects model was used. No significant difference between groups was observed (pooled SMD = − 1.49, 95% CI − 3.71 to 0.72) (Fig. 2B).

CRP

Four studies were included in the analysis of CRP level [18, 19, 23, 24]. Large heterogeneity was present (I² = 92%); thus, a random effects model was used. The analysis showed no significant differences between groups (pooled SMD = − 1.65, 95% CI − 3.88 to 0.58) (Fig. 2C).

Sensitivity analyses

Sensitivity analyses were performed using a leave-one-out approach, and the results are summarized in Table 2. For the ESR, the I² decreased to 79% after removing the study by Amalraj et al. [18] (Table 3), with the pooled estimate changing to − 0.30 (− 0.83, 0.24). For the CRP level, the I² value decreased to 0.0% after excluding the study by Amalraj et al. [18], with the new pooled estimate of − 0.34 (− 0.675, 0.004). The I² remained around 94% when other studies were removed (Table 4). However, the pooled estimates remained consistent across analyses, indicating that the leave-one-out sensitivity analysis did not substantially alter the overall findings.

Table 2.

Sensitivity analysis for disease severity score

Leave-one-out	SMD (95% CI)	p-value	tau2	tau	I2
Omitting Pourhabibi-Zarandi [21]	− 1.88 (− 5.21, 1.44)	0.267	11.078	3.328	93.6%
Omitting Rezaieyazdi [22]	− 3.07 (− 6.25, 0.10)	0.058	9.999	3.162	96.3%
Omitting Javadi [20]	− 2.88 (− 6.25, 0.48)	0.093	11.322	3.365	97.0%
Omitting Amalraj [18]	− 1.01 (− 2.79, 0.77)	0.267	3.162	1.778	95.1%
Omitting Chandran [19]	− 2.96 (− 6.26, 0.33)	0.078	10.821	3.290	97.0%
Pooled estimate	− 2.34 (− 5.13, 0.44)	0.099	9.687	3.113	96.0%

Table 3.

Sensitivity analysis for erythrocyte sedimentation rate

Leave-one-out	SMD (95% CI)	p-value	tau2	tau	I2
Omitting Sedighi [23]	− 1.85 (− 4.50, 0.81)	0.173	8.796	2.966	93.1%
Omitting Rezaieyazdi [22]	− 1.89 (− 4.51, 0.73)	0.157	8.556	2.925	92.5%
Omitting Pourhabibi− Zarandi [24]	− 1.54 (− 4.28, 1.20)	0.272	9.418	3.069	91.2%
Omitting Javadi [20]	− 1.78 (− 4.47, 0.92)	0.197	9.087	3.014	93.4%
Omitting Amalraj [18]	− 0.30 (− 0.83, 0.24)	0.276	0.280	0.530	79.5%
Omitting Chandran [19]	− 1.82 (− 4.49, 0.85)	0.181	8.893	2.982	93.3%
Pooled estimate	− 1.49 (− 3.71, 0.72)	0.186	7.338	2.709	91.7%

Table 4.

Sensitivity analysis for C-reactive protein

Leave-one-out	SMD (95% CI)	p-value	tau2	tau	I2
Omitting Sedighi [23]	− 2.22 (− 5.04, 0.61)	0.124	5.877	2.424	93.6%
Omitting Pourhabibi-Zarandi [24]	− 2.07 (− 5.03, 0.89)	0.170	6.473	2.544	94.4%
Omitting Amalraj [18]	− 0.34 (− 0.675, 0.004)	0.053	0.000	0.000	0.0%
Omitting Chandran [19]	− 2.09 (− 5.03, 0.84)	0.162	6.383	2.527	94.4%
Pooled estimate	− 1.65 (− 3.88, 0.58)	0.148	4.901	2.214	91.6%

Quality assessment

Detailed results of the risk of bias assessment for each study are presented in Supplementary Figs. 1A and B. Overall, the included studies demonstrated a generally low risk of bias, particularly in domains such as incomplete outcome data and blinding of participants and personnel; however, a few studies showed unclear or high risk in areas such blinding of outcome assessment, and selective reporting. (Supplementary Fig. 1A and B).

Discussion

This meta-analysis of seven studies—one focusing on SLE and six on RA—found no significant effect of curcumin or Curcuma longa extract on DAS-28, ESR, or CRP levels. Substantial heterogeneity was observed across outcomes, necessitating the use of random-effects models. Sensitivity analyses using a leave-one-out approach confirmed the robustness of the overall findings. Due to the limited sample size, subgroup analyses by disease type, dosage, and formulation could not be thoroughly conducted; however, these factors likely contribute to the observed heterogeneity and warrant further investigation. In summary, this meta-analysis provides the most up-to-date, disease-specific synthesis of curcumin’s effects, integrating recent RCTs to clarify key inflammatory outcomes and thereby strengthen the evidence base for clinical practice and future research.

Recent meta-analyses have examined curcumin in autoimmune diseases with inconsistent results. For example, Ebrahimzadeh et al. [25] and Kou et al. [26] reported significant improvements in inflammatory biomarkers in rheumatoid arthritis, whereas Zeng et al. [27] and Marton et al. [28] found limited or disease-specific benefits. In contrast, our study uniquely focused on rheumatoid arthritis and systemic lupus erythematosus with updated randomized controlled trials up to 2025, demonstrating inconsistent effects across key biomarkers (DAS-28, ESR, and CRP). These findings provide a disease-specific synthesis that refines the current evidence base and highlights the need for standardized, high-quality trials.

Inflammatory states have been implicated as an important mechanism of action in many diseases ranging from cardiovascular disease (CVD) to diabetes, to autoimmune conditions such as SLE and RA [29]. Current drug treatments for many of these conditions can have minimal effectiveness in a large number of patients, as well as side effects that can become intolerable. As a result, there is growing interest in exploring natural compounds such as curcumin (including Curcuma longa extract) as potential supplementary therapies to enhance disease management and reduce inflammatory burden.

Curcumin is one of the most studied natural compounds, and studies have indicated that it may have promise as a treatment or adjuvant treatment for many conditions such as inflammatory diseases [29], various malignancies [30], slowing of the aging process [31], weight loss [32], and diabetes mellitus [33]. However, despite the breadth of research, there is still no clear consensus regarding its clinical effectiveness across these indications.

Meta-analyses and systematic reviews performed over the past 5 years examining the effectiveness of curcumin on various conditions have provided inconsistent results. In a study published in 2021, Ebrahimzadeh et al. [25] performed a systematic review and meta-analysis of the effects of curcumin supplementation on inflammatory biomarkers in patients with RA and ulcerative colitis (UC). Six studies were included in their analysis, and they reported that curcumin supplementation of 250–1500 mg/day over 8–12 weeks was associated with significant decreases in CRP level and ESR in patients with RA and UC. Published in 2022, Zeng et al. [27] performed a systematic review and meta-analysis of 31 RCTs that examined the effectiveness of curcumin and Curcuma longa extract for the treatment of 10 different autoimmune diseases. The overall results showed that curcumin and Curcuma longa extract exhibited effectiveness in the treatment of psoriasis, UC, and RA. On the other hand, neither compound was effective for the treatment of oral lichen planus, and there was insufficient data in the studies to perform a meta-analysis with respect to the effect on Takayasu arteritis, SLE, multiple sclerosis, ankylosing spondylitis, Behcet’s disease, or Crohn’s disease.

In another study published in 2022, Marton et al. [28] performed a systematic review examining the effectiveness of curcumin in the treatment of autoimmune and inflammatory diseases. The authors reported that curcumin was effective for reducing symptoms and/or inducing remission in IBD, OA, SLE, psoriasis, and sclerosis. Notably, curcumin had effects equal to those of ibuprofen and diclofenac without the common adverse effects reported by patients. A meta-analysis published in 2023 specifically examined the effects of curcumin for the treatment of RA [26]. Six studies with a total of 539 patients with RA were included in the analysis. The results showed that curcumin supplementation resulted in significant changes of ESR, DAS-28, swollen joint count, and tender joint count compared to control patients.

Expanding beyond autoimmune conditions, broader evidence has also been synthesized to evaluate curcumin’s impact on general health-related outcomes. A review published in 2024 performed by Jafari et al. [34] examined the effects of curcumin on “human health” by performing a systematic review and meta-analysis of 103 RCTs. There were 42 outcomes reported in the 103 RCTs, and the total number of patients was 7,216. Overall, 55% of the outcomes had statistically significant effect sizes, and the credibility of evidence was rated high for fasting blood sugar, CRP, high-density lipoprotein (HDL), and weight. The credibility of evidence was rated as moderate for outcomes of waist circumference, hip circumference, and BMI. For other outcome measures the quality of evidence was rated as low or very low.

An important consideration of our study, as well as most studies examining the effects of natural compounds on human diseases, in the substantial heterogeneity across outcomes and studies. The heterogeneity limits the strength of pooled estimates and warrants careful interpretation of the results. There are many sources of the heterogeneity such as variations in curcumin formulations used indifferent studies (e.g., bioavailability-enhanced formulations, standardized extracts), which can the pharmacokinetics and clinical response. Dosage and treatment duration also varied considerably in the studies included in our meta-analysis, with curcumin dosages ranging from low to high and intervention periods ranging from 8 to 12 weeks. Differences in study populations can also influence the results: in our study, after applying the selection criteria, only 1 of the included studies was of SLE patients, the others were studies of patients with RA. In addition, there was inconsistent reporting of baseline disease activity or inflammatory marker levels among the studies included.

Implications for clinical practice and future research

While curcumin and Curcuma longa extract appear to be safe and are biologically plausible as adjunct therapies for autoimmune and inflammatory conditions, current clinical evidence does not support their reliable impact on inflammatory biomarkers in RA or SLE. Future RCTs should standardize curcumin formulations, optimize dosing strategies, and include longer follow-up durations. Efforts to stratify patients by disease subtype and baseline inflammatory burden may also help identify subgroups more likely to benefit.

Strengths and limitations

This meta-analysis provides a timely and focused synthesis of current evidence on the effects of curcumin and Curcuma longa extract on inflammatory biomarkers in patients with RA and SLE. Despite its strengths, several limitations should be noted. The small number of included trials (n = 7), with only one addressing SLE, limits the generalizability of findings—particularly to the SLE population. Significant heterogeneity across studies likely stems from variations in study design, formulations, dosages, treatment durations, and baseline disease activity. However, as the overall number of included studies was small, we were unable to perform reliable subgroup analyses—particularly those stratified by dosage and duration—which would have provided additional insights. This underscores the need for future stratified pooled analyses once more original studies become available. Additionally, while one study [18] appeared to influence pooled estimates, sensitivity analyses confirmed the overall robustness of results. Inconsistent biomarker measurement methods and the lack of standardized outcome reporting further challenged data synthesis. Although interleukins (e.g., IL-6, IL-1β) and TNF-α are key inflammatory mediators in autoimmune diseases, they were insufficiently reported among the included trials (only one reported IL-6 and none reported TNF-α). Therefore, a pooled meta-analysis of these cytokines was not feasible. Future studies should standardize biomarker reporting to allow for broader synthesis. Further, the generally short trial durations may not have captured the full therapeutic potential of these interventions. Although no significant benefits were observed for DAS-28, ESR, or CRP, these null findings are valuable in refining clinical expectations and reducing publication bias. Finally, although interleukins and TNF-α are important inflammatory markers that would be expected as outcome measures in studies of curcumin and autoimmune diseases, they were insufficiently reported among the included trials, making a pooled analysis infeasible.

Conclusions

This systematic review and meta-analysis indicates that curcumin and Curcuma longa extract may have limited and inconsistent effects on key inflammatory biomarkers in patients with RA and SLE. Further high-quality, large-scale RCTs are warranted, and updated meta-analyses will be necessary as more robust data become available to better clarify their therapeutic potential.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Ang-Jun Liu led the conceptual design, conducted the investigation, and drafted the initial manuscript. Po-Chang Wu was responsible for data curation, formal analysis, software implementation, and visualization. Yu-Pei Chen contributed to the development of methodology, validation of results, and participated in manuscript review and editing. Hsueh-Ting Chu played a key role in conceptualization, methodological guidance, supervision, and critical manuscript revision. Hen-Hong Chang provided leadership in project administration, resource coordination, funding acquisition, and overall supervision. All authors reviewed and approved the final version of the manuscript.

Funding

This study was supported by the Research and Development Incentive Program of the Affiliated Hospitals, Ministry of Health and Welfare, Taiwan (Grant No. 11336), and Tainan Hospital, Ministry of Health and Welfare.This work was also financially supported by National Science and Technology Council in Taiwan (NSTC 112-2320-B-039-046), National Research Institute of Chinese Medicine, Ministry of Health and Welfare, Taiwan (MOHW114-NRICM-M-124-000002) and China Medical University Hospital DMR-114-041) and the “Chinese Medicine Research Center, China Medical University” from The Featured Areas Research Center Program within the framework of the Higher Education Sprout Project by the Ministry of Education (MOE) in Taiwan (CMRC-CMA-0).

Data availability

All data used were extracted from previously published studies, which are cited within the manuscript.

Declarations

Conflict of interest

The authors declare no competing interests.

Ethical approval

Not applicable.

Patient consent

Not applicable.