Effect of curcumin supplement or placebo in delayed onset muscle soreness: a systematic review and meta-analysis

Abstract

Background

There are multiple strategies that have been suggested to attenuate delayed onset muscle soreness (DOMS). Curcumin has been shown to reduce exercise-induced oxidative stress (OS) and inflammation. However, currently, there is still controversy.

Main body of the abstract

We conduct this meta-analysis according to the PRISMA guidelines. Relevant studies were included from Medline and Scopus from the date of inception to May 04th, 2021 that reported VAS score, blood markers (creatinine kinese (CK), tumor necrotic factor (TNF)-α and interleukin (IL)-6) and range of motion of either group. There were total of 13 studies including 202 and 176 persons in curcumin and placebo group. The unstandardized mean difference (UMD) of VAS muscle soreness in post-exercise, 1, 2, 3 and 4 days was − 0.12 (95% CI − 0.46, 0.22), − 0.38 (− 0.83, 0.08), − 0.67 (− 1.19, − 0.16), − 0.86 (− 1.38, − 0.34), − 0.81 (− 1.27, − 036) and − 1.24 (− 1.50, − 0.99) scores lower in curcumin when compared to placebo. The UMD of CK was − 11.07 (95% CI − 24, 1.86), − 37.51 (− 68.04, − 6.97), − 45.40 (− 95.67, 4.86), − 53.33 (− 128.11, 21.45), − 90.98 (− 173.45, − 8.51) and 117.84 (− 338.69, 574.37) lower in curcumin when compared to placebo. No statistically significantly differences were noted for IL-6, TNF-α and ROM between two groups.

Short conclusion

This meta-analysis suggested that curcumin supplement reduced delayed onset muscle soreness and CK after exercise in 1, 2, 3, and 4 days when compared to placebo. However, TNF and IL were not affected by curcumin ingestion.

Level of evidence I.

Background

Delayed onset muscle soreness (DOMS) normally occurs approximately 1 to 2 days after the unaccustomed activity and eccentric muscle contraction, with a symptom of muscle soreness and discomfort (Kim and So 2019; Mizumura and Taguchi 2016; Lewis et al. 2012). Normally, DOMS is one symptom of exercise-induced muscle damage (EIMD) (Howatson and Someren 2008). Researchers have proposed several hypotheses for the etiology of DOMS previously such as lactic acid, muscle spasms, muscle damage, and inflammatory response, whereas novel hypotheses include Toll-like receptor 4 activation, increased levels of neurotrophic factors such as nerve growth factors and glial cell line-derived neurotrophic factors; however, the exact cause of DOMS remains unclear (Urai et al. 2013; Ota et al. 2013; Cheung et al. 2003; Smith 1991; Yoon et al. 2020). Delayed onset muscle soreness (DOMS) is a combination of muscle pain and stiffness occurring several hours after unaccustomed exercise, particularly when eccentric muscle activity is involved, it can induce muscle damage (Tanabe et al. 2019; Clarkson et al. 1992). An inflammatory response and the reactive oxygen species (ROS) production were triggered by this mechanical stress. The reason behind this process is that the mechanical stress promotes the activation of transcription factors such as nuclear factor-κB (NF-κB), and it would result in limiting an athlete’s performance and daily activities (García-López et al. 2007). EIMD can be manifested by prolonged decline in muscle strength, reduction in range of motion (ROM), swelling, DOMS and an increase in blood muscle proteins, including creatine kinase (CK) activity (Clarkson et al. 1992; García-López et al. 2007; Tiidus 1998; Warren et al. 1999; Takahashi et al. 1994). Muscle function impairment caused by damaged and its subsequent inflammatory responses may reduce the ability to perform daily training and reduce athletic performance. Therefore, post-exercise muscle damage should be prevented or minimized (Tanabe et al. 2019). At present, multiple strategies have been treated to attenuate DOMS such as hyperbaric oxygen, anti-inflammatory medication, massage, cryotherapy, homeopathy, stretching, electrotherapeutic modalities, ultrasound, rest, and light exercise. Nutritional interventions with antioxidant and anti-inflammatory properties are frequently reported to reduce DOMS.

Curcumin is a natural polyphenolic substance extracted from turmeric. It has various physiological effects, such as membrane protective effects, as well as anti‐inflammatory and antioxidant responses (Thapa et al. 2013; Hatcher et al. 2008). These mechanisms have been reported to suppress the activity of NF‐κB, thereby suppressing the expression of IL-6 and tumor necrotic factor TNF-α (Cho et al. 2007; Aggarwal et al. 2006), which is a modulating factor for cytokines and cyclooxygenase (COX) (Thapa et al. 2013; Hatcher et al. 2008; Cho et al. 2007; Aggarwal et al. 2006; Chun et al. 2003). Therefore, these actions are considered to have a positive effect on DOMS. However, outcomes (muscle soreness, CK, TNF-α, IL-6 and ROM) of the curcumin and placebo are inconsistent. Some studies show benefit with intake of curcumin supplement (Tanabe et al. 2019, 2015; Hillman et al. 2021), while other studies do not (Nicol et al. 2015; McFarlin et al. 2016; Drobnic et al. 2014; Mallard and Briskey 2020; Jäger et al. 2019; Ms et al. 2020). A previous meta-analysis (Fang and Nasir 2021) found that the efficacy of curcumin in reducing CK serum levels and muscle soreness index among adults, however, it included a study that combined curcumin and piperine supplement and compared to placebo, which should be excluded. Moreover, the previous meta-analyses did not assess other relevant outcomes (TNF-α, IL-6 and ROM), importance sources of heterogeneity (e.g., preoperative exercise, age, and dosage of curcumin supplement) with subgroup analysis and publication biases were not assessed. Furthermore, some of the other randomized controlled trial studies (RCTs) (Hillman et al. 2021; Mallard and Briskey 2020; Amalraj et al. 2020; Cardaci et al. 2020) that have since been published were not included. Therefore, this meta-analysis of curcumin supplement versus placebo was conducted. This study aims to update the comparisons between curcumin supplement and placebo of muscle soreness and CK and adds comparisons of TNF-α, IL-6 and ROM.

Main text

Search strategy and data sources

This review was conducted according to the transparent reporting of systematic reviews and meta-analyses (PRISMA guideline 2009). The search was performed in two databases: PubMed and Scopus; from the date of inception to May 04^th, 2021. The keywords were used as following search terms: ((Delayed onset muscle soreness) OR (DOMS)) AND (curcumin)). Manual search for reference lists of all included studies was screened for further eligible articles identification. The studies were screened independently by two authors (J.M. and P.K.) against the eligibility criteria based on titles and abstracts using bibliographical software package, EndNote version X7. Disagreements were resolved regarding inclusion and exclusion criteria of a study with a third author (J.K.)

Inclusion criteria

Studies were included if they met the following criteria: (a) RCT and quasi-RCT studies; (b) studies that reported outcomes as following: muscle soreness index (VAS), creatinine kinase (CK); tumor necrotic factor-alpha (TNF-α), interleukin-6 (IL-6) and range of motion (ROM); (c) studies that compared clinical outcomes between curcumin supplement and placebo; (d) There had adequate data for extracted and pooled. We excluded studies if they were using a combination of intervention besides curcumin supplement, experimental and animal studies, reviews, letters to editor, or case reports and non-English languages studies.

Data extraction and methodology quality assessment

The data were extracted from each study through structured data extraction forms by two reviewers (N.R. and P.P.), using data extraction forms. The items extracted were baseline characteristics of the study that included average age, sex, study design, mean follow-up time and dose of curcumin. Clinical outcomes data (number of subjects, means, and SD of VAS and CK) between groups were extracted and followed by data extraction of frequencies (adverse effect) between treatment groups. In case of any disagreements in opinion, a third author (J.M.) decision would be required. Quality assessment was performed by two authors (N.R. and P.P.) according to the Cochrane Collaboration tool for evaluating the risk of bias in order to avoid the distortion of the meta-analysis outcomes (Liberati et al. 2009). RCT studies were assessed by risk of bias followed by the PRISMA guideline recommendation based on sequence generation, allocation concealment, blinding, incomplete outcome data, selective outcome reporting, and other sources of bias. Any conflicts between reviewers for quality assessment were settled by third reviewer (J.K.).

Outcomes of interest

The outcomes were VAS of muscle soreness, CK, TNF-α, IL-6 and ROM. The measurement of those outcomes was the same as reported in the original studies, which were VAS of muscle soreness (0–10); lower values are equivalent to better outcomes. For CK, TNF-α, IL-6, lower values are equivalent to better outcomes. For ROM, higher values are equivalent to better outcomes.

Statistical methods

For continuous data, data were pooled as unstandardized mean difference (UMD) with 95% confidence interval (CI). The heterogeneity across the studies was assessed using Q statistic and I2 statistic for quantifying the degree of heterogeneity. I² value equal to 0% is considered as no, 25% as low, 50% as moderate and 75% as high heterogeneity. The statistical significance for heterogeneity was set with a P value < 0.10. A random-effect model was used if I² > 25, otherwise a fixed effects model was applied. In order to explore the cause of heterogeneity, meta-regression was applied in the meta-regression model. According to the results of meta-regression, sensitivity analyses were performed by leave-one-out to assess the robustness of a pooled conclusion. The funnel plots and Egger test were used to assess the publication bias (Palmer et al. 2020; Egger et al. 1997). The metatrim and fill method was used to estimate the number of studies that might be missing and to adjust the pooled estimate (Duval and Tweedie 2000). Data were analyzed using STATA version 15.0 (StataCorp. 2017). P value < 0.05 was considered statistically significant. P value of < 0.10 was considered as a threshold of significant heterogeneity.

Results

We identified 16 and 399 studies from Medline and Scopus, respectively (Fig. 1); 15 studies were duplicated, leaving 400 studies to review titles and abstracts. Of these, 12 full papers and a study, which was identified from the reference list, were reviewed and leaving a total of 13 studies for data extraction. Characteristics of the 13 studies (Tanabe et al. 2019, 2015; Hillman et al. 2021; Nicol et al. 2015; McFarlin et al. 2016; Drobnic et al. 2014; Mallard and Briskey 2020; Jäger et al. 2019; Ms et al. 2020; Amalraj et al. 2020; Cardaci et al. 2020; Nakhostin-Roohi et al. 2016) are described in Table 1. All studies were RCTs. Nine studies were parallel design, while four studies were crossover design. All 13 studies (Tanabe et al. 2019, 2015; Hillman et al. 2021; Nicol et al. 2015; McFarlin et al. 2016; Mallard and Briskey 2020; Jäger et al. 2019; Ms et al. 2020; Amalraj et al. 2020; Cardaci et al. 2020; Nakhostin-Roohi et al. 2016) were reported of post-exercise muscle soreness using VAS. Indirect markers of muscle damage were reported using CK in 11 studies (Tanabe et al. 2019, 2015; Hillman et al. 2021; Nicol et al. 2015; McFarlin et al. 2016; Drobnic et al. 2014; Mallard and Briskey 2020; Ms et al. 2020; Amalraj et al. 2020; Nakhostin-Roohi et al. 2016), TNF-α in six studies (Tanabe et al. 2019, 2015; Nicol et al. 2015; McFarlin et al. 2016; Mallard and Briskey 2020; Ms et al. 2020) and IL-6 in five studies (Tanabe et al. 2019, 2015; Nicol et al. 2015; McFarlin et al. 2016; Mallard and Briskey 2020). Post-exercise ROM was reported in three studies (Tanabe et al. 2019, 2015). Mean age and BMI of participants varied from 19.5 to 36 years and 22 to 26.2, respectively. Eight studies reported percentages of male gender 100 percent. While four studies reported percentages of male gender from 36 to 77 percent. Eight studies were active exercise, while five studies were resistive exercise.

Fig. 1

Flow of study selection

Table 1 Characteristics of included studies

Risk of bias in included studies

Risk of bias is described in Table 2. Among 13 studies, no selective outcome reporting and incomplete outcome report, followed by blinding (9/13), sequence generation (8/13) and allocation concealment (8/13).

Table 2 Risk of bias assessment

Outcomes

VAS of muscle soreness at pre- and post-exercise, 24, 48, 72 and 96 h

The mean values of VAS of muscle soreness between curcumin and placebo in post-exercise person at pre- and post-exercise, 24, 48, 72 and 96 h are shown in Table 3 and Fig. 2. The pooled UMD was − 0.12 (95% CI − 0.46, 0.22), − 0.38 (− 0.83, 0.08), − 0.67 (− 1.19, − 0.16), − 0.86 (− 1.38, − 0.34), − 0.81 (− 1.27, − 036) and − 1.24 (− 1.50, − 0.99), i.e., mean VAS muscle soreness was approximately − 0.7, − 0.9, − 0.8 and − 1.2 score statistically significant difference between two groups at 24, 48, 72 and 96 h (Table 3; Fig. 2). Separately fitting preoperative exercise, age, sex, and body mass index and dosage of curcumin supplement at baseline in a meta-regression analysis and none of the co-variables could explain the heterogeneity. Egger’s test and a contour funnel plot did not suggest any evidence of publication bias.

Table 3 Comparisons of VAS between curcumin supplement and placebo

Fig. 2

Forest plot for comparison of VAS score at pre-, post-, 24, 48, 72 and 96 h after exercise between curcumin supplement versus placebo

CK at pre- and post-exercise, 24, 48, 72 and 96 h

The mean values of CK between curcumin and placebo in post-exercise person at pre- and post-exercise, 24, 48, 72 and 96 h are shown in Table 4 and Fig. 3. The UMD was − 11.07 (95% CI − 24, 1.86), − 37.51 (− 68.04, − 6.97), − 45.40 (− 95.67, 4.86), − 53.33 (− 128.11, 21.45), − 90.98 (− 173.45, − 8.51) and 117.84 (− 338.69, 574.37) U/L, i.e., mean CK was approximately − 38 and − 91 U/L statistically significant difference between two groups at post-exercise and 72 h (Table 4; Fig. 3). None of the co-variables could explain the heterogeneity. Egger’s test and a contour funnel plot did not suggest any evidence of publication bias.

Table 4 Comparisons of CK between curcumin supplement and placebo

Fig. 3

Forest plot for comparison of CK score at pre-, post-, 24, 48, 72 and 96 h after exercise between curcumin supplement versus placebo

TNF-α at pre- and post-exercise, 24, 48, 72 and 96 h

The mean values of TNF-α between curcumin and placebo in post-exercise person at pre- and post-exercise, 24, 48, 72 and 96 h are shown in Table 5 and Fig. 4. The UMD was − 0.04 (95% CI − 0.00, 0.08), 0.01 (− 0.11, 0.13), − 0.40 (− 0.85, 0.06), 0.49 (− 0.33, 1.30), − 0.08 (− 0.47, 0.31) and − 0.51 (− 1.52, 0.50) pg/mL, i.e., mean TNF-α was not significant difference between two groups at post any time point. Egger’s test and a contour funnel plot did not suggest any evidence of publication bias.

Table 5 Comparisons of TNF between curcumin supplement and placebo

Fig. 4

Forest plot for comparison of TNF score at pre-, post-, 24, 48, 72 and 96 h after exercise between curcumin supplement versus placebo

IL-6 at pre- and post-exercise, 24, 48, 72 and 96 h

The mean values of IL-6 between curcumin and placebo in post-exercise person at preexercise, 24, 48, 72 and 96 h are shown in Table 6 and Fig. 5. UMD was − 0.03 (95% CI − 0.35, 0.32), − 0.46 (− 1.12, 0.02), − 0.49 (− 1.52, 0.54), − 0.29 (− 0.72, 0.14) and − 1.00 (− 1.83, − 0.16) pg/mL, i.e., mean IL-6 was approximately − 1 pg/mL statistically significant difference between two groups at 96 h. None of the co-variables could explain the heterogeneity. Egger’s test and a contour funnel plot did not suggest any evidence of publication bias.

Table 6 Comparisons of IL between curcumin supplement and placebo

Fig. 5

Forest plot for comparison of IL score at pre-, 24, 48, 72 and 96 h after exercise between curcumin supplement versus placebo

ROM at pre- and post-exercise, 24, 48, 72 and 96 h

The mean values of ROM between curcumin and placebo in post-exercise (eccentric elbow flexion) person at pre- and post-exercise, 24, 48, 72 and 96 h are shown in Table 7 and Fig. 6. The UMD was 1.18 (95% CI − 1.30, 3.67), − 0.30 (− 3.71, 3.11), 2.30 (− 3.78, 8.37), 4.57 (− 4.85, 14.00), 9.84 (− 0.02, 19.70) and 8.45 (− 0.84, 17.74) pg/mL, i.e., mean ROM had no statistically significant difference between two groups. Egger’s test and a contour funnel plot did not suggest any evidence of publication bias.

Table 7 Comparisons of ROM between curcumin supplement and placebo

Fig. 6

Forest plot for comparison of ROM score at pre-, post-, 24, 48, 72 and 96 h after exercise between curcumin supplement versus placebo

Discussion

This review suggested that persons who took curcumin supplement before exercise have pain score of about 1 score lower than those who took placebo in and post-exercise at 1, 2, 3 and 4 days. For indirect markers of muscle damage, the persons who took curcumin supplement before exercise have lower CK, TNF and IL score than those who took placebo. However, TNF and IL have no statistically significant difference between those two groups. In terms of ROM, no difference between the two groups in all follow-up time was shown. After all outcomes were pooled and the sources of heterogeneity were explored by meta-regression analysis, the differences of all co-variables were shown to have no effect on all outcomes.

From a review of previous meta-analysis (Fang and Nasir 2021), curcumin has efficacy in reducing CK serum levels and muscle soreness index among adults. Therefore, curcumin may be known as a priority EIMD recovery agent in interventions. However, previous meta-analyses have limitations as following: Included one study (Delecroix et al. 2017) that reported a combination of curcumin and piperine, which is not equal to curcumin alone; did not explore the outcomes such as IL, TNF and ROM of curcumin when compared to placebo.

This study has several strengths. First of all, this study included 13 studies in the pooling of all important clinical outcomes of curcumin supplement and placebo. Secondly, we explore the possible causes of heterogeneity, when covariate data at baseline are available. Publication bias for each outcome was assessed. However, there are some limitations in this study. The scope of this study does not include other important outcomes such as adverse effects and cost-effective analysis, since there are incomplete data. Finally, only English publications were considered in this study. At present, more than one supplement is consumed before exercise to reduce pain after playing sports (EIMD or DOMS). This may be more effective than using one alone, but there is no comparative research yet. In the future, there should be more RCT studies to responding to playing sports effectively. Moreover, further RCT studies that assess combine effect of cost-effective analysis should be performed.

Conclusion

To conclude, curcumin supplement has reduced muscle soreness and CK after exercise after 1, 2, 3, and 4 days when compared to placebo. On the other hand, TNF and IL were not affected by curcumin ingestion. Further RCT studies that assess cost-effective analysis should be performed.