This study enrolled middle and high school athletes receiving sub-special training, such as wrestling, soft tennis, and soccer. Participants who received regular training for more than 20 h per week for more than 1 year were included. Participants who were regularly taking supplements, such as turmeric, vitamins, and zinc, were excluded. Participants who were allergic to curcumin or felt uncomfortable with curcumin supplements were also excluded. We also excluded athletes who cannot receive regular training due to sports injuries to minimize bias from training levels. After enrollment, the participants were provided the option of joining either the curcumin group or the control group. The participants in the curcumin group received the curcumin supplement, whereas those in the control group did not. All participants performed exercise training in the Chiayi County Yung Ching Senior High School and recorded the amount of daily exercise training in an exercise training log. We explained the study project to the participants and their guardians and obtained a signed informed consent form. This study conformed to the tenets of the Declaration of Helsinki for medical research involving human subjects and was approved by the Institutional Review Board of Chang Gung Memorial Hospital (IRB No. 202100069A3C601).

The participants’ demographics, including age (years), height (cm), weight (kg), sports training experience (years), details of training undergone, nutritional supplement status, and medical history, were recorded. In addition, all participants underwent body composition, blood, urine, fitness, muscle fatigue, and muscle soreness assessments at the baseline and end of this 12-week study. No exercise training was scheduled on the test day.

The participants in the curcumin group received a Chinese curcumin supplement named Jiang Huang Powder Ko Da (KO DA Pharmaceutical Co., Ltd., Taoyuan, Taiwan) after being evaluated by a Chinese medical practitioner. Each pack contains 500 mg of Jiang Huang Powder, which is made by 300 mg of curcumin, 190 mg of starch, and 10 mg of sodium carboxymethyl cellulose. The participants received a 1.5-g sachet a day for 90 consecutive days; therefore, the curcumin intake of each participant was 1,200 mg per day.

We measured the participants’ body composition by using a body composition analyzer (InBody 720, Biospace, Seoul, Republic of Korea), which simultaneously records body weight, fat-free mass, body fat mass, and basal metabolic rate (BMR) (28).

We determined the serum TNF-α and CK levels to monitor the inflammatory index and muscle damage. In addition, we assessed the participants’ oxidative stress index by determining the levels of 8-hydroxy-2 deoxyguanosine (8-OHdG) and malondialdehyde (MDA) in urine.

Spot urine samples were collected and stored below –80°C until the analysis. Urinary MDA was quantified using a thiobarbituric acid reactive substances assay kit (Oxford Biomedical Research Inc., Rochester Hills, MI, USA. Product codes: FR40). Urinary 8-OHdG was quantified using an ultraperformance liquid chromatography–tandem mass spectrometer (Waters Xevo TQ-XS, Milford, MA, USA. Product codes: WAT058951). Urinary creatinine (CRE) concentration was used for urinary MDA and 8-OHdG concentration adjustments.

For determining the levels of CK and TNF-α, 5 mL of blood samples were collected into a serum separation tube through cubital fossa venipuncture in the dominant arm. Serum TNF-α was measured using chemiluminescent immunoassay on a Siemens IMMULITE 1000 system (Siemens Healthcare Diagnostics Inc., Camberley, Surrey, UK. Product Codes: LKNF1). In addition, serum CK activity was determined using the enzymatic rata method with a Hitachi LST008 (Hitachi, Japan) and a test kit (Cica LaboFit CK, Japan).

The concentration of CK, TNF-α, and MDA was determined at the Department of Laboratory Medicine, Chang Gung Memorial Hospital, Chiayi. The 8-OHdG quantification was performed at the Department of Laboratory Medicine, Chang Gung Memorial Hospital, Linkou. Both laboratories are accredited to ISO 15189 by Taiwan Accreditation Foundation.

The participants’ fitness level was evaluated using the HELMAS physical fitness management system (O2RUN CO., LTD., Seoul, Republic of Korea) at our institution. Several dimensions of health-related fitness, including muscular strength (grip and back strength), balance (closed-eyes foot balance), flexibility (sitting trunk flexion and trunk extension), muscle endurance (sit-ups), power (vertical jump), and agility (reaction times and side steps), were evaluated. The participants warmed up for 5 min before the test. To minimize the influence of fatigue, the participants rested for 1 min between each test. Each test was performed twice (with a minimum interval of 30 s), and the higher performance was selected as a measurement of individual physical fitness (29).

We used the visual analog scale to assess the participants’ muscle fatigue and muscle soreness status, with 0 and 10 cm denoting no fatigue or soreness and maximum fatigue or soreness, respectively (30). The participants filled out the assessment form daily after the training.

The effect size (Cohen’s d) was 1.13, as determined using G*Power 3.1.9.7 (Heinrich Heine University, Düsseldorf, Germany). The input parameters used for the two-tailed t test were alpha = 0.05, power = 0.8, the sample size of the control group was 15, and the sample size of the curcumin group was 13.

All data were analyzed using SPSS 26.0 (SPSS Inc., Chicago, IL, USA). Means (standard deviation) were calculated for the participants’ parameters. The Mann–Whitney U test was used to compare the parameters between the groups. The Wilcoxon test was used to compare the baseline and 12-week parameters in the same group. In addition, the change in each item was calculated using the following formula:

The percentage change between baseline and 12th week was compared in both groups. We used the chi-square test to compare the percentage change in parameters across the study period between the control and curcumin groups; p < 0.05 was considered significant.

Female athletes had a lower fat-free mass and BMR than the males in both groups (Supplementary Table 1). In the control group, no significant change in body composition was observed after 12 weeks. However, in the curcumin group, a significant increase in fat-free mass (p = 0.003) and BMR (p = 0.001) was observed after 12 weeks (Table 1). Furthermore, the significant increase in BMR was constant through box gender, whereas increase of fat-free mass was only significant in males (Supplementary Table 1).

We assessed the levels of CK, MDA, 8-OHdG, and TNF-α (Table 2). In the curcumin group, a significant reduction in the urinary level of 8-OHdG was observed after 12 weeks, with a decrease in mean value from 4.79 to 3.86 ng/mg CRE (p < 0.036). However, no significant changes were noted in the levels of CK, MDA, and TNF-α (all p > 0.05). Furthermore, the subgroup analysis according to gender did not find the significant difference of change as above (Supplementary Table 2).

In the curcumin group, the muscle fatigue score and muscle soreness score decreased significantly from 6 ± 1 to 7 ± 2, respectively, at baseline to 4 ± 2 and 4 ± 2, respectively, after 12 weeks of curcumin supplementation (all p = 0.005; Table 2). These significant changes were still noted in males, but not in females (Supplementary Table 2). A significant difference in reaction time, muscle fatigue score, and muscle soreness score was observed between the groups. The mean percentage change in reaction time of the control and curcumin groups was 21 and -14%, respectively (both p < 0.05; Table 3). Furthermore, significant differences in percentage change in muscle fatigue score and muscle soreness score were observed between the groups (p = 0.007 and p = 0.001, respectively, Table 3). Similarly, these significant changes were still noted in males, but not in females (Supplementary Table 3).

We used the chi-square test to compare the percentage change in parameters before and after supplementation between the control and curcumin groups (Figure 2). A significant difference in percentage change (decrease) in BMR, muscle fatigue score, and muscle soreness score (p = 0.044, 0.004, and 0.022, respectively) was observed between the groups.

In this study, some strengths may be highlighted, such as the first study investigating the effect of curcumin on postexercise recovery in adolescent athletes and a more extended intervention period (12 weeks) than ever. However, this study has several limitations. First, because of not a randomized, double-blind study with limited samples, inherent biases may exist. Although the baseline characteristics of the participants were similar, the participants were provided the option of joining either the curcumin group or the control group, thereby leading to a selection bias. Moreover, the participants and assessors were not blinded to the treatment, contributing to performance bias and detection bias. Limited samples jeopardized the interpretation of subgroup analyses by gender. These biases reduced the certainty of the results. Second, we only checked some inflammatory markers; therefore, the effect of curcumin on other inflammatory markers, such as IL-6, IL-8, nuclear factor kappa B, LDH, and lactate, in adolescent athletes could not be determined. Third, we did not develop a standardized protocol for sports training but continued the training the players were already undergoing. Although the inclusion of different sports-based exercises could have influenced the interpretation of the results, we used the exercise training log that recorded parameters, including the session rating of perceived exertion, to monitor the training load, which may somewhat limit this bias (36). Fourth, we did not standardize the participants’ nutritional intake or record each participant’s diet. Some diets containing natural curcumin, such as yellow mustard and curry, might potentially interfere with the results of this study. To eliminate the possible bias from unbalanced nutritional intake and diet, an athlete should be requested to design a diet tailored to their situation and record them in future investigations.