Comparison of two different Pilates exercise methods in women with dysmenorrhea: a randomized controlled trial.

This study is a prospective, single-blind, randomized controlled trial with a parallel-group design conducted between January 2025 and June 2025. Ethical approval was obtained from the Non-Interventional Research Ethics Committee of Üsküdar University (61351342/020-742). The study was conducted in accordance with the Declaration of Helsinki, and all participants signed informed consent forms. The protocol is registered with http://clinicaltrials.gov/ (08/01/2025, Clinical Trial, NCT06774521 ). The G*Power (v3.1) program was used to determine the required number of participants. An effect size of 0.90 (large effect, according to Cohen’s d, d ≥ 0.80 for large effect) for the primary outcome, pain intensity (VAS), was used. The significance level was set at α = 0.05, and the test power (1-β) was 0.90, the minimum sample size required for a significant difference between groups was obtained as 42 participants in total, with at least 14 participants in each group. The study population consisted of individuals diagnosed with dysmenorrhea at the Fizyorapt Rehabilitation and Pilates Center in Istanbul 21 . A total of 45 subjects who had volunteered to participate in the study and who met the inclusion criteria were divided into three groups using a simple randomization method. Participants were distributed into three groups: control (no intervention) ( n  = 15), Reformer Pilates ( n  = 15), and Mat Pilates ( n  = 15). The randomization process was executed by an independent researcher using a computer-generated random number table. Throughout the study, the data collector was unaware of the participants’ group assignment. The evaluator determined the patients’ eligibility for the study and also collected the study results. Data were collected at baseline and 12 weeks after randomization. Due to the nature of the interventions, it was not possible to blind the participants or physical therapists. Prior to the commencement of the experiment, a series of baseline measurements were obtained for all three groups. These measurements were obtained using the same methodology after randomization. The Pilates exercises (Group 1: Reformer Pilates exercises; Group 2: Mat Pilates exercises) were administered to the other two groups for a period of 12 weeks, with the exclusion of the control group. Initially, a total of 57 participants were evaluated for their eligibility to participate in the study. The study was concluded with a total of 45 participants. As illustrated in Fig.  1 , the participants’ flow through the study is depicted. The inclusion criteria encompassed individuals between the ages of 18 and 50 years, with menstrual cycles ranging from 3 to 9 days, and who were mobile and generally healthy but suffering from primary dysmenorrhea. The following criteria serve as exclusionary factors: the presence of secondary dysmenorrhea, the utilization of medication, a diagnosis of polycystic ovary syndrome, bleeding disorders, positive pregnancy tests, a history of abdominal surgery, the presence of active infection, or any pathology that would prevent exercise 9 . Fig. 1 Consort flow chart. Consort flow chart. The Reformer and Mat Pilates group participated in a total of 24 Pilates sessions over a period of 12 weeks, with sessions taking place twice weekly. The duration of each session was approximately 60 min. All exercise sessions in both groups were conducted as supervised group sessions. Group sizes were kept similar to ensure comparable levels of instruction and supervision between the Mat and Reformer Pilates groups. The exercise sessions comprised three distinct segments: a warm-up phase (10 min), a main exercise phase (30–40 min), and a cool-down/stretching phase (10 min). The control group was not administered any treatment during the 12-week period; it was used as an untreated comparison group. In the course of designing the exercise program, equivalent objectives were identified for both groups. On the first day of training, the fundamental principles of Pilates were imparted to both groups that encompassed proper breathing techniques, concentration, rib cage position, shoulder alignment, and head and neck position. During the exercises, participants were asked to continue the movements while paying attention to these basic principles. In order to ensure that the exercises were being performed correctly, the physical therapist demonstrated each movement prior to its execution by the patient. Throughout the training process, the physical therapist observed the participants’ movements and made corrections when necessary, using verbal guidance, tactile cues, and visualization techniques. Online Appendix 1 provides detailed information concerning both Pilates exercise programs. Mat Pilates exercises in the warm-up section are designed to prepare the body for exercise through diaphragmatic breathing and pelvic floor activation exercises. In the primary exercise segment, deliberate movements were executed to enhance core muscle activation, refine spinal mobility, and fortify postural equilibrium. As the weeks progressed, the exercises became increasingly challenging and demanding, with the incorporation of resistance materials in certain sessions. The subsequent cool-down and stretching phase were designed to induce relaxation of the muscles, enhance flexibility, and cultivate bodily awareness. In the Reformer Pilates exercises, the difficulty level is balanced by way of various position changes. Progress is attained by increasing the resistance of the springs. In the warm-up section, the emphasis was placed on diaphragmatic breathing and pelvic floor activation, with the muscles being warmed up through the execution of basic movements at a low resistance level. In the primary exercise segment, Reformer exercises are performed to enhance core muscle activation, provide support for pelvic stability, and promote improved spinal mobility. Throughout the 12-week intervention period, the intensity and resistance of the exercises were progressively increased, with TheraBands and medium-resistance springs used as part of the loading protocol. In the subsequent cool-down and stretching phase, stretching movements that induced relaxation in the pelvic and lumbar regions, guided breathing exercises, and controlled postures aimed at increasing spinal mobility were applied. The primary outcomes of this randomized controlled trial were the change in pain intensity and menstrual symptom severity from baseline to week 12. Pain intensity was assessed using the visual analog scale, a validated measure of subjective pain intensity. Menstrual pain and associated symptoms were evaluated using the Menstrual Symptom Scale. Secondary outcomes included fatigue, changes in depressive symptoms, and sleep quality during the 12-week intervention period. Fatigue was measured using the Fatigue Impact Scale, depressive symptoms were assessed using the Beck Depression Inventory, and sleep quality was evaluated using the Pittsburgh Sleep Quality Index. All outcome measures were reported as scale scores (points). All outcome measures were collected at baseline and at the end of the intervention (week 12). Participants were instructed to report dysmenorrhea-related outcomes specifically on the first day of menstruation to standardize assessments. The VAS was selected for its simplicity and reliability in measuring pain intensity, as demonstrated in both clinical and occupational health settings 22 . This scale ranges from zero to ten; zero indicates no pain, while ten indicates the most severe pain the participant has ever experienced. Pain intensity on the first day of the cycle was recorded 23 . The MSS was developed by Chesney and Tasto in 1975 to assess pain and symptoms that occur during menstruation 24 . The scale was adapted into Turkish and its psychometric properties were examined by Güvenç, Seven, and Akyüz in 2014 25 . The scale is structured as a five-point Likert scale, consisting of 22 items. Participants are invited to evaluate the intensity of their menstrual symptoms on a scale ranging from 1 (not at all) to 5 (extremely). Items 1–13 of the scale constitute the subdimension “Negative Effects/Somatic Complaints,” items 14–19 constitute the subdimension “Menstrual Pain Symptoms,” and items 20–22 constitute the subdimension “Coping Methods.” The scale ranges from 22, representing the lowest possible score, to 110, representing the highest possible score. The total score is calculated by taking the average of the scores obtained from all items; an increase in the average score indicates an increase in the severity of menstrual symptoms. Subscale scores are determined by taking the arithmetic mean of the relevant subscale items. An increase in the subscale average indicates that symptoms related to that dimension are being experienced more intensely. In the original form of the scale, Cronbach’s Alpha internal consistency coefficients were reported as 0.90 for “Negative Effects/Somatic Complaints,” 0.81 for “Menstrual Pain Symptoms,” and 0.74 for “Coping Methods.” The BDI is a self-report scale developed by Beck in 1961 to measure emotional, cognitive, somatic, and motivational components 26 . The scale comprises 21 items, with two items addressing emotions, eleven items addressing cognition, two items addressing behavior, five items addressing physical symptoms, and one item addressing interpersonal symptoms. Each question is scored 0, 1, 2, or 3, yielding scores ranging from 0 to 63. The results of the study are interpreted as follows: 0–9 no/minimal depression, 10–18 mild depression, 19–29 moderate depression, and 30–63 severe depression 26 . The validity and reliability study of the BDI, a tool employed to determine the severity of depression, was conducted by Hisli 27 . FIS is a Likert-type assessment tool that requests from participants an evaluation of their fatigue levels over the preceding month, with each item assigned a score ranging from 0 (no problem) to 4 (extreme problem). The scale under consideration comprises a total of 40 items. The FIS comprises three subscales: cognitive (10 items), physical (10 items), and psychosocial (20 items). The cognitive section of the test is designed to assess problems related to concentration, memory, thinking, and organizing thoughts. In contrast, the physical section of the test examines elements such as motivation, effort, endurance, and coordination. The psychosocial section examines the effects of fatigue on social isolation, emotional state, workload, and coping strategies. The total score is obtained by summing all items 28 . Modified versions of this scale are also available in the literature. The scale was adapted into Turkish, and its validity and reliability were established by Armutlu and colleagues 29 . The scale was developed by Buysse et al. 30 to assess sleep quality and disturbance and was adapted into Turkish by Ağargün et al. 31 . The scale comprises a total of 24 questions, including 19 self-report questions and five questions to be answered by a spouse/roommate. The following seven components are covered: The subjective quality of sleep, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, sleep medication use, and daytime dysfunction. Each component is assigned a rating on a scale of 0–3, with the total score ranging from 0 to 21. A total score greater than 5 indicates poor sleep quality 31 . All analyses were performed using the Statistical Package for Social Sciences (SPSS) version 26.0 for Windows software (SPSS, Inc., Chicago, IL, USA). The results were presented as mean ± standard deviation or median (interquartile range (IQR) for quantitative data and as frequency (n) and percentage (%) for categorical data. The normality of the data distribution was assessed by examining the kurtosis and skewness coefficients. The distribution was considered normal when kurtosis was within ± 8 and skewness was within ± 3 32 . For the purpose of conducting comparisons between groups of categorical variables, the chi-square test or Fisher’s Exact test was employed, with the selection of the latter being determined by the expected observation values 33 . Multiple comparisons between percentages were examined using the Bonferroni-corrected Z test. In instances where data exhibited a normal distribution, one-way analysis of variance (ANOVA) was employed for the purpose of conducting group comparisons among the three groups. Depending on the variance homogeneity, either the Tukey HSD or the Tamhane T2 test was applied to examine statistical differences between the two groups as post-hoc comparisons. For normally distributed data with unequal variances, the Welch’s test was applied. Non-normally distributed data were analyzed using the Kruskal–Wallis test, and significant differences were evaluated using the Bonferroni-corrected Dunn test as post-hoc comparisons. Paired sample t-tests were used to make pre-test-post-test comparisons within groups. The statistical significance was set at p  < 0.05.

A total of 45 participants completed the study, including 15 Mat Pilates participants, 15 Reformer Pilates participants and 15 control group participants. Their socio-demographic characteristics are shown in Table 1 . Table 1 Participants’ socio-demographic characteristics. Variables Mat pilates group ( n  = 15) Reformer pilates group ( n  = 15) Control group ( n  = 15) p Age, Mean (SD) 29.33 ± 6.02 28.87 ± 6.81 28.60 ± 5.29 0.080 a BMI (kg/m 2 ), Mean (SD) 22.25 ± 3.14 24.48 ± 4.79 23.51 ± 4.29 0.340 a Marital status, n (%)  Married 10 (66.7) 10 (66.7) 11 (73.3) 0.850 b  Single 5 (33.3) 5 (33.3) 4 (26.7) Educational status, n (%)  High school 0 (0) 1 (6.7) 1 (6.7) 1.000 b  University 15 (100) 14 (93.3) 14 (93.3) Superscript letters indicate statistical tests used for group comparisons: a: one-way analysis of variance (ANOVA); b: Fisher’s exact test. SD, standard deviation; n, number of subjects; BMI, body mass index; kg, kilogram; m, meter; %, percent. Bold values represent p  < 0.05. Participants’ socio-demographic characteristics. Superscript letters indicate statistical tests used for group comparisons: a: one-way analysis of variance (ANOVA); b: Fisher’s exact test. SD, standard deviation; n, number of subjects; BMI, body mass index; kg, kilogram; m, meter; %, percent. Bold values represent p  < 0.05. Table 2 shows the changes in VAS, BDI, and PSQI scores within and between the groups before and after the intervention. No significant differences were observed in VAS ( p  = 0.248), BDI ( p  = 0.250), and PSQI ( p  = 0.670) scores at baseline ( p  > 0.05). However, post-intervention assessments revealed significant differences between the groups in VAS ( p  < 0.001), BDI ( p  < 0.001), and PSQI ( p  = 0.001) scores. Table 2 Visual analog scale, beck depression scale, and Pittsburgh sleep quality index measurements. Variables Mat Pilates group ( n  = 15) Reformer Pilates group ( n  = 15) Control group ( n  = 15) Between-group p Pairwise comparisons p (AI) Mean ± SD Mean ± SD Mean ± SD VAS BI 5.93 ± 2.34 6.00 ± 2.48 6.87 ± 1.06 0.248 b AI 4.93 ± 1.22 2.33 ± 0.90 6.20 ± 1.47 < 0.001 a C-M: 0.019 C-R: <  0.001 M-R: <  0.001 Within-group p 0.051 < 0.001 0.086 BDI BI 26.40 ± 10.67 22.00 ± 13.85 25.53 ± 11.36 0.250 a AI 22.60 ± 8.52 11.40 ± 6.59 26.60 ± 10.49 < 0.001 a C-M: 0.003 C-R: <  0.001  M-R: 0.003 Within-group p 0.001 0.003 0.206 PSQI BI 12.13 ± 2.80 15.53 ± 1.51 11.93 ± 1.94 0.670 a AI 12.67 ± 1.29 13.73 ± 1.44 11.87 ± 1.19 0.001 a C-M: 0.227 C-R: 0.001  M-R: 0.077 Within-group p 0.334 < 0.001 0.827 Superscript letters indicate statistical tests used for group comparisons a: one-way analysis of variance (ANOVA), b: Welch’s test. SD: Standard Deviation, n: number of subjects, BI: Before Intervention, AI: After Intervention. Bold values represent p  < 0.05. Pairwise comparisons were performed using Tukey’s HSD test for post hoc analysis (AI). C: Control group; M: Mat Pilates group; R: Reformer Pilates group. C-M, C-R, and M-R indicate pairwise group comparisons. All outcome measures were reported as scale scores (points). Visual analog scale, beck depression scale, and Pittsburgh sleep quality index measurements. C-M: 0.019 C-R: <  0.001 M-R: <  0.001 C-M: 0.003 C-R: <  0.001  M-R: 0.003 Superscript letters indicate statistical tests used for group comparisons a: one-way analysis of variance (ANOVA), b: Welch’s test. SD: Standard Deviation, n: number of subjects, BI: Before Intervention, AI: After Intervention. Bold values represent p  < 0.05. Pairwise comparisons were performed using Tukey’s HSD test for post hoc analysis (AI). C: Control group; M: Mat Pilates group; R: Reformer Pilates group. C-M, C-R, and M-R indicate pairwise group comparisons. All outcome measures were reported as scale scores (points). Within-group analyses demonstrated significant decreases in BDI scores in the Mat Pilates group ( p  = 0.001), whereas changes in PSQI ( p  = 0.334) and VAS ( p  = 0.051) scores were not significant. In the Reformer Pilates group, significant decreases were observed in BDI ( p  = 0.003), PSQI scores ( p  < 0.001) and VAS ( p  < 0.001) scores from baseline to post-intervention. No significant changes were found in the Control group. Figure  2 shows the changes in the VAS, BDI, and PSQI scores before and after the intervention (values are expressed as mean ± SD or median [IQR]). Statistically significant differences were observed between the groups for VAS ( p  = 0.002), BDI ( p  = 0.012), and PSQI ( p  < 0.001) scores. Fig. 2 ( A, B, C ) Comparison of 12-week changes in visual analog scale (VAS) ( A ), beck depression inventory (BDI) ( B ), and Pittsburgh sleep quality index (PSQI) ( C ) before and after the 12-week study period in the control, Mat Pilates, and Reformer Pilates groups. ( A ) Bars show Mean ± SD. Pairwise p values derived from Tamhane T2. Colored columns represent group means; vertical error bars represent standard deviations (SD). Because the mean 12-week change values are negative, the columns extend below the zero line. These figures are bar charts and do not contain box and whisker plot elements. ( B ) Box and whisker plots show median and IQR; points represent individual participants. Pairwise p values were derived from Bonferroni-corrected Dunn tests. ( C ) Box and whisker plots show median and IQR; points represent individual participants. Pairwise p values were derived from Bonferroni-corrected Dunn tests. ( A, B, C ) Comparison of 12-week changes in visual analog scale (VAS) ( A ), beck depression inventory (BDI) ( B ), and Pittsburgh sleep quality index (PSQI) ( C ) before and after the 12-week study period in the control, Mat Pilates, and Reformer Pilates groups. ( A ) Bars show Mean ± SD. Pairwise p values derived from Tamhane T2. Colored columns represent group means; vertical error bars represent standard deviations (SD). Because the mean 12-week change values are negative, the columns extend below the zero line. These figures are bar charts and do not contain box and whisker plot elements. ( B ) Box and whisker plots show median and IQR; points represent individual participants. Pairwise p values were derived from Bonferroni-corrected Dunn tests. ( C ) Box and whisker plots show median and IQR; points represent individual participants. Pairwise p values were derived from Bonferroni-corrected Dunn tests. As illustrated in Table 3 , a comparative analysis of the groups’ FIS scores before and after the intervention is presented. No significant differences were found in the pre-intervention FIS scores ( p  > 0.05). Post-intervention assessments revealed significant difference between the groups in the Physical ( p  = 0.001), Cognitive ( p  < 0.001), and Social Fatigue Effect ( p  = 0.001) subscales, as well as in the FIS Total Score ( p  = 0.001). Table 3 Findings regarding fatigue impact scale. Variables Mat Pilates group ( n  = 15) Reformer Pilates group ( n  = 15) Control group ( n  = 15) Between-group p Pairwise comparisons p (AI) FIS-subparameters Mean ± SD Mean ± SD Mean ± SD FIS-Physical BI 24.80 ± 7.13 23.00 ± 9.61 26.53 ± 7.82 0.509 a AI 21.73 ± 5.19 16.33 ± 4.25 25.73 ± 7.71 0.001 b C-M: 0.291 C-R: 0.001 M-R: 0.013 Within-group p 0.008 0.006 0.976 FIS-Cognitive BI 24.33 ± 6.83 22.40 ± 9.49 26.33 ± 7.37 0.410 a AI 23.20 ± 5.49 16.27 ± 4.30 25.07 ± 7.71 < 0.001 b C-M: 0.836 C-R: 0.003 M-R: 0.002 Within-group p 0.021 0.010 0.906 FIS-Social BI 48.40 ± 14.74 45.60 ± 18.39 52.07 ± 15.08 0.551 a AI 44.27 ± 11.76 33.67 ± 9.49 51.13 ± 14.49 0.001 a C-M: 0.276 C-R: <  0.001 M-R: 0.053 Within-group p 0.001 0.011 0.974 FIS-Total BI 97.53 ± 28.26 91.00 ± 37.27 104.93 ± 29.72 0.496 a AI 89.20 ± 21.45 66.27 ± 17.43 101.93 ± 29.17 0.001 b C-M: 0.459 C-R: 0.001 M-R: 0.010 Within-group p 0.002 0.009 0.964 Superscript letters indicate statistical tests used for group comparisons a: one-way analysis of variance (ANOVA); b: Welch’s test. SD: Standard Deviation, n: number of subjects, BI: Before Intervention, AI: After Intervention. Bold values represent p  < 0.05. Pairwise comparisons were performed using Tamhane’s T2 test for FIS-Physical, FIS-Cognitive, and FIS-Total, and Tukey’s HSD test for FIS-Social (AI). C: Control group; M: Mat Pilates group; R: Reformer Pilates group. C-M, C-R, and M-R indicate pairwise group comparisons. All outcome measures were reported as scale scores (points). Findings regarding fatigue impact scale. C-M: 0.291 C-R: 0.001 M-R: 0.013 C-M: 0.836 C-R: 0.003 M-R: 0.002 C-M: 0.276 C-R: <  0.001 M-R: 0.053 C-M: 0.459 C-R: 0.001 M-R: 0.010 Superscript letters indicate statistical tests used for group comparisons a: one-way analysis of variance (ANOVA); b: Welch’s test. SD: Standard Deviation, n: number of subjects, BI: Before Intervention, AI: After Intervention. Bold values represent p  < 0.05. Pairwise comparisons were performed using Tamhane’s T2 test for FIS-Physical, FIS-Cognitive, and FIS-Total, and Tukey’s HSD test for FIS-Social (AI). C: Control group; M: Mat Pilates group; R: Reformer Pilates group. C-M, C-R, and M-R indicate pairwise group comparisons. All outcome measures were reported as scale scores (points). Within-group analyses demonstrated significant decrease in the Mat Pilates group in Physical ( p  = 0.008), Cognitive ( p  = 0.021), and Social Fatigue Effect ( p  = 0.001) subscales, as well as in the FIS Total Score ( p  = 0.002). In the Reformer Pilates group, significant decrease was observed in Physical ( p  = 0.006), Cognitive ( p  = 0.010), and Social Fatigue Effect ( p  = 0.011) scores, as well as in the FIS Total Score ( p  = 0.009). No significant changes were observed in any FIS scores of the Control group ( p  > 0.05). Figure  3 shows the changes in the FIS scores before and after the intervention (values are expressed as mean ± SD or median [IQR]). Statistically significant differences were observed between the groups in Physical Fatigue ( p  = 0.012), Social Fatigue ( p  = 0.033), and FIS Total Score ( p  = 0.027). No significant differences were found between the groups in Cognitive Fatigue scores ( p  = 0.088). Fig. 3 ( A, B, C ) Comparison of 12-week changes in fatigue impact scale (FIS)-Physical ( A ), fatigue impact scale (FIS)-social ( B ), and fatigue impact scale (FIS)-total ( C ) before and after the 12-week study period in the control, mat Pilates, and reformer Pilates groups. ( A ) Bars show Mean ± SD. Pairwise p values derived from Tamhane T2. Colored columns represent group means; vertical error bars represent standard deviations (SD). Because the mean 12-week change values are negative, the columns extend below the zero line. These figures are bar charts and do not contain box and whisker plot elements. ( B ) Box and whisker plots show median and IQR; points represent individual participants. Pairwise p values were derived from Bonferroni-corrected Dunn tests. ( C ) Bars show Mean ± SD. Pairwise p values derived from Tamhane T2. Colored columns represent group means; vertical error bars represent standard deviations (SD). Because the mean 12-week change values are negative, the columns extend below the zero line. These figures are bar charts and do not contain box and whisker plot elements. ( A, B, C ) Comparison of 12-week changes in fatigue impact scale (FIS)-Physical ( A ), fatigue impact scale (FIS)-social ( B ), and fatigue impact scale (FIS)-total ( C ) before and after the 12-week study period in the control, mat Pilates, and reformer Pilates groups. ( A ) Bars show Mean ± SD. Pairwise p values derived from Tamhane T2. Colored columns represent group means; vertical error bars represent standard deviations (SD). Because the mean 12-week change values are negative, the columns extend below the zero line. These figures are bar charts and do not contain box and whisker plot elements. ( B ) Box and whisker plots show median and IQR; points represent individual participants. Pairwise p values were derived from Bonferroni-corrected Dunn tests. ( C ) Bars show Mean ± SD. Pairwise p values derived from Tamhane T2. Colored columns represent group means; vertical error bars represent standard deviations (SD). Because the mean 12-week change values are negative, the columns extend below the zero line. These figures are bar charts and do not contain box and whisker plot elements. Table 4 presents the comparison results of the groups’ MSS scores before and after the intervention. In the pre-intervention assessments, statistically significant differences were identified in the Negative Effects/Somatic Complaints ( p  = 0.017) scores. No significant differences were identified in the other MSÖ subscales and total scores ( p  > 0.05). In the post-intervention assessments, significant differences were identified between the groups in Negative Effects/Somatic Complaints ( p  < 0.001) and Menstrual Symptom Total Score ( p  = 0.001). No statistically significant differences were identified between the groups with regard to menstrual pain symptoms and coping methods scores ( p  > 0.05). Within-group analyses demonstrated significant decreases in the Mat Pilates group in Negative Effects/Somatic Complaints ( p  = 0.007), Coping Methods ( p  = 0.017), and MSS Total Score ( p  = 0.025). In the Reformer Pilates group, significant decreases were observed in Negative Effects/Somatic Complaints ( p  < 0.001), Menstrual Pain Symptoms ( p  = 0.001), Coping Methods ( p  = 0.007), and MSS Total Score ( p  < 0.001). No significant changes were observed in the Control group. Table 4 Findings regarding menstrual symptom scale. Variables Mat Pilates group ( n  = 15) Reformer Pilates group ( n  = 15) Control group ( n  = 15) Between-group p Pairwise comparisons p (AI) MSS- subparameters Mean ± SD Mean ± SD Mean ± SD Negative effects/somatic complaints BI 34.07 ± 11.51 38.80 ± 10.09 37.20 ± 14.54 0.017 a C-M: 0.014 C-R: 0.155 M-R: 0.542 AI 28.47 ± 11.36 25.00 ± 7.98 45.73 ± 13.28 < 0.001 a C-M: <  0.001 C-R: <  0.001  M-R: 0.671 Within-group p 0.007 < 0.001 0.109 Menstrual pain symptoms BI 16.67 ± 11.75 21.00 ± 5.99 18.40 ± 8.44 0.425 a AI 15.00 ± 14.92 13.93 ± 5.82 16.80 ± 7.17 0.738 a Within-group p 0.504 0.001 0.054 Coping methods BI 7.20 ± 4.66 8.80 ± 4.81 7.53 ± 4.55 0.489 a AI 6.13 ± 3.64 5.27 ± 3.33 7.20 ± 4.21 0.375 a Within-group p 0.017 0.007 0.136 MSS-total BI 57.93 ± 22.25 68.60 ± 17.70 63.13 ± 19.50 0.112 a AI 49.60 ± 23.41 44.20 ± 14.10 61.73 ± 15.87 0.001 a C-M: 0.012 C-R: 0.001 M-R: 0.699 Within-group p 0.025 < 0.001 0.730 Superscript letters indicate statistical tests used for group comparisons a: one-way analysis of variance (ANOVA). SD: Standard Deviation, n: number of subjects, BI: Before Intervention, AI: After Intervention. Bold values represent p  < 0.05. Pairwise comparisons were performed using Tukey’s HSD test for Negative Effects/Somatic Complaints (BI and AI) and MSS-Total (AI). C: Control group; M: Mat Pilates group; R: Reformer Pilates group. C-M, C-R, and M-R indicate pairwise group comparisons. All outcome measures were reported as scale scores (points). Findings regarding menstrual symptom scale. C-M: 0.014 C-R: 0.155 M-R: 0.542 C-M: <  0.001 C-R: <  0.001  M-R: 0.671 C-M: 0.012 C-R: 0.001 M-R: 0.699 Superscript letters indicate statistical tests used for group comparisons a: one-way analysis of variance (ANOVA). SD: Standard Deviation, n: number of subjects, BI: Before Intervention, AI: After Intervention. Bold values represent p  < 0.05. Pairwise comparisons were performed using Tukey’s HSD test for Negative Effects/Somatic Complaints (BI and AI) and MSS-Total (AI). C: Control group; M: Mat Pilates group; R: Reformer Pilates group. C-M, C-R, and M-R indicate pairwise group comparisons. All outcome measures were reported as scale scores (points). Figure  4 shows the changes in the MSS scores before and after the intervention (values are expressed as mean ± SD or median [IQR]). Statistically significant differences were observed between the groups in Negative Effects/Somatic Complaints ( p  < 0.001), Coping Methods ( p  = 0.044), and MSS Total Score ( p  < 0.001). No significant differences were found between the groups in Menstrual Pain Symptoms ( p  = 0.050). Fig. 4 A, B, C Comparison of 12-week changes in negative effects/somatic complaints ( A ), coping methods ( B ), and menstrual symptom scale (MSS)-total ( C ) before and after the 12-week study period in the control, mat Pilates, and reformer Pilates groups. ( A ) Bars show Mean ± SD. Pairwise p values derived from Tamhane T2. Colored columns represent group means; vertical error bars represent standard deviations (SD). Because the mean 12-week change values are negative, the columns extend below the zero line. These figures are bar charts and do not contain box and whisker plot elements. ( B ) Box and whisker plots show median and IQR; points represent individual participants. Pairwise p values were derived from Bonferroni-corrected Dunn tests. ( C ) Bars show Mean ± SD. Pairwise p values derived from Tamhane T2. Colored columns represent group means; vertical error bars represent standard deviations (SD). Because the mean 12-week change values are negative, the columns extend below the zero line. These figures are bar charts and do not contain box and whisker plot elements. A, B, C Comparison of 12-week changes in negative effects/somatic complaints ( A ), coping methods ( B ), and menstrual symptom scale (MSS)-total ( C ) before and after the 12-week study period in the control, mat Pilates, and reformer Pilates groups. ( A ) Bars show Mean ± SD. Pairwise p values derived from Tamhane T2. Colored columns represent group means; vertical error bars represent standard deviations (SD). Because the mean 12-week change values are negative, the columns extend below the zero line. These figures are bar charts and do not contain box and whisker plot elements. ( B ) Box and whisker plots show median and IQR; points represent individual participants. Pairwise p values were derived from Bonferroni-corrected Dunn tests. ( C ) Bars show Mean ± SD. Pairwise p values derived from Tamhane T2. Colored columns represent group means; vertical error bars represent standard deviations (SD). Because the mean 12-week change values are negative, the columns extend below the zero line. These figures are bar charts and do not contain box and whisker plot elements.

This study compared the effects of Mat and Reformer Pilates on pain, menstrual symptoms, depression, fatigue, and sleep quality in women with dysmenorrhea. Both methods resulted in reductions in depression, fatigue, menstrual pain, and related symptoms. Reformer Pilates produced greater improvements across all outcomes, including pain intensity, sleep quality, and overall symptom reduction, compared with Mat Pilates. Few studies have examined the effects of Pilates on women with dysmenorrhea. Evidence shows reductions in menstrual discomfort and symptom severity following Pilates 16 , 34 , likely related to increased pelvic blood flow and metabolism 35 . Pilates has been shown to effectively reduce pain and other symptoms of primary dysmenorrhea and improve quality of life, supporting its role as a viable management option 36 . A systematic review reported that Pilates produced the greatest improvement in dysmenorrhea-related pain tests compared to aerobic dance, yoga, running, and brisk walking 8 . Consistent with these findings, another study noted significant decreases in pain intensity after Pilates 37 . In our study, similar benefits were observed, with reductions in depression, fatigue, menstrual pain, and symptoms in both Pilates groups. These outcomes may result from improved physical function and relaxation, which help reduce pain perception 38 . Additionally, core Pilates principles; centering, concentration, control, and conscious breathing; may have further supported physical and mental well-being 9 , 39 . Research comparing Mat and Reformer Pilates in women with dysmenorrhea is limited. However, studies in other populations show differing effects between these modalities. In individuals with chronic low back pain, Reformer Pilates led to faster clinical improvement than Mat Pilates 40 . Similarly, in fibromyalgia patients, Reformer Pilates improved clinical status and muscle strength, while Mat Pilates effectively reduced pain, enhanced clinical status, improved biopsychosocial well-being, and increased physical quality of life 41 . In the present study, both Mat and Reformer Pilates reduced menstrual pain and symptoms. Pilates has been shown to enhance posture awareness and concentration 42 , which may decrease pain perception by improving mechanoreceptor signaling to the central nervous system 43 . Reformer Pilates produced greater improvements in pain intensity, menstrual discomfort, and related symptoms, potentially due to its spring-based resistance system, which demands greater balance, coordination, and muscle activation. This enhanced neuromuscular control may further reduce pain and improve mind–body awareness 40 supporting the idea that Reformer Pilates may offer superior symptomatic relief. Wang et al. reported a link between poor sleep quality and increased risk of primary dysmenorrhea 44 . Studies have shown that individuals participating in Reformer Pilates report improved sleep quality 16 , 42 , and a recent meta-analysis confirmed that Pilates reduces PSQI scores across various populations, including postmenopausal, middle-aged, postpartum women, and hemodialysis patients 45 . In the present study, sleep quality improved only in the Reformer Pilates group. Although prior research is not specific to dysmenorrhea, these findings support the potential of Reformer Pilates to enhance sleep quality in different populations. Anxiety, depression, and stress are known to influence primary dysmenorrhea 44 . Prior research has shown that both Mat and Reformer Pilates can reduce depression and fatigue symptoms 6 , a finding consistent with the results of the present study. The deep breathing and rhythmic movements of Pilates may regulate the central nervous system and increase endorphin levels, promoting relaxation and reducing depressive symptoms 46 . The breathing techniques fundamental to Pilates may also reduce fatigue by enhancing respiratory control and activating the parasympathetic nervous system 47 , which has been reported to positively affect physical and mental fatigue 47 . In this study, Reformer Pilates produced greater improvements in depression and fatigue than Mat Pilates. The spring-based equipment in Reformer Pilates enables a wider range of controlled movements and may enhance motivation and exercise adherence. Increased enjoyment and engagement could contribute to reduced fatigue, depression, and improved sleep through psychological relaxation. Individualizing exercise strategies is essential for effectively managing menstrual symptoms. This approach offers clinicians and researchers a practical framework for selecting the most suitable Pilates method and planning treatment. A key strength of this study is that it is the first to directly compare Reformer and Mat Pilates in women with dysmenorrhea, making a meaningful contribution to the existing literature. However, this study has several limitations. The evaluator was involved in both participant eligibility assessment and outcome data collection, which may introduce detection bias. To mitigate this risk, all outcomes were collected using validated and standardized instruments, and participants reported outcomes independently on the first day of menstruation. Statistical analyses were performed blinded to group allocation. Additionally, the limited exercise variety in the Mat Pilates group, small sample size, and short intervention period may have influenced outcomes and restrict generalizability. No direct physiological measures, such as heart rate or energy expenditure, were conducted to compare the exercise intensity of Mat Pilates and Reformer Pilates, which should be considered a further limitation. The use of a non-intervention control group also limits comparisons with conventional exercise programs. Future studies should include larger samples, longer intervention periods, follow-up assessments to evaluate long-term effects, and comparisons with other exercise modalities. Despite these limitations, the findings provide important preliminary evidence on the effects of Mat and Reformer Pilates in women with dysmenorrhea.

In conclusion, both Mat and Reformer Pilates have been effective in reducing depression, fatigue, menstrual pain, and related symptoms. Reformer Pilates has been associated with more significant improvements in pain intensity, sleep quality, depression, fatigue, and menstrual discomfort. Exercise selection should be made considering the patient’s preference and access to equipment; when Reformer Pilates is not available, Mat Pilates remains a practical alternative in clinical settings. This protocol appears suitable for women with dysmenorrhea, and larger-scale studies are encouraged to validate these findings.

Dysmenorrhea is a condition characterized by the presence of intense cramping pain in the lower abdomen during menstruation, a condition that is particularly common in young women and that has been demonstrated to have a negative impact on quality of life 1 . Dysmenorrhea is classified into two primary categories: primary, which occurs in the absence of pelvic pathology, and secondary, which is associated with pelvic diseases such as endometriosis and uterine fibroids 2 . Primary dysmenorrhea, characterized by menstrual cramps, occurs in the absence of pelvic pathology and is more prevalent among adolescents 1 , 3 . Risk factors for this condition include biological factors, such as a family history of dysmenorrhea and early menarche; lack of physical activity; sleep disorders; and psychological conditions, including stress and depression 1 , 3 , 4 . The prevalence of primary dysmenorrhea is 71% in young women and 93% in adolescents, and these individuals often experience severe symptoms 5 , 6 . A substantial body of research has demonstrated that primary dysmenorrhea can result in various mental health complications, including but not limited to depression, anxiety, and post-traumatic stress disorder (PTSD). Furthermore, this condition can lead to limitations in daily activities, school and work absenteeism, and broader economic problems 6 , 7 . Consequently, there is an imperative for safer and more economical treatment methods to alleviate dysmenorrhea 8 . Although nonsteroidal anti-inflammatory drugs (NSAIDs) are widely used, their use is limited by adverse effects such as headaches, indigestion, and dizziness. Consequently, there is a growing interest in non-pharmacological methods, such as exercise-based approaches, to mitigate adverse effects associated with drug use 9 . Pilates, an exercise method that emphasizes the development of core muscle strength, flexibility, and body awareness, has gained popularity as a holistic approach to promoting overall health and well-being 10 . This mind–body exercise system contributes to maintaining physical balance by emphasizing breath control, core stability, and proper posture 11 . It has been posited that periods of vigorous physical exertion may stimulate the release of beta-endorphins, which possess inherent analgesic properties. This phenomenon could potentially mitigate the discomfort associated with primary dysmenorrhea 12 . In recent years, there has been a marked increase in the number of individuals practicing Pilates, and this method has also been found to be effective in reducing different types of pain, such as back and neck pain 13 , 14 . However, the number of studies investigating the effects of Pilates on dysmenorrhea is quite limited 15 . The positive effects of Pilates on dysmenorrhea may be mediated through improvements in sleep quality and psychological state 16 . A recent study reported that 12 weeks of Pilates training reduced depression and anxiety and improved sleep quality in postmenopausal women 17 . In addition, Pilates can be adapted to different needs. It aims to maintain the spine’s natural alignment and improve body awareness. Its widespread use in physical therapy programs is attributable to its positive effects on strength, flexibility, and pain control 18 . Joseph Pilates, a renowned authority in the field of physical rehabilitation, advocated for the execution of movements in a horizontal plane to mitigate joint stress. Building upon this principle, Pilates developed a piece of equipment known as the “Reformer,” which was designed to facilitate the execution of these movements. This apparatus, comprising springs, ropes, and a sliding platform, facilitates resistance exercises while mitigating the effects of gravity 19 . Furthermore, Joseph Pilates expanded the scope of the method by developing various pieces of equipment, including the Reformer, Cadillac, Ladder Barrel, and Step Chair 20 . At present, the majority of the Pilates repertoire consists of Mat and Reformer exercises, and these two applications form the basis of the system. Reformer and other Pilates equipment offer increased exercise variety by providing different application possibilities. To the best of our knowledge, there are no studies that have investigated whether this variety provides additional benefits in terms of pain intensity, menstrual pain and symptoms, depression, fatigue, and sleep quality in women with dysmenorrhea. Consequently, there is a compelling need for randomized controlled trials to investigate the effects of different Pilates methods in women with dysmenorrhea. The objective of this study is to systematically analyze and compare the effects of Mat Pilates and Reformer Pilates methods on pain intensity, menstrual pain and symptoms, depression, fatigue, and sleep quality in women with dysmenorrhea. The hypothesis of our study is that both Pilates methods will reduce symptoms associated with dysmenorrhea, but due to its equipment-supported structure, the Reformer Pilates method will provide more pronounced improvements compared to Mat Pilates.

Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1