The sample size for this study was determined during the research design phase using G*Power 3.1.3 software. A one-way analysis of variance (ANOVA) was planned to compare differences among three groups (mat Pilates, reformer Pilates, and control). Statistical parameters were set with an effect size (f) of 0.25 (medium), a significance level (α) of 0.05, and a power (1-β) of 0.80, resulting in a minimum required sample of 57 participants (19 per group). However, due to the exploratory nature of the study and practical constraints related to recruitment and program administration, the final sample consisted of 21 participants, with 7 assigned to each group. Although the reduced sample size may limit generalizability, the study serves as a foundational investigation, providing pilot data and assessing the feasibility of a larger-scale trial. Participants were recruited from among residents of Seoul, South Korea. Inclusion criteria were healthy pregnant women in their second trimester with no known musculoskeletal, cardiovascular, or metabolic disorders, and who had received medical clearance for physical activity. All participants were provided with a detailed explanation of the study’s objectives and procedures and voluntarily submitted written informed consent. Although this study was not reviewed by an institutional review board, all procedures were conducted in adherence to established ethical standards for research involving human subjects, including voluntary participation, informed consent, and the protection of personal information. The physical characteristics of participants in each group (mat Pilates, reformer Pilates, and control) are presented in <Table 1>.

This study was conducted from February to October 2023. Participants were randomly assigned to one of three groups: the mat Pilates group, the reformer Pilates group, or the control group. Prior to the intervention, all participants underwent baseline assessments to evaluate calf circumference, perceived stress using the Perceived Stress Scale (PSS), and antenatal depression using the Edinburgh Postnatal Depression Scale (EPDS). These pre-tests were administered to establish participants’ initial physical and psychological status. The intervention program was implemented twice a week over a total period of eight weeks. Each session lasted 50 minutes and consisted of a 10-minute warm-up, a 30-minute main exercise phase, and a 10-minute cool-down. Exercise intensity was adjusted based on the subjective Rate of Perceived Exertion (RPE), taking into account the physical condition and safety of the pregnant participants. Sessions during weeks 1 to 4 were conducted at an RPE level of 11-13, and sessions during weeks 5 to 8 at an RPE level of 13-15. The exercise protocols were specifically designed for pregnant women, considering their physiological characteristics and safety requirements. The mat Pilates and reformer Pilates programs were structured to have equivalent duration and frequency, and each group followed the assigned program accordingly. Upon completion of the 8-week intervention, post-tests were conducted using the same instruments and procedures as in the pre-test phase. The pre- and post-intervention data were compared to assess intraand inter-group changes. The control group did not receive any exercise intervention and was instructed to maintain their regular daily activities. All experimental conditions were carried out in a controlled environment to minimize the influence of external variables <Figure 1>.

Participants were seated upright on a chair with a backrest, maintaining an erect posture with their feet flat on the floor and knees flexed at a 90° angle. To objectively assess lower extremity edema, calf circumference was measured at the point of maximum girth on the posterior aspect of the lower leg—specifically, the most prominent region over the medial and lateral heads of the gastrocnemius muscle. A non-elastic tape measure was gently wrapped around this site, taking care not to compress the soft tissue. Measurements were recorded in centimeters (cm) and performed twice under identical conditions, with the average of the two values used for analysis [20]. To minimize the influence of environmental variables, all assessments were conducted in a quiet, temperature-controlled room (22-24°C) at a consistent time of day (between 9:00 a.m. and 11:00 a.m.). Prior to measurement, participants rested in a seated position for at least 10 minutes to stabilize circulatory dynamics and reduce the potential effects of recent physical activity or orthostatic fluid shifts. All assessments were performed by the same trained examiner using standardized instructions to ensure measurement reliability and consistency [21].

The instrument used to assess prenatal stress in this study was originally developed in Korean by Ahn Hwang-ran [22] and later revised by Cho Jae-aeng and Kim Young-hee [23], who validated the tool through psychometric testing among Korean pregnant women. The final version consists of 27 items designed to evaluate stress related to pregnancy across three subdomains: fetus-related factors, self-related factors, and partner-related factors. Each item is rated on a 5-point Likert scale ranging from 1 (“Not at all concerned”) to 5 (“Always concerned”), yielding a total possible score between 27 and 135. Higher scores indicate greater levels of perceived prenatal stress. The internal consistency of this scale, as measured by Cronbach’s alpha, was reported to be α = 0.90 [24].

Prenatal depressive symptoms were assessed using the Edinburgh Postnatal Depression Scale (EPDS), originally developed by Cox, Holden, and Sagovsky [25], and adapted into Korean by Han Gui-won, Kim Myung-jung, and Park Je-min through a translation and back-translation procedure followed by a validation study [26]. The scale comprises 10 items, each scored on a 4-point Likert scale. Response options are scored from 0 to 3, with reverse scoring applied to items 1, 2, and 4. Total scores range from 0 to 30, with higher scores indicating more severe depressive symptoms. The reliability of the Korean version of the EPDS was reported with a Cronbach’s alpha of α = 0.85.

This study was conducted on pregnant women in their second trimester. Throughout the exercise intervention, the safety of both the mother and fetus was prioritized above all else. Prior to each session, a certified instructor assessed participants’ baseline condition, including blood pressure, fatigue level, pain, and dizziness, to determine readiness for physical activity. To accommodate the physiological changes and subjective fatigue commonly experienced during pregnancy, exercise intensity was self-regulated using the Rating of Perceived Exertion (RPE) scale. Exercise intensity was adjusted in an individualized manner based on each participant’s gestational age and current physical condition, ensuring a personalized and responsive approach. Initially, the intensity corresponded to approximately 50-60% of the participants’ maximal heart rate (MHR), aligning with an RPE score of 11 to 13, which represents a perceived effort ranging from “light” to “somewhat hard.” As the intervention progressed, the intensity was gradually increased to an RPE of 13 to 15 during weeks 5 to 8, based on individual participants’ physiological adaptations and fatigue levels. Although the number of sets and repetitions remained consistent throughout the program, the intensity was modulated by adjusting the difficulty of Pilates movements— such as modifications in body position, range of motion, and degree of muscle engagement. This approach allowed for a progressive and individualized increase in perceived exertion while maintaining participant safety. Safety guidelines were strictly implemented during all sessions. Participants were instructed to immediately cease exercise and consult medical personnel if any abnormal symptoms—such as hyperventilation, abdominal pain, vaginal bleeding, dizziness, chest pain, or muscle weakness—occurred during the session. This precautionary framework was established based on the 2021 American College of Obstetricians and Gynecologists (ACOG) guidelines for physical activity during pregnancy, ensuring that the program design minimized potential risks and prioritized maternal-fetal safety at every stage[27].

The Pilates program was conducted twice weekly over an eight-week period, with each session lasting 50 minutes. The structure of each session was as follows:

• Warm-up (10 minutes): Light stretching and breathing exercises

• Main exercise (30 minutes): Mat- or equipment-based Pilates focusing on the abdominal muscles, lumbar region, lower extremities, and pelvic floor muscles

• Cool-down (10 minutes): Gentle stretching

The intervention protocol was primarily adapted from the prenatal Pilates program developed by Noh and Kim [28], and further revised in accordance with the framework proposed by Jeon and Yang [29]. The program aimed to prevent postpartum musculoskeletal pain, strengthen core musculature, stabilize the pelvis, and promote recovery of the abdominal muscles. A detailed breakdown of the components for both mat- and equipment-based Pilates interventions is provided in <Table 2>.

All statistical analyses were conducted using the SPSS for Windows version 22.0 software. Descriptive statistics, including the mean (M) and standard deviation (SD), were calculated for each variable. To evaluate the intervention effects across the three groups (mat Pilates, Reformer Pilates, and control group), change scores (Δ = post-test value − pre-test value) were computed for each outcome variable. This approach was adopted to control for baseline differences among groups and to more accurately isolate the pure effect of the intervention. Given the small sample size and the violation of normality assumptions, a nonparametric statistical approach centered on change scores was deemed appropriate. Tests for normality revealed that none of the variables followed a normal distribution. Therefore, differences in change scores among the three groups were analyzed using the Kruskal-Wallis H test, a non-parametric alternative to one-way ANOVA. When the Kruskal-Wallis test indicated statistically significant group differences, post hoc analyses were conducted using the Mann-Whitney U test with the Bonferroni correction applied to adjust for multiple comparisons (α = 0.05/3 = 0.017). The level of statistical significance for all tests was set at α = 0.05.

Analysis of mean changes in bilateral calf circumference revealed a statistically significant difference among the three groups (p < 0.01). Post hoc comparisons using the Mann-Whitney U test with Bonferroni correction (adjusted p < 0.01) showed that both the mat Pilates and reformer Pilates groups demonstrated significant reductions in calf circumference compared to the control group (p < 0.01) <Table 3>.

Significant differences in perceived stress levels were observed among the groups (p < 0.001). Post hoc analyses revealed that the mat Pilates group experienced significantly greater reductions in stress levels compared to the control group (p < 0.01), while the reformer Pilates group showed even greater reductions than the mat Pilates group (p < 0.01) <Table 4>.

Statistically significant differences in antenatal depression scores were found among the three groups (p < 0.01). Post hoc comparisons indicated that the reformer Pilates group exhibited significantly greater reductions in EPDS scores than both the control and mat Pilates groups (p < 0.01) <Table 5>.

Lower extremity edema is a common symptom reported by approximately 80% of pregnant women, typically attributed to increased venous pressure and fluid retention during pregnancy [4]. In this study, calf circumference was measured using a standard anthropometric tape to assess changes in peripheral edema. Participants were assigned to one of three groups—mat Pilates, reformer Pilates, or a control group that continued with routine daily activities—for eight weeks. The analysis revealed that both Pilates intervention groups demonstrated significantly greater reductions in calf circumference compared to the control group. These results are consistent with previous findings by Choi and Pyo [8], who reported that exercise interventions were effective in alleviating lower extremity edema during pregnancy. Numerous other studies have also documented the benefits of regular physical activity in reducing edema in pregnant populations [30,31]. The underlying mechanisms by which exercise alleviates edema include enhanced venous return facilitated by muscle contractions [32] and improved blood and lymphatic circulation [33].

Specifically, Pilates has been reported to improve core stability and promote lower limb circulation, thereby contributing to reduced edema [34]. The findings of the present study support this physiological rationale. Moreover, comparative analysis between groups indicated that the equipment-based Pilates group exhibited a greater reduction in calf circumference than the mat-based Pilates group. This may be attributed to the resistance provided by Pilates apparatus, which facilitates increased muscle activation and broader ranges of motion [35]. Exercises using equipment offer simultaneous support and resistance, which may enhance the efficacy of muscle pump action in the lower limbs [36]. In addition, previous studies have suggested that equipment-based exercises may provide more structured movement guidance, enabling better postural alignment and muscle recruitment, which are essential during pregnancy for both safety and effectiveness [37]. These structural differences between mat and reformer modalities may partly explain the differential outcomes observed in this study.

Prenatal stress encompasses a wide range of psychological responses, including anxiety and fear concerning the self, the fetus, and one’s partner, stemming from the physiological and emotional changes associated with pregnancy [3]. In the present study, prenatal stress was evaluated using a validated pregnancy-specific stress scale. Results demonstrated significant reductions in stress levels in both the mat and reformer Pilates groups following the 8-week intervention. Previous literature has linked prenatal stress to fatigue, depressive symptoms, and adverse outcomes such as preterm birth and disrupted maternal-fetal attachment [12,38]. Thus, effective stress management during pregnancy is considered essential [39]. The present findings are in line with prior research indicating that physical activity can serve as an effective means of mitigating psychological stress during pregnancy [40].

Nonetheless, some studies, such as that by Cho et al. [41], have reported no significant effect of mat-based Pilates on prenatal stress, suggesting variability in outcomes depending on the intervention type, structure, and adherence. In the current study, reformer Pilates showed a more pronounced reduction in stress compared to mat-based Pilates. This may be explained by the immersive and rhythmical nature of reformer-based sessions, which may offer greater sensory feedback, muscular coordination, and mind-body connection—all of which are known to influence psychological stability [14,18]. Moreover, reformer Pilates often incorporates controlled breathing, proprioceptive stimulation, and guided movements in a semi-reclined or supine position, which may enhance relaxation and improve sleep quality and fatigue—factors closely tied to stress modulation [41,42].

However, it is important to consider that prolonged supine positioning in pregnant women, especially beyond mid-second trimester, may compress the inferior vena cava and reduce venous return, potentially causing dizziness or hypotension [27]. Therefore, Pilates programs for pregnant women must ensure modifications such as side-lying or elevated torso positions when needed, and instructors should be trained to monitor maternal responses during sessions. These considerations are essential for safe and effective exercise prescription.

Antenatal depression is a mood disorder induced by biological changes during pregnancy and can have serious implications, including third-trimester depression, postpartum depression, preterm birth, and low birth weight [43]. Thus, early intervention is critical to safeguarding both fetal development and maternal-infant health. In this study, antenatal depression was assessed using a validated depression screening tool. Notably, a significant reduction in depressive symptoms was observed exclusively in the reformer Pilates group after the 8-week intervention. This suggests that reformer Pilates may offer psychological benefits by promoting emotional stability in pregnant women.

Previous research has supported the role of physical activity in alleviating depressive symptoms among pregnant women and has specifically highlighted the efficacy of Pilates in reducing depression and anxiety in female populations [44]. Oh [45] also found that reformer Pilates enhanced both exercise efficacy and self-efficacy in pregnant participants. Compared to mat-based interventions, reformer exercises provide more external feedback and neuromuscular engagement, which may foster a greater sense of physical competence and emotional confidence. The significant improvement observed in the reformer group suggests that this modality may be more effective in reducing fear and anxiety, thereby leading to improved mood states. In contrast, no significant change was observed in the mat Pilates group, indicating that the mental health benefits of prenatal exercise may depend on the type, duration, and intensity of the intervention, as well as participants’ psychological characteristics [44]. Moreover, consistent with clinical recommendations, exercise programs targeting pregnant women must incorporate careful supervision and avoid prolonged static postures, particularly in the supine position, to prevent maternal discomfort and adverse hemodynamic responses [27]. These safety considerations are essential to ensure both physiological benefit and psychological reassurance during prenatal exercise interventions.

This study has several limitations. Edema was measured solely by calf circumference, without objective assessments such as bioelectrical impedance or ultrasonography, which may affect measurement precision. The sample was limited to pregnant women in their second trimester, restricting generalizability. The small sample size and lack of control over confounding lifestyle factors, such as physical activity and dietary intake, may have influenced the outcomes. Additionally, reliance on self-reported questionnaires and perceived exertion introduces potential bias and variability. Future research should incorporate objective physiological and behavioral measures, larger and more diverse samples, and improved experimental controls.