Nine Months of Pilates Grew the Abs and Left the Glutes Alone

What the answer rests on. Two meanings of replace are kept apart. For the UK NHS guideline of strengthening activity on at least 2 days a week, pilates counts: it is one of the listed examples. For building muscle size and maximum strength, no source here shows pilates doing lifting's job, and none compares pilates with heavy lifting (80% of one-repetition maximum or more) in adults under 60. The pilates evidence is two small single-group MRI studies, one small two-group class comparison (abstract only) and two reviews whose certainty ratings are mostly low or very low. The lifting rules come from a 2017 meta-analysis, a 2023 meta-analysis and the 2026 ACSM position stand, none of which studied pilates.

The NHS guideline. NHS, Physical activity guidelines for adults aged 19 to 64 (nhs.uk; page last reviewed 22 May 2024; next review due 22 May 2027; Crown copyright). Adults should "do strengthening activities that work all the major muscle groups (legs, hips, back, abdomen, chest, shoulders and arms) on at least 2 days a week". The page's examples of muscle-strengthening activities include pilates and lifting weights, alongside carrying heavy shopping bags, yoga, tai chi, resistance bands, body-weight exercises and heavy gardening. Its effort cue is a health threshold ("to the point where you need a short rest before repeating the activity"), not a failure or reps-in-reserve target. The page cites no evidence, gives no amounts for any listed activity and ranks none of them.

Hip and glute muscles on MRI. Dorado et al. Hypertrophy of Lumbopelvic Muscles in Inactive Women: A 36-Week Pilates Study. Sports Health 2020;12(6):547–551. doi:10.1177/1941738120918381; PMID 32396038; PMC7785896. One pilates group measured before and after (the journal's label is "Controlled laboratory study"; no comparison group appears in the pasted article): 12 inactive, healthy premenopausal women (35.7 ± 5.4 years) without previous pilates experience; Table 1 reports n = 9 per side. A 36-week supervised modern studio program, twice a week, 55 minutes a session, designed to Pilates Method Alliance criteria. MRI volumes (cm3) of the quadratus lumborum, iliopsoas, piriformis and gluteus maximus, medius and minimus showed "small nonsignificant changes (P = 0.10-0.97)", negative for the quadratus lumborum and gluteus muscles and positive for the iliopsoas and piriformis. No strength was measured. The authors' hedged reading: "Thus, it seems that the lack of hypertrophy of the lumbopelvic muscles in response to Pilates is more likely due to the fact that Pilates does not activate these muscles with enough intensity to elicit significant muscle hypertrophy." They call the protocol "the reality of a typical Pilates training program for beginners". Body fat did not change significantly (34.6 ± 0.7 vs 36.7 ± 1.3%, P = 0.22). The authors report no conflicts of interest; no funding statement appears in the pasted text.

The abdominal muscles, same program. Dorado C, Calbet JA, Lopez-Gordillo A, Alayon S, Sanchis-Moysi J. Marked effects of Pilates on the abdominal muscles: a longitudinal magnetic resonance imaging study. Med Sci Sports Exerc. 2012;44:1589-1594. doi:10.1249/MSS.0b013e31824fb6ae; PMID 22357302; abstract only. Nine nonactive healthy women, 36 weeks of a standardized Modern Pilates program: rectus abdominis total volume +21% (P < 0.05), 21% on the dominant and 20% on the non-dominant side; obliques plus transversus abdominis +8% on the dominant side (P < 0.05) and +2% on the non-dominant side (P = 0.58). One group; no program details, apparatus or strength data in the abstract. The 2020 paper attributes this result to its own volunteers: "In fact, the volunteers experienced marked hypertrophy of the rectus abdominis and a little less hypertrophy of the obliques and transversus abdominis, as previously reported." The recommendation that Modern Pilates "can be recommended as an effective method to reinforce the muscles of the abdominal wall" is the authors'. The 2012 record lists the Physical Education Department of the University of Las Palmas de Gran Canaria, and the 2020 study was approved by that university's ethics committee.

Pooled evidence in adult women. Bajčetić et al. Effects of Pilates Programs on Body Composition, Flexibility, Muscular Strength, Balance, Functional Performance and Respiratory Function in Adult Women: A Systematic Review and Meta-Analysis. Sports 2026;14(9):407. doi:10.3390/sports14090407 (published 15 September 2026; no PMID in the locked record). 12 studies (2007–2025), approximately 540 women; programs of 5 to 12 weeks, one to four sessions a week; mat, reformer and combined programs pooled; controls "most frequently involved no exercise, usual lifestyle, or usual daily activity". No significant pooled effect on lean body mass (2 studies, 30 vs 30 participants; MD = 0.46 kg; 95% CI: −2.35 to 3.27; p = 0.75), handgrip strength (3 studies, 83 vs 80; SMD = 0.04; 95% CI: −0.63 to 0.72; p = 0.90), vertical jump (2 studies, 28 vs 28; MD = −0.91 cm; 95% CI: −5.26 to 3.44; p = 0.68) or body-fat percentage. Significant for sit-and-reach flexibility (SMD = 0.69; 95% CI: 0.43 to 0.94; p < 0.001), body weight (MD = −3.99 kg) and body mass index (MD = −1.22). Certainty: "the overall GRADE score ranged from very low to moderate certainty, with the majority of anthropometric and body-composition outcomes rated as very low or low." The authors note heterogeneous populations (young women, college students, postmenopausal women, obese and sedentary women). Narrative synthesis, attributed here to the review as a whole because some of its narrative citation numbers do not match its reference list: between-group advantages for pilates in the curl-up test, knee flexion/extension strength, back strength, the static plank and the Functional Reach Test, while 10-RM knee extension, lumbar strength and trunk-lift gains also appeared in control arms. The authors' hedged explanation for the muscle and power results: "One possible explanation may be the relatively low level of external loading that characterizes most Pilates programs." Funding: Provincial Secretariat for Higher Education and Scientific Research, grant number 142-451-3098; no conflicts declared.

One class against another. Evangelou et al. The effect of a three month, low-load- high-repetitions group-based exercise program versus pilates on physical fitness and body composition in inactive women. J Bodyw Mov Ther 2021;26:18-23. doi:10.1016/j.jbmt.2020.08.017; PMID 33992241; abstract only. 26 women (39.8 ± 9.1 y) were assigned to a low-load, high-repetition (LLHR) group program and 16 (39.1 ± 12.2 y) to pilates, both 3 times per week for 3 months; the abstract says "assigned" and does not say randomized. Field tests measured aerobic fitness, flexibility, handgrip strength and lower-limb explosiveness; total and trunk fat were measured by bioelectrical impedance; no muscle measure is reported. Within groups, aerobic fitness, lower-limb explosive power, left-arm handgrip strength and body composition improved with LLHR, and flexibility improved only with pilates. Between groups: "LLHR was superior to the Pilates program in improving aerobic fitness and body composition; whilst Pilates was superior in improving flexibility (p < 0.05)." The authors call LLHR a "medium-intensity form of exercise". The abstract gives no magnitudes, loads or equipment, and no funding or conflict statement appears on the pasted page.

Pilates against other exercise, for strength. Pinto et al. Is pilates better than other exercises at increasing muscle strength? A systematic review. Heliyon 2022;8(11):e11564. doi:10.1016/j.heliyon.2022.e11564; PMID 36439725; PMC9681646. 16 randomized trials reviewed and 11 meta-analysed; 689 participants (354 pilates, 335 other exercise); mean age 60 ± 19 years; programs of 12 ± 7 weeks at 3 ± 1 sessions a week. Comparators: strength exercises, stretching, physical education classes, balance training, PNF, physiotherapy, gymnastics, pool exercise, vibration platforms, elastic bands and yoga. No statistically significant difference for dynamic strength (SMD = -0.29; 95%CI -0.69; 0.10; I2 = 87%), isometric strength (SMD = 0.20; 95%CI -0.06; 0.47; I2 = 43%) or strength endurance (SMD = -0.19; 95%CI -0.46; 0.07; I2 = 0%), at low to very low certainty. Strength in the pilates groups rose 27% ± 16 from pre to post, against 21% ± 24 in the other-exercise groups (Table 1): a before-to-after change, not an effect against a control. In the review's Table 2, the two trials with a strength-training arm, both in older women (Carrasco-Poyatos 2019; Filho 2016), show dynamic strength rising more with strength training; in Carrasco-Poyatos, isometric force measured on an isokinetic dynamometer rose more with pilates. Mat and apparatus were not separated. Two internal inconsistencies: the Discussion calls the strength-endurance comparison statistically significant, while the Results and Abstract report no significant difference (the Results are used here); and a sensitivity analysis describes references 26 and 27 as the stretching trials, though those references are Carrasco-Poyatos and Filho. Funding: CAPES [001] and FCT [UIDB/04585/2020]; no conflicts declared.

The lifting rules. Schoenfeld, BJ, Grgic, J, Ogborn, D, and Krieger, JW. Strength and hypertrophy adaptations between low- vs. high-load resistance training: a systematic review and meta-analysis. J Strength Cond Res 31(12): 3508–3523, 2017. doi:10.1519/JSC.0000000000002200 (FitChef study page: fitchef.com/studies/light-vs-heavy-weights-muscle/). 21 studies, every set in every protocol taken to momentary muscular failure, at least 6 weeks, participants without medical conditions impairing training. Hypertrophy: 41 effect sizes from 10 studies; a trend toward heavier loads in mean effect size (p = 0.10) but no load effect at study level (p = 0.56); mean gains 8.3% vs 7.0%. 1RM strength favoured high loads (p = 0.003; effect sizes 1.69 vs 1.32; mean gains 35.4% vs 28.0%), mostly in untrained participants, and the authors suggest "heavier loading may become increasingly more important for maximal gains in isotonic strength as one garners training experience". Isometric strength did not differ significantly. The authors: comparable results "cannot reasonably be assumed for submaximal, nonfailure training". American College of Sports Medicine Position Stand. Resistance Training Prescription for Muscle Function, Hypertrophy, and Physical Performance in Healthy Adults: An Overview of Reviews. Med Sci Sports Exerc 2026. doi:10.1249/MSS.0000000000003897; PMID 41843416; PMC12965823. 137 systematic reviews (>30,000 participants): momentary failure is "not necessary for benefits to occur"; sufficient effort means "near-failure" or "a target of 2–3 repetitions in reserve (RIR)"; voluntary strength was enhanced by heavier loads (≥80% one-repetition maximum); equipment type was among the variables that did not consistently change outcomes; elastic-band training increased strength and hypertrophy against no-exercise controls. Pilates is not mentioned in the excerpted text. Haugen et al. Effect of free-weight vs. machine-based strength training on maximal strength, hypertrophy and jump performance – a systematic review and meta-analysis. BMC Sports Sci Med Rehabil 2023;15:103. doi:10.1186/s13102-023-00713-4; PMID 37582807; PMC10426227 (FitChef study page: fitchef.com/studies/haugen-2023-machines-vs-free-weights/). 13 studies, 1016 participants (789 men, 219 women), adults 18–60: "strength changes are specific to the training modality". It compared free weights with pin-loaded, fixed-path machines and excluded haptic-based, pneumatic, variable-resistance and non-fixed-path equipment, so spring apparatus is outside its evidence. No funding received; no conflict of interest declared.

The reformer. No source here isolates reformer pilates. Pinto et al. did not differentiate mat and apparatus ("We did not differentiate between Pilate’s mat and its apparatus method"); Bajčetić et al. pooled mat, reformer and combined programs; the pasted Dorado and Evangelou texts do not name the equipment. The spring question is answered only through the rules above: Pinto et al. describe pilates strength training as using "the practitioner’s own body weight or machines equipped with springs and elastic bands", and ACSM reports elastic-band training increasing strength and hypertrophy against no-exercise controls. No source here measures spring loads.

FitChef's reading, marked as ours. Linking the MRI result to the effort rule is FitChef's reading, not the researchers'. Dorado et al. write about intensity of activation and Bajčetić et al. about external loading; neither says sets stopped short of failure, and no source here measures how close a pilates class comes to failure. Populations: the MRI and class studies were in inactive women (mid-30s; around 40), the 2026 review in adult women of mixed profiles, and Pinto et al. in participants averaging 60 ± 19 years; the lifting rules come mostly from untrained participants (Schoenfeld et al.) and from adults 18–60, mostly men (Haugen et al.).