Exercise improves insulin and cholesterol in PCOS, on shaky evidence

PCOS is typically managed first through lifestyle change, aiming to ease symptoms and lower the raised risks of type 2 diabetes and cardiovascular disease that accompany the condition. Exercise sits at the centre of that advice, yet it has been unclear how well it actually performs against usual care, against diet alone, or when combined with diet. This review set out to pull the trial evidence together and test those comparisons directly. Knowing which outcomes exercise genuinely moves would help women with PCOS and their clinicians decide where to put their effort.

The authors searched databases in June 2017 with no time limit, looking for randomised or quasi-randomised trials measuring the chronic effects of exercise, or exercise plus diet, in women with PCOS. From 2390 articles returned, 27 papers from 18 trials were included. Outcomes spanned insulin and glucose measures, blood lipids, blood pressure, body measurements and cardiorespiratory fitness. Results were pooled as mean differences, and the review was registered in advance on an international register.

Compared with control, exercise lowered fasting insulin by 2.44 μIU/mL and improved insulin resistance, while reducing total cholesterol by 5.88 mg/dL, LDL cholesterol by 7.39 mg/dL and triglycerides by 4.78 mg/dL. Aerobic fitness rose by 3.84 ml/kg/min, waist circumference fell by 2.62 cm and body fat by 1.39 percentage points. No effect emerged for blood pressure, fasting glucose, HDL cholesterol or waist-to-hip ratio. Improvements were greatest in overweight or obese participants and when programmes were supervised, aerobic in nature, or shorter. On limited data, exercise combined with diet was no different from diet alone.

The results back the standard advice that exercise supports metabolic health in PCOS, but they also expose how thin the trial base is: the evidence was graded low or very low quality throughout, effects were modest with wide uncertainty, and several findings depended on individual trials. The subgroup signals, favouring supervision and aerobic work, are worth testing properly rather than treating as settled. The comparisons the field most needs, exercise versus diet and combined approaches versus diet alone, could not be made. The authors call for rigorously designed, well-reported trials to fill those gaps.

If you have PCOS, this is a reason to think regular exercise can shift meaningful markers: insulin, cholesterol, waist size and fitness all moved in the right direction across pooled trials. It is equally a reason to hold expectations lightly, because the studies behind those numbers were small and of low quality, and blood pressure and fasting glucose did not budge. The pattern hinting that supervised and aerobic programmes worked best is informational, not a prescription. Better trials will be needed before PCOS-specific exercise guidance can be written with confidence.