Lifting weights nudged the 'exercise hormone' most, but any training raised it

Irisin is a messenger protein released during exercise, one of a family known as myokines, and it has been proposed to carry some of exercise's benefits for metabolic health. For adults carrying excess weight, that raises a practical question: does training actually raise circulating irisin, and does the type of training matter? Individual studies have pointed in different directions, with some finding rises and others finding none, leaving the field genuinely inconsistent. Pooling all the randomised trials is the cleanest way to establish whether the signal is real and whether any one modality moves it most.

The authors searched six databases, including PubMed, EMBASE, Scopus and the Cochrane Library, up to 22 April 2025 for randomised controlled trials testing aerobic, resistance, concurrent or high-intensity interval training against passive controls in adults with overweight or obesity, with circulating irisin as the entry requirement. Fifty studies qualified, covering 1780 participants, of whom 1104 exercised and 676 served as controls. Where trials also reported follistatin, myostatin, FGF-21 or markers of blood sugar control and blood fats, those data were extracted as secondary outcomes. Results were pooled using random-effects models, with subgroup analyses by training modality.

Exercise training was associated with a significant rise in circulating irisin compared with doing nothing. In the subgroup analyses the rise was significant after resistance training, high-intensity interval training and concurrent training, and although resistance training showed the largest numerical effect, the differences between modalities were not statistically significant. Follistatin and FGF-21 also increased, while myostatin did not change significantly. Metabolic markers moved favourably as well: fasting blood glucose, insulin, insulin resistance, HbA1c and LDL cholesterol all fell, and HDL cholesterol rose.

This analysis pulls a scattered and contradictory literature into a reasonably coherent answer: regular structured exercise does appear to raise irisin in adults with overweight or obesity, whatever the modality. It also places the irisin response within a broader pattern of favourable myokine and metabolic changes, which strengthens the case that the signal reflects genuine adaptation rather than noise. That said, the included trials disagreed with each other considerably on the size of the effect, and whether a higher irisin level translates into better long-term health remains an open question. Direct head-to-head trials would be needed before any modality could be crowned the winner.

If you train regularly while carrying extra weight, this is a reason to think the sessions are doing more than burning calories: they appear to shift a whole set of muscle-derived messengers and metabolic markers in a favourable direction. It is also a reason not to agonise over picking the one correct type of exercise, since lifting, intervals and combined programmes all raised irisin and none proved superior. The honest unknown is whether irisin itself matters for health, so the improvements in blood sugar handling and cholesterol are arguably the more meaningful part of the result.