In type 2 diabetes, exercise tames inflammation but leaves cholesterol untouched

Type 2 diabetes and cardiometabolic disease feed each other, and the connection runs through the progressive inflammation that underpins both. Exercise training is considered an effective treatment strategy for type 2 diabetes, but that broad endorsement leaves a more specific question open: which biomarkers actually respond to training in this population, and which do not? Knowing whether exercise moves inflammatory signals, insulin measures or cholesterol tells us what benefit people with the condition can realistically expect to see in their blood work. This review pooled the randomised trial evidence to find out.

The researchers searched seven electronic databases, including PubMed, Embase, the Cochrane Library and Web of Science, for randomised controlled trials from inception to January 2022. Twenty-five trials met the criteria, covering 1257 participants with a mean age of 52 years. The included studies were of moderate to good overall methodological quality. The team used random-effects models to estimate the weighted average differences between exercise and control groups across inflammatory and cardiometabolic risk biomarkers, and ran subgroup analyses by type of training to see whether aerobic, resistance or other formats performed differently.

Exercise training significantly increased adiponectin, a hormone produced by fat tissue, and decreased fasting insulin and insulin resistance as measured by the homeostatic model. The pro-inflammatory markers tumour necrosis factor alpha, interleukin-6 and C-reactive protein all fell. In the subgroup comparisons, aerobic exercise showed the most consistent beneficial effects relative to other training types, although the studies disagreed with each other substantially. The clear null sits alongside those wins: the effects did not extend to the more commonly measured cardiometabolic risk factors, meaning lipid profiles did not benefit.

The inflammatory findings align with the broader evidence that training reduces low-grade inflammation across metabolic conditions, and they extend it specifically to people living with type 2 diabetes. The lipid null complicates a common expectation: readers often assume exercise improves cholesterol across the board, and in these pooled trials it did not for this population. The considerable disagreement between studies, and the authors' own call for additional trials to confirm the results, mean the exact size of each effect remains uncertain even where the direction looks consistent.

For anyone managing type 2 diabetes, this is a reason to think training can genuinely improve the inflammatory and insulin side of the ledger, even when a cholesterol panel barely moves. It reframes what success looks like: unchanged lipids after months of exercise would not mean the work was wasted, because the markers that did respond here sit close to the core of the condition. The evidence is pooled from trials that varied a lot, so the precise gains for any individual remain uncertain.