RESUMO
AIMS: Type 1 diabetes mellitus (T1DM) is characterised by insulin deficiency. Due to perceived physical activity (PA)-related hypoglycaemia, a minority of people with T1DM exercise regularly. However, the relationship between T1DM and PA remains poorly understood. Our aim was to summarise the existing literature on the effects of PA on short-term glucose control (glycated haemoglobin or time in range) in people with T1DM. METHODS: We searched seven electronic databases (PubMed, Embase, Cochrane library, Cinahl, SPORTDiscus, PEDro and Web Of Science) and two sources of the grey literature (ClinicalTrials.gov and ICTRP). All reviews were screened via title/abstract and full text by two independent reviewers (LE and HT), conflicts were solved by a third independent reviewer (DDC). We excluded animal studies, case reports, non-English articles, qualitative studies, conference abstracts and articles without full-text access. A meta-analysis using random effects model was performed to study the effect of PA on haemoglobin A1c (HbA1c) levels in people with T1DM. RESULTS: We obtained 19,201 unique references across nine different electronic databases. After screening and snowballing, 68 articles were found investigating the effect of PA on glycaemic control in people with T1DM. Overall, HbA1c levels in the PA group (mean difference = 0.29% (0.20%-0.39%)), were lower compared with the control group. CONCLUSION: An overall small beneficial effect of PA on glycaemic control in people with T1DM was found. Caution is advised when interpreting the results of this meta-analysis, given variations in study type, duration, frequency and intensity of physical activity across included studies.