The effect of aerobic exercise on circulating microRNA expression: A systematic review and meta-analysis

PROSPERO FORM 1. Review title. The effect of aerobic exercise on circulating microRNA expression: A systematic review and meta-analysis 2. Original language title. English 3. Anticipated or actual start date. 01/19/2020 4. Anticipated completion date. June 2021 5. Stage of review at time of this submission. The analysis has not yet started; preliminary searches and piloting of the selection process are completed; formal screening of search results against eligibility criteria, and data extraction have not been completed; risk of bias (quality) assessment or data analysis have not been performed. 6. Named contact. Andreas Kreutzer 7. Named contact email. a.kreutzer@tcu.edu 8. Named contact address. Texas Christian University, TCU Box 297730, 2800 S. University Dr, Fort Worth, TX 76129 10. Organizational affiliation of the review. 1. Department of Kinesiology, Texas Christian University 11. Review team members and their organizational affiliations. 1. Andreas Kreutzer, Texas Christian University; Meena Shah, Texas Christian University; Dennis Cheek, Texas Christian University; Ryan Porter, Texas Christian University; Joshua Carr, Texas Christian University 12. Funding sources/sponsors. There are no sources of funding. There are no grants for this research. 13. Conflicts of interest. There are no financial or academic conflicts of interest. 14. Review question. What is the effect of acute aerobic exercise on circulating microRNA expression in healthy adults? 16. Searches. The following databases will be searched 1. PubMed/MEDLINE, 2. CINAHL, 3. SPORTDiscus, 4. Web of Science. 5. Embase. The following Boolean search string will be employed: (exercise OR endurance OR “endurance exercise” OR aerobic OR “aerobic exercise” OR “continuous exercise” OR “moderate intensity continuous exercise” OR “moderate-intensity continuous exercise” OR “moderate exercise” OR “heavy exercise” OR “severe exercise” OR “intermittent exercise” OR “high intensity exercise” OR “high intensity interval exercise” OR “high intensity interval exercise” OR “high intensity intermittent exercise” OR CSE OR HIT OR HIIT OR HIIE) AND (“circulating microRNA” OR ci-miRNA OR c-miRNA OR “extracellular microRNA” OR “circulating microRNAs” OR ci-miRNAs OR c-miRNAs OR “extracellular microRNAs”) 17. URL to search strategy. N/A 18. Condition or domain being studied. This review will focus on the ci-miRNA response to aerobic exercise. 19. Participants/population. Healthy adults. 20. Intervention(s), exposure(s). Interventions will include baseline measurement of ci-miRNA and measurement following endurance exercise. 21. Comparator(s)/control. Pre- to post-intervention comparison. 22. Types of study to be included. Experimental or quasi-experimental studies with measurements of ci-miRNA before and after aerobic exercise. 23. Context. N/A 24. Main outcome(s). 1. Circulating miRNA expression in plasma, serum or whole blood. 25. Additional outcome(s). N/A 26. Data extraction (selection and coding). Search/screening will be carried out by the lead author, who will read all titles and abstracts. Full texts will be reviewed for papers deemed relevant based on title and abstract, and decisions then will be made as to whether a study warrants inclusion based on the following criteria: INCLUSION CRITERIA: a) Experimental or quasi-experimental studies of acute aerobic exercise in healthy adults; b) intervention is an aerobic exercise modality (e.g., cycling, running, rowing); c) incremental, continuous, or intermittent aerobic exercise; d) Published in a peer-reviewed English language journal or on a pre-print server. EXCLUSION CRITERIA: a) diseased or special populations; b) resistance exercise; c) continuous exercise duration < 30 minutes; d) intracellular miRNAs; e) circulating miRNAs from other bodily fluids (e.g., saliva or urine) or extracellular vesicles. After determining which studies meet inclusion, the following variables will be extracted for each study: Authors, title and year of publication, sample size, height, weight, BMI, training status, age, V̇O2max, description of the exercise intervention (incremental exercise, continuous steady state, intermittent exercise, duration, intensity, modality, rest intervals), sample type (whole blood, plasma, serum), ci-miRNAs reported, ci-miRNA expression (relative expression, fold change). If ci-miRNA data are only presented graphically, values will be extracted using WebPlotDigitizer. 27. Risk of bias (quality) assessment. Quality will be assessed using the Cochrane Collaboration’s Risk Of Bias In Non-Randomized Studies - of Interventions (ROBINS-I) tool. 28. Strategy for data synthesis. Both narrative and quantitative synthesis will be performed. All studies that meet the inclusion criteria will be reviewed critically. Meta-analyses will be performed on all ci-miRNAs that were reported in at least three studies with the metafor package (Viechtbauer, 2010) in R statistical software (R Core Team, 2020). Studies will be grouped by design (i.e., type, modality, duration, and intensity of exercise), sample type (whole blood, serum, plasma), ci-miRNAs analyzed/reported, and reporting convention of ci-miRNA response. For studies in which data were reported in a format that allow appropriate calculation of standardized effects, effect size (Hedges’ g), will be calculated using the metafor package. Grouping based on type and duration of exercise will be performed as follows: 1) incremental exercise tests; 2) continuous exercise in the laboratory (30-120 minutes), where nutrition and hydration was controlled; 3) field events (e.g., running or bicycle races), where nutrition, hydration, and environmental factors were not controlled; 4) intermittent exercise (high-intensity interval exercise [HIIE]). Restricted maximal likelihood estimation (REML) will be used for random-effects meta-analysis. For studies where sufficient data for the calculation of standardized effect sizes is not available, we will perform p-value based meta-analyses (Schulz et al., 2019; Willer et al., 2010). Briefly, we will convert the direction of the effect and the p-value observed in each dataset into a signed Z-score. Z-scores for each miRNA will be combined by calculating a weighted sum, with weights proportional to the square root of the effective sample size for each dataset. When p-values are only reported as greater than or less than a given cutoff value, we will use the following conversions: “p ≥ 0.05” and “p ≥ 0.01” will be converted to “p = 0.1”, “p < 0.05” to “p = 0.025”, “p < 0.01” to “p = 0.005”, “p < 0.001” to “p = 0.0005”, “p