Journal of Rehabilitation Medicine 51-7

474 X. Lin et al. Different combinations of key words and subject terms were used, including: [(‘exercise’) or (‘physical fitness’) or (‘exercise tolerance’) or (‘physical endurance’) or (‘sports’) or (‘dancing’) or (‘yoga’)] and [(‘telomere’)]. References to relevant reviews and included studies were also searched manually to screen for additional, potentially eligible, studies. The full search strategy using the 4 databases is shown in Appendix S1 1 . Eligibility and exclusion criteria Observational studies fulfilling the following criteria were included in the systematic review and meta-analysis: (i) study investigated the relationship between exercise and telomere length; (ii) study reported telomere length as mean (standard deviation; SD) or median (interquartile range; IQR); and (iii) sample size > 100. Studies with insufficient data, conference abstracts or reviews, and articles not published in English were excluded (see Appendix S1 1 ). Statistical analysis The aim of the current study was to investigate the relationship between physical exercise and telomere length. Heterogeneity among the studies was assessed using the I-squared test. When heterogeneity among studies was high (I 2 > 50%), a random- effects model was used (Review Manager version 5, Foundation for Statistical Computing, Vienna, Austria); otherwise, a fixed- effects model was used. Subgroup analyses were performed to analyse the source of heterogeneity, and according to country, sex, age, study type, tissue source, and type of exercise. RESULTS Search results Study selection The titles and abstracts of all the studies were independently exa- mined by 2 of the authors (XL and JZ). Disagreements between the reviewers were resolved by discussion and re-examination until a consensus was reached. Data extraction and risk of bias assessment sessment was independently conducted by 2 reviewers, and dis agreements were resolved through discussion and re-examination. Data extraction from the eligible studies was performed inde- pendently by 2 of the authors (XL and JZ). However, when data in potentially eligible original studies were lacking, the authors of these studies were contacted by e-mail to obtain information regarding mean (SD) values of telomere length. Methodological quality was evaluated using the Newcastle-Ottawa Scale (NOS) (12). The quality of the included studies was assessed according to 3 variables: (i) selection; (ii) comparability; and (iii) exposure/ outcome, with a total score ranging from 0 to 9. The quality as- A total of 6,763 relevant articles were retrieved using the search strategy (Appendix S1 1 ), of which 4,112 remained after removal of duplicates. Seventy-one studies were considered potentially eligible when title and abstract were screened, of which 60 were exclu- ded due to a sample size <100 (n = 10), 25 that did not provide mean (SD) telomere length, and 25 were either conference abstracts or review articles. A final total of 11 studies were included in the analysis (Fig. 1), including 1 cross-sectional, 7 case-control, and 3 cohort studies (Table I) (16–26). http://www.medicaljournals.se/jrm/content/?doi=10.2340/16501977-2560 1 Table I. Characteristics of studies examining the association between telomere length and exercise Author (ref.) Country Exercise Design Mason et al. (16) 2013 Year USA > 3.55 MET-h/week Cherkas et al. (17) Woo et al. (18) England Cohort study Case-control study Case-control study 2008 2008 Krauss et al. (19) 2011 Self-reported moderate or robust activity Hong Kong Robust: PASE Score > 112.89 USA Moderate: 61.04 < PASE Score < 112.88 Robust: > 7 METS Moderate: > 5–7 METS Denham et al. (20) Sun et al. (21) Garland et al. (22) Laine et al. (23) 2013 Australia Ultra-marathon runners 2012 USA Robust: > 2.5 MET-h/week 2014 USA 2015 Finland Tucker et al. (24) 2017 USA Shadyab et al. (25) USA 2017 Savela et al. (26) 2012 Finland Moderate: 1–1.5 MET-h/week Self-reported moderate or robust activity Robust: > 22.6–45.0 MET-h/week Male, Sample % number Age, years 0 202 50–75 Leukocyte PCR/TS ratio 8 11.6 1,477 18–81 Leukocytes TRF 7 48.7 1,999 ≥ 65 Leukocytes TRF 8 66.7 ± 10.7 Leukocyte PCR/TS ratio 8 43.6 ± 9.2 PBMC PCR/TS ratio 7 30–55 Leukocyte PCR/TS ratio 8 Cross- 83.3 944 sectional study Case-control 100 124 study Case-control 0 4,141 study Case-control 0 study Case-control 100 study Moderate: > 6.1–22.5 MET-h/week Robust: > 16.67 MET-h/week Case-control 47.9 Moderate: > 8.33–16.67 MET-h/week study Robust: ≥ 17.00 MET-h/week Cohort 0 Moderate: 1.25–17.00 MET-h/week study Physical activity determined by a 4-step scale Cohort study Tissue Method of evaluation of Quality* telomeres assessment 100 392 61.97 ± 10.36 Leukocytes TRF 8 599 72.3 ± 6.0 Leukocyte PCR/TS ratio 8 5,025 20–84 Leukocytes PCR/TS ratio 8 1,476 50–79 Leukocytes Southern blot 8 2,913 47.9 ± 4.1 Leukocytes TRF 7 PBMC: peripheral blood mononuclear cell; qPCR: real-time quantitative PCR detecting system; TRF: terminal restriction fragment; T/S ratio: telomere (T), single copy gene (S) ratio T/S; IPAQ: International Physical Activity Questionnaire; PACE: Physical Activity Scale for the Elderly; MET: metabolic equivalent tasks. www.medicaljournals.se/jrm

Journal of Rehabilitation Medicine 51-7 | Página 5