Table 3.
Sensitivity analysis of included randomized controlled trials for the outcome of peak workload and 6MWD.
| Outcomes | Number of studies | WMD (95% CI) | Heterogeneity | ||
|---|---|---|---|---|---|
| Pheterogeneity | I2, % | ||||
| Peak workload | Fixed effects model | 5 | 6.89 (3.97, 9.81) | 0.49 | 0 |
| Exclusion of Casaburi 2004 no training | 4 | 7.12 (4.18, 10.06) | 0.64 | 0 | |
| Exclusion of Casaburi 2004 training | 4 | 6.91 (3.98, 9.85) | 0.34 | 11 | |
| Exclusion of Creutaberg 2003 | 4 | 7.01 (4.03, 9.98) | 0.36 | 8 | |
| Exclusion of Piosn 2011 | 4 | 0.22 (−8.65, 9.09) | 0.81 | 0 | |
| Exclusion of Sharma 2008 | 4 | 7.26 (4.28, 10.25) | 0.56 | 0 | |
| 6MWD | Fixed effects model | 5 | 16.88 (−3.27, 37.04) | 0.21 | 31 |
| Exclusion of Dage 2014 | 4 | 5.16 (−18.47, 28.78) | 0.51 | 0 | |
| Exclusion of Ferreira 1998 | 4 | 21.83 (0.89, 42.76) | 0.41 | 0 | |
| Exclusion of Pison 2011 | 4 | 20.85 (−9.63, 51.32) | 0.13 | 47 | |
| Exclusion of Sharma 2008 | 4 | 17.36 (−3.03, 37.74) | 0.13 | 48 | |
| Exclusion of Svartberg 2004 | 4 | 18.78 (−2.13, 39.69) | 0.15 | 44 |
WMD, weighted mean difference; CI, confidence interval; 6MWD, 6–min walking distance.