172 6 Chapter 6 1.9. General health-related QoL General HRQoL was measured with the SF-36 in six studies (Aiken 2006; Kruis 2014; Lilholt 2017; Öztürk 2020; Rea 2004; Vianello 2016), or with the shorter SF-12 in two studies (Fan 2012; Kalter-Leibovici 2018). Aiken 2006 did not provide us with su cient information and did not respond to our emails. Rea 2004 and Öztürk 2020 reported only on the separate dimensions of the SF-36 and therefore could not be used for pooling. For the remaining studies, we pooled composite scores from the SF-36 and the SF-12. Hence, we pooled the data from studies for the Mental Component Summary (MCS) score with a total population of 3699 participants and of the Physical Component Summary (PCS) score with a total population of 3704 participants. Pooled MD on the MSC score showed no signi cant di erences between both groups (MD 0.36, 95% CI -0.38 to 1.11; I² = 0%). Also no signi cant di erences were observed on the PCS score (MD 1.06, 95% CI -0.67 to 2.79; I² = 84%). Substantial heterogeneity observed for the PCS score was due in part to di erences in the quality of the studies. Sensitivity analysis excluding Vianello 2016 and Lilholt 2017 showed similar non-signi cant e ects (see Table 7). Two studies measured QoL with the Sickness Impact Pro le (SIP) (Engstrom 1999; Littlejohns 1991) (Analysis 1.12). No between-group di erences were found in any domain of the SIP. 2. Exercise capacity Twenty-eight studies measured functional or maximum exercise capacity. Functional exercise capacity was measured through the 6MWD (26 studies) or the shuttle test (one study). Maximal exercise capacity was measured using the cycle ergometer test expressed as W-max (five studies), leg fatigue score (one study), and grip strength (one study). The MCID on the 6MWD is estimated at 35 meters (Puhan 2008). No MCID for the cycle ergometer test is reported in the current literature. Results are shown in Figure 4 . 2.1 Functional exercise capacity - short-term We pooled data from 17 studies using the 6MWD including 1390 participants (Bendstrup 1997; Bernocchi 2017; Boxall 2005; Cambach 1997; Gottlieb 2011; Güell 2000; Güell 2006; Jimenez-Reguera 2020; Khan 2019; Mendes 2010; Tabak 2014; Theander 2009; van Wetering 2010; Wakabayashi 2011; Wang 2017; Wijkstra 1994; Zhang 2020). One study could not be pooled, as study authors reported no data because there was no signi cant di erence between groups at 12 months’ follow-up (Bourbeau 2003). The pooled MD on the 6MWD outcome was 52.56 in favour of IDM (95% CI 32.39 to 72.74) and exceeded the MCID of 35. In other words, patients treated in an IDM programme were able to walk 52 meters more, on average, than those who received usual care. Pooling did indicate considerable heterogeneity (I² = 90%). Sensitivity analysis performed on high-quality studies showed a smaller but still statistically and clinically signi cant e ect in favour of IDM (MD 41.00, 95% CI 4.40 to 77.60, I² = 92%).
RkJQdWJsaXNoZXIy MTk4NDMw