Morihiro Katsura, Akira Kuriyama, Taro Takeshima, Shunichi Fukuhara, Toshi A Furukawa

Postoperative pulmonary complications (PPCs) have an impact on the recovery of adults after surgery. It is therefore important to establish whether preoperative respiratory rehabilitation can decrease the risk of PPCs and to identify adults who might benefit from respiratory rehabilitation. Our primary objective was to assess the effectiveness of preoperative inspiratory muscle training (IMT) on PPCs in adults undergoing cardiac or major abdominal surgery. We looked at all-cause mortality and adverse events. We searched the Cochrane Central Register of Controlled Trials (CENTRAL; 2014, Issue 10), MEDLINE (1966 to October 2014), EMBASE (1980 to October 2014), CINAHL (1982 to October 2014), LILACS (1982 to October 2014), and ISI Web of Science (1985 to October 2014). We did not impose any language restrictions. We included randomized controlled trials that compared preoperative IMT and usual preoperative care for adults undergoing cardiac or major abdominal surgery. Two or more review authors independently identified studies, assessed trial quality, and extracted data. We extracted the following information: study characteristics, participant characteristics, intervention details, and outcome measures. We contacted study authors for additional information in order to identify any unpublished data. We included 12 trials with 695 participants; five trials included participants awaiting elective cardiac surgery and seven trials included participants awaiting elective major abdominal surgery. All trials contained at least one domain judged to be at high or unclear risk of bias. Of greatest concern was the risk of bias associated with inadequate blinding, as it was impossible to blind participants due to the nature of the study designs. We could pool postoperative atelectasis in seven trials (443 participants) and postoperative pneumonia in 11 trials (675 participants) in a meta-analysis. Preoperative IMT was associated with a reduction of postoperative atelectasis and pneumonia, compared with usual care or non-exercise intervention (respectively; risk ratio (RR) 0.53, 95% confidence interval (CI) 0.34 to 0.82 and RR 0.45, 95% CI 0.26 to 0.77). We could pool all-cause mortality within postoperative period in seven trials (431 participants) in a meta-analysis. However, the effect of IMT on all-cause postoperative mortality is uncertain (RR 0.40, 95% CI 0.04 to 4.23). Eight trials reported the incidence of adverse events caused by IMT. All of these trials reported that there were no adverse events in both groups. We could pool the mean duration of hospital stay in six trials (424 participants) in a meta-analysis. Preoperative IMT was associated with reduced length of hospital stay (MD -1.33, 95% CI -2.53 to -0.13). According to the Grades of Recommendation, Assessment, Development and Evaluation (GRADE) Working Group guidelines for evaluating the impact of healthcare interventions, the overall quality of studies for the incidence of pneumonia was moderate, whereas the overall quality of studies for the incidence of atelectasis, all-cause postoperative death, adverse events, and duration of hospital stay was low or very low. We found evidence that preoperative IMT was associated with a reduction of postoperative atelectasis, pneumonia, and duration of hospital stay in adults undergoing cardiac and major abdominal surgery. The potential for overestimation of treatment effect due to lack of adequate blinding, small-study effects, and publication bias needs to be considered when interpreting the present findings.