Conservative management of occult pneumothorax in mechanically ventilated patients

An occult pneumothorax is defined as air within the pleural cavity diagnosed with a computed tomography (CT) scan which has not been suspected on the basis of clinical findings or chest X-ray. The best management strategy has remained unclear, with inconsistencies in the guidelines, literature and speciality opinion. As a high percentage of trauma patients require mechanical ventilation either for general anaesthesia or intensive care stay due to the nature of their injuries, the question of how to manage occult pneumothoraces in this population continues to be raised. The aim of the research presented in this thesis was to investigate the safety and effectiveness of conservative management versus intercostal catheter (ICC) insertion for the management of occult pneumothoraces in mechanically ventilated patients. JBI systematic review methodology and methods were employed to address this aim. A search for published and unpublished literature included PubMed, Embase, CINAHL, Web of Science, Cochrane Central Register of Controlled Trials, ICTR, ANZCTR and ClinicalTrials.gov. Following the database search, hand searching of reference lists from included articles was conducted. Studies were included if they explored the effectiveness of conservative management versus ICC insertion for the management of occult pneumothoraces in mechanically ventilated patients. Randomised controlled trials (RCTs) and cohort studies were included. Eligible studies were critically appraised by two reviewers using appropriate JBI tools to assess methodological quality. Where required, contact was attempted with corresponding authors for clarifications and further data. RCTs and cohort studies, where appropriate, were analysed in separate meta-analyses using mixed-methods logistic regression. Sensitivity analyses were performed using Mantel-Haenszel and Peto models. The search yielded 2230 unique citations. Following screening of titles and abstracts, 20 articles were retrieved for full-text screening. Of these, one trial was ongoing and could not be included. Two additional studies were identified through hand searching. Twenty-one full-text articles were screened; eight were ineligible. Two articles were from the same study, leaving 12 included studies (three RCTs and nine cohort studies) involving 311 participants (135 in RCTs and 176 in cohort studies). One RCT had high methodological quality, while aspects of the remaining two trials were unclear. Overall, the cohort studies fulfilled the majority of the quality appraisal criteria. For the primary outcomes, analysis of RCTs revealed with conservative management versus ICC insertion: progression of pneumothorax OR 2.36 (95% CI 0.81-6.8, 3 RCTs), ICC insertion (any reason) OR 4.2 (95% CI 0.33-52.5, 2 RCTs). No result was statistically significant. Similarly, considering the remaining outcomes, there were no statistically significant differences, except for ICC insertion (progression to simple pneumothorax); OR 4.8 (95% CI 1.01-23.6, 3 RCTs). Observational data confirmed these trends in the majority of outcomes; however, contradictory results were seen in the outcomes of pneumonia/empyema and ICC insertion (non-pneumothorax reasons). Adverse events included tension pneumothorax and ICC complications. Incidence of tension pneumothorax was 2.5% in the conservative management group and 0.7% in the ICC group. The incidence of ICC complications in the ICC group was 20% versus 3.8% of patients requiring an ICC with conservative management. ICC complications were significantly lower in the conservative management group when an ICC was required in the RCTs (OR 0.12, 95% CI 0.01 – 2.26). In conclusion, conservative management and ICC insertion appeared equally effective for the management of occult pneumothorax in mechanically ventilated patients. Conservative management can be seen as a safe alternative to ICC insertion, with a low percentage of failure of conservative management reported, a low tension pneumothorax rate and a lower ICC complication rate when an ICC is subsequently required.Thesis (MClinSc) -- University of Adelaide, The Joanna Briggs Institute, 202