Patient-reported outcomes after pelvic exenteration for colorectal cancer: A systematic review

Aim Pelvic exenteration (PE) carries high morbidity. Our aim was to analyse the use of patient-reported outcome measures (PROMs) in PE patients. Method Search strategies were protocolized and registered in PROSPERO. PubMed, Embase, Cochrane Library, Google Scholar, Web of Science and ClinicalTrials.gov were searched with the terms 'patient reported outcomes', 'pelvic exenteration' and 'colorectal cancer'. Studies published after 1980 reporting on PROMs for at least 10 PE patients were considered. Study selection, data extraction, rating of certainty of evidence (GRADE) and risk of bias (ROBINS-I) were performed independently by two reviewers. Results Nineteen of 173 studies were included (13 retrospective, six prospective). All studies were low to very low quality, with an overall moderate/serious risk of bias. Studies included data on 878 patients with locally advanced rectal cancer (n = 344), recurrent rectal cancer (n = 411) or cancer of unknown type (n = 123). Thirteen studies used validated questionnaires, four used non-validated measures and two used both. Questionnaires included the Functional Assessment of Cancer Therapy-Colorectal questionnaire (n = 6), Short Form Health Survey (n = 6), European Organization for Research and Treatment for Cancer (EORTC) Quality of Life Questionnaire C30 (n = 6), EORTC-CR38 (n = 4), EORTC-BLM30 (n = 1), Brief Pain Inventory (n = 2), Short Form 12 (n = 1), Assessment of Quality of Life (n = 1), Short Form Six-Dimension (n = 1), the Memorial Sloan Kettering Cancer Center Sphincter Function Scale (n = 1), the Cleveland Global Quality of Life (n = 1) or other (n = 4). Timing varied between studies. Conclusions Whilst the use of validated questionnaires increased over time, this study shows that there is a need for uniform use and timing of PROMs to enable multicentre studies.