Variation in outcome measurement and reporting in studies of pelvic exenteration for locally advanced and recurrent rectal cancer: a systematic review

Abstract

AIM: There is increasing research interest in pelvic exenteration for locally advanced and recurrent rectal cancer. Heterogeneity in outcome reporting can prevent meaningful interpretation and valid synthesis of pooled data and meta-analyses. The aim of this study was to assess homogeneity in outcome measures in the current pelvic exenteration literature. METHOD: MEDLINE, Embase, CENTRAL, CINAHL and Scopus databases were searched from 1990 to 25 April 2023 to identify studies reporting outcomes of pelvic exenteration for locally advanced or recurrent rectal cancer. All reported outcomes were extracted, merged with those of similar meaning and assigned a domain. RESULTS: Of 4137 abstracts screened, 156 studies met the inclusion criteria. A total of 2765 outcomes were reported, of which 17% were accompanied by a definition. There were 1157 unique outcomes, merged into 84 standardized outcomes and assigned one of seven domains. The most reported domains were complications (147 studies, 94%), survival (127, 81%) and surgical outcomes (123, 79%). Resection margins were reported in 122 studies (78%): the definition of a clear resection margin was not provided in 45 studies (37%), it was unclear in 11 studies (9%) and not specified beyond microscopically 'clear' or 'negative' in 31 (28%). Measurements of 2, 1, 0.5 mm and any healthy tissue were all used to define R0 margins. CONCLUSION: There is significant heterogeneity in outcome measurement and reporting in the current pelvic exenteration literature, raising concerns about the validity of comparative or collaborative studies between centres and meta-analyses. Coordinated international collaboration is required to define core outcome sets and benchmarks.