Background: Treatment of fistulising perianal Crohn's disease (pCD) is challenging, and it can be difficult to ascertain gold standard interventions from the literature. One factor limiting development of a robust evidence-base is heterogeneity of reported outcomes, being prone to reporting bias and being difficult to meta-analyse.
Core outcome sets are used to reduce heterogeneity of outcome reporting. The aim of this systematic review was to make an objective assessment of outcome reporting heterogeneity in the literature, and through identifying outcomes currently reported, inform development of a fistulising pCD core outcome set.
Methods: A systematic review of studies assessing medical, surgical and combination treatment of pCDf was performed. Search strategy was restricted to prospective studies from 2010-onwards. Studies of luminal Crohn's disease were included only if there was a defined perianal fistula subgroup.
Our primary aim was to list the type and frequency of outcome measures used in each study. The secondary aim was to assess the quality of reporting was assessed using Harman's criteria  and assess the methodical quality of the studies using the GRADE criteria.
Results: In total, 49 studies were included which reported a median of 6 (IQR 3–7) outcomes. The five most commonly investigated outcomes were: ≥50% tracts not draining on clinical examination (22 studies; 45%); Perianal Disease Activity Index (20 studies; 41%); Crohn's Disease Activity Index (19 studies; 39%); closure of external opening (17 studies; 35%); and an absence of drainage either spontaneously or on gentle finger pressure (12 studies; 24%). No single outcome was reported in every study.
11 out of the 49 studies were classified as being of high quality, with GRADE scores ≥4. The primary outcome was clearly stated in 73% (36 studies) and defined 76% (37 studies). Secondary outcomes were clearly stated 57% (28 studies) and defined 63% (31 studies). Authors only explained the rationale for using the outcome measures they selected in 33% (16 studies).
“Fistula healing” from the ACCENT II study was defined at more than one time-point. We found the definition of “fistula healing” to be particularly heterogeneous with only 9 of 49 studies (18%) reporting sustained healing on ≥2 clinical visits.
Conclusions: This review demonstrates the heterogeneity of outcomes used in fistulising pCD studies. Researchers should consider the views of all stakeholders, including patients, when determining which outcomes should be reported in clinical studies. This review represents the first step in informing a Delphi consensus with the aim of developing a Core Outcome Set for fistulising pCD.
 Harman NL, Bruce IA, Callery P, et al. (2013), MOMENT–Management of Otitis Media with Effusion in Cleft Palate: protocol for a systematic review of the literature and identification of a core outcome set using a Delphi survey. Trials, 14:70