Abstract
It is well accepted that the randomized controlled trial (RCT) is the best design for determining treatment effectiveness. Despite this, the surgical literature, particularly in the area of inflammatory bowel disease, is replete with case series while there is a paucity of RCTs assessing surgical interventions. There certainly is no lack of questions to be answered and there is a lack of consensus on many issues. Why then are there so few RCTs? Unfortunately, it appears that surgeons in general are more accepting of evidence from case series. However, it also appears that there are certain issues that are unique to, or at least occur more frequently in, clinical trials assessing surgical interventions. The issues of special concern in surgical trials can broadly be categorized as methodological and feasibility issues. Methodological issues include difficulties in standardization of the procedure, deciding on the timing of the trial and prevention of bias when blinding may be difficult or impossible. All these are challenges, but do not preclude the performance of RCTs. On the other hand, feasibility issues may pose more of a challenge. Patients may refuse entry, particularly if they view the two treatments as ‘unequal’ (eg, medical versus surgical therapy) and irreversible (as they usually are in surgery). Patient accrual may be difficult given the variable patterns of disease and indications for surgery. Finally, unlike medical therapies which are controlled by regulating agencies, surgical procedures can simply be performed and there may be little incentive, and even some disincentives, for the surgeon to participate in a trial. Despite these difficulties, better evaluation of surgical therapies is required. Where possible, RCTs should be performed, but in situations where this is not possible other trial designs must be employed that may lack some of the rigour of an RCT but that are more rigorous than the uncontrolled case series.