Design of Studies for Antibiotic Treatment of Acute Appendicitis in Children: In Support of RCTs.

and surgery groups in our trial had similar demographics, preoperative characteristics, and preoperative diagnostic evaluations. Additionally, the patient choice design may have also minimized the negative effects of surgeon/parent preference bias on outcomes as there have been no appendectomies carried out without a clinical presentation consistent with recurrent appendicitis. A second concern with the RCT by Svensson et al is their primary outcome. In trials of appendicitis comparing nonoperative management to surgery, there is no clear primary outcome that should be used. Similar to other trials, Svensson et al used a variation of comparing ‘‘complication rates’’ between the groups, specifically ‘‘resolution of symptoms without significant complications’’ in which they included failures of nonoperative management as complications. These trials compare 2 therapies with very different risk– benefit profiles. Medical therapy with antibiotics alone allows for many patients to avoid ever having to undergo surgery but has an associated failure rate and recurrence rate. In contrast, appendectomy is a curative invasive procedure that has associated risks for complications and is associated with pain and disability. Several inflammatory intraabdominal diseases, including acute diverticulitis, intraabdominal abscesses from Crohn’s disease, and tubo-ovarian abscesses, have transitioned from primarily surgical management to primarily medical management with surgery reserved for failures of medical management. These practices have evolved because medical management is effective with the majority of patients being successfully treated without surgery and minimal harm occurring in patients who fail medical management and subsequently undergo surgery. Data from previous trials of nonoperative management of appendicitis support its safety as there is no increase in the rate of complicated appendicitis in patients who fail nonoperative management compared with patients who undergo surgery as their initial therapy. However, the previous trials suggest that there is likely to be a 20%– 30% failure rate at 1 year. If a trial’s primary outcome is a comparison of complications between groups and failures of nonoperative management are counted as complications, then an increase in sample size alone will demonstrate a significantly higher rate of complications in the nonoperative group. A primary outcome comparing complications between groups may undermine the potential benefits of nonoperative management. With a 75% success rate, 3 out of every 4 of patients managed nonoperatively would never need surgery, would never experience the pain and disability that occurs after surgery, and would not be put at risk for a complication from surgery. We believe that the question that needs to be answered in future trials of nonoperative management is: if nonoperative management does not lead to a higher rate of complicated appendicitis, then what success rate is necessary to offer it as an initial therapy for appendicitis? Our approach to answer this question in children is to use a patient choice design with the primary outcome being the success rate of nonoperative management, defined as not undergoing an appendectomy. As compared with an RCT, the patient choice design should minimize the potential effects of patient/parent preference bias on outcomes and enrollment by aligning the therapy with the patient’s and family’s preferences. The results from this type of trial will estimate the effectiveness of offering nonoperative management to children with appendicitis in clinical practice.