LAUSR

Nonoperative Treatment of Appendicitis To the Editor We appreciate the meta-analysis1 and Editorial2 evaluating nonoperative treatment of uncomplicated appendicitis in children in this issue of JAMA Pediatrics. However, we propose an alternative interpretation of the evidence and future direction of this work.1,2 The meta-analysis demonstrates consistent safety and efficacy of antibiotics alone to treat an initial episode of acute uncomplicated appendicitis across 5 pediatric studies.1 These results are similar to a recent meta-analysis performed across 10 pediatric studies and across the multiple randomized clinical trials, observational studies, and systematic reviews performed in adults.3,4 Therefore, we agree with the authors’ conclusions that antibiotics are feasible and effective for the initial treatment of acute uncomplicated appendicitis. In contrast, the related Editorial concludes that “nonoperative treatment remains an experimental proposition.”2 The editorialists call for higher-quality clinical trials with more robust study designs, such as randomized clinical trials, to ensure the internal validity of the results. However, the efficacy of nonoperative treatment of appendicitis has already been described in numerous randomized clinical trials and meta-analyses.1,3,4 While we agree with the need for additional studies of nonoperative treatment to assess long-term success and patient-reported outcomes, we believe these future studies should determine the effectiveness of nonoperative treatment in clinical practice. As acknowledged by the editorialists, antibiotics alone and surgery are 2 very different treatment strategies with discordant risks and benefits. These different risk/benefit profiles lead to preferences and biases in patients, families, and physicians for 1 treatment vs another. Therefore, the next logical step is to study the effectiveness of nonoperative treatment outside of the tightly controlled randomized clinical trial, where strict criteria and limited enrollment can affect external validity. Patient-choice trials are associated with higher enrollment rates of a broader patient population; therefore, they are an efficient way to assess the effectiveness of nonoperative treatment in a more diverse population. Because they facilitate informed treatment choice, similar to what occurs in clinical practice, they intrinsically support aligning patient preferences with their treatment and subsequently their outcomes. Available data already clearly support the efficacy and safety of antibiotics alone as an initial treatment for uncomplicated appendicitis in both children and adults. Therefore, additional rigorously controlled trials with high internal validity but potentially limited external validity are not warranted.5 Patients and families need trials that generate results reflective of outcomes in clinical practice. This is best accomplished by enrolling broad patient populations and incorporating the inherent biases that exist in clinical care.