The original APPAC was one of the first trials systematically testing an antibiotics-first strategy for appendicitis, demonstrating its feasibly and safety as an alternative to immediate surgery. Based on these and other data, reasonable differences of opinion exists regarding the favored approach. In Finland, however, where the original APPAC was performed, they’ve already moved on from the “if” question and onto “how best”.
In APPAC II, the “how best” question involves whether the initial treatment for uncomplicated appendicitis (no perforation, appendicolith, or tumor) need be intravenous, or whether a completely oral antibiotic strategy is noninferior. The intravenous strategy was comprised of two days of ertapenem, followed by five days of levofloxacin plus metronidazole, while the oral strategy was comprised of seven days of moxifloxacin monotherapy. All patients were hospitalized for observation for at least 20 hours, the minimum time necessary for two doses of intravenous antibiotics. The primary endpoint was treatment success at 1 year, defined as avoidance of surgery or recurrent appendicitis.
There were approximately 300 patients enrolled in each group, based on sample size estimates derived from their non-inferiority margin of -6% and an expected success rate of 73%. At one year follow-up, the success rate for the intravenous cohort was 73%, as compared with 70% for oral antibiotics. However, this did not meet their pre-defined margin for non-inferiority, as the difference of −3.6% had a one-sided 95% CI lower bound of -9.7%. This leads us into our favorite statistical wasteland, the land of not-non-inferior, yet also not inferior, nor equivalent.
These are interesting data, and, cutting through the statistical chicanery, it is most likely the outcomes in each arm are virtually indistinguishable. It is hard to tell, however, the advantage of adopting the oral strategy, as implemented, in the face of even a small amount of potential harm. Because all patients were hospitalized and observed inititally, the oral strategy does not avoid unnecessary bed utilization. It is not clear whether this initial hospitalization could be avoided; this initial timeframe constituted the greatest percentage of treatment failures, although this can also be potentially confounded by conservative clinical judgement and readily available operative resources.
The choice of moxifloxacin monotherapy as the comparator is interesting, as it is not strictly equivalent to levofloxacin plus metronidazole with respect to its anaerobic efficacy. It is rather baffling not to simply use levofloxacin plus metronidazole as the oral therapy in each group. The authors cite several publications demonstrating the viability of moxifloxacin monotherapy for intra-abdominal infections, but it seems to muddle the comparison unnecessarily.
In the end, these represent yet another interesting permutation in the approach to the non-surgical management of appendicitis. From a pragmatic standpoint, it seems rather mooted until such data exist showing patients can be managed without hospitalization. Then, if an acute emergency department evaluation is being performed, this provides plenty of opportunity to give at least a single intravenous dose of antibiotics, if warranted, rather than hewing dogmatically to oral-only – admitting fluoroquinolones have identical oral and intravenous bioavailability. These data raise as many follow-up questions as answers, unfortunately.
Finally, tucked into this publication is an even more interesting tidbit: APPAC III. Characterizing diverticulitis as “left-sided appendicitis”, and noting the relative inessential nature of antibiotics for diverticulitis, this currently-enrolling trial tests antibiotics versus placebo for uncomplicated appendicitis. In a world where others are slow to move beyond mandatory operative intervention, this group is testing zero intervention at all – fascinating!
“Effect of Oral Moxifloxacin vs Intravenous Ertapenem Plus Oral Levofloxacin for Treatment of Uncomplicated Acute Appendicitis”
https://jamanetwork.com/journals/jama/fullarticle/2775227