
Methods: Retrospective cohort study comparing outcomes following post-discharge intravenous (IV) or oral therapy for perforated appendicitis at a large, tertiary care children’s hospital. Patients < 19 years hospitalized between Jan 2005 and Mar 2015 were identified by ICD-9 codes. Patient eligibility and outcomes in the 6 weeks post-discharge were determined through manual chart review. Patients with immunocompromising conditions, non-surgical management, hospitalization>48 hours prior to surgery, or appendicitis without acute perforation were excluded. Treatment failure was defined as abscess or wound infection. Secondary complications included central venous catheter complications, medication adverse effects, or ongoing abdominal symptoms that were not infectious.
Results: Of 838 children with perforated appendicitis, 782 were prescribed antibiotic therapy at hospital discharge; 203 received IV antibiotics, range 0-75% across 26 surgeons (Figure 1) while 579 received oral antibiotics. Treatment failure occurred in 27 (4.7%) children receiving oral and 6 (3.0%) receiving IV therapy, and secondary complications occurred in 16 (2.8%) children receiving oral and 37 (18.2%) receiving IV therapy. In a multivariable logistic regression model adjusted for sex, age, length of stay, race, initial white blood cell count, C-reactive protein, creatinine, parenteral nutrition, inpatient complications, and inclusion of an anti-pseudomonal agent at discharge, treatment failure was not significantly associated with use of with IV antibiotics (aOR 0.22; 95% CI 0.04, 1.35) but was associated with secondary complications (aOR 5.16; 95% CI 1.94, 13.68).
Conclusion: In children with perforated appendicitis, treatment failure did not differ between children treated with IV versus oral antibiotics after discharge. However, secondary complications were significantly increased in patients with IV therapy after discharge.

L. Handy,
None
M. Bryan, None
D. Dona, None
E. Spyridakis, None
A. Kyriakousi, None
T. A. Metjian, None
J. S. Gerber, None