Predicting appendiceal tumors among patients with appendicitis

Abstract

As non-operative management of appendicitis gains popularity, vigilance for appendiceal tumors becomes increasingly important. We hypothesized that among patients presenting with acute appendicitis, those with advanced age, multiple comorbidities, atypical presentation, and complicated appendicitis would be more likely to have underlying appendiceal tumors. We performed a 4-year retrospective cohort analysis of 677 consecutive adult patients who underwent appendectomy for appendicitis at our tertiary care center. Patients with an appendiceal tumor on their final pathology report were compared to patients with no tumor. Conditions present on admission were used to create a multivariate logistic regression model to predict appendiceal tumor. Risk factors were reported as odds ratio (OR) [95% confidence interval]. Model strength was assessed by area under the receiver operating characteristic curve (AUROC). Seventeen patients (2.5%) had an appendiceal tumor. Within this group, fourteen underwent immediate appendectomy, two initially had non-operative management but failed to improve on antibiotics and underwent appendectomy during the initial admission, and one had successful non-operative management and elective appendectomy 19 days after discharge. Four variables contributed to the multivariate model to predict the presence appendiceal tumor: age ≥50 (OR 3.6 [1.1–11.4]), outpatient steroid/immunosuppressant use (OR 12.1 [2.0–72.5]), the absence of migratory right lower quadrant pain (OR 4.7 [1.2–18.1]), and the appearance of a phlegmon on CT scan (OR 7.0 [1.6–30.2]); model AUROC: 0.860 [0.705–0.969]. For patients presenting with acute appendicitis, conditions present on admission may predict underlying appendiceal tumor. Patients with advanced age, multiple comorbidities, atypical presentation, and complicated appendicitis should be considered for appendectomy during the index admission or at earliest convenience if non-operative management is necessary. level III prognostic study