Hypothesis: Elevated serum bilirubin is associated with perforated appendicitis. Design: Retrospective observational descriptive study using multiple logistic regression expressed as odds ratios. Setting: Large public teaching hospital. Patients: One hundred and seventy consecutive patients with a diagnosis of appendicitis. Patients were included in the study if they had liver function tests on admission and underwent appendectomy. Main Outcome Measures: 1) Incidence of hyperbilirubinemia. 2) Dependent variable: Presence of pathologically defined gangrenous / perforated appendicitis (vs. acute appendicitis). 3) Independent variables: Age (≤ and > 18 years), duration of symptoms (≤ and > 24 hours), admission temperature (≤ and > 38 ^o C), white blood cell count (WBC ≤ and > 10.3 K/cumm) , and systematic inflammatory response score (SIR score ≤ and > 2). Results: Appendicitis was found and confirmed pathologically in 157 (92 %) of 170 patients. Elevated total bilirubin levels (>1mg/dL) were found in 59 (38 %) of 157 patients. Patients with pathologically defined gangrene or perforation were significantly (p=.004) more likely to have hyperbilirubinemia than those with acute appendicitis (see table). No statistical differences were observed for any of the other variables (see table). On logistic regression the only significant relationship between the presence or absence of appendiceal gangrene and perforation was the presence of hyperbilirubinemia (p = 0.031, 95% confidence interval 1.11-7.6). None of the other variables showed any significant relationship. The odds of appendiceal perforation are 3 times higher (Odds ratio of 2.96) for patients with hyperbilirubinemia compared to those with normal bilirubin levels. Conclusions: Hyperbilirubinemia is frequently associated with appendicitis. Elevated bilirubin levels have a predictive potential for the diagnosis of appendiceal perforation.

* A pathological normal appendix was found in 13 patients.