Diagnostic accuracy of hyperbilirubinaemia in anticipating appendicitis and its severity.

BACKGROUND

Hyperbilirubinaemia is reported to be a positive predictor in diagnosing appendicitis and especially appendiceal perforation. We, therefore, analysed the diagnostic accuracy of serum bilirubin in anticipating appendicitis and its severity.

METHODS

All consecutive patients undergoing appendectomy for suspected appendicitis from May 2009 to August 2011 were analysed. Patients were classified based on final histopathological findings into the groups: no appendiceal inflammation, non-perforated appendicitis and perforated appendicitis. Primary outcome was the diagnostic accuracy of serum bilirubin levels in discriminating between no appendiceal inflammation and any appendicitis (perforated and non-perforated appendicitis) and non-perforated and perforated appendicitis.

RESULTS

Of 493 analysed patients, 125 (25%) had no appendiceal inflammation, 312 (64%) had non-perforated appendicitis and 56 (11%) had perforated appendicitis. The proportion of patients with bilirubin elevation (>1.1 mg/dL) was different between those with no appendiceal inflammation (14%) and any appendicitis (36%) (p<0.0001), and between non-perforated appendicitis and perforated appendicitis 48% (p=0.04). However, the positive and negative likelihood ratios (LRs) for an elevated bilirubin were poor at discriminating the groups: no appendiceal inflammation versus any appendicitis (LR+ 2.62 (95% CI 1.65 to 4.16) and LR- 0.75 (95% CI 0.67 to 0.83)) and non-perforated appendicitis versus perforated appendicitis (LR+ estimate 1.74 (95% CI 1.28 to 2.38) and LR- 0.72 (95% CI 0.55 to 0.93)).

CONCLUSIONS

Hyperbilirubinaemia is present in acute appendicitis but has a low diagnostic accuracy in discriminating between any appendicitis versus no appendiceal inflammation and perforated versus non-perforated appendicitis and is, therefore, of limited value in clinical routine.

TRIAL REGISTRATION NUMBER

NCT01698099.