Probability of appendicitis before and after observation

Abstract

Study objective: To examine patients with abdominal pain for changes in probability of appendicitis during observation.

Study design: Retrospective cohort study.

Setting: University-affiliated community hospital.

Methods: 252 patients with abdominal pain who were examined underwent short-term (10.4 hours) observation (95% confidence interval [CI], 8.7, 12.1) before the decision to operate during a one-year period. Alvarado's scoring system and a probability-of-diagnosis nomogram were used to assign scores and estimate probability of appendicitis.

Measurements and results: In the study group, mean score of patients with appendicitis increased after observation from 6.8 (95% CI, 6.2, 7.4) to 7.8 (95% CI, 7.3, 8.3), corresponding to a change in probability of appendicitis from 50% to 65%. Mean score of patients without appendicitis decreased from 3.8 (95% CI, 3.5, 4.1) to 1.6 (95% CI, 1.58, 1.62), corresponding to a change in probability from 35% to 22%. The difference between mean scores for patients with and without appendicitis increased from 2.6 (95% CI, 2.0, 3.2) to 6.2 (95% CI, 6.15, 6.25) during observation. The study group initially had intermediate probability of appendicitis (score, 4.35; 95% CI, 4.04, 4.66) compared with high probability for patients who went directly to surgery after their initial evaluation (63 patients; score, 7.59; 95% CI, 7.05, 8.73) and low probability for patients with abdominal pain who were sent home after their initial evaluation without observation or surgery (2,097 patients; score, 1.87; 95% CI, 1.48, 2.26).

Conclusion: In this group of patients with intermediate initial probability of appendicitis, observation improved the ability to distinguish patients with from those without appendicitis.