Prognostic indicators in acute pancreatitis: CT vs APACHE II

To investigate the correlation between established contrast-enhanced computed tomography (CECT) criteria of disease severity in acute pancreatitis and the APACHE (Acute Physiology and Chronic Health Evaluation) II score and to assess the utility of each as prognostic indicators in acute pancreatitis.

Over a 1-year period, prospective, consensus interpretation of the CECTs of 35 consecutive inpatients was performed with determination of the CECT grade, degree of necrosis, and severity index. The APACHE II score was calculated within 24 h of CECT. Multiple clinical endpoints were recorded: local complications (pseudocyst, abscess, or acute fluid collections requiring urgent surgical or radiological intervention), systemic disease (intensive care unit admission), and duration of hospitalization. Statistical analysis was performed to determine correlations.

No statistically significant correlation existed between the APACHE II score and CECT grade, the degree of necrosis, or the CECT severity index. Only the CECT grade and severity index correlated significantly with the occurrence of local complications (P = 0.0035 and 0.0048, respectively). The APACHE II score was superior to the CECT grade as a predictor of the need for ICU admission (P = 0.022 vs P = 0.035), and no other CECT criteria was a significant predictor of ICU admission.

The preferred clinical and imaging prognostic measures in acute pancreatitis, the APACHE II score and CECT criteria, do not correlate with one another in the commonly encountered, mixed primary and tertiary care population. The mathematical integration of CECT criteria and the APACHE II score fails to yield a predictor of outcome superior to the use of any one measure alone. CECT criteria best define local anatomic abnormality, and are superior to the APACHE II score as predictors of local complications. The APACHE II score is superior to all CECT criteria as an indicator of systemic disease severity (reflected in the need for ICU admission). The most effective initial triage would be immediate APACHE II calculation. Further use of imaging vs clinical assessment can then be individualized.