C-reactive protein prognostic accuracy in acute pancreatitis: timing of measurement and cutoff points

OBJECTIVESC-reactive protein (CRP) has been used widely in the early risk assessment of patients with acute pancreatitis. This study evaluated the prognostic accuracy of CRP for severe acute pancreatitis (SAP), pancreatic necrosis (PNec), and in-hospital mortality (IM) in terms of the best timing for CRP measurement and the optimal CRP cutoff points. MATERIALS AND METHODSThis was a single-center retrospective cohort study including 379 patients consecutively admitted with acute pancreatitis. CRP determinations at hospital admission, 24, 48, and 72 h after hospital admission were collected. Discriminative and predictive abilities of CRP for SAP, PNec, and IM were assessed by the area under the receiver-operating characteristic curve and the Hosmer–Lemeshow test, respectively. To determine the optimal CRP cutoff points for SAP, PNec, and IM, the minimum P-value approach was used. RESULTSIn total, 11% of patients had SAP, 20% developed PNec, and 4.2% died. The area under the receiver-operating characteristic curves of CRP at 48 h after hospital admission for SAP, PNec, and IM were 0.81 [95% confidence interval (CI) 0.72–0.90], 0.77 (95% CI 0.68–0.87), and 0.79 (95% CI 0.67–0.91), respectively. The Hosmer–Lemeshow test P-values of CRP at 48 h after hospital admission for SAP, PNec, and IM were 0.82, 0.47, and 0.24, respectively. The optimal CRP at 48 h after hospital admission cutoff points for SAP, PNec, and IM derived were 190, 190, and 170 mg/l, respectively. CONCLUSIONCRP at 48 h after hospital admission showed a good prognostic accuracy for SAP, PNec, and IM, better than CRP measured at any other timing. The optimal CRP at 48 h after hospital admission cutoff points for SAP, PNec, and IM varied from 170 to 190 mg/l.