Background/Aim: Despite advances in intensive care, the mortality rate in severe acute pancreatitis (SAP) is still high. As special therapies, continuous regional arterial infusion of protease inhibitor and antibiotics (CRAI), and enteral nutrition (EN) are now utilized in Japan. Since 1999, we have performed CRAI in patients with pancreatic necrosis and EN in patients with SAP according to the Japanese criteria. We recently reported that the mortality rate was lower in CRAI (+) group (37%) than that in CRAI (-) group (54%) and that the mortality rate was lower in EN (+) group (19%) than that in EN (-) group (35%). However, there have been no analyses about the indications for CRAI and EN. This study aimed to clarify the optimum severity score for CRAI and EN.
Methods: We evaluated 125 patients with SAP according to the Japanese criteria between 1990 and 2006. Severity scores (Ranson and APACHE II) were estimated on admission. CRAI administered a protease inhibitor (nafamostat mesilate: 250 mg/day) and an antibiotic (imipenem: 1.0 g/day) via the celiac artery and SMA for 5-7 days after admission. EN was started through the N-J tube within 3-7 days after admission and was continued until day 21. We analyzed the relationships between severity scores and the mortality rates in patients with and without CRAI and EN, respectively.
Results: In patients with Ranson ≦4, there was no significant difference between the mortality rates in CRAI (+) and CRAI (-) group. In patients with Ranson ≧5, the mortality rate was lower in CRAI (+) group (12/25=48%) than that in CRAI (-) group (15/19=79%) (P <0.05). In patients with APACHE II 0-7, 8-14, and ≧22, there were no significant differences between the mortality rates in CRAI (+) and CRAI (-) group, respectively. In patients with APACHE II 15-21, the mortality rate was lower in CRAI (+) group (5/13=38%) than that in CRAI (-) group (9/12=75%) (P =0.07). In patients with Ranson ≦2, there was no significant difference between the mortality rates in EN (+) and EN (-) group. In patients with Ranson ≧3, the mortality rate was lower in EN (+) group (9/38=24%) than that in EN (-) group (27/57=47%) (P <0.05). In patients with APACHE II 0-7, 8-14, and ≧22, there were no significant differences between the mortality rates in EN (+) and EN (-) group, respectively. In patients with APACHE II 15-21, the mortality rate was lower in EN (+) group (3/10=30%) than that in EN (-) group (11/15=73%) (P <0.05).
Conclusions: In patients with Ranson ≧5 or APACHE II 15-21, CRAI was effective in reducing the mortality rate. In patients with Ranson ≧3 or APACHE II 15-21, EN was effective in reducing the mortality rate.