Objectives: To evaluate the efficacy and safety of ultrasound-guided percutaneous catheter drainage (PCD) treatment for severe acute pancreatitis compared to conservative and conventional surgical treatments. Methods: Eighty-one patients with severe acute pancreatitis (SAP) were admitted and divided into three groups: forty-nine cases in the conservative therapy group; nineteen cases in the surgery group; and thirteen cases in the PCD therapy group. Forty-five patients with a CT severity index (CTSI) ≤ 8.0 received conservative treatment. One patient with CTSI = 7.0 underwent surgery. Thirty-five patients with a CTSI > 8.0 were randomly selected for surgery or PCD treatment. After randomization, six patients (four patients in the surgery group and two patients in the PCD group) were dropped from the study. The total number of patients included in the surgery and PCD groups was sixteen and thirteen, respectively. Results: Four patients (8.2%) in the conservative therapy group died, five patients (31.3%) in surgery group with a CTSI > 8.0 died, and all patients in the PCD group survived. The mortality rate was lower in the PCD group than in the surgery group (P = 0.048). The serum C-reactive protein (CRP) level recovered more quickly in patients in the PCD group compared to those in the surgery group (P < 0.001). Conclusions: Patients with SAP and a CTSI ≤ 8.0 could be treated with conservative therapy, while patients with a CTSI > 8.0 should be treated with surgery or PCD therapy if the life-threatening complications of extensive fluid collection or necrosis are a factor. However, PCD therapy used in a timely manner for drainage may decrease mortality in patients with SAP, decrease inflammatory mediator release, and avoid incidence of severe sepsis or acute respiratory distress syndrome (ARDS) and emergency surgery. © 2009 Springer.