Abstract: Objective To investigate the optimal timing of continuous venovenous hemofiltration (CVVH) combined with hemoperfusion (HP) for the treatment of patients with acute pancreatitis (AP). Methods A total of 50 AP patients treated in ICU with the disease conditions not improved after conventional treatment for 48 hours and the presence of secondary organ dysfunction were recruited for this study. They were randomly divided into group A (n=25) and group B (n=25). Group A received CVVH+HP in the presence of one organ dysfunction, and group B received CVVH+HP when dysfunctions of two organs emerged. The disappearance
of abdominal pain and the improvement of organ functions (Marshall score increased for at least one score) were the indicators for the discontinuing of CVVH + HP. Changes of physiological parameters, clinical efficacy and prognosis before and after CVVH+ HP were compared between the two groups. Results In group A, the period from onset to the treatment was significantly shorter than that in group B [(69.3±5.4)h vs. (89.4±6.9)h, t=11.446, P=0.000]. APACHE II score had no difference between the two groups before the treatment [(15.7±3.2) vs. (17.4±3.3), χ2=1.893, P=0.064]. The score became decreased after the treatment in both groups [(10.5±1.9) vs. (15.7±3.2), t=10.157, P=0.000 for group A; (13.6±3.9) vs. (17.4±3.3), t=7.192, P=0.000 for group B], and was lower in group A than in group B after the treatment [(10.5±1.9) vs. (13.6±3.9), t=2.388, P=0.024]. The mortality with 28 days was lower in group A than in group B (12.0% vs. 36.0%, χ2=3.947, P=0.047), but the rate of surgical treatment within 28 days had no difference between the two groups (12.0% vs. 32.0%, χ2= 2.914, P=0.088). Compared between groups A and B, length of stay in ICU [(7.8±1.8)d vs. (9.5±2.6)d, t=2.659, P=0.011], duration of mechanical ventilation [(5.6±1.6)d vs. (7.0±2.4)d, t=2.385, P= 0.021] and blood purification [(65.4±14.6)h vs. (78.8±4.7)h, t=4.388, P=0.000] were significantly shorter in group A. After the treatment, oxygenation index (OI) (t=10.715, P=0.000 for group A; t=11.634, P=0.000 for group B), respiratory rate (RR) (t= 11.974, P=0.000 for group A; t=9.721, P=0.001 for group B), blood serum creatinine (Scr) (t=13.910, P=0.000 for group A; t=13.939, P=0.000 for group B) and blood lactic acid (BLA) (t=11.886, P=0.000 for group A; t=9.494, P=0.000 for group B) improved in both groups, and the improvement was more significant in group A than in group B (t=2.388, P=0.024 for OI; t=4.211, P=0.008 for RR; t= 4.823, P=0.006 for Scr; t=4.950, P=0.004 for BLA). The mean arterial pressure (MAP) and heart rate had no differences before and after the treatment in both groups (P＞0.05). Conclusion CVVH+HP is an effective method for the treatment of patients with AP. The treatment should be initiated as soon as possible when conventional treatment fails and dysfunction of one organ appears.

Key words: Hemofiltration, Hemoperfusion, Acute pancreatitis, timing