Efficacy of hemoperfusion combined with continuous veno-venous hemofiltration on the treatment of paraquat poisoning: a Meta-analysis

Abstract

To evaluate the efficacy of hemoperfusion (HP) combined with continuous veno-venous hemofiltration (CVVH) on the treatment of acute paraquat (PQ) poisoning. Prospective randomized controlled trials and retrospective studies on the efficacy of HP combined CVVH in patients with oral PQ poisoning (poisoning time ≤ 24 hours) were found by searching from PubMed, Embase, Cochrane Library, Web of Science, SinoMed, CNKI and Wanfang databases before November 1st, 2019. The experimental group was treated with HP+CVVH, and the control group was treated with HP. Data included the general information of the literature, mortality, survival time, the incidence of respiratory failure and circulatory failure. The bias risk and the data were analyzed using the RevMan 5.3 software. A total of 1 041 literatures were retrieved, and 7 literatures were finally enrolled, including 1 199 patients, with 735 patients in the control group and 464 patients in experimental group. Meta-analysis showed that compared with HP alone, HP+CVVH could significantly reduce the short-term mortality [4-day mortality: hazard ratio (HR) = 0.52, 95% confidence interval (95%CI) was 0.38-0.71, P < 0.000 1], but no significant improvement in long-term mortality was found (28-day or 30-day mortality: HR = 0.68, 95%CI was 0.39-1.21, P = 0.19; 90-day mortality: HR = 1.13, 95%CI was 0.61-2.10, P = 0.07; total mortality: HR = 0.96, 95%CI was 0.72-1.29, P = 0.78). The survival time of patients treated with HP+CVVH was significantly longer than that of HP patients [mean difference (MD) = 2.02, 95%CI was 0.81-3.22, P = 0.001], but the heterogeneity between studies was large. According to the type of literature, a subgroup analysis showed that the survival time of patients treated with HP+CVVH in prospective randomized controlled trials and retrospective studies were significantly longer than that of HP patients (prospective studies: MD = 1.53, 95%CI was 0.94-2.12, P < 0.000 01; retrospective studies: MD = 2.40, 95%CI was 0.08-4.73, P = 0.04). Compared with HP group, HP+CVVH could significantly reduce the incidence of circulatory failure [relative risk (RR) = 0.40, 95%CI was 0.30-0.52, P < 0.00 01], but the incidence of respiratory failure significantly increased (RR = 2.75, 95%CI was 2.18-3.48, P < 0.000 01). HP combined with CVVH can reduce the short-term mortality and the incidence of circulatory failure, prolong the survival time, and save time for further rescue, but it can't improve the long-term prognosis of patients.