腹膜透析相关性细菌性腹膜炎反复发作的危险因素及预后分析

Risk factors and outcomes of multiple episodes of peritoneal dialysis related bacterial peritonitis

目的探讨腹膜透析相关性细菌性腹膜炎反复发作的危险因素及预后。方法回顾性分析浙江大学医学院附属第一医院2006年1月至2016年9月期间腹膜透析相关性细菌性腹膜炎患者的临床资料，随访时间截止至2016年12月。根据腹膜炎发生频次将患者分为单次组、反复发作组。用Logistic回归及生存分析法分析患者腹膜炎反复发作及长期预后差的危险因素。根据历次细菌性腹膜炎治疗疗效分为初次治疗成功组（治疗后腹膜炎无复发、再发和重现）和初次治疗失败组（包括治疗后复发、再发和重现），采用Logistic回归法分析疗效因素影响；Cox回归模型法分析患者初次治疗失败的影响因素。结果559例细菌性腹膜炎患者纳入本研究，单次组339例，反复发作组220例。Logistic回归分析结果提示，基线血白蛋白（Alb）水平低（OR=0．787，P〈0．001）、主观综合性评估为营养不良（OR=0．422，P〈0．001）是患者反复发作腹膜炎的独立危险因素。LogRank检验结果提示，单次组与反复发作组患者生存率的差异无统计学意义（48．2％比24．1％，P=0．592）；单次组技术生存率高于反复发作组（75．2％比36．2％，P=0．001）。78例初次治疗失败组患者共发生细菌性腹膜炎147例次（复发88例次、再发16例次、重现43例次）；253例初次治疗成功组患者发生细菌性腹膜炎366例次（包括反复发作组上一次腹膜炎治疗成功4周后再次发生的致病菌不同的腹膜炎）。Logistic回归分析结果显示，治疗应答时间晚（OR=1．200，P〈0．001）、革兰阳性菌感染（OR=1．736，P=0．022）、超敏c反应蛋白（hs—CRP）水平高（OR=1．004，P=0．013）、血Alb水平低（OR=0．935，P=0．008）是疗效的独立危险因素。多因素Cox回归结果显示，治疗应答时间晚（HR=1．120，P=0．032）、hs-CRP水平高（HR=1．007，P=0．009）、革兰阳性菌感染（HR=2．462，P=0．002）为细菌性腹膜炎复发、再发、重现的独立危险因素；血Alb水平高（HR=0．942，P=0．048）为腹膜炎复发、再发、重现的独立保护因素。结论基线血Alb水平较低、营养不良是腹膜透析患者反复发作腹膜炎的独立危险因素；治疗应答时间晚、革兰阳性菌感染、血hs—CRP水平高为腹膜透析相关性细菌性腹膜炎复发、再发、重现的独立危险因素，而血Alb水平高则为其独立保护因素。

Objective To investigate the risk factors and prognosis of peritoneal dialysis （PD） related bacterial peritonitis. Methods The clinical data of patients with PD related bacterial peritonitis from January 2006 to September 2016 in our hospital were retrospectively analyzed and followed up until December 2016. Patients were divided into two groups according to the frequency of peritonitis, single episode group and multiple episodes group （no less than two episodes of peritonitis）. According to efficacy of therapy, the episodes of peritonitis were divided into two groups, cured group （no relapse, recurrence or repeat episodes） and failure group （relapse, recurrent or repeat infection after the therapy of initial episode）. Logistic regression and Cox regression were used to analyze the risk factors for outcomes. Results Five hundred and fifty- nine patients had PD related bacterial peritonitis, including 339 patients in the single episode group and 220 patients in the multiple episodes group. Logistic analysis showed low serum albumin level （OR=787, P 〈 0.001） and malnutrition （OR= 0.422, P 〈 0.001） at baseline were independent risk factors for multiple episodes （P 〈 0.001）. The technical survival was better in the single episode group than that in the multiple episodes group （75.2% vs 36.2%, P=0.001） while the difference of survival rate was not significant between the two groups （48.2% vs 24.1%, P=0.592）. Five hundred and thirteen episodes of peritonitis were analyzed, including 147 episodes in failure group （88 relapse episodes, 16 recurrent episodes and 43 repeat episodes） and 366 episodes in cured group. There were 78 patients in failure group and 253 patients in cured group. Logistic analysis showed prolonged response time （OR=1.200, P 〈 0.001）, Gram-positive bacteria infection （OR=1.736, P=0.022）, higher hs- CRP level （OR=1.004, P=0.013）, lower serum albumin level （OR=0.936, P=0.008） were independent risk factors for failure of therapy. Multivariate Cox regression showed prolonged response time （HR=1.120, P=0.032）, Gram-positive bacteria infection （HR=2.462, P=0.002）, higher hs- CRP level （HR=1.007, P=0.009） were independent risk factors for failure of therapy and higher serum albumin level （HR=0.942, P=0.048） was an independent protection factor. Conclusions Low serum albumin level and malnutrition at baseline are independent risk factors for patients with multiple peritonitis episodes. Prolonged response time, Gram-positive bacteria infection, the high hs-CRP level are independent risk factors for relapse or recurrent or repeat episodes while high serum albumin level was an independent protection factor.