Abstract: Background　Rituximab（RTX） has been used to treat glomerular diseases for nearly 20 years.Although there have been accumulated results from evidence-based studies of RTX for primary and secondary nephropathy，there is no research on RTX salvage therapy for severe renal insufficiency.Objective　To study the efficacy and safety of RTX salvage therapy in patients with secondary refractory nephropathy.Methods　Total 48 adult inpatients with refractory nephropathy，who received RTX treatment in Peking University International Hospital from 2018 to 2019，were enrolled in this retrospective study.They were divided into primary nephropathy （PN） group and secondary nephropathy （SN） group according to renal biopsy diagnosis of primary disease.Those in SN group were further divided into SN with non-salvage therapy group （SN-NST subgroup，Scr<300 μmol/L） and SN with salvage therapy group （SN-ST subgroup，Scr≥300 μmol/L） according to the Scr level before RTX treatment.General information （including gender，age，weight，height，body surface area and primary diagnosis），and laboratory tests results before and after using RTX，such as 24-hour urine protein （Upro），white blood cell （WBC） count，red blood cell （RBC） count，platelet （PLT） count，serum albumin （Alb），Scr，and CD4+ cell count as well as calculated CD4+ cell count/WBC ratio were collected.The serious adverse events during RTX treatment 〔including serious infection （treated either by 7 or more days of systemic antibiotic therapy or by more than one antibiotic agents），renal replacement therapy，multiple organ dysfunction syndrome，and death〕，were recorded.3-month renal outcome after the final RTX therapy（namely，whether requiring maintenance hemdialysis） was also recorded.Results　There were 19 （39.6%） patients in PN group，14 （29.2%）in SN-NST subgroup，and 15 （31.2%）in SN-ST subgroup.The mean age of SN-ST subgroup was less than that of PN group and SN-NST subgroup （P<0.05）.PN group showed higher mean baseline levels of 24 hUpro，RBC and PLT，and lower mean Alb than both subgroups SN-NST and SN-ST before using RTX（P<0.05）.SN-ST subgroup had higher mean level of Scr and less mean CD4+ cell count than PN group and SN-NST subgroup before using RTX（P<0.05）.The following several parameters were found with significant pre- and post-RTX treatment differences（P<0.05）：24 hUpro decreased in PN group and SN-NST subgroup，WBC count decreased in subgroups SN-ST and SN-NST，RBC count increased in SN-ST subgroup，PLT count decreased in PN group but increased in SN-ST subgroup，and Alb level increased in PN group and SN-NST subgroup.Post-RTX treatment intergroup comparison found that the mean level of Scr in SN-ST subgroup was higher than that of PN group and SN-NST subgroup （P<0.05）.The accumulated doses of RTX for PN group and SN-NST and SN-ST subgroups were （437±322）mg?month-1?（m2）-1，（503±344）mg?month-1?（m2）-1，and （331±312）mg?month-1?（m2）-1，respectively，showing no significant differences（F=1.048，P=0.359）.SN-ST subgroup had higher rates of serious infection and renal replacement therapy during RTX treatment as well as higher rate of maintenance hemodialysis 3 months after RTX treatment than PN group（P<0.05）.Moreover，compared to SN-NST subgroup，SN-ST subgroup had higher rates of renal replacement therapy during RTX treatment and maintenance hemodialysis 3 months after RTX treatment than PN-NST subgroup（P<0.05）.Conclusion　In those with Scr≥300 μmol/L，rather than improving the remission of nephropathy effectively and safely，RTX therapy may significantly increase the risks of serious infections and renal failure，so it should not be used as salvage therapy in them.

Key words: Nephrosis, Rituximab, Secondary glomerulonephritis, Salvage therapy, Prognosis, Infection, Drug toxicity

摘要： 背景　利妥昔单克隆抗体（RTX）用于肾小球疾病的治疗已近20年，近年来其在治疗原发性和继发性肾病方面积累了一些循证医学证据，但对于较为明显的肾功能不全患者是否适合应用RTX进行挽救性治疗，目前尚无相关研究。目的　观察RTX挽救性治疗继发性难治性肾病患者的有效性和安全性。方法　回顾性选取2018—2019年北京大学国际医院收治的所有住院使用RTX治疗的成年难治性肾病患者48例。根据原发病诊断结果将患者分为原发性肾病（PN）组和继发性肾病（SN）组，再将SN组依据RTX治疗前血肌酐（Scr）分为SN-NST亚组（Scr<300 μmol/L）和SN-ST亚组（Scr≥300 μmol/L）。收集所有患者的一般资料（包括性别、年龄、体质量、身高、体表面积、原发病），RTX治疗前、后实验室检查结果〔24 h尿蛋白（Upro）、白细胞计数（WBC）、红细胞计数（RBC）、血小板计数（PLT）、白蛋白（Alb）、Scr、CD4+细胞计数〕，计算CD4+细胞计数/WBC比例，记录RTX治疗期间严重不良事件〔包括严重感染全身（应用抗生素≥7 d或联合应用2种及以上抗感染药物）、肾脏替代治疗（RRT）、多脏器功能衰竭综合征（MODS）、死亡等〕，RTX末次治疗后3个月肾脏转归情况（即是否行维持性血液透析）。结果　48例使用RTX治疗的难治性肾病患者均进行肾穿刺活检；PN组19例（39.6%），SN-NST亚组14例（29.2%），SN-ST亚组15例（31.2%）。SN-ST亚组患者年龄小于PN组和SN-NST亚组（P<0.05）。PN组患者RTX治疗前24 hUpro、RBC、PLT高于SN-NST亚组、SN-ST亚组，Alb低于SN-NST亚组、SN-ST亚组（P<0.05）；SN-ST亚组患者RTX治疗前Scr高于PN组、SN-NST亚组，CD4+细胞计数低于PN组、SN-NST亚组（P<0.05）。SN-ST亚组患者RTX治疗后Scr高于PN组、SN-NST亚组（P<0.05）。PN组、SN-NST亚组患者RTX治疗后24 hUpro较同组治疗前降低（P<0.05）。SN-NST亚组、SN-ST亚组患者RTX治疗后WBC较同组治疗前降低（P<0.05）。SN-ST亚组患者RTX治疗后RBC较同组治疗前升高（P<0.05）。PN组患者RTX治疗后PLT较同组治疗前降低，SN-ST亚组患者RTX治疗后PLT较同组治疗前升高（P<0.05）。PN组、SN-NST亚组患者RTX治疗后Alb较同组治疗前升高（P<0.05）。三组患者RTX单位时间体表面积的累积剂量比较〔PN组、SN-NST亚组、SN-ST亚组依次为（437±322）、（503±344）、（331±312）mg?月-1?（m2）-1〕，差异无统计学意义（F=1.048，P=0.359）。SN-ST亚组患者RTX治疗期间严重感染发生率、行RRT者比例及RTX末次治疗后3个月行维持性血液透析者比例高于PN组（P<0.05）；SN-ST亚组患者RTX治疗期间行RRT者比例、RTX末次治疗后3个月行维持性血液透析者比例高于SN-NST亚组（P<0.05）。结论　Scr≥300 μmol/L的继发性难治性肾病患者应用RTX治疗对其肾病改善作用不明显，且会明显增加其严重感染和肾衰竭发生风险，故RTX不应作为这部分患者的挽救治疗措施。

关键词: 肾病, 利妥昔单克隆抗体, 继发性肾小球肾炎, 挽救治疗, 预后, 感染, 药物毒性