摘要目的:探讨连续性肾脏替代治疗（continuous renal replacement therapy，CRRT）净超滤强度与危重症急性肾损伤（acute kidney injury，AKI）患者生存预后的相关性，为CRRT期间最佳净超滤的目标设定提供参考。方法:该研究为回顾性观察性研究。收集2021年5月至2023年9月在四川大学华西医院重症监护病房接受CRRT的AKI危重症患者的人口学资料及临床资料。净超滤定义为CRRT前72 h的每小时液体清除量，将该变量转换为分类变量，即低净超滤<1.01 ml·kg -1·h -1，中净超滤1.01～1.38 ml·kg -1·h -1，高净超滤>1.38 ml·kg -1·h -1，并比较三组患者基线特征和临床治疗情况的差异。采用Kaplan-Meier生存曲线和Log-rank检验比较三组患者CRRT后28 d和60 d的生存情况。采用Logistic回归分析法分析患者CRRT后28 d及60 d死亡的相关因素。 结果:该研究共纳入661例CRRT时长≥72 h的危重症AKI患者，年龄56.00（43.00，68.00）岁，男性488例（73.83%），净超滤为1.36（0.94，1.89）ml·kg -1·h -1。其中，低净超滤组188例（28.44%），中净超滤组152例（23.00%），高净超滤组321例（48.56%）。三组患者在性别分布（ χ2=17.81， P<0.001）、体重指数（ H=32.37， P<0.001）、上机前24 h尿量（ H=9.41， P=0.009）、液体过负荷（ H=6.02， P=0.049）、血小板（ H=13.49， P=0.001）、脑钠肽前体（ H=14.18， P<0.001）、血肌酐（ H=9.66， P=0.008）、乳酸（ H=9.83， P=0.007）、AKI分期分布（ χ2=15.51， P=0.004）、上机指征（ P<0.001）、CRRT总时长（ H=8.45， P=0.015）、超滤（ H=456.10， P<0.001）、净超滤（ H=561.20， P<0.001）以及CRRT治疗72 h血管活性肌力评分（ H=10.42， P=0.005）方面的差异均有统计学意义。Kaplan-Meier生存分析结果显示，三组患者CRRT后28 d（Log-rank检验， χ2=10.89， P=0.004）和60 d（Log-rank检验， χ2=8.55， P=0.014）生存率的差异均有统计学意义。多因素Logistic回归分析结果显示，年龄（ OR=1.03，95% CI 1.02～1.04， P<0.001）、平均动脉压（ OR=0.98，95% CI 0.97～1.00， P=0.011）、胆红素（ OR=3.02，95% CI 1.39～5.59， P=0.006）、CRRT治疗72 h血管活性肌力评分（ OR=1.01，95% CI 1.00～1.02， P=0.004）、低净超滤组（以中净超滤组为参考， OR=1.66，95% CI 1.02～2.72， P=0.042）及高净超滤组（以中净超滤组为参考， OR=1.78，95% CI 1.14～2.78， P=0.011）是CRRT后28 d死亡的独立相关因素；年龄（ OR=1.02，95% CI 1.01～1.04， P<0.001）、平均动脉压（ OR=0.98，95% CI 0.97～1.00， P=0.016）、液体过负荷（ OR=1.10，95% CI 1.02～1.19， P=0.012）、胆红素（ OR=4.96，95% CI 1.00～17.80， P=0.013）、CRRT治疗72 h血管活性肌力评分（ OR=1.02，95% CI 1.01～1.03， P=0.003）及高净超滤组（以中净超滤组为参考， OR=1.91，95% CI 1.22～3.00， P=0.005）是CRRT后60 d死亡的独立相关因素。 结论:在CRRT前72 h，净超滤>1.38 ml·kg -1·h -1和净超滤<1.01 ml·kg -1·h -1与CRRT后28 d或60 d病死率较高相关，净超滤1.01~1.38 ml·kg -1·h -1可能是相对安全的范围。

abstractsObjective:To explore the correlation between the intensity of net ultrafiltration in continuous renal replacement therapy (CRRT) and the survival prognosis in critically ill patients with acute kidney injury (AKI), and provide evidence-based references for establishing optimal net ultrafiltration target during CRRT.Methods:This was a retrospective observational study. Demographic and clinical data of critically ill AKI patients who received CRRT in the Intensive Care Unit of West China Hospital, Sichuan University from May 2021 to September 2023 were collected. Net ultrafiltration was defined as the hourly fluid clearance volume in the 72 hours prior of CRRT. This variable was converted into a categorical variable, including low net ultrafiltration <1.01 ml·kg -1·h -1, moderate net ultrafiltration 1.01-1.38 ml·kg -1·h -1 and high net ultrafiltration >1.38 ml·kg -1·h -1, and the differences of baseline characteristics and clinical treatment conditions among the three groups were compared. Kaplan-Meier survival curve and log-rank test were used to compare the survival conditions among the three groups in patients at 28 days and 60 days after CRRT. Logistic regression analysis method was used to analyze the related factors of mortality in patients 28 days and 60 days after CRRT. Results:This study included a total of 661 critically ill AKI patients who underwent CRRT for more than 72 hours. The age was 56.00 (43.00, 68.00) years, and 488 patients (73.83%) were males. The net ultrafiltration rate was 1.36 (0.94, 1.89) ml·kg -1·h -1. Among them, 188 patients (28.44%) were in the low net ultrafiltration group, 152 patients (23.00%) were in the medium net ultrafiltration group, and 321 patients (48.56%) were in the high net ultrafiltration group. There were statistically significant differences among the three groups in terms of gender distribution ( χ2=17.81, P<0.001), body mass index ( H=32.37, P<0.001), urine volume 24 hours before admission ( H=9.41, P=0.009), fluid overload ( H=6.02, P=0.049), platelets ( H=13.49, P=0.001), pro-B type natriuretic peptide ( H=14.18, P<0.001), serum creatinine ( H=9.66, P=0.008), lactate ( H=9.83, P=0.007), AKI stage distribution ( χ2=15.51, P=0.004), admission indication ( P<0.001), total CRRT duration ( H=8.45, P=0.015), ultrafiltration ( H=456.10, P<0.001), net ultrafiltration ( H=561.20, P<0.001), and vasoactive-inotropic score at 72 hours of CRRT treatment ( H=10.42, P=0.005). Kaplan-Meier survival analysis showed that there were statistically significant differences in the 28-day (Log-rank test, χ2=10.89, P=0.004) and 60-day (Log-rank test, χ2=8.55, P=0.014) survival rates among the three groups in patients after CRRT. Multivariate logistic regression analysis showed age ( OR=1.03, 95% CI 1.02-1.04, P<0.001), mean arterial pressure ( OR=0.98, 95% CI 0.97-1.00, P=0.011), bilirubin ( OR=3.02,95% CI 1.39-5.59, P=0.006), 72-hour vasoactive-inotropic score ( OR=1.01, 95% CI 1.00-1.02, P=0.004), low net ultrafiltration group (medium net ultrafiltration group as a reference, OR=1.66, 95% CI 1.02-2.72, P=0.042), and high net ultrafiltration group (medium net ultrafiltration group as a reference, OR=1.78, 95% CI 1.14-2.78, P=0.011) were independent correlated factors of 28-day mortality after CRRT. Age ( OR=1.02，95% CI 1.01-1.04， P<0.001), mean arterial pressure ( OR=0.98，95% CI 0.97-1.00， P=0.016), fluid overload ( OR=1.10, 95% CI 1.02-1.19, P=0.012), bilirubin ( OR=4.96，95% CI 1.00-17.80， P=0.013), 72-hour vasoactive-inotropic score ( OR=1.02，95% CI 1.01-1.03， P=0.003), and high net ultrafiltration group (medium net ultrafiltration group as a reference, OR=1.91，95% CI 1.22-3.00， P=0.005) were independent correlated factors of 60-day mortality after CRRT. Conclusions:During the first 72 hours of CRRT, net ultrafiltration > 1.38 ml·kg -1·h -1 and net ultrafiltration < 1.01 ml·kg -1·h -1 are associated with a higher mortality rate at 28 days or 60 days after CRRT. Net ultrafiltration of 1.01-1.38 ml·kg -1·h -1 may be a relatively safe range.