血磷清除率对接受持续性肾脏替代治疗脓毒性急性肾损伤患者预后的影响

Effect of Serum Phosphorus Clearance Rate on the Prognosis of Patients with Septic Acute Renal Injury Receiving Continuous Renal Replacement Therapy

摘要: 目的：探讨血磷清除率与接受持续性肾脏替代治疗(CRRT)的脓毒性急性肾损伤(SAKI)患者预后的关系。方法：收集接受CRRT治疗的SAKI患者的临床资料进行回顾性分析。根据患者行CRRT后28天转归为研究终点，将患者分为存活组和死亡组，比较两组患者的临床信息、生化指标及血磷清除率，并采用二元Logistic回归分析接受CRRT治疗的SAKI患者28天死亡的影响因素，再根据影响因素建立列线图预测模型并对模型进行评价。结果：本研究发现BMI ≥ 24 kg/m2、血肌酐 ≥ 186 μmol/L、白蛋白 ≥ 25 g/L、血磷清除率 ≥ 20.91%是接受CRRT治疗的SAKI患者28天死亡的保护因素，而APACHEII评分 > 35分、SOFA评分 > 5分是危险因素(P < 0.05)。将上述影响因素建立列线图预测模型，预测接受CRRT治疗的SAKI患者28天死亡的ROC曲线分析结果显示，AUC为0.746 [95% CI (0.718~0.775)]说明列线图预测模型区分度好。采用Bootstrap法重复抽样1000次，结果显示一致性指数为0.746，校准曲线显示列线图预测模型有较好的准确度。决策曲线显示，当列线图预测模型的概率阈值为0.20~0.95时，净获益率较高。结论：血磷清除率对接受CRRT治疗的SAKI患者病情判断及预后评估有着重要的价值。

Abstract: Objective: To investigate the relationship between serum phosphorus clearance rate and prognosis of septic acute kidney injury (SAKI) patients receiving continuous renal replacement therapy (CRRT). Methods: The clinical data of SAKI patients who received CRRT treatment were collected and analyzed retrospectively. According to 28 days after CRRT, the patients were divided into sur-vival group and death group. The clinical information, biochemical indexes and blood phosphorus clearance rate of the two groups were compared. Binary Logistic regression analysis was used to analyze the influencing factors of 28-day death in SAKI patients treated with CRRT. A nomogram prediction model was established and evaluated according to the influencing factors. Results: In this study, it was found that BMI ≥ 24 kg/m2, serum creatinine ≥ 186 μmol/L, albumin ≥ 25 g/L and serum phosphorus clearance rate ≥ 20.91% were protective factors for 28-day death in SAKI pa-tients treated with CRRT, while APACHEII score > 35 and SOFA score > 5 were risk factors (P < 0.05). The above influencing factors were used to establish a nomogram prediction model to predict the 28-day death of SAKI patients treated with CRRT. The results of ROC curve analysis showed that the AUC was 0.746 [95% CI (0.718~0.775)], indicating that the nomogram prediction model had a good differentiation. The Bootstrap method was used to repeat sampling for 1000 times, and the results showed that the consistency index was 0.746. The calibration curve showed that the nomogram prediction model had better accuracy. The decision curve shows that when the probability thresh-old of the nomogram prediction model is 0.20~0.95, the net benefit rate is higher. Conclusion: The serum phosphorus clearance rate is of great value in judging the condition and prognosis of SAKI patients treated with CRRT.