改良版危重症营养风险评分对脓毒症病人发生持续炎症 -免疫抑制 -分解代谢综合征的预测价值

目的探讨改良版危重症营养风险( mNUTRIC)评分对脓毒症病人发生持续炎症 -免疫抑制 -分解代谢综合征( PICS)的预测价值。方法纳入 2021年 1月至 2022年 8月在海南医学院第二附属医院重症监护病房(ICU)接受治疗的 285例脓毒症病人为研究对象。收集病人入 ICU时的年龄、查尔森合并症指数评分、急性生理学与慢性健康状况评价 Ⅱ(APACHE Ⅱ)评分、贯器官衰竭估计( SOFA)评分、 mNUTRIC评分以及实验室指标和器官功能支持治疗情况。根据是否发生 PICS将病人分为序PICS组( n=102)与非 PICS组( n=183)比较两组间各指标的差异。应用二分类 logistic回归模型和受试者操作特征( ROC)曲线分析 mNUTRIC评分与脓毒症病人发生，PICS的关系。结果脓毒症病人 PICS发生率为 35.79%。PICS组年龄、查尔森合并症指数评分、 APACHEⅡ评分、 SOFA评分、 mNUTRIC评分、应用血管活性药物时间 ≥72 h占比、机械通气时间 ≥72 h占比、连续肾脏替代治疗( CRRT)时间 ≥72 h占比以及血清 C-反应蛋白、乳酸水平分别为( 73.49±8.64)岁、［2.00(1.00，3.00)］分、(21.67±6.62)分、(8.69±2.63)分、［5.00(5.00，6.00)］分、 53.92%、71.57%、48.04%、［32.85(20.83，74.98)］ mg/L、［3.33(2.30，4.78)］ mmol/L，均高于非 PICS组的( 67.72±8.63)岁、［1.00(0.00，2.00)］分、(17.83±6.05)分、(7.45±2.43)分、［3.00(2.00，4.00)］分、 29.51%、41.53%、23.50%、［27.00(18.50，38.30)］ mg/L、［2.60(2.02，4.00)］ mmol/L，差异有统计学意义( P<0.05)。 PICS组的血清白蛋白水平为(37.17±2.92)g/L，低于非 PICS组的( 38.22±2.88)g/L，差异有统计学意义( P<0.05)。 mNUTRIC评分 ≥5分的脓毒症病人 PICS发生率为 73.64%，高于 mNUTRIC评分 <5分的 12.00%(χ2=111.66，P<0.001)。 logistic回归分析结果显示， mNUTRIC评分值升高［OR=2.61，95％CI:(1.73，3.94)P<0.001］机械通气时间 ≥72 h［OR=3.24，95％CI:(1.60，6.56)P=0.001］、 CRRT时间 ≥72 h［OR=3.45，95％CI:(1.68，7.08)症病人发生 PICS的独立危险因素。 mNUTRIC预测脓毒症病人发生 PICS的 ROC曲线下面积为 0.85， CI:(0.81，0.89)最佳截断值为 4分时，预测脓毒症病人发生 PICS的灵敏度为 79.41%、特异度为 P=0.00，1］是脓毒，评分，95%，84.15%和约登指数为 0.64。结论发生 PICS脓毒症病人入 ICU时 mNUTRIC评分明显高于非 PICS病人，同时 mNUTRIC评的，分是脓毒症病人发生 PICS的独立预测因子，并且 mNUTRIC评分在预测脓毒症病人 PICS方面具有更好的区分能力。

Predictive value of modified NUTrition risk in the critically ill score for persistent inflammation, immunosuppression and catabolism syndrome in patients with sepsis

Objective To investigate the predictive value of modified NUTrition Risk in the Critically ill (mNUTRIC) score for persis-tent inflammation, immunosuppression and catabolism syndrome (PICS) in patients with sepsis. Methods Two hundred and eighty-five patients with sepsis who were treated in the intensive care unit (ICU) of the Second Affiliated Hospital of Hainan Medical Universi-ty from January 2021 to August 2022 were enrolled as the research objects. The age, chalson comorbidity index score, acute physiologyand chronic health evaluation Ⅱ (APACHE Ⅱ) score, sequential organ failure assessment (SOFA) score, mNUTRIC score, laboratoryindexes at ICU admission, and organ function supportive treatment were collected. Patients were assigned into the PICS group (n=102) and the non-PICS group (n=183) according to the presence or absence of PICS, and the differences of each index between the twogroups were compared. Binary logistic regression model and receiver operating characteristic (ROC) curve were used to analyze the rela-tionship between mNUTRIC score and the occurrence of PICS in patients with sepsis.Results In this study, the incidence of PICS inpatients with sepsis was 35.79%. The age, chalson comorbidity index score, SOFA score, APACHE Ⅱ score, mNUTRIC score, the pro-portion of duration of vasoactive drug use more than 72 hours, the proportion of duration of mechanical ventilation more than 72 hours,the proportion of duration of continuous renal replacement therapy (CRRT) more than 72 hours, and the serum levels of C-reactive pro-tein and lactate in the PICS group were (73.49±8.64) years, [2.00 (1.00, 3.00)] points, (21.67±6.62) points, (8.69±2.63) points, [5.00(5.00, 6.00)] points, 53.92%, 71.57%, 48.04%, [32.85 (20.83, 74.98)] mg/L and [3.33 (2.30, 4.78)] mmol/L, respectively, which werehigher than those in the non-PICS group { (67.72±8.63) years, [1.00(0.00, 2.00)] points, (17.83±6.05) points, (7.45±2.43) points, [3.00(2.00, 4.00)] points, 29.51%, 41.53%, 23.50%, [27.00 (18.50, 38.30)] mg/L and [2.60 (2.02, 4.00)] mmol/L}(all P<0.05). The serum al-bumin level in the PICS group was (37.17±2.92) g/L, which was lower than that in the non-PICS group [(38.22±2.88) mmol/L] (P<0.05).The incidence of PICS in patients with sepsis with mNUTRIC score≥5 points was 73.64%, which was higher than that in patients withsepsis with mNUTRIC score<5 points (12.00%)(χ2=111.66, P<0.001). Logistic regression analysis showed that the elevated value of mNUTRIC score [OR= 2.61, 95%CI: (1.73,3.94), P<0.001], duration of mechanical ventilation more than 72 hours [OR=3.24, 95%CI: (1.60, 6.56), P=0.001], duration of CRRT more than 72 hours [OR=3.45, 95%CI: (1.68, 7.08), P=0.001] were the independent risk fac-tors for PICS in patients with sepsis. The area under the ROC curve of mNUTRIC score in predicting PICS in patients with sepsis was0.85, 95%CI: (0.81, 0.89), and the best cut-off value was 4 points to predict PICS in patients with sepsis, with a sensitivity of 79.41%, aspecificity of 84.15% and a Youden index of 0.64.Conclusions The mNUTRIC score of patients with sepsis with PICS is significantlyhigher than that of patients without PICS at ICU admission. At the same time, the mNUTRIC score is an independent predictor of PICSin patients with sepsis, and the mNUTRIC score has a better discriminatory ability for PICS prediction in patients with sepsis.