Investigation and model prediction of early risk factors for ICU refeeding syndrome
-
摘要: 目的 探讨首次收入重症医学科(ICU)患者初次喂养后发生再喂养综合征(RFS)的早期危险因素，并构建预测模型，为入院早期的急危重症患者营养选择策略提供参考依据。方法 回顾性分析2020年1月1日—2021年7月31日首次收住蚌埠医学院第一附属医院ICU且符合病例入排标准患者的病例资料。根据是否并发RFS，分为RFS组和非RFS组，分析比较2组的临床数据指标，并构建RFS早期预测模型。结果 纳入分析的患者200例，分为RFS组134例，非RFS组66例，发病率为66.7%。单因素分析显示：NRS2002营养评分，APACHEⅡ评分，入院3 d是否诊断脓毒症，3 d内胰岛素应用病史，前白蛋白，尿素氮，C反应蛋白等与RFS发生相关(P < 0.05)。筛选患者入院初期即可获得的危险因素并经二元logistic回归分析显示：前白蛋白、尿素氮、C反应蛋白为RFS的独立危险因素(P < 0.05)。构建入院早期RFS预测模型Logit(P)=0.114X1+0.011X2-0.005X3，模型预测急危重症患者RFS发病率的ROC曲线下面积为0.742(95%CI：0.670 8~0.813 2)，灵敏度为81.34%，特异度为57.58%。模型拟合优度显示：χ2=3.808(P>0.05)。结论 在剔除了入院初始合并肝肾功能不全的急危重症患者后，其首次喂养获得了更高的RFS发病率，脏器功能的损害或抑制了电解质水平的下降。前白蛋白、尿素氮、C反应蛋白是急危重症患者早期即可获得的危险因素指标，基于此构建的RFS风险预测模型有着良好的区分度及校准度，可以有效评估急危重症患者RFS发病风险，为医师营养策略选择提供参考意见。Abstract: Objective To investigate the early risk factors of refeeding syndrome (RFS) after initial feeding in ICU patients admitted for the first time, and to construct a prediction model, so as to provide a reference for nutritional selection strategies of acute critically ill patients in early admission.Methods The medical records of patients who were admitted to ICU of the First Affiliated Hospital of Bengbu Medical College for the first time from January 1, 2020 to July 31, 2021 and met the criteria for case inclusion were analyzed retrospectively. The patients were divided into RFS group and non-RFS group according to whether there were concurrent RFS. The clinical data indicators of the two groups were analyzed and compared, and the early prediction model of RFS was constructed.Results A total of 200 cases were included in the analysis, which were divided into RFS group (134 cases) and non-RFS group (66 cases), with an incidence of 66.7%. Univariate analysis showed that NRS2002 nutrition score, APACHE Ⅱ score, diagnosis of sepsis within 3 days after admission, history of insulin use within 3 days, prealbumin, urea nitrogen and C-reactive protein were correlated with RFS (P < 0.05). The risk factors obtained at the early stage of admission were screened and binary logistic regression analysis showed that prealbumin, urea nitrogen and C-reactive protein were independent risk factors for RFS (P < 0.05). The early admission RFS prediction model Logit(P)=0.114X1+0.011X2-0.005X3 was constructed. The area under the ROC curve of the model to predict the incidence of RFS in critically ill patients was 0.742 (95%CI: 0.670 8-0.813 2), the sensitivity was 81.34%, and the specificity was 57.58%. Model goodness-of-fit shows: χ2=3.808 (P>0.05).Conclusion Excluding the acute critically ill patients with hepatic and renal insufficiency at the initial admission, there is a higher incidence of RFS after the first feeding, and the impairment of organ function may inhibit the decline of electrolyte level. Prealbumin, urea nitrogen and C-reactive protein are risk factors that can be obtained at an early stage of critically ill patients. The RFS risk prediction model built on this basis has a good discrimination and calibration degree, which can effectively assess the risk of RFS in critically ill patients, and provide reference for physicians to choose nutrition strategies.
-
-
表 1 一般情况及预后分析结果
例(%)，X±S 组别 总例数(200) RFS组(n=134) 非RFS组(n=66) χ2/t P 性别(男/女) 110/90 75/59 35/31 0.154 0.694 年龄/岁 60.83±16.946 62.33±16.419 57.77±17.705 1.798 0.074 NRS2002/分 4.02±1.034 4.25±1.095 3.53±0.684 4.913 < 0.001 APACHEⅡ/分 18.92±6.726 20.78±5.911 15.121±6.722 6.084 < 0.001 原发病/入院初步诊断 14.421 0.071 脑出血/脑梗死 78(39.0) 52(38.8) 26(39.4) 多发伤(不含颅脑创伤) 25(12.5) 13(9.7) 12(18.2) 重型颅脑损伤 8(4.0) 6(4.5) 2(3.0) 重症肺炎 15(7.5) 8(6.0) 7(10.6) 感染性休克 15(7.5) 15(11.2) 0 急性呼吸衰竭 15(7.5) 9(6.7) 6(9.1) 恶性肿瘤 10(5.0) 9(6.7) 1(1.5) 急性心力衰竭 7(3.5) 5(3.7) 2(3.0) 其他 27(13.5) 17(12.7) 10(15.2) 喂养方式 2.244 0.326 TPN 96(48.0) 68(50.7) 28(42.4) TEN 17(8.5) 9(6.7) 8(12.1) PN+EN 87(43.5) 57(42.5) 30(45.5) 28 d心动过速 149(74.5) 121(90.3) 28(42.4) 53.349 0.001 28 d新插管/机械通气 57/143 52/82 5/61 21.165 < 0.001 机械通气时间/d 4.91±7.43 5.91±7.595 2.86±6.675 2.78 0.006 住院时间/d 22.52±18.81 23.11±19.498 21.30±17.409 0.641 0.522 住院总费用/元 93 032±88 779 101 308±85 130 76 229±94 201 1.891 0.06 28 d内死亡 46(23.0) 39(29.1) 7(10.6) 8.544 0.003 注：APACHE Ⅱ为急性生理与慢性健康状况评分；NRS2002为营养风险评估量表 2002；TPN：全肠外喂养；TEN：全肠内喂养；PN+EN：肠外+肠内联合喂养。 
下载: 导出CSV
表 2 ICU患者发生再喂养综合征的单因素分析
例(%) 因素 RFS组 非RFS组 χ2/Z P 频数 百分比/% 频数 百分比/% 既往饮酒史 0.712 0.399 是 35 26.12 21 31.82 否 99 73.88 45 68.18 初次热量摄入/(kcal·kg-1·d-1) -0.866 0.386 < 5 54 40.3 22 33.33 5~10 18 13.43 10 15.15 >10 62 46.27 34 51.52 初次氨基酸(静脉)/(g·kg-1·d-1) -1.763 0.078 0 51 38.06 34 51.52 0~0.5 39 29.1 16 24.24 >0.5 44 32.84 16 24.24 入院3 d诊断脓毒症 17.384 < 0.001 是 30 22.39 0 0 否 104 77.61 66 100.00 入院诊断糖尿病 2.031 0.154 是 32 23.88 10 15.15 否 102 76.12 56 84.85 喂养前应用预防性维生素B1 0.269 0.604 是 74 55.22 39 59.09 否 60 44.78 27 40.91 3 d内应用糖皮质激素 0.260 0.610 是 64 47.76 29 43.94 否 70 52.24 37 56.06 3 d内应用胰岛素 5.499 0.019 是 106 79.10 42 63.64 否 28 20.90 24 36.36 3 d内应用利尿剂 1.306 0.253 是 58 43.28 23 34.85 否 76 56.72 43 65.15 3 d内胃肠降压 1.475 0.225 是 87 64.93 37 56.06 否 47 35.08 29 43.94 ALB/(g·L-1) 0.002 0.963 < 30 39 29.10 19 28.79 ≥30 95 70.90 47 71.21 PALB/(mg·L-1) 13.999 < 0.001 < 170 90 67.16 26 39.39 ≥170 44 32.84 40 60.61 CRP/(μg·mL-1) 7.785 0.005 < 85 72 53.73 49 74.24 ≥85 62 46.27 17 25.76 尿素氮/(mmol·L-1) 4.083 0.043 < 7.1 80 59.70 49 74.24 ≥7.1 54 40.30 17 25.76 注：ALB为白蛋白；PALB为前白蛋白；CRP为C反应蛋白。
下载: 导出CSV
表 3 自变量赋值方式
自变量 序号 赋值 尿素氮/(mmol·L-1) X1 未赋值/连续性变量 C反应蛋白/(μg·mL-1) X2 未赋值/连续性变量 前白蛋白/(mg·L-1) X3 未赋值/连续性变量
下载: 导出CSV
表 4 ICU患者发生再喂养综合征影响因素二元logistic回归分析
自变量 β 标准误差 瓦尔德 P OR 95%CI 尿素氮 0.114 0.047 5.748 0.017 1.120 1.021~1.230 CRP 0.011 0.003 13.329 < 0.001 1.011 1.005~1.018 前白蛋白 -0.005 0.002 5.677 0.017 0.995 0.991~0.999 常量 -0.029 0.531 0.003 0.957 0.972
下载: 导出CSV
-
[1] da Silva JSV, Seres DS, Sabino K, et al. ASPEN consensus recommendations for refeeding syndrome[J]. Nutr Clin Pract, 2020, 35(2): 178-195. doi: 10.1002/ncp.10474
[2] Koerner T, Haas V, Heese J, et al. Outcomes of an accelerated inpatient refeeding protocol in 103 extremely underweight adults with anorexia nervosa at a specialized clinic in prien, Germany[J]. J Clin Med, 2020, 9(5): 1535. doi: 10.3390/jcm9051535
[3] Yoshida M, Izawa J, Wakatake H, et al. Mortality associated with new risk classification of developing refeeding syndrome in critically ill patients: a cohort study[J]. Clin Nutr, 2021, 40(3): 1207-1213. doi: 10.1016/j.clnu.2020.07.034
[4] Parker EK, Flood V, Halaki M, et al. A standard enteral formula versus an iso-caloric lower carbohydrate/high fat enteral formula in the hospital management of adolescent and young adults admitted with anorexia nervosa: a randomised controlled trial[J]. J Eat Disord, 2021, 9(1): 160. doi: 10.1186/s40337-021-00513-6
[5] Reintam Blaser A, van Zanten ARH. Electrolyte disorders during the initiation of nutrition therapy in the ICU[J]. Curr Opin Clin Nutr Metab Care, 2021, 24(2): 151-158. doi: 10.1097/MCO.0000000000000730
[6] Drysdale C, Matthews-Rensch K, Young A. Further evidence to throw caution to the wind: outcomes using an assertive approach to manage refeeding syndrome risk[J]. Eur J Clin Nutr, 2021, 75(1): 91-98. doi: 10.1038/s41430-020-0676-6
[7] Kanduri SR, Ramanand A, Varghese V, et al. Refractoriness of Hyperkalemia and Hyperphosphatemia in Dialysis-Dependent AKI Associated with COVID-19[J]. Kidney360, 2022, 3(8): 1317-1322. doi: 10.34067/KID.0001632022
[8] Häberle J. Clinical and biochemical aspects of primary and secondary hyperammonemic disorders[J]. Arch Biochem Biophys, 2013, 536(2): 101-108. doi: 10.1016/j.abb.2013.04.009
[9] Weiner SM. Nephrologisches management und medikamentendosierung Bei rheumapatienten mit niereninsuffizienz[J]. Z Rheumatol, 2022, 81(10): 811-828. doi: 10.1007/s00393-022-01283-5
[10] 龙兴霞, 姚梅琪, 姚金兰, 等. ICU肠内营养患者再喂养综合征发生现状及影响因素研究[J]. 中华护理杂志, 2021, 56(6): 818-823. doi: 10.3761/j.issn.0254-1769.2021.06.003
[11] 倪军喜, 周树生, 刘宝. 重症监护室患者再喂养综合征风险评估和预后的关系[J]. 中国老年学杂志, 2017, 37(20): 5109-5111. doi: 10.3969/j.issn.1005-9202.2017.20.069
[12] Rinninella E, D'Angelo M, Borriello R, et al. Incidence and impact of refeeding syndrome in an internal medicine and gastroenterology ward of an Italian tertiary referral center: a prospective cohort study[J]. Nutrients, 2022, 14(7): 1343. doi: 10.3390/nu14071343
[13] Kondrup J, Rasmussen HH, Hamberg O, et al. Nutritional risk screening(NRS 2002): a new method based on an analysis of controlled clinical trials[J]. Clin Nutr, 2003, 22(3): 321-336. doi: 10.1016/S0261-5614(02)00214-5
[14] National Collaborating Centre for Acute Care(UK). Nutrition support in adults: oral nutrition support, enteral tube feeding and parenteral nutrition[R]. National Institute for Health and Clinical Excellence: Guidance, 2006.
[15] Liu N, Zhao XL, Xiong RQ, et al. The performances of SNAQ, GLIM, mNICE, and ASPEN for identification of neurocritically ill patients at high risk of developing refeeding syndrome[J]. Nutrients, 2022, 14(19): 4032. doi: 10.3390/nu14194032
[16] Singer P. ESPEN guideline on clinical nutrition in the intensive care unit[J]. Clin Nutr, 2019, 38(1): 48-79. doi: 10.1016/j.clnu.2018.08.037
[17] Ni TT, Wen Y, Wang YH, et al. Association between albumin or prealbumin levels at different stages and prognosis in severe acute pancreatitis: a 5-year retrospective study[J]. Sci Rep, 2022, 12(1): 16792. doi: 10.1038/s41598-022-21278-1
[18] 赖震宇, 赵展庆, 余秉昌, 等. BISAP评分联合YKL-40和GIP对重症急性胰腺炎患者的预后评估[J]. 临床急诊杂志, 2022, 23(9): 649-654. https://lcjz.whuhzzs.com/article/doi/10.13201/j.issn.1009-5918.2022.09.008
-
图(1)
表(4)
计量
- 文章访问数: 919
- PDF下载数: 390
- 施引文献: 0