低体重对不同年龄重症患者短期临床结局的影响

Effects of low body weight on short-term clinical outcome of intensive care patients of differentages

目的探讨低体重对不同年龄重症患者28 d死亡及其他临床结局的影响。方法从MIMIC-Ⅲ重症医学数据库中提取17134例重症监护病房(ICU)的成年患者,提取其人口统计学特征、基础疾病、生命体征、实验室检查结果及疾病严重程度评分等数据。根据入住ICU时患者的年龄分成两组:低龄组(<65岁)7986例与高龄组(≥65岁)9148例。根据患者体重指数(BMI)分成两组:低体重组(BMI<18.5 kg/m2)458例与非低体重组(BMI≥18.5 kg/m2)16676例。比较不同年龄组及不同体重组患者的临床资料,采用单因素及多因素logistic回归及交互作用分析重症患者28 d死亡的影响因素。结果与非低体重组比较,低体重组患者年龄大[(66.6±18.0)岁vs.(64.7±15.9)岁],近期体重下降比例高(19.7%vs.3.7%),急诊入院比例高(83.0%vs.73.0%),慢性阻塞性肺疾病(COPD)、类风湿关节炎、癌症、脓毒症、急性肾功能不全患病率高,但高血压、糖尿病患病比例低;机械通气时间长[29.2(12.0,106.0)h vs.18.7(5.7,93.8)h],差异均有统计学意义(P<0.05)。在全队列组及不同年龄分组中均可见,低体重组28 d病死率较非低体重组高(P<0.05)。单因素logistic分析结果显示,高龄、低体重、体重下降、急诊入院、心功能不全、COPD、癌症、序贯器官衰竭评分(SOFA)、急性肾功能不全、脓毒症、脓毒症休克及最高白细胞均为28 d死亡的影响因素。多因素logistic分析结果显示,癌症及SOFA评分高为28 d死亡的独立危险因素。交互效应分析结果显示,相对于非低体重的低龄组,在调整多因素回归中P<0.05及有临床意义的变量后,低体重低龄(OR=1.77,95%CI 1.24~3.14,P=0.004)、非低体重高龄(OR=1.84,95%CI 1.64~2.08,P<0.001)及低体重高龄(OR=2.95,95%CI 2.09~4.17,P<0.001)均为重症患者28 d死亡的危险因素。结论低体重是不同年龄重症患者短期死亡的危险因素,其不良预后可能与低体重患者人群中癌症患病率高、病情重及机械通气时间长有关。

Objective To explore the effects of low body weight and different ages on the 28-day death and other clinical outcomes of intensive care patients.Methods Data were extracted from Medical Information Mart for Intensive CareⅢ(MIMIC-Ⅲ)intensive care database,and adult patients admitted to the intensive care unit(ICU)were screened.The body mass index(BMI)is calculated according to the initial weight and height when entering the ICU.According to their age,they were divided into two groups when they entered the ICU:the younger age group<65 years old,and the older age group≥65 years old.BMI was divided into two groups:low BMI group<18.5 kg/m2 and non-low BMI group≥18.5 kg/m2.Univariate analysis,multivariate logistic regression,and interaction analysis analyzed the impact of low body weight and different ages on the 28-day death and other clinical outcomes of intensive care patients.Results A total of 17134 patients met the inclusion criteria.According to their age,they were divided into 7986 patients in the low-age group and 9148 patients in the high-age group.According to BMI,they were divided into 458 patients with low BMI and 16676 patients with non-low BMI.Compared with non-low BMI,patients with low BMI were older(66.65±18.02 vs.64.73±15.98,P=0.012),had a higher proportion of recent weight loss(19.7%vs.3.7%,P<0.001)and emergency admissions(83.0%vs.73.0%,P<0.001),had a higher prevalence of chronic obstructive pulmonary disease(COPD),rheumatoid arthritis,cancer,sepsis,and acute renal insufficiency.However,patients with low BMI had a lower prevalence of hypertension and diabetes.On the final clinical outcome,patients with low BMI had a higher prevalence of sepsis and acute renal insufficiency and longer mechanical ventilation time[29.2(12.0,106.0)h vs.18.7(5.7,93.8)h,P<0.001].In the whole cohort and different age groups,it could be seen that the 28-day mortality rate of patients with low BMI was higher than that of non-low BMI.Univariate analysis showed all the characteristics,including high age,low BIM,weight loss,emergency hospital admission,cardiac insufficiency,COPD,cancer,sequential organ failure assessment(SOFA)score,acute renal insufficiency,sepsis,septic shock,and highest white blood cell,were risk factors for 28-day death.Multivariate analysis showed that cancer and SOFA score were independent risk factors for 28-day death.In the interaction analysis,compared with the low-age group with non-low BMI,after adjusting the variables of P<0.05 in the multivariate regression and clinically significant variables,the low-age group with low BMI(OR=1.77,95%CI 1.24-3.14,P=0.004),high-age group with non-low BMI(OR=1.84,95%CI 1.64-2.08,P<0.001)and high-age group with low BMI(OR=2.95,95%CI 2.09-4.17,P<0.001)were risk factors for death at 28 days.Conclusions Low BMI is a risk factor for short-term death in intensive care patients of different ages.Its poor prognosis may be related to low BMI patients with high cancer prevalence,severe illness,and longer mechanical ventilation time.