Study Of Obesity And Relationship With Metabolic Disease And With Prognosis In Acute Ischemic Stroke Patients

Background：Obesity is an important public concern all around the world. In both developedand developing countries, the number of obesity is increasing rapidly. The generalpopulation researches showed that obesity was an epidemic concern in China andincreased the prevalence of metabolic disease. In general population, obesityincreases the risk of stroke and early death. However, obese stroke patients havebetter prognosis and they are likely to survive after the stroke, which is calledobesity paradox. The prevalence of obesity in stroke patients is unknown and thereis scare data on the relationship of obesity and metabolic disease. There is no data ofobesity paradox in Chinese stroke population. The results of researches on obesityparadox are conflicting. Most of the available researches focused on the associationbetween body mass index (BMI) and mortality. The data on the relationship of otherobesity parameter and other stroke prognosis is lacking.Objectives：We aimed to evaluate the obesity prevalence in the ischemic stroke patients inChina, relationship of obesity and metabolic disease, predictive value of obesity formetabolic disease, and association between obesity and stroke prognosis. Methods：We analyzed the data of acute ischemic stroke patients from CNSR (The ChinaNational Stroke Registry) which was a prospective and multicentre registry research.CNSR registered acute cerebrovascular disease patients in132hospitals in China.Data of social characteristics, risk factors, clinical features and treatment werecollected and patients were followed up for all-cause mortality, functional recoveryand stroke recurrence. General obesity was defined by the World HealthOrganization’s recommendation for Asian population: underweight (BMI<18.5kg/m2),normal weight (18.5-22.9kg/m2), overweight (23-27.4kg/m2), obese (27.5-32.4kg/m2) and severely obese (≥32.5kg/m2). Central obesity was defined by waistcircumference (WC) or waist-to-height ratio (WHtR). The cut-off point of WC was85cm for male and80cm for female. The cut-off point of WHtR was0.5for bothgender. Logistic regression was performed to analyze the association between obesityand metabolic disease. Receiver operating characteristic (ROC) curve was used toanalyze the predictive value of obesity parameter for metabolic disease. Logisticregression was performed to analyze the association between obesity and strokeprognosis.Results：CNSR recruited22216acute cerebrovascular disease patients in132hospitals.10033eligible acute ischemic stroke patients were included in our study, including6210(61.9%) men and3823(38.1%) women.According to BMI,403(4.0%) were underweight,3126(31.2%) werenormal-weight,4932(49.2%) were overweight and1572(15.7%) were obese.According to WC,6272(62.5%) were central obesity. According to WHtR,6147(61.3%) were central obesity. Obesity rate was higher in younger male patients thanmale patients of other age groups and female patients of the same age group. The rateof overweight/obesity differed among different areas, economic conditions andeducation levels: higher in north than south(69.2%vs.56.7%); east area highest,middle area moderate, and west lowest (66.9%,63.5%and58.9%respectively); Advanced economic areas higher (66.2%vs.63.5%); high average salary incomehigher (66.1%vs.62.6%); middle school and above education level was higher thanpreliminary school and below (69.4%vs.59.3%).With the increase of BMI, WC and WHtR, the prevalence of hypertension,diabetes and dyslipidemia increased. With the increase of BMI groups, the prevalenceof hypertension was49.1%,55.9%,65.2%and76.0%（P<0.001），the prevalence ofdiabetes was16.6%,20.7%,27.4%and33.1%（P<0.001），the prevalence ofdyslipidemia was38.2%,47.4%,52.1%and58.1%（P<0.001）. With the increase ofWC groups, the prevalence of hypertension was55.5%,61.7%,66.7%and70.7%（P<0.001），the prevalence of diabetes was19.4%,26.5%,27.3%and31.1%（P<0.001），the prevalence of dyslipidemia was47.4%,49.5%,51.4%and56.6%（P<0.001）；With the increase of WC groups, the prevalence of hypertension was54.8%,61.8%,65.3%and71.3%（P<0.001），the prevalence of diabetes was19.0%,25.9%,26.4%and31.7%（P<0.001），the prevalence of dyslipidemia was46.3%,48.1%,53.1%and56.5%（P<0.001）. The odds ratio of higher BMI group forhypertension, diabetes and dyslipidemia was2.69（95%CI2.35-3.06）,1.97（95%CI1.73-2.25）and1.45（95%CI1.29-1.63）respectively. The odds ratio of higher WCgroup was2.00（95%CI1.78-2.26）,1.92（95%CI1.68-2.18）and1.44（95%CI1.29-1.61）respectively. The odds ratio of higher WHtR group was2.03（95%CI1.81-2.28）,1.96（95%CI1.72-2.23）and1.50（95%CI1.35-1.68）respectively. Formale patients, the area under ROC: BMI0.604（95%CI0.591-0.616），WC0.582（95%CI0.569-0.594）and WHtR0.583（95%CI0.570-0.595）. For female patients,the area under ROC: BMI0.629（95%CI0.613-0.644），WC0.609（95%CI0.593-0.624）and WHtR0.610（95%CI0.594-0.626）. There was no statisticaldifference among the three parameters。Among the10033ischemic stroke patients,9342（93.1%）were followed up at12month and691（6.9%）were lost to the follow-up. Among the survivors, higher BMIpatients had better3month functional recovery (mRS0-1): underweight51.7%，normal weight54.9%，overweight60.0%，obese59.3%and extremely obese59.5%（P<0.001）.12month functional recovery (mRS0-1): underweight57.0%，normalweight63.1%，overweight65.1%，overweight66.4%and extremely obese66.4% （P=0.018）. Multivariate logistic regression showed that overweight was associatedwith3month functional recovery（OR=1.26;95%CI,1.13-1.39）and BMI was notassociated with12month functional recovery. All-cause mortality was higher inunderweight and extremely obese group.3month mortality: underweight15.7%，normal weight8.3%，overweight7.7%，obese7.3%and extremely obese12.1%（P<0.001）.12month mortality: underweight25.3%，normal weight14.2%，overweight12.3%， obese11.3%and extremely obese16.7%（P<0.001）.Multivariate logistic regression showed that extremely obesity was associated with3month mortality（OR2.13;95%CI1.15-3.68）and12month mortality（OR1.46;95%CI1.09-2.50）. Central obesity was not associated with stroke prognosis.Conclusions：1．The prevalence of overweight or obesity was high in ischemic stroke patients.Obesity rate was higher in younger male patients. Obesity was prevalent in north area,high economic condition and high education level population. Education andintervention should be conducted in ischemic stroke patients, especially thepopulation and area mentioned above.2．Obesity increases the prevalence of hypertension, diabetes and dyslipidemia inischemic stroke patients. Obesity is associated with metabolic diseases but thepredictive value is limited. It’s important to screen the risk factors.3．Overweight predicts short term favorable functional recovery in ischemicstroke patients and extremely obesity increases all-cause death risk. More researchesare needed to illustrate the association between obesity and stroke prognosis to guidethe secondary prevention.