目的 探讨体检人群脂肪肝的检出率及其相关因素.方法 以在我院体检资料完整的健康体检者15452例作为对象,记录其脂肪肝检出率,分析脂肪肝的发生与性别、年龄、体重指数的关系,对脂肪肝的相关危险因素进行Logistic多因素回归分析.结果 本组对象中,轻度脂肪肝所占比例最大;脂肪肝的检出率以50～59岁组最高,男性明显多于女性,超重、肥胖者的检出率明显高于正常体重和体重偏瘦者;多因素回归分析提示,年龄、空腹血糖、甘油三酯、低密度脂蛋白、高密度脂蛋白、胆固醇、尿酸、体质指数与脂肪肝的发生密切相关.结论 肥胖、高血糖、高血脂、高血尿酸是脂肪肝发病的主要危险因素,其可为脂肪肝的早期干预提供依据.

目的 探讨腰高比对2型糖尿病(T2DM)高危人群筛查的价值.方法 选择首钢4个社区的无糖尿病病史的T2DM普查人群,根据口服葡萄糖耐量试验(OGTT)结果,分层随机抽取6072例患者,男、女各3036例,其中,糖耐量正常组3080例,糖调节受损组2992例(包括空腹血糖受损及糖耐量减低).采用Logistic回归分析和受试者工作特征(ROC)曲线对比分析肥胖评价指标体重指数(BMI)、腰臀比(WHR)和腰高比(WHtR)对T2DM高危人群筛查的价值.结果 Logistic回归分析显示WHR、WHtR、BMI与T2DM的高危人群呈显著正相关(P＜0.05).WHtR作为预测变量时,其对应的ROC曲线下面积(AUC)较大,提示筛查效果较好.结论WHtR筛查T2DM高危人群的效果优于BMI和WHR.

目的:探讨体检人群中高尿酸血症与心血管危险因素的相关性.方法:选取2016年～2017年连续在我院体检的健康体检者,资料完整的15118人,筛选出高尿酸血症组(HUA)4060例,正常血清尿酸组(NUA)11058例,比较两组血压、血脂、空腹血糖(FBG)、体质指数(BMI)等心血管危险因素水平,分析血清尿酸水平与心血管危险因素的关系.结果:①HUA组收缩压(SBP)、舒张压(DBP)、空腹血糖(FBG)、总胆固醇(TC)、甘油三酯(TG)、低密度脂蛋白胆固醇(LDL-C)水平及BMI高于NUA组,HUA组HDL-C低于NUA组,两组间比较均有统计学差异(P<0.01或P<0.05).根据血清尿酸水平四分位数比较心血管危险因素聚集情况,结果:显示:随着血清尿酸水平的增加,危险因素聚集显著增加(p<0.01);②Logistics回归显示,性别、年龄、吸烟、FPG、TG、TC、血压、超重与肥胖及脂肪肝与高尿酸血症发生密切相关(P<0.01或P<0.05);男性是女性发病的2.576倍,吸烟是不吸烟的1.158倍,血压不达标是正常血压的1.307倍,血糖水平的影响更为显著,空腹血糖受损、新发血糖增高、糖尿病血糖控制达标及不达标,分别是正常血糖的1.801倍、2.144倍、1.161倍及1.833倍,说明从空腹血糖受损就对血尿酸有一定的影响,而且血糖越高影响越明显.结论:高尿酸血症与心血管危险因素密切相关,增加心血管事件的危险性.

目的 通过对比分析健康管理对心血管危险因素的聚集情况,评价健康管理的干预效果.方法 以471名体检人群为研究对象,分析12种(吸烟、饮酒、少活动、体质指数、血压、血脂、血糖、血尿酸、同型半胱氨酸、脂肪肝、骨密度异常及动脉硬化检测异常)心血管危险因素的分布与聚集情况,同时对健康管理干预前后的效果进行分析,并提出相应的健康管理方案建议.结果 (1)干预前、后均存在严重的心血管危险因素聚集现象,研究人群中0～2项危险因素分别为21.44％、19.32％,同时具有3种及以上危险因素的分别占有78.56％、80.68％;(2)干预后空腹血糖受损、血压正常高值检出率明显降低,高血压治疗达标率显著增高,均P＜0.05;高尿酸、活动少略有增加,但差异无统计学意义(P＞0.05);高同型半胱氨酸增加(P＜0.05),其他吸烟、饮酒、脂肪肝、新发高血压、高血糖等干预前后均无统计学意义(P＞0.05).结论 开展健康管理是预防和遏制心血管危险因素发生、发展的有效途径,尤其对血压、血糖的改善更为显著.

Objective This study investigated the application value of waist-to-height ratio (WHtR), body mass index (BMI), and waist-to-hip ratio (WHR) in screening for type 2 diabetes mellitus (T2DM) and high risk of T2DM. Methods A total of 7 582 subjects aged 40-75 years were randomly selected and stratified based on the results of an oral glucose tolerance test (OGTT) administered during a standard screening for T2DM in the Shougang community.Three anthropometric indices(WHtR,BMI,and WHR) were compared, with the optimal cutoffs for WHtR, BMI, and WHR identified using receiver operating characteristic curve (ROC) analysis. Using multivariate logistic regression analysis and the area under the curve(AUC)of ROC,the associations between BMI,WHR,and WHtR and T2DM were analyzed by group: normal glucose tolerance group (n= 3 080), T2DM high-risk group (n= 2 992 cases), and T2DM group(n=1 510).Results Logistic regression analysis showed that BMI,WHR,WHtR,and family history of diabetes were positively correlated with T2DM and high risk of T2DM(P<0.05 and P<0.01,respectively);WHtR was most significant, with odds ratios of 90.409 and 69.285, respectively. WHtR had the greatest AUC under the ROC in men,whereas BMI had the greatest AUC in women.The optimal cutoffs values for the detection of T2DM were 0.51, 25.47 kg/m2, and 0.91 for WHtR, BMI, and WHR in men, respectively, and 0.52, 24.95 kg/m2, and 0.86 for WHtR, BMI, and WHR in women, respectively. WHtR was more efficient than WHR and BMI based on the AUC. The optimal cutoff values for detecting a high risk for T2DM were 0.51,25.30 kg/m2,and 0.91 for WHtR,BMI,and WHR in men and 0.51,24.81 kg/m2,and 0.86 for WHtR, BMI, and WHR in women, respectively. Conclusion Waist-to-height ratio may be a more effective index for diagnosing type 2 diabetes mellitus and identifying individuals at high risk for T2DM than BMI or WHR.

随着信息技术的发展,医疗和健康咨询服务已经成为互联网经济中的新生力量,加快实现了健康管理进程,各种健康管理软件、健康信息平台应运而生[1～3],但目前尚缺少适宜企业的健康管理模式与体系.首钢总公司党委高度重视集团员工的健康状况,提出了建立覆盖全集团12万职工的健康管理体系;首钢集团下属北京首钢自动化信息技术有限公司、北京大学首钢医院及集团公司相关部门,充分发挥资源的综合优势,共同设计开发了首钢职工健康管理信息系统.

Objective To analyze the result of glycated hemoglobin (HbA1c) measurement and related factors in persons with normal glucose tolerance (NGT).Methods A 75 g glucose tolerance test (OGTT) was conducted in all participants from diabetes screening population of Shougang four communities.HbA1c,fasting plasma glucose (FPG),2 hour plasma glucose (2 hPG) after OGTT and blood biochemical tests were conducted.Anthropometric measurements were performed to all the subjects.A total of 9 711 individuals with normal FPG and 2 hPG were included for data analysis.All the subjects were divided into three groups by levels of HbA1c,and the clinical characteristics of the three groups were compared.Logistic regression was used to analyze the correlation of HbA1c to related factors.Results ① HbA1c was normally distributed in NGT populations.The range of HbA1c was 4.9％-6.5％ (5.69±0.44)％.There wassignificant difference in HbA1c between men (5.68 ± 0.44)％ and women (5.70 ± 0.44)％ (P＜0.05).Especially,there was significant difference (P＜0.01) between the older group (60 to 79 years old) and the other grooups.②HbA1c increased with age.There was significant differences among the young group (16 to 44 years old),middle-aged group (45-59 years),the elderly groups and very elderly group (≥80 years) in HbA1c,FPG and 2 hPG (P＜0.01),whereas there was no significant difference between the elderly groups and the very elderly group.③There were statistically significant differences between HbA1c≤5.6％ group and HbA1c 5.7％-6.4％ group,HbA1c≥6.5％ group in body mass index (BMI),waist-hip ratio (WHR),systolic blood pressure (SBP),serum total cholesterol (TC),triglyceride (TG),low density lipoprotein cholesterol (LDL-C) levels,FPG and 2 hPG (P＜001 or P＜005),whereas there was no significant difference between HbA1c 5.7％-6.4％ group and ≥the 6.5％ group.④ Logistic regression analysis showed that HbA1c was positively correlated with gender,age,BMI,WHR,SBP,TC,FPG,and was negatively correlated with DBP,HDL-C (OR:1.377,1.018,1.033,22.867,1.011,1.3,4.049,0.973,0.835,respectively).Conclusion HbA1c is positively correlated with age and FPG in the population with normal glucose tolerance.The non-diabetic population with high risk HbA1c should undergo monitoring for FPG,2 hPG and related cardiovascular risk factors.

Objective Toanalyze the result of glycated hemoglobin (HbA1c) measurement and related factors in normal glucose tolerance (NGT) population.MethodsA 75 g glucose tolerance test (OGTT) was administered in all participants from diabetes (DM) screening population of Shougang four community. HbA1c, fasting plasma glucose (FPG), 2 hour plasma glucose (2 h PG) after OGTT and blood biochemical test were conducted. Anthropometric measurements were performed to all the subjects. A total of 9 711 individuals with normal FPG and 2 h PG were included for data analysis. All the subjects were divided into three groups by levels of HbA1c, and the clinical characteristics of the three groups were compared. Logistic regression was used to analyze the correlation of HbA1c to related factors.ResultsHbA1c was normally distributed in NGT populations. The range of HbA1c was 4.9%-6.5%, (5.69±0.44)%. There was difference in HbA1c between men and women (P<0.05). Especially, there was significant difference (P<0.01) in the older group (60 to 79 years old). HbA1c increased with age. There was significant&nbsp;differences among the young group (16 to 44 years old), middle-aged group (45-59 years), the elderly groups (60 to 79 years) and very elderly group (≥80 years) in HbA1c, FPG and 2 h PG (P<0.01), whereas there was no significant difference between the elderly groups and the very elderly group. There were statistically significant between HbA1c≤5.6% group and HbA1c 5.7%-6.4% group, HbA1c≥6.5% group in body mass index (BMI), waist-hip ratio (WHR), systolic blood pressure (SBP), serum cholesterol (TC), triglyceride (TG), low density lipoprotein cholesterol (LDL-C) levels, FPG and 2 h PG (P<0.01 orP<0.05), whereas there was no significant difference between HbA1c 5.7%-6.4% group and≥the 6.5% group. Logistic regression analysis showed that HbA1c was positively correlated with gender, age, BMI, WHR, SBP, TC, FPG, and was negatively correlated with DBP, HDL-C.ConclusionHbA1c is positively correlated with age and FPG in the population with normal glucose tolerance. The non-diabetic population with high risk HbA1c should be monitor FPG, 2 h PG and related cardiovascular risk factors.

Objective To explore the suitable ways screening of type 2 diabetes mellitus (T2DM) and DM related high-risk population. Methods The 16 056 subjects from Shougang community were studied. The oral glucose tolerance test (OGTT) was performed and the HbA1c, FPG and 2 h PG were examined in all subjects in the screening. Receiver operating characteristic (ROC) curve analysis was used to examine the sensitivity and specificity of FPG, HbA1c and 2 h PG for diagnosing T2DM and DM related high-risk population, which was defined by the standard from WHO in 1999. Results According to the OGTT standard of screening for DM, which was FPG≥7.0 mmol/L or 2 h PG≥11.1 mmol/L, the detection rates were 12.52% for T2DM and 26.99% for DM related high-risk population. With HbA1c≥6.5% for the screen of DM, the rate of missed diagnosis was 46.59%. With FPG≥7.0 mmol/L, the rate of missed diagnosis was 50.02%. Based on the ROC curve, the optimal cut-points of FPG related to the OGTT-diagnostic criteria for DM was 6.1 mmol/L which was associated with a sensitivity and specificity of 75.5% and 90.3% respectively. The optimal cut-points of HbA1c related to T2DM diagnosis by OGTT&nbsp;was 6.2%, which was associated with a sensitivity and specificity of 71.4% and 84.5% respectively. The optimal cut-points of 2 h PG related to T2DM diagnosis by OGTT was 11.1%, which was associated with a sensitivity and specificity of 87.7% and 100% respectively. The optimal cut-points of FPG related to the OGTT-diagnostic criteria for T2DM and DM related high-risk population was 5.7 mmol/L which was associated with a sensitivity and specificity of 62% and 87.2% respectively. The optimal cut-points of HbA1c related to T2DM and DM related high-risk population diagnosis by OGTT was 6.0%, which was associated with a sensitivity and specificity of 57.3% and 77.4% respectively. The optimal cut-points of 2 h PG related to T2DM and DM related high-risk population diagnosis by OGTT was 7.8%, which was associated with a sensitivity and specificity of 86.7% and 100% respectively. The screening model using FPG≥6.1 mmol/L or HbA1c≥6.2% had sensitivities 85.3% for detecting T2DM and using FPG≥5.6 mmol/L or HbA1c≥6.1% had sensitivities 74.4% for detecting T2DM and DM related high-risk population. Conclusions The simultaneous measurement of FPG and HbA1c might be a more sensitive and specific screening tool for identifying high-risk individuals for T2DM at an early stage.

Objective To investigate changes in bone mineral density (BMD) and analyze its related factors in community populations to provide the early diagnosis of osteoporosis (OP) and give right guidance to prevent osteoporosis. Methods The quantitative ultrasound BMD analyzer was used to measure BMD of heel in 8 711 adults in community. At the same time a questionnaire survey was conducted among these subjects. The data were analyzed by Logistic stepwise regression analysis. Results With the age changing, males and females tended to have different BMD variation. Male and female's BMD peak values were respectively in 35-age group, 30-age group and decreased as age increased. T value of BMD was different between male and female (-1.40(-2.0--0.7)vs-1.3(-2.0--0.5))(P<0.001). Both men and women had a higher incidence of low bone mass (53.34%vs 47.46%), and the difference was significant (P<0.001), whereas the incidence of osteoporosis after 55 years of age between men and women was statistically significance (P<0.05). BMD was the lowest for female than for male in 50-age group, and the prevalence rate of OP was remarkably increased. Analysis of related factors of BMD showed that body mass index (BMI), age, diabetes and menopause were risk facts. Conclusion The BMD among adults is related to many factors, in which the age, BMI and menopause are the most important factors. Much attention should be paid to low bone mass phenomenon, timely monitoring, timely intervention, develop healthy working and living habits, is important to the prevention of osteoporosis and its complications.

目的 探讨不同血压水平人群的颈动脉硬化改变分布情况及与血压水平的相关性.方法 按血压值＜120/80 mmHg、(120～129)/(80～84)mmHg、(130～139)/(85～89)mmHg)和≥140/90 mmHg分层随机抽取1600例社区居民为观察对象,分别为血压正常组、血压正常高值1期组、血压正常高值2期组和高血压组各400例,均应用彩色多普勒超声检测仪检测双侧颈动脉并测量颈动脉内-中膜厚度(IMT)及斑块形成情况,了解各组观察对象的颈动脉硬化程度及病灶分布情况.结果 (1)颈动脉硬化好发部位依次为颈动脉窦部、颈总动脉、颈内动脉和颈外动脉,硬化病灶分布间比较,差异均有统计学意义(P＜0.01).(2)不同血压水平的观察对象其颈动脉硬化检出率不同,正常血压组、血压高值1期组、血压高值2期组和高血压组检出率分别为42.5％、53.3％、55.5％和67.3％;除血压正常高值1期组与血压正常高值2期组之间的颈动脉硬化检出率无统计学差异外,其余各组间的颈动脉硬化检出率比较差异均有统计学意义(P＜0.01).结论 颈动脉硬化好发部位主要在颈动脉窦部和颈总动脉;颈动脉粥样硬化程度与血压水平呈明显正相关关系,颈动脉粥样硬化程度随着血压水平的增高而增高;可将IMT增厚和斑块形成作为全身动脉硬化的一个重点筛查指标.

目的 探讨以亚洲人骨质疏松自我筛查工具(OSTA)确定的不同骨质疏松风险与骨定量超声T值(QUS-T)的相关性.方法 采用定量超声骨密度仪,对8711例成年人进行左跟骨骨密度检测,获得QUS-T,完成身高、体质量测定.依据OSTA指数划分为低危组、中危组和高危组.依据WHO骨质疏松诊断标准分为骨量正常组、低骨量组和骨质疏松组.比较各组间的差异,分析OSTA与QUS-T值的相关性.结果 QUS-T值在低危组、中危组和高危组患者中逐渐降低,男性低于女性,分别为(-1.3874士1.2063)、(-1.4950±1.2126)、(-1.913±1.2472)和(-1.1228±1.8720)、(-1.5983±1.2572)、(-2.0094±1.2920),男性组间无差异,女性组间有极显著差异(P＜0.001).在骨量正常组、低骨量组、骨质疏松组QUS-T值逐渐降低,女性骨质疏松的发生率随着OSTA的降低显著增高,分别为9.80％、23.50％、45.65％.结论 随着OSTA指数降低、骨质疏松风险增高,QUS-T值逐渐下降,OSTA指数与QUS-T值呈线性正相关.

高血压是一种严重危害人类健康的常见病和多发病,而且是心脑血管病主要危险因素之一.已有大量的证据表明,降压治疗可以降低心血管病事件的发生率.我国也制定了相应的包含高血压患者合理用药的高血压防治指南,并不断修订[1,2],旨在通过降压治疗降低患者心血管病的发病和死亡率.

目的：探讨在职体检人群中高尿酸血症与心血管危险因素的关系。方法选取某公司在职人员健康体检资料，筛选出高尿酸血症组（HUA）468例，随机选取同期体检的正常血清尿酸组（ NUA）508例，比较2组血压、血脂、空腹血糖（FBG）、体质量指数（BMI）等心血管危险因素水平，并分析血清尿酸水平与心血管危险因素的关系。结果 HUA 组收缩压（SBP）、舒张压（DBP）、三酰甘油（ TG）、低密度脂蛋白胆固醇（ LDL - C）水平及 BMI 均明显高于NUA 组（P 均﹤0.01）。多元回归分析显示血清尿酸水平与 DBP、BMI、TG 和 FBG 密切相关（P ﹤0.05或0.01）。结论血清尿酸水平与心血管危险因素密切相关，并增加心血管病致病风险。

目的：分析在职体检人群中血清尿酸浓度及其与心血管疾病危险因素的关系。方法以2422名在职体检人员为研究对象，测量受检者的身高、体质量、体质量指数、收缩压、舒张压，同时检测血清尿酸、总胆同醇、三酰甘油、高密度脂蛋白胆固醇、低密度脂蛋白胆固醇、空腹血糖浓度。统计分析血清尿酸浓度及其与心血管病危险因素的关系。结果该组在职人群血清尿酸浓度、男性血清尿酸值明显高于女性（P均＜0．01）。各心血管危险因素在控制其他危险因素后与血清尿酸浓度进行偏相关分析，男性：体质量指数（r’＝0．193，P＜0．01）、舒张压（r＝0．043，P＜0.01）、三酰甘油（r’＝0．141，P＜0．01）均与血清尿酸浓度呈正相关；而年龄（r’＝-0．202，P＜0．01）、空腹血糖（r’＝-0．071，P＜0．01）均与血清尿酸浓度呈负相关；总胆固醇（r’＝0．002，P＞0．05）、低密度脂蛋白胆固醇（r’＝-0．01， P＞0．05）、高密度脂蛋白胆固醇（r’＝-0．035，P＞0．05）、收缩压（r’＝-0．008，P＞0．05）与血清尿酸浓度无相关性；女性：除体质量指数（r’＝0．212，P＜0．01）和低密度脂蛋白胆固醇（r’＝0．142，P＜0．05）与血清尿酸浓度呈正相关，其余指标均无相关性。结论本组体检人群血清尿酸浓度较高，尤其男性血清尿酸浓度升高与多种心血管危险因素相关。

目的 探讨非糖尿病人群中糖化血红蛋白(HbA1c)与心血管危险因素的关系,明确HbA1c对非糖尿病人群心血管危险因素的预测意义.方法 采用横断面研究方法,选择古城社区自愿参加糖尿病筛查的资料完整并符合入选条件的2 007例(男650例,女1 357例).以HbA1c 6.5％为糖尿病诊断界值,每0.5为级差,以高血压、血脂紊乱、超重或肥胖、年龄(男＞45岁、女＞55岁)、HbA1c≥6.0％及空腹血糖(FBG)≥6.1 mmol/L为心血管危险因素,对不同HbA1c组中的个体所拥有的危险因素进行个数统计分析.同时将受检者按照所拥有的危险因素个数分组为0,1,2,3及4个以上组,比较不同组的危险因素的均值.通过Logistic回归分析HbA1c与心血管危险因素的相关性.结果 ①随着HbA1c的升高,有3个以上危险因素的频数逐渐增多.②HbA1c在不同危险因素个数组的均值随危险因素个数的增加而增加,并且组间比较有显著性差异(P ＜0.001).③Logistic回归分析显示HbA1c与FBG受损及糖耐量减低相关性最强,年龄次之.结论 6.0％＜ HbA1c ＜6.5％是非糖尿病人群心血管危险因素高发人群;HbA1c水平与FBG受损及糖耐量减低高度相关,可能是独立于其他心血管危险因素以外的非糖尿病人群的心血管危险因素预测指标.

Objective To investigate the status of cardiovascular disease risk factors cluster in on-the-job check-up personnel and discuss comprehensive health management scheme.Methods 2422workers were made physical examinations.The cardiovascular disease risk factors were analyzed,and the suggestions about health management were proposed.Results The study indicated that 69.2%of the subjects have at least one of the following cardiovascular disease risk factors.Clustering of 2and more than 3risk factors was 22.4% and 16.1%respectively.Men had higher prevalence of overweight,obesity,high blood pressure,high blood sugar and high uric acid than women.Conclusion The risk of the cardiovascular disease would increase with age,especially overweight and obesity.So it was important significance of management and intervention the health program of lifestyle early would decrease the risk and delay the occur of cardiovascular disease.

为了全面掌握劳动者在岗的健康情况,了解相关健康基础资料,及时发现职业禁忌证.本文针对2008-2012年从事粉尘作业的企业在岗员工进行职业健康管理,连续5年高仟伏胸片结果追踪. 1 对象与方法 1.1 对象 以2008年5、6月分别体检的两企业粉尘作业在岗员工为基线,合计3790人.入选标准:2008-2012年连续5年在岗员工,并在我院进行职业健康体检的人员.排除标准:(1)离岗、换岗、新入及退休人员;(2)纵隔增宽、肺门增大、肺间质纤维化建议进一步检查以及治疗者.企业1为2112人,年龄21～53岁,平均(34.46土9.54)岁,男1997人,女115人;企业2为1678人,年龄19～54岁,平均(33.07±8.76)岁,男1459人,女219人.

Objective To explore combined application value of non-invasive detection methods of vascular function and status among adult smokers.Methods A total of 2739people over 18years who were residents from Shijingshan District of Beijing were involved in the follow-up survey.The smoking history and proportion of male and female were taken as investigation indicators.PWV and ABI as well as carotid artery were determined.Results Among 2739participants,952were males and 1787females.The smoker group included 462smokers with a smoking rate of 16.9%.In the smoker group,there were 417males with a smoking rate of 43.8%and 45females with a smoking rate of 2.5%.There were more males than females(P0.01).The detection rate of carotid IMT thickening,plaque formation and stenosis was higher in the smoker group than in non-smoking group.There was no difference in the PWV of two groups(P0.05),while there was difference in the ABI of two groups(P0. 05).Correlation analysis showed it was a positive correlation in smoking with carotid intimal thickening,carotid artery plaque and carotid artery stenosis(P0.01),after controlling for other risk factors partial correlation analysis was still a positive correlation(P0.01).There was a negative correlation in smoking with ABI(P0.01),and an irrelevant between smoking and PWV.Conclusion Smoking was an independent risk factors of carotid atherosclerosis change,therefore,it might be a more important significance for early prevention and treatment of atherosclerosis among smokers,early combined detect PWV,ABI and carotid indicators examination.

目的 综合、定量评价糖尿病(DM)与肝癌发生的相关关系.方法 全面检索Medline、Cochrane、Embase、CNKI和万方等中英文数据库,将符合研究标准的DM与肝癌发生相关的前瞻性研究文献进行荟萃分析,采用RevMan 5.1统计软件进行异质性检验和统计分析.结果 共有8项研究文献入选,共获得945975例研究对象,其中DM患者190006例,发生肝癌1439例.荟萃分析结果显示,DM组发生肝癌的风险是非DM组的2.09倍(95％CI为1.57～2.79);敏感性分析结果显示,在自然人群中,DM患者的肝癌发生风险较非DM者增加(RR=2.37,95％CI为2.03～2.77).结论 在自然人群中,DM是导致肝癌发生风险增加的危险因素.