【作者】 王燕；

【导师】 何燕霞；

【作者基本信息】 郑州大学 ， 临床检验诊断学， 2012， 硕士

【摘要】 心力衰竭(heart failure, HF)是各种类型心脏病病情严重时的临床表现,临床许多疾病如冠心病、高血压、肺气肿都可以导致心脏病变,疾病发展到终末期便会出现心力衰竭的表现。高血压、冠心病是老年人的常见病和多发病,随着社会老龄人口不断增加,这些疾病的发病率也逐年上升；另外,医疗技术的发展和医疗水平的提升,急性心梗后患者存活机会增加,这些都使得全世界范围内心力衰竭患者人数显著增加。心力衰竭是一个慢性的连续发展过程,患者的短期和长期预后差,有报道称,慢性心力衰竭患者5年内的死亡率可达到60%,而病情严重患者1年的死亡率就能达到50%。一旦患上心力衰竭,患者常因症状复发而反复住院,给患者本人和家人带来了沉重的经济和心理负担,患者生活质量明显下降,有些患者甚至生活不能自理。心力衰竭已经成为严重影响全世界公众健康的疾病,因此寻找有效的预后预测指标,在疾病发展早期,对心力衰竭患者进行准确分层,早期发现预后差的患者,并对其进行积极有效的干预治疗,改善患者预后具有极其重要的临床意义。外周血中红细胞的形态、体积改变往往与血液系统疾病的发生和发展有关系,但在最近,研究者发现其与心血管系统疾病的预后也存在一定的关系,红细胞体积分布宽度(red cell distribution width, RDW)这个反映外周血中红细胞体积变化的指标,可能对心力衰竭患者的短期和长期预后有一定的预测价值。目的本研究通过回顾性分析心力衰竭住院患者入院RDW与出院后两年内预后的关系,来证实RDW与心力衰竭患者长期预后的相关性,分析RDW对心力衰竭患者两年期死亡风险的预测价值,通过比较不同病因心力衰竭患者中RDW预测价值的差异,为解释RDW预测患者预后的机制提供参考依据,并对心力衰竭患者RDW升高的原因进行初步探讨。材料与方法1研究对象选择2009年10月至2011年12月因心力衰竭在北京阜外心血管病医院心衰病房住院的患者1021例。排除病历资料不完整,年龄不足18周岁,有贫血、白血病、恶性肿瘤、严重肝肾疾病,消化道出血等疾病和其他可能改变RDW的疾病患者,并去除感染性心内膜炎,主动脉夹层,缩窄性心包炎,心包积液,肺血栓栓塞等患者。所选患者年龄在18岁至90岁,男性720例,女性301例。2研究方法2.1采集患者入院时的临床信息根据患者病历资料,收集患者入院时的年龄、性别、BMI、合并疾病(高血压、高脂血症、糖尿病)等。记录患者入院时的NYHA’心功能分级,超声心动图检查结果。2.2采集患者血液标本入院后次日清晨采集患者空腹血液。标本采集后及时送至临床检验中心,并尽快进行血液常规,生化和NT-proBNP等检查。血液常规和NT-proBNP检查用EDTA抗凝管,生化检查用无抗凝剂的干燥管采集。2.3检验仪器RDW、红细胞计数(RBC),血红蛋白(Hb)用Sysmex XE-2100(?)血细胞分析仪及配套试剂检测；谷丙转氨酶(ALT),天冬氨酸转氨酶(AST),总胆红素(TBil),血尿素氮(BUN),血肌酐(Cr),高敏C反应蛋白(hs-CRP)用OlmplusAU5100生化分析仪以及中生北控公司的试剂盒检测；NT-proBNP用Biomedica公司的酶联免疫试剂盒以及美国Bio-Tek公司的ELx800(?)标仪检测；左室射血分数(]LVEF)用Philips公司的iE33全数字心脏彩超诊断仪检测。2.4随访随访患者预后：对住院死亡患者,记录死亡时间和死亡原因；对病情好转出院患者,记录出院时间、出院前最后一次RDW检测结果,电话随访患者本人或家人,记录患者出院后的再住院和死亡情况。3统计学分析采用SPSS17.0统计软件进行分析。RDW用中位数[第一四分位数,第三四分位数,M(Q1,Q3)]表示。两均数比较用t检验或Mann-Whitney U检验,率比较用χ2检验,多样本比较用Kruskal Wallis H检验,两两比较用Bonferroni法。患者生存率的估计用Kaplan-Meier法,组间比较用Log-rank检验。ROC曲线分析RDW对心力衰竭患者死亡风险是否有预测价值,曲线下面积比较采用Z检验。Cox比例风险模型评估心力衰竭患者死亡风险的独立影响因素。RDW与其他指标间的相关性用Spearman分析,RDW的影响因素用多重线性逐步回归分析。P<0.05具有统计学意义。结果1.入选患者的一般情况：符合入选标准的患者有1021例,入院时中位数RDW为13.5%,RDW≥15%患者占17.0%。中位随访693天后,有116例失访,137例死亡,死亡率为15.1%,死亡患者中位数RDW为14.8%,明显高于生存者的13.3%(P<0.001)。死亡患者BMI和血压偏低,心率稍快,心功能较差,RDW和肝肾功能指标异常的人数较多。2.RDW升高与心力衰竭患者两年期死亡风险增加有关：RDW异常(≥15.0%)患者死亡率为34.8%,再住院率为31.6%,正常者死亡率为11.0%,再住院率33.6%。RDW异常患者的死亡率明显升高,生存时间明显缩短。Cox回归分析,RDW与心力衰竭患者两年期预后有关,RDW异常患者的死亡风险约为正常者的4倍(校正HR=3.77)。3.RDW对心力衰竭患者两年期预后有一定的价值：ROC曲线分析,曲线下面积(AUC)为0.716(P<0.001)。约登指数最大时,RDW截断值为13.45%,预测的敏感性为76.3%,特异性为56.7%,阳性预测值为24.2%,阴性预测值为93.7%。4.不同病因心力衰竭患者RDW、死亡率和生存时间不同：选取冠心病,扩张型心肌病和瓣膜病这三种主要病因患者进行研究。三组患者的RDW,死亡率和生存时间明显不同,其中瓣膜病和扩张型心肌病患者的RDW和死亡率明显高于冠心病,生存时间明显短于冠心病,而瓣膜病和扩张型心肌病患者间无差异。5.RDW对不同病因心力衰竭患者两年期死亡风险的预测价值不同：ROC曲线分析,RDW对冠心病心衰和扩张型心肌病心衰死亡有预测价值,AUC分别为0.704、0.753(P均<0.001),对瓣膜病心衰死亡无预测价值,AUC为0.593(P=0.168)。约登指数最大时,RDW对冠心病心衰死亡预测的敏感性是59%,特异性是75%,对扩张型心肌病心衰的敏感性是60%,特异性是79%。6.RDW升高的相关影响因素分析：随着RDW升高,患者的BMI和LVEF逐渐降低,心功能分级增加。RDW与BMI、LVEF和心功能分级均相关,相关系数分别为-0.230、-0.261、0.357(P均<0.001)。RDW与Hb、TBil、hs-CRP、和NT-proBNP相关,相关系数分别为-0.092、0.363、0.205、0.442(P均<0.01)。多重线性逐步回归分析,RDW升高与BMI下降,NYHA心功能分级上升,NT-proBNP、TBil和hs-CRP增加有关,其中以胆红索代谢异常对RDW的影响最大。结论1.RDW可以作为心力衰竭患者的预后指标,对两年期死亡风险的预测敏感性与NT-proBNP相似。2.不同病因心力衰竭患者,RDW预测预后的价值不同。RDW可作为冠心病心衰和扩张型心肌病心衰的预后指标,而对瓣膜病心衰预后没有预测价值,这可为RDW在临床的具体应用提供依据。3.心力衰竭患者RDW升高受很多因素的影响,总胆红素升高是一个重要的影响因素,但心力衰竭预后不良患者RDW升高的原因,还需要后续的研究进行阐述。更多还原

【Abstract】 Heart failure (HF) is the terminal stage of heart disease caused by various reasons. Many clinical diseases such as coronary heart disease, hypertension and pulmonary emphysema can lead to pathological changes of the heart. When diseases progress in the end-stage, they would turn out symptoms of heart failure. Along with the number of aged people increasing, incidences of coronary heart disease and hypertension, which is common and frequently-occurring in old people, increase year after year. In addition, as the medical technology developping, diagnosis and treatment improving, there are marked inceases in survival rates of heart disease patient, which leads to the increasing significantly in the patients of the world. Heart failure is a chronic and continuous developing process with poor short and long-term prognosis, and with high mortality. It is reported that mortality of patients with chronic heart failure can reach60%in five years;1-year mortality of patients with worse illness can achieve50%. Once suffering from heart failure, patients have to repeatedly hospitalize due to recurrence of heart failure symptoms. The heavy economic and psychological burdens on patients themselves and their families decrease patients’life quality. Some patients can even not take care of themselves. Heart failure becomes a serious disease that impacts on public health around the world, so it is necessary to find effective prognostic indicators that can accurately stratify patients with poor prognosis in the early development of the disease, and then clinicians can execute positive and effective intervention treatments to improve patient outcomes.The changes in the shape and size of red blood cells in the peripheral blood are often associated with the occurrence and development of blood system diseases. However, researchers have found that it has a certain relationship with the prognosis of cardiovascular diseases recently. Red blood cell distribution width (RDW), an index reflecting the red blood cell size changes in peripheral blood, may have a certain value in predicting short and long-term prognosis in patients with heart failure.ObjectiveThe purpose of this thesis is to confirm the relationship between RDW and long-term prognosis of heart failure, and to analyze the predictive value of RDW on two-year mortality risk by a retrospective analysis of patients admitted to hospital for heart failure. Then in order to explain the mechanism of RDW predicting the prognosis, the differences of predictive value of RDW in different causes of heart failure are compared. At last, the reasons why RDW increased in heart failure patients are discussed.Materials and Methods1Object1021consecutive patients with heart failure hospitalized in heart failure ward from October2009to December2011. Excluding the following patients:with incomplete medical records, under18years old, with diseases such as anemia, leukemia, malignant tumor, serious liver or kidney disease, digestive tract hemorrhage, and other diseases that could possibly change RDW. The patients with infectious endocardium inflammation, aortic dissection, constriction pericarditis, hydropericardium and pulmonary thromboembolism were also excluded. Selected patients were aged from18to90years old, with720males and301females. 2Methods2.1Collect patients’clinical information on admissionAccording to the medical records of patients, information of age, gender, BMI and complication diseases, such as hypertension, hyperlipidemia and diabetes, were collected on admission. NYHA heart function classification, ultrasonic cardiogram results on admission, as well as the earlist routine blood test and biochemical test results after admission was recorded.2.2Collect patients’blood samplesFasting blood were collected from patients by highly trained nurses the next morning after admission. Specimen was timely sent to the Clinical Laboratory Center after collection and executed routine blood, biochemical, amino-terminal B-type natriuretic peptide (NT-proBNP) and other tests as soon as possible. Routine blood and NT-proBNP tests were collected in EDTA anticoagulant tubes, and biochemical tests in tubes without anticoagulant2.3Testing equipmentRDW, Red blood cell count (RBC), Hemoglobin (Hb) were tested by Sysmex XE-2100blood cell analyzers and mating reagents; Alanine Aminotransferase (ALT), Aspartate Aminotransferase (AST), Total Bilirubin (TBil), blood urea nitrogen (BUN), serum creatinine (Cr); high-sensitivity C-reactive protein (hs-CRP) by Olmplus AU5100biochemistry analyzer and the reagents produced by Biosino bio-technology and science incorporation; NT-proBNP with enzyme-linked immunosorbent kit produced by Biomedica company and ELx800microplate reader producted by Bio-Tek; left ventricular ejection fraction (LVEF) with Philips iE33all-digital echocardiography.2.4Follow-upPrognosis of patients was followed-up:as for patients who died in hospital, time of death and causes of death were recorded; for patients discharged, time of discharge and the last RDW test results before discharge were recorded; patient himself or his family was followed-up by telephone, situation of re-hospitalization or death were recorded after discharge. Readmission is that patients hospitalized with heart failure or symptoms appear again (including inpatients and emergency hospitalization). Death is all causes of death after discharge.3Statistics analysisThe data was analyzed statistically by SPSS17.0version.RDW was non-normal distribution data and representated as median [first quartile, third quartile, M (Q1, Q3)]. Two group compared by t test or Mann-Whitney U test or Kruskal Walks’H test, further comparisons by Bonferroni. The survival rate of patients estimated by Kaplan-Meier and log-rank tested the difference between two groups. ROC curve analysed the predictive value of RDW for death risk in heart failure patients, areas under the curve were compared by Z test. Cox proportional hazards models assessed the independent predictors of death risk in patients with heart failure. Spearman analysed the correlation between RDW and other indicators. The RDW impact factors were analysed by multiple linear stepwise regression. P<0.05means statistically significance.Results1General information of selected patients:1021cases meeting the admission criteria, the median RDW in these patients was13.5%, RDW≥15%patients account for17%. After follow-up a median693days,116cases lost, loss rate was11.4%.137cases died during follow-up, mortality rate was15.1%, RDW in patients of death was14.8%, signifficantly higher than that(13.3%) of survivors(P<0.001). Death patients have lower BMI and blood pressure, faster heart rate, poorer heart function, higher RDW and more patients with liver and kidney abnormal.2Increased RDW related to the increased two-year risk of death and rehospitalization:The mortality rate of patients with abnormal RDW (≥15.0%) was34.8%, readmission rate was31.6%, mortality rate of normal patients was11.0%, re-hospitalization rate was33.6%. The mortality rate of patients with abnormal RDW was significantly higher than that of patients with normal RDW, and the survival time was shorter. Cox regression analysis showed that RDW was related to two-year prognosis in heart failure patients and the risk of death with abnormal RDW was about4times higher than that of normal patients(adjusted HR=3.77).3RDW had a certain predictive value for two-year prognosis in patients with heart failure:ROC curve analysis, area under the curve (AUC) was0.716(P<0.001). When Youden index reached the maximum, cutoff value of RDW was13.45%, the predicted sensitivity76.3%, specificity56.7%, positive predictive value24.2%and negative predictive value93.7%.4RDW, mortality and survival time in patients with different causes of heart failure were different:Coronary heart disease, dilated cardiomyopathy and valvular heart disease patients, three major causes of heart failure, were selected as investigated subject. RDW, mortality and survival time were significantly different in three groups. RDW and mortality in valvular heart disease and dilated cardiomyopathy patients were significantly higher than that of coronary heart disease, survival time was shorter, while there was no difference between patients of valvular heart disease and dilated cardiomyopathy.5The predictive value of RDW for the two-year risk of death in different causes of heart failure patients was different:ROC curve analysis showed that RDW had predictive value for death in heart failure caused by coronary heart disease and dilated cardiomyopathy, AUC was0.704,0.753, respectively (allP<0.001); no value in valvular heart disease, AUC was0.593(P=0.168). When Youden index reached the maximum, the sensitivity of RDW in coronary heart disease was59%, specificity75%, the sensitivity in dilated cardiomyopathy was60%and a specificity of79%.6The related risk factors of elevated RDW:As RDW elevated, the patient’s BMI and LVEF decreased and heart function become worse. The correlation coefficients of RDW with BMI, LVEF and cardiac function classification were0.230,-0.261,0.357, respectively (allP<0.001). RDW related to Hb, TBil, hs-CRP and NT-proBNP, correlation coefficients were-0.092,0.363,0.205,0.442, respectively (all P<0.01). Multiple linear stepwise regression analysis showed that increased RDW was affected by BMI decline, NYHA functional classification rise, NT-proBNP, TBil and hs-CRP increase, but the biggest impact on RDW was bilirubin metabolism abnormity. Conclusions1. RDW can be used as a prognostic indicator in patients with congestive heart failure, forecasts sensitivity for the two-year mortality risk is similar to NT-proBNP.2. In patients with different etiology, predicting values of RDW are different. RDW can be used as prognostic indicators of heart failure caused by coronary heart disease and dilated cardiomyopathy, but have no predictive value on valvular heart disease, which may provide the basis for RDW in specific clinical application.3. Higher RDW in patients with heart failure under the influence of many factors, elevated total bilirubin is an important factor. Causes of RDW elevated in patients with congestive heart failure with poor prognosis, follow-up studies are needed to explain.更多还原