【作者】 龙曼；

【导师】 陈爱华；

【作者基本信息】 南方医科大学 ， 临床医学（专业学位）， 2013， 博士

【摘要】 研究背景及目的冠心病(Coronary heart disease, CHD)是多数发达国家45岁以上人群常见的心血管疾病,在发展中国家发病率也越来越高。世界卫生组织(world health organization, WHO)预测,在世界范围内的传染性疾病病死率得到有效控制之后,到2020年CHD将成为世界第一死因。急性冠脉综合征(acute coronary syndrome, ACS)是CHD中最严重临床表现,包括ST段抬高型心肌梗死(ST-segment elevation myocardial infarction、非ST段抬高型心肌梗死(non-ST-segment elevation myocardial infarction, NSTEMI)和不稳定型心绞痛(unstable angina pectoris, UAP)以及心源性猝死。CHD患者一旦发生ACS,危害极大,未来发生心血管死亡、非致死性心肌梗死、因心绞痛再次入院、再次接受血管重建、心功能恶化、脑卒中等不良心血管终点事件的机率也明显上升。传统的危险因素及心肌损伤的标志物如肌钙蛋白(cardiac troponin, cTn)尚不能准确的评估ACS预后,而无传统危险因素或cTn阴性的ACS患者未来心血管不良事件发生的风险常被低估。cTn阴性患者在6个月发生死亡的风险为5.9%,复发心肌梗死的风险约为5.9%。大量研究致力于寻找新的可准确评估ACS患者预后的标志物。妊娠相关蛋白A (pregnancy-associated plasma protein-A, PAPP-A)是新发现的冠状动脉粥样斑块不稳定性的生物标志物之一。动物及人体解剖结果显示PAPP-A在破裂的斑块局部高表达,ACS患者血浆PAPP-A水平与稳定型冠心病(stable angina pectoris, SAP)及健康人群相比显著升高。然而,对于PAPP-A能否预测ACS患者未来不良心血管终点事件的发生仍存在争议,有的研究认为PAPP-A可预测ACS患者远期不良预后,而有的研究则得出相反的结论,认为PAPP-A不能预测ACS患者远期预后。本项研究通过Meta分析方法对PAPP-A应用于ACS患者预后评估的文献进行汇总,为全面了解ACS患者外周血PAPP-A水平升高与不良心血管终点事件的危险性之间的关系。材料与方法本文主要遵循观察性流行病学研究Meta分析的相关指南进行资料收集及分析,并遵守系统综述和Meta分析报告规范。我们对Pubmed, EMBASE, OVID, Web of Knowledge, the Cochrane library和相关综述的参考文献进行了全面检索,时间截止到2012年9月30日。文献入选标准为：(1)队列研究或者随机对照试验(Randomized controlled trail, RCT)的事后分析,RCT的事后分析是评价研究人群基线资料中干预措施以外的某一个暴露因素(PAPP-A的高低)与预后的关联的研究；(2)随访时间不少于1个月；(3)研究对象为确诊ACS患者及以急性胸痛入院可疑ACS患者；(4)结局指标为：死亡、心血管死亡、因心绞痛再次入院、再次进行血管重建及非致死性心肌梗死；(5)文献中提供了PAPP-A对ACS患者预后关联的风险效应指标(HR. RR或OR值)及其95%CIs；(6) PAPP-A检测样本采取时间在胸痛症状出现后72小时内。采用观察性研究Newcastle-Ottawa Scale (NOS)质量评分标准对纳入文献进行质量评价。采用随机效应模型,RR值为效应指标合并PAPP-A在ACS预后的相关研究。采用I2指数进行异质性检验,并进行敏感性分析观察单个临床试验研究对总体分析结果的影响。按照研究变量(研究设计类型、来源国家、随访时间长短、患者平均年龄、PAPP-A检测是否受到肝素影响、风险效应指标的选择、是否对多因素进行调整)分组进行亚组分析以探索可能的异质性来源。采用漏斗图及Egger提出的线性回归法进行发表偏倚的识别和评估。采用Revman5.1及Stata10.0软件进行数据处理与分析。所有统计学指标均为双侧,P<0.05认为差异有统计学意义。结果本研究共纳入14篇文献,涉及到9413例ACS患者,总的Meta分析结果显示早期PAPP-A升高的人群未来发生死亡、心血管死亡、因心绞痛再次入院、再次进行血管重建及非致死性心肌梗死的相对风险是低水平PAPP-A人群的1.97倍(RR1.97,95%CI:1.49-2.60, P<0.00001)。PAPP-A升高与ACS患者未来不良心血管终点事件发生相关,PAPP-A可能是ACS患者预后的预测因子。1.不同研究设计类型亚组分析的结果纳入的14篇文献中研究设计类型为RCT的事后分析有3篇,11篇为队列研究。在RCT的事后分析研究涉及的5962例ACS患者中,亚组分析结果显示早期PAPP-A升高的人群未来发生死亡、心血管死亡、因心绞痛再次入院、再次进行血管重建及非致死性心肌梗死的相对风险是低水平PAPP-A人群的1.30倍(RR1.30,95%CI:1.08-1.57, P=0.006)。在队列研究涉及3451例ACS患者中,亚组分析结果显示早期PAPP-A升高的人群未来发生不良心血管终点事件的相对风险是低水平PAPP-A人群的2.32倍(RR2.32,95%CI:1.66-3.26, P<0.00001)。在队列研究中,PAPP-A升高与未来不良心血管终点事件发生之间的关联强度显著高于其在RCT的事后分析研究中与未来不良心血管终点事件发生的关联强度(RR2.32vs1.30,P=0.003)。2.不同来源国家亚组分析的结果纳入的14篇文献中在欧洲国家进行研究有8篇涉及3562例ACS患者,在非欧洲国家进行的研究有6篇涉及5851例ACS患者。在欧洲国家中,早期PAPP-A升高的人群未来发生死亡、心血管死亡、因心绞痛再次入院、再次进行血管重建及非致死性心肌梗死的相对风险是低水平PAPP-A人群的1.79倍(RR1.79,95%CI:1.18-2.71, P=0.006)。在非欧洲国家中,早期PAPP-A升高的人群未来发生不良心血管终点事件的相对风险是低水平PAPP-A人群的2.26倍(RR2.26,95%CI:1.48-3.44,P=0.0001)。在欧洲国际及非欧洲国家中,PAPP-A升高与未来不良心血管终点事件发生之间的关联强度无显著差异(RR1.79vs2.26,P=0.44),PAPP-A具有类似的预测作用。3.不同随访时间亚组分析的结果纳入的14篇文献中随访时间不足1年的研究有5篇涉及1367例ACS患者,随访时间超过1年的研究有9篇涉及8046例ACS患者。早期PAPP-A升高的人群1年内发生死亡、心血管死亡、因心绞痛再次入院、再次进行血管重建及非致死性心肌梗死的相对风险是低水平PAPP-A人群的3.33倍(RR3.33,95%CI:2.51-4.42,P<0.00001)。而早期PAPP-A升高的人群1年以后发生不良心血管终点事件的相对风险是低水平PAPP-A人群的1.42倍(RR1.42,95%CI:1.13-1.79,P=0.003)。PAPP-A升高与ACS患者短期预后(<1年)的关联强度显著高于其与ACS患者长远预后(≥1年)的关联强度(RR3.33vs1.42, P<0.00001)。4.不同平均年龄亚组分析的结果依据ACS患者人群平均年龄进行分组,平均年龄不足65岁的研究有9篇涉及4738例ACS患者,平均年龄大于65岁的研究有3篇涉及818例ACS患者。在平均年龄不足65岁亚组人群中,早期PAPP-A升高的人群未来发生死亡、心血管死亡、因心绞痛再次入院、再次进行血管重建及非致死性心肌梗死的相对风险是低水平PAPP-A人群的1.89倍(RR1.89,95%CI:1.26-2.84,P=0.002)。而在平均年龄大于65岁亚组人群中,早期PAPP-A升高的人群未来发生不良心血管终点事件的相对风险是低水平PAPP-A人群的2.26倍(RR2.26,95%CI:1.42-3.61,P=0.0006)。在平均年龄大于65岁人群及不足65岁人群中,PAPP-A升高与未来不良心血管终点事件发生之间的关联强度无显著差异(RR2.26vs1.89,P=0.57),PAPP-A具有类似的预测作用。5.是否受肝素影响亚组分析的结果纳入的14篇文献中PAPP-A的检测可能受到肝素干扰的研究有8篇涉及7023例ACS患者,PAPP-A的检测未受到肝素干扰的研究有6篇涉及2390例ACS患者。在使用肝素后采血检测的PAPP-A水平的升高可预测ACS患者预后,升高者未来发生死亡、心血管死亡、因心绞痛再次入院、再次进行血管重建及非致死性心肌梗死的相对风险是低水平PAPP-A人群的2.18倍(RR2.18,95%CI:1.50-3.16, P<0.0001)。采血前未使用肝素所检测的PAPP-A水平的升高同样可预测ACS患者预后,PAPP-A升高者未来发生不良心血管终点事件的风险是低水平人群的1.71倍(RR1.71,95%CI:1.06-2.75, P=0.03)。不论是否使用肝素,PAPP-A升高与未来心血管终点事件发生之间的关联强度无显著差异(RR2.18vs1.71,P=0.43)。6.不同风险效应指标亚组分析的结果14篇研究中以HR、RR为风险效应指标的研究各有6篇分别涉及7095例、1137例ACS患者, OR为风险效应指标的研究有2篇涉及1181例ACS患者。亚组分析结果显示PAPP-A在以OR为效应指标的研究中不能预测ACS患者预后,尽管PAPP-A升高患者未来发生心血管终点事件的相对风险是低水平PAPP-A人群的1.88倍(RR1.88,95%CI:0.69-5.16, P=0.22),但差异不具有统计学意义。而在以HR、RR为效应指标的研究中PAPP-A均可预测ACS患者预后,且在以RR为效应指标的研究中PAPP-A升高与未来心血管终点事件发生之间的关联强度更强(RR1.37vs3.15, P=0.0002),差异具有统计学意义。在以HR为效应指标的研究中PAPP-A升高患者未来发生不良心血管终点事件的相对风险是低水平PAPP-A人群的1.37倍(RR1.37,95%CI:1.06-1.77,P=0.02)。在以RR为效应指标的研究中PAPP-A升高患者未来发生不良心血管终点事件的相对风险是低水平PAPP-A人群的3.15倍(RR3.15,95%CI:2.33-4.26, P<0.00001)。7.是否对多因素调整亚组分析的结果14篇文献中有7篇研究对未调整的效应值及经多因素调整后的效应值均进行了报道,共涉及7442例ACS患者。对未调整的效应值进行Meta合并后显示PAPP-A升高患者未来发生不良心血管终点事件的相对风险是低水平PAPP-A人群的1.86倍(RR1.86,95%CI:1.43-2.43, P<0.00001)。在对年龄、性别、吸烟、高血压、高血脂、糖尿病等危险因素进行调整后的效应值进行Meta合并显示,在排除上述传统危险因素的干扰后,PAPP-A升高患者未来发生不良心血管终点事件的相对风险是低水平PAPP-A人群的1.66倍(RR1.66,95%CI:1.14-2.41, P=0.008)。PAPP-A对ACS患者预后的预测作用不受传统的危险因素等影响,多因素调整后合并RR值与调整前的合并RR值相比并无显著降低(RR1.86vs1.66, P=0.61), PAPP-A可能是ACS预后的独立预测因子。结论1.对纳入14篇文献总的Meta分析结果显示在9413名ACS患者中,PAPP-A升高患者未来1个月至未来47个月内发生死亡、心血管死亡、因心绞痛再次入院、再次进行血管重建及非致死性心肌梗死的相对风险是低水平PAPP-A人群的1.97倍(RR1.97,95%CI:1.49-2.60, P<0.00001)。PAPP-A升高与ACS患者未来不良心血管终点事件发生相关,提示PAPP-A可能是ACS患者预后的预测因子。2. PAPP-A升高除了在选用OR为风险效应指标的亚组研究中与ACS预后无显著相关,而在其他不同的亚组研究中PAPP-A升高与ACS患者未来不良心血管终点事件的发生均显著相关,PAPP-A升高的患者发生不良预后的风险可能增加。3. PAPP-A在欧洲及非欧洲国家、平均年龄大于65岁及不足65岁ACS人群中具有类似的预测作用,且其预测作用与PAPP-A测量是否受肝素影响、是否对传统危险因素进行调整无关,PAPP-A可能是ACS患者预后独立的预测因子。但随着随访时间的延长,PAPP-A升高与预后的关联强度减弱,且其关联强度在RCT的事后分析和队列研究中、在选用HR和RR研究中具有显著差异。随访时间的长短、研究设计类型、风险效应指标值的不同是不同研究之间异质性的主要来源。更多还原

【Abstract】 Background and Objection:Coronary heart disease (CHD) is one of the most common cardiovascular disease in people over45years in most developed and developing cpumrries.World health organization forecast that CHD will become the first cause of death all.over world by the year2020, after effective controls of severe infectious doseases Acute coronary syndrome (ACS) including ST-segment elevation myocamdial infarciton (STEMI), non-ST-segment elevation myocardial infarction (NSTEMI),unstable angina pectoris (UAP) and sudden cardiac death is the most serious clinical manifestations of CHD. After the onset of ACS, the future risk of adverse cardiovascular outcomes such as cardiovascular death, non-fatal myocardial infarction, rehospitalisation and revascularization owing to angina, heart failure, stroke obviously increase.Traditional cardiovascular risk factors and biomarkers have been used to predict the occurrence of adverse outcomes after ACS. However, there are still patients whose risk of adverse cardiovascular outcomes is underestimated. The mortality rate of cardiac troponin (cTn) negative patients in6month after ACS was5.9%and the risk rate of recurrent myocardial infarction was alomst5.9%. A lot of studies are dedicating in finding new biomarkers which could accurately predict the prognosis of patients with ACS.New markers of plaque instability have been identified in recent years, among which, pregnancy-associated plasma protein-A (PAPP-A) appears to offer an interesting profile. Data from animal and human histological study suggest that PAPP-A is markedly enriched in ruptured and eroded unstable plaques of coronary arteries. Circulating levels of PAPP-A have been shown to be elevated in patients with ACS compared with patients with chronic stable angina (SAP) or healthy subjects. A large number of studies have examined the prognostic value of PAPP-A in patients with ACS but the results were conflicting. Some studies found that PAPP-A was a valuable predictor of adverse cardiovascular outcomes in patients with ACS, while others found no prognostic value of PAPP-A. There is an increasing need to synthesise all the evidence. Herein, we present this meta-analysis to assess the association between higher PAPP-A levels and risk of long-term adverse cardiovascular outcomes after ACS occurs.Methods and materialsThis study was conducted following the meta-analysis of observational studies in epidemiology guidelines, simultaneously conforming to most of the preferred reporting items for systematic reviews and meta-analyses guidelines.We performed an electronic search of PubMed, EMBASE, OVID, Web of Knowledge and the Cochrane library from inception of each database to September,2012. We included studies if they met all the following criteria:(1) cohort studies or post hoc analyses of randomized controlled trial(RCT);(2) studies with a follow-up duration of at least1month;(3) subjects were patients with angiographically vaidated actue coronary syndrome and patients with acute chest pain in a heterogeneous emergency room;(4) outcomes were death, cardiovascular death, rehospitalisation and revascularization owing to angina and non-fatal myocardial infarction;(5) risk estimates with95%confidence intervals(CIs) were reported or could be calculated.(6) blood samples for PAPP-A assays were obtained within72h from the onset of symptoms or on admission to hospital. The quality of included studies were assessed based on the Newcastle-Ottawa Scale for quality of observational studies in meta-analyses. Random effects meta-analysis and relative risk (RR) were used to estimate the association between PAPP-A levels and adverse cardiovascular outcomes after ACS. Heterogeneity was assessed using the Higgins I-squared statistic (I2). Sensitivity analyses were carried out to characterize possible sources of statistical heterogeneity, by excluding studies one-by-one.Subgroup analyses based on study design, study location, duration of follow-up, mean age, heparin use, measurement of association chosen and confounder factors adjustments were conducted to identify the risk-subgroup interactions that could explain the inter-study differences.Publication bias was assessed using funnel spot and the Egger regression asymmetry test. Software Review Manager (RevMan5.1, Copenhagen) and Stata (Statal0.0, Texas) were used for the statistical analyses. All the P values were two tailed, and P values<0.05were considered statistically significant.Results:Fourteen studies were included in our meta-analysis involving9413patients. The overall meta-analysis showed that patients with higher PAPP-A values had1.97times the risk of death, cardiovascular death, rehospitalisation and revascularization owing to angina and non-fatal myocardial infarction for patients with lower PAPP-A levels (RR1.97,95%CI:1.49-2.60, P<0.00001). Higher PAPP-A values were associated with increased adverse cardiovascular outcomes, suggesting that PAPP-A could be a valuable predictor of adverse cardiovascular outcomes in patients with ACS.1. Subgroup analysis based on study designThere are3post hoc analyses of RCT involving5962patients and11cohort studies involving3451patients. In post hoc analyses of RCT, patients with higher PAPP-A values had1.30times the risk of adverse cardiovascular outcomes for patients with lower PAPP-A levels (RR1.30,95%CI:1.08-1.57, P=0.006). In cohort studies, patients with higher PAPP-A values had2.32times the risk of adverse cardiovascular outcomes for patients with lower PAPP-A levels (RR2.32,95%CI:1.66-3.26, P<0.00001). The association between higher PAPP-A values and increased adverse cardiovascular outcomes in post hoc analyses of RCT was statistically bigger than that in cohort studies (RR2.32vs1.30, P=0.003).2. Subgroup analysis based on study locationEight studies were from Europe countries involving3562patients and6studies were from non-Europe countries involving5851patients. In studies from Europe countries, patients with higher PAPP-A values had1.79times the risk of adverse cardiovascular outcomes for patients with lower PAPP-A levels (RR1.79,95%CI:1.18-2.71, P=0.006). In studies from non-Europe countries, patients with higher PAPP-A values had2.26times the risk of adverse cardiovascular outcomes for patients with lower PAPP-A levels (RR2.26,95%CI:1.48-3.44, P=0.0001). The associations between higher PAPP-A values and increased adverse cardiovascular outcomes in European studies was not significantly different from that in non-European studies (RR1.79vs2.26, P=0.44), suggesting that PAPP-A had similar prognostic values in both European and non-European studies.3. Subgroup analysis based on duration of follow-upThere are5studies with a duration of follow-up less than1year involving1367patients and9studies with a duration of follow-up more than1year involving8046patients. The future risk of adverse cardiovascular outcomes within1year in patients with higher PAPP-A values is3.33times as compared with patients with lower PAPP-A levels (RR3.33,95%CI:2.51-4.42, P<0.00001). The future risk of adverse cardiovascular outcomes more than1year in patients with higher PAPP-A values is1.42times as compared with patients with lower PAPP-A levels (RR1.42,95%CI:1.13-1.79, P=0.003). The association between higher PAPP-A values and increased short-term(<1year) outcomes was statistically bigger than the association between higher PAPP-A values and increased long-term(≥1year) outcomes (RR3.33vs1.42, P<0.00001).4. Subgroup analysis based on mean age There were9studies in which mean age of patients involved were younger than65years and3studies in which mean age of patients involved were older than65years. In subgroup analysis of4738patients younger than65years, those with higher PAPP-A values had1.89times the risk of adverse cardiovascular outcomes for those with lower PAPP-A levels (RR1.89,95%CI:1.26-2.84, P=0.002). In subgroup analysis of818patients older than65years, those with higher PAPP-A values had2.26times the risk of adverse cardiovascular outcomes for those with lower PAPP-A levels (RR2.26,95%CI:1.42-3.61, P=0.0006). The associations between higher PAPP-A values and increased adverse cardiovascular outcomes in patients older than65years was not significantly different from that in patients younger than65years(RR2.26vs1.89, P=0.57), suggesting that PAPP-A had similar prognostic values in both subgroups.5. Subgroup analysis based on heparin useThere were8studies involving7023patients in which boold samples of PAPP-A were obtained after heparin use and6studies involving2390patients in which boold samples of PAPP-A were obtained without heparin use. In studies in which blood levels of PAPP-A were affected by heparin use, patients with higher PAPP-A values had2.18times the risk of adverse cardiovascular outcomes for patients with lower PAPP-A levels (RR2.18,95%CI:1.50-3.16, P<0.0001). In studies in which blood levels of PAPP-A were not affected by heparin use, patients with higher PAPP-A values had1.71times the risk of adverse cardiovascular outcomes for those with lower PAPP-A levels (RR1.71,95%CI:1.06-2.75,P=0.03). No significant differences were found in the risk estimates between the two subgroups with and without heparin us (RR2.18vs1.71, P=0.43).6. Subgroup analysis based on measurement of association chosenThere were respectively6studies used HR and RR as the measurement of rsik estimates and the other2studies chose OR as the measurement of rsik estimates. In studies involving1181patients in which OR was estimated, subgroup analysis suggested that PAPP-A could not predict adverse cardiovascular outcomes(RR1.88, 95%CI:0.69-5.16, P=0.22). PAPP-A had lower prognostic value in subgroup analysis of studies in which HR was estimated as compared with that in subgroup analysis of studies in which RR was used(RR1.37vs3.15, P=0.0002). In studies in which HR was estimated, patients with higher PAPP-A levels were associated with1.37-fold higher risks of adverse cardiovascular outcomes as compared with those with lower PAPP-A levels(RR1.37,95%CI:1.06-1.77, P=0.02). In studies in which RR was estimated, patients with higher PAPP-A levels were associated with3.15-fold higher risks of adverse cardiovascular outcomes as compared with those with lower PAPP-A levels(RR3.15,95%CI:2.33-4.26, P<0.00001).7. Subgroup analysis based on confounder factors adjustmentSeven studies reported both unadjusted and adjusted risk estimates involving7442patients. The pooled RR and its95%CI of unadjusted risk estimates was1.86(RR1.86,95%CI:1.43-2.43, P<0.00001). The pooled RR and its95%CI of adjusted risk estimates was1.66(RR1.66,95%CI:1.14-2.41, P=0.008). The prognostic value of PAPP-A was not significantly reduced after adjustment for confounder factors such as age, sex, smoke, hypertension, hyperlipidemia and diabetes mellitus (RR1.86vs1.66, P=0.61). The prognostic role of PAPP-A was independent of traditional risk factors and other markers.Conclusion1. This meta-analysis of14studies support an association between higher PAPP-A levels and increased risk of adverse cardiovascular outcomes in9413patients with ACS. In the follow-up duration that ranged from1to47months after ACS occurred, patients with higher PAPP-A values had1.97times in the risk of death, cardiovascular death, rehospitalisation and revascularization owing to angina and non-fatal myocardial infarction of patients with lower PAPP-A levels (RR1.97,95%CI:1.49-2.60, P<0.00001). PAPP-A could be a valuable predictor of adverse cardiovascular outcomes in patients with ACS.2. Higher PAPP-A values were statistically associated with increased adverse cardiovascular outcomes in all subgroup studies except the one using OR as the measurement of association chosen.3. The prognostic value of PAPP-A were similar in European and non-European countries as well as in patients older and younger than65years, and its prognostic value was not affected by heparin use and confounder factors. However, its predicting value decreased as time went by. Significant subgroup differences were observed in subgroups stratified by follow-up duration, study design and the measurement of association chosen, suggesting that those are the main sources of heterogeneities.更多还原