论文部分内容阅读
目的研究急性心肌梗死(AMI)患者高敏心肌肌钙蛋白T(hs-cTnT)水平与发病时间的关系。方法以2012年1月至2013年12月因胸痛于四川大学华西医院急诊科就诊的3 096例患者为研究对象,最终确诊为AMI患者1 082例,男性861例,女性221例,以胸痛发生至急诊就诊采血为发病时间,按hs-cTnT水平变化趋势将发病时间分为<1h、1~<2h、2~<3h、3~<4h、4~<5h、5~<6h、6~<7h、7~<8h、8~<12h、12~<16h、16~<24h、24~<48h、48~<72h、72~<96h、96~<144h、≥144h共16组,分析不同发病时间段患者人数及hs-cTnT水平,计算各发病时间段患者hs-cTnT水平低于传统hs-cTnT阈值(14ng/L)的比例。通过研究各发病时间段胸痛患者hs-cTnT水平及ROC曲线,比较新确定的cut-off值(分段阈值)与传统阈值14ng/L在AMI诊断中的价值。结果 AMI发生5h内,血清hs-cTnT水平缓慢升高,5h以后升幅较大,48h左右到达峰值水平(2 000ng/L左右),48~<96hhs-cTnT处于一个相对稳定期,随后迅速降低。1 082例AMI患者中,胸痛发生1h内就诊的患者hs-cTnT水平小于14ng/L的比例为53.3%,1~<2h为19.1%,2~<3h为9.4%,3~<4h为6.1%,4~<5h为2.8%。各时间段诊断AMI的cut-off值为:胸痛时间<3h,hs-cTnT cut-off值为13.5ng/L时,敏感性为81.8%,特异性为80.1%;胸痛时间为3~<6h,cut-off值为17.8ng/L时,敏感性为94.6%,特异性为84.3%;胸痛时间6~<12h,cut-off值为30.0ng/L时,敏感性为95.9%,特异性为85.5%;胸痛时间≥12h,cut-off值为58ng/L时,敏感性92.7%,特异性为93.3%。4条ROC曲线的曲线下面积均>0.5,均有诊断价值,且以≥12h组和6~<12h组诊断价值最高,3~<6h组诊断价值居中,<3h组诊断价值最低。6~<12h组分段阈值的敏感性与传统阈值相当,但特异性更高(P<0.05);≥12h组的分段阈值牺牲了部分敏感性(P<0.05),换取特异性的大幅提高(P<0.05)。结论胸痛时间<3h的hs-cTnT水平不足以诊断和排除AMI,应至少动态监测5h。胸痛时间≥6h者应使用较传统阈值更高的阈值以减少误诊。
Objective To study the relationship between the level of hs-cTnT and the onset time in patients with acute myocardial infarction (AMI). Methods From January 2012 to December 2013, 3 096 patients with chest pain in the emergency department of West China Hospital of Sichuan University were enrolled in the study. Among them, 1 082 were diagnosed as AMI patients, 861 males and 221 females, with chest pain According to the trend of hs-cTnT, the onset time was divided into <1h, 1 ~ <2h, 2 ~ 3h, 3 ~ 4h, 4 ~ 5h, 5 ~ 6h, <7h, 7 ~ <8h, 8 ~ <12h, 12 ~ <16h, 16 ~ <24h, 24 ~ <48h, 48 ~ <72h, 72 ~ <96h, 96 ~ <144h, The number of patients and the level of hs-cTnT at different stages of onset were calculated. The proportion of patients with hs-cTnT at each stage of onset was lower than the threshold of traditional hs-cTnT (14ng / L). Through the study of hs-cTnT levels and ROC curve in patients with chest pain at various time points, the value of the newly determined cut-off value (segmentation threshold) and the traditional threshold of 14 ng / L in the diagnosis of AMI were compared. Results Serum hs-cTnT level increased slowly within 5 hours after AMI. After 5 hours, the level of hs-cTnT increased sharply. The peak level reached about 48 hours (about 2 000 ng / L), and the phase of 48 h <96 hhs-cTnT was at a relatively stable phase, then decreased rapidly. 1 082 cases of AMI patients, the hs-cTnT level of less than 14ng / L in patients treated within 1h after chest pain was 53.3%, 19.1% in 1 ~ <2h, 9.4% in 2 ~ <3h and 6.1% in 3 ~ <4h %, 4 ~ <5h is 2.8%. The cut-off values of AMI in each time period were: chest pain time <3h, hs-cTnT cut-off value 13.5ng / L, the sensitivity was 81.8%, specificity was 80.1%; chest pain time was 3 ~ <6h , the sensitivity was 94.6% and the specificity was 84.3% when the cut-off value was 17.8ng / L; the sensitivity was 95.9% when the cut-off value was 30.0ng / L when the chest pain time was 6 ~ <12h, For 85.5%, chest pain time≥12h, cut-off value 58ng / L, the sensitivity was 92.7% and the specificity was 93.3%. The area under the curve of the four ROC curves were all> 0.5, all of which had diagnostic value. The diagnostic value was the highest in≥12h group and 6 ~ <12h group, and the diagnostic value in 3 ~ <6h group was the lowest. The sensitivities of segment thresholds in 6 ~ <12h group were the same as the traditional thresholds, but the specificity was higher (P <0.05); the partial thresholds of ≥12h group sacrificed partial sensitivities (P <0.05) Increase (P <0.05). Conclusions The level of hs-cTnT in chest pain time <3h is not enough to diagnose and exclude AMI. It should be monitored dynamically for at least 5h. Chest pain time> 6h should use a higher threshold than the traditional threshold to reduce misdiagnosis.