B型钠尿肽对社区获得性肺炎危险分层的价值

B-type natriuretic peptide in risk stratification of community-acquired pneumonia

目的探讨B型钠尿肽(BNP)的动态变化水平与社区获得性肺炎(CAP)患者病情危重情况之间的关系,从而探索BNP对CAP患者危险分层的价值。方法选取2017年1月至2018年1月首都医科大学附属复兴医院急诊科收治的58例CAP患者为研究对象,测定患者的BNP、C反应蛋白(CRP)水平,以及肺炎严重度评分(PSI)所需相关指标,计算PSI评分。根据病情程度将患者分为重症肺炎组(n=12)和非重症肺炎组(n=46),分析不同组患者的BNP水平;根据患者预后分为幸存组(n=49)和死亡组(n=9),比较两组患者间BNP差异、BNP水平与PSI评分的受试者工作特征(ROC)曲线。结果就诊时,38例患者有呼吸急促,25例患者血氧饱和度低于92%。PSI评分为(92±19)分,CRP中位值为110 mg/L(38,206),BNP中位值为170 pg/ml(72,528)。BNP水平随着CAP严重程度(PSI评分)增加而升高(r=0. 787,P <0. 01),重症肺炎组患者BNP水平显著高于非重症肺炎组[(459. 62±27. 83) pg/ml vs.(210. 35±23. 41) pg/ml,P <0. 05]。相比幸存组的BNP值[中位值为160 pg/ml(61,291)],死亡组的BNP值明显升高[中位值为372 pg/ml(291,1 300),P=0. 002]。BNP能预测死亡(AUC=0. 821),其预测死亡最佳截点为282. 0 pg/ml,灵敏度为89%,特异性为76%,阳性预测值为40%,阴性预测值为97%。使用BNP预测死亡的ROC曲线下面积(0. 821,95%CI:0. 697～0. 945)高于PSI评分的曲线下面积(0. 553,95%CI:0. 362～0. 745)。结论 BNP预测CAP严重程度具有一定的实用性,并且可以早期识别高危患者,优化医疗资源利用。

Objective To investigate B-type natriuretic peptide regarding risk stratification of patients with community acquired pneumonia( CAP). Methods A total of 58 patients with CAP admitted to the emergency department of our hospital from January 2017 to January 2018 were enrolled. All patients were tested for BNP level,C-reactive protein( CRP),as well as other component markers of pneumonia severity index( PSI). They were divided into severe pneumonia group( n = 12) and non-severe pneumonia group( n = 46) as well as survivor group( n =49) and death group( n = 9). BNP levels between the two groups were compared,and ROC curve analysis was performed regarding BNP levels versus PSI. Results On admission,38 patients had shortness of breath. 25 patients had less than 92% of blood oxygen saturation. The PSI was92 ± 19,median CRP being 110 mg/L( 38,206) and median BNP 170 pg/ml( 72,528). BNP levels increased with CAP severity( r = 0. 787,P < 0. 01). BNP level in severe group was significantly higher than in non-severe group [( 459. 62 ± 27. 83) pg/ml vs.( 210. 35 ± 23. 41) pg/ml,P < 0. 05]. Compared with survivor group [n = 49; median 160 pg/ml( 61,291) ],those died in 30 days had significantly higher BNP level[n = 9; median 372 pg/ml( 291,1300),P = 0. 002]. BNP level had definite accuracy in predicting mortality( AUC = 0. 821) with optimal cut-off point 282. 0 pg/ml. Sensitivity: 89%. Specificity: 76%. Negative predictive value: 97%. Positive predictive value: 40%. The AUC( 0. 821; 95% CI,0. 697-0. 945) under BNP mortality prediction curve was higher than AUC of PSI curve( 0. 553; 95% CI,0. 362-0. 745).Conclusion BNP is positively correlated with CAP risk stratification and can directly reflect CAP severity. Early identification of high-risk patients can optimize resource allocation.