外周血细胞参数对早期识别骨髓增生异常综合征患者诊断价值的研究

The Diagnostic Value of Peripheral Blood Cell Parameters for Early Recognition of Myelodysplastic Syndrome

目的:评估外周血细胞参数对早期识别骨髓增生异常综合征(MDS)患者的诊断价值。方法:回顾性选取2015年1月1日至2017年12月31日首诊于河北医科大学第二医院的MDS患者86例和非恶性克隆性贫血患者72例的临床及实验室检查数据,比较分析两组患者的外周血细胞参数,将有统计学差异的参数制作出受试者工作特性曲线(receiver operator characteristic curve,ROC curve),并建立二元Logistic模型,分别计算并比较联合多个及单个指标的ROC曲线下面积(AUC)和灵敏度、特异度、阳阴性似然比及诊断正确率以评估其诊断效能。结果:与非恶性克隆性贫血组患者相比,MDS组患者的白细胞数(WBC)、中性粒细胞百分比(NE%)、嗜酸性粒细胞百分比(E%)、嗜酸性粒细胞绝对值(E#)、血小板计数(PLT)、血小板比容(PCT%)明显减低,而淋巴细胞百分比(LY%)、嗜碱性粒细胞百分比(B%)、血小板分布宽度(PDW)明显升高。将上述指标制作成ROC曲线,AUC大于0.5的有AUC_(LY%)=0.718(P=0.040),AUC_(PDW=)0.674(P=0.044),AUC_(B%)=0.650(P=0.044)。建立二元Logistic回归模型,联合LY%、PDW、B%这3个指标所得的AUC_(PRE-4)为0.777(P=0.037),明显高于单个指标的AUC。取ROC曲线上的一点,当敏感度77.91%,特异度61.11%时,对应的临界值为0.47时,阳性似然比为2.00,阴性似然比为0.36,正确分类的案例比为54.40%。结论:外周血细胞参数中PDW、B%、LY%对早期识别MDS有一定的诊断价值。

Objective: To evaluate the diagnostic value of peripheral blood cell parameters for early recognition of myelodysplastic syndrome(MDS) patients. Methods: The clinical and laboratory data of 86 patients with MDS and 72 patients with non-malignant clonal anemia treated in first diagnosed in the Second Hospital of Hebei Medical University from January 1, 2015 to December 31, 2017 was retrospectively analyzed. The peripheral blood cell parameters of the patients in two groups were analyzed, generated the receiver operator characteristic curve(ROC curve) from the statistically significant parameters, the binary logistic model was build to calculate and compare the area under the ROC curve(AUC) combined with multiple indicators and individual indicators, sensitivity, specificity, positive and negative likelihood ratio, and diagnostic accuracy, the diagnostic efficacy of the patients was analyzed. Results: Compared with patients in the non-malignant clonal anemia group,white blood cell count(WBC), neutrophil percentage(NE%),eosinophil percentage(E%), eosinophil absolute value(E#), platelet count(PLT), platelet specific volume(PCT%) in the MDS patients were significantly reduced;while percentage of lymphocytes(LY%), basophilic percentage(B%), and the width of platelet distribution(PDW) significantly increased. The several ROC curves with the above indicators were established, which showed that AUC_(LY%)=0.718(P=0.040);AUC_(PDW)=0.674(P=0.044);AUC_(B%)=0.650(P=0.044) were>0.5. After established a binary logistic regression model, the AUC_(PRE-4) obtained by combining the three indicators of LY%, PDW and B% was 0.777(P=0.037), which was significantly higher than the AUC of any indicator alone. When the sensitivity was 77.91% and the specificity was 61.11%, the corresponding threshold value was 0.47, the positive likelihood ratio was 2.00, the negative likelihood ratio was 0.36, and the case ratio of correct classification was 54.40%.Conclusion: PDW, B% and LY% in peripheral blood cell parameters have certain diagnostic value for early recognition of MDS.