应用国际血细胞复检规则中可疑警示条款筛查初诊白血病的探讨

目的Beckman Coulter LH755血液分析仪显示的形态学可疑警示信息结合国际血细胞复检规则中相应条款，探讨其用于初诊白血病筛查的价值，为制定本室血细胞复检规则提供依据。方法1984例患者标本分为初诊非白血病组（1832例），初诊白血病组（73例）和复诊组（79例），所有标本同时进行仪器分析和手工白细胞分类及细胞形态观察，参照国际复检规则，对初诊非白血病组和A血病组标本涉及的国际血细胞可疑警示条款进行评价，观察白血病和非白血病患者治疗前后变化；运用受试者工作特征（ROC）曲线，观察其对初诊白血病筛查的价值。结果本组初诊病例724例涉及可疑警示条款，其中非白血病组655例，白血病组69例，46例髓系白血病均涉及可疑警示条款，且显微镜镜检阳性；其中条款34（左移报警）和37（原始细胞报警），敏感性和阳性预期值均较高，条款30[血小板（PLT）聚集报警]，31（PLT报警）和35[不典型和（或）变异Lym]阳性预期值较低；治疗后患者检出假阴性病例28例，其中8例检出原始细胞，这些标本均来自于白血病治疗后患者（28．57％，8／28）；采用ROC曲线分析，条款37（原始细胞报警）对初诊白血病价值最高（AUC=0．821），其次为条款34（左移报警），32（IG报警），26（白细胞结果不可靠）和35[不典型和（或）变异Lym]，而条款30（PLT聚集报警）和31（PLT报警）诊断价值较小。结论参照国际血细胞复检可疑警示条款对于初诊白血病筛查有重要价值，白血病患者治疗后细胞形态更为多样和复杂，仪器示警不能直接提供血细胞形态变化的确切信息，需进一步用显微镜镜检血涂片进行核实；当标本涉及条款37（原始细胞报警）时应予以显微镜镜检，以排除白血病可能，涉及条款26（白细胞结果不可靠），32（IG报警），34（左移报警），35[不典型和（或）变异Lym]时也应引起重视，以免造成漏诊和误诊．

Investigation of screening preliminary diagnosed leukemia according to the warning informations of suspicious morphology of the international blood cell review rules

Objective To investigate the value of screening preliminarily diagnosed leukemia according to the warning informations of suspicious morphology of the international blood cell review rules by Beckman Coulter LH755 hematology analyzer, and to provide the reference for re-examination of blood cells. Methods 1 984 cases were classified into non-leukemia group （1 832 cases）, preliminarily diagnosed leukemia group （73 cases） and further consultation group （79 cases）. All specimens were detected for leucocyte differentiation and cell morphology observation by instrumental analysis and manual examination. According to the international blood cell review rules, the relative international blood cell review rules were evaluated in non-leukemia and preliminarily diagnosed leukemia groups. The changes of suspicious warning between leukemia and non-leukemia groups before and after treatment were observed. Using receiver operating characteristic （ROC） curve, the value of diagnosing leukemia prefiminarily was observed. Results A total of 724 preliminarily diagnosed leukemia cases involved suspicious warning rules, including 655 cases of non-leukemia group, 69 cases of leukemia group and 46 myeloid leukemia patients. The microscopy results were positive. The terms of 34 （ left alarm） and 37 （blast cell alarm）were more sensitive and positively predictive. The terms of 30 [ platelet （PLT） aggregation alarm], 31 （PLT alarm） and 35 [ non-typical and （or） variation Lym] had lower positive predictive value. The false negative cases were detected in 28 cases. There were 8 cases detected blast ceils, and these samples were all from the leukemia patients after treatment （28.57%, 8/28 ）. According to the ROC curve, the term of 37 （blast cell alarm） had the highest value （AUC = 0.82l ）to the preliminarily diagnosed leukemia patients, followed by the 34 （left alarm）, 32 （IG alarm）, 26 （white blood result unreliability） and 35 [non-typical and （or） variation Lym~. The terms of 30 （PLT aggregation alarm） and 31 （PLT alarm） had low value. Conclusions The international blood cell suspicious warning review rules has an important value in diagnosing leukemia preliminarily. The cell morphology after treatment is more varied and complex in leukemia patients, and the instruments can not directly provide the accurate information of blood cell change, which is needed to be further verified by manual microscopy. The case involving term 37 （blast cell alarm） needs microscopy, and the case involving term 26 （white blood result unreliability）, 32 （IG alarm）, 34 （left alarm） and 35 [ non-typical and （or） variation Lym] , also should be paid attention to avoid misdiagnosis.