Abstract: Objective　To analyze the platelet count of newly diagnosed patients with acute myeloid leukemia（AML） after initial induction chemotherapy for the first time，and to access its relationship with prognosis and importance，and to explore simple and easy indices of predicating AML prognosis.Methods　110 AML patients who received explicit diagnosis and treatment initially in Tianjin First Central Hospital from May 2011 to December 2014 were selected，all of them accepted classic DA or similar chemotherapy regimens to perform remission induction chemotherapy.Three weeks after the first time of induction chemotherapy，platelet count was detected.The patients were divided into four groups according to platelet count:45 patients with platelet count <100×109/L group，19 patients with platelet count of（100-150）×109/L group，25 patients with platelet count of（151-300）×109/L group，and 21 patients with platelet count >300×109/L group.Gender，age，white blood cell count during initial treatment，poor prognostic chromosomes，whether obtained complete remission of bone marrow after initial induction chemotherapy or not，and recurrence rate of patients in 4 groups were compared.Taking patients’ age of 60 as a dividing line，there were 41 patients aged ≥60 and 69 patients aged<60;taking patients’ white blood cell count of 30×109/L during initial treatment as a dividing line，there were 28 patients with white blood cell count ≥30×109/L，and 82 patients with white blood cell count<30×109/L；taking patients had poor prognosis chromosomes，16 patients had poor prognosis chromosomes and 94 patients no poor prognosis chromosomes；taking patients had obtained complete remission of bone marrow after initial induction chemotherapy or not，70 patients had obtained complete remission and 40 patients had obtained non-complete remission；taking patients’ platelet count of 300×109/L after initial induction chemotherapy as a dividing line，21 patients with platelet count >300×109/L，and 89 patients with platelet count ≤300×109/L.Kaplan-Meier survival curve was applied，and overall survival and disease-free survival were taken as the end point of observation respectively.Log-rank test and multivariate Cox proportional hazards regression model was used to analyze the influencing factors of prognosis.Results　There was no significant difference in gender，the proportion of poor prognosis chromosomes and obtaining complete remission of bone marrow after initial induction chemotherapy among groups of different platelet count after initial induction chemotherapy（P>0.05）;the age of patients in platelet count <100×109/L group and platelet count of（100-150）×109/L group was greater than that in platelet count >300×109/L group（P<0.05）.The proportion of white blood count ≥30×109/L during initial treatment and recurrence rate of patients in platelet count <100×109/L group during initial treatment were higher than those in platelet count >300×109/L group（P<0.007）.Log-rank test showed that there were significant difference in overall survival and disease-free survival among patients of different ages，white blood count during initial treatment，with or without obtaining of complete remission and different platelet count after initial induction chemotherapy（P<0.05）;there were no significant difference in overall survival and disease-free survival between patients with or without poor prognosis chromosomes（P>0.05）.Multivariate Cox proportional hazards regression analysis revealed that there were regression correlation between age，the obtaining of complete remission and platelet count after initial induction chemotherapy and overall survival，disease-free survival of patients（P<0.05）.Conclusion　The increasing of platelet count of AML patients after initial induction chemotherapy leads to relatively good prognosis of patients and can independently affect prognosis of AML patients.

Key words: Leukemia, myeloid, acute, Antineoplastic combined chemotherapy protocols, Platelet count, Prognosis

摘要： 目的　分析初治急性髓系白血病（AML）患者首次诱导化疗后血小板计数情况，评估其与预后的关系及重要性，探索预测AML预后的简单易行的指标。方法　选取2011年5月—2014年12月天津市第一中心医院明确诊断为初治AML的患者110例，均采用经典DA化疗方案或类似化疗方案进行诱导缓解化疗。首次诱导化疗结束后3周时检测血小板计数，根据血小板计数分为4组：<100×109/L组45例、（100～150）×109/L组19例、（151～300）×109/L组25例、>300×109/L组21例。比较4组患者的性别、年龄、初治时白细胞计数、预后不良染色体、首次诱导化疗是否获得骨髓完全缓解、复发率。以患者年龄60岁为界，分为≥60岁41例，<60岁69例；以患者初治时白细胞计数30×109/L为界，分为≥30×109/L 28例，<30×109/L 82例；以患者是否预后不良染色体为界，分为预后不良染色体16例，非预后不良染色体94例；以患者首次诱导化疗是否获得骨髓完全缓解为界，分为完全缓解70例，未完全缓解40例；以患者首次诱导化疗后血小板计数300×109/L为界，分为>300×109/L 21例，≤300×109/L 89例。采用Kaplan-Meier法绘制生存曲线，分别以总生存率和无病生存率作为观察终点，采用Log-rank检验和多因素Cox比例风险回归模型分析预后的影响因素。结果　首次诱导化疗后不同血小板计数组患者性别、预后不良染色体、首次诱导化疗获得骨髓完全缓解所占比例比较，差异均无统计学意义（P>0.05）；血小板计数<100×109/L组、（100～150）×109/L组患者年龄高于>300×109/L组（P<0.05）。血小板计数<100×109/L组患者初治时白细胞计数≥30×109/L所占比例、复发率高于>300×109/L组（P<0.007）。Log-rank检验示，不同年龄、初治时白细胞计数、首次诱导化疗是否获得骨髓完全缓解、首次诱导化疗后血小板计数患者总生存率、无病生存率比较，差异均有统计学意义（P<0.05）；是否预后不良染色体患者总生存率、无病生存率比较，差异均无统计学意义（P>0.05）。多因素Cox比例风险回归分析结果显示，年龄、首次诱导化疗获得骨髓完全缓解、首次诱导化疗后血小板计数与患者总生存率、无病生存率有回归关系（P<0.05）。结论　初治AML患者首次诱导化疗后血小板计数增多者预后相对较好，且可独立影响AML患者的预后。

关键词: 白血病, 髓样, 急性, 抗肿瘤联合化疗方案, 血小板计数, 预后