论文部分内容阅读
目的:探讨利奈唑胺诱导血小板减少（LIT）的危险因素及其预警价值。方法:收集苏州市立医院2019年7月至2021年10月收治的医院获得性肺炎（HAP）患者的病历资料进行回顾性分析。提取患者一般情况、合并疾病、利奈唑胺应用情况、实验室检查结果和利奈唑胺稳态血药谷浓度（利奈唑胺谷浓度）等数据，根据LIT发生与否将患者分为LIT组和无LIT组，比较2组患者的临床特征，采用二元logistic回归模型分析HAP患者发生LIT的危险因素，计算比值比（n OR）及其95%置信区间（n CI）。采用受试者工作特征（ROC）曲线分析危险因素对LIT发生的预测价值。n 结果:纳入分析的患者共74例，男性55例，女性19例；年龄82（73，88）岁；发生LIT者25例（33.8%）。LIT组患者年龄、利奈唑胺谷浓度高于无LIT组[88（81，92）岁比79（70，86）岁，n P=0.001；（19.6±10.3）mg/L比（9.8±6.4）mg/L，n P<0.001]，基础血小板计数（PLT）、基础肌酐清除率（Ccr）低于无LIT组[181（162，212）×10n 9/L比229（169，289）×10n 9/L，n P=0.025；31（19，44）ml/（min·1.73 mn 2）比46（27，65）ml/（min·1.73 mn 2），n P=0.018]。二元logistic回归分析结果显示，基础Ccr低（n OR=0.974，95n %CI：0.951~0.998，n P=0.035）、利奈唑胺谷浓度高（n OR=1.156，95n %CI：1.059~1.261，n P=0.001）是HAP患者发生LIT的独立危险因素。ROC曲线分析结果显示，HAP患者发生LIT的年龄、利奈唑胺谷浓度、基础PLT及基础Ccr折点分别为87岁（敏感度56.0%，特异度83.7%）、15.4 mg/L（敏感度64.0%，特异度87.8%）、189×10n 9/L（敏感度67.3%，特异度68.0%）和45 ml/（min·1.73 mn 2）（敏感度53.1%，特异度80.0%）。以上述折点为警戒值将患者各分为2组比较其LIT发生率，结果显示年龄、利奈唑胺谷浓度超过警戒值组和基础PLT、基础Ccr、低于和等于警戒值组LIT发生率均明显高于相比较的另一组（均n P<0.01）。n 结论:基础Ccr、利奈唑胺谷浓度、年龄和基础PLT是HAP患者发生LIT的危险因素，其警戒值分别为45 ml/（min·1.73 mn 2）、15.4 mg/L、87岁和189×10n 9/L，对LIT的发生有较好预测价值。n “,”Objective:To explore the risk factors of linezolid-induced thrombocytopenia (LIT) and evaluate their predictive value.Methods:Medical records of hospital acquired pneumonia (HAP) patients who admitted in Suzhou Municipal Hospital from July 2019 to October 2021 and received linezolid were collected and retrospectively analyzed. Clinical data including general information, comorbidities, linezolid application, laboratory test results, and trough concentration of linezolid was extracted. Patients were divided into LIT group and non-LIT group according to the occurrence of LIT. Clinical characteristics were compared between the 2 groups; risk factors of LIT in patients with HAP were analyzed using a binary logistic regression model, and the odds ratio (n OR) and its 95% confidence interval (n CI) were calculated; the predictive value of the risk factors for LIT were evaluated using receiver operating characteristic (ROC) curve.n Results:A total of 74 patients were included in the study, including 55 males and 19 females, aged 82 (73, 88) years. LIT occurred in 25 patients (33.8%). Compared with the non-LIT group, the age and trough concentration of linezolid in patients in the LIT group were higher [88 (81, 92) years n vs. 79(70, 86) years, n P=0.001; (19.6±10.3) mg/L n vs. (9.8±6.4) mg/L, n P<0.001], and the baseline platelet count and baseline creatinine clearance rate were lower [181(162, 212) ×10n 9/L n vs. 229 (169, 289) ×10n 9/L, n P=0.025; 31(19, 44) ml/(min·1.73 mn 2) n vs. 46 (27, 65) ml/(min·1.73 mn 2), n P=0.018]. Binary logistic regression analysis showed that the lower baseline creatinine clearance rate (n OR=0.974, 95n %CI: 0.951-0.998, n P=0.035) and higher trough concentration of linezolid (n OR=1.156, 95n %CI: 1.059-1.261, n P=0.001) were independent risk factors for LIT in HAP patients. ROC curve analysis showed that the threshold of the age, trough concentration of linezolid, baseline platelet count, and baseline creatinine clearance rate were 87 years (sensitivity 56.0%, specificity 83.7%), 15.4 mg/L (sensitivity 64.0%, specificity 87.8%), 189×10n 9/L (sensitivity 67.3%, specificity 68.0%), and 45 ml/(min·1.73 mn 2) (sensitivity 53.1%, specificity 80.0%), respectively. Patients were respectively divided into 2 groups according to the thresholds and the incidences of LIT were compared. The results showed that the incidences of LIT in patients with age and trough concentration of linezolid exceeding the thresholds and in patients with baseline plate count and baseline creatinine clearance rate lower than or equal to the thresholds were significantly higher than those in the other patients (all n P<0.01).n Conclusions:Baseline creatinine clearance rate, trough concentration of linezolid, age, and plate count are risk factors for LIT in HAP patients and their thresholds are 45 ml/(min·1.73 mn 2), 15.4 mg/L, 87 years, and 189×10n 9/L, respectively. These risk factors have good predictive value for the occurrence of LIT.n