目的 分析急性淋巴细胞白血病（acute lymphoblastic leukemia, ALL）患儿入住儿童重症监护室（pediatric intensive care unit, PICU）的危险因素及其在PICU接受生命支持治疗（life-sustaining therapy, LST）的危险因素。 方法 回顾性分析2016年6月—2021年6月中南大学湘雅二医院儿童医学中心收治的ALL患儿的临床资料，根据是否入住PICU分为PICU组和非PICU组，采用多因素logistic回归分析入住PICU的危险因素。采用Kaplan-Meier曲线评估PICU累积入住概率。根据是否接受LST将PICU患儿分为LST组和非LST组，采用多因素logistic回归分析接受LST的危险因素。 结果 共纳入200例ALL患儿，其中42例（21.0%）患儿至少入住过一次PICU，总入住次数为48次。多因素分析显示，诊断时伴高白细胞血症和乳酸脱氢酶>500 U/L是ALL患儿入住PICU的独立危险因素（均P<0.05）。Kaplan-Meier曲线显示T淋巴细胞白血病和高白细胞血症均具有相对较高的累积PICU入院率。C反应蛋白、白蛋白水平及呼吸衰竭与患儿接受LST显著相关（均P<0.05）；进一步行多因素logistic回归分析显示呼吸衰竭的PICU患儿接受LST的风险显著升高（OR=13.254，P=0.027）。 结论 诊断时伴高白细胞血症和乳酸脱氢酶>500 U/L的ALL患儿入住PICU的风险较高；呼吸衰竭是PICU内ALL患儿接受LST的独立危险因素。

Objective To investigate risk factors for pediatric intensive care unit (PICU) admission among children with acute lymphoblastic leukemia (ALL) and risk factors for receipt of life-sustaining therapy (LST) in the PICU. Methods Clinical data of ALL patients treated at the Children's Medical Center of the Second Xiangya Hospital from June 2016 to June 2021 were retrospectively reviewed. Patients were categorized into PICU and non-PICU groups according to PICU admission. Multivariable logistic regression was applied to identify risk factors for PICU admission. The cumulative probability of PICU admission was estimated using Kaplan-Meier curves. PICU patients were further stratified into LST and non-LST groups according to whether LST was received, and multivariable logistic regression was used to identify risk factors for receiving LST. Results A total of 200 children with ALL were included; 42 (21.0%) were admitted to the PICU at least once, with 48 total admissions. Multivariable logistic regression analysis showed that hyperleukocytosis at diagnosis and lactate dehydrogenase (LDH) >500 U/L were independent risk factors for PICU admission (both P<0.05). Kaplan-Meier curves demonstrated that T-cell ALL and hyperleukocytosis were associated with higher cumulative PICU admission rates. Univariate analysis showed that C-reactive protein, albumin, and respiratory failure were significantly associated with the receipt of LST (all P<0.05). Further multivariable logistic regression analysis revealed that respiratory failure was significantly associated with an increased risk of receiving LST (OR=13.254, P=0.027). Conclusions Children with ALL who have hyperleukocytosis at diagnosis and LDH >500 U/L have a higher risk of PICU admission; respiratory failure is an independent risk factor for receipt of LST among PICU-admitted ALL patients.