摘要

第一部分 N末端B型利钠肽原和高敏肌钙蛋白T对小儿先天性

心脏病的预后价值

 

背景

目前，关于利钠肽和肌钙蛋白在小儿先天性心脏病（先心病）领域的临床研究仍较为有限，且研究结论存在一定争议。本研究旨在基于一个包含超过千名患者的前瞻性队列，分析术后测定的N末端B型利钠肽原（N-terminal pro-B-type natriuretic peptide，NT-proBNP）与高敏肌钙蛋白T（High-sensitivity troponin T，hs-TnT）对于不同年龄段、接受体外循环（Cardiopulmonary bypass，CPB）下心脏手术的小儿先心病患者的预后价值。

 

方法

本研究收集了2022年1月至2023年5月期间，接受手术治疗的小儿先心病患者的临床数据。在术后转入重症监护病房后的6小时内采集患儿的血液样本，并同时检测NT-proBNP和hs-TnT。复合不良事件包括院内死亡、非计划再干预（包括心包开窗术和开胸探查术）、机械通气时间超过10天、重症监护病房住院时间超过30天，以及术后需要体外膜肺氧合或腹膜透析治疗。采用多因素Logistic回归模型结合最小p值法以确定心脏生物标志物的最优风险分层阈值。

 

结果

本研究最终纳入了1015名15岁以下的先心病患儿（女性占比46.6%），其中中位年龄为2.33岁【四分位距（Interquartile range，IQR）：0.60-5.12岁】。共74位患儿发生了不良事件。1岁以下患者（NT-proBNP：952.0 ng/L，IQR：355.0-3023.0 ng/L；hs-TnT：5730.0 ng/L，IQR：3265.0-10000 ng/L）的NT-proBNP和hs-TnT水平分别是1岁以上患者（NT-proBNP：94.5 ng/L，IQR：47.3-178.5 ng/L；hs-TnT：1132.5 ng/L，IQR：693.3-2077.5 ng/L）的大约10倍和5倍。在各个年龄亚组中，两种心脏生物标志物的水平均随着胸外科医师协会-欧洲心胸外科协会死亡风险分级的提高而上升。hs-TnT与CPB持续时间的关联性（r=0.61，p<0.001）强于NT-proBNP与CPB持续时间的关联性（r=0.37，p<0.001）。NT-proBNP和hs-TnT均与机械通气治疗时间紧密相关（r=0.63，p<0.001；r=0.62，p<0.001），并与重症监护病房住院时间呈中度相关（r=0.54，p<0.001；r=0.52，p<0.001）。NT-proBNP对于1岁以下患儿【曲线下面积（Area under the curve，AUC）：0.772，95%置信区间（Confidence interval，CI）：0.689-0.844】和1岁以上患儿（AUC：0.840，95% CI：0.754-0.914）均表现出良好的预后预测能力。然而，hs-TnT仅对1岁以上患儿具有良好的预后价值（AUC：0.784，95% CI：0.716-0.844），其对1岁以下患儿的预后价值有限（AUC：0.611，95% CI：0.522-0.695）。对于1岁以下的患儿，hs-TnT无适合的风险分层阈值。与NT-proBNP低于2000 ng/L的1岁以下患儿相比，NT-proBNP水平处于2000-10000 ng/L【比值比（Odds ratio，OR）：3.79，95% CI：1.47-9.76，p=0.006】和高于10000 ng/L（OR：12.21，95% CI：3.66-40.80，p<0.001）的患儿发生不良事件的风险显著增加。NT-proBNP超过500 ng/L（OR：15.09，95% CI：6.05-37.66，p<0.001）或hs-TnT高于1200 ng/L（OR：5.50，95% CI：1.47-20.59，p=0.011）的1岁以上患儿发生不良事件的风险显著增加。

 

结论

术后6小时内检测的NT-proBNP和hs-TnT对于1岁以上的先心病患儿具有显著的不良事件预测价值。然而，在1岁以下的先心病患儿中，只有NT-proBNP对术后不良事件展现出较好的预测性能。建议根据先心病患儿的年龄，在术后6小时内进行心脏生物标志物的检测，以便及时评估早期不良事件风险，并尽早实施预防性治疗措施。

 

 

第二部分 年龄校正的N末端B型利钠肽原对小儿复杂先天性心脏病的预后价值

 

背景

复杂先天性心脏病（先心病）约占所有先心病病例的30%，其预后较简单先心病显著更差。第一部分的研究结果显示N末端B型利钠肽原（N-terminal pro-B-type natriuretic peptide，NT-proBNP）对于各年龄段的小儿先心病患者均具有良好的预后预测价值。然而，由于NT-proBNP浓度随年龄增长呈现显著的生理性变化，仅以1岁为界将患者划分为两个年龄段来评估NT-proBNP的预后价值存在一定局限性。为此，本研究计划首次探索年龄校正的NT-proBNP（zlog-proBNP）对中国复杂先心病患儿的预后意义。

 

方法

本研究对2018年1月至2022年12月期间接受心脏手术治疗的复杂先心病患儿的临床数据进行了回顾性分析。患者在术后24小时内接受的NT-proBNP检测数据被详细记录。体重、NT-proBNP检测值及其他实验室项目检测值均根据不同年龄段和性别的正常参考区间进行了相应校正。主要终点事件为全因死亡，次要终点则是由全因死亡和非计划心脏再干预组成的复合不良事件。使用限制性立方样条拟合zlog-proBNP水平与不良事件风险之间的关系。此外，采用预先设定的多因素Cox回归模型，对zlog-proBNP作为连续变量或分类变量与不良事件的关联进行进一步分析。

 

结果

本研究共纳入了2681例复杂先心病患者，中位年龄为10.6月【四分位距（Interquartile range，IQR）：4.9-39.2月】，男性患者占比为59%。术后NT-proBNP的中位值为3970 ng/L（IQR：1352–9095 ng/L），zlog-proBNP的中位值为3.83（IQR：2.93–4.44）。患者的年龄和采血时间均对zlog-proBNP水平产生了显著影响。中位随访期为3.0年（IQR：1.5-4.6年；范围：1天-6.2年），随访共记录到121例死亡和260例复合不良事件。限制性立方样条分析显示，zlog-proBNP水平与死亡或复合不良事件风险之间存在非线性正相关性。当zlog-proBNP值约等于3.8时，死亡和复合不良事件的风险比（Hazard ratio，HR）均接近于1。多因素Cox回归分析显示，与zlog-proBNP水平较低（≤1.96）的患者相比，zlog-proBNP极高组【zlog-proBNP>5；HR：9.78，95%置信区间（Confidence interval，CI）：3.17-30.22，p<0.001】患者的死亡风险最高。其次是zlog-proBNP高水平组（3.8

 

结论

zlog-proBNP可作为独立于年龄的评估小儿先心病中长期预后的有效工具。死亡或非计划心脏再干预的风险均随着zlog-proBNP水平的升高而显著增加。临床应当密切关注zlog-proBNP值超过5的患者。

 

第三部分 小儿先天性心脏病患者的铁代谢及其预后关联

 

背景

缺铁是全球儿童中最常见的微量营养素缺乏症，且铁代谢失衡与心血管疾病发病率和死亡率的增加密切相关。然而，小儿先天性心脏病（先心病）患者的铁代谢状况及其临床意义尚未被充分阐明。本研究旨在系统地探讨先心病患儿的术前铁状态以及其与术后不良结局之间的潜在关系。

 

方法

本研究对2017年1月至2022年12月期间接受心脏手术治疗的1个月至5岁先心病患者的临床资料进行了回顾性分析。实验室检测数据（包括生化及血常规检查）均来源于患者入院时的首次检查结果。缺铁的诊断标准为满足以下任一条件：1）血清铁蛋白水平低于15 ng/mL；或2）转铁蛋白饱和度低于15%，且同时伴有红细胞分布宽度超过16%。院内死亡被定义为主要不良事件，而复合不良事件则包括院内死亡、因病危放弃治疗、非计划再干预（包括心包开窗术和开胸探查术）、机械通气时间超过2周、重症监护时间超过30天，以及需要体外膜肺氧合或腹膜透析治疗。通过限制性立方样条描绘铁蛋白水平与不良事件风险之间的关系。采用多因素Logistic回归分析来确定术前不同的铁状态与术后不良结局之间的相关性。

 

结果

本研究最终纳入8065名患者，中位年龄为17.3月【四分位距（Interquartile range，IQR）：7.7-34.7月】，其中男性患者占49.6%。共有1305位患者患有紫绀型先心病。分别有23%和12%的患者在术前患有缺铁和贫血，其中超过半数的贫血病例由缺铁引起的。紫绀型先心病患者的缺铁发生率（37.7%）显著高于非紫绀型患者（20.2%），但紫绀型患者的贫血发生率（9.6%）却低于非紫绀型患者（12.5%）。铁蛋白水平与不良事件风险之间呈右偏的“U形”关系，铁蛋白浓度处于15-33 ng/mL范围的患者发生不良事件的风险最低。高铁蛋白组患者【铁蛋白>100 ng/mL；比值比（Odds ratio，OR）：8.20，95%置信区间（Confidence interval，CI）：1.61-41.86，p=0.011】的院内死亡风险显著高于铁饱和组患者（15 ng/mL≤铁蛋白<33 ng/mL），而中等铁蛋白组（33 ng/mL<铁蛋白<100 ng/mL；OR：3.09，95% CI：0.65-14.69，p=0.156）和缺铁组（OR：3.68，95% CI：0.77-17.59，p=0.102）患者的死亡风险虽有所增加，但无统计学意义。相较于铁饱和组，高铁蛋白组（OR：4.21，95% CI：2.65-6.68，p<0.001）、中等铁蛋白组（OR：1.64，95% CI：1.11-2.44，p=0.014）和缺铁组（OR：1.80，95% CI：1.18-2.73，p=0.006）的复合不良事件风险均显著升高。此外，高铁蛋白组患者的机械通气时间和重症监护病房住院时间均长于其他三组，其后依次为缺铁组、中等铁蛋白组和铁饱和组。缺铁（OR：1.28，95% CI：1.08-1.52，p=0.005）和高铁蛋白（OR：1.71，95% CI：1.23-2.37，p=0.002）均是围手术期需要接受红细胞输注的独立风险因素。

 

结论

在计划接受心脏手术治疗的小儿先心病患者群体中，近四分之一存在缺铁问题，且超过一成的患者患有贫血。紫绀型先心病患者的缺铁状况更为普遍，但贫血的发生率相对较低。与铁饱和的患者相比，缺铁和高铁蛋白负荷均是术后不良结局的独立危险因素。

 

第四部分 小儿先天性心脏病患者于重症监护期间发生的血小板减少与预后的关联

 

背景

血小板减少与重症监护病房（Intensive care unit，ICU）患者的出血风险、输血需求及死亡风险密切相关。然而，对于接受体外循环（Cardiopulmonary Bypass，CPB）支持的心脏手术的先天性心脏病（先心病）患儿，其在ICU中出现的血小板减少情况及其对预后的潜在影响尚未得到充分研究。

 

方法

本研究回顾性分析了2017年1月至2022年12月期间，接受CPB支持的心脏手术治疗的18岁以下先心病患者的临床数据。研究收集了患者在ICU住院前7天的血小板计数检测结果。血小板计数低于150×10^⁹/L被定义为血小板减少，并根据严重程度进一步分为轻度（100-150×10^⁹/L）、中度（50-100×10^⁹/L）和重度（<50×10^⁹/L）三个等级。本研究以术后30天死亡作为主要终点事件。通过Logistic回归模型识别血小板减少的风险因素，并采用Cox回归模型探讨不同严重程度血小板减少与术后30天死亡之间的相关性。

 

结果

本研究共纳入11761名小儿先心病患者，男性占51.5%，患者的中位年龄为1.7岁【四分位距（Interquartile range，IQR）：0.7-3.7岁】。约20%的患者被诊断为紫绀型先心病。共有4007例（34.1%）患者在ICU中出现血小板减少，其中轻度、中度和重度血小板减少的患者分别有2773例（23.6%）、987例（8.4%）和247例（2.1%）。术前基线血小板水平约为300×10⁹/L，但在术后第一天血小板水平急剧下降，并在接下来的两天内保持较低水平，随后逐渐回升。总体而言，非紫绀型先心病患者在术前及术后各检测时间点的血小板水平均高于紫绀型先心病患者。年龄较小、紫绀型先心病、CPB持续时间较长、术前红细胞计数较低、术前血小板减少、术前凝血功能障碍、术前红细胞分布宽度较高以及术前红细胞比容较高是各种严重程度血小板减少的危险因素。多因素Cox回归分析显示，中度ICU血小板减少【风险比（Hazard ratio，HR）：11.38，95%置信区间（Confidence interval，CI）：3.02-42.87，p<0.001】、重度ICU血小板减少（HR：49.54，95% CI：13.11-187.14，p<0.001）、CPB持续时间（HR：1.01，95% CI：1.01-1.01，p<0.001）、紫绀型先心病（HR：2.59，95% CI：1.13-5.93，p=0.024）、术前红细胞比容（HR：0.95，95% CI：0.91-1.00，p=0.030）、术前血清钠水平（HR：0.89，95% CI：0.80-0.99，p=0.025）和术前超敏C反应蛋白大于3 mg/L（HR：2.46，95% CI：1.17-5.17，p=0.017）均与术后30天死亡风险独立相关。Kaplan-Meier生存曲线显示，中度和重度ICU血小板减少患者的30天生存率显著低于无血小板减少患者。然而，轻度ICU血小板减少患者的30天生存率与无血小板减少患者相近。此外，随着ICU血小板减少严重程度的增加，成分血的输血总量和比例、术后严重出血和血栓事件的发生率、ICU住院时间和机械通气时间均呈“剂量依赖性”升高。

 

结论

超过三分之一的小儿先心病患者在术后重症监护期间会发生血小板减少。中度或重度血小板减少是术后30天死亡的独立危险因素。因此，建议临床对先心病患者进行术后血小板动态监测，并及时关注血小板检测结果对不良预后的警示作用。

 

Abstract

Part 1. Prognostic Value of N-Terminal Pro-B-Type Natriuretic Peptide and High-Sensitivity Troponin T in Pediatric Congenital Heart Disease

 

Background

Currently, there is a lack of research on the clinical application of natriuretic peptides and troponins in pediatric congenital heart disease (CHD). This study aims to analyze the prognostic value of postoperative N-terminal pro-B-type natriuretic peptide (NT-proBNP) and High-sensitivity troponin T (hs-TnT) in children with CHD undergoing cardiac surgery with cardiopulmonary bypass (CPB), across different age groups, through a prospective cohort study involving more than a thousand patients.

 

Methods

This study enrolled pediatric patients with CHD who underwent surgical treatment from January 2022 to May 2023. Blood samples were drawn within 6 hours of the patients being transferred to the intensive care unit (ICU) after surgery, and NT-proBNP and hs-TnT were measured simultaneously. The composite adverse events include in-hospital death, unplanned re-intervention (including fenestration pericardium and thoracotomy), mechanical ventilation for more than 10 days, ICU stay for more than 30 days, and the need for postoperative extracorporeal membrane oxygenation and peritoneal dialysis treatment. A multivariate Logistic regression model combined with the minimum p-value method was used to determine the optimal risk threshold.

 

Results

This study ultimately included 1015 children under the age of 15 (46.6% female), with a median age of 2.33 years [IQR (Interquartile range): 0.60-5.12 years], and a total of 74 children experienced adverse events. Compared to children over 1 year old (NT-proBNP: 94.5 ng/L, IQR: 47.3-178.5 ng/L; hs-TnT: 1132.5 ng/L, IQR: 693.3-2077.5 ng/L), children under 1 year old had approximately 10 times higher postoperative NT-proBNP levels (952.0 ng/L, IQR: 355.0-3023.0 ng/L) and approximately 5 times higher hs-TnT levels (5730.0 ng/L, IQR: 3265.0-10000 ng/L). In each age subgroup, the levels of both cardiac biomarkers significantly increased with the elevation of the Society of Thoracic Surgeons-European Association for Cardio-Thoracic Surgery Mortality Categories. The correlation between hs-TnT and CPB duration (r=0.61, p<0.001) was stronger than that between NT-proBNP and CPB duration (r=0.37, p<0.001). Both NT-proBNP and hs-TnT were closely related to the duration of mechanical ventilation (r=0.63, p<0.001; r=0.62, p<0.001) and moderately correlated with ICU stay (r=0.54, p<0.001; r=0.52, p<0.001). NT-proBNP showed good prognostic performance for children under 1 year old [AUC (Area under the curve): 0.772, 95% CI (Confidence interval): 0.689-0.844] and for those over 1 year old (AUC: 0.840, 95% CI: 0.754-0.914). However, hs-TnT only had good prognostic value for children over 1 year old (AUC: 0.784, 95% CI: 0.716-0.844) and limited prognostic value for children under 1 year old (AUC: 0.611, 95% CI: 0.522-0.695). For children under 1 year old, there was no appropriate risk stratification threshold for hs-TnT. Compared to children with NT-proBNP levels below 2000 ng/L, the risk of adverse events significantly increased in children with NT-proBNP levels between 2000-10000 ng/L [OR (Odds ratio): 3.79, 95% CI: 1.47-9.76, p=0.006] and above 10000 ng/L (OR: 12.21, 95% CI: 3.66-40.80, p<0.001). For children over 1 year old, the risk of adverse events significantly increased in those with NT-proBNP levels exceeding 500 ng/L (OR: 15.09, 95% CI: 6.05-37.66, p<0.001) or hs-TnT levels above 1200 ng/L (OR: 5.50, 95% CI: 1.47-20.59, p=0.011).

 

Conclusion

NT-proBNP and hs-TnT tested within 6 hours postoperatively has significant prognostic value for postoperative adverse events in patients over 1 year old with CHD. However, in patients under 1 year old, only NT-proBNP demonstrated commendable predictive performance for postoperative adverse events. Cardiac biomarker testing within 6 hours postoperatively should be conducted based on the patient’s age to timely assess potential early adverse events and preventive treatment measures should be implemented as soon as possible.

 

Part 2. Prognostic Value of Age-Adjusted N-Terminal Pro-B-Type Natriuretic Peptide in Pediatric Complex Congenital Heart Disease

 

Background

Complex congenital heart disease (CHD) accounts for approximately 30% of all CHD cases, with a much poorer prognosis compared to simple subtypes. The results of Part 1 indicated that NT-proBNP has good prognostic value for CHD patients of all ages. However, given that the concentration of NT-proBNP undergoes significant physiological changes with age, dividing patients into only two age groups based on a one-year cutoff to assess the prognostic value of NT-proBNP in pediatric CHD is clearly limited. Therefore, this study aims to explore for the first time the prognostic significance of age-adjusted NT-proBNP (zlog-proBNP) for Chinese children with complex CHD.

 

Methods

This study conducted a retrospective analysis of the clinical data of children with complex CHD who underwent cardiac surgery from January 2018 to December 2022. NT-proBNP test performed within 24 hours after surgery for all patients were recorded in detail. Body weight, NT-proBNP values, and other laboratory tests were adjusted according to the standard reference ranges for different age groups and genders. The primary endpoint of the study was set as all-cause death, while the secondary endpoint was a composite adverse event consisting of all-cause death and unplanned cardiac reintervention. The relationship between zlog-proBNP levels and the risk of adverse outcomes was fitted using restricted cubic splines. In addition, a pre-set multivariate Cox regression model was used to further analyze the association between zlog-proBNP as a continuous or categorical variable and adverse events.

 

Results

This study included 2681 patients with complex CHD, with a median age of 10.6 months [IQR (Interquartile range): 4.9-39.2 months], and 59% were male. The median follow-up time was 3.0 years (IQR: 1.5-4.6 years; Range: 1 day-6.2 years). The median postoperative NT-proBNP value was as high as 3970 ng/L (IQR: 1352–9095 ng/L), and the median zlog-proBNP level was 3.83 (IQR: 2.93–4.44). Both the age and the blood collection time-point have a significant impact on the zlog-proBNP levels. A total of 121 deaths and 260 composite events were recorded. Restricted cubic splines showed a nonlinear positive correlation between zlog-proBNP levels and the risk of death or composite adverse events, with both HR (Hazard ratio) approaching 1 when the zlog-proBNP level was approximately 3.8. Multivariate Cox regression analysis revealed that compared to patients with low zlog-proBNP (≤1.96), patients with very high zlog-proBNP [zlog-proBNP>5; HR: 9.78, 95% CI (Confidence interval): 3.17-30.22, p<0.001] had the highest risk of death, followed by patients with high zlog-proBNP levels (3.8

 

Conclusion

zlog-proBNP could serve as an excellent tool for assessing postoperative risk in children with complex CHD, independent of age. Patients with higher zlog-proBNP have a significantly increased risk of death or unplanned cardiac reintervention. In particular, clinicians should pay special attention to the patients whose zlog-proBNP levels exceed 5.

 

Part 3. Iron Metabolism in Pediatric Congenital Heart Disease and Its Association with Prognosis

 

Background

Iron deficiency is the most common micronutrient deficiency in children worldwide, and iron imbalance is associated with the increased incidence and mortality of cardiovascular disease. However, the iron status in children with CHD and its clinical significance have not been fully elucidated. This study aims to systematically investigate the preoperative iron status of CHD patients and its potential relationship with postoperative adverse outcomes.

 

Methods

We conducted a retrospective analysis of clinical data from patients aged 1 month to 5 years who were diagnosed with CHD and underwent surgery from January 2017 to December 2022. The laboratory data used in this study, including biochemical and hematological tests, were derived from the patients’ first test results upon admission. The diagnostic criteria for iron deficiency were defined as meeting either of the following conditions: 1) serum ferritin level less than 15 ng/mL; or 2) transferrin saturation less than 15%, accompanied by a red blood cell distribution width exceeding 16%. In-hospital death was defined as the primary adverse event, while the composite adverse events included in-hospital death, abandoned treatment due to critical illness, unplanned reintervention (including pericardial fenestration and thoracotomy), mechanical ventilation for more than 2 weeks, intensive care duration exceeding 30 days, and the need for extracorporeal membrane oxygenation or peritoneal dialysis treatment. The relationship between ferritin levels and prognosis was fitted using restricted cubic splines, and multivariate Logistic regression was employed to determine the correlation between different iron statuses and adverse outcomes.

 

Results

This study ultimately included 8065 patients with a median age of 17.3 months [IQR (Interquartile range): 7.7-34.7 months], and male patients accounted for 49.6% of the total. Among them, 1305 patients were diagnosed with cyanotic CHD. Preoperatively, 23% of patients had iron deficiency and 12% had anemia, with more than half of the anemia cases being attributed to iron deficiency. The incidence of iron deficiency in patients with cyanotic CHD (37.7%) was nearly twice that of acyanotic patients (20.2%), but the incidence of anemia in cyanotic patients (9.6%) was lower than that in acyanotic patients (12.5%). The relationship between ferritin levels and the risk of adverse outcomes showed a right-skewed U-shaped curve, with the lowest risk of adverse events in patients with ferritin levels in the range of 15-33 ng/mL. The risk of in-hospital death in the high ferritin group [ferritin > 100 ng/mL; OR (Odds ratio): 8.20, 95% CI (Confidence interval): 1.61-41.86, p=0.011] was significantly higher than that in the iron repletion group (15 ng/mL ≤ ferritin < 33 ng/mL), while the risk in the intermediate ferritin group (33 ng/mL < ferritin < 100 ng/mL; OR: 3.09, 95% CI: 0.65-14.69, p=0.156) and the iron deficiency group (OR: 3.68, 95% CI: 0.77-17.59, p=0.102) increased but was not statistically significant. Compared to the iron repletion group, the risk of composite adverse events was significantly higher in the high ferritin group (OR: 4.21, 95% CI: 2.65-6.68, p<0.001), intermediate ferritin group (OR: 1.64, 95% CI: 1.11-2.44, p=0.014), and iron deficiency group (OR: 1.80, 95% CI: 1.18-2.73, p=0.006). Additionally, the duration of mechanical ventilation and intensive care in the high ferritin group was much longer than that in the other three groups, followed by the iron deficiency group, intermediate ferritin group, and iron repletion group. Iron deficiency (OR: 1.28, 95% CI: 1.08-1.52, p=0.005), and high ferritin (OR: 1.71, 95% CI: 1.23-2.37, p=0.002) were both independent risk factors for perioperative red blood cell transfusion.

 

Conclusion

In pediatric patients with CHD who are scheduled to receive cardiac surgery, nearly one quarter are affected by iron deficiency, and more than 10% of the patients suffer from anemia. Iron deficiency is more prevalent in patients with cyanotic CHD, whereas the incidence of anemia is relatively lower. Compared to patients with iron repletion, both iron deficiency and high ferritin statuses are risk factors for adverse outcomes.

 

 

Part 4. Association Between Thrombocytopenia Occurring During Intensive Care and Prognosis in Pediatric Congenital Heart Disease

 

Background

Thrombocytopenia is associated with the risks of bleeding, blood transfusions, and death in intensive care unit (ICU). The existing studies on the issue of thrombocytopenia in children with congenital heart disease (CHD) after cardiac surgery with cardiopulmonary bypass, and the association between thrombocytopenia and prognosis, are extremely limited.

 

Methods

We conducted a retrospective analysis of patients under the age of 18 with CHD who underwent cardiac surgery with cardiopulmonary bypass from January 2017 to December 2022. The platelet count results during the first 7 days of ICU stay were collected. Platelet below 150×10^⁹/L was defined as thrombocytopenia, and it was further categorized into mild (100-150×10^⁹/L), moderate (50-100×10^⁹/L), and severe (<50×10^⁹/L). The primary endpoint event of this study was death at 30 days postoperatively. Logistic regression models were used to identify risk factors for thrombocytopenia, and Cox regression models were employed to explore the correlation between thrombocytopenia and 30-day postoperative mortality.

 

Results

This study included 11761 pediatric patients with CHD (51.5% male), with a median age of 1.7 years [IQR (Interquartile range): 0.7-3.7 years), and approximately 20% were diagnosed with cyanotic CHD. A total of 4007 (34.1%) patients developed thrombocytopenia in the ICU, with 2773 cases (23.6%), 987 cases (8.4%), and 247 cases (2.1%) having mild, moderate, and severe thrombocytopenia, respectively. The preoperative baseline platelet levels were generally around 300×10^⁹/L, but the platelet levels dropped sharply on the first day after surgery and remained low for the next two days before gradually recovering. Overall, patients with acyanotic CHD had higher platelet levels at each preoperative and postoperative testing time point compared to patients with cyanotic CHD. Younger age, cyanotic CHD, longer cardiopulmonary bypass duration, lower preoperative red blood cell count, preoperative thrombocytopenia, preoperative coagulation dysfunction, increased preoperative red blood cell distribution width, and higher hematocrit were risk factors for thrombocytopenia of all severity. Multivariate Cox regression analysis showed that moderate ICU thrombocytopenia [HR (Hazard ratio): 11.38, 95% CI (Confidence interval): 3.02-42.87, p<0.001], severe ICU thrombocytopenia (HR: 49.54, 95% CI: 13.11-187.14, p<0.001), duration of cardiopulmonary bypass (HR: 1.01, 95% CI: 1.01-1.01, p<0.001), cyanotic CHD (HR: 2.59, 95% CI: 1.13-5.93, p=0.024), preoperative hematocrit (HR: 0.95, 95% CI: 0.91-1.00, p=0.030), preoperative serum sodium (HR: 0.89, 95% CI: 0.80-0.99, p=0.025), and preoperative high-sensitivity C-reactive protein greater than 3 mg/L (HR: 2.46, 95% CI: 1.17-5.17, p=0.017) were independently associated with the risk of 30-day mortality. Kaplan-Meier curves showed that the 30-day survival rate of patients with moderate and severe ICU thrombocytopenia was significantly lower than that of patients without thrombocytopenia. However, the 30-day survival rate of patients with mild ICU thrombocytopenia was similar to that of patients without thrombocytopenia. Additionally, with the increase in the severity of ICU thrombocytopenia, there was a “dose-dependent” increase in the total volume and proportion of component transfusions, the incidence of severe postoperative bleeding and thrombotic events, length of ICU stay, and duration of mechanical ventilation.

 

Conclusion

More than one-third of children with CHD experience thrombocytopenia during intensive care. Moderate and severe thrombocytopenia are both independent risk factors for 30-day mortality. It is recommended that dynamic monitoring of postoperative platelet levels be conducted in clinical practice for patients with CHD, and additional attention should be paid to the warning role of platelet tests for poor prognosis.