<b>Weaning in Post-Cardiac Surgery</b>

<b>Background:</b>Weaning from mechanical ventilation is a crucial clinical hurdle for patients after cardiac surgery. Successfully liberating patients from the ventilator can considerably enhance their recovery and survival rates. The objective of this study is to create and validate a clinical prediction model that can assess the probability of successful extubation in patients following cardiac surgery, aiming to assist clinicians in making more accurate decisions during the weaning process from mechanical ventilation.<b>Method:</b>We conducted a single-center retrospective study enrolling 148 adult patients after cardiac surgery. All patients involved in the analysis were randomly assigned to a training set and a validation set. Univariate and multivariate analyses were performed to identify risk factors for weaning from mechanical ventilation in the training cohort. Based on these independent predictors, the nomograms were constructed to predict the probability of weaning and were validated in the validation cohort.<b>Results:</b> Among the 148 adults included in our study, 49 patients (33.1%) experienced delayed extubation. A predictive scoring system was derived based on 5 identified risk factors including Proportion of male, EuroscoreII, operation time, pump time, bleeding during operation and BNP level. According to the predictive system, the 5 independent predictors were used to construct a full nomogram. The model performed well in the validation set with AUC of 0.760 and a significant p-value of 0.956 in the Hosmer-Lemeshow test. The DCA curve and clinical impact curve showed a good clinical utility of this model.<b>Conclusion</b><b>：</b>We developed and validated nomogram model to predict early extubation after cardiac surgery. The nomogram may have clinical utility in risk estimation, risk stratification, and targeted potential preventive measures.<b>背景：</b>心脏手术后患者的机械通气脱机是一项关键的临床难题。成功协助患者脱离呼吸机可显著改善其康复结局与生存率。本研究旨在构建并验证一款临床预测模型，用于评估心脏手术后患者成功拔管的概率，以期辅助临床医师在机械通气脱机流程中做出更为精准的诊疗决策。 <b>方法：</b>本研究开展单中心回顾性研究，纳入148名心脏手术后的成年患者。所有纳入分析的患者被随机分配至训练集与验证集。在训练队列中，通过单因素与多因素分析筛选机械通气脱机相关危险因素。基于筛选得到的独立预测因子，构建列线图（nomogram）以预测脱机成功概率，并在验证队列中对该模型进行验证。 <b>结果：</b>本研究纳入的148名成年患者中，49例（33.1%）出现拔管延迟。基于5个筛选出的危险因素（包括男性比例、Euroscore II、手术时长、体外循环时间、术中出血量及B型钠尿肽（BNP）水平）构建预测评分系统。随后以该5个独立预测因子为基础，构建完整列线图。该模型在验证集中表现优异，受试者工作特征曲线下面积（AUC）为0.760，Hosmer-Lemeshow拟合优度检验的P值为0.956，具有统计学意义。决策曲线分析（DCA）曲线与临床影响曲线均证实该模型具备良好的临床应用价值。 <b>结论：</b>本研究构建并验证了一款可预测心脏手术后早期拔管的列线图模型。该列线图可在风险评估、风险分层及针对性潜在预防措施的制定中发挥临床应用价值。