Development and Validation of a Score to Identify Cardiac Surgery Patients at High Risk of Prolonged Mechanical Ventilation

doi:10.1053/j.jvca.2019.03.009

Journal of Cardiothoracic and Vascular Anesthesia

Volume 33, Issue 10, October 2019, Pages 2709-2716

https://doi.org/10.1053/j.jvca.2019.03.009 Get rights and content

Objective

To develop and validate a score for the early identification of cardiac surgery patients at high risk of prolonged mechanical ventilation (MV) who may be suitable targets for interventional trials.

Design

Retrospective analysis.

Setting

Tertiary intensive care unit.

Participants

Cardiac surgery patients.

Interventions

Observational study.

Measurements and Main Results

The study comprised 1,994 patients. Median age was 67 years, and 1,457 patients (74%) were male. Median duration of MV was 9.4 hours. A total of 229 (11%), 182 (9%), and 127 (6%) patients received MV for ≥24, ≥36, and ≥48 hours, respectively. In-hospital mortality was 13%, 15%, and 17%, respectively. For the study model, all preoperative, intraoperative, and early (first 4 hours) postoperative variables were considered. A multivariable logistic regression model was developed, and a predictive scoring system was derived. Using MV ≥24 hours as the primary outcome, the model performance in the development set was good with a c-index of 0.876 (95% confidence interval 0.846-0.905) and a Brier's score of 0.062. In the validation set, the c-index was 0.907 (0.867-0.948), Brier's score was 0.059, and the model remained well calibrated.

Conclusions

The authors developed a simple score to predict prolonged MV after cardiac surgery. This score, if externally validated, is potentially suitable for identifying a high-risk target population for future randomized controlled trials of postoperative care after cardiac surgery.

Section snippets

Study Design and Ethics Approval

Demographics, biochemical results, and blood product administration were obtained from local administrative, laboratory, and clinical databases. Also, data were obtained from the following 2 clinical quality registries—the Australian and New Zealand Society of Cardiac and Thoracic Surgeons Cardiac Surgery Database, which contains detailed information on all cardiothoracic surgery procedures, including preoperative, intraoperative, and postoperative data,¹² and the Australian and New Zealand

Patient Characteristics and Outcome

For the present study, 2,150 cardiothoracic surgery admissions were screened. Of these, 156 patients were excluded for any of the the following reasons: (1) some did not have sufficient biochemical data available (n = 38); (2) complete data on MV was not available (n = 3); (3) death during the surgical procedure (n = 7); or (4) the only data available related to their ICU admission were precardiac surgery (n = 93). In addition, 15 readmissions also were excluded. The final database consisted of

Key Findings

In a study of nearly 2,000 patients, 11%, 9%, and 6% of patients were mechanically ventilated for more than 24, 36, and 48 hours, respectively. Such prolonged MV carried a high mortality of 13%, 15%, and 17% for these respective ventilation cutoffs. Moreover, significant early (first 4 hours in ICU) predictors of prolonged ventilation were observed, and a score to help predict such ventilation early after ICU admission was developed. When using this score, good predictive performance was

Conclusions

In a study of a large cohort of cardiac surgery patients, prolonged MV of more than 24, 36, and 48 hours was observed to be common and associated with a high risk of death. The authors developed a predictive score to identify patients at risk of prolonged MV early in their ICU admission. This predictive score had good discrimination, has the potential of being used for risk stratification and inclusion of patients in future interventional studies, and now needs further validation in a large

Conflicts of Interest

The authors have no conflicts of interest to disclose.

References (39)

J.P. Jacobs et al.
What is operative mortality? Defining death in a surgical registry database: A report of the STS Congenital Database Taskforce and the Joins EACTS-STS Congenital Database Committee
Ann Thorac Surg
(2006)
J.L. Trouillet et al.
Prolonged mechanical ventilation after cardiac surgery: Outcome and predictors
J Thorac Cardiovasc Surg
(2009)
M.L. Bailey et al.
Risk factors and survival in patients with respiratory failure after cardiac operations
Ann Thorac Surg
(2011)
E. Durand et al.
Performance analysis of EuroSCORE II compared to the original logistic EuroSCORE and STS scores for predicting 30-day mortality after transcatheter aortic valve replacement
Am J Cardiol
(2013)
J.D. Puskas et al.
The Society of Thoracic Surgeons 30-day predicted risk of mortality score also predicts long term survival
Ann Thorac Surg
(2012)
V. Sharma et al.
A derived and validated score to predict prolonged mechanical ventilation in patients undergoing cardiac surgery
J Thorac Cardiovasc Surg
(2017)
P.J. Stow et al.
Development and implementation of a high-quality clinical database: The Australian and New Zealand Intensive Care Society Adult Patient Database
J Crit Care
(2006)
T.G. Coulson et al.
Acute risk change (ARC) identifies outlier institutions in perioperative cardiac surgical care when the standardized mortality rate cannot
Br J Anaesth
(2016)
E. Paul et al.
Risk prediction of hospital mortality of adult patients admitted to the Australian and New Zealand intensive care units: Development and validation of the Australian and New Zealand Risk of Death model
J Crit Care
(2013)
C. Wang et al.
A local risk prediction model for prolonged ventilation after adult heart valve surgery in a Chinese single center
Heart Lung
(2013)

F. Gumus et al.

Prolonged mechanical ventilation after CABG: Risk factor analysis

J Cardiothorac Vasc Anesth

(2015)

F. Filsoufi et al.

Predictors of early and late outcomes of respiratory failure in contemporary cardiac surgery

Chest

(2008)

N. Serrano et al.

Prolonged intubation rates after coronary artery bypass surgery and ICU risk stratification score

Chest

(2005)

S.L.C. Reddy et al.

Logistic risk model for prolonged ventilation after adult cardiac surgery

Ann Thorac Surg

(2007)

S.M. O'Brien et al.

The Society of Thoracic Surgeons 2018 adult cardiac surgery risk models: Part 2-statistical methods and results

Ann Thorac Surg

(2018)

R.R. Bartz et al.

Prolonged support after cardiac surgery: Incidence, risk factors and outcomes: A retrospective cohort study

J Crit Care

(2015)

S.L. Camp et al.

Can timing of tracheal extubation predict improved outcomes after cardiac surgery?

HSR Proc Intensive Care Cardiovasc Anesth

(2009)

S.A. Nashef et al.

European system for cardiac operative risk evaluation (EuroSCORE)

Eur J Cardiothorac Surg

(1999)

S.A. Nashef et al.

EuroSCORE II

Eur J Cardiothorac Surg

(2012)

Cited by (18)

Development and validation of a clinical prediction model for early ventilator weaning in post-cardiac surgery
2024, Heliyon
Weaning patients from mechanical ventilation is a critical clinical challenge post cardiac surgery. The effective liberation of patients from the ventilator significantly improves their recovery and survival rates. This study aimed to develop and validate a clinical prediction model to evaluate the likelihood of successful extubation in post-cardiac surgery patients.
A predictive nomogram was constructed for extubation success in individual patients, and receiver operating characteristic (ROC) and calibration curves were generated to assess its predictive capability. The superior performance of the model was confirmed using Delong's test in the ROC analysis. A decision curve analysis (DCA) was conducted to evaluate the clinical utility of the nomogram.
Among 270 adults included in our study, 107 (28.84%) experienced delayed extubation. A predictive nomogram system was derived based on five identified risk factors, including the proportion of male patients, EuroSCORE II, operation time, pump time, bleeding during operation, and brain natriuretic peptide (BNP) level. Based on the predictive system, five independent predictors were used to construct a full nomogram. The area under the curve values of the nomogram were 0.880 and 0.753 for the training and validation cohorts, respectively. The DCA and clinical impact curves showed good clinical utility of this model.
Delayed extubation and weaning failure, common and potentially hazardous complications following cardiac surgery, vary in timing based on factors such as sex, EuroSCORE II, pump duration, bleeding, and postoperative BNP reduction. The nomogram developed and validated in this study can accurately predict when extubation should occur in these patients. This tool is vital for assessing risks on an individual basis and making well-informed clinical decisions.
Development and Validation of Scores to Predict Prolonged Mechanical Ventilation after Cardiac Surgery
2024, Journal of Cardiothoracic and Vascular Anesthesia
To optimize the early prediction of prolonged postoperative mechanical ventilation after cardiac surgery (>24 hours postoperatively).
The authors performed a retrospective analysis.
The Australian and New Zealand Society of Cardiac and Thoracic Surgeons (ANZSCTS) database was utilized.
All patients included in the ANZSCTS database between January 2015 and December 2018 were analyzed.
No interventions were performed in this observational study.
A previously developed model was modified to allow retrospective risk calculation and model assessment (Modified Hessels score). The database was split into development and validation sets. A new risk model was developed using forward and backward stepwise elimination (ANZ-PreVent score). The authors assessed 48,382 patients, of whom 5004 (10.3%) were ventilated mechanically for >24 hours post-operatively. The Modified Hessels score demonstrated good performance in this database, with a c-index of 0.78 (95% CI 0.77-0.78) and a Brier score of 0.08. The newly developed ANZ-PreVent score demonstrated better performance (validation cohort, n = 12,229), with a c-index of 0.84 (95% CI 0.83-0.85) (p < 0.0001) and a Brier score of 0.07. Both scores performed better than the severity of illness scores commonly used to predict outcomes in intensive care.
The authors validated a modified version of an existing prediction score and developed the ANZ-PreVent score, with improved performance for identifying patients at risk of ventilation for >24 hours. The improved score can be used to identify high-risk patients for targeted interventions in future randomized controlled trials.
Developing an explainable machine learning model to predict the mechanical ventilation duration of patients with ARDS in intensive care units
2023, Heart and Lung
Acute respiratory distress syndrome (ARDS) is common in intensive care units with high mortality rate and mechanical ventilation (MV) is the most important related treatment. Early prediction of MV duration has benefit for patients risk stratification and care strategies support.
To develop an explainable model for predicting mechanical ventilation (MV) duration in patients with ARDS using the machine learning (ML) approach.
The number of 1,148, 1,697, and 29 ARDS patients admitted to intensive care units (ICU) in the MIMIC-IV, eICU-CRD, and AmsterdamUMCdb databases were included in the study. Features at MV initiation from the MIMIC-IV dataset were used to train prediction models based on seven supervised machine learning algorithms. After 5-fold cross-validation for hyperparameters tuning, the hyperparameters- optimized model of different algorithms was tested by external datasets extracted from eICU-CRD and Amsterdamumcdb. Finally, three descriptive machine learning explanation methods were conducted for the model explanation.
The XGBoosting model showed the most stable and accurate performance among two testing datasets (RMSE= 5.57 and 5.46 days in eICU-CRD and AmsterdamUMCdb) and was selected as the optimal model. The model explanation based on SHAP, LIME, and DALEX results showed a consistent result, vasopressor, PH, and SOFA score had the highest effect on MV duration prediction.
ML models with features at MV initiation can accurate predict MV duration in patients with ARDS in ICUs. Among seven algorithms, XGB models showed the best performance (RMSE= 5.57 and 5.46 in two external datasets). LIME, SHAP, and Breakdown methods showed good performance as AXI methods.
Development and Validation of a Predictive Score for Prolonged Mechanical Ventilation After Cardiac Surgery
2022, Journal of Cardiothoracic and Vascular Anesthesia
Citation Excerpt :
The study authors analysis identified five major variables associated with the occurrence of PMV. These variables are known to be associated with the occurrence of a PMV.2,5,6,13-15 In the study by Sharma et al,2 the authors found two preoperative values associated with the occurrence of PMV: prior cardiac surgery and the New York Heart Association Class IV.
The authors aimed to identify risk factors associated with prolonged mechanical ventilation (PMV) after scheduled cardiac surgery under cardiopulmonary bypass (CPB).
A single-center, observational study.
Tertiary hospital.
All adult patients who underwent scheduled cardiac surgery under cardiopulmonary bypass between January 2017 and December 2017.
None.
Among the 568 patients included, 68 (12.0%) presented a PMV. The median ventilation time was 5.7 hours in the group without PMV and 85.2 hours in the group with PMV. A logistic regression found five variables independently associated with the occurrence of PMV: (1) prior cardiac surgery, (2) preoperative congestive heart failure, (3) preoperative creatinine clearance <30 mL/min/1.73 m², (4) intraoperative implantation of extracorporeal membrane oxygenation, and (5) serum lactate >4 mmol/L on admission. A predictive score to allow the authors to anticipate PMV was developed from the regression coefficient of perioperative factors independently associated with PMV. With a threshold of 2/13, the score had a sensitivity of 80.9%, a specificity of 80.5%, a positive predictive value of 37.2%, and a negative predictive value of 96.7%. The score then was validated in a distinct cohort.
The study authors have developed a simple score to predict PMV in patients undergoing cardiac surgery with CPB. This score could allow clinicians to identify a high-risk population that might benefit from specific management upon arrival in the intensive care unit.
Continuous Magnesium Infusion to Prevent Atrial Fibrillation After Cardiac Surgery: A Sequential Matched Case-Controlled Pilot Study
2020, Journal of Cardiothoracic and Vascular Anesthesia
Citation Excerpt :
The variables adjusted for were European System for Cardiac Operative Risk Evaluation, type of surgery, age, sex, duration of CPB, diagnosis of diabetes, peripheral vascular disease, and cerebrovascular disease. Matching was performed using independently and prospectively collected data from the routinely collected database from the authors’ hospital cardiac surgery program, which is part of the Australian Society of Cardiothoracic Surgery database.22-24 All matching was conducted by staff who were blinded to the subsequent development or non-development of AF.
The authors aimed to test whether a bolus of magnesium followed by continuous intravenous infusion might prevent the development of atrial fibrillation (AF) after cardiac surgery.
Sequential, matched, case-controlled pilot study.
Tertiary university hospital.
Matched cohort of 99 patients before and intervention cohort of 99 consecutive patients after the introduction of a continuous magnesium infusion protocol.
The magnesium infusion protocol consisted of a 10 mmol loading dose of magnesium sulphate followed by a continuous infusion of 3 mmol/h over a maximum duration of 96 hours or until intensive care unit discharge.
The study groups were balanced except for a lower cardiac index in the intervention cohort. The mean duration of magnesium infusion was 27.93 hours (95% confidence interval [CI]: 24.10-31.76 hours). The intervention group had greater serum peak magnesium levels: 1.72 mmol/L ± 0.34 on day 1, 1.32 ± 0.36 on day 2 versus 1.01 ± 1.14 and 0.97 ± 0.13, respectively, in the control group (p < 0.01). Atrial fibrillation occurred in 25 patients (25.3%) in the intervention group and 40 patients (40.4%) in the control group (odds ratio 0.49, 95% CI, 0.27-0.92; p = 0.023). On a multivariate Cox proportional hazards model, the hazard ratio for the development of AF was significantly less in the intervention group (hazard ratio 0.45, 95% CI, 0.26-0.77; p = 0.004).
The magnesium delivery strategy was associated with a decreased incidence of postoperative AF in cardiac surgery patients. These findings provide a rationale and preliminary data for the design of future randomized controlled trials.
The Effects of Escalation of Respiratory Support and Prolonged Invasive Ventilation on Outcomes of Cardiac Surgical Patients: A Retrospective Cohort Study
2020, Journal of Cardiothoracic and Vascular Anesthesia
Citation Excerpt :
Future studies should aim to obtain such data as they may be significant predictors of respiratory complications after surgery, and if so whether they are modifiable. Several authors have reported outcomes related to prolonged invasive ventilation after cardiac surgery and developed prediction models mainly using 24-, 36-, or 48-hour thresholds for prolonged invasive ventilation.26-33 The standard definition of prolonged invasive ventilation according the Society of Thoracic Surgeons is a duration exceeding 24 hours.13
The aim of this study was to determine the effects of escalation of respiratory support and prolonged postoperative invasive ventilation on patient-centered outcomes, and identify perioperative factors associated with these 2 respiratory complications.
A retrospective cohort analysis of cardiac surgical patients admitted to the cardiothoracic intensive care unit (ICU) between August 2015 and January 2018. Escalation of respiratory support was defined as “unplanned continuous positive airway pressure,” “non-invasive ventilation,” or “reintubation” after surgery; prolonged invasive ventilation was defined as “invasive ventilation beyond the first 12 hours following surgery.” The primary endpoint was the composite of escalation of respiratory support and prolonged ventilation.
Tertiary cardiothoracic ICU.
A total of 2,098 patients were included and analyzed.
None.
The composite of escalation of support or prolonged ventilation occurred in 509 patients (24.3%). Patients who met the composite had higher mortality (2.9% v 0.1%; p < 0.001) and longer median [interquartile range] length of ICU (2.1 [1.0-4.9] v 0.9 [0.8-1.0] days; p < 0.0001) and hospital (10.6 [8.0-16.0] v 7.2 [6.2-10.0] days; p < 0.0001) stay. Hypoxemia and anemia on admission to ICU were the only 2 factors independently associated with the need for escalation of respiratory support or prolonged invasive ventilation.
Escalation of respiratory support or prolonged invasive ventilation is frequently seen in cardiac surgery patients and is highly associated with increased mortality and morbidity. Hypoxemia and anemia on admission to the ICU are potentially modifiable factors associated with escalation of respiratory support or prolonged invasive ventilation.

View all citing articles on Scopus

View full text

Original ArticleDevelopment and Validation of a Score to Identify Cardiac Surgery Patients at High Risk of Prolonged Mechanical Ventilation

Objective

Design

Setting

Participants

Interventions

Measurements and Main Results

Conclusions

Section snippets

Study Design and Ethics Approval

Patient Characteristics and Outcome

Key Findings

Conclusions

Conflicts of Interest

Ann Thorac Surg

J Thorac Cardiovasc Surg

Ann Thorac Surg

Am J Cardiol

Ann Thorac Surg

J Thorac Cardiovasc Surg

J Crit Care

Br J Anaesth

J Crit Care

Heart Lung

J Cardiothorac Vasc Anesth

Chest

Chest

Ann Thorac Surg

Ann Thorac Surg

J Crit Care

Can timing of tracheal extubation predict improved outcomes after cardiac surgery?

HSR Proc Intensive Care Cardiovasc Anesth

European system for cardiac operative risk evaluation (EuroSCORE)

Eur J Cardiothorac Surg

EuroSCORE II

Eur J Cardiothorac Surg

Original Article
Development and Validation of a Score to Identify Cardiac Surgery Patients at High Risk of Prolonged Mechanical Ventilation