scholarly journals Adverse Outcomes Prediction for Congenital Heart Surgery: A Machine Learning Approach

2021 ◽  
pp. 215013512110071
Author(s):  
Dimitris Bertsimas ◽  
Daisy Zhuo ◽  
Jack Dunn ◽  
Jordan Levine ◽  
Eugenio Zuccarelli ◽  
...  
Keyword(s):  
Machine Learning ◽  
Risk Factors ◽  
Congenital Heart ◽  
Heart Surgery ◽  
Classification Trees ◽  
Congenital Heart Surgery ◽  
Assessment Tools ◽  
Hospital Length Of Stay ◽  
Total Quality ◽  
Optimal Classification

Objective: Risk assessment tools typically used in congenital heart surgery (CHS) assume that various possible risk factors interact in a linear and additive fashion, an assumption that may not reflect reality. Using artificial intelligence techniques, we sought to develop nonlinear models for predicting outcomes in CHS. Methods: We built machine learning (ML) models to predict mortality, postoperative mechanical ventilatory support time (MVST), and hospital length of stay (LOS) for patients who underwent CHS, based on data of more than 235,000 patients and 295,000 operations provided by the European Congenital Heart Surgeons Association Congenital Database. We used optimal classification trees (OCTs) methodology for its interpretability and accuracy, and compared to logistic regression and state-of-the-art ML methods (Random Forests, Gradient Boosting), reporting their area under the curve (AUC or c-statistic) for both training and testing data sets. Results: Optimal classification trees achieve outstanding performance across all three models (mortality AUC = 0.86, prolonged MVST AUC = 0.85, prolonged LOS AUC = 0.82), while being intuitively interpretable. The most significant predictors of mortality are procedure, age, and weight, followed by days since previous admission and any general preoperative patient risk factors. Conclusions: The nonlinear ML-based models of OCTs are intuitively interpretable and provide superior predictive power. The associated risk calculator allows easy, accurate, and understandable estimation of individual patient risks, in the theoretical framework of the average performance of all centers represented in the database. This methodology has the potential to facilitate decision-making and resource optimization in CHS, enabling total quality management and precise benchmarking initiatives.

Benchmarking in Congenital Heart Surgery Using Machine Learning-Derived Optimal Classification Trees

2021 ◽  
pp. 215013512110512
Author(s):  
Dimitris Bertsimas ◽  
Daisy Zhuo ◽  
Jordan Levine ◽  
Jack Dunn ◽  
Zdzislaw Tobota ◽  
...  
Keyword(s):  
Machine Learning ◽  
Congenital Heart ◽  
Heart Surgery ◽  
Ventilatory Support ◽  
Classification Trees ◽  
Length Of Hospital Stay ◽  
Congenital Heart Surgery ◽  
Case Mix ◽  
Specific Patient ◽  
Optimal Classification

Background: We have previously shown that the machine learning methodology of optimal classification trees (OCTs) can accurately predict risk after congenital heart surgery (CHS). We have now applied this methodology to define benchmarking standards after CHS, permitting case-adjusted hospital-specific performance evaluation. Methods: The European Congenital Heart Surgeons Association Congenital Database data subset (31 792 patients) who had undergone any of the 10 “benchmark procedure group” primary procedures were analyzed. OCT models were built predicting hospital mortality (HM), and prolonged postoperative mechanical ventilatory support time (MVST) or length of hospital stay (LOS), thereby establishing case-adjusted benchmarking standards reflecting the overall performance of all participating hospitals, designated as the “virtual hospital.” These models were then used to predict individual hospitals’ expected outcomes (both aggregate and, importantly, for risk-matched patient cohorts) for their own specific cases and case-mix, based on OCT analysis of aggregate data from the “virtual hospital.” Results: The raw average rates were HM = 4.4%, MVST = 15.3%, and LOS = 15.5%. Of 64 participating centers, in comparison with each hospital's specific case-adjusted benchmark, 17.0% statistically (under 90% confidence intervals) overperformed and 26.4% underperformed with respect to the predicted outcomes for their own specific cases and case-mix. For MVST and LOS, overperformers were 34.0% and 26.4%, and underperformers were 28.3% and 43.4%, respectively. OCT analyses reveal hospital-specific patient cohorts of either overperformance or underperformance. Conclusions: OCT benchmarking analysis can assess hospital-specific case-adjusted performance after CHS, both overall and patient cohort-specific, serving as a tool for hospital self-assessment and quality improvement.

Reducing readmissions following paediatric cardiothoracic surgery: a quality improvement initiative

2014 ◽  
Vol 25 (5) ◽  
pp. 935-940 ◽  
Author(s):  
Brian Kogon ◽  
Kim Woodall ◽  
Kirk Kanter ◽  
Bahaaldin Alsoufi ◽  
Matt Oster
Keyword(s):  
Risk Factors ◽  
Hospital Stay ◽  
Congenital Heart ◽  
Heart Surgery ◽  
Failure To Thrive ◽  
Congenital Heart Surgery ◽  
Cardiothoracic Surgery ◽  
Hispanic Ethnicity ◽  
Multivariate Risk ◽  
Initial Hospital

AbstractBackground: We have previously identified risk factors for readmission following congenital heart surgery – Hispanic ethnicity, failure to thrive, and original hospital stay more than 10 days. As part of a quality initiative, changes were made to the discharge process in hopes of reducing the impact. All discharges were carried out with an interpreter, medications were delivered to the hospital before discharge, and phone calls were made to families within 72 hours following discharge. We hypothesised that these changes would decrease readmissions. Methods: The current cohort of 635 patients underwent surgery in 2012. Demographic, preoperative, operative, and postoperative variables were evaluated. Univariate and multivariate risk factor analyses were performed. Comparisons were made between the initial (2009) and the current (2012) cohorts. Results: There were 86 readmissions of 77 patients during 2012. Multivariate risk factors for readmission were risk adjustment for congenital heart surgery score and initial hospital stay >10 days. In comparing 2009 with 2012, the overall readmission rate was similar (10 versus 12%, p=0.27). Although there were slight decreases in the 2012 readmissions for those patients with Hispanic ethnicity (18 versus 16%, p=0.79), failure to thrive (23 versus 17%, p=0.49), and initial hospital stay >10 days (22 versus 20%, p=0.63), they were not statistically significant. Conclusions: Potential risk factors for readmission following paediatric cardiothoracic surgery have been identified. Although targeted modifications in discharge processes can be made, they may not reduce readmissions. Efforts should continue to identify modifiable factors that can reduce the negative impact of hospital readmissions.

Junctional Ectopic Tachycardia: Recognition and Modern Management Strategies

2020 ◽  
Vol 40 (1) ◽  
pp. 46-55
Author(s):  
Kirsti G. Catton ◽  
Jennifer K. Peterson
Keyword(s):  
Risk Factors ◽  
Congenital Heart ◽  
Heart Surgery ◽  
Early Recognition ◽  
Treatment Options ◽  
Age Groups ◽  
Atrioventricular Node ◽  
Congenital Heart Surgery ◽  
Junctional Ectopic Tachycardia ◽  
Perioperative Morbidity

Junctional ectopic tachycardia is a common dysrhythmia after congenital heart surgery that is associated with increased perioperative morbidity and mortality. Risk factors for development of junctional ectopic tachycardia include young age (neonatal and infant age groups); hypomagnesemia; higher-complexity surgical procedure, especially near the atrioventricular node or His bundle; and use of exogenous catecholamines such as dopamine and epinephrine. Critical care nurses play a vital role in early recognition of dysrhythmias after congenital heart surgery, assessment of hemodynamics affecting cardiac output, and monitoring the effects of antiarrhythmic therapy. This article reviews the underlying mechanisms of junctional ectopic tachycardia, incidence and risk factors, and treatment options. Currently, amiodarone is the pharmacological treatment of choice, with dexmedetomidine increasingly used because of its anti-arrhythmic properties and sedative effect.

Beyond 30 Days: Analysis of Unplanned Readmissions During the First Year Following Congenital Heart Surgery

2020 ◽  
Vol 11 (2) ◽  
pp. 177-182 ◽  
Author(s):  
Entela B. Lushaj ◽  
Joshua Hermsen ◽  
Glen Leverson ◽  
Susan G. MacLellan-Tobert ◽  
Kari Nelson ◽  
...  
Keyword(s):  
Risk Factors ◽  
High Risk ◽  
Congenital Heart ◽  
Heart Surgery ◽  
Hospital Readmissions ◽  
Congenital Heart Surgery ◽  
First Year ◽  
Term Survival ◽  
Viral Illness ◽  
First 90 Days

Objective: We investigated the incidence and etiologies for unplanned hospital readmissions during the first year following congenital heart surgery (CHS) at our institution and the potential association of readmissions with longer term survival. Methods: We retrospectively reviewed 263 patients undergoing CHS at our institution from August 2011 to June 2015. Scheduled readmissions were excluded. Results: Seventy patients accrued a total of 120 readmissions (1.7 readmission/patient) within one year after surgery. The first readmission for 57% of the patients was within 30 days postdischarge. Twenty-two patients were first readmitted between 31 and 90 days postdischarge. Eight patients were first readmitted between 90 days and 1 year postdischarge. Median time-to-first readmission was 21 days. Median hospital length of stay at readmission was two days. Causes of 30-day readmissions included viral illness (25%), wound infections (15%), and cardiac causes (15%). Readmissions between 30 and 90 days included viral illness (27%), gastrointestinal (27%), and cardiac causes (9%). Age, STAT category, length of surgery, intubation, intensive care unit, and hospital stay were risk factors associated with readmissions based on logistic regression. Distance to hospital had a significant effect on readmissions ( P < .001). Patients with higher family income were less likely to be readmitted ( P < .001). There was no difference in survival between readmitted and non-readmitted patients ( P = .68). Conclusions: The first 90 days is a high-risk period for unplanned hospital readmissions after CHS. Complicated postoperative course, higher surgical complexity, and lower socioeconomic status are risk factors for unplanned readmissions the first 90 days after surgery. Efforts to improve the incidence or readmission after CHS should extend to the first 3 months after surgery and target these high-risk patient populations.

Alternative strategies in newborns and infants with major co-morbidities to improve congenital heart surgery outcomes at an emerging programme

2015 ◽  
Vol 26 (3) ◽  
pp. 485-492 ◽  
Author(s):  
Jannika Dodge-Khatami ◽  
Ali Dodge-Khatami ◽  
Jarrod D. Knudson ◽  
Samantha R. Seals ◽  
Avichal Aggarwal ◽  
...  
Keyword(s):  
Risk Factors ◽  
High Risk ◽  
Congenital Heart ◽  
Heart Surgery ◽  
Congenital Heart Surgery ◽  
Hospital Length ◽  
Surgical Strategies ◽  
Prolonged Intubation ◽  
Co Morbidity ◽  
Neonates And Infants

AbstractIntroductionDebilitating patient-related non-cardiac co-morbidity cumulatively increases risk for congenital heart surgery. At our emerging programme, flexible surgical strategies were used in high-risk neonates and infants generally considered in-operable, in an attempt to make them surgical candidates and achieve excellent outcomes.Materials and methodsBetween April, 2010 and November, 2013, all referred neonates (142) and infants (300) (average scores: RACHS 2.8 and STAT 3.0) underwent 442 primary cardiac operations: patients with bi-ventricular lesions underwent standard (n=294) or alternative (n=19) repair/staging strategies, such as pulmonary artery banding(s), ductal stenting, right outflow patching, etc. Patients with uni-ventricular hearts followed standard (n=96) or alternative hybrid (n=34) staging. The impact of major pre-operative risk factors (37%), standard or alternative surgical strategy, prematurity (50%), gestational age, low birth weight, genetic syndromes (23%), and major non-cardiac co-morbidity requiring same admission surgery (27%) was analysed on the need for extracorporeal membrane oxygenation, mortality, length of intubation, as well as ICU and hospital length of stays.ResultsThe need for extracorporeal membrane oxygenation (8%) and hospital survival (94%) varied significantly between surgical strategy groups (p=0.0083 and 0.028, respectively). In high-risk patients, alternative bi- and uni-ventricular strategies minimised mortality, but were associated with prolonged intubation and ICU stay. Major pre-operative risk factors and lower weight at surgery significantly correlated with prolonged intubation, hospital length of stay, and mortality.DiscussionIn our emerging programme, flexible surgical strategies were offered to 53/442 high-risk neonates and infants with complex CHDs and significant non-cardiac co-morbidity, in order to buffer risk and achieve patient survival, although at the cost of increased resource utilisation.

scholarly journals Risk Factors for Death After Adult Congenital Heart Surgery in Pediatric Hospitals

2011 ◽  
Vol 4 (4) ◽  
pp. 433-439 ◽  
Author(s):  
Yuli Y. Kim ◽  
Kimberlee Gauvreau ◽  
Emile A. Bacha ◽  
Michael J. Landzberg ◽  
Oscar J. Benavidez
Keyword(s):  
Risk Factors ◽  
Congenital Heart ◽  
Heart Surgery ◽  
Congenital Heart Surgery ◽  
Pediatric Hospitals ◽  
Adult Congenital

Abstract 220: Technical Performance Scores Predict Outcomes Following Congenital Heart Surgery: Multicenter Validation

2014 ◽  
Vol 7 (suppl_1) ◽  
Author(s):  
Meena Nathan ◽  
Hua Liu ◽  
Steven D Colan ◽  
Lazaros Kochilas ◽  
Geetha Raghuveer ◽  
...  
Keyword(s):  
Adverse Events ◽  
Hospital Stay ◽  
Congenital Heart ◽  
Heart Surgery ◽  
Assessment Tool ◽  
Multivariable Analysis ◽  
Congenital Heart Surgery ◽  
Hospital Length Of Stay ◽  
Atrioventricular Canal ◽  
Technical Performance

BACKGROUND: In previous work from a single center, Technical Performance Score (TPS), a tool that assesses technical adequacy of repair, has been shown to be strongly associated with outcomes in congenital cardiac surgery. We sought to validate the efficacy of TPS in a multicenter environment. METHODS: All patients (1 day to 62 years) who were discharged from January 1 to December 31, 2011; and who underwent 9 congenital cardiac procedures (Arterial switch operation [84], Bidirectional Glenn [75], Atrioventricular canal repair [135], Fontan [97], Arch repair on pump [58], Stage I Procedure [85], Pulmonary valve replacement [116], Tetralogy of Fallot repair [112], and Ventricular septal defect repair [163]); from 5 centers were included. Based on echocardiograms (echo) prior to discharge or death, and unplanned reinterventions at surgical site; TPS was assigned using previously established criteria. Case complexity was determined by RACHS-1 category. Outcomes included (a) major postoperative adverse events (AE) excluding unplanned reinterventions, (b) length of ventilation, and (c) postoperative hospital stay. Adjusted analysis used logistic/linear regression to determine odds ratio (OR) and regression coefficient (b) for each outcome. RESULTS: There were 925 hospital discharges: 418 (45%) were RACHS-1 category 2, 295 (32%) category 3, 85 (9%) category 4, 86 (9%) category 6 and the cohort included 41 (4%) adults. TPS were as follows: 491 (53%) class 1-optimal, 263 (28%) class 2-adequate, 131 (14%) class 3-inadequate and 40 (4%) had no TPS assigned because of a lack of or incomplete echos (NA). There were 26 (2.8%) deaths (81% of deaths were in class 3) and 105 (11%) adverse events. Occurrence of major adverse events, ventilation time and hospital length of stay were all significantly higher in class 3 (Figure). On multivariable analysis adjusting for age, RACHS-1, prematurity, and presence of non-cardiac anomalies; Class 3 TPS was associated with a higher odds of AE (OR 7.4, CI 4.1-13.2, p<0.001), longer ventilation (b 1.9, CI 1.6-2.2, p<0.001), and hospital stay (b 1.6, CI 1.4 to 1.8, p<0.001). CONCLUSION: TPS predicts outcomes after congenital heart surgery in a multicenter cohort, and can serve as quality assessment tool. Outcomes may be favorably influenced by focusing on technical excellence.

A retrospective study of perioperative clinical seizures and epilepsy in children after operation for CHD

2021 ◽  
pp. 1-7
Author(s):  
Takeshi Ikegawa ◽  
Shin Ono ◽  
Kouji Yamamoto ◽  
Mikihiro Shimizu ◽  
Sadamitsu Yanagi ◽  
...  
Keyword(s):  
Risk Factors ◽  
Cardiac Surgery ◽  
White Matter ◽  
Risk Adjustment ◽  
Congenital Heart ◽  
Heart Surgery ◽  
Congenital Heart Surgery ◽  
White Matter Integrity ◽  
Partial Regression Coefficient ◽  
Clinical Seizure

Abstract This study investigated the incidence and risk factors of perioperative clinical seizure and epilepsy in children after operation for CHD. We included 777 consecutive children who underwent operation from January 2013 to December 2016 at Kanagawa Children’s Medical Center, Kanagawa, Japan. Perinatal, perioperative, and follow-up medical data were collected. Elastic net regression and mediation analysis were performed to investigate risk factors of perioperative clinical seizure and epilepsy. Anatomic CHD classification was performed based on the preoperative echocardiograms; cardiac surgery was evaluated using Risk Adjustment in Congenital Heart Surgery 1. Twenty-three (3.0%) and 15 (1.9%) patients experienced perioperative clinical seizure and epilepsy, respectively. Partial regression coefficient with epilepsy as the objective variable for anatomical CHD classification, Risk Adjustment in Congenital Heart Surgery 1, and the number of surgeries was 0.367, 0.014, and 0.142, respectively. The proportion of indirect effects on epilepsy via perioperative clinical seizure was 22.0, 21.0, and 33.0%, respectively. The 15 patients with epilepsy included eight cases with cerebral infarction, two cases with cerebral haemorrhage, and three cases with hypoxic-ischaemic encephalopathy; white matter integrity was not found. Anatomical complexity of CHD, high-risk cardiac surgery, and multiple cardiac surgeries were identified as potential risk factors for developing epilepsy, with a low rate of indirect involvement via perioperative clinical seizure and a high rate of direct involvement independently of perioperative clinical seizure. Unlike white matter integrity, stroke and hypoxic-ischaemic encephalopathy were identified as potential factors for developing epilepsy.

Incidence, risk factors, and outcomes of acute kidney injury in adults undergoing surgery for congenital heart disease

2016 ◽  
Vol 27 (6) ◽  
pp. 1068-1075 ◽  
Author(s):  
David M. Kwiatkowski ◽  
Elizabeth Price ◽  
David M. Axelrod ◽  
Anitra W. Romfh ◽  
Brian S. Han ◽  
...  
Keyword(s):  
Risk Factors ◽  
Acute Kidney Injury ◽  
Congenital Heart Disease ◽  
Cardiac Surgery ◽  
Heart Disease ◽  
Congenital Heart ◽  
Heart Surgery ◽  
Kidney Injury ◽  
Congenital Heart Surgery ◽  
Left Ventricular

AbstractBackgroundAcute kidney injury after cardiac surgery is a frequent and serious complication among children with congenital heart disease (CHD) and adults with acquired heart disease; however, the significance of kidney injury in adults after congenital heart surgery is unknown. The primary objective of this study was to determine the incidence of acute kidney injury after surgery for adult CHD. Secondary objectives included determination of risk factors and associations with clinical outcomes.MethodsThis single-centre, retrospective cohort study was performed in a quaternary cardiovascular ICU in a paediatric hospital including all consecutive patients ⩾18 years between 2010 and 2013.ResultsData from 118 patients with a median age of 29 years undergoing cardiac surgery were analysed. Using Kidney Disease: Improving Global Outcome creatinine criteria, 36% of patients developed kidney injury, with 5% being moderate to severe (stage 2/3). Among higher-complexity surgeries, incidence was 59%. Age ⩾35 years, preoperative left ventricular dysfunction, preoperative arrhythmia, longer bypass time, higher Risk Adjustment for Congenital Heart Surgery-1 category, and perioperative vancomycin use were significant risk factors for kidney injury development. In multivariable analysis, age ⩾35 years and vancomycin use were significant predictors. Those with kidney injury were more likely to have prolonged duration of mechanical ventilation and cardiovascular ICU stay in the univariable regression analysis.ConclusionsWe demonstrated that acute kidney injury is a frequent complication in adults after surgery for CHD and is associated with poor outcomes. Risk factors for development were identified but largely not modifiable. Further investigation within this cohort is necessary to better understand the problem of kidney injury.

Sign in / Sign up

Export Citation Format

Share Document