CCAS Meeting Reviews
Afternoon Session Review
By Richard Hubbard, MD and Nischal K. Gautam, MD
McGovern Medical School
UTHealth Houston
Session IV: Selected Posters for Oral Presentation
Moderator: Susan C. Nicolson, MD (Children’s Hospital of Philadelphia, Philadelphia)
All of the abstracts submitted to CCAS were reviewed and scored. Three outstanding abstracts were selected for the Oral Presentation Session. A brief description of each presentation is described below.
____________________
Title: Electrocardiography Signal Processing and Serum Amyloid A Biomarker Improves Detection of Coronary Allograft Vasculopathy After Pediatric Heart Transplantation.
Authors: Vu E; Rusin C; Dreyer W; Devaraj S; Easly B; Andropoulos, D; Brady K.
Description: Dr. Vu began his presentation by discussing coronary allograft vasculopathy, a complication which will eventually affect approximately half of the pediatric patients receiving an orthotopic heart transplant. Coronary vasculopathy is associated with sudden death in this population. Despite the prevalence and severity of this complication, diagnosis is difficult. This prospective, case-control study attempted to test the validity of a multivariate test of coronary allograft vasculopathy (CAV). The first component of the test was based upon electrocardiograms performed at the time of follow-up coronary angiography. ST segment vectors were collected, analyzed, and quantified. This signal processing allowed the researchers to examine ST variability.
In patients with diagnosed CAV, ST segment variability was significantly lower than in matched controls with a ROC area under the curve of 0.78. This data was then matched to the level of Serum Amyloid A, a biomarker associated with coronary vasculopathy. A logistic regression analysis using both ST segment variability and biomarker levels increased the ROC area under the curve to 0.87. Dr. Vu and his colleagues, therefore, concluded that the combination of these two diagnostic variables might help to accurately diagnose this disease in the future.
____________________
Title: Associations Between Anthropometric Indices and Outcomes of Congenital Heart Operations in Infants and Young Children: An Analysis of Data from the Society of Thoracic Surgeons Congenital Heart Surgery Database.
Authors: Ross F; Radman M; Jacobs M; Sassano-Miguel C; Joffe D; Hill K; Chiswell K; Feng L; Jacobs J; Vener D; Latham G
Description: Dr. Ross and her colleagues performed a retrospective database study attempting to analyze the relationship between anthropometric indices and outcomes in patients undergoing congenital heart surgery. The Society of Thoracic Surgeons database was queried for patients between the ages of one month to 10 years of age, resulting in 73,417 patients for inclusion in the analysis. Patient Z-scores for weight-for-age, height-for-age, and weight-for-height were recorded.
The outcomes analyzed included mortality, a composite of mortality and major complications, and infection. Mixed effects logistic regression was performed adjusting for procedural and patient factors. Patients with lower weight-for-age and height-for-age had significantly worse outcomes in all three categories analyzed. In addition, patients with both extremes of weight-for-height were noted to have worse outcomes. Interestingly, the associations held for double and single ventricle patients. This study suggests that anthropometric indices may be useful tools for identifying children at increased risk for poor outcomes following cardiac surgery.
____________________
Title: Risk Prediction for One-year Transplant-Free Survival following Norwood Operation for Single Ventricles Using Neural Networks
Authors: Do N; Peck J; Jalali A; Ghazarian S; P. J Ahumada L; Gupta M; Rehman M
Description: Demographic and clinical factors which influence the individual risk for mortality in single ventricle patients are poorly understood. Based upon information from the Pediatric Heart Network’s (PHN) Single Ventricle Reconstruction Trial, this retrospective study of 550 neonates attempted to develop an accurate predictive model for determining individual one-year mortality risk in this patient population. A machine-learning based algorithm using the inputs of 24 variables from the PHN database was developed. Next, a feed-forward neural network classifier and a random forest model were both tested. The predictive strength of different variables was analyzed. Thirteen of the 24 variables were noted to predict 90% of the variations in the data. The individual variables were given different weights in the analysis based on their predictive power.
Of note, the strongest predictors of mortality included poverty, socioeconomic factors, low-birth weight, surgeon volume, and gestational age. The neural network classifier outperformed the predictive power of the random forest modifier with an accuracy of 85% versus 61% and a ROC area under the curve of 0.93 versus 0.52. The ultimate goal is to create a customized, individual-specific mortality prediction for use by both clinicians and parents.
Session V: What Happens When the Patient Goes to the CICU?
Moderator: John Costello, MD, MPH (Medical University of South Carolina Children’s Hospital, Charleston)
Dr. Costello, President of the Pediatric Cardiac Intensive Care Society (PCICS), moderated Session V, which was the joint session between CCAS and PCICS. He emphasized that the multidisciplinary interactions between the overlapping cross-specialties serve to accelerate academic opportunities and expand on the concept of the continuum of care with an overall goal to improve the outcomes of patients with congenital heart disease.
R. Blaine Easley, MD (Texas Children’s Hospital, Houston)
Critical Transitions
Dr. Easley was the first speaker of the session and reviewed the importance of safe transitions or hand-offs between the cardiac operating rooms and intensive care units. He highlighted the glaring absence of formal education in the science and art of critical communications in our specialty.
It is well known that the lack of critical communications is the root cause of many preventable medical errors and sentinel events. The Joint Commission, as part of their hospital accreditation process, reviews these events and will begin to inquire about mechanisms in place to ensure safe critical transitions in hospitals. To achieve this goal of effective communication, a scripted hand-off tool is of paramount importance to reduce the errors in technical and information transmission while maintaining the duration of time taken for process completion.
The successful elements of an effective hand-off tool include identification of caregivers who are leading the hand-off process and their respective roles, completion of urgent care tasks, and maintenance of patient-specific conversations only. The team receiving the patient acknowledges all patient transfer information, asks pertinent questions, and verbalizes a formalized plan of care to ensure completion of this process.
Akin to the hand-off tools, pre-procedural checklists are also valuable guides that warrant the teams acquiring patient responsibility from the intensive care unit to review patient data, plan for the safety of the intervention or procedure, and anticipate the post-intervention trajectory. The implementation of these hand-off tools and checklists reduces patient harm, reduces patient dissatisfaction, improves the quality and the content of patient-driven healthcare, promulgates a safety culture at the workplace, and overall improves patient outcomes.
In the discussion that followed, it was notable by a show of hands that only a few institutions have tools built into electronic medical records that assist in easily documentable transitions. A question was asked about how to minimize distractions during hand-offs. Dr. Easley stated that this is an ongoing struggle at all busy institutions and would require a robust team effort from all disciplines with a shared collective vision for safe transition built towards patient safety.
____________________
Mjaye L. Mazwi, MBChB, MD (The Hospital for Sick Children, Toronto)
Unplanned Reinterventions: Epidemiology and Outcomes
An unplanned reintervention is defined as the need for at least one surgical or transcatheter intervention during the same hospitalization as the index operation. Dr. Mazwi began his talk by emphasizing the impact of residual or undiagnosed anatomic lesions on postoperative outcomes after congenital heart surgery. In their institutional study spanning seven years on 943 neonates, Dr. Mazwi reported that 11% of the patients required an unplanned reintervention, which was similar to multicenter database reports.
The residual anatomic lesions are either from preventable causes such as an incomplete preoperative diagnosis, a failed therapeutic plan when an alternative existed, inadequate surgical technique performance for a correctly planned surgery, or simply, an unpreventable de novo sequelae. The neonates at risk for reintervention and poor outcomes were the smaller neonates requiring preoperative mechanical ventilation and having higher RACHS scores.
Although the number of years of experience by a cardiac surgeon (< 9 years) had a significant impact on postoperative outcomes, the case volume at a surgery center showed no effects. This is perhaps due to variability in a surgical center’s decision for early rescue reintervention that leads to improved outcomes as opposed to failure to rescue or delay reintervention.
In summary, Dr. Mazwi opined that as anesthesiologists, we should anticipate and plan for additional anesthetic management difficulties encountered when caring for high risk neonates, be able to forecast their postoperative course, and maximize our capability to contribute for early rescue reintervention when needed.
During the Q&A session, when asked about current research in pattern recognition to assist with patient trajectory, Dr. Mazwi responded that his institution is at the forefront of big data analytics to capture all postoperative data, compare it to aggregate population norms, and predict outcomes. Another question was “the data regarding unplanned reintervention is striking, the condition is not rare, somewhat predictable, and associated with poor outcomes. Now what?”
Dr. Mazwi highlighted the importance of risk stratification and pre-surgical conferences using the collective experience of an institution’s heart center for these high-risk patients, which may lend to create innovative solutions, find alternatives, and improve outcomes. When asked about the impact of the years of experience of the cardiac surgeon on unplanned reintervention, Dr. Mazwi clarified that surgical experience was associated with a lower threshold to re-investigate the details of a repair, a higher unplanned reintervention rate, but not higher mortality.
____________________
Patricia Bastero, MD (Texas Children’s Hospital, Houston)
Evolving Indication for Postcardiotomy ECMO
Dr. Bastero highlighted that 1.8 to 3% of congenital heart disease patients undergoing cardiac surgery require mechanical circulatory support immediately after cardiac surgery. ECMO is often instituted for conditions such as low cardiac output syndrome (LCOS), failure to wean from cardiopulmonary bypass, and pulmonary hypertensive crises. Over the last decade, despite the significant strides made in improving the quality of care on ECMO, the survival rates for patients needing ECMO after cardiac surgery has not improved. This statistic is perhaps due to expanding the inclusion criteria and increasing application for patients with higher case complexity, patients with low birth weight (less than 3 Kg), or for patients undergoing rescue cardiopulmonary resuscitation. Survival is low when ECMO is re-initiated for recurrent cardiorespiratory failure unless obvious anatomical correctable lesions are detected.
However, early recognition of low cardiac output states and early support with ECMO show higher survival rates when compared to ECMO utilization for rescue cardiopulmonary resuscitation (CPR) and pulmonary hypertensive crises. Dr. Bastero explained the special considerations that apply to use of ECMO in single ventricle patients.
For patients on ECMO with systemic to pulmonary artery shunts, strategies that allow adequate unloading of the single ventricles and maintaining pulmonary blood flow on ECMO (systemic to pulmonary shunt open versus clamped) favor improved survival. However, there are ongoing challenges when ECMO is used for the 2nd and 3rd stage palliated single ventricle patients secondary to difficulties in cannulation methods for complete cardiac unloading and inability to maintain cerebral perfusion pressure.
During the Q&A session, Dr. Bastero was asked that in the setting of a patient on multiple inotropes with a high lactate but stable cardiac function, which factor has the prognostic implication to proceed to postcardiotomy ECMO? Dr. Bastero responded that pattern recognition and trends in multiple clinical and laboratory factors may help in decision making rather than a single factor alone. Another question: When conventional therapy fails during weaning from bypass, does one proceed to rest on postcardiotomy ECMO for 24 hours or immediately visit the cardiac catheterization lab to evaluate residual anatomic lesions? Dr. Mazwi and Dr. Bastero concluded that it’s a difficult scenario. There are merits to both strategies and should be based on the likelihood of possible residual lesions and be on a case-by-case basis.
____________________
Session VI: Pro-Con Debates
Dr. Scott Walker, MD (Riley Children’s Hospital, Indianapolis) moderated Pro-Con debates concerning three specific clinical conundrums. Descriptions of each discussion are included below.
Title: Extubation after the Arterial Switch Operation
Debate Speakers: James DiNardo, MD(Boston Children’s Hospital, Boston) & Susan Nicolson, MD (Children’s Hospital of Philadelphia, Philadelphia)
Pro Debate: Dr. Nicolson presented the “pro” side of the early extubation debate. She started from a viewpoint that all patients should be a candidate for early extubation until “they prove themselves ineligible.” She described early extubation as extubation in the operating room or the Pediatric intensive care unit (ICU) before connecting to a ventilator. She argued that the advantages of early extubation include fewer complications from the endotracheal tube or the ventilator, lower requirements for sedative and analgesic medications, and earlier use of non-pharmacologic comfort measures. A benefit may also be shorter ICU and hospital lengths of stay and, therefore, less exposure to nosocomial infections. She reported that patients presenting for correction of dextro-Transposition of the Great Arteries are typically robust newborns born at term with few other comorbidities and who rarely require intubation before surgery. She further argued that rates of re-intubation after early and late extubation are nearly identical.
Con Debate: Dr. Jim DiNardo presented the “con” side of the debate. He suggested that although the concept of early extubation has merit, the decision to extubate in the early post-operative period requires that all team members be committed to such a strategy. This specifically includes the anesthesia team, surgeons, the ICU staff, and the floor team. In addition, the motivation for such an initiative should be explored with all stakeholders setting mutual goals for the early postoperative period. Dr. DiNardo also took up the question of cost-savings stating that though costs may be saved in the ICU, the added operating room time required to wean to extubation may cancel out any such savings. Staffing requirements of the anesthesia and ICU teams must also be considered as busy anesthesia teams may be needed to start other cases. Finally, such initiatives may be hard to maintain in the long term and practice may drift back to previous baselines over time.
Questions:
- A question was asked of Dr. Nicolson about specific anesthetic techniques to facilitate extubation. She responded that the use of lower doses of narcotics along with IV acetaminophen and dexmedetomidine helps to promote this process. Patients are typically extubated to a nasal cannula.
- Dr. DiNardo was asked if he felt that his perspective on early extubation in the operating room was different when he was acting as an ICU physician or an anesthesiologist. He reported that the goal should be the same for both practitioners and that the skill to successfully wean to extubation is an area of overlap between the two types of practitioners.
____________________
Title: Early Use of Recombinant Factor VIIa and Prothrombin Complex Concentrate (PCC) Prevents Bleeding
Debate Speakers: Nina Guzzetta, MD (Children’s Healthcare of Atlanta, Atlanta) & Jumbo Glyn Williams, MD, ChB (Stanford University School of Medicine, Stanford)
Pro Debate: Dr. Williams took on the “pro” side of the debate. Dr. Williams argued that both Factor VIIa and PCC’s help to reduce bleeding, but PCC’s may be more efficacious and safer. PCC’s may also be more efficacious and safer than fresh frozen plasma administration. PCC’s may also help in situations where bloodless surgery is the goal. At his institution, 80% of the patients who received PCC’s (specifically FEIBA) received the medication in the early post-protamine period. He admits that randomized controlled trials are required to fully elucidate the effectiveness of these interventions. He suggested that PCC’s should be administered early in patients who are at high risk of post-bypass bleeding. In patients who are at low risk for bleeding, the administration of PCC’s may be withheld until excess bleeding occurs. It may also be used as a rescue therapy for life-threatening hemorrhage.
Con Debate: Dr. Guzzetta took the “con” side of the debate. She began with a brief overview of the various forms of factor concentrates which are in existence. She reported that both blood product transfusions and excessive bleeding are associated with morbidity and mortality. Despite the potential benefits of factor concentrates, she argued that many of the in-vivo effects of these products are incompletely understood. Besides, ex-vivo data is lacking. Translational clinical data is significantly lacking. Appropriate dosing is similarly unknown. Risks are also insufficiently characterized and the possible side effect of thrombosis is not insignificant. Therefore, providers should proceed with caution in using such products.
Questions:
- Why may Factor VIIa create a higher risk for thrombosis? Dr. Williams suggested that Factor VIIa may be given at relatively high doses, while FEIBA is given at lower doses. He also indicated that thromboelastography during the bypass run would help guide the utilization of PCC’s after the bypass run.
- How is thrombosis being monitored and reported? Dr. Guzzetta stated that existing studies are retrospective and are looking at clinically relevant thromboses that were diagnosed in the ICU.
____________________
Title: Idiopathic Pulmonary Hypertension in a 10-Year-Old Child for Diagnostic Cath: General Anesthesia versus Sedation.
Debate Speakers: Chandra Ramamoorthy, MBBS, FRCA (Stanford University School of Medicine, Stanford) & Mark Twite, MD, BCh (Children’s Hospital Colorado, Denver)
Pro Debate: Dr. Twite presented the argument in favor of general anesthesia with endotracheal intubation in patients with pulmonary hypertension. He argued that precisely controlled ventilatory parameters are essential for the success of the procedure. Accurate control of inspired oxygen, inhaled Nitric Oxide, and nebulized agents require endotracheal intubation and mechanical ventilation. In addition, a sedation technique places the patient at higher risk for entering a variety of anesthetic depths. If the patient is in a light plane of sedation, needle puncture by the proceduralist may cause agitation leading to worsening pulmonary vascular resistance and desaturation. If the anesthetic is too deep, the patient may have worsening hypoxia and hypercarbia, which causes worsening the underlying pulmonary hypertension. The fact that these patients are at a higher risk for cardiac arrest in the catheterization laboratory means that adequate titration of these parameters is critical.
Con Debate: Dr. Ramamoorthy argued against intubation in this clinical situation. She explained that if the goal of catheterization is to replicate the patient’s physiology, a natural airway with the patient breathing spontaneously makes the most sense. Also, induction and maintenance of anesthesia will drop the systemic vascular resistance more than the pulmonary vascular resistance in patients with pulmonary hypertension with potentially disastrous physiologic consequences. An endotracheal tube may provide a sense of false security rather than increasing safety. The Global TOPP Registry of 908 catheterizations found 5.9% of catheterizations had adverse events with associated factors including general anesthesia and a high functional class. Finally, the practice should be tailored to the clinical environment and not all institutions may be oriented in such a way to successfully perform these cases under sedation.
Following this exciting and informative pro con debate, attendees moved on to a moderated poster session.