2015 annual meeting

Meeting Review

By Faith J. Ross, MD
Pediatric Anesthesiology Fellow
Seattle Children’s Hospital

Morning Sessions

This year the Congenital Cardiac Anesthesia Society celebrated its tenth anniversary with a fascinating and informative program that highlighted the history and the remarkable progress of congenital cardiac anesthesia over the past few decades. Program Director Nina Guzetta, MD (Children’s Healthcare of Atlanta, Atlanta) and Co-directors Scott Walker, MD (Riley Hospital for Children, Indianapolis) and Mark Twite, MD (Colorado Children’s Hospital, Aurora) put together an impressive lineup of speakers addressing a variety of topics from neuromonitoring and hemostasis to outcomes data and quality improvement in pediatric cardiac anesthesia.

Session I: Controversial Aspects of Pediatric Cardiopulmonary Bypass (CPB)

Moderator Victor C. Baum, MD (University of Virginia Children’s Hospital, Charlottesville) introduced the first session of the day, a discussion of controversial topics in pediatric cardiopulmonary bypass. Chandra Ramamoorthy, MBBS, FRCA (Lucile Packard Children’s Hospital, Stanford) opened the session by posing the question “Should near infrared spectroscopy (NIRS) be standard care for pediatric CPB?” The vast majority of the audience indicated that NIRS (cerebral oximetry) is routinely used for bypass cases at their institution and 73% felt that it should be standard care for these cases. Cerebral oximetry relies on the differential absorption of oxy- and deoxy-hemoglobin to provide a venous-weighted estimate of cerebral oxygen saturation. Commercially available devices differ in the type of light source (LED vs. laser), number of wavelengths, and proprietary algorithm. Normal cerebral oxygen saturation ranges from 60-70% on room air or 2/3 of the systemic arterial saturation, and values less than 33-44% are associated with biochemical and electrical evidence of tissue damage. Longer duration and extent of cerebral desaturation is associated with worse neurologic outcome.

There is a growing body of literature to support the utility of NIRS monitoring in pediatric CPB cases. A retrospective study by Austin et al demonstrated improved neurologic outcome in children using cerebral oximetry to optimize cannula position and hemodynamic management during CPB. There are also several studies suggesting an association between low NIRS values and postoperative MRI abnormalities.  Dr. Ramamoorthy concluded the session by drawing a comparison to pulse oximetry. Despite the lack of concrete evidence that use of pulse oximetry decreases mortality, most anesthesiologists would agree that it is a vital monitor for anesthetized patients. Similarly, NIRS monitoring may play an important role in operative care even in the absence of level 1A evidence for benefit. NIRS monitoring may also prove useful in non-cardiac cases for patients with high risk of hemodynamic compromise.

Michael P. Eaton, MD (Golisano Children’s Hospital, Rochester) continued this engaging session with a review of antifibrinolytics. Infants with congenital heart disease are particularly vulnerable to the inflammatory and hemodilutional effects of CPB and are prone to significant coagulopathy in the postoperative period. As has been demonstrated in adult patients, exaggerated fibrinolysis may play a role in this coagulopathy and may prove amenable to improvement with antifibrinolytic drugs.

Recent pediatric clinical trials indicate that all of the recently available antifibrinolytic drugs (E-aminocaproic acid (EACA), tranexamic acid (TA), and aprotinin) are superior to placebo in reducing bleeding after CPB. Studies comparing the drugs to each other have had mixed results, but TA appears to have a slight edge in efficacy. Notable side effects of antifibrinolytic medications include anaphylactic reactions to bovine-derived aprotinin and seizures with high dose TA in adults. Despite the fact that aprotinin was removed from the US market due to concerns about renal dysfunction in adult patients, the pediatric literature does not support this association in children.  Dosing for these medications is highly variable among institutions, and pharmacokinetic and pharmacodynamic studies are needed to further define rational dosing schemes. Neonates require lower plasma levels of TA and EACA to inhibit fibrinolysis, so dosing may need to be adjusted in this population. Loading doses of antifibrinolytics for pump prime are most logically dosed based on pump prime volume rather than patient weight. Dr. Eaton concluded that EACA, TA, and aprotinin all appear to be effective with some slight differences in side effect profile and drug shortages may be a determining factor for drug choice at individual institutions.

The session continued with a talk by Gina Whitney, MD (Children’s Hospital of Colorado, Aurora) addressing the optimal hematocrit (HCT) for pediatric CPB. Red blood cell transfusion is associated with reduced survival after adult coronary artery bypass surgery and prolonged ventilation after CPB in children. The risks of transfusion must be weighed against the risk of inadequate oxygen delivery, particularly in infants and children with high baseline oxygen consumption. The Boston Hematocrit Trials attempted to answer the question of optimal hematocrit in pediatric CPB by comparing infants undergoing biventricular repair managed with a goal HCT of 20% or 30%.  The group with hematocrit of 20% was found to have lower psychomotor and motor development indices than the group with HCT of 30%. A follow-up trial comparing infants randomized to HCT 25% or 35% found no difference in developmental indices. Combining the trials, there was a linear increase in psychomotor development indices up to a hematocrit of 23.5%. Thus, the optimal hematocrit for pediatric CPB (with hypothermia and pH-stat management) appears to be around 25%. Dr. Whitney cautioned that these results might not apply to normothermic or alpha-stat bypass cases or single-ventricle patients.

• Questions for Dr. Ramamoorthy:
• What is your recommendation regarding use of NIRS outside of the cardiac OR?
  Answer: NIRS can be helpful in any patient at risk of significant hemodynamic compromise. It may be particularly useful in pre-term neonates.
• Is there a NIRS value that you would consider too high? Answer: NIRS > 60-70% may indicate hypercarbia with increased cerebral blood flow. It is unclear whether there is an association with worse outcomes and elevated NIRS.
• Question for All:
• How do you manage FiO2 on CPB? Answer (Dr. Eaton): I aim for PaO2 around 200 mmHg.
  Answer: I aim for PaO2 150-200 mmHg during hypothermia and would consider dropping this further to prevent free radical damage. (The audience brought up the OXIC-1 and OXIC-2 Trials which address the issue of hyperoxia during CPB in cyanotic children.)
• Questions for Dr. Eaton:
• The doses of antibrinolytics suggested in your talk are lower than are typically used in the community. Have you changed your dosing based on these studies?
  Answer: Yes. However, it is possible that there is a dose response beyond the dose that inhibits fibrinolysis, so these medications may have other effects at higher doses that have not yet been appreciated.

Session II: Basic Science and Clinical Application

Dr. Guzetta introduced two fascinating lectures on right heart failure focusing on interventricular interactions and novel treatment strategies. Andrew Redington, MBBS, MRCP, MD, FRCP, Chief of pediatric cardiology at Cincinnati Children’s Hospital (Cincinnati), explained the physiology of RV-LV interactions in the normal and failing heart. RV and LV pressure-volume loops demonstrate a significant difference between LV and RV physiology. The right ventricle is subject to decreased stroke volume with minimal pressure change and has a limited ability to adapt to chronic stress. The RV is also four to five times more sensitive than the LV to changes in afterload.

Interventricular interactions play a strong role in biventricular function due in part to continuity of superficial myocardial fibers from the RV to the LV. Pacing of the LV alone can create robust PA flow even in the presence of an inert RV free wall; thus, by increasing LV contractility, RV contractility may be improved. This hypothesis has been tested by using aortic banding which mimics the effect of systemic vasoconstrictors on LV contractility and appears to have a beneficial effect on RV function. This somewhat surprising finding challenges the traditional role of vasodilators in RV failure. An advantageous effect on LV function may be similarly induced by placing a pulmonary artery band to increase RV contractility. This treatment was found to improve LV function in a small series of neonates with dilated cardiomyopathy. The interaction between the RV and LV is likely not solely related to myocardial restraint and preload changes as the effect persists in the absence of intact pericardium. In chronic heart failure, molecular signaling cross talk between the ventricles may play a role in fibrosis and remodeling.

Dr. Redington wrapped up his talk with a discussion of pulmonary arterial hypertension (PAH). RV ejection is prolonged in PAH, and diminished diastolic time decreases both RV and LV filling, leading to worsening RV and LV performance. Tachycardia is especially problematic as it further diminishes diastolic time and decreases cardiac output. The need for inotropes is associated with a high risk of mortality in patients with PAH and RV failure, and vasoconstrictors without tachycardia-inducing effects (for example vasopressin) may be more useful in improving coronary perfusion without impairing ventricular filling.

Jeffrey R. Fineman, MD (UCSF Benioff Children’s Hospital, San Francisco) continued with a presentation of novel therapeutic strategies for the management of RV failure. Dr. Fineman reinforced the concept that the RV and LV have different responses to physiologic stress and respond to distinct pathologic mediators of hypertrophy and failure. Many treatments for LV failure have not demonstrated utility in RV failure, but novel treatment options are beginning to show some promise for this difficult to treat condition. Important goals in management include treatment of the underlying disease process (ARDS, pulmonary embolism), optimization of preload while avoiding RV dilation, maintenance of systemic vascular resistance, avoidance of tachycardia, reduction of RV afterload, improvement of coronary perfusion pressure, and improvement in RV contractility.

To these ends, systemic vasodilators may be detrimental as they can lead to tachycardia and decreased coronary perfusion. Beta-blockers have a theoretically beneficial effect on RV remodeling and pathologic gene expression in RVH but should be used with caution to avoid decompensation in tenuous patients. Renin-angiotensin-aldosterone inhibitors have not shown utility in human trials. Histone deacetylases (HDAC) play a role in cell proliferation, inflammation, and fibrosis, and HDAC inhibitors are useful in treatment of LVH. However, HDAC inhibitors are associated with RV dysfunction and deleterious remodeling in animal models of RV failure. More selective class I HDAC inhibitors cause a modest reduction in RVH and suppress pathologic gene expression and may be more helpful. Antiproliferative agents such as Rho-kinase inhibitors, pyruvate dehydrogenase kinase inhibitors, and fatty acid oxidation inhibitors are other promising new agents of interest.

Ongoing research indicates a potential role for endothelin receptor antagonists and phosphodiesterase inhibitors in treatment of RV failure. Cardiac resynchronization and invasive surgical treatments (atrial septostomy, ventricular assist devices, and lung transplantation to name a few) can be effective in selected patients. Hopefully some of these promising new therapies will prove useful in the management of this challenging condition.

• Question for Dr. Redington:
• What is the rationale for the frequent use of milrinone after repair of Tetralogy of Fallot?
  Answer: I have abandoned routine use of milrinone for this procedure. There is better rationale for starting a beta-blocker than a beta-agonist/vasodilator. Vasoconstriction can be helpful in these patients.

Session III: Celebrating 10 Years of the Congenital Cardiac Anesthesia Society

Dr. Twite introduced two remarkably productive and influential anesthesiologists, with a combined total of greater than 70 years of practice and 150 peer-reviewed publications, who treated us to a whirlwind review of the growth of congenital cardiac anesthesia and the progress of research since the inception of the specialty in the 1940s.

Robert H. Friesen, MD (Children’s Hospital of Colorado, Aurora) presented a fascinating review of landmark articles that have been influential in the development of the field over the past several decades. (The extensive list of articles will not be reviewed here but can be viewed at www.ccasociety.org.) In 1946, Harmel and Lamont published the first paper addressing congenital cardiac anesthesia describing the management and complications in a series of patients undergoing Blalock-Taussig shunt for congenital pulmonic stenosis. In 1958, Keats reported his experience with anesthetic complications for a series of 200 patients undergoing cardiopulmonary bypass. He reported a mortality rate of 23% overall and 48% in patients less than 1 year old – dismal results by current standards. Dr. Friesen highlighted several influential studies demonstrating the hemodynamic effects of fentanyl, propofol, and ketamine that paved the way for more hemodynamically stable cardiac anesthesia. More recent areas of investigative interest include bypass-induced coagulopathy and neurodevelopmental outcomes after pediatric heart surgery.

Paul R. Hickey, MD (Boston Children’s Hospital, Boston) presented a timeline of the development of cardiac surgery and anesthesia. Cardiac anesthesia essentially began with the first PDA ligation in 1938 (performed with mask cyclopropane). With the advent of CPB in the 1950s, intracardiac repair became possible in adult patients; however, outcomes in children after CPB continued to be poor throughout the 1960s and 1970s due to the large volume of the pump and difficulty scaling down technology from the adult to pediatric population. Thus, pediatric surgeons continued to use hypothermic circulatory arrest commonly in the 1960s to avoid the morbidity associated with CPB. Outcomes improved in the late 1970s, paving the way for the successful palliation of single-ventricle lesions, beginning with the Norwood procedure in the early 1980s. Expanding surgical options have brought new ethical concerns as providers struggle with the balance between innovation and safe patient care.

Surgical and anesthetic complexity have increased in tandem, and specialized pediatric cardiac anesthesia providers have become essential for the care of these complex patients. Anesthesia training has evolved to meet the demand for highly skilled providers with the development of advanced training in pediatric cardiac anesthesia. Congenital cardiac surgery and anesthesia are dynamic and growing fields that are likely to benefit from rapidly expanding technology. Development of new medications, microbubble oxygenators, bioengineered autologous valves, and stem cell treatment for hypoplastic ventricles promise to provide exciting challenges in the coming years.

Dr. Hickey concluded by sharing the lessons that he has learned in 45 years of practicing anesthesia: Teams are better than the sum of the individuals. Cardiac surgeons and cardiologists are people too. Today’s dogma will become yesterday’s mistakes. And the privilege of meeting the challenges of these patients is secondary only to the privilege of learning from one’s colleagues.

Afternoon Sessions
Session IV: Selected Posters for Oral Presentation

Scott Walker, MD (Riley Hospital for Children, Indianapolis) began the afternoon session by introducing the three top posters chosen for oral presentation.

Sarah Franklin, MS4 (Emory University School of Medicine, Atlanta) presented research on the optimization of thrombin generation with 4-factor prothrombin complex concentrates (PCCs) Kcentra and FEIBA. Ms. Franklin and her colleagues evaluated baseline, post-CPB, and post-transfusion thrombin generation curves from 15 neonates undergoing complex cardiac repairs with cardiopulmonary bypass.  They found that ex-vivo addition of Kcentra or FEIBA to the samples increased peak thrombin generation in the post-CPB and post-transfusion samples, and FEIBA (but not Kcentra) significantly shortened the lag time for thrombin generation in the post-transfusion samples. She concluded that the addition of 4-factor PCCs to neonatal plasma has the potential to enhance thrombin generation after CPB, and PCCs may be useful in augmenting hemostasis in neonates undergoing complex cardiac procedures. Further studies are needed to confirm the clinical role of PCC in neonates.

Jake Scott, MD (Children’s Hospital of Wisconsin, Milwaukee) presented an analysis of the incidence and risk factors for acute kidney injury following Fontan procedure. He noted that AKI is apparent in 15-20% of patients after cardiopulmonary bypass and is associated with increased morbidity and mortality. Children undergoing Fontan procedure are likely at increased risk of AKI due to elevated systemic venous pressure. Dr. Scott and his colleagues found that AKI is common after Fontan completion (54%), and half of Fontan patients who develop AKI meet criteria for severe AKI. Risk factors for AKI in this population include decreased renal perfusion pressure, longer CPB time, high vasoactive medication requirement, decreased urine output, and large volume albumin administration.

The final poster, presented by Roby Sebastian, MD (Nationwide Children’s Hospital, Columbus) examined the impact of storage techniques on platelet count and function after acute normovolemic hemodilution (ANH). The protocol involved withdrawal of approximately 10-20% of the circulating volume from 23 pediatric Jehovah’s Witness patients prior to surgical incision. Platelet count and function were measured after blood collection and just prior to reinfusion. Nineteen of the 23 patients showed no change in platelet function, and the average decrease in platelet number was 13 +/- 41 x 1000/mm3. The duration and type of storage container had no impact on platelet count or function. Dr. Sebastian concluded that ANH has little effect on platelet count or function, and patients who are treated with ANH are less likely to require allogeneic blood products.

• Questions for Dr. Scott:
• Which Fontan Technique is favored at your institution and was it homogenous throughout the study?
  Answer: The vast majority of Fontan procedures during this time frame were extracardiac.
• Did any patients require renal replacement therapy (RRT)?
  Answer: Two patients required RRT but there were no permanent sequelae.
• Did you look at the difference between cerebral and renal somatic NIRS?
  Answer: We saw no widening of the cerebral-somatic saturation difference or arterial-somatic saturation difference or change in the absolute value of the renal NIRS which was surprising.
• If you were planning a prospective study, is the renal resistive index something you would use in the future?
  Answer: We have ongoing data collection with other biomarkers, we have not looked into renal resistive index specifically.
• Questions for Dr. Sebastian:
• Did you have any trouble maintaining normal vital signs in any patients during blood withdrawal?
  Answer: We occasionally gave small doses of crystalloid or small doses of epinephrine as needed.
• Did you look at the effect of ANH on coagulation factors?
  Answer: We plan to look into that in the future.
• Did all patients receive antifibrinolytics? When was blood collected relative to antifibrinolytic administration?
  Answer: Not all patients received antifibrinolytics. Hemodilution occurred prior to antifibrinolytic administration.
• Did you do ANH in both cyanotic and noncyanotic children?  What was the lowest HCT?
  Answer: Yes and we accepted a lower HCT coming off bypass for two-ventricle patients.

Session V: The Congenital Cardiac Database and Clinical Practice

Emad Mossad, MD (Texas Children’s Hospital, Houston) moderated an informative discussion of the Congenital Cardiac Database and the role it plays in clinical practice. The session began with the following very important and clinically relevant question: (audience responses noted in parentheses)

• Question: How tall is David Vener?
• 6’2’’ (15%)
• 6’4’’ (39%)
• We need more data (46%)
• The true answer, kidded Dr. Vener, is that he is shorter than he used to be as he is beaten down daily by Dr. Mossad.

Sara Pasquali, MD, MHS (Mott Children’s Hospital, Ann Arbor) began the session with an excellent discussion of data-driven approaches to quality improvement in health care. Recent changes in health care policy, particularly the development of a pay for performance reimbursement structure, designation of centers for excellence, and public quality reporting initiatives have increased the need for quality data analysis and active collaboration between centers to improve outcomes. The STS database has revealed significant variation among centers in practice patterns and patient outcomes. Dr. Pasquali presented an example of mortality and length of stay after Norwood procedure, which ranges nationally from 7% to 42% and 17.5 to 67.5 days respectively. Interestingly, the hospitals with the lowest mortality tend to have lower costs and on average treat a more complex group of patients.

By focusing on the variability in practice and its effect on outcomes, groups can identify effective practices of top-performing centers and disseminate this information nationally. As an example, the Northern New England Cardiovascular Disease Study group looked at outcomes after adult coronary artery bypass. They found that low-output cardiac failure accounted for 80% of the mortality variation among hospitals. By analyzing the data and providing feedback to decrease variability between participating sites, they were able to decrease the mortality from low-output failure from 5% to less than 2%. Data collection alone is not sufficient to improve outcomes, and active collaboration from participating centers is a key component of outcome optimization.

Data analysis has traditionally been plagued by the confounding effect of procedural case mix and heterogeneity of patient characteristics. Recent efforts to correct STS data for these factors have proved useful in providing more meaningful feedback to participating sites.  When patient characteristics and case mix were factored, 30% of hospitals changed mortality ranking by more than 10 positions and 12% changed by more than 20 positions. There are currently several databases, including STS, that are collecting data relevant to the care of patients with congenital cardiac disease. Dr. Pasquali cited as an example the ongoing Collaborative Learning Study from the Pediatric Heart Network, which has been collecting data and performing site visits to assess variation in practice and outcomes in patients undergoing repair of aortic coarctation and Tetralogy of Fallot. The investigators identified a high level of variability in timing of extubation and found that early extubation is associated with shorter length of stay. Utilization of feedback from these data sources along with inter-site collaboration has the potential to improve patient care and reduce cost.
David Vener, MD (Texas Children’s Hospital, Houston) presented an overview of the evolution and current state of the STS database and plans for future development. From January 2010 through December 2014, 115 of 122 US congenital cardiac surgery programs have participated in the STS database.  Currently 48 anesthesia programs participate, and much of the data submitted consists of surgical and demographic information only. Participation in the database provides contributing sites the opportunity to receive feedback regarding adverse events and outcomes and may provide valuable information regarding practice patterns at other institutions. The database has grown significantly over the past few years, from 20,000 cases in the spring of 2012 to almost 60,000 in the fall of 2014. Bypass cases are most heavily represented, but non-cardiac surgery and cardiac catheterization cases are increasingly reported. Upcoming changes to the database include consolidation of blood product reporting, revision of medication reporting, and the establishment of national standards of care.

• Question for Dr. Pasquali and Dr. Vener:
• Is there a trend toward better outcome in larger centers?
  Answer: Not necessarily. Answer (Dr. Pasquali): The relationship between volume and outcome tends to be more prominent in more complex lesions, but there is tremendous variation.
• Questions for Dr. Vener:
• Are you capturing any data from outside the US on cardiac mission trips?
  Answer: Because of the fees involved and the reporting mechanisms, STS is resistant to taking mission trip data. There is a move to pool data with the Society of European Thoracic Surgeons as well as Japanese and Australian groups to look at international data.
• Do you think that you are missing much of the potential data due to reporting bias? Would you get more accurate data from electronic records with automated data collection? 
  Answer: Neither Epic nor Cerner allows direct data input into the database but we are working on making that data available in the future.

 

Session VI: Focus on a Lesion: Total Anomalous Pulmonary Venous Connection (TAPVC).

Moderator Peggy McNaull, MD (University of North Carolina, Chapel Hill) kicked off the Focus on a Lesion Session with a series of audience questions about total anomalous pulmonary venous connection.

Timothy Slesnick, MD, a pediatric cardiologist and director of cardiac MRI at Children’s Healthcare of Atlanta (Atlanta), began the discussion of TAPVC with a description of the embryology and pathophysiology. TAPVC results when the common pulmonary vein fails to connect with the pulmonary venous plexus and the primitive embryologic connections to the common cardinal and umbilicovitelline systems persist. Thus all pulmonary venous blood flow returns anomalously to the systemic veins or directly to the right atrium. Four types of TAPVC – Supracardiac, Cardiac, Infracardiac, and Mixed - are classified based on the location of the pulmonary venous connection with the systemic venous circulation. An open atrial connection is absolutely essential as it provides the only supply of blood to the systemic circulation via the left atrium and left ventricle.

In supracardiac TAPVC (43-50% of cases) a vertical vein connects the pulmonary venous confluence to the superior vena cava, typically via the innominate vein. The vertical vein may run between the left pulmonary artery and the left mainstem bronchus, and 50% of cases present with obstruction of pulmonary venous return. Obstruction of pulmonary venous return results in profound instability and circulatory shock and requires expeditious surgical management. In cardiac TAPVC (18-20% of cases), pulmonary venous blood returns to the right atrium, most commonly via the coronary sinus. This type may present after the neonatal period and is less likely to be obstructed. In infracardiac TAPVC (20-27% of cases) a descending vertical vein passes inferiorly to connect with the portal vein, IVC, hepatic vein, or ductus venosus and is almost universally obstructed at birth. Patients with mixed type TAPVC (10-12% of cases) have pulmonary venous communications at more than one level.

Transthoracic echocardiography (TTE) is the mainstay of diagnostic imaging for TAPVC. Transesophageal echocardiography is rarely used as the echo probe in the midesophageal position lies directly behind the pulmonary venous confluence and may cause life-threatening pulmonary venous obstruction. All pulmonary veins should be visualized by TTE, and spectral doppler should be used to evaluate for obstruction. Loss of phasic flow and elevated mean pressure on doppler are suggestive of venous obstruction. Because only 5% of  blood flows through the pulmonary veins in-utero, prenatal echo diagnosis of TAPVC is very challenging and often missed. CT angiography and cardiac MRI may also provide valuable information particularly in cases where TTE visualization is difficult, especially mixed type TAPVC. CT angiography can be used as a template for 3-D printed models that can help clarify anatomy. Cardiac catheterization is not typically used unless intervention (e.g. stenting of a vertical vein) is planned or anatomy is not well defined by other modalities.

Daniel Velez, MD a congenital cardiac surgeon from Phoenix Children’s Hospital (Phoenix), presented his perspective on the surgical management of TAPVC. Surgical repair is typically performed with circulatory arrest in critically ill infants. Surgical goals include the elimination of anomalous connections (i.e. ligation of the vertical vein), establishment of pulmonary venous drainage into the left atrium via anastomosis of the pulmonary venous confluence to the left atrium, and closure of intra-cardiac shunts. Early postoperative complications include pulmonary edema, pulmonary hypertensive crisis, and phrenic nerve damage. Pulmonary venous obstruction, anastomotic stricture, and pulmonary vein stenosis are relatively common long-term sequelae. Patients with pre-operative acidosis, pulmonary venous obstruction, high pulmonary vascular resistance, young age, small left ventricle, major associated anomalies, or infracardiac or mixed type connection are at increased risk for morbidity and mortality. Management of mixed TAPVC is particularly challenging due to the need for two or more smaller anastomoses that are prone to obstruction.

Andrew Wolf, MB, BChir, MD, FRCA (Rush University Medical Center, Chicago) then described the major anesthetic considerations for these very challenging patients. He stressed the importance of understanding the anatomy and physiology of individual patients and the lack of evidence-based data to guide clinical practice. Timing and nature of presentation are extremely variable depending on the location of drainage and degree of obstruction. Patients with obstructed TAPVC present in extremis with shock and respiratory failure in the neonatal period. They have significant interstitial edema with lymphatic dilation and pulmonary hypertension. Nitric oxide and inotropic support may be initiated preoperatively; however, the ability to improve the condition of these patients with preoperative support is limited, and they should be taken urgently to the operating room for surgical repair. Anesthetic care should be tailored to minimize pulmonary edema and pulmonary hypertension and maintain adequate blood pressure and coronary perfusion.  Pulmonary hypertensive crisis is not uncommon in the postoperative period, even with adequate surgical repair, and nitric oxide, high inspired oxygen concentration, mild hypocapnea, and phosphodiesterase inhibition are key components of postoperative management. ICU ventilators may provide superior respiratory support in tenuous patients. Dr. Wolf also elicited discussion of some of the more controversial aspects of management, including placement of pulmonary artery pressure lines, use of ECMO, and the utility of leaving a PFO as a “pop-off.” There was a lack of consensus among the presenters regarding the advisability of a “pop-off” in these patients.

• Question for Dr. Slesnick:
• What is the role of breath-holding in cardiac MRI?
  Answer: At our institution we do not typically employ breath-holding.
• Questions for Dr. Velez:
• Do you leave a PFO or other connection as a “pop-off?”
  Answer: I always ligate the connecting vein and I never leave a PFO given the potential for left to right shunting.
• Are you using 3D printed models in your practice right now?
  Answer: I have been using this for complex lesions.
• What have you found to be the predictors for re-stenosis?
  Answer: Heterotaxy appears to be associated with re-stenosis. I also try to avoid surgical manipulation or clamping of veins whenever possible.
• Questions for Dr. Wolf:
• What is the best ventilation strategy for these patients and how far would you push the ventilation to normalize the ABG?
  Answer: You have to recognize that the lungs are filled with pulmonary edema and limiting the time the lungs are exposed to this prior to CPB may help in minimizing damage. The oscillator may have a role for post-op lung rest.
• Do you use TEE for these patients? Answer: We typically do not use TEE. We use epicardial echo if needed, but it may not produce quality images of the back of the heart.
  Answer: Agree. There are significant risks of TEE and it typically does not add a lot of information to TTE.
• Questions for All:
• For patients who present in extremis, are there any you would not feel comfortable taking to the OR?
  Answer (Dr. Slesnick): We recently had two patients with pulmonary vein atresia who were not operative candidates.
  Answer (Dr. Velez): I would consider pre-op ECMO in some cases.
• How does mitral valve size impact outcome?
  Answer (Dr. Slesnick): We consider aortic and mitral valve sizes, but typically the left ventricle is just compressed by the right ventricle and the mitral valve is actually normal in size.
  Answer (Dr. Velez): Agreed. If the valve truly is small it does contribute to poor outcome.

The meeting concluded with a review of posters and a reception celebrating the 10th anniversary of the CCAS. (Posters are available for viewing at www.ccasociety.org.) Overall, this informative and fascinating meeting demonstrated remarkable development in the field of congenital cardiac anesthesia. I look forward to learning more about the exciting new developments in this growing field at the next annual meeting in Colorado.