Update on Pediatric Cardiac Anesthesia

George M. Hoffman, MD, pediatric anesthesiologist

Anesthesia for children undergoing congenital heart defect repair involves all aspects of medical (non-surgical) care during the perioperative period. In this article, anesthesia refers to the application of medications, devices and environmental manipulations to produce optimum perioperative outcomes.

Appropriate anesthesia care accounts for the complex perioperative stressors (surgical trauma, acute vascular volume shifts, wide temperature changes, cardiopulmonary bypass, variable periods of partial or total ischemia), the predictable responses to those events, modification of those responses by different anesthetic techniques and multiple drug-drug technique interactions.

A variety of pharmacologic regimens can produce the quadrad of analgesia, amnesia (and prevention of avoidance conditioning), autonomic stability and muscle relaxation that characterize the anesthetic state.

• Potent anesthetic vapors (halothane, isoflurane, sevoflurane) uniquely preserve spontaneous respiration while producing dose-dependent effects on the anesthetic quadrad. Because uptake and elimination occur through the lungs, which normally receive all of the cardiac output, predictable pharmacokinetics result. These drugs also depress cardiac contractility (especially halothane) and alter brain stem baroreflex responses, which can be desirable or undesirable in different clinical conditions.
• Opioid analgesics reduce autonomic and behavioral arousal to noxious stimulation and are commonly used to supplement vapor techniques. In high doses, potent synthetic opioids (fentanyl and analogues) can induce profound hyporeflexia of the autonomic nervous system, with characteristic vital sign stability and preservation of intrinsic myocardial contractility and baroreflex function. This hemodynamic profile made high-dose 'pure' narcotic techniques popular. Their application has been widely tested and published.

Opiate-based anesthesia trials demonstrated that surgical stress causes metabolic derangements in neonates; derangements are proportional to the severity of the physiologic trespass at surgery; fentanyl added to nitrous oxide ameliorates these responses in infants undergoing PDA ligation with a salutary effect on perioperative outcome; and high-dose fentanyl (>100 mcg/kg for operation with a perioperative infusion of 10 mcg/kg/hr, or equivalent) results in postoperative hemodynamic stability, better organ function and less late mortality in neonates undergoing repairs on CPB.

At least 50 human and animal trials have demonstrated the linkage between autonomic responses, splanchnic ischemia, translocation of gut bacteria, systemic inflammatory response syndrome, multi-system organ dysfunction and ICU mortality. Because of these hemodynamic relationships, high-dose potent-opioid-based anesthesia remains the standard for high-risk surgery, despite disadvantages related to pharmacokinetics, development of tolerance, impaired gut motility and respiratory depression.

Opioids alone, however, do not reliably induce the anesthetic quadrad. Therefore, we almost always supplement even high-dose opioid techniques with vapors or other sedative-hypnotic drugs (benzodiazepines, barbiturates or propofol) to induce a balanced anesthetic state while maintaining the hemodynamic advantages.

For patients at lower risk of vital organ compromise (older children or those having a less complex repair) the hemodynamic advantages of opioid techniques may be offset by the disadvantages of required postoperative mechanical ventilation with high-dose opioids resulting in sedation-respiratory depression entanglement. Over the last decade, a variety of techniques have evolved to facilitate early tracheal extubation, most notably the application of epidural anesthesia and analgesia to the low-risk patient population. Rosen and Rosen (then at Ann Arbor) in 1989 published a study of patients managed by single-shot epidural morphine analgesia at the conclusion of a bypass run. They found adequate criteria for tracheal extubation, lower pain scores and no significant problems, compared to a vapor-parenteral opioid control group.

In an adult series, epidural analgesia with a morphine infusion produced more reliable extubation criteria and more autonomic stability (lower levels of cortisol, beta-endorphin, and glucose) than did parenteral morphine. Shayevitz, Gutsein and colleagues at Ann Arbor recently compared lumbar epidural morphine infusions to parenteral fentanyl infusions in children (not neonates) undergoing cardiac repairs and found epidural morphine facilitated earlier extubation and shorter ICU stay without significant serious side effects. Their study is especially relevant because the surgical and anesthetic techniques most closely resemble those used here at Children's Hospital.

We have adopted anesthetic techniques to facilitate early extubation in lower-risk patients, with preliminary data showing more reliable early extubation with epidural techniques. Because of complex interactions between anesthetic state, site of surgery, lung volume maintenance, effects of spontaneous ventilation on metabolic rate and oxygen transport demands, appropriate patient selection is critical to avoid hemodynamic or respiratory failure in higher-risk patients.

The more widespread use of epidural techniques may be limited by the recent warning from the FDA regarding the risk of epidural hematoma when neuraxial puncture is performed in the presence of low-molecular-weight heparin. LMWH has more effects on factor X than factor II, compared to unfractionated heparin, resulting in significant antithrombotic activity without effect on the PTT, and thus dose-response data for this side effect are missing. The relative risk of neuraxial block in patients receiving standard heparin for CPB anticoagulation is unknown, but presumably higher. It is our practice to perform epidural puncture prior to anticoagulation, but heparin on CPB may increase the risk of bleeding around an indwelling epidural catheter. The issue needs further examination. Typically, we employ "single-shot" morphine in the 50 mcg/kg range without catheter insertion, to afford 12 hours of perioperative analgesia, facilitate tracheal extubation and avoid hemorrhagic complications.

Epidural anesthesia is a special case of the general principle of selective drug delivery. Opioid effects on nociceptive signal transduction occur widely throughout central gray matter, including the spinal cord. Epidural administration maximizes spinal effects of analgesia while minimizing brain stem effects of sedation and respiratory depression. Various other innovations can help maximize drug effect and minimize side effect, but clinical data are largely missing. Drug combinations can result in a more desirable side-effect profile when those combinations have additive or potential effects on a desired outcome but not on an undesirable outcome; this principle is applied in the various forms of balanced anesthetic techniques.

New drug development is intended to produce the "magic bullet" resulting in all desirable aspects of the anesthetic state without complications. As an example, use of remifentanyl, an ultra-short-acting opioid metabolized by plasma pseudocholinesterase (like esmolol and succinylcholine), presumably permits the advantages of deep narcotic anesthesia to coexist with the pharmacokinetic advantages of vapors, but acute tolerance and autonomic activation on withdrawal may be problematic without the co-administration of more standard opioids. Modification of host responses by non-anesthetic drugs is becoming increasingly important, the almost universal use of vasoactive drugs being the most common example. We have shown a beneficial effect of phenoxybenzamine on hemodynamics after single-ventricle palliation of HLHS when co-administered with a high-dose opioid balanced anesthetic. Theoretically, such specific targeting of the vasoconstrictor side of the stress response might obviate the need for high-dose opioid anesthesia to control autonomic responses, with the potential for decreased opioid side effects, but this relationship has not been proven.

Specific protection of the splanchnic circulation with dopexamine may have similar advantages. Modification of the inflammatory response to bypass and surgery may also change the time course of hemodynamic risk, with subsequent push for alteration in anesthetic technique. However, currently anesthetic techniques have such low risk (when appropriate monitoring, personnel, and postoperative care are provided) that further innovation will, by necessity, need to closely examine multiple outcomes, early and late, for intended and unintended consequences.

Double Outlet Right Ventricle

Peter Frommelt, MD, pediatric cardiologist

Double outlet right ventricle (DORV) is a rare congenital cardiac defect in which all of one great artery and at least half of the other great artery arise from the right ventricle.

This defect is found only in about 1.5 percent of all patients with congenital heart disease and frequently is associated with other congenital cardiac abnormalities. For example, since both great arteries arise from the right ventricle, a ventricular septal defect (VSD) almost always is present to allow egress of blood flow from the left ventricle (otherwise it would have no outlet to empty). The position of the great arteries relative to the VSD determines the physiology of each patient. The clinical presentation and surgical options can vary dramatically depending on the relationship of the great arteries with the VSD and whether there is obstruction to either great artery.

Anatomy and physiology
The three most common anatomic subtypes of double outlet right ventricle are:

1. The VSD sits below the aortic valve without pulmonary stenosis.
2. The VSD sits below the pulmonary artery without pulmonary stenosis.
3. The VSD sits below either great artery in association with pulmonary stenosis.

Subaortic VSD without pulmonary stenosis
In this situation, oxygenated blood from the left ventricle is directed appropriately into the aorta, so there is no systemic desaturation. Since pulmonary stenosis is not present, the majority of right ventricular output, as well as some of the left ventricular output that passes across the VSD, will shunt into the pulmonary arteries, similar to the patient with a large VSD. Because of this left to right shunt, these patients typically present with congestive heart failure, failure to thrive and normal oxygen saturations.

Subpulmonary VSD without pulmonary stenosis
In this anatomical subset, called the Taussig-Bing form of double outlet right ventricle, oxygenated blood from the left ventricle is directed back into the pulmonary artery, while deoxygenated blood that returns to the right ventricle is directed into the aorta. This physiology is identical to patients with complete transposition of the great arteries. Cyanosis generally is the first clinical sign of heart disease and usually is recognized early in the neonatal period. Some patients with a subpulmonary VSD also develop obstructive subaortic muscle and aortic outflow tract obstruction in combination with coarctation of the aorta. They can present with a shock-like state and poor lower extremity pulse and perfusion.

Subaortic VSD with pulmonary stenosis
In patients with DORV, an atrial septal defect occurs 25 percent of the time. Additional muscular ventricular septal defects and a patent ductus arteriosus occasionally are found. Mitral valve abnormalities, including mitral valve straddle, parachute mitral valve and cleft mitral valve also are seen. In addition, coronary artery anomalies frequently are associated with DORV. If there is pulmonary stenosis, the left anterior descending coronary artery arising from the right main coronary artery and crossing anteriorly across the outflow tract occasionally occurs. If the aorta is in an anterior position relative to the pulmonary artery, coronary anomalies similar to the ones found in D-transposition of the great arteries are common.

DORV generally occurs as an isolated entity, but it also is seen in association with other complex defects such as total anomalous pulmonary venous return, atrioventricular septal defect, mitral stenosis or atresia, juxtaposition of the right atrial appendage and a persistent left superior cava. These complex patients often also have situs anomalies, particularly heterotaxy syndromes (asplenia and polysplenia).

Surgical treatment
Subaortic VSD without pulmonary stenosis
Since their defect acts like a large VSD with significant congestive heart failure and puts these patients at risk for early pulmonary vascular obstructive disease, surgery generally is performed by age 1. This is the simplest DORV to repair, as the VSD already is situated in close proximity to the aortic valve. The VSD is closed to include the aortic valve as part of the left ventricle, creating a tunnel that excludes the right ventricle from the systemic circulation. Since this intraventricular tunnel is relatively simple to create, it is rare to find postoperative tunnel obstruction. Long-term prognosis is excellent, and no further surgery generally is required.

Subpulmonary VSD
Two surgical approaches are possible for these patients. If there is no pulmonary or aortic stenosis, the simplest repair is closure of the VSD, so the left (closer) ventricle ejects into the pulmonary artery, combined with an arterial switch repair. This operation generally is performed during the first one to two months of life, depending on the degree of cyanosis. This procedure carries the same risks as an arterial switch repair for transposition of the great arteries, although the positioning of the great arteries (generally side by side rather than anterior-posterior) can result in distortion of the branch pulmonary arteries after the arterial switch, with late development of branch pulmonary artery stenosis.

Figure 1 - DORV with side-by-side great arteries and pulmonary stenosis.

If the pulmonary outflow tract is obstructed (figure 1), the arterial switch repair is not an option. Under these circumstances, a Rastelli repair is performed, with creation of an intraventricular tunnel to baffle left ventricular blood to the aorta and placement of a right ventricular-to-pulmonary artery conduit (figure 2). Because the aorta sits farther away from the VSD in this anatomic subset, the creation of the tunnel is more complex and the incidence of tunnel obstruction or leakage is significantly higher than when the VSD is closely related to the aortic valve. A valved homograft generally is used to connect the right ventricle and pulmonary artery, so this procedure is often delayed until the child is at least 1. This step allows for placement of a larger homograft, by delaying homograft replacement until the child is school-age. If the patient has excessive cyanosis during the first year of life, a palliative Blalock-Taussig shunt is placed to augment pulmonary blood flow.

Figure 2 - DORV S/P Rastelli

Subaortic VSD with pulmonary stenosis
These patients are physiologically and anatomically similar to those with tetralogy of Fallot. VSD closure, to allow left ventricular ejection only into the aorta in association with patch augmentation of the right ventricular outflow tract to relieve the subvalvar and valvar pulmonary stenosis generally is performed at 6 to 12 months of age. Excessive cyanosis during early infancy again can be palliated by a Blalock-Taussig shunt prior to the definitive surgical operation. As with tetralogy of Fallot repair, the most common postoperative complication is poor right ventricular systolic and diastolic function, with low cardiac output and systemic venous congestion. Rarely, residual right ventricular outflow tract or branch pulmonary artery obstruction can complicate the postoperative course. More commonly, these patients have significant pulmonary insufficiency related to patch augmentation of the right ventricular outflow tract, and many require placement of a right ventricular-to-pulmonary artery homograft as older children or young adults when right ventricular dysfunction develops secondary to the chronic right ventricular volume overload.

Complex double outlet right ventricle
DORV with complex associated defects (atrioventricular valve anomalies, anomalous venous drainage, ventricular hypoplasia) frequently must be palliated as a functional single ventricle. If no pulmonary stenosis exists in these complex cases, pulmonary artery banding is performed early in life to control congestive heart failure and protect the pulmonary vascular bed. If pulmonary stenosis exists, then a Blalock-Taussig shunt is placed to augment pulmonary blood flow. As within any single ventricle patient, a bidirectional Glenn anastomosis is created between 6 and 12 months of age, and complete Fontan palliation usually is performed between ages 2 and 4.

Transition Project:
Improving the Transition from Intensive Care to Intermediate Care

Kathy Mussatto, RN, clinical research nurse

Transfer anxiety is a state in which an individual experiences physiological and/or psychological disturbances as a result of transfer from one environment to another (Leith, 1998).

Potential reasons for transfer anxiety include:

• Little or no preparation for the transfer.
• Sudden reduction in monitoring of the patient at the time of transfer.
• Loss of security from lack of monitors.
• Lack of explanation of differences between the ICU and the general hospital unit.
• Little or no time between when the patient is notified of transfer and the actual move.
• Change in care routines and care givers.
• Lack of opportunity for closure with ICU staff.

Signs of transfer anxiety include demonstrations of dependency, insecurity, lack of trust, need for excessive reassurance, unfavorable comparison of staff, vigilance, withdrawal, restlessness, feelings of abandonment, loneliness, disappointment or anger (Carpenito, 1995).

Until the Transition Project, although a great deal of time and effort was invested in preparing patients and their families prior to surgery and their ICU stay, little information was provided regarding their transition to intermediate care.

A core group, including Child Life specialists, ICU and IICU nurses and social workers, met in October to identify the initial goal of the project - 80 percent of cardiovascular patients and their families cared for in the PICU for more than three days will verify they know what to expect following a transfer to intermediate care. To measure this goal, families will be surveyed about their experiences and asked what would have helped them to better prepare for their child's transfer.

The Transition Project team also has identified several potential interventions to smooth transfers. These include early initiation of discharge planning, the ICU presented to families as a temporary location, a brochure for parents which addresses transfer issues, tours of the IICU, introduction of staff to families prior to transfer, and reinforcement that the transfer is a positive step in the child's recovery. These ideas will be implemented in the next few months.