A medical strategy to address persistent chest tube drainage after the Fontan operation

Joseph Cava, MD, pediatric cardiologist, Children's Hospital of Wisconsin; assistant professor, Pediatrics, Medical College of Wisconsin.

The Fontan operation was devised to surgically correct single ventricles, specifically tricuspid atresia. Dr. Fontan first described the procedure in 1971 and initially proposed using the right atrium as a pumping chamber. Initial surgeries consisted of a classic Glenn to the right lung and a valved conduit from the right atrium to the pulmonary artery to supply the left lung with blood.

On the theory that the right atrium would make a poor pump, Dr. Kreitzer modified the original design and directly anastomosed the right atrium to the main pulmonary artery. This allowed passive blood flow from the right atrium directly to both branch pulmonary arteries. The basic concept of the Fontan procedure, to bypass the ventricle, has remained the same, but the procedure itself has undergone several modifications. The procedure now is used for many types of congenital heart disease that have single-ventricle physiology.

Despite the success of the Fontan procedure, there continues to be significant morbidity for these patients. Newer modifications have evolved to help control complications from the Fontan, such as arrhythmias, protein-losing enteropathy, atrio-venous malformations and recurrent pleural effusions. The most recent operations employ either the lateral tunnel or extracardiac conduit.

The lateral tunnel uses a Gore-Tex patch sewn into the right atrium that baffles IVC blood directly to the pulmonary arteries. Part of this tunnel consists of right atrial wall. Therefore, it has the capacity to grow with the patient. However, construction of the lateral tunnel involves fairly extensive manipulation and suturing of the atrium, increasing the potential for complications.

The extracardiac conduit circumvents these potential problems because the atrium is bypassed. This conduit supplies IVC blood directly to the pulmonary arteries. The extracardiac conduit, however, does not have potential for growth.

Patients who undergo the Fontan procedure have elevated central venous pressures (CVP), which is to be expected as a result of their new physiology. A significant number of patients have an elevated systemic venous pressure that is not well tolerated shortly after the procedure. In order to address this potential problem, the fenestrated Fontan was introduced. The fenestration allows right to left shunting of blood at the atrial level to decrease central venous pressure at the expense of arterial oxygen saturation. With these new techniques and modifications, the Fontan operation is very successful with extremely low mortality.

Unfortunately, morbidity related to persistent chest tube drainage and prolonged hospital stay continues to be a problem. Traditionally, persistent chest tube drainage was treated with a low fat diet and, in more difficult cases, NPO status with intravenous hyperalimentation. Failure of diet therapy resulted in the need for pleural sclerosis. These treatments required the use of central venous catheters and their potential complications, additional surgical procedures and anesthesia.

Here at Children's Hospital of Wisconsin, we have developed a medical strategy to address persistent chest tube output. Initial results are encouraging and a standardized protocol now is used. The medical strategy addresses the elevated central venous pressure, which is an important issue in post-Fontan patients. Despite the use of a fenestration, it is felt the increase in CVP significantly contributes to pleural effusions and drainage.

There are three main components to the medical strategy. The first, and most important, component is fluid management, accomplished mainly through the use of diuretic therapy. Diuretics typically are started when the patient is hemodynamically stable on the first postoperative day. Diuretic therapy is aggressive with special attention to monitoring electrolytes and renal function. The patient also is maintained on a total fluid restriction of 80 percent maintenance until discharge. The second component is afterload reduction. An ACE inhibitor is typically used. These medications directly inhibit vasoconstriction by blocking the formation of the vasoconstrictor angiotensin II. Secondly, aldosterone release is inhibited preventing excessive salt and water retention. Finally, ACE inhibitors indirectly increase the production of nitric oxide and prostacyclin, potent pulmonary vasodilators, by blocking bradykinin inactivation. The third component is oxygen, which is a pulmonary vasodilator. Theoretically, oxygen therapy should decrease pulmonary vascular resistance and allow increased flow through the Fontan baffle into the lungs. Supplemental oxygen is used until the chest tubes are removed. All Fontan patients are kept on a low-fat diet, which likely is most helpful if the chest tube drainage is chylous.

A total of 25 patients have been placed on the protocol since August 2000. They were compared to a historical control group of patients consisting of 35 patients treated from January 1997 to July 2000. The average length of stay for the protocol group was significantly reduced compared to the control. In addition, none of the protocol patients needed to be made NPO and have hyperalimentation started, or required pleural sclerosis. In contrast, 10 control group patients were placed on hyperalimentation, with nine eventually undergoing pleural sclerosis. Thus far, the patients have tolerated the new protocol very well. Although patient care, at times, needs to be highly individualized, the relatively standardized care using the post-Fontan protocol initially has proved to be successful in reducing morbidity from persistent chest tube drainage.

Growth velocity of infants with hypoplastic left heart syndrome: A comparison of enteral feeding strategies

Nancy Rudd, CPNP, Children's Hospital of Wisconsin

Background
Failure to thrive is a well-documented finding in infants with hemodynamically significant heart disease. Although the relationship between growth failure and abnormal hemodynamics is not clearly understood, there are many likely contributing factors.

Multifactorial causes include inadequate caloric intake, increased metabolic rate, malabsorption and severity of hemodynamic impairment. Infants with hypoplastic left heart syndrome (HLHS) who undergo stage I Norwood palliation have additional factors affecting their nutritional status, such as hypoxemia or poor cardiac output. There also can be postoperative complications such as prolonged intubation or infection.

Over the past several years, some pediatric cardiac centers have attempted to address the problem of failure to thrive by electively placing a gastrostomy tube (GT) in all Norwood patients prior to discharge from the hospital. Other institutions have utilized nasogastric tubes (NG) for supplemental feeds. The practice of NG supplementation in the Norwood population no longer is utilized at Children's Hospital of Wisconsin. There is potential risk for dislodgement, formula aspiration or vagal response with NG placement. At Children's Hospital, if feeding tube supplementation is needed, a GT is placed.

Our cardiac nurses and physicians have struggled with the nutritional management of the Norwood population. There often were inconsistencies in the nutritional management strategies utilized for this population of infants with significant congenital heart disease. It was decided that a systematic review of our nutritional management practices would help optimize nutritional interventions for this high-risk group.

Purpose
The purpose of this study was to define the postoperative nutritional support patterns and to determine their impact on growth velocity in infants with HLHS from birth to bidirectional Glenn.

Methods
A retrospective chart review of 50 consecutive patients undergoing the Norwood procedure and surviving to bi-directional Glenn palliation was done. Patients were born between March 1994 and December 2000. Patients not surviving to bi-directional Glenn were not included in the analysis. Nutritional management strategies and growth measurements from birth to bidirectional Glenn were recorded. Based on feeding method at time of discharge s/p Norwood, patients were classified into the following groups.

• Oral fed (OF) , either by bottle or breast.
• Combination oral and tube fed (OTF) via NG (before the change in protocol) or GT.
• Tube fed (TF) via NG or GT.

Groups were compared on clinical and demographic variables. Growth velocities, assessed as percent change in weight relative to birth weight, were compared by linear regression.

Patient Population
Diagnostic categories for the 50 infants revealed 70 percent had aortic atresia (AA) with either mitral atresia (MA) or mitral stenosis (MS), and 30 percent had aortic stenosis (AS) with either MA or MS. At the time of discharge from the Norwood palliation, 28/50 (56 percent) were entirely oral fed, 15/50 (30 percent) were combination oral/tube fed, and 7/50 (14 percent) were entirely tube fed (Figure 1).

Results
At birth, the three groups did not vary significantly in weight, height, gestational age, sex or diagnostic category. At time of their Norwood procedures, groups again did not vary in respect to age, BSA, indexed shunt size, cardiopulmonary bypass time, circulatory arrest time, or number of days to first oral feeding attempt. And, at time of discharge, caloric density of formula (20-30 cal/ounce, mean 24) did not vary significantly between groups. Most noteworthy was the lack of relationship between indexed shunt size and method of feeding since it has long been hypothesized that infants with larger shunts have more pulmonary blood flow resulting in tachypnea, which hinders the infant's ability to feed orally.

Several significant differences between the three groups were noted. Hospital length of stay, (Figure 2), was longer for the TF group (27±10 days) than either the OTF group (39±12 days) or the TF group (78±44 days. Nutritional intake, reported as kcal/kg/day, at discharge s/p Norwood procedure (Figure 3) also varied significanltly among groups (OF 104±18, OTF 120±15, TF 112±20). Age at bidirectional Glenn (Figure 4) was significantly different as well with the tube fed group being the oldest (OF 4.9±1.5 months, OTF 6.3±2.1 months, TF 7.6±2.7 months). The final significant difference was growth velocity from birth to bidirectional Glenn (Figure 5). The oral fed group averaged the greatest grams per day weight gain (OF 16±4.2, OTF 13±5.3, TF 10±4.4, [ANOVA p<0.05]).

When growth velocities of each group were compared, the oral fed group demonstrated the greatest and most consistent weight change over time from Norwood discharge to bidirectional Glenn (Figure 6).

We concluded the ability to achieve full oral feeds is related to shorter lengths of stay and greater growth velocity in infants with HLHS following the Norwood procedure. The ability to feed orally also appears to be an important indicator of wellness in this population as these patients grew best (Figure 5) despite receiving the lowest caloric intake at time of discharge (Figure 3). Even with nutritional strategies such as supplemental tube feedings and high calorie formulas, optimal weight gain cannot be guaranteed in the Norwood population. Finally, we speculate that the need for supplemental tube feedings after the Norwood procedure identifies a population who may benefit from earlier bidirectional Glenn intervention rather than prolonged tube supplementation because of suboptimal weight gain in spite of aggressive caloric supplemention.

Clinical Implications
The nutritional management of infants with hemodynamically significant heart disease has been positively influenced as the result of this retrospective study. Nutritional issues are identified and addressed throughout the postoperative course.

Great attention is given to optimizing both the caloric density and volume of feeds prior to discharge from the hospital. A caloric intake of greater than 120kcal/kg/day is typically required to see adequate and consistent growth. Whereas healthy infants typically gain 30 grams per day or 6 to 8 ounces per week, our patients with severe congenital heart disease have a goal of 15-20 grams per day weight increase. To see this velocity of growth, often 24-, 27-, or 30-calorie-per-ounce formulas or fortified breast milk are needed.

Unless an anatomical contraindication to oral feeding exists, oral feedings always are the goal for discharge. Nasogastric supplementation is frequently utilized during the early postoperative period, but rarely continued at home as mentioned earlier. While the NG tube is in place, the patient is encouraged to orally feed first and then any remaining volume is gavaged. When a patient has shown the ability to consume greater than half of his or her daily enteral requirement by mouth, the NG tube is removed for a by- mouth only trial. Emphasis is put on recognizing the infant's hunger signs and feeding on demand. Parents and grandparents are encouraged to "room in" if possible, providing the infant with consistent feeding techniques and a caregiver who always is available to offer the feed in a timely manner. If an infant is not able to consume the necessary volume to ensure adequate hydration and positive weight gain, GT placement is indicated. However, even with a GT in place, the infant continues to be offered oral feeds first. Only volumes not taken by mouth are given via GT.

With a well-organized discharge feeding plan in place, consistent weight gain still is not ensured. Surveillance of growth parameters and ongoing adjustments to the feeding plan are necessary. Input from parents, primary care physicians, and cardiac nurses and physicians is needed. With this coordinated effort, the outcome of a nutritionally stable infant with adequate growth, despite their hemodynamically significant heart disease, can be achieved.

A primary pediatric care provider perspective on comprehensive cardiac care for children

Michael Gutzeit, MD, primary care pediatrician, Children's Medical Group, member of Children's Health System

"Doctor, my obstetrician just told me that my fetal ultrasound showed an abnormal four-chamber view. What do I do now?"

This scenario, or one similar to it, may very well present itself to you in the near future, if it hasn't already. With the advent of new and better fetal echocardiography techniques and the routine use of fetal ultrasonography in the general obstetrician's office, the prenatal diagnosis of congenital heart disease is much more common today than in the past. At the present time, the elective evaluation of the fetus via echocardiography optimally occurs at 16-22 weeks' gestation. Evaluation can be initiated by any number of individuals: primary care physicians, perinatologists, obstetricians, or even families with previously affected children.

The evaluation of the newborn infant with a murmur sometimes may present a dilemma depending on the place of birth and the availability of specialists and appropriate equipment. The infant in distress may need urgent transfer to an appropriate facility for comprehensive evaluation and support, both medically and surgically. Less urgent evaluations also may occur at a facility outside of Children's Hospital of Wisconsin through consultation with physicians at the Heart Center.

The ability to fax ECGs and, in some facilities, to provide tele-echocardiography, can help to clarify the diagnosis for the primary care physician with questions regarding the ongoing care of an infant in the nursery. In some circumstances, urgent privileges may be granted to the pediatric cardiologist for isolated cases if there is no staff member capable of providing the necessary care.

One of the most important components of primary care, and often the one that is most daunting, is the care of infants and children once they have been discharged following a cardiac procedure. Recent standardization of discharge forms from the cardiac service have made this task easier and also provide useful information when coordinating care. An excellent resource (for those of us who have not performed cardiac surgery recently) is available from the Heart Center in CD form.

In most cases, primary care visits, and especially immunizations, may proceed on schedule. Some adjustments may be necessary in rare circumstances due to planned multi-stage surgical procedures.

One concern of all involved in the care of children with surgically-corrected congenital heart disease is the risk of developmental problems as the child grows. This may vary in scope and is multi-factorial in origin. These children need careful developmental monitoring and early intervention to address these concerns, if warranted. Currently, there is no comprehensive clinic that specifically follows developmental parameters, so it is incumbent upon the primary care physician to keep this in mind.

Pediatricians also can help prevent some serious cardiac episodes. It has been shown that the use of automated external defibrillators (AEDs) may have a significant impact on surviving an episode of sudden cardiac arrest. Recent cases of sudden cardiac death of young athletes have focused more attention on this potentially avoidable tragedy. Pediatricians can ask about the availability of AEDs in their local high schools. If they are not available, we can become active advocates for their placement in the schools. It also may be possible to advocate for basic CPR instruction at the high-school level.

Another issue impacting pediatricians is, as children become adults and the long-term survival of children with congenital heart disease improves, who will care for these individuals? Certainly most adult cardiologists were not trained in this capacity and the pediatric cardiologist and pediatrician may feel uncomfortable providing care to young adults in their 20s and 30s. Recent comprehensive clinics on the West Coast have begun to develop a medical model in these patients as they pass the usual limits of pediatric care, but in the meantime, communication by Pediatric and Internal Medicine disciplines will be important to coordinate ongoing care.

Is rectal acetaminophen (Tylenol®) an effective postoperative analgesic in patients after cardiac surgery?

Eckehard A.E. Stuth, MD, director of Pediatric Cardiac Anesthesia, Children's Hospital of Wisconsin; associate professor, Anesthesiology, Medical College of Wisconsin.

Rectal acetaminophen frequently is given prophylactically for postoperative pain management in infants and small children in order to reduce postoperative opioid requirements.

In recent years, the pharmacokinetics (what the body does with the drug) and pharmacodynamics (what the drug does to the body) of rectal acetaminophen have been revisited in pediatric anesthesia and pain literature. It was discovered that doses of 10 to 30 mg/kg acetaminophen yielded only subtherapeutic peak plasma concentrations (effective concentrations often < 10 µg/ml). Not surprisingly, the majority of recent well-controlled postoperative pediatric studies using 15-35 mg/kg rectal acetaminophen did not demonstrate any significant analgesic or opioid-sparing effects. The rare exceptions were operations that appeared to cause very little postoperative pain, such as inguinal herniorrhaphies.

A recent (2001) well-controlled, double-blind, prospective study using 40 mg/kg rectal acetaminophen given to infants for cleft palate repair at the beginning of surgery did not show any postoperative opioid-sparing effects and two hours post application maximum plasma concentrations of 20 µg/ml were considered subtherapeutic for analgesia. Results were equivocal for adenotonsillectomies. Only about 50 percent of patients receiving 40 mg/kg oral acetaminophen were considered to have acceptable analgesia at acetaminophen concentrations of 17 µg/ml. The typical antipyretic therapeutic plasma concentration range for acetaminophen is thought to be 10-20 µg/ml, while adequate analgesia for operations causing moderate pain seems to require higher concentrations (20-25 µg/ml). These concentrations cannot be achieved with less than 40 mg/kg acetaminophen. In fact, a dose of 45 mg/kg rectal acetaminophen achieved a mean peak plasma concentration of less than 15 µg/ml, which was not reached until more than three hours post application. The literature clearly shows that the absorption of rectal acetaminophen is quite unpredictable, the time to peak plasma concentrations is in the range of many hours (two to four hours) and the bioavailability is much lower than for oral acetaminophen (about 30 to 50 percent of an oral dose at best).

Pharmacokinetic dynamic simulations suggest that only very large doses of rectal acetaminophen would reliably lead to therapeutic plasma concentrations for adequate analgesia (>20 µg/ml) in children with mild to moderately severe pain. This would require a loading dose of 70 mg/kg and maintenance of 50 mg/kg every eight hours.

However, it has not been shown that such a high- dose regimen would be safe. The maximum daily cumulative recommended dose is 90 mg/kg, since hepatotoxic effects may be possible at doses above 150 mg/kg/day.

In conclusion, the minimum effective rectal loading dose of acetaminophen for prophylactic antipyresis (fever control) is 40 mg/kg given at least two to three hours before the anticipated need. Repeat doses should be 20-25 mg/kg every eight hours. Thus the recommended daily limit of 90 mg/kg will not be exceeded. However, there is no convincing evidence that these doses will provide significant analgesia or significant opioid-sparing effect in postoperative cardiac patients.