Laparoscopic Antireflux Surgery in Infants with Single Ventricle Physiology: A Review

Introduction

Since the early 1990s, laparoscopic fundoplication and gastrostomy tube placement have grown in popularity within the pediatric population for treatment of gastroesophageal reflux disease and failure to thrive, respectively.^1–5 Numerous studies have documented positive outcomes with these procedures in healthy infants.^2–4 Collins et al. ³ found that laparoscopic fundoplication and gastrostomy tube placement led to decreased postoperative stays and shorter time to feeding tolerance than in patients who had these procedures performed in the traditional open fashion. One population that has an increased need for gastric surgery in infancy is patients with congenital heart defects, especially those with “single ventricle physiology.”^6–10 Single ventricle physiology defines a group of congenital heart defects in which one of the two ventricles is either absent or severely hypoplastic. ¹¹ This physiology can be recognized in infants born with critical aortic and pulmonary stenosis and interrupted aortic arch; however, hypoplastic left heart syndrome (HLHS) is the most common defect, accounting for 2–3% of all congenital heart defects and affecting 2.5–3.5 of every 10,000 live births.^11–14 These infants suffer severe hypoxia secondary to mixing of their systemic and pulmonary venous return, requiring staged surgical reconstruction or cardiac transplantation.^11,15 There is concern that laparoscopic surgery and use of CO₂ insufflation for pneumoperitoneum in this patient population may have negative effects on cardiopulmonary function.^16,17

Management of Infants with Single Ventricle Physiology

Single ventricle physiology encompasses several congenital heart defects, defined by the mixing of systemic and pulmonary venous return due to the absence or hypoplastic nature of one of the two ventricles. ¹¹ Treatment of this condition involves either a three-staged reconstruction or cardiac transplantation.^15,18 Because of the low availability of donor hearts and risk of rejection, the staged reconstruction is often the preferred management for these patients. ¹⁸ Reconstruction refers to the reorganization of the cardiovascular system so the pumping ventricle supplies only the systemic circulation while the pulmonary system relies on passive flow from venous return.^11,15 The first successful initial palliation for HLHS ¹⁹ was performed in 1983 by Norwood et al. ²⁰ There has since been a dramatic improvement in overall mortality following reconstruction, with 70% of patients currently diagnosed expected to reach adulthood. ²¹ Table 1 summarizes the major characteristics of each stage.

The goal of initial palliation is to establish unobstructed venous return from both the pulmonary and systemic systems, provide unobstructed systemic outflow, and limit pulmonary blood flow and pressures. ¹¹ The operation is usually performed within the first week of life. ¹⁹ Numerous procedures can be used, depending on the initial anatomy. These include the Norwood procedure, a modified Blalock–Taussig shunt (MBTS), a Sano shunt, or a pulmonary artery (PA) band. ¹¹ The Norwood procedure involves the use of autologous pulmonary homograft material to create a neo-aorta, which forms an attachment with both the proximal aorta, the PA, and the distal aorta. ²² The distal PA is left in discontinuity, usually closed by a synthetic patch, and pulmonary arterial flow is then provided by either an MBTS or a Sano shunt.^22,23 An MBTS is a synthetic shunt connection from the right innominate artery to the right PA.^21,22 The Sano shunt connects the right ventricle to the left PA.^22,23 Postoperative mortality following the Norwood procedure is reported to be between 2% and 29%, with decreased mortality at specialized centers and possibly with those using the Sano shunt over the MBTS.^{19,21,24–27}

The second stage of reconstruction, the hemi-Fontan or bidirectional Glenn, involves an anastomosis of the superior vena cava (SVC) to the PA. Although more dependent on the patient's status, this stage usually occurs between 4 and 6 months of age.^15,18 Specific factors that play into the timing of this operation include findings on echocardiography and cardiac catheterization. Degree of polycythemia and arterial oxygen saturation are also important determinants.^18,19 The bidirectional Glenn involves ligation of the SVC from the right atrium and an end-to-side anastomosis of the SVC to the right PA, allowing blood to flow “bidirectionally” within the PA. ²² The hemi-Fontan involves the same initial procedure, along with a second anastomosis of the PA to the proximal SVC (closer to the right atrium). ¹⁵ The right atrium/PA junction is then temporarily occluded with an allograft patch. Further allograft patching is used to enlarge the pulmonary arteries. The MBTS or Sano shunt created during the first stage of palliation is ligated in this second stage. ²² Stage 2 reconstruction results in the SVC becoming the only provider of pulmonary blood flow, facilitating the eventual third stage of reconstruction.^15,22 Mortality of these procedures is remarkably lower than the first stage of palliation, around 3–5%. ¹¹

The third and final stage of reconstruction, known as the Fontan procedure, channels the blood flow from the inferior vena cava into the PA, thus completing a single series.^15,18,22 This procedure also has a variable range in timing but usually occurs at a minimum of 6 months following Stage 2, often between 12 and 24 months of age. ²² It is most commonly prompted by worsening hypoxemia, but other concerns include the risk of stroke due to the persisting direct venous–arterial connection, as well as the formation of arteriovenous malformations due to chronic hypoxemia.^15,18 The two approaches commonly used in the Fontan procedure include the lateral caval tunnel and an extracardiac Fontan connection. The lateral caval tunnel involves placement of synthetic graft tubing within the right atrium, connecting the inferior vena cava to the PA. An extracardiac Fontan connection is an anastomosis of a conduit (either synthetic or allograft) between the inferior vena cava and the right PA placed externally to the heart. ²² The Fontan procedure carries an estimated mortality similar to the second stage of palliation. ²⁸

Interstage Concerns

Interstage physiology and related issues are very important to the surgeon who may be required to operate on patients with HLHS. Each stage provides a different list of concerns. Stage 1 palliation occurs very early in life, and it is rare to perform procedures prior to this. At this point, there is often a mismatch in pulmonary and systemic flow depending on the exact malformation. It is important to keep the ductus arteriosus patent with prostaglandins to balance these systems. ²² Following the Norwood procedure, a major concern is shunt thrombosis.^22,24 These infants are often kept on acetylsalicylic acid post-procedure. Other issues include polycythemia secondary to chronic hypoxemia, as well as a risk of stroke or other ischemic events secondary to a direct venous-arterial connection that could lead any venous emboli (either air or solid) to travel directly into the systemic system. Because of these concerns and the overall fragility of the cardiovascular system, it is not recommended to perform any elective operations during this interval. ²²

Following Stage 2 reconstruction, the patient is often more stable, and elective noncardiac operations may be considered. Perioperative management includes judicious fluid administration and avoidance of inotropic agents to prevent increased pulmonary resistance. These patients still have some amount of persistent cyanosis; therefore polycythemia and its related sequelae are still concerns. Passage of venous emboli into the systemic circulation remain a major concern. ²²

After the Fontan procedure, many patients may eventually undergo an elective operation, although there are still concerns that must be considered. Intraoperatively, hypovolemia may result in decreased preload, which will affect these patients more dramatically than in those with a biventricular circulation. ²² Positive pressure ventilation can increase pulmonary vascular resistance; therefore spontaneous respiration is recommended if possible for elective operations.^22,29 The cardiopulmonary circulation following Stage 3 reconstruction has also been linked with more chronic concerns. Protein-losing enteropathy has been seen in between 3% and 10% of post-Fontan patients, possibly secondary to increased hepatic venous pressure. ³⁰ Still other concerns relate to coagulation issues in these children. Although Raffini et al. ³¹ analyzed post-Fontan and non-Fontan children for hypercoagulability and found no significant differences, large numbers of cases of thromboembolism are found found in post-Fontan patients.^21,30,32,33

Laparoscopy and Single Ventricle Physiology

Over the last two decades, laparoscopic surgery has grown exponentially in both the adult and pediatric populations.^34–36 Laparoscopic surgery exposes a patient to different physiologic effects than traditional open surgery.^37,38 In adults, studies have shown that the increased intraabdominal pressure necessary in laparoscopic surgery can have a significant impairment on cardiac septal wall motion and causes a decrease in cardiac index.^38–41 Few studies have attempted to determine the effect of increased intraabdominal pressure in infants. Gueugniaud et al. ³⁸ found that peritoneal CO₂ insufflation at a pressure of 10 mm Hg resulted in decreased aortic blood flow and increased systemic vascular resistance but no changes in mean arterial pressure. Huettemann et al. ⁴² found an intraabdominal pressure of 12 mm Hg resulted in an increase in arterial pressure, systemic vascular resistance, and systolic wall stress, as well as a decrease in cardiac index.

Patients with single ventricle physiology have an increased need for gastric surgery compared with otherwise healthy infants.^6–10 Jeffries et al. ⁶ described their experience with HLHS patients after the Norwood procedure; they found 18% of their patients required gastrostomy tube placement, and 15% required an antireflux procedure. Skinner et al. ¹⁰ found an aspiration rate of 24% after the Norwood procedure, often necessitating gastrostomy tube placement. Garcia et al. ⁷ explored whether patients should be given preemptive gastrostomy tubes prior to the Norwood procedure to decrease postoperative morbidity; patients with preemptive gastrostomy tubes had a significantly higher rate of transplant-free survival to discharge at 90% versus 81% in the group without preemptive gastrostomy tube insertion.

Few studies have explored the safety of laparoscopic gastric surgery in infants with single ventricle physiology. Slater et al. ¹⁷ reviewed their institution's HLHS patients who had undergone laparoscopic surgery between September 2002 and March 2005; of the 12 patients reviewed, 11 had a total of 12 laparoscopic gastric surgeries, and 1 had a laparoscopic Ladd's procedure. There were no intraoperative or postoperative complications attributed to the laparoscopic approach. Postoperative complications were seen in 6 patients, including cellulitis, bowel obstruction, urinary tract infections, and postoperative sepsis that did not appear to be secondary to wound or intraabdominal infections. ¹⁷ In a study of 39 HLHS patients who underwent open gastric surgery, there was a 41% complication rate, including three intraoperative complications (hemodynamic instability and a pulmonary hypertensive crisis requiring extracorporeal membrane oxygenation) and several postoperative complications, including sepsis, necrotizing enterocolitis, bowel ischemia, cardiac failure, deep venous thrombosis, and cerebral infarction. ⁴³ Mariano et al. ¹⁶ reviewed their institution's experience with 5 HLHS patients who underwent laparoscopic Nissen fundoplication; all 5 patients had pneumoperitoneum pressures up to 12 mm Hg and had no intraoperative or postoperative complications, although they did recommend more invasive monitoring (i.e., arterial line) and admission to a cardiac intensive care unit for postoperative care.

Norén et al. ⁴⁴ performed a retrospective review of 31 patients at their institution with congenital heart disease who underwent laparoscopic gastrostomy placement. Fourteen of these patients had single ventricle physiology (6 HLHS patients). No patient suffered from perioperative complications, although 2 required conversion to an open approach because of difficult anatomy and adhesions from previous surgery. In the patients with single ventricle physiology, 2 had postoperative wound infections, and 8 had stoma-related problems that were managed conservatively. ⁴⁴

The group at Emory University ⁸ reviewed 114 Nissen fundoplications in patients with congenital heart disease performed at Children's Healthcare of Atlanta, 104 of which were performed laparoscopically. Thirty-seven of these patients had single ventricle physiology. Thirty-one patients (28%) suffered from postoperative complications, including recurrent gastroesophageal reflux disease, mechanical ventilator dependence, necrotizing enterocolitis, iatrogenic hemothorax, and aspiration. Overall mortality was 26%, although only one death was attributed directly to the procedure: aspiration upon extubation with immediate cardiopulmonary arrest. ⁸ The authors explained that this mortality rate is comparable to the average mortality for patients with congenital heart defects, quoting Tweddell and Spray, ⁴⁵ who estimated a mortality of 31.4% for HLHS patients following the Norwood procedure.

Conclusions

With advances in technology and scientific knowledge, today patients with congenital heart disease, especially those with single ventricle physiology, have increased survival compared with decades past.^18,45 Because of increasing survival, noncardiac surgery is being required more frequently in this patient population.^6–10 Laparoscopic gastric surgery has been found to be safe and effective in the general pediatric population.^1–5 Within the cohort of infants with single ventricle physiology, there is concern that the insufflation pressures required for pneumoperitoneum may have adverse cardiopulmonary effects despite a lack of evidence.^16,17 A multidisciplinary evidence-based protocol for perioperative management of these infants undergoing laparoscopic gastric surgery could potentially decrease morbidity and provide a foundation for more definitive prospective studies.