Evaluation of Cardiac Function During Laparoscopic Gastrostomy in Pediatric Patients with Hypoplastic Left Heart Syndrome Using Intraoperative Transesophageal Echocardiography

Introduction

Hypoplastic left heart syndrome (HLHS) refers to a type of congenital heart disease where the left ventricle is underdeveloped and does not contribute to cardiac output; it is universally fatal without surgical intervention. ¹ Concerns exist as to whether or not laparoscopic surgery is safe in patients with HLHS, given required increases in intraabdominal pressure (IAP) and subsequent effects of cardiac function. Increased IAP in healthy children has been shown to decrease cardiac index, aortic blood flow, and stroke volume at pressures as low as 10 mm Hg.^2,3 Two small case series have reported safe use of laparoscopy in patients with HLHS.^4,5 Measurements of cardiac function have not been reported during laparoscopic surgery in patients with congenital heart disease.

Patients and Methods

After approval from the Internal Review Board (IRB number 1111-193E), patients with single ventricle physiology (SVP) who underwent laparoscopic gastrostomy tube placement between August 2011 and February 2013 were prospectively identified. Demographic data and reason for gastrostomy tube were recorded. All patients were monitored by a cardiac anesthesiologist with standard American Society of Anesthesiologists monitors plus invasive arterial monitors and cerebral and somatic near-infrared spectroscopy. All patients had central venous access. Patients were transferred to the pediatric intensive care unit postoperatively. Transesophageal echocardiography (TEE) was performed by a cardiologist in all cases. Given the inability to obtain routine measurements of cardiac function secondary to anatomic differences, a novel measure was created. Fractional shortening (FS) of the single ventricle was calculated by subtracting end-systolic chamber length from diastolic chamber length and dividing the difference by the diastolic chamber length. FS was compared before, during, and after insufflation of the abdomen. In order to assess clinical changes, hemodynamics were compared at the same time points. Continuous variables were compared using one-way analysis of variance. Significance was defined as a value of P ≤.05.

Results

Laparoscopic gastrostomy tube placement was performed in 6 patients with HLHS and 5 patients with SVP. Demographics, operative time, maximum insufflation pressure, and concomitant procedures are reported in Table 1. All HLHS patients had completed first-stage palliation; in addition, 1 patient in each group had undergone second-stage palliation with the Glenn procedure. One patient with HLHS and 1 with SVP also underwent laparoscopic fundoplication, which explains their longer operative times. Insufflation pressure was set to the lowest possible level to allow for adequate visualization.

Table 2 reports mean arterial pressure and FS of the single ventricle before and during insufflation and after desufflation of the abdomen. Six patients received dopamine infusion during the operation at the discretion of the anesthesiologist, who was aware of changes in the FS. Mean arterial pressure was maintained during insufflation despite a trend toward decreased FS, which returned to baseline after desufflation. No postoperative complications were noted. There was no operative mortality.

Discussion

Patients with HLHS show impaired growth prior to their second-stage operation, and preemptive gastrostomy tube placement has been studied. ⁶ However, operative risk for noncardiac cases is very high. ⁷ Concerns exist as to whether or not laparoscopic surgery is safe in patients with congenital heart disease, especially those with HLHS, given required increases in IAP and subsequent effects on cardiac function. ⁸ In healthy children, an IAP of 12 mm Hg, but not 6 mm Hg, during laparoscopic hernia repair has been shown to decrease the cardiac index using TEE. ² Use of echo-Doppler has shown decreases in aortic blood flow, stroke volume, and cardiac index in children at pressures as low as 10 mm Hg. ³ These significant changes have been well tolerated in healthy patients but have the potential to cause catastrophic problems in SVP patients, such as reversal of flow in the Blalock–Taussig shunt with resultant life-threatening occlusion. Additionally, the increase in arterial pressure of CO₂ induced by laparoscopy needs to be considered and managed in these high-risk patients, as elevations in CO₂ levels have been shown to increase pulmonary vascular resistance in healthy patients. ⁹ A significant increase in the arterial pressure of CO₂ and end-tidal CO₂ gradient after abdominal insufflation has shown end-tidal CO₂ to be an insensitive monitor in healthy infants. ¹⁰

The successful use of laparoscopic surgery in patients with HLHS has been reported in two small case series. Slater et al. ⁴ reported experience of 12 patients with HLHS undergoing 13 laparoscopic operations with no mortality. A second report demonstrated no complications after five laparoscopic Nissen fundoplications using an IAP of 12 mm Hg in this patient population. ⁵ Our study is the first to demonstrate the real-time changes in cardiac function in these patients. FS of the ventricle appears to be reduced in this small study during insufflation but returns to baseline shortly after desufflation. In addition, no patient demonstrated change in mean arterial pressure.

Infants with HLHS and SVP are heterogeneous as there are 21 first-stage palliative procedures. ¹¹ Each of these has advantages and disadvantages and causes some variation in cardiac physiology. For example, the Sano modification eliminates diastolic runoff into the pulmonary circulation and potentially improves coronary perfusion pressure. ¹² Selection of the specific procedure is patient and institution specific. With each surgical option, there is frequently an imbalance in the ratio of pulmonary to systemic blood flow, which is determined by the balance between the pulmonary and systemic vascular resistances. ⁵ This ratio changes over time because of factors such as shunt calcification/stenosis, infant growth, and the development of pulmonary hypertension. Our patients underwent various first-stage palliative procedures, and 2 had undergone second-stage palliation. Therefore, it is not possible to determine the effects of pneumoperitoneum on cardiac function for all of these approaches. This would be an area for future investigation in collaboration with other institutions.

Acute myocardial depression during insufflation, as evidenced in this study, is important to immediately recognize in order to manage reductions in shunt flow due to alterations in the ratio of pulmonary to systemic blood flow and myocardial function, as these may result in reversal of shunt and catastrophic occlusion. ¹³ The use of TEE to monitor function before, during, and after insufflation may allow adjustments to be made to peak insufflation pressure and inotrope administration in order to mitigate the decrease in myocardial function. This may be important, especially in longer procedures, as function may deteriorate over longer periods of time. It is interesting that inotropic support was not required in the 2 patients who underwent longer operations for fundoplication. We concur with previous recommendations to use careful hemodynamic monitoring and a multidisciplinary approach, which includes a skilled pediatric anesthesia team and postoperative management in a pediatric intensive care unit.

This prospective pilot study demonstrates that TEE is able to detect real-time changes in cardiac function of children with SVP undergoing laparoscopic gastrostomy tube placement and that those changes do not affect hemodynamics. Based on our study, laparoscopic gastrostomy tube placement is safe in this patient population.