Mesenteric Oxygen Desaturation in an Infant with Congenital Heart Disease and Necrotizing Enterocolitis

Abstract

Congenital heart disease is a risk factor for the development of necrotizing enterocolitis, although the exact mechanism of development remains unclear. Herein, we report the case of an infant with pulmonary atresia, an intact ventricular septum, and multiple aortopulmonary collateral vessels. At 4 weeks of age, the infant developed necrotizing enterocolitis in association with significant mesenteric oxygen desaturation, as measured by means of near-infrared spectroscopy. With bowel rest and antibiotic therapy, the patient's mesenteric oxygen saturation and clinical status improved.

This case highlights the importance of impaired mesenteric oxygen delivery consequential to congenital heart disease as a possible risk factor for necrotizing enterocolitis, and the use of near-infrared spectroscopy to measure tissue perfusion noninvasively in high-risk patients. To our knowledge, this is the 1st report of mesenteric oxyhemoglobin desaturation in association with necrotizing enterocolitis in a patient who also had congenital heart disease.

The pathophysiology of necrotizing enterocolitis (NEC) is not completely understood, but it is thought to involve a complex interplay between several factors: immaturity of the neonatal gut mucosa, mesenteric ischemia, tissue hypoxia, enteral alimentation, and the presence of infectious agents or toxins. Necrotizing enterocolitis is most commonly seen in preterm neonates; however, 10% of neonates who are diagnosed with NEC are term or near-term. These term patients often have additional risk factors that can predispose them to bowel ischemia, such as congenital heart disease (CHD), intrauterine growth restriction, or birth asphyxia.^1–4 Of those patients with CHD, factors associated with an elevated risk of developing NEC include aortic arch obstruction, episodes of poor systemic perfusion, and lesions with significant systemic-to-pulmonary runoff, including hypoplastic left-heart syndrome and truncus arteriosus.¹

Near-infrared (NIR) spectroscopy is a noninvasive means of measuring regional tissue oxygenation. The technique relies on the differential absorption of NIR light by oxyhemoglobin and deoxyhemoglobin. Measuring the attenuation of light at different wavelengths and different distances between the emitter and the diode enables real-time estimation of deep-tissue oxygenation, which is expressed as a weighted average primarily of venous but also of arterial circulation.⁵ Because impaired tissue oxygen delivery results in greater oxygen extraction and therefore in venous oxygen desaturation, the NIR spectroscopic signal can be used to detect regional tissue hypoxia.

We report a case of mesenteric oxyhemoglobin desaturation in association with NEC in a neonate with CHD.

Case Report

In April 2004, a 2.4-kg term infant was transferred to our facility when 2 weeks old for further evaluation of CHD. He had been mechanically ventilated during his 1st week of life and had been successfully extubated before transfer. A balloon atrial septostomy had been performed when he was 2 days old, and adequate oxygen saturation had been maintained by means of prostaglandin infusion since his 3rd day of life. On arrival at our institution, the patient's pulse oximetry on room air was 88% to 96%. His systolic blood pressure ranged from 65 to 90 mmHg; his diastolic pressure ranged from 30 to 45 mmHg.

Echocardiography showed pulmonary valve atresia with an intact ventricular septum. The branch pulmonary arteries appeared to be confluent but severely hypoplastic. Multiple aortopulmonary collateral vessels arose from the brachiocephalic trunk and from the posterolateral and anterior surfaces of the proximal descending aorta. The left main coronary artery was moderately dilated, with suspected fistulous connection to the right ventricular outflow tract. The right ventricle was hypoplastic and hypertrophied, and the interventricular septum bowed into the left ventricle. Because this patient had adequate and stable pulmonary blood flow, our clinical management plan was to encourage somatic growth until possible later surgical intervention, which would entail pulmonary valve perforation or single-ventricle palliation.

Oral feedings resulted in severe gastroesophageal reflux. The patient initially tolerated nasogastric feedings of fortified formula (24 kcal/oz) at a daily volume of120 cc/kg, and he gained weight adequately. At the ageof 34 days, he developed hematochezia. Upright and lateral decubitus abdominal radiography showed extensive intramural pneumatosis of the ascending, transverse, and proximal descending colon without intraperitoneal or portal venous air. The diagnosis of NEC was made. The infant developed neither acidosis nor thrombocytopenia, and he remained hemodynamically stable.

Approximately 48 hours after the diagnosis of NEC, a NIR spectroscopic probe (INVOS® 5100B, Soma-netics Corporation; Troy, Mich) was placed on the patient's abdomen in the midline below the umbilicus and above the pubic symphysis in order to measure regional mesenteric oxygenation, and a 2nd probe was placed on the right side of the patient's forehead to measure cerebral oxygen saturation. The examination revealed significant mesenteric desaturation (24.5% ± 9.3%) in comparison with cerebral saturation (53.4% ± 4.0%) during the initial 24 hours of monitoring (P <0.0001). With intravenous antibiotics and bowel rest over the next 24 hours, the patient's mesenteric saturation significantly improved (43.1% ± 7.6%, P <0.0001) in comparison with initial mesenteric saturation (Fig. 1). The improvement of the mesenteric-to-cerebral ratio to 0.87 during the latter half of the monitoring period is shown in Figure 2. Oral feeding with nasogastric supplementation was restarted when the patient was 48 days old, and he tolerated it well. The infant recovered uneventfully and was discharged from the hospital when he was 63 days old.

Fig. 1 Regional cerebral and mesenteric oxygen saturations in our patient, as measured by near-infrared spectroscopy, after the diagnosis of necrotizing enterocolitis.

Fig. 2 The mesenteric-to-cerebral oximetry ratios in our patient throughout the monitoring period.

Discussion

Near-infrared spectroscopy has been used extensively to monitor cerebral perfusion in neonates, especially during cardiac surgery and cardiopulmonary bypass. More recently, reported uses of NIR spectroscopy have included the measurement of perfusion in other tissues, such as those of the liver, the kidneys, and the lower abdomen.^6–8

Regional mesenteric oxygen desaturation in association with NEC has been described by Fortune and colleagues.⁸ They monitored a series of neonates who had acute surgical abdomen (including 5 with NEC) together with decreased mesenteric-to-cerebral oxygenation ratios. In that study, the control group had a median mesenteric-to-cerebral oxygenation ratio of 0.96, whereas the acute-abdomen group had a significantly lower ratio of 0.66 (P <0.001). The authors also reported that a ratio of less than 0.75 was predictive of intestinal ischemia (positive predictive value) and that a ratio of 0.96 or more excluded the diagnosis. Notably, the poorest correlation between mesenteric desaturation and abdominal pathology was found in a single patient who had systemic arterial oxygen desaturation and NEC. None of the patients in the Fortune group's study had cyanotic CHD.

It is unknown whether gut ischemia or hypoxia is a primary or secondary factor in the development of NEC. The finding that our patient experienced significant mesenteric desaturation during the early stages of NEC supports the hypothesis that tissue hypoxia is central to the pathophysiology of NEC, especially in term neonates with CHD. The development of mesenteric desaturation may be due to decreased regional oxygen delivery that is secondary to decreased cardiac output or increased flow to the pulmonary vascular bed, increased oxygen extraction, or both. As shown previously,¹ decreased cardiac output is likely most significant in patients who have large systemic-to-pulmonary runoff lesions that may “steal” blood flow from the mesenteric tissue bed.

Our case supports the use of noninvasive techniques (such as NIR spectroscopy) to measure tissue perfusion in order to evaluate at-risk patients for tissue hypoxia, and it highlights the possibility of monitoring patients over time to determine their response to medical therapy. Researchers who conducted a porcine study detected an increased risk of neurologic impairment when regional cerebral saturation as measured by NIR spectroscopy was less than 45%; however, the authors did not suggest a regional mesenteric saturation threshold below which there is an increased risk for hypoxic or ischemic bowel injury.⁹ Perhaps when the response of early NEC to therapy is better understood, more aggressive intervention may be sought for patients who do not respond positively to therapy, and earlier progression to enteral feeding may be considered safe for those who improve rapidly.

To our knowledge, our report is the first to document mesenteric oxyhemoglobin desaturation in association with NEC in a patient who also had CHD. The case of our patient highlights the importance of impaired mesenteric oxygen delivery in the presence of CHD as a possible risk factor for NEC, and the use of NIR spectroscopy to measure tissue perfusion noninvasively in high-risk patients. As experience with this technology grows, it is likely that the understanding and management of patients such as ours will improve.