Infants born with heart problems are at risk of developing gut disease due to reduced blood flow to the intestines which can result in poor weight gain, surgery and even death. At present, doctors are often unaware of any gut problems until clinical symptoms present (poor feed tolerance, blood stained stools or bloated stomach) which is often too late to prevent gut damage.

Earlier diagnosis of gut disease may now be possible; calprotectin is produced when the gut is inflamed and can be found in faeces and blood. Calprotectin levels have been shown to be a reliable marker in diagnosing gut disease in premature infants. To date, calprotectin levels have not been monitored in infants with cardiac defects, who like premature infants are at high risk of gut disease but the cause of gut disease is different to that seen in premature infants and therefore requires specific monitoring.

This study will implement a high risk feeding protocol which has been adapted from current feeding practices from the United States; the aim being to promote weight gain without increasing the risk of gut inflammation. Furthermore, the study will validate whether faecal calprotectin is a useful non-invasive marker in identifying gut disease in infants with cardiac defects. Currently, infants are diagnosed with necrotising enterocolitis by an abdominal X-ray (current 'Gold Standard'); infants who have a positive diagnosis will have faecal calprotectin levels cross-checked. From this data, cut-off values will be established which will provide data to diagnose necrotising enterocolitis eliminating the need for X-rays (radiation).

Secondly, faecal calprotectin levels will be measured at strategic time points (longitudinal data) linked to increased risk of gut damage (following cardiac surgery and feeding) which will then be cross-checked against infants that developed NEC to identify whether high risk infants had raised calprotectin levels earlier.

Background:

Congenital heart disease is the most common birth defect in the U.K., affecting between 0.3-1.5% (9 in every 1000 live births) of infants (Hoffman and Kaplan 2002), and requires immediate surgical intervention. Infants with heart defects are known to exhibit early and progressive falls in their growth trajectory compared to healthy infants, increasing risk of death post-surgery (Eskedal, Hagemo et al. 2008). Lifesaving cardiac surgery and feeding can increase the infant's risk of developing gut inflammation and necrotising enterocolitis (NEC). NEC is a severe gastrointestinal disorder associated with high morbidity and mortality (Kelleher, Laussen et al. 2006).

Infants with cyanotic heart defects are at particular high risk of developing NEC due to impaired cardiac output which reduces blood flow to the gut, resulting in poor tissue oxygen perfusion, leading to gut ischemia (Neu and Walker 2011). It has been proposed that enteral feeding and lifesaving cardiac surgery - in particular the length of cardio-pulmonary bypass can further compromise the integrity of the gut wall (Golbus, Wojcik et al. 2011); however, the extent to which these interventions cause gut damage is unclear.

The initial clinical manifestations of NEC are nonspecific and indistinguishable from other type of sepsis. As a precautionary measure feed is often stopped in infants with suspected gut inflammation until symptoms have resolved, further contributing to poor growth. A six stage classification system has been devised to identify differing severity of NEC (Bell's Classification), stage six being the most severe level (Bell, Ternberg et al. 1978). A definitive diagnosis (Gold standard) of NEC is confirmed radiographically from abdominal X-Ray - signs include intramural gas, pneumatosis, and portal venus gas. The disease often has a rapid onset and therefore early detection of gut inflammation could avert severe gut damage and prolonged periods of parenteral nutrition on the cardiac intensive care unit (Sharma and Hudak 2013).

The relationship between feeding and the development of NEC has been well described in preterm infants with few incidences of NEC developing before feeding commences. However, this distinction is less clear in infants with cyanotic heart defects as compromised blood flow to the gut and cardiopulmonary bypass can result in damage before feeding commences (Iannucci, Oster et al. 2013). In light of the link between feeding and NEC it is imperative that the gastrointestinal tract is not overloaded by over feeding infants. Overfeeding can cause stasis of milk substrate in the gastrointestinal tract due to dysmotility leading to intestinal dilatation with fluid and gas and possibly to the impairment of the intestinal mucosal barrier. Intestinal dilatation in the presence of abnormal microbial colonisation (dysbiosis) can distort normal signal transduction across the intestinal wall barrier resulting in excessive inflammation and intestinal wall death (necrosis) (Ravindranath, Yoshioka et al. 1997).

Despite improvements in survival rate of infants undergoing palliative cardiac surgery, one issue that remains is poor weight gain. The pattern of poor growth following surgery has been well described with the poorest growth occurring during the early post-operative phase but continuing up until discharge. Poor growth and longer hospital stay are risk factors for death in infants with congenital heart disease (Medoff-Cooper, Irving et al.

2011). S100/ A8 - Calprotectin Recently, a non-invasive faecal biomarker (calprotectin) has been used to diagnose NEC and is also a potential predictive marker of NEC in preterm infants (Pergialiotis, Konstantopoulos et al. 2016). However, to date, faecal calprotectin has not been used to diagnose or predict infants at risk of NEC with cardiac defects, who have a different aetiology compared to preterm infants. The manifestation of NEC in infants with cardiac defects is driven by compromised blood perfusion to the splanchnic area which is further exacerbated by cardiac surgery, particularly the length of cardiopulmonary bypass (Typpo, Larmonier et al. 2015). Whereas preterm infant's aetiology for NEC is associated with gut immaturity, alimentation, microbiota and host defence mechanisms (Pergialiotis, Konstantopoulos et al. 2016). In light of this differing pathogenesis of NEC in preterm and cardiac infants, it is essential to establish specific validation for infants with cardiac defects.

Calprotectin is a neutrophil activation marker and therefore distinguishes between acute bacterial and viral infection. Calprotectin (36.5kDa) is a calcium and zinc binding protein of the S100/ calgranulin family. It is mainly exhibited in the cytoplasm of neutrophils (about 5% of their total protein content) and expressed on activated monocytes and macrophages (Yui, Nakatani et al. 2003). Furthermore, it participates in leukocyte interactions with the endothelium and cellular adhesions, leading to the recruitment of leukocytes to inflamed intestinal tissue Therefore, elevated levels are evident in infectious and inflammatory diseases such as NEC (Stroncek, Shankar et al. 2005).

A systematic review of the current evidence (13 studies) suggests that faecal calprotectin is elevated in preterm infants suffering from NEC. (Pergialiotis, Konstantopoulos et al. 2016). Five studies evaluated the efficacy of faecal calprotectin as a diagnostic marker which ranged from 792ug/g (76% sensitivity; 92% specificity; P-value <0.001) (Aydemir, Aydemir et al. 2012) to 480ug/g (100% sensitivity; 84% specificity) (Bin-Nun, Booms et al. 2015).

Serum calprotectin has also been shown to be an accurate marker of gut inflammation in Crohn's disease (Lehmann, Burri et al. 2015). Furthermore, a study by Reisinger et al (2014) monitored serum calprotectin (Amyloid A) concentrations in 29 neonates with a diagnosis of NEC and monitored its usefulness in identifying disease severity and established a cut-off value of > 27.8ng/ml (71% sensitivity; 83% specificity) (Reisinger, Kramer et al. 2014). However, infants with heart defects pose a unique clinical presentation in that their gut may be inflamed from birth due to compromised gut perfusion and therefore warrant specific investigation and monitoring.

Retrospective chart review I performed a retrospective chart review at Great Ormond Street Hospital (GOSH) on infants with single ventricle heart defects and identified a 20% incidence rate of NEC Bell's stage 1 and a 12% incidence rate of NEC Bell's stage 2. In all but one case NEC developed within seven days post-surgery. Furthermore, infants that developed NEC on average stayed on CICU 16 days longer than infants that did not. Moreover, 56% of infants had their feed temporally stopped (24-72hrs) at least once due to suspected gut inflammation. This has a huge implication on the nutritional intake of these already nutritionally compromised infants. By differentiating between what is potentially bacterial gut inflammation and other systemic issues (non-specific viral infection) could dramatically reduce the unnecessary stopping of infants feeding.

HYPOTHESIS

Primary Hypothesis

H0: Faecal calprotectin levels will not be different among infants with or without necrotising enterocolitis

HA: Faecal calprotectin levels are different among infants with and without necrotising enterocolitis.

Secondary Hypotheses

Baseline faecal calprotectin levels are not associated with the development necrotising enterocolitis in infants with heart defects

To investigate the above hypotheses a cross-sectional study will be performed to validate whether faecal calprotectin can be used as a diagnostic tool for necrotising enterocolitis. Additionally, a longitudinal study will measure faecal calprotectin a specific time points to ascertain whether there is an association between calprotectin levels after surgery and during feeding with the development of necrotising enterocolitis.

This will involve reviewing information relating to diagnosis and plan for surgical procedure.

If the parents wish to find out more about the study they can discuss with the principle investigator; written Parent/ legal guardian information sheet will be provided.

If participants are happy to proceed, the formal consent process will be initiated. Signed copies will be placed in the medical notes and research file

Inclusion Criteria:

• Term infants (>37 weeks gestation) delivered vaginally (bacterial colonisation)
• Birth weight > 2.0kg (low birth weight classification)
• High risk infants - cyanotic heart defect (univentricular heart - hypoplastic left heart syndrome and hypoplastic right heart, or truncus arteriosus or coarctation of Great arteries).

Exclusion criteria:

• Any gastroenterological complications such as gastro schisis
• Mother or infant received antibiotics 2 weeks prior to delivery

Methods Specific data relating to cardiac surgery will be collected

Data collected during surgery include:

• Infant age and weight at time of surgery
• type of surgery performed (Norwoods procedure or modified Blalock- Taussig shunt [mBTS] or right ventricle to right pulmonary artery conduit (Sano)) including the size of mBTS or Sano
• length of time on cardiopulmonary bypass and cross-clamp time
• length of stay on CICU after surgery.

As per current feeding practice infants will commence intravenous nutrition post cardiac surgery with the introduction of enteral nutrition within 72hrs post-surgery. However, this is highly dependable on the medical state of the infant and will be controlled as much as possible.

Expressed breast milk (EBM) is the preferred feed of choice and once milk is expressed bottles will be labelled and stored in the ward fridge and used as required. If EBM is not available then a hydrolysed feed will be used called Pepti- Junior (standard concentration 13.8% Cow&Gate) which will be prepared by a technician in the hospitals special feeds unit. Feed bottles will be changed every four to six hour.

Routine treatment - Infants will commence on a high risk feeding protocol (adapted from published guidelines (Slicker, Hehir et al. 2013). Nasogastric tube feeds will start at 0.5ml/kg for 8hrs. If aspirates are less than 5ml/ kg feed will be increased by 0.5ml/kg and Continued to increase by 0.5ml/kg every 8hrs until achieved fluid allowance of 100ml/ kg (worked example below). The number of days taken to meet 100ml/ kg will be recorded.

Worked Example - 4kg Infant:

Start at 0.5ml/kg = 2ml x 8hr = total 16ml in 8hrs Increase by 0.5ml/ kg/ = 4ml x 8hrs = 32ml; totals 48ml in 16hrs Increase by 0.5ml/ kg = 6ml x 8hrs = 48ml; total 96ml in 24hrs Increase by 0.5ml/ kg = 8ml x 8hrs = 64ml; total 160ml in 32hrs Increase by 0.5ml/ kg = 10ml x 8hrs = 80ml; total 240ml in 40hrs Increase by 0.5ml/ kg = 12ml x 8hrs = 96ml; total 336ml in 48hrs (84ml/ kg) Continue increasing until at fluid allowance (100ml/ kg)

Routine treatment - Stomach contents will be aspirated every 4hrs using a 20ml syringe and volume will be measured and recorded. Aspirated contents will be replaced into the stomach once recorded http://www.gosh.nhs.uk/health-professionals/clinical-guidelines/nasogastric-and-orogastric-tube-management.

Routine treatment - Weights will be measured on an infant-weighing scale (kg), which will be calibrated weekly for accuracy. Infant's occipital frontal circumference (OFC) should not be performed before 36 hours of age. It should be done after 36 hours of age or preferably at 7-10 days (Lindley, Benson et al. 1999). To obtain the measurement, loop the measuring tape and place over the child's head. The measuring tape should be placed above the ears and midway between the eyebrows and the hairline to the occipital prominence at the back of the head. Pull the measuring tape so that any hair is compressed. The measurement should be taken to the nearest millimeter.

Methods - Gut inflammation monitoring

Intervention - To monitor the impact of cardiac surgery and enteral feeding on gut inflammation, biomarkers (faecal Calprotectin) will be measured 24-48hrs after surgery and 24-48hrs after enteral feed commences. Additionally, infants that are diagnosed with NEC will have calprotectin levels measured.

Faecal calprotectin - 50-100mg of faeces will be collected in plastic containers from infant's nappies. Samples will then be homogenised by shaking, and supernatants and then sent to laboratories for immediate analysis.

Inclusion Criteria:

• - Term infants (>37 weeks gestation) delivered vaginally (bacterial colonisation)
• Birth weight > 2.0kg (low birth weight classification)
• High risk infants - cyanotic heart defect (univentricular heart - hypoplastic left heart syndrome and hypoplastic right heart, or truncus arteriosus or coarctation of Great arteries).

Exclusion Criteria:

• Any gastroenterological complications such as gastro schisis
• Mother or infant received antibiotics 2 weeks prior to delivery

This study will be performed at Great Ormond Street Children's Hospital NHS Foundation Trust on the cardiac intensive care unit.