Necrotizing Enterocolitis in Neonates: Has the Brain Taken a Hit 10 Years Later?

BACKGROUND: The neonate with necrotizing enterocolitis (NEC) is at risk of developing poor neurodevelopmental outcomes. There is a dearth of long-term follow-up studies in this field, with a majority of studies reporting a follow-up duration of 2 years. The aim of this study was to assess neurodevelopment of babies diagnosed with NEC more than a decade ago. This study was carried out in a tertiary hospital with neonatal surgery and intensive care units.
MATERIALS AND METHODS: Retrospective review of notes and telephone interviews with parents of babies diagnosed with NEC between January 2007 and December 2008 was conducted. Evidence of motor, cognitive, and sensory impairment was recorded. Fisher's exact, χ2, and unpaired t-tests were used. P-values <0.05 were considered significant.
RESULTS: Overall mortality in this cohort was 31%. Eighteen patients were followed up to an average age of 11.2 years. Of the 18 patients, 11 (61%) had a neurological impairment. Of the 15 surgically managed patients, 10 (67%) had an impairment and, of the 3 medically managed patients, 1 (33%) had an impairment. Cognitive impairment was the most common (10/18, 56%), followed by motor (6/18, 33%). Ten of 18 (56%) had special education needs, 9 of 18 (50%) had learning difficulties, 6 of 18 (33%) had speaking difficulties, and 4 of 18 (22%) had cerebral palsy. Patients also had behavioral conditions (3/18, 17%), visual impairment (2/18, 11%), and seizures (2/18, 11%).
CONCLUSION: In the field of NEC, there is a hidden neurological burden that neonatal surgeons bequeath to the community. Sixty-one percent of patients are neurologically impaired, affecting the quality of life and function in the long-term. There should be appropriate parent counseling at the point of diagnosis and regular development checks for children with NEC. Copyright:

INTRODUCTION

Pediatric surgery literature on necrotizing enterocolitis (NEC) tends to concentrate on gastrointestinal morbidity such as short gut syndrome, often neglecting neurological status.[12] Long-term neurological impairment has been noted in preterm and low birth weight infants, per se. Severe systemic inflammation, as seen in NEC, can additionally compound this problem by impairing oligodendrocyte development.[3]

NEC is often fatal, limiting the feasibility of large-scale, long-term neurodevelopmental assessment studies. Validated neurodevelopment assessment tools also tend to focus on infant and toddler development with a seminal check at 2 years, further resulting in generally shorter-term follow-up of neurodevelopment in survivors. This paper thus seeks to follow-up patients diagnosed with NEC as neonates, 10 years down the line. A review of the current literature on neurodevelopment in NEC was also carried out.

MATERIALS AND METHODS

NICU register was searched for patients diagnosed with NEC between January 2006 and December 2007. A retrospective review of case notes and electronic patient records (EPR) was carried out. The presence of the following conditions, used in previous studies to illustrate “neurodevelopment impairment,” was noted where applicable:

Epilepsy, blindness, deafness, speaking difficulties, cerebral palsy (CP), special education needs (SEN), learning disabilities (LD), behavioral issues such as autism spectrum disorder and attention deficit and hyperactivity disorder, and dyspraxia[456789]

Telephone interviews with parents were conducted to confirm EPR findings for alive patients. A template for the telephone interview can be seen as in Figure 1. Impairment was classified as cognitive (SEN, LD, and behavioral issues), motor (CP, dyspraxia, and speaking difficulties), sensory (blindness and hearing loss), or others (epilepsy).

Figure 1

Template used for follow-up telephone questionnaire

The following search was conducted on PubMed for literature review: (((((outcome) OR impairment) OR cognitive) OR neurodevelopment)) AND (((NEC) OR necrotizing enterocolitis) OR necrotizing enterocolitis). Original research articles written in English were included, with no restrictions placed on the date of publication. Only studies which controlled for factors such as prematurity and low birth weight were included.

Ethics approval

The case series was registered as a clinical governance project. All information was confidential and anonymized.

Statistics

Fisher’s exact tests, χ2 tests, and unpaired t-tests were used to compare groups. SPSS (Version 25; IBM Corp., Armonk, NY, USA) was used for analyses. All P-values <0.05 were considered statistically significant.

RESULTS

In total, 67 patients were diagnosed with NEC from January 2006 to December 2007. The mean birth weight was 1216g (470–3580g). The mean gestational age was 29 weeks (23–41 weeks). There was 31% overall mortality (21/67). The mean age at the time of death was 40 days (1–209 days).

Of 46 alive patients, 18 (39%) were contacted. The mean age of the patients followed up was 11.2 (10.5–12.3) years. There were no significant differences in gestational age and birth weight for the group which was followed up and the group which was lost to follow-up (P = 0.9, 0.65, respectively). Prevalence of surgical management between the two groups was also not significantly different (P = 1).

Of 18 patients contacted, 11 (61%) had some form of neurological impairment. Sex, ethnicity, incidence of multiple births, gestational age, and birth weight of the babies were not seen to be statistically different between those who had a neurological impairment and those who did not (P = 0.3, 0.1, 1, 0.8, and 0.9, respectively).

The most common form of neurological impairment was cognitive impairment [Table 1]. Note that some forms of impairment result in a total higher than the total number of children with impairment, as some children had multiple forms of impairment.

Table 1

Neurological impairment at follow-up

Motor impairment (n =15)	Cognitive impairment (n =22)	Sensory impairment (n =3)	Others (n =2)
Speaking difficulties (n =6)	Special Education Needs (n =10)	Blindness (n =2)	Current seizures (n =2)
Cerebral Palsy (n =4)	Learning difficulties (n =9)	Hearing impairment (n =1)
• Quadriplegia (n =2), hemiplegia (n =1), diplegia (n =1)
Dyspraxia (n =1)	Behavioural conditions (n =3)

Ten of 15 (67%) patients with NEC requiring surgical management (sNEC) and one of three (33%) patients with NEC requiring only medical management (mNEC) had neurological impairment. Though not statistically significant (P = 0.528), patients with sNEC within our cohort were 3.143 times more likely to be neurologically impaired than their mNEC counterparts (RR = 3.143, 95% CI 0.346–26.520).

Literature review

Literature review yielded 25 articles which met the inclusion/exclusion criteria. Thirteen were cohort studies and 12 were case–control studies. Nineteen (78%) had a maximum follow-up duration of 24 ± 2 months. The median number of patients in a study was 20, and the mode was 12.

There was a lack of consensus in literature review regarding independent effects of NEC on neurodevelopment. While more studies[10111213141516] found significantly higher rates of neurological impairment in children with NEC, some papers found the differences to be insignificant.[1718192021] All papers which compared babies with sNEC to those with no NEC, however, found significantly higher levels of impairment.[222324252627] Conversely, when analyzing studies which compared babies with mNEC with babies who did not develop NEC, we found that more studies found that differences in neurodevelopment between the two cohorts were not statistically significant.[1522,2326,27]

DISCUSSION

Neurodevelopment refers to the development and maturation of the central nervous system. At its basic, it involves the delicated balance of processes such as neuron generation and growth, synaptogenesis, and neurohormone production.[282930] Starting in utero, neurodevelopment continues well into adult life, as reflected by the Diagnostic and Statistical Manual of Mental Disorders Edition 5 (DSM-5) definition of “neurodevelopment disorders” being deficits in “personal, social, academic, or occupational functioning.”[31]

Premature neonates undergo a seminal check at 24 months age. Correspondingly, most validated forms of neurodevelopment assessment have a major checkpoint at this age. Therefore, most studies on neurodevelopment outcomes in NEC have reported a mean follow-up of 2 years. Our study, however, has a mean follow-up of 11.2 years. As mentioned, neurodevelopment continues beyond infancy, and this much longer follow-up duration allows for a better understanding of the long-term effects of NEC on neurodevelopment.

The majority (61%) of our children diagnosed with NEC as neonates had neurodevelopment impairment persisting into late childhood. This is a high percentage, and while this study does not directly compare with similar children who did not develop NEC, the suggested pathophysiology of NEC—an inflammatory condition with large cytokine surges—can go toward explaining the large percentage of children with neurological impairment. The cytokine storm seen in inflammatory conditions like NEC may add insult to the oftentimes premature brain, causing damage and resultant neurological impairment.

When looking closer at the subgroup analysis, our study showed that 67% of patients with sNEC had neurological impairment, which is comparable to values ranging from 55% to 57% in literature.[2425] This was higher than the prevalence in babies with mNEC (33%). While not statistically significant due to the small sample size, this suggests that patients with sNEC are at higher risk than their mNEC counterparts. This is not surprising because neonates needing surgery are sicker with severe metabolic acidosis, presence of shock, and respiratory embarrassment needing ventilation being the norm.[416] There is also a higher incidence of intraventricular bleeds in surgical patients.[2123]

Referring back to our literature review, it was noticed that all studies comparing impairment in babies with sNEC with babies who had no NEC found statistically significant differences in impairment between the two groups. In contrast, studies comparing babies with mNEC vs no NEC, mostly found differences in impairment to be insignificant. This suggests that sNEC perhaps has a clearer impact on worsening neurodevelopment. Various hypotheses have been put forward to try and explain this observation. For instance, intraoperative hypothermia and cerebral hypotension have been suggested as possible reasons for the observed worse neurological outcomes in babies undergoing surgery.[32] Surgery is indicated in those with more extensive disease, as judged by using “scoring” systems such as the Bell’s criteria, where a combination of escalating radiological and clinical signs are used to determine whether a baby will require surgical intervention. Surgical intervention for NEC has been shown to also result in higher levels of inflammation and risk of sepsis[34]—further exacerbating cerebral insult. Indeed, it is likely that it is the amalgamation of these factors which result in long-term neurodevelopmental consequences.

There has been, in recent years, interest in the possible role of anesthesia in damaging the brains of neonates undergoing even elective surgery.[33] Neurotoxicity due to repeated exposure to anesthetic agents has been suggested and research is underway to identify which of the anesthetic agents are kinder to the developing brain.[33]

On follow-up, cognitive impairment was the most common form of impairment. Previous studies have also found this to be the pattern,[5] but a few report motor impairment to be more prevalent.[35] Interestingly, a study by Simon et al.[36] describes that differences in motor development between children with sNEC and children with no NEC become less significant with increasing age. A further study by Waugh et al.[21] describes how minor signs of CP seemed to “disappear” in their older patients. The higher rates of cognitive rather than motor impairment in our series could thus perhaps be explained by the long follow-up duration.

With the increase in survival of small, preterm babies, it becomes ever more important to have adequate follow-up and recognize that even after these children have made it past their neonatal hurdles, morbidity can extend long into their lives, affecting both themselves as well as the family unit. Studies like this help in decision-making, for it is not only the volatile short term that needs to be considered for these babies.

Limitations

The number of patients followed up was small, as is often the case with rare, fatal conditions such as NEC (the median number of participants in literature was 20). A large number of patients were lost to follow-up (61%). A myriad of social reasons—babies moving into foster care, families moving away—complicates efforts to follow patients up. While there were no statistically significant differences between the contacted and lost to follow-up groups, there remains the possibility of follow-up bias.

CONCLUSION

Given that studies have suggested the resolution of some forms of impairment with age, our length of follow-up (11.2 years mean, with the shortest follow-up duration of 10.5 years) is especially useful. 61% of our patients had some form of neurodevelopment impairment—56% had SEN requirements, 22% had CP and 11% were blind, amongst others.

This is the hidden burden that neonatal surgeons bequeath to the community. Economic studies estimate that the American healthcare system spends $1 billion annually on the neurological care of patients with NEC.[2]

We now mention this study to the parents when we go through the consent process for surgery or later when prognosis is discussed weekly during the usual stormy postoperative period. We also keep this information in perspective when we contemplate the end of life scenario in selected cases.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.