Necrotising enterocolitis and spontaneous intestinal perforation — branching viva

Branch 1 — Diagnosis and staging

Examiner: "What is your working diagnosis, and how do you stage it?" [1]

The working diagnosis is necrotising enterocolitis. The bubbly bowel-wall lucency is pneumatosis intestinalis, the hallmark radiological sign, and the branching liver lucency is portal venous gas, a marker of severe disease. Using the modified Bell staging (Bell 1978, Walsh and Kliegman 1986), the combination of pneumatosis, portal gas, the systemic instability (apnoea) and the biochemical deterioration places this at Stage IIB — definite, moderate NEC. [1] [2]

I would move to Stage III if the infant developed shock with acidosis and peritonitis (IIIA, intact bowel) or a pneumoperitoneum (IIIB, perforation). The A/B suffix is the surgical discriminator: A is an intact bowel managed medically; B is perforation or gas requiring surgery. [1]

Examiner: "If the infant were instead a day-6 ELBW baby, well apart from an incidental pneumoperitoneum on a routine film, how would your diagnosis change?" [7]

That is the picture of spontaneous intestinal perforation (SIP), not NEC. The discriminator is that the SIP infant is haemodynamically stable, has no pneumatosis and no portal gas, and carries the risk profile of an ELBW infant exposed to early postnatal indomethacin combined with glucocorticoid. The distinction matters because SIP is often managed with peritoneal drainage alone and carries a markedly better prognosis than perforated NEC. [7] [11]

Branch 2 — Pathophysiology (if probed)

Examiner: "Walk me through the mechanism of NEC." [3]

NEC is the convergence of an immature gut, dysbiosis, and a substrate that feeds the wrong organisms, igniting an innate immune response that destroys the bowel wall. [3] The preterm mucosa is thin, the mucus sparse, secretory IgA low, and motility disordered; the immature enterocyte expresses high TLR4. Formula substrate and dysbiosis — overgrowth of pro-inflammatory Proteobacteria and loss of protective bifidobacteria — deliver PAMPs that activate TLR4, driving NF-κB-mediated inflammation and release of TNF-α, IL-6, IL-1β and iNOS. [3]

The downstream signs fall out of this. Epithelial apoptosis and tight-junction disruption let gas-forming organisms into the wall, producing pneumatosis intestinalis; intramural gas entering the mesenteric veins produces portal venous gas; and vasoconstriction with ischaemia produces full-thickness necrosis and perforation. [3]

Examiner: "And SIP?" [11]

SIP is mechanical, not inflammatory. A focal thinning or defect of the muscularis propria at the terminal ileum, compounded by early postnatal indomethacin (impaired mucosal restitution and mesenteric vasoconstriction) and postnatal glucocorticoid, produces a single, clean, isolated perforation. That is why there is no pneumatosis, no portal gas, and often minimal systemic sepsis. A common misconception is that antenatal steroids contribute — they do not; the signal is for postnatal indomethacin plus postnatal steroid. [11]

Branch 3 — Immediate management

Examiner: "What is your immediate management of this infant?" [3]

I would start the medical bundle immediately and not wait for further radiographic confirmation. Make the infant nil by mouth for 7 to 14 days, pass a large-bore nasogastric tube for free drainage and decompression, take blood, urine and CSF cultures, and start broad-spectrum antibiotics with anaerobic cover — a gram-negative agent plus metronidazole or clindamycin, adjusted to local resistance. [1] [3]

In parallel I would resuscitate: isotonic fluid boluses for hypotension, inotropes if needed, correction of acidosis and coagulopathy, and respiratory support for the apnoeic infant. The goal is a perfused, stable infant so that, if surgery becomes necessary, the infant is a viable candidate. [1]

Examiner: "When do you call the surgeon?" [1]

I call the surgeon at the surgical trigger, in parallel with resuscitation — not after the antibiotics are complete. The triggers are: a new pneumoperitoneum; a fixed and persistent loop with clinical deterioration (a loop that does not change on serial films is dead bowel); abdominal wall discolouration with shock; or a positive diagnostic paracentesis (brown or bile-stained fluid, bacteria on Gram stain). I would re-examine and re-image on a schedule to catch the slide. [1] [13]

Branch 4 — The surgical decision and the evidence

Examiner: "The infant perforates. Peritoneal drainage or laparotomy?" [4]

I would individualise, and I would cite the NET trial (Rees 2008), which randomised ELBW infants with perforated NEC to primary peritoneal drainage or laparotomy and found that PD did not improve survival or short-term outcomes versus laparotomy, and the follow-up (Rees 2010) showed PD did not stabilise ELBW infants with perforated bowel. [4] [5]

In practice I would favour primary peritoneal drainage as a bedside option for the smallest or most unstable infant — it evacuates the free air and meconium and can be definitive — and laparotomy with resection and stoma for the infant with established necrotic segments or who fails drainage. NEC totalis, pan-intestinal necrosis, has a grim prognosis and prompts a comfort-care discussion with the family. For a well ELBW infant with SIP, PD alone is often definitive. [4] [13]

Examiner: "Name the prevention bundle." [3]

The single most evidence-based prevention is own mother's milk, with a dose-dependent reduction in NEC; donor human milk also reduces NEC versus formula. I would add standardised feeding protocols (which reduce practice variation and NEC), antenatal corticosteroids, cautious feed advancement with avoidance of hypertonic feeds, and probiotics — though probiotic choice is strain-specific and region-specific, because probiotic sepsis has driven caution in ANZ and Canada. [7] [13]