Necrotizing enterocolitis (NEC) — neonatal / preterm Bell-staged
NEW Phase C dossier — authored 2026-05-15 for shard-5-obped-id wave 4. Covers necrotizing enterocolitis (NEC) in neonates / preterm infants — the most common surgical GI emergency in NICU; ~ 7-10% of VLBW infants; mortality ~ 20-30% surgical NEC. Bell modified staging (Walsh & Kliegman 1986; AAP + JNPM 2024 update) drives medical-vs-surgical decision branching. Manifest field intentionally blanked (manifest: "") per shard-5 precedent on peds.febrile-infant.core.v1 + id.influenza.core.v1 + id.neonatal-sepsis.early-late.v1 — seed manifest authoring at prisma/seed/manifests/peds.nec.v1.{ts,atoms.ts} is out-of-shard scope and deferred to a future shard once the manifest + atoms cycle ships. Audit may report broken_pointers cleared but next-tier "missing manifest pointer" until manifest lands. Refined Phase-C-wave-4 pattern: 3-file new-dossier batch (TS + brief + research bundle) WITHOUT touching _registry.ts — registration will be picked up in a wave-roll-up commit (parallel-agent staging conflict avoidance per shard refined pattern). Distinct from id.neonatal-sepsis.early-late.v1 (parent sepsis context; bidirectional routing with carryover of CBC, CRP, blood culture, current empiric regimen, gestational age, ventilator status, abdominal exam, Bell stage). NEC owns Bell-staged GI emergency; neonatal-sepsis owns parent neonatal-sepsis context. Sibling differentiation explicitly encoded. Phenotype matrix (6-axis Bell-stage × gestational-age × age-at-onset × NEC-vs-SIP × management-arm × outcome — 960 cells collapsed to 10 anchor combinations) encoded indirectly via regimen_axes.nec_empirics_by_bell_stage.steps (stage_i_ii_medical / stage_iii_severe_or_surgical / vasoactive_neonate) + severity_triggers (10 phenotype-specific triggers) + setting playbooks (icu = NICU / inpatient = surgical-service or level-2-nursery / outpatient = peds-GI + peds-surgery + nutrition + neurodev follow-up). First-class TS phenotype field is schema-blocked. Severity triggers (10): nec_stage_iiib_with_perforation (life_threatening — Walsh Kliegman 1986; pneumoperitoneum → emergent peritoneal drain vs laparotomy per weight + stability), nec_stage_iiia_severe_no_perforation (life_threatening — ventilator + hypotension + DIC + neutropenia; surgical readiness), nec_stage_ii_with_pneumatosis_intestinalis (severe — definite NEC; full antibiotic course + NPO 7-14 d), nec_with_portal_venous_gas (severe — Pediatric Radiology LR+ ~ 20; relative surgical indication), nec_in_term_or_late_preterm_infant_atypical (severe — broader DDx: malrotation/volvulus, CHD with mesenteric ischemia, sepsis with translocation), recurrent_nec_post_initial_resolution (severe — stricture / missed perforation), nec_with_short_bowel_syndrome_emerging (severe — post extensive resection; lifelong IFALD prevention), nec_stricture_post_medical_management (severe — ~ 10-30% post Stage II-III medical; surgical resection), sip_vs_nec_differential_clarification (moderate — Gordon 2007; early < day 7 + no pneumatosis + indomethacin/steroids → SIP > NEC), human_milk_feeding_protocol_prevention_eligible (moderate — Quigley & McGuire Cochrane 2019; human milk reduces NEC ~ 50%; lactoferrin + probiotics adjuncts). Bayesian linkage (per §5.5.2): pre-test priors documented in _research-bundles/peds.nec.v1.md — NEC overall ~ 1-3% NICU admissions; VLBW ~ 7-10%; extreme preterm < 28 wk ~ 10-15%; term ~ 0.1% (atypical; consider underlying anomaly). Key LRs: pneumatosis intestinalis on KUB LR+ ~ 50; pneumoperitoneum essentially diagnostic of perforation; portal venous gas LR+ ~ 20; bloody stools + distention + thrombocytopenia in preterm combined LR+ ~ 5-8; fixed loop > 24 h LR+ ~ 10 for progression; abdominal-wall erythema LR+ ~ 8 for advanced NEC. Conditional dependencies modeled: pneumatosis-on-KUB ↔ onset-timing coupling (SIP suspicion if < day 7 + no pneumatosis), bloody-stools ↔ feeding-stage coupling, thrombocytopenia ↔ sepsis/Candida-coinfection coupling, CRP-trajectory ↔ surgical-indication coupling. Decision thresholds: T_treat_medical low (NPO + IV abx + serial KUB at first Stage I suspicion in preterm); T_treat_surgical at Stage IIIB OR clinical deterioration OR fixed mass/loop OR portal venous gas (relative). Cross-dossier routing: id.neonatal-sepsis.early-late.v1 (bidirectional), id.candidemia.core.v1 (Candida coinfection), peds.aki.v1 (drug nephrotoxicity / shock-related ATN), id.sepsis.peds.v1 (rare > 28 d late presentation). ROS/DDx LR seed data NOT touched (cross-cutting; not in shard scope). Settings (4): prehospital (rare — most NEC presents in-hospital; inter-facility transport of deteriorating preterm with pre-notify NICU; first-class "prehospital" DossierSetting value is schema-blocked, encoded implicitly via flow.entry_points), NICU = "icu" in dossier-setting vocabulary (primary venue — all Stage I-III), pediatric surgical service = "inpatient" (Stage IIIB or rapidly deteriorating Stage IIIA → peritoneal drain or laparotomy ± resection ± ostomy; post-op NICU return; level-2 nursery for stable late-stage Stage I-II with feeding-advance), outpatient peds-GI + peds-surgery + nutrition + neurodev follow-up. Drug guidance grounded in Walsh & Kliegman 1986 + AAP NEC clinical guidance + Cantey 2018 stewardship + IDSA candidiasis 2016 + Mermel IDSA CRBSI 2009 + Solomkin IDSA intra-abdominal 2010. RxCUIs referenced (all RxNav-live-verified 2026-05-25 as TTY=IN ingredients except L-AmB=PIN): ampicillin (733), gentamicin (1596450), vancomycin (11124), cefepime (20481), meropenem (29561), piperacillin-tazobactam (74169), metronidazole (6922), micafungin (325887), liposomal amphotericin B (236594), epinephrine (3992), norepinephrine (7512), dopamine (3628), hydrocortisone (5492). NOTE: prior codes 4921/477391/1665005/358258/1721538/5489 were fabricated/wrong (resolved to invalid OR levofloxacin/ceftriaxone/bortezomib/hydrocodone) and were corrected via RxNav reverse-lookup. Open gaps: (1) Phenotype matrix not first-class TS field — schema-blocked. (2) Bayesian LR seed data not encoded — lives in narrative + research bundle only this pass; ROS/DDx seed edit cross-cutting. (3) Prehospital not a DossierSetting value — schema-blocked. (4) calc.bell_nec_stage not yet registered in clinical-tools-registry.ts — pending registry addition (could be first-class Bell-staging calculator). (5) Manifest file not authored this pass — shard precedent for manifest: "" with seed deferred. (6) Co-located test file (peds.nec.v1.test.ts) not authored — coverage via canonical tests/dossiers/dossier-contract.test.ts only. (7) _registry.ts NOT modified this commit — refined Phase-C-wave-4 pattern; registration will be picked up in a wave-roll-up commit. (8) Walsh & Kliegman 1986 + Bell 1978 + Gordon 2007 SIP + Pediatric Radiology pneumatosis + Quigley McGuire Cochrane 2019 + Pammi Suresh Cochrane 2020 + ESPGHAN 2020 probiotic + Pichler 2014 SBS/IFALD PMIDs deferred to next research:pubmed loop. Status declared PLANNED with manifest: "" matching audit-resolved actual_status — audit gate honored. Per shard precedent, this is acceptable for new Phase C dossiers awaiting manifest authoring in a future shard.
Entry points (7)
- symptomPreterm neonate with feeding intolerance (residuals, emesis), abdominal distention, occult/bloody stools (Walsh & Kliegman 1986; AAP NEC guidance)feeding_intolerance_with_distention_preterm
- symptomPreterm neonate with abdominal wall erythema / induration / palpable mass — advanced NEC (Stage IIB+) featuresabdominal_wall_erythema_or_induration_neonate
- lab_abnormalityPreterm neonate with declining platelets + metabolic acidosis (Stage IIB criteria; Walsh & Kliegman 1986)thrombocytopenia_with_metabolic_acidosis_neonate
- imagingPneumatosis intestinalis on KUB — pathognomonic of definite NEC (Stage IIA+) (Walsh & Kliegman 1986; LR+ ~ 50)pneumatosis_intestinalis_on_kub
- imagingPneumoperitoneum on KUB (free air under diaphragm, football sign, Rigler sign) — Stage IIIB surgical emergencypneumoperitoneum_on_kub_or_us
- imagingPortal venous gas on KUB or abdominal US — LR+ ~ 20 for NEC; relative surgical indicationportal_venous_gas_on_imaging
- demographicVLBW (< 1500 g) preterm post-feeding-advance week 2-4 of life — high-pretest NEC cohort (~ 7-10% incidence)vlbw_preterm_in_nicu_post_feeding_advance
Required inputs (25)
- gestational_age_weeksrequireddemographic • used at CONTEXTExtreme preterm < 28 wk vs very preterm 28-32 wk vs late preterm 32-36 wk vs term ≥ 37 wk drives pretest probability + surgical decision (peritoneal drain vs laparotomy) + atypical-presentation broader DDx in term
- birth_weight_gramsrequireddemographic • used at CONTEXTVLBW < 1500 g and ELBW < 1000 g cohorts drive NEC pretest probability ~ 7-10%; weight-based dosing for all neonatal drugs; ELBW favours peritoneal drain over laparotomy in perforation
- chronologic_age_in_daysrequireddemographic • used at FRAMEOnset < day 7 raises SIP suspicion (no pneumatosis + indomethacin/steroid exposure); onset week 2-4 is classic NEC; onset > 4 wk is uncommon classic NEC
- temperaturerequiredvital • used at RED_FLAGSHypothermia in preterm with NEC is concerning for sepsis coinfection; fever rare in NEC alone
- hr_neonaterequiredvital • used at CONTEXTBradycardia + apnoea episodes in Stage IIIA NEC; tachycardia in early NEC with dehydration
- rr_neonaterequiredvital • used at CONTEXTApnoea / respiratory failure in Stage IIIA NEC drives ventilator-dependence
- spo2_neonaterequiredvital • used at CONTEXTHypoxaemia → intubation decision in advanced NEC
- sbp_neonaterequiredvital • used at RED_FLAGSHypotension in Stage IIIA-B NEC (SBP < gestational-age threshold) drives vasoactive
- feeding_status_and_advance_historyrequiredhistory • used at CONTEXTMother's-own-milk vs donor-human-milk vs formula; standardised feeding protocol; recent feeding-advance episode drives NEC pretest probability
- indomethacin_or_steroid_recent_exposurerequiredhistory • used at CONTEXTRecent indomethacin (PDA management) OR antenatal/postnatal steroids → SIP > NEC differential (Gordon 2007)
- central_line_or_tpn_durationrequiredhistory • used at CONTEXTPICC / umbilical line / surgical drain + TPN > 5-7 d in preterm → candidemia risk (IDSA candidiasis 2016)
- prior_antibiotic_exposure_daysrequiredhistory • used at CONTEXTProlonged empiric antibiotics > 5-7 d in preterm is a NEC risk factor (Cantey 2018 PMID 30172430 dose-response)
- cbc_with_diff_neonaterequiredlab • used at INITIAL_WORKUPWBC < 5K or > 20K + neutropenia + thrombocytopenia < 100K supports Stage IIB-III NEC; serial trend more informative than single value
- blood_culturerequiredlab • used at INITIAL_WORKUPSepsis coinfection workup; ≥ 1 mL minimum per bottle (Schelonka 1996); positive culture at 36-48 h LR+ > 100
- crprequiredlab • used at INITIAL_WORKUPCRP > 10 mg/L LR+ ~ 3-5; serial 24 + 48 h pair more informative than single value; rising CRP despite ≥ 48 h antibiotics is a relative surgical indication
- metabolic_panel_neonaterequiredlab • used at INITIAL_WORKUPMetabolic acidosis (HCO3 < 18) supports Stage IIB; hyponatremia + hypoglycemia common; AKI marker (BUN / creatinine elevation)
- lactate_neonaterequiredlab • used at INITIAL_WORKUPLactate > 4 mmol/L suggests perfusion failure / bowel ischaemia; serial trend for resuscitation response + surgical decision
- platelets_neonaterequiredlab • used at INITIAL_WORKUPThrombocytopenia < 100K supports Stage IIB; declining trend raises Candida coinfection suspicion in TPN + broad-abx preterm
- coag_panel_neonaterequiredlab • used at INITIAL_WORKUPDIC in Stage IIIA NEC; INR / aPTT / fibrinogen / D-dimer baseline + serial
- abg_or_vbg_neonaterequiredlab • used at INITIAL_WORKUPMetabolic acidosis severity + pCO2 + lactate; serial trend for resuscitation response
- stool_occult_blood_or_visible_bloodrequiredlab • used at INITIAL_WORKUPStage I criterion: occult blood; bloody stools more advanced; differential includes allergic colitis + swallowed maternal blood in non-NEC
- kub_xray_serialrequiredimaging • used at INITIAL_WORKUPSerial KUB q6-12h for Bell-staging progression detection; AP + left-lateral-decubitus views; pneumatosis intestinalis LR+ ~ 50 for NEC; pneumoperitoneum essentially diagnostic of perforation; portal venous gas LR+ ~ 20; fixed loop > 24 h LR+ ~ 10 for progression
- abdominal_ultrasoundimaging • used at BRANCHING_WORKUPBedside US for free fluid + bowel-wall thickening + perfusion + portal venous gas; complementary to KUB; especially useful in extremely preterm / unstable infants
- contrast_study_upper_or_lower_giimaging • used at FOLLOWUPReserved for post-recovery stricture evaluation (~ 10-30% after Stage II-III medical); NOT acute NEC workup
- fungal_culture_or_beta_d_glucanlab • used at BRANCHING_WORKUPTPN + broad-abx + preterm + thrombocytopenia / persistent fever despite antibacterial cover → fungal cultures + 1,3-β-D-glucan if Candida coinfection suspected (IDSA candidiasis 2016)
12-phase flow (12)
- 1FRAMEFrame the preterm-NICU NEC presentation: gestational age + birth weight + chronologic age in days + onset relative to feeding-advance. Onset < day 7 + no pneumatosis + recent indomethacin/steroids → SIP suspicion (pivot to differential); onset week 2-4 in VLBW preterm post-feeding-advance → classic NEC pretest 7-10%; onset in term/late-preterm → atypical, broaden DDx (malrotation/volvulus, CHD with mesenteric ischemia, sepsis with translocation).inputs: gestational_age_weeks, birth_weight_grams, chronologic_age_in_daysadvance: NEC vs SIP vs broader-DDx framing set; gestational-age cohort tagged
- 2ENTRYRecognise via clinical features (feeding intolerance + abdominal distention + occult/bloody stools) OR imaging (pneumatosis intestinalis, pneumoperitoneum, portal venous gas) OR lab (thrombocytopenia + metabolic acidosis). Stage I-IIA Bell features anchor entry; stage IIIB pneumoperitoneum is immediate surgical-emergency entry.inputs: feeding_status_and_advance_history, stool_occult_blood_or_visible_bloodadvance: NEC suspected by Bell-stage feature combination
- 3CONTEXTGestational age + birth weight + feeding status (mother's-own-milk vs donor-human-milk vs formula; recent advance) + indomethacin/steroid history (raises SIP differential) + central line / TPN duration (candidemia risk) + prior antibiotic exposure (Cantey 2018 dose-response).inputs: gestational_age_weeks, birth_weight_grams, feeding_status_and_advance_history, indomethacin_or_steroid_recent_exposure, central_line_or_tpn_duration, prior_antibiotic_exposure_daysadvance: Risk-factor profile + feeding context + medication history captured
- 4RED_FLAGSPneumoperitoneum on KUB (Stage IIIB — surgical emergency) → immediate pediatric surgery consult; hemodynamic instability (SBP < gestational-age threshold, lactate ≥ 5, vasoactive medication, capillary refill > 3 s, mottling) → ICU + max support; respiratory failure / apnoea + bradycardia (Stage IIIA features) → intubation + ventilator; bulging abdomen + fixed loop + abdominal-wall erythema + induration → advanced NEC (Stage IIB-III).inputs: temperature, sbp_neonate, spo2_neonateactions: protocol.septic_shockadvance: Red flags actioned; surgical / ICU escalation within minutes if Stage IIIB
- 5INITIAL_WORKUPNPO + NG decompression + IV access + bowel rest; CBC with diff (age-adjusted), CRP, blood culture (≥ 1 mL), metabolic panel, ABG/VBG, lactate, platelets, coag panel, stool occult / visible blood; KUB AP + left-lateral-decubitus views (pneumatosis, pneumoperitoneum, portal venous gas, fixed loop); empiric antibiotics within 1 h: Stage I-II → ampicillin + gentamicin (Gram-positive + Gram-negative + anaerobe baseline) OR pip-tazo; Stage III → pip-tazo OR meropenem + vancomycin (broader gram-negative + MRSA cover); + micafungin / L-AmB if Candida risk factors.inputs: cbc_with_diff_neonate, blood_culture, crp, metabolic_panel_neonate, lactate_neonate, platelets_neonate, coag_panel_neonate, abg_or_vbg_neonate, stool_occult_blood_or_visible_blood, kub_xray_serialactions: panel.cbc, panel.renal, panel.coag, panel.inflammation, panel.abgadvance: NPO + NG; cultures sent; empiric antibiotics in; KUB obtained + Bell-staged
- 6BRANCHING_WORKUPSource-directed: abdominal US for free fluid + bowel-wall thickening + portal venous gas (complementary to KUB); fungal cultures + 1,3-β-D-glucan if Candida risk (TPN + broad-abx + preterm + thrombocytopenia); CRBSI workup if central line — differential time-to-positivity (Mermel IDSA 2009); LP if sepsis coinfection with CNS features; serial KUB q6-12h for Bell-stage progression detection. SIP differential workup if onset < day 7 + no pneumatosis + recent indomethacin/steroids.inputs: abdominal_ultrasound, fungal_culture_or_beta_d_glucanactions: workup.crbsi, workup.bacterial_meningitisadvance: Source identified or empirically covered; serial imaging plan in place
- 7DIFFERENTIALSpontaneous intestinal perforation (SIP) — early < day 7 + no pneumatosis + indomethacin/steroid exposure (Gordon 2007 pivot). NEC in term infant → broader DDx (malrotation/volvulus → upper GI series; congenital heart disease with mesenteric ischemia → echo; sepsis with bacterial translocation). Hirschsprung-associated enterocolitis (history of delayed meconium passage). Allergic enterocolitis (milk-protein-induced). Volvulus / intussusception (late-preterm and term). Coexistence pairs: NEC + sepsis (most cases); NEC + AKI; NEC + DIC; NEC + Candidemia.advance: Mimics excluded or co-managed; SIP-vs-NEC pivot resolved if early-onset
- 8RISK_STRATIFICATIONBell modified staging (Walsh & Kliegman 1986): Stage I (suspected) → 48-72 h rule-out medical; Stage IIA (pneumatosis) → confirmed NEC + full 7-10 d course; Stage IIB (pneumatosis + acidosis + thrombocytopenia) → severe systemic involvement; Stage IIIA (advanced no-perforation; ventilator + hypotension + DIC + neutropenia) → ICU + max support + surgical readiness; Stage IIIB (pneumoperitoneum) → surgical emergency. Mortality: Stage I-IIA ~ 5%; Stage IIB ~ 15%; Stage III ~ 30-50%.inputs: kub_xray_serial, platelets_neonate, abg_or_vbg_neonate, sbp_neonateadvance: Bell stage assigned + mortality stratified + surgical-vs-medical pathway chosen
- 9TREATMENTMedical management for Stage I-II: NPO 7-14 d + NG decompression + IV crystalloid maintenance + TPN + central venous access + bowel rest; antibiotics 7-10 d (Stage I-II) — ampicillin + gentamicin OR pip-tazo OR meropenem (broader); + acyclovir if HSV features ≤ 28 d; + micafungin / L-AmB if Candida risk. Surgical management for Stage IIIB: pediatric surgery consult; peritoneal drain (extremely preterm < 1000 g, unstable) vs laparotomy (larger / definitive); resection of necrotic bowel ± ostomy formation. Antibiotic course 10-14 d for Stage III. Vasoactive — epinephrine 0.05-0.3 µg/kg/min OR dopamine 5-15 µg/kg/min OR norepinephrine 0.05-0.5 µg/kg/min; hydrocortisone 1 mg/kg q8h if catecholamine-resistant.inputs: birth_weight_grams, gestational_age_weeks, sbp_neonate, kub_xray_serialactions: protocol.septic_shockadvance: NPO + NG + antibiotics + fluid/TPN + surgical-vs-medical plan in place; vasoactive titrated if shock
- 10DISPOSITIONNICU for all NEC (Stage I-III); pediatric surgical service for Stage IIIB or rapidly deteriorating Stage IIIA; step-down to level-2 special care nursery only after antibiotic course complete + tolerating feed advances + Bell-staging resolved. PICU rare (most centres NICU-manages all neonates).inputs: sbp_neonate, spo2_neonateadvance: Level of care set; surgical service involved if Stage IIIB
- 11MONITORINGVitals q4h (q1h initially in ill neonate), abdominal girth q4h, NG output + stool output + emesis q4h, weight daily, lactate trend q2-4h until normalised, CBC + CRP + platelets at 24 + 48 h, blood culture follow-up at 24 + 48 h, vancomycin / gentamicin levels per pharmacy (trough or AUC), serial KUB q6-12h during acute phase, daily reassessment of antibiotic + duration with goal de-escalation per Cantey 2018 stewardship; nutrition + feeding-advance plan; developmental + feeding follow-up.inputs: cbc_with_diff_neonate, crp, blood_culture, platelets_neonate, kub_xray_serialactions: panel.cbc, panel.renal, panel.inflammationadvance: Response confirmed; antibiotics narrowed or completed; KUB resolution; tolerating feed advances
- 12FOLLOWUPOutpatient peds within 24-48 h of discharge for high-risk; 1-week visit for all; growth + feeding tracking + developmental milestones; peds-GI surveillance for stricture (~ 10-30% after Stage II-III medical; symptoms — feeding intolerance, distention, emesis post-recovery; contrast study + surgical resection if symptomatic stricture); peds-surgery follow-up for post-op (ostomy management, takedown timing); nutrition follow-up for SBS / IFALD if extensive resection (TPN management, lipid strategy, central line care, intestinal transplant evaluation if non-rehabilitatable); neurodevelopmental peds at 6 + 12 + 24 mo (Bayley III / ASQ-3 — preterm + NEC survivors at elevated risk); immunization catch-up per cause (PCV / Hib + hepatitis B); hearing screen if meningitis precipitant. Family education on return precautions; vaccination of family contacts (Tdap, influenza, COVID-19); breastfeeding support.advance: Outpatient plan documented; peds-GI + peds-surgery + nutrition + neurodev follow-up scheduled; family education delivered