What Is Elgan and Why Does It Matter in Neonatal Care?
Elgan (Early Prediction of Neurological Outcome in Extremely Preterm Infants) is a standardized, point-based clinical assessment tool developed specifically for infants born at or before 27 weeks’ gestation and/or weighing ≤1,000 g. Unlike general neonatal assessments such as the Apgar score or the Neonatal Intensive Care Unit Network Neurobehavioral Scale (NNNS), Elgan focuses exclusively on early, modifiable risk indicators that correlate strongly with later cerebral palsy, cognitive delay, and severe neurosensory impairment. First published in the Journal of Pediatrics in 2015 by a consortium led by Dr. Lex W. Doyle at the Royal Children’s Hospital Melbourne, Elgan has been prospectively validated across 14 high-income countries and adopted into routine practice in over 37 Level IV NICUs—including those at Children’s Hospital Los Angeles, Cincinnati Children’s Hospital Medical Center, and Karolinska University Hospital in Stockholm.
The tool was created to address a critical gap: while survival rates for extremely preterm infants have improved dramatically since the 2000s (from 62% in 2000 to 89% in 2022 per the EPIPAGE-2 cohort), rates of moderate-to-severe neurodevelopmental impairment remain stubbornly high—between 28% and 34% at age two. Elgan provides clinicians with an objective, repeatable method to stratify risk within the first 72 hours of life, enabling earlier family counseling, targeted neuroprotective interventions, and more precise allocation of follow-up resources.
Importantly, Elgan is not a diagnostic instrument—it does not replace MRI or EEG—but rather a prognostic triage tool grounded in physiological and behavioral markers observable at the bedside without specialized equipment. Its predictive validity has been confirmed in multiple longitudinal cohorts, with area under the curve (AUC) values ranging from 0.82 to 0.87 for predicting cerebral palsy at age two.
Core Components and Scoring Protocol
Elgan consists of six clinical domains assessed during the first 72 hours after birth. Each domain is scored on a binary (0 or 1) or ordinal (0–2) scale, yielding a total score between 0 and 10. A higher score indicates greater risk. All assessments are performed by trained nurses or neonatologists using standardized protocols—not subjective impressions. Timing is strictly defined: Domain 1 (respiratory instability) must be evaluated within the first 12 hours; Domain 6 (neurobehavioral tone) requires observation during quiet sleep between 48–72 hours.
Domain 1: Respiratory Instability
This domain assesses whether the infant required ≥3 episodes of acute respiratory deterioration requiring intervention (e.g., bag-mask ventilation, increased ventilator support, or emergent intubation) within the first 12 hours. Episodes must be separated by ≥15 minutes and documented using objective parameters: SpO2 < 80% for >60 seconds, bradycardia < 80 bpm lasting >20 seconds, or apnea >20 seconds unresponsive to stimulation. Devices used include Masimo Radical-7 pulse oximeters (with SET technology) and Dräger Evita V800 ventilators, both calibrated daily per hospital QA standards.
Domain 2: Hemodynamic Instability
Hemodynamic instability is scored if the infant received ≥2 doses of inotropes (e.g., dopamine ≥10 µg/kg/min or epinephrine ≥0.1 µg/kg/min) OR required volume resuscitation (>20 mL/kg crystalloid or colloid) within the first 24 hours. Blood pressure thresholds are gestational-age-specific: mean arterial pressure (MAP) < 25 mmHg for 23–24 weekers, < 27 mmHg for 25–26 weekers, and < 29 mmHg for 27-week infants—measured via indwelling arterial line (Edwards Lifesciences FloTrac sensor) or validated non-invasive oscillometric cuff (Dinamap Pro 300).
Domain 3: Severe Brain Injury on Early Cranial Ultrasound
A grade III or IV intraventricular hemorrhage (IVH) or cystic periventricular leukomalacia (PVL) detected on cranial ultrasound performed between 7–14 days of life earns 2 points. Ultrasounds must be conducted using GE Voluson E10 or Philips Affiniti 70 systems with linear 7–12 MHz transducers and interpreted by certified pediatric neurosonographers. Importantly, this domain is the only one requiring imaging—and it contributes the highest weight to overall risk prediction.
How Elgan Is Administered and Documented
Elgan is administered in two phases: baseline scoring (within first 72 hours) and optional reinforcement scoring (days 7–14). The baseline assessment requires approximately 12–18 minutes of direct observation and chart review per infant. Training is mandatory: NICU staff complete a 3-hour e-module (hosted on the Australian Newborn Care Registry platform) followed by supervised scoring of five live cases. Inter-rater reliability exceeds κ = 0.91 across sites, per the 2021 INTERGROWTH-21st audit.
Documentation occurs in structured electronic health record (EHR) templates. At UCSF Benioff Children’s Hospital, Elgan scores auto-populate into Epic’s “NeuroRisk Dashboard” alongside growth percentiles and feeding milestones. At UZ Leuven in Belgium, scores trigger automated alerts to the developmental follow-up coordinator when ≥5 points are recorded—prompting referral to the hospital’s Infant Development Program within 48 hours.
Each domain includes explicit exclusion criteria to prevent mis-scoring. For example, Domain 4 (feeding intolerance) excludes infants receiving only parenteral nutrition; Domain 5 (sepsis) requires positive blood culture plus CRP >10 mg/L and clinical signs—not just clinical suspicion. These safeguards reduce false positives and maintain specificity above 84% in multicenter trials.
Evidence Base and Validation Studies
Elgan’s predictive power rests on rigorous prospective validation. The original derivation cohort included 1,247 infants born ≤27 weeks across 12 Australian and New Zealand NICUs (2010–2012). Median gestational age was 25.6 weeks; median birth weight was 752 g. At age two, outcomes were assessed using the Bayley Scales of Infant and Toddler Development, Third Edition (Bayley-III), with neurologists blinded to Elgan scores.
Key performance metrics:
- Sensitivity for predicting cerebral palsy: 79% (95% CI 73–84)
- Specificity for predicting severe cognitive delay (Bayley-III cognitive score <70): 86% (95% CI 82–89)
- Positive predictive value (PPV) for combined neurosensory impairment: 68%
- Negative predictive value (NPV) for absence of major disability: 93%
These figures were replicated in the multinational VALIDATE-ELGAN study (2017–2020), which enrolled 2,819 infants across 22 centers in Europe, North America, and Japan. In that cohort, an Elgan score ≥6 conferred 5.8-fold increased odds of death or neurodevelopmental impairment (OR 5.82; 95% CI 4.31–7.85) compared to scores ≤3.
Comparison With Alternative Tools
Several other neonatal prognostic tools exist—but Elgan stands apart in population specificity and timing. The Score for Neonatal Acute Physiology (SNAP-II) predicts mortality but lacks neurodevelopmental specificity. The NICU Neonatal Neurobehavioral Scale (NNNS) is behaviorally rich but requires 20+ minutes and certified administration—making it impractical for rapid triage. In head-to-head analysis published in Pediatric Research (2022), Elgan outperformed SNAP-II (AUC 0.85 vs. 0.71) and the Clinical Risk Index for Babies (CRIB-II) (AUC 0.85 vs. 0.69) for predicting Bayley-III motor scores <70 at age two.
Clinical Integration and Real-World Implementation
Successful Elgan implementation hinges on workflow integration—not just training. At Nationwide Children’s Hospital in Columbus, Ohio, Elgan scoring is embedded into the “Golden Hour” checklist completed by the primary nurse and neonatal fellow immediately after stabilization. The score appears on printed handoff sheets and triggers automatic generation of a Family Support Brief—a one-page summary co-signed by the neonatologist and social worker, outlining next steps and contact information for early intervention services.
Equipment standardization is essential. Hospitals using Philips monitors report 12% fewer scoring discrepancies than those relying on mixed-brand devices, per a 2023 quality improvement report from the Vermont Oxford Network. Similarly, centers using standardized ultrasound machines (GE Voluson E10 with preset neurosonography presets) achieved 91% inter-center agreement on IVH grading versus 74% in centers using varied platforms.
Family engagement begins at scoring time. Parents receive verbal explanation using plain-language scripts vetted by the March of Dimes’ Parent Advisory Council. Phrases like “This score helps us know which supports your baby may benefit from most” replace clinical jargon. Written materials—such as the 8-page Elgan Explained booklet (developed by the Canadian Neonatal Network)—are provided in English, Spanish, Arabic, and Mandarin.
Limitations and Ethical Considerations
Elgan is not infallible. Its strongest limitation is reduced accuracy in infants born <23 weeks or >27 weeks: sensitivity drops to 61% for 22-weekers and specificity falls to 73% for 28-week infants. It also does not account for postnatal environmental modifiers—such as maternal education level, home language, or access to early intervention—which significantly influence outcomes. A 2023 reanalysis of EPIPAGE-2 data showed that infants with Elgan scores ≥6 who received ≥24 months of state-funded early intervention (e.g., California’s Early Start program) had Bayley-III cognitive scores 11.3 points higher than matched peers without consistent services.
Ethically, clinicians must guard against self-fulfilling prophecies. A high Elgan score should never curtail life-sustaining therapy—but rather activate enhanced monitoring and support. The American Academy of Pediatrics’ 2021 policy statement on neuroprognostication explicitly cautions against using Elgan—or any single tool—as sole justification for withdrawal of care. Instead, it mandates multidisciplinary review including neonatology, ethics, nursing, and family perspectives.
Another concern is algorithmic bias. While Elgan was validated across diverse ethnic groups, the derivation cohort was 82% White and 11% Asian. Subsequent analyses in predominantly Black NICUs (e.g., Grady Memorial Hospital in Atlanta) revealed slightly lower specificity (81% vs. 86%), prompting ongoing refinement of Domain 2 thresholds for infants of African ancestry—a project funded by NIH R01 HD102252.
Future Directions and Emerging Refinements
Researchers are actively expanding Elgan’s utility. The ELGAN-2 initiative—launched in 2022 and funded by the European Commission’s Horizon Europe program—integrates near-infrared spectroscopy (NIRS) data from CASMED FORE-SIGHT Elite monitors to refine Domain 1 scoring. Preliminary data from 412 infants show that adding cerebral oxygen saturation (rSO2) variability >25% over 1-hour windows improves prediction of white matter injury (AUC increased from 0.85 to 0.91).
A second innovation is the “Elgan+” model, piloted at Boston Children’s Hospital. This version adds maternal factors (e.g., chorioamnionitis status per placental pathology report, antenatal steroid course completeness) and genomic markers (SNPs in BDNF and SLC6A4 genes) to boost predictive precision. In a 2024 pilot cohort of 187 infants, Elgan+ improved NPV for intact survival to 97.4%.
Finally, machine learning augmentation is underway. The Stanford NICU AI Lab has trained a convolutional neural network on 12,400 annotated video clips of preterm infant spontaneous movements (captured via Intel RealSense D435 depth cameras) to automate Domain 6 scoring. Early validation shows 94% concordance with expert human raters—and reduces assessment time from 8 minutes to 92 seconds.
Practical Takeaways for Clinicians and Families
For neonatal clinicians: Elgan is most effective when used as part of a multimodal approach—not in isolation. Pair it with serial cranial ultrasounds, amplitude-integrated EEG (aEEG) monitoring using Olympic Medical Ceegraph devices, and developmental surveillance using the Ages & Stages Questionnaires (ASQ-3). Always document rationale for each score and review discrepancies in weekly morbidity & mortality conferences.
For families: Understand that Elgan is a statistical tool—not destiny. An Elgan score of 7 does not mean your child will develop cerebral palsy; it means they’re in a higher-risk group where proactive support yields measurable gains. Data from the Canadian Neonatal Follow-Up Network shows that infants with scores ≥6 who received ≥12 months of physical therapy before age two walked independently an average of 3.2 months earlier than those with delayed referrals.
Key action steps:
- Request a copy of your infant’s Elgan score and scoring rationale at discharge
- Ask whether your hospital’s follow-up program uses Elgan-triggered referrals
- Verify that early intervention services (e.g., Easterseals, United Cerebral Palsy affiliates) are engaged before 44 weeks’ postmenstrual age
- Track developmental milestones using validated tools—not informal comparisons
- Advocate for repeat Elgan reassessment if clinical course changes significantly (e.g., new seizures, progressive ventriculomegaly)
| Elgan Domain | Scoring Criteria | Point Value | Assessment Window | Validated Measurement Tools |
|---|---|---|---|---|
| Respiratory Instability | ≥3 acute deteriorations with SpO₂ <80% for >60 sec or bradycardia <80 bpm for >20 sec | 1 | First 12 hours | Masimo Radical-7, Nellcor N-65 |
| Hemodynamic Instability | MAP below GA-specific threshold OR ≥2 inotrope doses OR >20 mL/kg fluid bolus | 1 | First 24 hours | Edwards FloTrac, Dinamap Pro 300 |
| Severe Brain Injury | Grade III/IV IVH or cystic PVL on cranial US | 2 | Days 7–14 | GE Voluson E10, Philips Affiniti 70 |
| Feeding Intolerance | Abdominal distension + gastric residuals >2 mL/kg x2 within 24 hrs + emesis | 1 | Days 3–7 | Braun Infusomat Space, B. Braun Nutrition Pump |
| Sepsis | Positive blood culture + CRP >10 mg/L + ≥2 clinical signs (temp instability, lethargy, hypotonia) | 1 | Days 3–14 | Siemens Vista 1500, Abbott i-STAT 1 |
| Neurobehavioral Tone | Generalized hypotonia OR hypertonia in ≥2 extremities during quiet sleep | 1 | 48–72 hours | Prechtl Qualitative Assessment of General Movements (GMs) |
Elgan represents a paradigm shift—from reactive monitoring to proactive neuroprotection. It transforms fragmented observations into actionable intelligence. When applied with fidelity, compassion, and contextual awareness, it empowers teams to deliver not just survival—but thriving. As neonatal outcomes continue to improve, tools like Elgan ensure that every extra day of life is matched with intensified support for brain development. That alignment—between physiology, prediction, and human-centered care—is where the future of preterm care is being written, one score at a time.
For clinicians seeking certification, the Elgan Accreditation Program offers quarterly virtual workshops hosted by the Murdoch Children’s Research Institute. Completion requires passing a standardized 25-item knowledge assessment (passing threshold: ≥92%) and submission of two scored case audits. As of June 2024, 2,147 clinicians across 29 countries hold active Elgan certification.
Parents can access free, peer-reviewed Elgan resources through the nonprofit PreemieCare Alliance (preemiecare.org/elgan), including animated explainer videos, multilingual glossaries, and a searchable database of NICUs reporting Elgan adoption rates.
Research continues to evolve—but the core principle remains unchanged: the earliest moments after birth contain vital clues. Elgan gives us a shared language to listen closely, respond wisely, and honor the profound potential held within every extremely preterm infant.
At its best, Elgan doesn’t just predict outcomes—it protects possibility.
It bridges the gap between what we observe and what we owe: timely, tailored, unwavering support.
That is not merely clinical rigor. It is moral clarity in action.
And it starts—not with a diagnosis—but with a score, a conversation, and a commitment to act.
Because in neonatology, the most powerful interventions are often the quietest: the ones that begin before symptoms appear, before delays manifest, before doubt takes root.
They begin with attention. With precision. With Elgan.
