Introduction: The Challenge of Cerebral Autoregulation in the Preterm Brain
Extremely preterm infants, born before 29 weeks of gestation, face a precarious transitional period during their first days of life. One of the most significant clinical challenges in the Neonatal Intensive Care Unit (NICU) is the management of cerebral hemodynamics. Unlike full-term infants or adults, extremely preterm neonates often lack robust cerebral autoregulation, making their brain perfusion highly dependent on systemic blood pressure and arterial oxygen content. This vulnerability places them at high risk for germinal matrix-intraventricular hemorrhage (IVH) and periventricular leukomalacia (PVL), both of which are strongly associated with long-term neurodevelopmental impairments.
While peripheral oxygen saturation (SpO2) is routinely monitored, it provides an incomplete picture of tissue-level oxygen delivery, particularly in the brain. Near-infrared spectroscopy (NIRS) has emerged as a non-invasive tool to measure regional cerebral oxygen saturation (rcSO2), offering a more direct window into the balance between oxygen delivery and consumption. However, despite its potential, the clinical utility of NIRS-guided treatment has remained a subject of debate due to variability in device manufacturers, sensor types, and the lack of standardized intervention protocols.
The Rational for the Study: Standardizing Cerebral Oximetry
A recent randomized clinical trial published in JAMA Network Open (2026) sought to address these gaps. The study, led by Jani et al., investigated whether a dedicated treatment guideline, when combined with a specific NIRS device manufacturer (Nonin Medical Inc) and neonatal-specific sensors, could improve the stability of cerebral oxygenation in this fragile population. By narrowing the technological variables and providing clinicians with a clear roadmap for intervention, the researchers aimed to determine if the burden of hypoxia and hyperoxia could be significantly mitigated.
Study Design and Methodology
This single-blinded, two-arm randomized clinical trial was conducted between October 2021 and July 2024. The study spanned five tertiary NICUs across Australia, New Zealand, and the United States, providing a multi-center perspective on the intervention’s efficacy.
Participants and Stratification
The trial included 100 infants born at less than 29 weeks’ gestation who were younger than six hours at the time of enrollment. To ensure balanced groups, random allocation (1:1) was stratified by gestational age (sub-groups of <26 weeks and ≥26 weeks) and by study site. Of the 149 infants screened, 50 were analyzed in the intervention group and 50 in the standard care (control) group. The median gestational age was 27 weeks, and the median birth weight was 883 grams, representing a high-risk cohort of extremely low birth weight (ELBW) infants.
The Intervention: A Targeted Treatment Guideline
Infants in the intervention group were monitored using NIRS, and clinicians were provided with a standardized treatment guideline. The target range for cerebral oxygenation was set at 65% to 90%. If the readings fell outside this range, clinicians were directed to follow a structured physiological algorithm. This algorithm typically involves assessing and adjusting factors such as:
1. Fractional inspired oxygen (FiO2) to address hypoxia or hyperoxia.
2. Mean arterial pressure (MAP) to ensure adequate cerebral perfusion pressure.
3. Carbon dioxide levels (PaCO2), which act as a potent cerebral vasodilator or vasoconstrictor.
4. Hemoglobin levels and cardiac output, ensuring sufficient oxygen-carrying capacity.
In contrast, the control group underwent blinded cerebral oximetry. While the data were recorded for analysis, the bedside clinicians were not privy to the NIRS values, and treatment was guided solely by standard clinical monitoring, such as SpO2, heart rate, and blood pressure.
Primary and Secondary Endpoints
The primary outcome was the “burden” of cerebral hypoxia and hyperoxia during the first five days (120 hours) after birth. This burden was expressed as percentage hours—a composite measure of the duration and depth of oxygenation deviations outside the 65% to 90% target range. Secondary outcomes included mortality, common neonatal morbidities (such as IVH, necrotizing enterocolitis, and retinopathy of prematurity), and safety concerns, specifically NIRS-related skin injury.
Key Findings: A Dramatic Reduction in Oxygenation Instability
The results of the trial demonstrated a profound statistical and clinical difference between the two groups.
Primary Outcome Results
The intervention group experienced a significantly lower median burden of cerebral hypoxia and hyperoxia compared to the control group. Specifically, the intervention group showed a median burden of 5.7% hours (IQR, 2.8% to 15.0%), whereas the standard care group showed a median burden of 39.6% hours (IQR, 6.5% to 82.3%).
After adjusting for stratification factors, the study reported a 42.8% reduction in the burden of oxygenation instability (95% CI, 35.6% to 53.3%; P < .001). This suggests that providing clinicians with real-time cerebral oxygenation data and a structured response plan allows for much tighter control of brain oxygenation than standard care alone.
Secondary Outcomes and Safety
Importantly, the study found that the intervention was safe. There were no significant differences in NIRS-related skin injuries between the groups, addressing a common concern regarding the prolonged use of sensors on the fragile skin of extremely preterm infants.
Regarding clinical outcomes, mortality and morbidities before discharge were comparable between the intervention and control groups. While the study was not powered to detect differences in these long-term clinical outcomes, the stability of these metrics suggests that the intervention does not introduce unintended harm. The median duration of monitoring was approximately 115 hours in both groups, ensuring that the data captured the most critical window of transitional physiology.
Expert Commentary: Mechanistic Insights and Clinical Implications
The findings of this trial represent a significant step forward in neonatal neuroprotection. The massive reduction in the burden of hypoxia and hyperoxia (from nearly 40% of the time to less than 6%) highlights the limitations of relying solely on systemic markers like SpO2.
The Importance of Device Standardization
One of the unique strengths of this study was the use of a single NIRS manufacturer and sensor type. Previous trials, such as the SafeBoosC-III trial, have sometimes shown mixed results, which some experts attribute to the inherent variability between different NIRS technologies. By standardizing the equipment, Jani et al. have provided a more controlled look at the efficacy of the treatment guideline itself. This suggests that for NIRS to be effective in clinical practice, NICUs may need to adopt specific, validated hardware-software combinations rather than treating all NIRS devices as interchangeable.
Biological Plausibility
The biological plausibility for this intervention is strong. Cerebral hypoxia is a known precursor to energy failure and cellular apoptosis in the developing brain, while hyperoxia can lead to oxidative stress and the formation of free radicals, which are particularly damaging to pre-oligodendrocytes. By maintaining the “Goldilocks zone” of 65-90% saturation, clinicians can theoretically prevent the fluctuations that lead to reperfusion injury and hemorrhage.
Limitations and Future Directions
Despite the impressive reduction in oxygenation burden, several questions remain. The trial was relatively small (100 infants), and while it achieved its primary physiological endpoint, it was not designed to assess long-term neurodevelopmental scores at two years of age. Skeptics may argue that “stabilizing a number” on a monitor is only meaningful if it translates to better functional outcomes for the child. Furthermore, the intensive nature of following a NIRS-guided guideline requires significant nursing and medical staff engagement, which may vary across different clinical settings.
Conclusion: Moving Toward Precision Neonatology
In conclusion, the study by Jani et al. provides compelling evidence that cerebral oximetry-guided treatment, supported by a dedicated guideline and standardized technology, significantly improves cerebral oxygenation stability in extremely preterm infants. The dramatic 42.8% reduction in hypoxia and hyperoxia burden suggests that our current standard of care may be leaving infants exposed to significant periods of cerebral oxygenation imbalance that go undetected by traditional monitoring.
While larger multicenter trials are necessary to confirm whether this physiological stabilization leads to improved survival without neurodevelopmental impairment, this trial establishes a clear methodology for future research. For now, these findings support the consideration of NIRS as a valuable adjunct in the NICU, moving us closer to a model of precision medicine where the individual needs of the preterm brain are monitored and addressed in real-time.
Trial Registration and Funding
This study was registered with the Australian New Zealand Clinical Trials Registry (ACTRN12621000778886). The trial was supported by various clinical research grants and institutional funds across the participating sites in Australia, New Zealand, and the US.
References
1. Jani PR, Goyen TA, Balegar KK, et al. Cerebral Oximetry-Guided Treatment and Cerebral Oxygenation in Extremely Preterm Infants: A Randomized Clinical Trial. JAMA Netw Open. 2026;9(2):e2557620. doi:10.1001/jamanetworkopen.2025.57620.
2. Hyttel-Sorensen S, Pellicer A, Alderliesten T, et al. Cerebral near infrared spectroscopy oximetry in extremely preterm infants: phase II randomised clinical trial. BMJ. 2015;350:g7635.
3. Hansen ML, Pellicer A, Greisen G, et al. Cerebral oximetry monitoring in extremely preterm infants: a systematic review and meta-analysis. Seminars in Fetal and Neonatal Medicine. 2023;28(1):101416.
4. Wong FY, Leung TS, Austin T, et al. Impaired autoregulation in preterm infants identified by using spatially resolved spectroscopy. Pediatrics. 2008;121(3):e604-e611.