Evaluation of a deuterated triarylmethyl spin probe for in vivo R2 ∗-based EPR oximetric imaging with enhanced dynamic range.

PURPOSE: In this study, we compared two triarylmethyl (TAM) spin probes, Ox071 and Ox063 for their efficacy in measuring tissue oxygen levels under hypoxic and normoxic conditions by R2 *-based EPR oximetry. METHODS: The R2 * dependencies on the spin probe concentration and oxygen level were calibrated using deoxygenated 1, 2, 5, and 10 mM standard solutions and 2 mM solutions saturated at 0%, 2%, 5%, 10%, and 21% of oxygen. For the hypoxic model, in vivo imaging of a MIA PaCa-2 tumor implanted in the hind leg of a mouse was performed on successive days by R2 *-based EPR oximetry using either Ox071 or Ox063. For the normoxic model, renal imaging of healthy athymic mice was performed using both spin probes. The 3D images were reconstructed by single point imaging and multi-gradient technique was used to determine R2 * maps. RESULTS: The signal intensities of Ox071 were approximately three times greater than that of Ox063 in the entire partial pressure of oxygen (pO2 ) range investigated. The histograms of the tumor pO2 images were skewed for both spin probes, and Ox071 showed more frequency counts at pO2 > 32 mm Hg. In the normoxic kidney model, there was a clear delineation between the high pO2 cortex and the low pO2 medulla regions. The histogram of high-resolution kidney oximetry image using Ox071 was nearly symmetrical and frequency counts were seen up to 55 mm Hg, which were missed in Ox063 imaging. CONCLUSION: As an oximetric probe, Ox071 has clear advantages over Ox063 in terms of sensitivity and the pO2 dynamic range.

A characterization of cardiac-induced noise in R2 * maps of the brain.

PURPOSE: Cardiac pulsation increases the noise level in brain maps of the transverse relaxation rate R2 *. Cardiac-induced noise is challenging to mitigate during the acquisition of R2 * mapping data because its characteristics are unknown. In this work, we aim to characterize cardiac-induced noise in brain maps of the MRI parameter R2 *. METHODS: We designed a sampling strategy to acquire multi-echo 3D data in 12 intervals of the cardiac cycle, monitored with a fingertip pulse-oximeter. We measured the amplitude of cardiac-induced noise in this data and assessed the effect of cardiac pulsation on R2 * maps computed across echoes. The area of k-space that contains most of the cardiac-induced noise in R2 * maps was then identified. Based on these characteristics, we introduced a tentative sampling strategy that aims to mitigate cardiac-induced noise in R2 * maps of the brain. RESULTS: In inferior brain regions, cardiac pulsation accounts for R2 * variations of up to 3 s-1 across the cardiac cycle (i.e., ∼35% of the overall variability). Cardiac-induced fluctuations occur throughout the cardiac cycle, with a reduced intensity during the first quarter of the cycle. A total of 50% to 60% of the overall cardiac-induced noise is localized near the k-space center (k < 0.074 mm-1 ). The tentative cardiac noise mitigation strategy reduced the variability of R2 * maps across repetitions by 11% in the brainstem and 6% across the whole brain. CONCLUSION: We provide a characterization of cardiac-induced noise in brain R2 * maps that can be used as a basis for the design of mitigation strategies during data acquisition.

Evaluation of Leading Smartwatches for the Detection of Hypoxemia: Comparison to Reference Oximeter.

Although smartwatches are not considered medical devices, experimental validation of their accuracy in detecting hypoxemia is necessary due to their potential use in monitoring conditions manifested by a prolonged decrease in peripheral blood oxygen saturation (SpO2), such as chronic obstructive pulmonary disease, sleep apnea syndrome, and COVID-19, or at high altitudes, e.g., during sport climbing, where the use of finger-sensor-based pulse oximeters may be limited. The aim of this study was to experimentally compare the accuracy of SpO2 measurement of popular smartwatches with a clinically used pulse oximeter according to the requirements of ISO 80601-2-61. Each of the 18 young and healthy participants underwent the experimental assessment three times in randomized order-wearing Apple Watch 8, Samsung Galaxy Watch 5, or Withings ScanWatch-resulting in 54 individual experimental assessments and complete datasets. The accuracy of the SpO2 measurements was compared to that of the Radical-7 (Masimo Corporation, Irvine, CA, USA) during short-term hypoxemia induced by consecutive inhalation of three prepared gas mixtures with reduced oxygen concentrations (14%, 12%, and 10%). All three smartwatch models met the maximum acceptable root-mean-square deviation (≤4%) from the reference measurement at both normal oxygen levels and induced desaturation with SpO2 less than 90%. Apple Watch 8 reached the highest reliability due to its lowest mean bias and root-mean-square deviation, highest Pearson correlation coefficient, and accuracy in detecting hypoxemia. Our findings support the use of smartwatches to reliably detect hypoxemia in situations where the use of standard finger pulse oximeters may be limited.

Non-invasive assessment of oxygenation status using the oxygen reserve index in dogs.

BACKGROUND: The oxygen reserve index (ORi) is a real-time, continuous index measured with multi-wavelength pulse CO-oximetry technology. It estimates mild hyperoxemia in humans, which is defined as a partial pressure of oxygen (PaO2) level between 100 and 200 mmHg. The objectives of this study were to assess the correlation between ORi and PaO2, as well as to determine its ability in detecting mild hyperoxemia in dogs. METHODS: This prospective observational study enrolled 37 anaesthetised and mechanically ventilated dogs undergoing elective procedures. Simultaneous measurements of ORi and PaO2 were collected, using a multi-wavelength pulse CO-oximeter with a probe placed on the dog's tongue, and a blood gas analyser, respectively. A mixed-effects model was used to calculate the correlation (r2) between simultaneous measurements of ORi and PaO2. The trending ability of ORi to identify dependable and proportional changes of PaO2 was determined. The diagnostic performances of ORi to detect PaO2 ≥ 150 mmHg and ≥ 190 mmHg were estimated using the area under the receiver operating characteristic curve (AUROC). The effects of perfusion index (PI), haemoglobin (Hb), arterial blood pH and partial pressure of carbon dioxide (PaCO2) on AUROC for PaO2 ≥ 150 mmHg were evaluated. RESULTS: A total of 101 paired measurements of ORi and PaO2 were collected. PaO2 values ranged from 74 to 258 mmHg. A strong positive correlation (r2 = 0.52, p < 0.001) was found between ORi and PaO2. The trending ability ORi was 90.7%, with 92% sensitivity and 89% specificity in detecting decreasing PaO2. An ORi value ≥ 0.53 and ≥ 0.76 indicated a PaO2 ≥ 150 and ≥ 190 mmHg, respectively, with ≥ 82% sensitivity, ≥ 77% specificity and AUROC ≥ 0.75. The AUROC of ORi was not affected by PI, Hb, pH and PaCO2. CONCLUSIONS: In anaesthetised dogs, ORi may detect mild hyperoxaemia, although it does not replace blood gas analysis for measuring the arterial partial pressure of oxygen. ORi monitoring could be used to non-invasively assess oxygenation in dogs receiving supplemental oxygen, limiting excessive hyperoxia.

Detailed statistical analysis plan for a secondary Bayesian analysis of the SafeBoosC-III trial: a multinational, randomised clinical trial assessing treatment guided by cerebral oximetry monitoring versus usual care in extremely preterm infants.

BACKGROUND: Extremely preterm infants have a high mortality and morbidity. Here, we present a statistical analysis plan for secondary Bayesian analyses of the pragmatic, sufficiently powered multinational, trial-SafeBoosC III-evaluating the benefits and harms of cerebral oximetry monitoring plus a treatment guideline versus usual care for such infants. METHODS: The SafeBoosC-III trial is an investigator-initiated, open-label, randomised, multinational, pragmatic, phase III clinical trial with a parallel-group design. The trial randomised 1601 infants, and the frequentist analyses were published in April 2023. The primary outcome is a dichotomous composite outcome of death or severe brain injury. The exploratory outcomes are major neonatal morbidities associated with neurodevelopmental impairment later in life: (1) bronchopulmonary dysplasia; (2) retinopathy of prematurity; (3) late-onset sepsis; (4) necrotising enterocolitis; and (5) number of major neonatal morbidities (count of bronchopulmonary dysplasia, retinopathy of prematurity, and severe brain injury). The primary Bayesian analyses will use non-informed priors including all plausible effects. The models will use a Hamiltonian Monte Carlo sampler with 1 chain, a sampling of 10,000, and at least 25,000 iterations for the burn-in period. In Bayesian statistics, such analyses are referred to as 'posteriors' and will be presented as point estimates with 95% credibility intervals (CrIs), encompassing the most probable results based on the data, model, and priors selected. The results will be presented as probability of any benefit or any harm, Bayes factor, and the probability of clinical important benefit or harm. Two statisticians will analyse the blinded data independently following this protocol. DISCUSSION: This statistical analysis plan presents a secondary Bayesian analysis of the SafeBoosC-III trial. The analysis and the final manuscript will be carried out and written after we publicise the primary frequentist trial report. Thus, we can interpret the findings from both the frequentists and Bayesian perspective. This approach should provide a better foundation for interpreting of our findings. TRIAL REGISTRATION: ClinicalTrials.org, NCT03770741. Registered on 10 December 2018.

The incidence, risk factors and outcomes of impaired cerebral autoregulation in aortic arch surgery: a single-center, retrospective cohort study.

BACKGROUND: Impairment of cerebral autoregulation (CA) has been observed in patients undergoing cardiopulmonary bypass (CPB), but little is known about its risks and associations with outcomes. The cerebral oximetry index (COx), which is a moving linear correlation coefficient between regional cerebral oxygen saturation (rScO2) and mean blood pressure (MAP), may reflect CA function. When COx approaches 1, it implies that CA is damaged, whereas the CA is functional when the COx value approaches 0. The objective of this study was to analyze the incidence and risks of impaired CA, based on COx assessment, in patients undergoing total aortic arch replacement under systemic moderate hypothermia and circulatory arrest of the lower body (MHCA). We also evaluated the association between impaired CA and patient outcomes. METHODS: One hundred and fifty-four adult patients who underwent total aortic arch replacement with stented elephant trunk implantation under MHCA at our hospital were retrospectively analyzed. Patients were defined as having new-onset impaired CA if pre-CPB COx < 0.3 and post-CPB COx > 0.3. Pre- and intraoperative factors were tested for independent association with impaired CA. Postoperative outcomes were compared between patients with normal and impaired CA. RESULTS: In our 154 patients, 46(29.9%) developed new-onset impaired CA after CPB. Multivariable analysis revealed a prolonged low rScO2 (rScO2 < 55%) independently associated with onset of impaired CA, and receiver operating charactoristic curve showed a cutoff value at 40 min (sensitivity, 89.5%; specificity, 68.0%). Compared with normal CA patients, those with impaired CA showed a significantly higher rates of in-hospital mortality and postoperative complications. CONCLUSIONS: Prolonged low rScO2 (rScO2 < 55%) during aortic arch surgery was closely related to onset of impaired CA. Impaired CA remained associated with the increased rates of postoperative complications and in-hospital mortality. TRIAL REGISTRATION: ChiCTR1800014545 with registered date 20/01/2018.

Enhanced SpO2 estimation using explainable machine learning and neck photoplethysmography.

Reflectance-based photoplethysmogram (PPG) sensors provide flexible options of measuring sites for blood oxygen saturation (SpO2) measurement. But they are mostly limited by accuracy, especially when applied to different subjects, due to the diverse human characteristics (skin colors, hair density, etc.) and usage conditions of different sensor settings. This study addresses the estimation of SpO2 at non-standard measuring sites employing reflectance-based sensors. It proposes an automated construction of subject inclusion-exclusion criteria for SpO2 measuring devices, using a combination of unsupervised clustering, supervised regression, and model explanations. This is perhaps among the first adaptation of SHAP to explain the clusters gleaned from unsupervised learning methods. As a wellness application case study, we developed a pillow-based wearable device to collect reflectance PPGs from both the brachiocephalic and carotid arteries around the neck. The experiment was conducted on 33 subjects, each under totally 80 different sensor settings. The proposed approach addressed the variations of humans and devices, as well as the heterogeneous mapping between signals and SpO2 values. It identified effective device settings and characteristics of their applicable subject groups (i.e., subject inclusion-exclusion criteria). Overall, it reduced the root mean squared error (RMSE) by 16%, compared to an empirical formula and a plain SpO2 estimation model.

Can machine learning methods be used for identification of at-risk neonates in low-resource settings? A prospective cohort study.

INTRODUCTION: Timely identification of at-risk neonates (ARNs) in the community is essential to reduce mortality in low-resource settings. Tools such as American Academy of Pediatrics pulse oximetry (POx) and WHO Young Infants Clinical Signs (WHOS) have high specificity but low sensitivity to identify ARNs. Our aim was assessing the value of POx and WHOS independently, in combination and with machine learning (ML) from clinical features, to detect ARNs in a low/middle-income country. METHODS: This prospective cohort study was conducted in a periurban community in Pakistan. Eligible live births were screened using WHOS and POx along with clinical information regarding pregnancy and delivery. The enrolled neonates were followed for 4 weeks of life to assess the vital status. The predictive value to identify ARNs, of POx, WHOS and an ML model using maternal and neonatal clinical features, was assessed. RESULTS: Of 1336 neonates, 68 (5%) had adverse outcomes, that is, sepsis (n=40, 59%), critical congenital heart disease (n=2, 3%), severe persistent pulmonary hypertension (n=1), hospitalisation (n=8, 12%) and death (n=17, 25%) assessed at 4 weeks of life. Specificity of POx and WHOS to independently identify ARNs was 99%, with sensitivity of 19% and 63%,respectively. Combining both improved sensitivity to 70%, keeping specificity at 98%. An ML model using clinical variables had 44% specificity and 76% sensitivity. A staged assessment, where WHOS, POx and ML are sequentially used for triage, increased sensitivity to 85%, keeping specificity 75%. Using ML (when WHOS and POx negative) for community follow-up detected the majority of ARNs. CONCLUSION: Classic screening, combined with ML, can help maximise identifying ARNs and could be embedded in low-resource clinical settings, thereby improving outcome. Sequential use of classic assessment and clinical ML identifies the most ARNs in the community, still optimising follow-up clinical care.

Provider Perceptions of Oxygenation Strategies for Critically Ill Trauma Patients With and Without Moderate-to-Severe Traumatic Brain Injury.

BACKGROUND: Hypoxia and hyperoxia (pulse oximetry [SpO2] > 96%) are associated with increased mortality in critically ill patients. However, provider practices regarding oxygenation in traumatic brain injury (TBI) patients are unknown. This study assesses views on oxygenation of critically ill trauma patients with and without TBI and how this varies between Neurological ICU (NeuroICU) and Surgical-Trauma ICU (STICU) providers. METHODS: This is a cross-sectional survey of Level I trauma center's NeuroICU and STICU providers. We used Likert scales, yes-no questions, and multiple-choice case-based scenarios to characterize provider views on oxygenation with descriptive statistics to characterize responses. Significant differences regarding TBI and non-TBI patients or NeuroICU and STICU providers were determined using Fisher's exact test and a P-value of .05. RESULTS: A total of 83 providers initiated the survey, and 53 providers completed it. Most providers identified a threshold SpO2 < 92% for the administration of supplemental oxygen in critically ill TBI patients. A total of 9% of providers "somewhat or completely agreed" that they were more likely to give supplemental oxygen to a critically ill trauma patient with TBI than one without TBI and the same SpO2. A total of 48% of providers selected an SpO2 < 90% as the point at which supplemental oxygen should be initiated in patients without TBI, compared to 27% of providers in patients with TBI (P < .01). This threshold for supplemental oxygen use varied by provider type for non-TBI patients, but not for TBI patients (30% NeuroICU and 69% STICU providers selected SpO2 < 90% in non-TBI, P < .05; 30% NeuroICU and 35% STICU providers selected SpO2 < 90% in TBI, P = .85). CONCLUSIONS: Critical care providers at UCHealth University of Colorado Hospital approach the oxygenation of critically ill trauma patients with and without TBI differently. Specifically, critical care respondents accepted a different lower oxygen saturation threshold for TBI and non-TBI patients. NeuroICU and STICU respondents differed in their threshold for the down-titration of supplemental oxygen. Targeted education for critical care providers may reduce these discrepancies and optimize oxygen use.

Increased Oxygen Saturation in Retinal Venules During Isometric Exercise Is Accompanied With Increased Peripheral Blood Flow in Normal Persons.

Purpose: A recent study has shown that an increase in the arterial blood pressure of approximately 10 mm Hg in healthy persons can increase the oxygen saturation in venules from the retinal periphery but not from the macular area. The purpose of the present study was to investigate whether a higher increase in blood pressure has further effects on oxygen saturations and whether this is accompanied with changes in retinal blood flow. Methods: In 30 healthy persons, oxygen saturation, diameter, and blood flow were measured in arterioles to and venules from the retinal periphery and the macular area. The experiments were performed before and during an experimental increase in arterial blood pressure of (mean ± SD) 18.3 ± 6.2 mm Hg. Results: A higher number of venules than arterioles branching from the temporal vascular arcades to the macular area was balanced by a smaller diameter of the venules. Isometric exercise induced significant contraction of both peripheral and macular arterioles (P < 0.01 for both comparisons) and significant increase in oxygen saturation in both peripheral and macular venules (P < 0.001 for both comparisons). This was accompanied with a significant increase in the blood flow in the peripheral arterioles and venules (P = 0.4 for both comparisons), but not in their macular counterparts (P > 0.06 for both comparisons). Conclusions: Increased systemic blood pressure leading to arterial contraction and increased venous oxygen saturation in the retina in normal persons can increase peripheral blood flow without significant effects on macular blood flow. This may contribute to explaining regional differences in the response pattern of retinal vascular disease.

Incidence of postoperative opioid-induced respiratory depression episodes in patients on room air or supplemental oxygen: a post-hoc analysis of the PRODIGY trial.

BACKGROUND: Supplemental oxygen (SO) potentiates opioid-induced respiratory depression (OIRD) in experiments on healthy volunteers. Our objective was to examine the relationship between SO and OIRD in patients on surgical units. METHODS: This post-hoc analysis utilized a portion of the observational PRediction of Opioid-induced respiratory Depression In patients monitored by capnoGraphY (PRODIGY) trial dataset (202 patients, two trial sites), which involved blinded continuous pulse oximetry and capnography monitoring of postsurgical patients on surgical units. OIRD incidence was determined for patients receiving room air (RA), intermittent SO, or continuous SO. Generalized estimating equation (GEE) models, with a Poisson distribution, a log-link function and time of exposure as offset, were used to compare the incidence of OIRD when patients were receiving SO vs RA. RESULTS: Within the analysis cohort, 74 patients were always on RA, 88 on intermittent and 40 on continuous SO. Compared with when on RA, when receiving SO patients had a higher risk for all OIRD episodes (incidence rate ratio [IRR] 2.7, 95% confidence interval [CI] 1.4-5.1), apnea episodes (IRR 2.8, 95% CI 1.5-5.2), and bradypnea episodes (IRR 3.0, 95% CI 1.2-7.9). Patients with high or intermediate PRODIGY scores had higher IRRs of OIRD episodes when receiving SO, compared with RA (IRR 4.5, 95% CI 2.2-9.6 and IRR 2.3, 95% CI 1.1-4.9, for high and intermediate scores, respectively). CONCLUSIONS: Despite oxygen desaturation events not differing between SO and RA, SO may clinically promote OIRD. Clinicians should be aware that postoperative patients receiving SO therapy remain at increased risk for apnea and bradypnea. TRIAL REGISTRATION: Clinicaltrials.gov: NCT02811302, registered June 23, 2016.

Perioperative monitoring of the oxygen reserve: where do we stand?

The Oxygen Reserve Index (ORi) is an advanced plethysmography-derived variable that may help to quantify the degree of hyperoxia in patients receiving supplemental oxygen administration. ORi is a (relative) indicator of the actual partial pressure of oxygen dissolved in arterial blood (PaO2). As such, it may help in the titration of oxygen administration or it may help to warn the clinician of a deterioration of oxygen status of the patient.In this issue of the journal, Fadel et al. provide a 'classical' clinical validation study by assessing the correlation between ORi and PaO2 in patients about to undergo open-heart surgery. Within the moderate hyperoxic range (100-200 mmHg PaO2), there is a sound correlation between ORi and PaO2. This editorial discusses the clinical implications of this validation study and elaborates on the possible role of ORi monitoring in addition to SpO2 (peripheral arterial oxygen saturation) monitoring alone.

[Pulse Oximetry Values in Preterm infants at 34-36 weeks postmenstrual age]. / Valores de Oximetría de Pulso en Prematuros a las 34-36 semanas post-concepcional.

Pulse oximetry (SpO2) is essential for guiding oxygen therapy in preterm infants. Data on SpO2 values before discharge are limited. OBJECTIVE: To establish SpO2 values in asymptomatic premature infants at 34, 35, and 36 weeks of postmenstrual age (PMA). SUBJECTS AND METHOD: Longitudinal, multicen ter study carried out from May 2018 to May 2019 in three neonatal intensive care units in Santiago, Chile (altitude 579 m) which included premature infants born ≤ 32 weeks of gestational age, healthy, with clinical stability, and without respiratory morbidity at the moment of the study or until dis charge. The following parameters were analyzed: mean accumulated SpO2 and SD, minimum SpO2 value, SpO2 time percentage < 90%, SpO2 time percentage < 80%, DI4, and DI80. Continuous over night SpO2 was obtained with Masimo Radical-7 or Rad-8 (USA), mean artifact-free-recording-time (AFRT) ≥ 6 hours. RESULTS: 101 SpO2 recordings were registered in 44, 33, and 24 studies at 34, 35, and 36 weeks of PMA, respectively, from 62 preterm infants, twenty-eight (45%) were male, median gestational age at birth 30 weeks (range 26-32), and median birth weight 1480 g (range 785-2700g). Oximetry variables were mean AFRT (± SD) 8.6 (± 1.5) hours; median mean SpO2 96.9% (range 93.3-99.3%); median minimum SpO2 74% (range 51-89%); median time of SpO2 < 90% 2% (range 0-10.6%); median time of SpO2 < 80% 0.1% (range 0-1.3%); median desaturation event ≥ 4% (DI4) within ≥ 10 seconds per hour sampled 15 (range 3.5-62.5); and median desaturation event <80% (DI80) 0.58 (range 0-10.8). We found no differences in SpO2 values between the different PMA wee ks. CONCLUSIONS: We described SpO2 values in very preterm infants, asymptomatic at 34-36 PMA weeks. These values could be used as a reference for guiding oxygen therapy.

Abnormally low regional cerebral oxygen saturation after induction of anesthesia without neurological abnormality: A case report.

Measurement of regional cerebral oxygen saturation (rSO2) using near-infrared spectroscopy (NIRS) in cardiac surgery is known to be useful in reducing postoperative neurological complications. We here present a case of a 71-year-old man in whom severe decrease in unilateral rSO2 was observed after induction of general anesthesia for percutaneous mitral valve clipping, although no neurological abnormalities were found. NIRS does not always predict postoperative neurological complications.

The Capnography Project.

Capnography is an essential tool used in the monitoring of patients during anesthesia and in critical care which, while required in most high-income countries, is unavailable in many low- and middle-income countries. Launched in 2020, the Smile Train-Lifebox Capnography Project aimed to find a "capnography solution" for resource-poor settings. The project was specifically interested in a capnography device that would meet the needs of the Smile Train partner hospitals to help monitor children requiring airway or cleft surgery. Project advisory and technical groups were formed and included representation from anesthesia practitioners from a balanced representation from all level of income countries, technical experts in capnography, and representatives from the Global Capnography Project (GCAP), the University of California at San Francisco Center for Health Equity in Surgery & Anesthesia (CHESA), and the World Federation of Societies of Anaesthesiologists (WFSA). Built upon the WFSA minimum capnometer specifications, a human centered design approach was used to develop a Target Product Profile. Seven manufacturers submitted 13 devices for consideration and 3 devices were selected for the testing phase. Each of these devices was evaluated for build quality, and clinical and usability performance. Based on the findings from the overall testing process, a combined capnography and pulse oximetry device by Zug Medical Systems was chosen. To accompany the new Smile Train-Lifebox capnograph, an international team of experienced anesthesiologists and educators came together to develop the necessary education materials. These materials were piloted in Ethiopia, subsequently modified, and endorsed by the education team. The device is now ready for distribution, with the accompanying education package, to the Smile Train network and beyond. In addition, a study is being planned to measure the impact of capnography introduction into operating rooms in resource-constrained settings.

Cerebral oximetry monitoring versus usual care for extremely preterm infants: a study protocol for the 2-year follow-up of the SafeBoosC-III randomised clinical trial.

BACKGROUND: In the SafeBoosC-III trial, treatment guided by cerebral oximetry monitoring for the first 72 hours after birth did not reduce the incidence of death or severe brain injury in extremely preterm infants at 36 weeks' postmenstrual age, as compared with usual care. Despite an association between severe brain injury diagnosed in the neonatal period and later neurodevelopmental disability, this relationship is not always strong. The objective of the SafeBoosC-III follow-up study is to assess mortality, neurodevelopmental disability, or any harm in trial participants at 2 years of corrected age. One important challenge is the lack of funding for local costs for a trial-specific assessment. METHODS: Of the 1601 infants randomised in the SafeBoosC-III trial, 1276 infants were alive at 36 weeks' postmenstrual age and will potentially be available for the 2-year follow-up. Inclusion criteria will be enrollment in a neonatal intensive care unit taking part in the follow-up study and parental consent if required by local regulations. We aim to collect data from routine follow-up programmes between the ages of 18 and 30 months of corrected age. If no routine follow-up has been conducted, we will collect informal assessments from other health care records from the age of at least 12 months. A local co-investigator blinded to group allocation will classify outcomes based on these records. We will supplement this with parental questionnaires including the Parent Report of Children's Abilities-Revised. There will be two co-primary outcomes: the composite of death or moderate or severe neurodevelopmental disability and mean Bayley-III/IV cognitive score. We will use a 3-tier model for prioritisation, based on the quality of data. This approach has been chosen to minimise loss to follow-up assuming that little data is better than no data at all. DISCUSSION: Follow-up at the age of 2 years is important for intervention trials in the newborn period as only time can show real benefits and harms later in childhood. To decrease the risk of generalisation and data-driven biased conclusions, we present a detailed description of the methodology for the SafeBoosC-III follow-up study. As funding is limited, a pragmatic approach is necessary. TRIAL REGISTRATION: ClinicalTrials.gov NCT05134116 . Registered on 24 November 2021.