Transforming neonatal care with artificial intelligence: challenges, ethical consideration, and opportunities
Brynne A. SullivanKristyn BeamZachary A. VesoulisKhyzer B. AzizAmeena HusainLindsey A. KnakeAlvaro MoreiraThomas A. HoovenElliott Mark WeissNicholas CarrGeorge El-FerzliRavi M. PatelKelsey A. SimekA. J. HernándezJames S. BarryRyan M. McAdams
13
Citation
104
Reference
10
Related Paper
Citation Trend
Keywords:
Neonatology
Supplemental oxygen
Childhood blindness
Supplemental oxygen
Premature birth
Cite
Citations (3)
In the United States, 3.9 million babies are born each year, of which approximately 28,000 (0.7%) weigh less than 1250 g at birth. Approximately half of these small preterm infants become affected, to a lesser or greater extent, by retinopathy of prematurity (ROP), and therefore screening examinations are part of routine care.1 ROP, previously known as retrolental fibroplasia, is a disorder of disorganized growth of developing retinal blood vessels and can result in fibrovascularization and retinal detachment. In approximately 90% of neonates who develop ROP, the disorder resolves, requires no specific treatment, and leaves no permanent damage. However, the 10% with the most severe forms of ROP go on to have impaired vision or even blindness.2 In fact, 1100 to 1500 infants annually in the United States develop ROP severe enough to require specific treatment and 400 to 600 of these become legally blind.3 The development of ROP historically has involved both elements of high oxygen saturation and relative hypoxia.4 Besides these oxygen stresses, fluctuating oxygen levels and poor infant growth are also now recognized as contributing elements, but many gaps exist in understanding pathogenesis and susceptibility factors. Randomized, masked trials in the United States, Australia, New Zealand, Canada, and the United Kingdom indicated that lower targets of oxygen saturation (85%-89%) using pulse oximeters reduced the rate of treatment for ROP (10.6% vs. 13.5%; RR, 0.79; 95% CI, 0.63-1.00; p = 0.045). However the lower-target group had a higher rate of death than those in the higher-target group (saturations 91%-95%; 23.1% vs. 15.9%; RR in the lower-target group, 1.45; 95% confidence interval [CI], 1.15-1.84; p = 0.002).5 Evidence for genetic susceptibility to ROP is strong, but the identity of the gene(s) involved and the molecular mechanisms remain uncertain. In a recent meta-analysis including seven studies, Liu and colleagues6 concluded that advanced ROP is significantly associated with VEGF (vascular endothelial growth factor) gene polymorphisms. Analyzing monozygotic versus dizygotic twins using mixed-effects logistic regression, Bizzarro and colleagues7 concluded that 30% of the pathogenesis of ROP can be accounted for by clinical factors (including hyperoxia) with genetic factors accounting for 70%. Sanghi and colleagues8 studied 35 pairs of identical twins where both developed ROP and found that 20% had the same or approximately the same severity of ROP while 80% were widely discordant in severity, suggesting that both genetic and environmental factors are relevant. Early dosing of recombinant erythropoietin (rEPO) to preterm infants was once thought to be a risk factor for ROP development,9 but animal models,10 human studies,11, 12 and a revision of the original meta-analysis data (Ohls and Widness, personal communication, 2013) all indicate that rEPO dosing is not a risk factor for developing ROP. Beginning in the 1980s many reports concluded that RBC transfusion was a risk factor for ROP.13-16 Recent publications also report this association,17-19 but it is difficult to sort out the effect of RBC transfusions from the fact that the most severely ill neonates receive more RBC transfusions. This increased transfusion requirement is largely due to the relationship between the number of blood tests needed for intensive care monitoring and the number of RBC transfusions given.20, 21 It has been postulated that damaging effects of transfusions on the immature retina are mediated by an increase in free iron.22 However, whether this is so and whether reducing RBC transfusions of neonates susceptible to ROP reduces their ROP risk are not known.23 In this issue of TRANSFUSION, Dani and colleagues24 from Florence, Italy, report an intriguing observation relevant to ROP pathogenesis and prevention. Reviewing their single-center data from 1999 through 2008, they found that neonates born at less than 29 weeks' gestation who received fresh-frozen plasma (FFP) infusions during their first week after birth were less likely to develop ROP. Their data indicated that receiving at least two infusions of FFP diminished the risk of ROP by approximately one-half (relative risk [RR], 0.46; 95% CI, 0.23-0.93). The FFP was administered to these neonates generally because of prolonged clotting studies or signs of bleeding. The authors list the appropriate cautions that the study was retrospective and that FFP administration was not administered with the intent of diminishing the ROP risk. Also, the neonates who received FFP differed in known and unknown ways from those who did not, and those differences may be relevant to the outcomes. If FFP has the capacity to supply preterm infants with something of value toward ROP prevention, that substance might be IGF-1 or IGFBP-3 as the authors postulate, but it might just as likely be other factors or properties currently untested or unknown. Perhaps their observation, although in need of validation, will help focus ROP research efforts toward preventive approaches supplemental to the current transcutaneous oxygen saturation control programs.1-3 Moreover, if preventive substances from FFP can be identified, perhaps supplying the specific factor(s) in reliable amounts could have substantial advantages over infusing FFP. Importantly, we judge that it would be imprudent to advise a widespread practice change on the basis of this observation; specifically, we do not advocate routinely infusing FFP to the smallest neonates with the hope of reducing their risk of severe ROP. Rather, we maintain that, like many associations discovered in retrospective data analysis, this provocative and potentially important finding should be followed by validation, prospective studies, and other rigorous means of testing efficacy, risk, and benefit. The authors report no conflicts of interest or funding sources.
Childhood blindness
Supplemental oxygen
Hypoxia
Cite
Citations (4)
Blinding
Supplemental oxygen
Cite
Citations (6)
Neonatology
Supplemental oxygen
Cite
Citations (59)
Retinopathy of prematurity (ROP) is a leading cause of childhood blindness. Not only do the epidemiologic determinants and distributions of patients with ROP vary worldwide, but clinical differences have also been described. The Third Edition of the International Classification of ROP (ICROP3) acknowledges that aggressive ROP (AROP) can occur in larger preterm infants and involve areas of the more anterior retina, particularly in low-resource settings with unmonitored oxygen supplementation. As sub-specialty training programs are underway to address an epidemic of ROP in sub-Saharan Africa, recognizing characteristic retinal pathology in preterm infants exposed to unmonitored supplemental oxygen is important to proper diagnosis and treatment. This paper describes specific features associated with various ROP presentations: oxygen-induced retinopathy in animal models, traditional ROP seen in high-income countries with modern oxygen management, and ROP related to excessive oxygen supplementation in low- and middle-income countries: oxygen-associated ROP (OA-ROP).
Childhood blindness
Supplemental oxygen
Cite
Citations (9)
Oxygen Saturation
Supplemental oxygen
Cite
Citations (45)
( JAMA . 2018;319(21):2173–2174) While preterm newborns were once routinely exposed to unrestricted oxygen therapy, in the 1950s it was suggested that there may be an association between this practice and development of retinopathy of prematurity. Since then, 2 randomized clinical trials have supported this association [Supplemental Therapeutic Oxygen for Prethreshold Retinopathy of Prematurity (STOP-ROP) and Benefits of Oxygen Saturation Targeting (BOOST)] and also revealed that higher SpO 2 ranges significantly increased the risk of severe pulmonary morbidity.
Supplemental oxygen
Oxygen Saturation
Oxygen therapy
Cite
Citations (1)
Supplemental oxygen
Oxygen Saturation
Pulse Oximetry
Cite
Citations (21)
Purpose Excessive oxygen supplementation increases risk of retinopathy of prematurity (ROP). While numerous oxygen parameters could be considered when predicting ROP (saturation targets, actual saturation, fraction of inspired oxygen, etc.), complicated measures are impractical as screening criteria. We sought to develop a simple, clinically useful measure of daily oxygen supplementation during ages 0–28 days to improve prediction of ROPMethods Secondary analysis of two Postnatal Growth and ROP (G-ROP) Study cohorts (G-ROP-1 and G-ROP-2) at 45 hospitals. Infants with a known ROP outcome and complete oxygen data were included. Associations between severe ROP and days on supplemental oxygen (FiO2 > 21%), during ages 0–28 days (DSO28) were assessed, controlling for birth weight (BW) and gestational age (GA). New screening criteria incorporating DSO were developed and compared to current guidelines.Results Among 8,949 studied infants, 459 (5.1%) developed type 1 ROP. DSO28 was associated with severe ROP (adjusted-OR 1.05 per day supplemental oxygen, 95%CI 1.03–1.07, p < .0001). The following criteria had 100% sensitivity for type 1 ROP and higher specificity than current guidelines: new BW/GA criteria with DSO (BW<901 g, GA<26 weeks, or DSO >3), 23.4% fewer infants examined; modified G-ROP criteria including DSO, 29.0% fewer infants; original G-ROP criteria, 31.8% fewer infants.Conclusion In high-level neonatal-care settings, incorporating DSO (a simple measure of oxygen supplementation) into screening criteria improves sensitivity and specificity for type 1 ROP over current BW-GA criteria, but does not perform as well as the validated G-ROP criteria.
Oxygen Saturation
Supplemental oxygen
Oxygen therapy
Cite
Citations (5)