Intermittent Hypoxemia in Preterm Infants Q1 3

Home , Hypoxemia

1 2 Intermittent Hypoxemia in 3 4 Preterm Infants 5 6 a, b b Juliann M. Di Fiore, BS *, Peter M. MacFarlane, PhD , Richard J. Martin, MD Q2 Q3 7 Q4 8 9 KEYWORDS 10 11 Intermittent hypoxemia Hypoxia Pulse oximetry Retinopathy of prematurity 12 Neurodevelopmental impairment Outcomes 13 14 KEY POINTS 15 16 Intermittent hypoxemia (IH) events are common in preterm infants during early postnatal 17 life. 18 In neonates, IH events have been associated with multiple morbidities, including retinop- 19 athy of prematurity, sleep disordered breathing, neurodevelopmental impairment, and death. 20 21 The relationship between IH and morbidity may depend on the pattern of the IH events, although this needs further investigation. 22 23 24 25 26 INTRODUCTION 27 28 Although maintaining adequate oxygenation is a fundamental aspect of newborn care, 29 clinicians are only beginning to appreciate how even subtle alterations in oxygen levels 30 can affect both short-term and long-term outcomes. Before the implementation of Q8 31 noninvasive technologies, oxygen assessment was limited to intermittent arterial sam- 32 pling, which only gave a glimpse of the true instability of oxygen levels that can occur 33 during early postnatal life. Current continuous recordings of oxygen saturation reveal a 34 much higher frequency of intermittent hypoxemia (IH) events that were previously un- 35 documented in medical charts and provide insight to high-risk patterns associated 36 with both short-term and long-term morbidity. This article summarizes what is 37 currently known about the technology used to assess oxygen levels, patterns of IH 38 during early postnatal life, underlying mechanisms associated with IH, and high-risk 39 IH patterns that may induce a pathologic cascade. 40 41 42 43 Disclosures: The authors have nothing to disclose. Q7 44 a Division of Neonatology, Case Western Reserve University, Rainbow Babies & Children’s 45 Hospital, Suite RBC 3100, Cleveland, OH 44106-6010, USA; b Case Western Reserve University, 46 Rainbow Babies & Children’s Hospital, 11100 Euclid Avenue, Suite RBC 3100, Cleveland, OH Q5 47 44106-6010, USA * Corresponding author. Q6 48 E-mail address: [email protected]

CLP1106_proof ■ 14 June 2019 ■ 12:34 pm 2 Di Fiore et al

49 HISTORICAL PERSPECTIVE 50 51 Before the mid-1970s, intermittent arterial sampling formed the basis for assessing 52 and managing supplemental oxygen administration. There was no clear evidence 53 that arterial oxygen tension (PaO2) or arterial oxygen saturation (SaO2) showed frequent Q9 54 fluctuations in preterm infants even though apnea of prematurity was known to be a 55 problem. This situation changed with the advent of transcutaneous PO2 (TcPO2) mon- 56 itors, which were widely used during the 1980s. It was remarkable to observe how po- 57 sitional changes and associated procedures, such as spinal taps, could cause TcPO2 1 58 to decrease. This finding led to the widespread acceptance of gentle care for fragile 59 preterm infants. 60 However, TcPO2 electrodes (later combined with TcPCO2 electrodes) were cumber- some, required frequent recalibration, and resulted in site erythema from heating to 61 62 43 Cto44C. There was also the realization in the 1980s that TcPO2 increasingly 63 underestimated PaO2 with advancing postnatal age; this was especially a problem 2–4 64 with the increasing incidence of bronchopulmonary dysplasia (BPD). These patients 65 were a new population of extremely low birth weight infants who no longer had arterial 66 access. A solution was found in pulse oximetry and this technology has dominated 67 neonatology since approximately 1990. As a result, the available literature on IH epi- 68 sodes in preterm infants is almost exclusively based on pulse oximetry. 69 70 PULSE OXIMETRY 71 Because stabilization of oxygenation is one of the primary challenges in the neonatal 72 intensive care unit (NICU), pulse oximetry plays an important role in patient care. 73 Bedside discussions often include oximetry-based histogram data to note percentage 74 time in any given oxygen saturation range and/or nursing notation of IH events. How- 75 ever, treatment decisions based on medical chart documentation may be problematic 76 because they significantly underestimate the true incidence of even prolonged events.5 77 Adding to the confusion is that there is currently no criteria for a clinically relevant IH 78 event and, therefore, corresponding pulse oximeter alarm settings. Most research trials 79 have defined an IH event as a decrease less than 85% or 80%, but in the clinical setting 80 a low threshold alarm is determined by the individual NICU oxygen saturation target. 81 There is also wide variation in practice pertaining to the clinical significance of the dura- 82 tion of an IH event. For example, health care workers in some NICUs use a long pulse 83 oximeter alarm delay as a tool to minimize nuisance alarms caused by short self- 84 resolving IH, whereas others consider that even short events require intervention. 85 Correspondingly, alarm delay criteria can vary widely between NICUs depending on 86 the manufacturer (ranging from 0 to 15 seconds) and staff perception of the duration 87 needed for a clinically relevant desaturation event. 88 Pulse oximeters have the advantage of obtaining longitudinal documentation of ox- 89 ygen levels in a noninvasive and rapidly responding manner, but there are limitations 90 that may affect measurement accuracy. The most widely acknowledged disadvantage 91 of pulse oximetry is in the inability to detect hyperoxemia at levels of SpO2 (oxygen satu- 92 ration via pulse oximetry) exceeding approximately 97%.6 Additional factors, including 93 probe position, motion and ambient light interference, low perfusion, skin pigmenta- 94 tion, and variations in hemoglobin level, may result in delayed waveform recognition 95 and/or underestimation of oxygen saturation levels.7,8 Most manufacturers report an 96 overall error of Æ2% to 3% of full scale but, even under ideal conditions, accuracy 97 9 may diminish with decreasing SpO2. In a study of 1664 preterm infants, overall 98 mean differences between SaO2 and SpO2 (Masimo) were À1.8% Æ 2.9% but less 99 than 40% of infants were within 3% of the corresponding SaO2 when SpO2 decreased

CLP1106_proof ■ 14 June 2019 ■ 12:34 pm Intermittent Hypoxemia in Preterm Infants Q1 3

100 less than 88%. During very low oxygen saturation levels of less than 70%, which are 101 impractical or unethical to target in infants, newborn lamb models have shown even 102 larger differences, ranging from 13% to 17%.10 Therefore, although pulse oximetry is 103 often used to identify periods of IH, treatment decisions based on the absolute SpO2 104 value should be made with caution, especially at low levels of oxygen saturation, as 105 can occur in cardiac patients. 106 107 INCIDENCE OF INTERMITTENT HYPOXEMIA EVENTS AND UNDERLYING 108 MECHANISMS 109 110 There is considerable current interest in the postnatal time course of IH events in pre- 111 term infants. This interest is precipitated by the fact that persistent IH events frequently 112 delay hospital discharge and may be perceived as increasing the vulnerability of these 113 infants. Barrington and colleagues11 documented more than 20 years ago that apneic 114 events of longer than 12 seconds are common in very low birth weight preterm infants 115 before discharge, many of which would, presumably, have been associated with IH. 116 Data for infants of 24 to 28 weeks’ gestation show a marked change in IH events 117 over time, with few hypoxemic episodes occurring during the first week of postnatal 118 life, a progressive increase in weeks 2 to 3, a plateau around weeks 4 to 6, and then 119 a decrease in weeks 6 to 812 (Fig. 1). It has been proposed that this postnatal increase 120 in IH events may be related to a documented postnatal increase in periodic breathing, 121 which is frequently associated with episodic desaturation.13 In a subsequent study, the 122 incidence of IH events was significantly increased in a low (85%–89%) versus high 14 123 (91%–95%) baseline SpO2 target. This finding is consistent with earlier data in pre- 124 term infants with BPD.15 More recently IH events (comprising a >10% decrease in 125 baseline SpO2) were reported in most preterm infants during home recordings of 16 126 SpO2, but declined between 36 and 44 weeks of postmenstrual age. 127 Immature respiratory control resulting in apnea and respiratory pauses, as well as 128 ineffective and/or obstructed inspiratory efforts, are the major precipitants of IH 129 events,17 but several physiologic parameters likely contribute to the resultant desatu- 130 ration. The most important of these is probably pulmonary oxygen stores, which may 131 reflect lung volume. Preterm infants are at risk for a low basal functional residual 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 Fig. 1. The progression of IH events during early postnatal life. Preterm infants have few IH 148 events during the first week of life, followed by an increase during weeks 2 to 4, and a decrease thereafter. (From Di Fiore JM, Bloom JN, Orge F, et al. A higher incidence of inter- 149 mittent hypoxemic episodes is associated with severe retinopathy of prematurity. The Jour- 150 nal of pediatrics. Jul 2010;157(1):69-73)

CLP1106_proof ■ 14 June 2019 ■ 12:34 pm 4 Di Fiore et al

151 capacity because of both atelectasis and high chest wall compliance. Therapy with 152 continuous positive airway pressure (CPAP) clearly benefits these infants by splinting 153 the upper airway, stabilizing lung volume, and increasing baseline SpO2, thereby mini- 154 mizing the risk of desaturation. Other physiologic parameters, including blood oxygen 155 capacity comprising blood volume and hemoglobin content,18,19 may also be impli- 156 cated in IH. Current neonatal practice attempts to minimize the use of mechanical 157 ventilation via endotracheal intubation and favors the widespread use of CPAP or 158 other noninvasive ventilation techniques. Synchronization methods during noninva- 159 sive ventilation to decrease IH events continue to be a challenge.20 However, IH epi- 160 sodes are also common even in intubated, ventilated infants as a consequence of 161 ineffective ventilator support and loss of the infant’s lung volume.21,22 Interestingly, in- 162 trauterine growth status also seems to be a factor. Small-for-gestational-age (SGA), 163 versus appropriately grown, preterm infants are more vulnerable at a low baseline 164 23 SpO2 target, as manifested by higher mortality and increased incidence of IH events. 165 The mechanism of this increased vulnerability to IH is unclear, but pulmonary hyper- 166 tension in SGA infants and resultant hypoxia-induced pulmonary vasoconstriction 167 may contribute. 168 169 170 INTERMITTENT HYPOXEMIA AND OUTCOMES 171 Stabilizing the extreme fluctuations in oxygenation in preterm infants often requires a 172 balance between the risks and benefits of oxygen supplementation and constant ad- 173 justments of fraction of inspired oxygen (FiO2), which can be labor intensive. Therefore, 174 with the current NICU focus on aggressive weaning protocols, knowledge of the asso- 175 ciation between IH and morbidity would be of great benefit in guiding clinical practice. 176 The evidence that suggests that IH may initiate a pathologic cascade is derived from 177 animal models that show that intermittent hypoxia during adulthood increases extra- 178 cellular superoxide concentration,24 induces hypoxia-inducible factor (HIF) 1a expres- 179 sion,25 degrades HIF-2a expression, and downregulates superoxide dismutase26 180 leading to overall pro-oxidant signaling. In addition, exposure to IH during early post- 181 natal life disrupts expression patterns of proteins involved with dopamine signaling27 182 and causes a proinflammatory response, including increased levels of tumor necrosis 183 factor alpha and interleukin-1b.28 Thus, in infants, IH may induce a pathologic cascade 184 via a pro-oxidant, proinflammatory, or neurotransmitter imbalance pathway (Fig. 2). 185 IH during early postnatal life has been associated with multiple poor outcomes, 186 including retinopathy of prematurity (ROP), growth restriction, sleep disordered 187 breathing, neurodevelopmental impairment, and mortality. Although it is well known 188 that early postnatal hyperoxic exposure is the major risk factor for severe ROP, there 189 is evidence that later IH causes rebound overexpression of growth factors (ie vascular 190 endothelial growth factor, erythropoietin) associated with HIFs that may play a role in Q10 191 neovascularization.29 Multiple animal models with various intermittent hypoxic para- 192 digms have induced retinopathy30 and have shown that the level of neovascularization 193 may depend on the pattern of intermittent hypoxic exposure.31 For example, rats 194 exposed to cycles of IH in a clustered (10 minutes apart) versus equally dispersed 195 (2 hours apart) pattern have more severe oxygen-induced retinopathy, including 196 vascular tufts, leaky vessels, retinal hemorrhage, and vascular overgrowth in the clus- 197 tered paradigm. Studies in preterm infants have shown similar findings with a higher 198 number of IH events during early postnatal life in infants with severe ROP requiring 199 laser therapy.12 Closer examination of IH patterns revealed an association between 200 ROP and IH of longer duration,32,33 less severity, and a specific time interval between 201 events of 1 to 20 minutes.32 The time between IH events may play an important role in

CLP1106_proof ■ 14 June 2019 ■ 12:34 pm Intermittent Hypoxemia in Preterm Infants 5

202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 Fig. 2. The relationship between IH and outcomes may depend on the pattern of the hypoxemic events. IH has been associated with morbidity in preterm infants but not all IH has 235 been shown to have deleterious effects. Therefore, these studies may represent severe IH in 236 terms of frequency, severity, duration, and timing that induce inflammation, neurotrans- 237 mitter imbalance, and reactive oxygen species (ROS) with a high risk of poor outcomes. In 238 contrast, the effect of mild IH on outcomes is unclear. 239 240 initiating a pathophysiologic response, because rodent studies have revealed tran- 241 sient alterations in pro-oxidant signaling occurring during the resolution of the hypox- 242 emic event.24,25 243 The relationship between IH events and growth restriction is currently limited to 244 neonatal rodent models. For example, repetitive exposure to IH during the first 245 week of life significantly reduced body weight by the third day, with subsequent 246 growth restriction with every day of exposure.34 After 21 days of recovery in room 247 air, the rat pups showed catch-up growth, suggesting that IH events commonly 248 seen in preterm infants during early postnatal life may play a role in weight gain, a com- 249 mon criterion for hospital discharge. However, to our knowledge, there are currently 250 no available data for the potential effect of IH on growth restriction in human neonates. 251 IH events can have both short-term and long-term effects on respiratory stability 252 and sleep disordered breathing. In rodents, early postnatal exposure to intermittent

CLP1106_proof ■ 14 June 2019 ■ 12:34 pm 6 Di Fiore et al

253 hypoxia enhanced the acute hypoxic chemoreflex immediately following intermittent 254 hypoxia exposure35 in males with no effect in females. In contrast, early postnatal 255 exposure to intermittent hypoxia followed by recovery to young adulthood reduced 256 the acute ventilatory response to hypoxia,36 suggesting that IH events during early 257 postnatal life may have divergent short-term versus long-term effects on respiratory 258 control. The early postnatal increase in peripheral chemoreceptor activity may be 259 one explanation why periodic breathing often develops within 2 to 4 weeks of life in 260 preterm infants,13 whereas the long-term reduction in peripheral chemoreceptor activ- 261 ity may be one component contributing to preterm birth being a risk factor for sleep 262 disordered breathing at 8 to 11 years of age.37 263 Human trials in neonates have revealed associations between both delayed res- 264 olution38,39 and increased frequency and/or severity of cardiorespiratory events40,41 265 and neurodevelopmental impairment. The findings of these single-center studies 266 have been confirmed by a recent retrospective analysis in the multicenter Canadian 267 Oxygen Trial (COT).33 Using continuous recordings of oxygen saturation in a large 268 cohort of more than 1000 infants, the investigators found a correlation between 269 time spent with hypoxemia during the first few months of life and adverse 18-month 270 outcomes, including late death or disability, cognitive language delay, and motor 271 impairment. Additional clinical trials have shown a relationship between BPD and 272 poor cognitive outcomes,42–45 which begs a question: is it the oxygen exposure 273 or the increased IH events that can occur with chronic lung disease that may put 274 these infants at risk for neurodevelopmental impairment? Data in rodents have 275 shown that neonatal exposure to hyperoxia (65% O2) alone had no effect on 276 long-term working memory, whereas hyperoxia plus IH resulted in neurofunctional 277 handicap.46 Taken together, these studies suggest that exposure to IH events dur- 278 ing a critical period of brain development may have long-term consequences on 279 brain dysfunction. 280 In addition, patterns of oxygenation during the first few days of life have been asso- 281 ciated with mortality. The Surfactant Positive Pressure and Oxygenation Trial (SUP- 282 PORT) randomized infants to a lower (85%–89%) versus higher (90%–95%) oxygen 283 saturation target to reduce the incidence of severe ROP (SUPPORT47). One unex- 284 pected finding of the trial was a higher mortality in the lower target group that was 285 limited to infants with intrauterine growth restriction.48 Closer examination of achieved 286 levels of oxygenation (as opposed to the randomized target) revealed an association 287 between lower oxygen saturation (92%) during the first 3 days of life and decreased 288 90-day survival in both appropriate gestational age (AGA) and SGA infants. However, 289 SGA infants also had an enhanced mortality associated with an increased frequency 290 of IH events that was not seen in the AGA infant cohort.23 291 Interestingly, not all patterns of IH may be deleterious and the effects of mild hyp- 292 oxemia are not clear. For example, mild IH has been associated with later impaired 293 sensorimotor performance in mice.49 In contrast, various paradigms in both human 294 and animal models have suggested that mild hypoxia/hypoxemia may be benign or 295 even beneficial. In rodents, exposure to a single brief (5-minute) cycle of hypoxia dur- 296 ing the first 24 hours after birth,50 or longer (4 hours) and milder cycles of hypoxia dur- 297 ing the first 3 to 4 weeks of life51 enhanced long-term spatial learning,51 memory,50 298 and structural changes in both the hippocampus50,51 and frontal cortex.50 In neonates, 299 short33 or tightly clustered (<1 minute apart)12 IH events were not associated with 300 morbidity, but the risk associated with such patterns may be confounded by other fac- 301 tors, including intrauterine growth restriction.23 In adults, various therapeutic mild IH 302 paradigms are currently being examined for sleep apnea, systemic hypertension, 303 depression, and neural inflammation.52 These combined findings suggest that the

CLP1106_proof ■ 14 June 2019 ■ 12:34 pm Intermittent Hypoxemia in Preterm Infants 7

304 compensatory versus maladaptive effects of IH may depend on the “dose” of IH. 305 Because premature infants show a wide array of IH patterns with increasing postnatal 306 age,32 it is possible that early postnatal IH configurations contain both pathologic and 307 beneficial components (see Fig. 2). 308 309 THERAPEUTIC OPTIONS 310 311 Management strategies for IH events focus primarily on their prevention. The aggres- 312 siveness of these approaches clearly depends on the likelihood of adverse effects of 313 IH, and unanswered questions remain. Desaturation events are a consequence of 314 immature respiratory control superimposed on an immature respiratory system and, 315 in many cases, there is probably vulnerability to pulmonary hypertension. Two main- 316 stays of management are, therefore, optimizing baseline oxygenation and enhancing 317 respiratory control (Fig. 3). 318 1. Optimizing baseline oxygenation. There seems to be widespread consensus that 319 baseline SpO2 should be maintained in the 90% to 95% range, and that risks asso- 320 ciated with restricting the upper limit to 89% outweigh the benefits.53,54 As already 321 noted, low baseline SpO2 significantly increases the risk of desaturation, especially 322 in SGA infants.14,23 Over several decades, CPAP has proved an important means to 323 stabilize oxygenation by supporting functional residual capacity and splinting the 324 upper airway to prevent its closure during apnea.The role of red blood cell transfu- 325 sions in decreasing apnea (and resultant desaturation) has been controversial. 326 Most recent data show improvement in apnea and IH after red blood cell transfu- 327 sion. Age-dependent improvement in frequency and severity of IH after transfusion 328 have been documented beyond the first week of life when IH events begin to 329 increase.18 Another study documented a decrease in apnea associated with 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 Fig. 3. Treatment strategies for IH include a multipronged approach; xanthines to enhance 353 respiratory control, supplemental oxygen and pressure support to optimize baseline level of 354 oxygenation, and red blood cell (RBC) transfusion to improve oxygen stores.

CLP1106_proof ■ 14 June 2019 ■ 12:34 pm 8 Di Fiore et al

355 desaturation and bradycardia in the 12 hours after red cell transfusion.55 It seems 356 that the benefits of increasing red cell mass on IH outweigh the risk of transfusion in 357 selected infants. 358 There is great interest in the rapid emergence of automated (vs manual) control of 359 supplemental oxygen delivery in high-risk neonates, as addressed elsewhere in this 360 issue. Initial automated systems focused on ventilated infants and had great suc- 361 cess in reducing hyperoxic, but not hypoxic, episodes.56 More recent studies by 362 several European groups have had success in reducing IH events with automated 363 versus manual FiO2 control and further documented this benefit in infants on nonin- 364 vasive as well as invasive ventilator support.57,58 This technology is being used clin- 365 ically in Europe and is undergoing further investigative trials in the United States. 366 Although the available data are encouraging, it should be noted that the percentage 367 58 of time that SpO2 was in range (91%–95%) only increased from 58% to 62%, 368 emphasizing the major challenge in keeping preterm infants in any given target 369 even with rapidly responding computer controlled feedback. 370 The application of near-infrared spectroscopy as a tool for assessing the relation- 371 ship between SpO2 and cerebral tissue oxygen saturation is still the subject of 372 ongoing research. Although interpretation of absolute values of cerebral tissue 373 oxygenation have yet to be determined, recent data have shown a greater adverse 374 impact from IH than bradycardic events.59 It remains to be seen whether this nonin- 375 vasive technique can be used as a prognostic marker.60In addition, other novel new 376 approaches are being explored. Two investigative groups are exploring afferent 377 stimulation at the body surface to decrease apnea and resultant IH. Smith and col- 378 leagues61 are using stochastic resonance stimulation via gentle mattress vibra- 379 tions, whereas Kesavan and colleagues are using extremity vibration devices to 380 enhance limb proprioceptive afferents.61,62 Although both approaches show prom- 381 ise in decreasing IH events, it is unclear whether stabilization of oxygenation is a 382 contributing mechanism. 383 2. Role of caffeine therapy. Xanthine therapy, notably caffeine, has become a mainstay 384 of neonatal care and improves not only respiratory but also neurodevelopmental out- 385 comes.63 Although its mechanisms of benefit may be multifactorial, enhanced 386 neonatal respiratory control with caffeine is well accepted.64 Earlier data questioned 387 whether caffeine improved hypoxemic episodes in preterm infants,65 but recent data 388 clearly show a reduction in IH.66 This is not surprising, given that caffeine reduces the 389 incidence of apnea of prematurity.There are still interesting issues regarding optimal 390 caffeine administration for preterm infants. The first issue is when this therapy should 391 begin. Available data support early (even prophylactic) dosing in high-risk neonates; 392 however, these data are largely based on retrospective reviews and associations 393 rather than randomized trials.67 The second issue is when this therapy should 394 stop. Rhein and colleagues66 showed a reduction in IH events with prolonged ther- 395 apy until approximately 36 weeks of postmenstrual age and even beyond. This 396 finding is under further study but runs the potential risk of either prolonging hospital- 397 ization or increasing home monitor use. The third and most challenging issue is 398 whether doses higher than those traditionally used should be used in either a loading 399 or maintenance mode. Available data comparing high versus standard dosing are 400 limited by low-quality outcome measures, small sample sizes, and diverse caffeine 401 dosing regimens.68 Benefit versus safety of such an approach must be carefully 402 studied because adenosine receptor subtype inhibition (the presumed main mecha- 403 nism of action for xanthines) is variable and dose dependent, raising potential safety 404 concerns.69,70 The challenge is to weigh the potential consequences of IH episodes 405 against the pros and cons of any therapeutic intervention.

CLP1106_proof ■ 14 June 2019 ■ 12:34 pm Intermittent Hypoxemia in Preterm Infants 9

406 SUMMARY 407 408 Intermittent hypoxemic events are ubiquitous in preterm infants and are a major chal- 409 lenge in clinical care. Early arterial blood gas sampling limited the ability to monitor 410 oxygen levels over small windows of time, whereas implementation of pulse oximetry 411 has increased knowledge of the transient progression of IH during early postnatal life. 412 Current treatment modalities used in the clinical setting have only been partially suc- 413 cessful in reducing the incidence of apnea and accompanying IH but the risks and 414 benefits of more aggressive interventions must include knowledge of the relationship 415 between IH and morbidity. Future clinical challenges that may assist in mitigating IH 416 and possible sequelae include identification of optimum oxygen saturation targets, 417 recognition of high-risk (and possibly beneficial) IH patterns, and implementation of 418 FiO2 automated controllers and other novel therapies to avoid periods of both hyper- 419 oxemia and hypoxemia. 420 421 ACKNOWLEDGMENTS 422 423 HL 056470, HL 138402 and Gerber Foundation Reference # 1082-4005. Q11 424 425 Best Practices 426 What is the current practice for IH? 427 Best practice/guideline/care path objectives 428 Continuous oxygen saturation monitoring at the bedside 429 Identify periods of hypoxemia 430 Interventions include 431 Caffeine 432 CPAP or nasal cannula support Cautious use of supplemental oxygen 433 434 What changes in current practice are likely to improve outcomes? 435 Identification of optimal oxygen saturation targets 436 Implementation of automated oxygen control systems to minimize fluctuations in 437 oxygenation and improve time in target 438 Optimization of caffeine therapy dosing and duration 439 440 Major recommendations/rating for the strength of the evidence 441 The current recommended oxygen saturation target is 90% to 95%, although strength of 442 evidence remains modest 443 Maintenance of adequate ventilation and lung volume to support oxygenation can be 444 provided by CPAP, nasal cannula or noninvasive ventilation 445 Caffeine may be administered prophylactically to reduce intermittent hypoxemic events and 446 minimize ventilatory support 447 Bibliographic sources: Refs.53,57,66 448 449 450 451 REFERENCES 452 1. Gleason CA, Martin RJ, Anderson JV, et al. Optimal position for a spinal tap in 453 preterm infants. Pediatrics 1983;71(1):31–5. 454 2. Rome ES, Stork EK, Carlo WA, et al. Limitations of transcutaneous PO2 and PCO2 455 monitoring in infants with bronchopulmonary dysplasia. Pediatrics 1984;74(2): 456 217–20.

CLP1106_proof ■ 14 June 2019 ■ 12:34 pm 10 Di Fiore et al

457 3. Solimano AJ, Smyth JA, Mann TK, et al. Pulse oximetry advantages in infants with 458 bronchopulmonary dysplasia. Pediatrics 1986;78(5):844–9. 459 4. Ramanathan R, Durand M, Larrazabal C. Pulse oximetry in very low birth weight 460 infants with acute and chronic lung disease. Pediatrics 1987;79(4):612–7. 461 5. Brockmann PE, Wiechers C, Pantalitschka T, et al. Under-recognition of alarms in 462 a neonatal intensive care unit. Arch Dis Child Fetal Neonatal Ed 2013;98(6): 463 F524–7. 464 6. Poets CF, Wilken M, Seidenberg J, et al. Reliability of a pulse oximeter in the 465 detection of hyperoxemia. J Pediatr 1993;122(1):87–90. 466 467 7. Poets CF, Southall DP. Noninvasive monitoring of oxygenation in infants and chil- 468 dren: practical considerations and areas of concern. Pediatrics 1994;93(5): 469 737–46. 470 8. Trivedi NS, Ghouri AF, Shah NK, et al. Effects of motion, ambient light, and hypo- 471 perfusion on pulse oximeter function. J Clin Anesth 1997;9(3):179–83. 472 9. Rosychuk RJ, Hudson-Mason A, Eklund D, et al. Discrepancies between arterial 473 oxygen saturation and functional oxygen saturation measured with pulse oximetry 474 in very preterm infants. Neonatology 2012;101(1):14–9. 475 10. Dawson JA, Bastrenta P, Cavigioli F, et al. The precision and accuracy of Nellcor 476 and Masimo oximeters at low oxygen saturations (70%) in newborn lambs. Arch 477 Dis Child Fetal Neonatal Ed 2014;99(4):F278–81. 478 11. Barrington KJ, Finer N, Li D. Predischarge respiratory recordings in very low birth 479 weight newborn infants. J Pediatr 1996;129(6):934–40. 480 12. Di Fiore JM, Bloom JN, Orge F, et al. A higher incidence of intermittent hypoxemic 481 episodes is associated with severe retinopathy of prematurity. J Pediatr 2010; 482 157(1):69–73. 483 13. Patel M, Mohr M, Lake D, et al. Clinical associations with immature breathing in 484 preterm infants: part 2-periodic breathing. Pediatr Res 2016;80(1):28–34. 485 14. Di Fiore JM, Walsh M, Wrage L, et al. Low oxygen saturation target range is asso- 486 ciated with increased incidence of intermittent hypoxemia. J Pediatr 2012;161(6): 487 1047–52. Q12 488 489 15. McEvoy C, Durand M, Hewlett V. Episodes of spontaneous desaturations in in- 490 fants with chronic lung disease at two different levels of oxygenation. Pediatr Pul- 491 monol 1993;15(3):140–4. 492 16. Hunt CE, Corwin MJ, Weese-Mayer DE, et al. Longitudinal assessment of hemo- 493 globin oxygen saturation in preterm and term infants in the first six months of life. 494 J Pediatr 2011;159(3):377–83.e1. 495 17. Di Fiore JM, Arko MK, Miller MJ, et al. Cardiorespiratory events in preterm infants 496 referred for apnea monitoring studies. Pediatrics 2001;108(6):1304–8. 497 18. Abu Jawdeh EG, Martin RJ, Dick TE, et al. The effect of red blood cell transfusion 498 on intermittent hypoxemia in ELBW infants. J Perinatol 2014;34(12):921–5. 499 19. Sands SA, Edwards BA, Kelly VJ, et al. Mechanism underlying accelerated arte- 500 rial oxygen desaturation during recurrent apnea. Am J Respir Crit Care Med 501 2010;182(7):961–9. 502 20. de Waal CG, van Leuteren RW, de Jongh FH, et al. Patient-ventilator asynchrony 503 in preterm infants on nasal intermittent positive pressure ventilation. Arch Dis 504 Child Fetal Neonatal Ed 2018;104(3):F280–4. 505 21. Dimaguila MA, Di Fiore JM, Martin RJ, et al. Characteristics of hypoxemic epi- 506 sodes in very low birth weight infants on ventilatory support. J Pediatr 1997; 507 130(4):577–83.

CLP1106_proof ■ 14 June 2019 ■ 12:34 pm Intermittent Hypoxemia in Preterm Infants 11

508 22. Esquer C, Claure N, D’Ugard C, et al. Role of abdominal muscles activity on dura- 509 tion and severity of hypoxemia episodes in mechanically ventilated preterm in- 510 fants. Neonatology 2007;92(3):182–6. 511 23. Di Fiore JM, Martin RJ, Li H, et al. Patterns of oxygenation, mortality, and growth 512 status in the surfactant positive pressure and oxygen trial cohort. J Pediatr 2017; 513 186:49–56.e1. 514 24. Fabian RH, Perez-Polo JR, Kent TA. Extracellular superoxide concentration in- 515 creases following cerebral hypoxia but does not affect cerebral blood flow. Int 516 J Dev Neurosci 2004;22(4):225–30. 517 518 25. Yuan G, Nanduri J, Khan S, et al. Induction of HIF-1alpha expression by intermit- 1 519 tent hypoxia: involvement of NADPH oxidase, Ca2 signaling, prolyl hydroxy- 520 lases, and mTOR. J Cell Physiol 2008;217(3):674–85. 521 26. Nanduri J, Wang N, Yuan G, et al. Intermittent hypoxia degrades HIF-2alpha via 522 calpains resulting in oxidative stress: implications for recurrent apnea-induced 523 morbidities. Proc Natl Acad Sci U S A 2009;106(4):1199–204. 524 27. Decker MJ, Rye DB. Neonatal intermittent hypoxia impairs dopamine signaling 525 and executive functioning. Sleep Breath 2002;6(4):205–10. 526 28. Del Rio R, Moya EA, Iturriaga R. Differential expression of pro-inflammatory cyto- 527 kines, endothelin-1 and nitric oxide synthases in the rat carotid body exposed to 528 intermittent hypoxia. Brain Res 2011;1395:74–85. 529 29. Chow LC, Wright KW, Sola A. Can changes in clinical practice decrease the inci- 530 dence of severe retinopathy of prematurity in very low birth weight infants? Pedi- 531 atrics 2003;111(2):339–45. 532 30. Penn JS, Henry MM, Wall PT, et al. The range of PaO2 variation determines the 533 severity of oxygen-induced retinopathy in newborn rats. Invest Ophthalmol Vis 534 Sci 1995;36(10):2063–70. 535 31. Coleman RJ, Beharry KD, Brock RS, et al. Effects of brief, clustered versus 536 dispersed hypoxic episodes on systemic and ocular growth factors in a rat model 537 of oxygen-induced retinopathy. Pediatr Res 2008;64(1):50–5. 538 32. Di Fiore JM, Kaffashi F, Loparo K, et al. The relationship between patterns of inter- 539 mittent hypoxia and retinopathy of prematurity in preterm infants. Pediatr Res 540 2012;72(6):606–12. 541 542 33. Poets CF, Roberts RS, Schmidt B, et al. Association between intermittent hypox- 543 emia or bradycardia and late death or disability in extremely preterm infants. 544 JAMA 2015;314(6):595–603. 545 34. Pozo ME, Cave A, Koroglu OA, et al. Effect of postnatal intermittent hypoxia on 546 growth and cardiovascular regulation of rat pups. Neonatology 2012;102(2): 547 107–13. 548 35. Julien C, Bairam A, Joseph V. Chronic intermittent hypoxia reduces ventilatory 549 long-term facilitation and enhances apnea frequency in newborn rats. Am J Phys- 550 iol Regul Integr Comp Physiol 2008;294(4):R1356–66. 551 36. Reeves SR, Mitchell GS, Gozal D. Early postnatal chronic intermittent hypoxia 552 modifies hypoxic respiratory responses and long-term phrenic facilitation in adult 553 rats. Am J Physiol Regul Integr Comp Physiol 2006;290(6):R1664–71. 554 37. Rosen CL, Larkin EK, Kirchner HL, et al. Prevalence and risk factors for sleep- 555 disordered breathing in 8- to 11-year-old children: association with race and pre- 556 maturity. J Pediatr 2003;142(4):383–9. 557 38. Janvier A, Khairy M, Kokkotis A, et al. Apnea is associated with neurodevelop- 558 mental impairment in very low birth weight infants. J Perinatol 2004;24(12):763–8.

CLP1106_proof ■ 14 June 2019 ■ 12:34 pm 12 Di Fiore et al

559 39. Pillekamp F, Hermann C, Keller T, et al. Factors influencing apnea and brady- 560 cardia of prematurity - implications for neurodevelopment. Neonatology 2007; 561 91(3):155–61. 562 40. Taylor HG, Klein N, Schatschneider C, et al. Predictors of early school age out- 563 comes in very low birth weight children. J Dev Behav Pediatr 1998;19(4):235–43. 564 41. Greene MM, Patra K, Khan S, et al. Cardiorespiratory events in extremely low 565 birth weight infants: neurodevelopmental outcome at 1 and 2 years. J Perinatol 566 2014;34(7):562–5. 567 42. Short EJ, Klein NK, Lewis BA, et al. Cognitive and academic consequences of 568 bronchopulmonary dysplasia and very low birth weight: 8-year-old outcomes. Pe- 569 diatrics 2003;112(5):e359. 570 43. Gray PH, O’Callaghan MJ, Rogers YM. Psychoeducational outcome at school 571 age of preterm infants with bronchopulmonary dysplasia. J Paediatr Child Health 572 2004;40(3):114–20. 573 44. Anderson PJ, Doyle LW. Neurodevelopmental outcome of bronchopulmonary 574 dysplasia. Semin Perinatol 2006;30(4):227–32. 575 45. Twilhaar ES, Wade RM, de Kieviet JF, et al. Cognitive outcomes of children born 576 extremely or very preterm since the 1990s and associated risk factors: a meta- 577 analysis and meta-regression. JAMA Pediatr 2018;172(4):361–7. 578 46. Ratner V, Kishkurno SV, Slinko SK, et al. The contribution of intermittent hypox- 579 emia to late neurological handicap in mice with hyperoxia-induced lung injury. 580 Neonatology 2007;92(1):50–8. 581 47. Carlo WA, Finer NN, Walsh MC, et al. Target ranges of oxygen saturation in 582 extremely preterm infants. N Engl J Med 2010;362(21):1959–69. 583 48. Walsh MC, Di Fiore JM, Martin RJ, et al. Association of oxygen target and growth 584 status with increased mortality in small for gestational age infants: further analysis 585 of the surfactant, positive pressure and pulse oximetry randomized trial. JAMA 586 Pediatr 2016;170(3):292–4. 587 49. Juliano C, Sosunov S, Niatsetskaya Z, et al. Mild intermittent hypoxemia in 588 neonatal mice causes permanent neurofunctional deficit and white matter hypo- 589 myelination. Exp Neurol 2015;264:33–42. 590 50. Martin N, Pourie G, Bossenmeyer-Pourie C, et al. Conditioning-like brief neonatal 591 hypoxia improves cognitive function and brain tissue properties with marked 592 gender dimorphism in adult rats. Semin Perinatol 2010;34(3):193–200. 593 51. Zhang JX, Chen XQ, Du JZ, et al. Neonatal exposure to intermittent hypoxia en- 594 hances mice performance in water maze and 8-arm radial maze tasks. 595 J Neurobiol 2005;65(1):72–84. 596 52. Navarrete-Opazo A, Mitchell GS. Therapeutic potential of intermittent hypoxia: a 597 matter of dose. Am J Physiol Regul Integr Comp Physiol 2014;307(10):R1181–97. 598 53. Manja V, Saugstad OD, Lakshminrusimha S. Oxygen saturation targets in preterm 599 infants and outcomes at 18-24 Months: a systematic review. Pediatrics 2017; 600 139(1) [pii:e20161609]. 601 54. Stenson BJ. Oxygen saturation targets for extremely preterm infants after the 602 NeOProM trials. Neonatology 2016;109(4):352–8. 603 55. Zagol K, Lake DE, Vergales B, et al. Anemia, apnea of prematurity, and blood 604 transfusions. J Pediatr 2012;161(3):417–421 e411. 605 56. Claure N, Bancalari E, D’Ugard C, et al. Multicenter crossover study of automated 606 control of inspired oxygen in ventilated preterm infants. Pediatrics 2011;127(1): 607 e76–83. 608 57. Waitz M, Schmid MB, Fuchs H, et al. Effects of automated adjustment of the 609 inspired oxygen on fluctuations of arterial and regional cerebral tissue

CLP1106_proof ■ 14 June 2019 ■ 12:34 pm Intermittent Hypoxemia in Preterm Infants 13

610 oxygenation in preterm infants with frequent desaturations. J Pediatr 2015;166(2): 611 240–244 e1. 612 58. van Kaam AH, Hummler HD, Wilinska M, et al. Automated versus manual oxygen 613 control with different saturation targets and modes of respiratory support in pre- 614 615 term infants. J Pediatr 2015;167(3):545–50.e1-2. 616 59. Schmid MB, Hopfner RJ, Lenhof S, et al. Cerebral oxygenation during intermittent 617 hypoxemia and bradycardia in preterm infants. Neonatology 2015;107(2): 618 137–46. 619 60. Korcek P, Stranak Z, Sirc J, et al. The role of near-infrared spectroscopy moni- 620 toring in preterm infants. J Perinatol 2017;37(10):1070–7. 621 61. Smith VC, Kelty-Stephen D, Qureshi Ahmad M, et al. Stochastic resonance ef- 622 fects on apnea, bradycardia, and oxygenation: a randomized controlled trial. Pe- 623 diatrics 2015;136(6):e1561–8. 624 62. Kesavan K, Frank P, Cordero DM, et al. Neuromodulation of limb proprioceptive 625 626 afferents decreases apnea of prematurity and accompanying intermittent hypox- 627 ia and bradycardia. PLoS One 2016;11(6):e0157349. 628 63. Schmidt B, Roberts RS, Anderson PJ, et al. Academic performance, motor func- 629 tion, and behavior 11 Years after neonatal caffeine citrate therapy for apnea of 630 prematurity: an 11-year follow-up of the CAP randomized clinical trial. JAMA Pe- 631 diatr 2017;171(6):564–72. 632 64. Abu-Shaweesh JM, Martin RJ. Caffeine use in the neonatal intensive care unit. 633 Semin Fetal Neonatal Med 2017;22(5):342–7. 634 65. Bucher HU, Duc G. Does caffeine prevent hypoxaemic episodes in premature in- 635 fants? A randomized controlled trial. Eur J Pediatr 1988;147(3):288–91. 636 637 66. Rhein LM, Dobson NR, Darnall RA, et al. Effects of caffeine on intermittent hyp- 638 oxia in infants born prematurely: a randomized clinical trial. JAMA Pediatr 639 2014;168(3):250–7. 640 67. Lodha A, Seshia M, McMillan DD, et al. Association of early caffeine administra- 641 tion and neonatal outcomes in very preterm neonates. JAMA Pediatr 2015; 642 169(1):33–8. 643 68. Vliegenthart R, Miedema M, Hutten GJ, et al. High versus standard dose caffeine 644 for apnoea: a systematic review. Arch Dis Child Fetal Neonatal Ed 2018;103(6): 645 F523–9. 646 69. Chavez Valdez R, Ahlawat R, Wills-Karp M, et al. Correlation between serum 647 648 caffeine levels and changes in cytokine profile in a cohort of preterm infants. 649 J Pediatr 2011;158(1):57–64, 64 e51. 650 70. McPherson C, Neil JJ, Tjoeng TH, et al. A pilot randomized trial of high-dose caffeine therapy in preterm infants. Pediatr Res 2015;78(2):198–204.

CLP1106_proof ■ 14 June 2019 ■ 12:34 pm