Promoting Sound Development Of Preterm Infants in The Name Of Developmental Neuroscience: Beyond Advanced Life Support And Neuroprotection

for more information:ali.ma@wecistanche.com

Osuke Iwata a,b,*, Sachiko Iwata a, Yung-Chieh Lin a,c,Shin Kato a, Yuko Mizutani a, Tadashi Hisano a,Masahiro Kinoshita c, Satoko Fukaya a, Koya Kawase a,Shinji Saitoh a

a Department of Pediatrics and Neonatology, Nagoya City University Graduate School of Medical Sciences, Nagoya, 467-8601, Japan

b Department of Paediatrics and Child Health, Kurume University School of Medicine, Fukuoka, 830- 0011, Japan

c Department of Pediatrics, National Cheng Kung University Hospital, College of Medicine, National Cheng-Kung University, Tainan, 70403, Taiwan

Received Oct 3, 2020; accepted Oct 30, 2020

Available online 5 December 2020

Click to Cistanche powder health benefits for neuroprotection

Despite the increased survival opportunities for extremely preterm infants, their long-term cognitive outcomes remain poor, with increased incidence of cognitive impairments in childhood and reduced opportunities to attend higher education in young adulthood compared to their term-born peers. Given that a considerable fraction of preterm infants develops cognitive impairments even without apparent sentinel events at birth and cerebral lesions on MRI as- sessed at term equivalent age, future strategies to improve the outcome may need to address cerebral dysfunction, which cannot be explained by the classical understanding of the injury cascade triggered by hypoxia-ischemia around birth. Developmental care has been proposed to minimize neurodevelopmental impairments related to preterm birth. However, consider-able modes of care, environmental settings and procedures provided by the developmental care of current style appear to offer little benefit to the sound development of infants. Although it is obvious that advanced life support and neuroprotective treatments fall far short of compensating for the burden of preterm birth, researchers need to make further effort to fill the knowledge gap in the cerebral function of fetuses and newborn infants before

1. What modern life support techniques and neuroprotective treatments leave behind

According to a study from the national registry of Japanese tertiary neonatal intensive care centers, during a period between 2003 and 2007, 41.6% of extremely preterm in- fants born at a gestational age of 22 and 23 weeks were discharged to home, which improved to 67.8% in the period between 2008 and 2012.1 In contrast to the increased sur-

vival opportunities for these vulnerable infants, their neu- rodevelopmental outcomes are generally poor, with 36.5% of the survivors developing moderate to severe neuro- developmental impairments for the period between 2003 and 2007, which remained unchanged at 34.9% for the period between 2008 and 2012.1 Long-term follow-up studies of very low birth weight infants demonstrated significantly lower opportunities of these infants to attend higher education in young adulthood than their peers with normal birth weight,2 highlighting the presence of a huge gap between survival and long-term functional outcomes of preterm infants.

2. Subtle cerebral injury and cognitive impairments

Imaging studies in the previous decades explored for up- stream events and clinical conditions responsible for neu- rodevelopmental impairments of preterm infants using ultrasound sonography and magnetic resonance imaging (MRI).3 These studies found robust relationships between destructive brain lesions, such as intraventricular haemor- rhage and periventricular leukomalacia, and neuro- developmental outcomes. However, in addition to poor neurodevelopmental outcomes reported in very preterm infants,4e6 recent studies suggested that even late preterm infants (born between 34 and 36 weeks gestation) who rarely develop destructive cerebral injury are at increased risk of developing higher cognitive dysfunctions, such as attention deficit hyperactivity disorder, autism spectrum disorders and learning difficulties.7 These findings suggest that cognitive dysfunction related to preterm birth is a continuum from severe brain injury in extremely preterm infants to more subtle injury in late preterm infants. Indeed, subtle but abnormal signal intensity on the head MRI has been linked to adverse neurodevelopmental out- comes of preterm infants in young childhood and school-age; composite MRI assessment scales, which incorporated these subtle qualitative findings, have now been established as a standard tool to estimate cognitive out- comes of preterm infants.8,9 For further improvement of cognitive outcomes of preterm infants, the focus of clinical investigations may need to be directed from the prevention of destructive cerebral injury to that of relatively more subtle brain injury.

3. Cerebral dysfunction without injury: what cannot be explained by classical injury cascade

As described above, cerebral injury and dysfunction are likely to develop even in the absence of sentinel events. Consistent with this hypothesis, a large prospective cohort study conducted in the United Kingdom demonstrated that newborn infants who required some form of resuscitation but were not hospitalized at neonatal intensive care centers were at increased risk of developing cognitive impair- ments at 8 years of age (odds ratio, 1.65; 95% confidence interval, 1.13e2.43).10 More recent studies further sug- gested the importance of accounting for other independent variables than “classical” causes of cerebral injury and dysfunction. Kim and colleagues investigated the relation- ship between gestational age at birth and school achieve- ment in math and reading at the age of 8 and found that a greater gestational age was consistently associated with a higher performance within the range of 36e41 weeks.10 Our clinical study in 189 preterm and term infants suggested that, in a considerable fraction of brain regions, including the corpus callosum and cerebellum, the regional brain size assessed by MRI was predominantly determined by gesta- tional age at birth rather than by post conceptional age at MRI scans and nutritional status after birth, suggesting that brain growth after preterm birth was considerably restricted or even became negligible compared with that in utero.11 Considering that this trend was consistently observed for extremely preterm infants and relatively more mature infants, the influence of growth restriction of the regional brain may need to be considered for a wide spectrum of preterm infants.

In addition to the trophic support for cerebral growth, pain and stress have increasingly been recognized as another potential independent variable of modified ce- rebral structure and cognitive outcomes of preterm infants. Miller and colleagues reported that preterm infants who experienced relatively more painful procedures had less complex white matter microstructure compared with their peers, when their brains were assessed using the diffusion tensor imaging.12,13 These studies highlighted that therapeutic regimen simply targeting the reduction of hypoxia-ischemia and other stressful events may not be sufficient to ameliorate cognitive impairments of preterm infants. More attention may need to be paid to incorporate novel independent variables of neurodevelopmental impairments, such as malnutrition, pain and stress, environment and circadian rhythms.

4. Current strategy to promote neurological development in preterm infants

To improve the outcome of high-risk newborn infants, a set of clinical care practices, or developmental care, have been proposed from the 1990s.14 During the preliminary phase of developmental care, clinicians attempted to pro- vide newborn infants hospitalized at neonatal intensive care centers with an environment which mimicked intra- uterine conditions.15,16 Within the uterus, oxygen and other energy substrates are provided via the placenta.17 Howev- er, human foetuses are allowed to use only limited amount of oxygen because the partial oxygen tension of the foetal arterial blood typically remains at less than 25 mmHg.18 It would be relevant that foetal cerebral function is attenu- ated to match extremely low oxygen supply, the level of which is even lower compared with that of newborn in- fants.19 The intrauterine environment is warm, soft, dark and calm, with limited noises from the maternal heart beat and blood flow through large vessels, which was believed, to be an ideal environment to accommodate preterm in- fants.15,16 In addition to warm and humid ambiance within the incubator, preterm infants were placed at a flexion position using positioning mattresses and nesting/posi- tioning tools.20 Conversations of ward staffs and visiting parents were controlled so that the ambient noise level did not exceed 45 db (i.e., similar to the level of a quiet library of 40 db but lower than normal conversation at 60 db),21 as recommended by the American Academy of Pediatrics.22 Based on the lack of the spontaneous diurnal secretion of melatonin from the pineal gland of foetuses and newborn infants before 12 weeks of life (and subsequent lack of diurnal cycles in physiological variables and activity pat- terns),23,24 continuous dim lighting was introduced to neonatal intensive care centers in 1980s and 1990s; ambient brightness within the incubator was typically kept under 5 Lux.25 Als and colleagues applied a carefully chosen set of developmental care practices, or the Newborn Individual- ized Developmental Care and Assessment Program (NID- CAP), to 38 very low birth weight infants, which resulted in a shorter duration of mechanical ventilation [63.8 (72.9) days vs. 28.3 (23.3) days; mean (standard deviation)], a better daily body weight gain [20 (6) g vs. 24 (7) g], a shorter hospital stay [151 (120) days vs. 87 (26) days] and a better mental developmental index at 9 months post- conceptional age [94 (23) vs. 118 (17)].14 Despite the dramatic effect of NIDCAP that was demonstrated in preliminary studies, more recent studies with large-enough study populations and reliable endpoints of long-term neurodevelopmental outcomes failed to confirm its benefit in improving the outcome of preterm infants.26 In the latest systematic review and meta-analysis of NIDCAP for preterm infants, Ohlsson and Jacobs did not find any evidence to support the benefit of the intervention to improve short-term medical outcomes and long-term neu- rodevelopmental outcomes. The authors concluded that the implementation of the care in its present form as standard care is not recommended in preterm infants, considering the resource, labour and expense to implement and maintain NIDCAP as well as the lack of evidence support.

5. Gaps between the clinical image of preterm infants and who they really are

It might seem counter-intuitive that an advanced strategy for preterm infants that provides them with a calm, natural and stress-free environment has not been linked with improved long-term outcomes of affected infants as compared with conventional care. However, the recent advances in developmental neuroscience highlighted the presence of a gap in the understanding of the in-utero life of the foetus, on which developmental care is based.

5.1. Auditory input

In the 1980s, Vince and colleagues performed a series of experiments using pregnant sheep to investigate how much auditory input was delivered to the foetus. The sound of the mother’s voice assessed using a hydrophone inside the amni- otic sac was slightly louder than that assessed using a micro- phone beside the mother, whereas sounds produced outside the ewe at levels similar to the normal conversation were deliv- ered to the fetus at a small but audible level.27,28 These findings suggest that the intrauterine environment for foetal lambs (and perhaps human foetuses) is neither quite nor monotonous, but it is likely to be equivalent or at least com- parable with the extrauterine environment in terms of the sound and vocal stimuli. Consistent with this speculation, newborn infants can distinguish their mother's voice from that of others from shortly after birth.29 Caskey and colleagues investigated the number of words spoken to preterm infants hospitalized at a NICU and found that the number of adult words spoken to the infants at 32 weeks post conceptional age accounted for 12% of the variance in language composite scores and 20% of the variance in expressive communication scores at 18 months postconceptional age.30 Pineda and col- leagues found that language composite scores assessed at two years of age were higher for preterm infants, who were hos- pitalized at an open ward, compared with their peers who werehospitalizedataprivateroom[91.9(11.4)vs.84.0(10.5), shown as mean (standard deviation)],31 supporting the hy- pothesis that consistent exposure to vocal and other auditory stimuli might be essential for the verbal development of pre- terminfants. Wearecurrentlyundertakingaprospectivestudy to investigate whether intermittent exposure of preterm in- fants to recorded maternal voice improves their verbal and cognitive outcomes.

5.2. Visual input and diurnal rhythm

The penetration of the ambient light to the foetus is esti- mated to be low unlikely to entrain the foetal circadian clock.32 Subsequently, foetal pineal gland does not acquire its spontaneous diurnal rhythm in the secretion of mela- tonin.33 However, maternal melatonin freely crosses the placenta, which, together with daytime dominant activ- ities and conversations of the mothers, may contribute to the development of foetal diurnal rhythms in heart rate and activity.34,35 Seron-Ferre and colleagues further identified the presence of a remarkable diurnal rhythm in the serum corticosterone level which was entrained in antiphase to the maternal rhythm.36 Considering these consistent observations supporting the presence of diurnal physiological rhythm in the foetus, an important question is raised regarding why previous studies did not find any meaningful diurnal changes in blood and salivary cortisol levels of newborns infants associated with the clock time.

To answer the question, our group performed a series of studies that addressed control variables of cortisol levels in newborn infants. These studies highlighted that serum cortisol levels were unlikely to serve as a surrogate marker of adrenal circadian clock because of the significant influ- ence of procedural pain at blood sampling on the cortisol level.37 In addition, even for non-invasively collected saliva samples, the timing of sampling needs to be carefully scheduled accounting for the cycle of feeding because of a robust influence of feeding on cortisol levels.38 Previous studies, which did not incorporate these crucial indepen- dent variables of cortisol levels, might be biased (and therefore underpowered) to detect diurnal rhythms of newborn infants shortly after birth. We then carefully monitored temporal changes in salivary cortisol levels in newborn infants who were hospitalized at a neonatal intensive care center. Subsequently, we observed the presence of a foetal-type diurnal rhythm of cortisol levels, which was entrained in antiphase to the adult-type rhythm.39 Interestingly, for newborn infants within the first week of life, a second, but relatively higher peak in the cortisol level was observed a few hours after the time of birth.39 We speculated that this transient peak in salivary cortisol might be the consequence of an entrainment at the time of birth and that this might have influenced the find- ings in previous studies which did not find the diurnal rhythm in newborn infants.

For infants hospitalized at a neonatal intensive care center, who are provided with cycled dim lighting (e.g., 100e200 Lux during the day time and 10e30 Lux during the night time), the foetal-type diurnal rhythm in cortisol levels was observed even after four weeks of birth.40 In contrast, when the sleep schedule of 1302 healthy newborn infants was investigated, the sleep duration was already night-time dominant at the age of one month, the level of which depended on the light-off time of the infants’ bedroom and season of birth.41 These findings suggest that newborn in- fants are able to use the information of both daily and seasonal light cycles to acquire a matured dayenight rhythm within the first month of life, although the day time lighting of 100e200 Lux appears to be insufficient to promote the process. Future studies need to address the potential benefit of relatively clearer lighting cycles for newborn in- fants to promote the early establishment of night time- dominant sleep patterns. Given that the night time sleep duration of infants is one of primary independent variables of their mothers’ sleep41 and that insufficient sleep duration of the mothers is an established independent variable of peripartum depression,42 the improvement in the light cycle may help accelerate the acquisition of a matured sleep cycle, and help prevent peripartum depression and subse- quent suicides and child neglect.