Perinatal tissue oxygenation and neurodevelopment in preterm and growth restricted infants
Richter, Anne
DOI:
10.33612/diss.122713783
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2020
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Richter, A. (2020). Perinatal tissue oxygenation and neurodevelopment in preterm and growth restricted infants. University of Groningen. https://doi.org/10.33612/diss.122713783
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
GENERAL DISCUSSION AND
FUTURE PERSPECTIVES
GD
General discussionPerinatal tissue oxygenation greatly influences survival and developmental outcome. Because the brain is particularly important in this matter, several mechanisms exist to spare the brain from hypoxic damage. The presence of cerebral autoregulation and fetal hemodynamic redistribution with brain-sparing mechanisms at the expense of other organs in fetal growth restriction (FGR) represent two of these mechanisms. Both underline the importance of adequate cerebral oxygenation for our existence and identity. However, in many preterm and FGR infants cerebral autoregulation is impaired, which predisposes them to fluctuations in perfusion and oxygenation. Moreover, fetal brain-sparing does not always occur in FGR and its neuroprotective role is controversial. In addition, immaturity and several exogenous factors may interfere with cardiopulmonary transition after birth and adequate oxygenation.
Because not only hypo- but also hyperoxia is associated with tissue injury, balancing between the two extremes can be particularly difficult. Moreover, what we refer to as tissue normoxia in terms of outcome for preterm and growth restricted neonates is poorly defined. Although monitoring tissue oxygenation with near-infrared spectroscopy (NIRS) in addition to the surveillance of arterial oxygenation and other vital parameters clearly has huge potential, its interpretation can be difficult. Up to date, this limits its large-scale use in routine clinical care. We first need to better define normoxic ranges within several circumstances and identify how perinatal factors and complications are related to changes in neonatal oxygenation.
With this thesis we hope to contribute to a better understanding of perinatal hemodynamics and in particular cerebral tissue oxygenation in preterm and FGR infants. In part I, we investigated the influence of prenatal medication with
vasodilative properties used for either maternal or fetal treatment on the cerebral and systemic tissue oxygenation in preterm and FGR infants. Moreover, we tried to entangle their effects from FGR-related placental pathology and fetal brain-sparing. In part II of this thesis, we investigated how fetal brain-sparing and cerebral oxygenation relate to outcome following preterm birth and FGR, including the development of retinopathy of prematurity and neurodevelopment at 4 years of age. Analysis of the methylation of neurodevelopmental genes at oxygen-dependent regulatory genomic regions aimed to find a pathophysiologic explanation on how perinatal oxygenation may influence neurodevelopment without causing visible tissue injury. The main findings of this thesis are summarized in Tables 1 and 2.
GD
Tab le 1. M ai n f in di n gs o f the t he si s, p ar t I (p re n at al p re d icto rs ). Ce re b ral o xy ge n ati o n & b lo o d fl o w R ena l o xyg en ati o n Spl an chni c o xyg en ati o n O the r vi tal p arame te rs La b eta lo l ± Mg SO 4 Ch ap te r 1, Pre te rm i n fa n ts , mi xe d SG A/ AG A Day 1, 2, an d 4: F TO E ↓ FT OE = Day 1 an d 2: F TO E ↑ N I N if e di pi ne ± M gS O4 Ch ap te r 1, Pre te rm i n fa n ts , mi xe d SG A /AG A Day 2: F TO E ↓ FT OE = FT OE = N I Mg SO 4 Ch ap te r 1 & 2, Pre te rm i n fa n ts , m ix ed SG A /AG A Day 2 ( an d 4) : F TO E ↓ PSV = , R I = A u to re gu lati o n = FT OE = FT OE = N I Pr ee cl a m p si a Ch ap te r 2, Pre te rm i n fa n ts , mi xe d SG A/ AG A Day 1-4 ( an d 5) : F TO E ↓ (D ay 2: R I ↓) Day 1( , 2 an d 5) : PSV ↓ (Day 1-5: A u to re gu lati o n ↓) Fe tal b rai n -s p ari n g as me d iato r N I N I N I S il d e n a fil Ch ap te r 3, Se ve re e arl y-o ns et F G R in fant s rSO 2 , F TOE = A u to re gu lati o n = Day 1-2: r SO 2 ↓, F TOE ↑ Day 3: rSO 2 ↑ N I Day 1: d ias to lic B P ↑, HR ↓ Day 3: d ia st o lic BP ↓ , HR ↑ A ll N IR S-rel at e d f in di n gs w er e a ss es sed w it h t h e I NV O S dev ic e an d t h e c o rres po n di n g neo na ta l s ens o r (M e dt ro ni c, D u bl in, I re la n d) . Fi n d in gs i n b racke ts d em o ns tra te no n -si gn ifi cant tre n d f in d in gs ( p -val u e b et w ee n 0.05 an d 0.1). A G A , ap p ro p ri ate -fo r-ge st at io na l a ge ; B P , bl o o d pr es sur e; F GR , fet al g ro w th re str ic ti o n ; FT O E, fr ac ti o n al ti ss u e o xyg en e xtr ac ti o n ; H R , h eart rate ; M gS O4 , magn es iu m su lfat e; N I, n o t in ve st ig ate d ; PSV , p eak -sy st o lic (b lo o d fl o w ) vel o ci ty; RI , re si stan ce in d e x; rSO 2 , r eg io na l t is sue o xy gen s at ur at io n; SG A, s m al l-fo r-ge stati o n al ag e.Tab le 2 . Ma in fi n d in gs o f th e th es is , p art II (p o st n at al o u tco me ). Ce re b ral o xy ge n ati o n & bl o o d fl o w R ena l o xyg en ati o n Spl an chni c o xyg en ati o n O the r vi tal p arame te rs R eti n o p a th y o f p rema tu ri ty Ch ap te r 4, Pre te rm i n fa n ts , mi xe d SG A/ AG A Th e mo re ti me an in fant s p en d s a t rSO 2 le ve ls > 8 5% w ith in th e fi rs t d ays afte r b irt h , the hi gh er th e ri sk to dev el o p severe re ti n o p ath y o f pre m atur ity N A N A N o ass o ci ati o n w it h t he ti me s p en t at S aO 2 lev el s >90/92/95% ( in a po pul at io n w it h n ar ro w Sa O2 targ et r an ge s) N e u ro d e ve lo p m e n t at 4 ye ar s Ch ap te r 5, FG R in fant s, mi xe d te rm/p re te rm 1. rS O2 le ve ls ˃ 71% o n d ay 1 a nd ˃76% o n d ay 2 are ass o ci ate d w it h bet ter be ha vi o r a n d e xec ut iv e fu n cti o n in g at 4 ye ars 2. r SO 2 le vel s ˃ 76% o n day 2 a re as so ci ate d w ith p o o re r PI Q 3. F etal b ra in -s p ari n g l ar ge ly me d iate d h ig h rS O2 le ve ls o n d ay 2 an d ass o ci ate d o u tco me s N A N A N I Meth y la ti o n o f n e u ro d e ve lo p m e n ta l g e n e s a t 4 y ea rs Ch ap te r 6, FG R i n fan ts , mi xe d te rm/p re te rm Fe tal b rai n -s pari ng is a ss o ci at ed w ith d iff ere n ti al me th yl ati o n o f o xyg e n -depe nd ent r eg ul at o ry r eg io ns o f (H IF 1A , V E G FA ,) BDN F , a n d N T R K 2 N A N A N I A ll N IR S-rel ate d fi n di n gs w ere a ss es se d w it h th e IN V O S de vi ce and th e co rresp o n di n g ne o nata l se ns o r ( Me dtro ni c, Du bl in, I re la nd ). Fi n din gs i n b racke ts d emo n str at e n o n -si gn if ic ant t re n d f in d in gs ( p -val u e b e tw ee n 0.05 an d 0.1) . A G A , ap p ro p ri ate -fo r-ge sta ti o n al ag e; B D N F , b rai n -der iv ed n eu ro tr o p h ic fac to r; F G R , fet al g ro w th re str ic ti o n ; FT O E, frac ti o n al ti ss u e o xyg en e xtr ac ti o n ; H IF 1A , hy p o xi a-in d u ci b le fac to r 1 al p h a; N A , no t ap p lic ab le ; N I, n o t i n ve st iga te d ; N T R K 2 , n eu ro tr o p h ic tyr o si n e k in as e, re ce p to r, typ e 2; P IQ , pe rf o rm anc e i nt el lig e nc e q uo ti ent ; rSO 2 , re gi o n al t issu e o xyg en s at u rati o n ; SaO 2 , arte ri al o xyg en s at u rati o n ; SG A , s mal l-fo r-ge stati o n al a ge ; V E G FA , va sc u la r e n d o th e lial g ro w th fa cto r A .
GD
Part I: Postnatal cerebral and systemic tissue oxygenation in relation toprenatal medication and placental dysfunction
In part I of this thesis, we found that preterm infants prenatally exposed to labetalol, nifedipine, and MgSO4 had a lower cerebral fractional tissue oxygen
extraction (FTOE) compared to unexposed infants (chapter 1). Since FTOE represents the relative difference between arterial oxygen saturation (SaO2) and
regional (largely venous) tissue oxygen saturation (rSO2), this suggested either
increased cerebral perfusion or decreased cerebral oxygen consumption. The effect was largest and endured longest for labetalol. Moreover, labetalol was associated with higher splanchnic FTOE on the first two days, which may reflect decreased intestinal perfusion. However, we hypothesized that underlying maternal preeclampsia (PE) and fetoplacental vascular underperfusion may have confounded our results observed for labetalol, as it is associated with fetal hemodynamic redistribution and preferential perfusion of the brain at the expense of other organs. Moreover, labetalol has a half-life of 24 hours in neonates and an effect of labetalol until day 4 seemed unlikely.1 Additionally,
many infants exposed to labetalol or nifedipine were also exposed to antenatal MgSO4, which is given as anticonvulsant therapy to their mother to prevent
eclamptic insults in PE or to protect the fetal brain in imminent preterm birth. To better understand the hemodynamic effects and entangle the contributions of MgSO4 and PE to our findings in chapter 1, we investigated the differential
influence of MgSO4 and PE on cerebral FTOE as measured by NIRS and cerebral
blood flow (CBF) as assessed by Doppler ultrasound (chapter 2). We found that MgSO4 independently reduced cerebral FTOE without affecting CBF or cerebral
autoregulation, suggesting that MgSO4 affects FTOE by merely reducing cerebral
oxygen consumption and not by a vasodilatory increase in oxygen supply. Moreover, MgSO4 did not appear to affect cerebral autoregulation. PE lowered
cerebral FTOE as well, but twice as much as it was observed for MgSO4. In
addition, PE was associated with a lower peak-systolic cerebral blood flow velocity, especially on day 1 after birth, which may suggest cerebral vasodilation and/or reduced left ventricular function, and tended to be associated with impaired postnatal cerebral autoregulation within the first 5 days after birth. These effects of PE on postnatal cerebral oxygenation, autoregulation and flow velocity seemed to be mediated by fetal brain-sparing.
The effect of maternal labetalol is difficult to entangle from underlying maternal disease
The exact mechanisms underlying pregnancy-induced hypertension and PE are poorly understood, but involve upregulation of anti-angiogenic factors.2 Associated high placental resistance and endothelial damage not only cause fetal hypoxia but also maternal hypertension and end-organ disease, including renal insufficiency.3,4 Labetalol effectively reduces maternal blood pressure and is the most frequent choice of maternal symptomatic treatment.5 However, it also accumulates in the neonate and raised concern about neonatal hypotension, bradycardia, and cerebral vasodilation through α1- and β-adrenergic inhibition.6
Indeed, we found in chapter 1 that labetalol was associated with lower cerebral oxygen extraction, which is supported by the findings of Verhagen et al. and may suggest cerebral vasodilation.7 However, in chapter 2, we demonstrated a
significant influence of PE on the cerebral oxygenation, which was mediated by fetal brain-sparing compensating placental insufficiency, while labetalol did not play a confounding role. This is in line with the findings of Thewissen et al., who demonstrated no difference in cerebral FTOE between infants with and without labetalol exposure in hypertensive pregnancy, but a trend towards lower FTOE in both groups when compared to infants born following normotensive pregnancy.8
GD
Similarly, in chapter 2, fetal brain-sparing and not labetalol significantlycontributed to impairment of blood pressure-reactive cerebral autoregulation in the neonate, leading to a pressure-passive cerebral rSO2. Although Caicedo et al.
did not find a significant additional effect of labetalol on the cerebral autoregulation using the same correlation method as well, the same study group also assessed autoregulation by means of transfer function analysis.9 This method evaluates frequency-dependent differences between the amplitude of simultaneously measured mean arterial blood pressure (MABP) and cerebral rSO2
oscillations (‘gain’, testing the damping effect of the autoregulatory system, which is usually better at low-frequency than high-frequency oscillations), time difference between oscillations (‘phase shift’, testing the delay of the autoregulatory response), and overall linearity between MABP and cerebral rSO2
oscillations (testing the ‘coherence’ between the two hemodynamic parameters).10-12 They found gain to be significantly higher at low-frequency oscillations on day 1, indicating less damping of oscillations and thus less effective cerebral autoregulation in infants born following a gestational hypertensive disorder and labetalol use compared to both controls and infants born following a gestational hypertensive disorder without labetalol use, which normalized by day 3.9 They found no difference in coherence or time-delay.9 In addition, gain between cerebral rSO2 andheart rate activity as a surrogate for autonomic control
of muscular tone was increased on days 1 and 2 in the group exposed to labetalol, suggesting interference with cerebral autoregulation by labetalol through adrenergic inhibition.13
It remains difficult to entangle the effect of labetalol and underlying placental insufficiency since both usually come together and maternal hypertension presents a consequence of increased placental resistance to flow.14 Although the
may be possible that their findings for labetalol were biased by more severe placental insufficiency in the respective patient population, requiring treatment with labetalol but also leading to stronger hemodynamic redistribution in the offspring. Compared to the effects of concomitant brain-sparing, our own findings suggest that any additional effects of labetalol on cerebral oxygenation may be relatively negligible, balancing them against their use in treating maternal hypertension and prolonging pregnancy. Under normal circumstances, sympathetic (neurogenic) control of the cerebral autoregulation has shown to play only a small and controversial role compared to a myogenic control mechanism caused by changes in smooth muscle cell membrane potential upon altered intraluminal pressure.15,16 Sympathetic control of the cerebrovasculature,
however, has shown to be upregulated under chronic hypoxic conditions, which may augment an adrenergic effect by labetalol.17 Moreover, the use of labetalol in chapter 2 was infrequent and related to the occurrence of PE, with only few PE-pregnancies not receiving labetalol, which does not allow us to draw any conclusions from our findings in chapter 2 regarding labetalol. Given our findings in chapter 1 and the findings by Caicedo et al., one should stay alert for cerebral vasodilation and impaired cerebral autoregulation during the first days after birth following labetalol and severe early-onset FGR, especially when there is also (induced) bradycardia and hypotension. Additionally, organs without autoregulatory capacity such as the intestines may be affected.
Nifedipine may potentiate the effects of other medications
Nifedipine is officially registered as a maternal antihypertensive drug, but at our institution is primarily being used as an off-label tocolytic agent. There is conflicting evidence whether the calcium channel blocker affects fetal CBF, either directly or by decreasing maternal blood pressure and uteroplacental flow.18-21
GD
Studies evaluating neonatal CBF and blood pressure following nifedipine arescarce. Although nifedipine is considered safer for tocolysis than betamimetics or MgSO4, well-implemented randomized controlled trials (RCTs) assessing efficacy
and safety of nifedipine versus placebo or other tocolytics such as atosiban are few and so far lack sufficient power to detect differences in perinatal outcome and brain injury.22-25 Moreover, although nifedipine crosses the placenta, the exact neonatal elimination half-life is unknown.26,27
In chapter 1, we found that tocolysis with nifedipine (with or without concomitant MgSO4) was associated with lower cerebral FTOE on day 2, which
was larger than the effect of MgSO4 alone, and a trend towards lower renal FTOE
on day 1. From our data it was difficult to conclude whether nifedipine may directly increase cerebral oxygenation by vasodilation or whether its effect was potentiated by MgSO4. Verhagen et al., however, reported no significantly altered
cerebral FTOE during the first 5 days after birth in 19 infants exposed to only nifedipine, comparing them to infants without maternal drug exposure.7 Combining our results, we suggest that tocolytic oral nifedipine alone does not have a clinically relevant effect on cerebral oxygenation but may do so in combination with other medication such as MgSO4.
Differential effect of antenatal MgSO4 and FGR-associated hemodynamic
redistribution
MgSO4 has calcium entry blocking properties and is therefore a potent smooth
muscle relaxant and vasodilator.28 Its high-dose use for tocolytic purposes,
however, has been abolished since hypermagnesemia has shown to increase the risk of IVH and mortality in the offspring.29 On the other hand, low-dose MgSO4 as
maternal anticonvulsive treatment in PE has shown to lower the risk of cerebral palsy, for which it has been adopted as a fetal neuroprotective agent in imminent
preterm birth.30,31 Because IVH and PVL represent major risk factors for cerebral
palsy but have not yet shown to be significantly reduced by MgSO4, its
neuroprotective effect has largely been attributed to NMDA-receptor blockade reducing glutamate excitotoxicity following ischemia.28,30-32 Because it remains unclear how MgSO4 affects cerebral oxygenation, blood flow, and autoregulation
independently from underlying PE and thereby alters the risk of PVL and IVH, we investigated the differential effects of MgSO4 and PE on these parameters in
chapter 2.
Our results suggest that MgSO4 increases cerebral oxygen availability
independently from concomitant PE without altering CBF. Instead, it may rather result from a lower cerebral oxygen consumption caused by reduced cerebral activity and metabolism of MgSO4, which has also been proposed by Stark et al.33
We hypothesized that this could contribute to the neuroprotective effect of MgSO4 in preterm neonates, in particular in those with impaired cerebral
autoregulation, hypotension, and/or difficult pulmonary transition, who are at increased risk of cerebral hypoxia. However, the estimated reduction of cerebral FTOE by MgSO4 was only about 3% in our population. It may therefore be argued
whether this effect is clinically relevant and can outweigh potentially negative effects associated with MgSO4-induced hypotension or respiratory depression.
In contrast to PE, MgSO4 did not affect cerebral autoregulation as assessed by
correlation between cerebral rSO2 and MABP. Given the debate on an in- or
decreased risk of IVH following MgSO4 and its contribution to a reduction of
cerebral palsy, we render our finding of unaltered cerebral autoregulation important.30 It suggests that MgSO4 is unlikely to alter the risk of IVH through an
effect on cerebrovascular reactivity. An increased risk of IVH in infants exposed to MgSO4 may rather be attributed to lenticulostriate vasculopathy, which was
GD
one has to bear in mind that cerebral autoregulation greatly depends on bloodpressure. Although, in chapter 2, MABP was not influenced by MgSO4, MgSO4 – in
particular at higher doses – may be able to cause hypotension and thereby still impair autoregulation. Additionally, it may be worthwhile to analyze a potential effect on the neurogenic aspect of cerebral autoregulation, since magnesium may enhance parasympathetic activity and thereby induce cerebral vasodilation.34,35 But because CBF parameters appeared unaffected by MgSO4, we expect this
effect to be negligible.
The cerebrovascular effect of fetal brain-sparing
Altered cerebrorenal oxygenation and impaired cerebral autoregulation have been observed in FGR infants.36,37 Tanis et al. and Polavarapu et al. recently suggested that this may be attributed to fetal brain-sparing.38,39 That fetal
hemodynamic redistribution occurs in severe placental insufficiency as a compensatory response to a hypoxic environment is well established. As mentioned in the introduction of this thesis, it involves peripheral vasoconstriction, cerebral vasodilation, and an increase in fetal left cardiac output, which causes more oxygenated blood to reach the brain and less to reach the periphery and ‘non-vital’ organs.40,41 Chapter 2 of this thesis supports the
association between fetal brain-sparing, (postnatal) cerebral vasodilation, increased (postnatal) cerebral oxygenation, and impaired (postnatal) cerebral autoregulation. We further demonstrate that fetal brain-sparing may strongly confound effects attributed to maternal labetalol or MgSO4. Although fetal
cerebral vasodilation in FGR increases cerebral oxygen supply and protects the brain from fetal hypoxia, it may also cause (postnatal) cerebral hyperoxia, which – depending on gestational age – may be harmful. Moreover, its interference with cerebral autoregulation predisposes to fluctuations in cerebral perfusion
associated with transition or maternal antihypertensive medication. This may be attributed to vascular remodeling and a loss of myogenic vasoreactivity caused by the downregulation of calcium-dependent pathways and upregulation of sympathetic control mechanisms upon chronic hypoxia as stated earlier.17,42 It has also been proposed that fetal brain-sparing and an associated increase in CBF may therefore also be the result of impaired cerebral autoregulation and not the other way around.42 Ongoing brain-sparing after birth despite resolved hypoxia may
support this theory.38,42 Moreover, cerebral autoregulation and fetal chemoreflex
responses to hypoxemia only mature in the last trimester of pregnancy, which raises the question whether early brain-sparing can indeed be an adaptive mechanism.43 However, the exact complexity of the fetal brain-sparing response
to chronic fetal hypoxia and whether it is a neuroprotective mechanism or not remains unknown.41
Sildenafil to improve placental function and perinatal oxygenation in FGR
As FGR due to placental insufficiency is associated with high fetal and neonatal morbidity and mortality, there is much interest in improving placental function, fetal growth, and maternal hypertension.44 The phosphodiesterase-5 (PDE-5)
inhibitor sildenafil has been promising in experimental and small clinical studies and treatment of FGR with sildenafil is already practice in some obstetric units.45,46 However, large-scale randomized controlled trials investigating whether sildenafil benefits or may even harm the fetus have only recently been performed as part of the international ‘STRIDER’ collaboration (Sildenafil TheRapy In Dismal Prognosis Early onset fetal growth Restriction).47 In part I, chapter 3 of this thesis, we investigated in a subset of infants whether antenatal sildenafil given within the Dutch STRIDER trial improved fetal hemodynamics in early-onset placental insufficiency or may otherwise alter neonatal hemodynamics, since PDE-5 is
GD
expressed in several tissues, including the brain, lungs, and kidneys.48 We did notfind any differences in cerebral oxygenation or autoregulation between sildenafil- versus placebo-exposed infants, suggesting that sildenafil did not affect preferential fetal hemodynamic redistribution to the brain and may therefore not significantly improve placental function, although this remains speculative. Renal rSO2 was, however, lower and renal FTOE higher on days 1 and 2 after birth, with
opposite findings on day 3. Our findings suggested a direct vasodilative effect of sildenafil on the renal (and systemic) vasculature but not the cerebral vasculature, which becomes visible on day 3 after an initial vasoconstrictive rebound. The presence of a rebound was supported by a higher (diastolic) blood pressure on day 1 after birth, which developed to be lower by day 3 in sildenafil-exposed infants. Moreover, the duration of sildenafil intake by the mother positively correlated with renal FTOE on days 2 and 3, but not day 1, suggesting a stronger but also later rebound following higher cumulative doses of sildenafil.
Since sildenafil seemed to alter renal oxygenation, kidney function may be altered as well. Observed increased urine production in these infants may be indeed caused by better perfusion and renal function in these FGR fetuses, who often present with reduced urine production and oligohydramnios.49 Likewise, altered
blood pressure may reflect altered renal function, although comparing the course of both blood pressure and renal oxygenation suggests that these have rather been two independent effects of tissue-specific sildenafil-induced vasodilation. However, urine production itself is insufficient to establish whether the hemodynamic effects of sildenafil were beneficial or not and follow-up in these infants should warrant thorough examination of renal function.50
It may be disappointing that sildenafil may not improve fetoplacental hemodynamics and cerebral autoregulation. However, unaffected cerebral
oxygenation and autoregulation may also be regarded reassuring since sildenafil crosses the blood-brain-barrier and may act on PDE-5 of the cerebral endothelium. Although hemodynamic redistribution reflects severity of FGR and fetal hypoxia, fetal brain-sparing is considered a neuroprotective response. Without simultaneous improvement of placental dysfunction, interference with fetal brain-sparing may be detrimental for neurodevelopmental outcome. Despite the fact that sildenafil has been proposed to improve neurologic function in elderly, in FGR fetuses further vasodilation of the cerebral vasculature may not be beneficial (or even possible).51 Moreover, a vasoconstrictive rebound in the brain following sildenafil withdrawal could cause severe cerebral ischemia. Therefore, a lack of a direct effect on the cerebral oxygenation and autoregulation in these infants may potentially be encouraging. Of major concern, however, was the side effect of sildenafil on the pulmonary vasculature. It is very likely that the vasoconstrictive rebound observed in our study also occurred in the lungs, interfering with pulmonary vasodilation and hemodynamic transition after birth. Accordingly, an increased incidence of persistent pulmonary hypertension (PPHN) following sildenafil was reported by the Dutch STRIDER trial (unpublished, personal communications). Furthermore, sildenafil may already cause pulmonary vasodilation in utero with the potential to interfere with normal fetal hemodynamics.
In conclusion, sildenafil may not alter or even improve fetoplacental hemodynamics in our small subcohort of the Dutch STRIDER trial as suggested by a lack of effect on the cerebrorenal oxygenation ratio. Furthermore, it seems to lack a positive effect on fetal growth but on the other hand increase the risk of PPHN and neonatal death (partially unpublished data, personal communications).52,53 Sildenafil, however, may improve renal perfusion and
GD
perfusion and autoregulation. Before any conclusions can be drawn on whethersildenafil presents a good and safe option for the treatment of placental insufficiency, a meta-analysis of these combined STRIDER trials needs to be awaited. If the results are promising, the high rates of mortality and morbidity in severe early-onset FGR may warrant further research in favor of sildenafil. First, long-term neurological, renal, and pulmonary function should, however, be evaluated. Additionally, and importantly, future trials should consider postnatal weaning off sildenafil to prevent a vasoconstrictive rebound.
Part II: Outcomes associated with fetal brain-sparing and postnatal cerebral tissue oxygenation
In part II of this thesis, we investigated the neurodevelopmental outcomes of fetal brain-sparing and postnatal cerebral oxygen saturation levels. In chapter 4, we found that a prolonged exposure to cerebral rSO2 levels above 85% were
associated with an increased risk of retinopathy of prematurity (ROP), a potentially blinding disease. Moreover, in infants with strict oxygen administration management as in our study population, cerebral rSO2 seemed
more sensitive than SaO2 to early identify infants with an increased risk of ROP.
With regards to neurodevelopmental outcome, we found that fetal brain-sparing was associated with better behavior and executive functioning at 4 years of age, but not better (or worse) IQ (chapter 5). The same applied to cerebral rSO2 equal
to or above 72% on day 1 and 77% on day 2, although cerebral rSO2 levels above
77% on day 2 were also associated with worse performance IQ. However, as particularly cerebral rSO2 levels on day 2 were a reflection of fetal brain-sparing,
higher cerebral oxygenation levels may have mediated some of the findings for fetal brain-sparing. In chapter 6, we found that fetal brain-sparing was associated with a trend towards hypermethylation of the hypoxia-response element in the
promoter regions of HIF1A (hypoxia-inducible factor 1 alpha) and VEGFA (vascular endothelial growth factor A). Moreover, we observed hypermethylation at a regulatory CREB (cAMP response element-binding protein) binding site in the promoter region of exon 4 of BDNF (brain-derived neurotrophic factor) and hypomethylation at the hypoxia-response elements of its receptor gene NTRK2. Hypermethylation of HIF1A, VEGFA, and BDNF were associated with a better verbal IQ (VIQ), lower performance IQ (PIQ), and better inhibitory self-control, respectively.
The effect of perinatal cerebral hyperoxia on neurodevelopment
Oxygen availability is not only important for cell survival but also a powerful driving force of early human development. In the fetus, a delicate balance exists between physiologic hypoxia, inducing angiogenesis, proliferation, and differentiation of cell lineages, and pathologic hypoxia, causing cell death and developmental arrest.54 With advancing development, this balance shifts, hypoxia becomes less stimulating, and higher oxygen levels are required to answer to growing tissue needs and prevent hypoxic injury.54 This has become particularly
evident for the developing retinal tissue, in which the vaso-obliterative effect of hyperoxia strongly depends on gestational age.55
This thesis supports a harmful effect of early hyperoxia on human brain development following preterm birth. Moreover, with respect to retinal development, the accepted upper limit of regional oxygen saturation seems to be around 85% as measured with a cerebral neonatal INVOS sensor in a population below a gestational age of 32 weeks (chapter 4). The 85% threshold was supported by our subanalysis, showing a weaker association between ROP and a threshold of 80%, but an even stronger association with ROP and a threshold of 90%. Verhagen et al. also previously demonstrated that cerebral oxygen
GD
saturations above 85% are related to poorer cognitive outcome following pretermbirth below 32 weeks gestational age.56 In chapter 5, the cerebral saturation threshold, above which the PIQ demonstrated to be significantly poorer, was with 77% somewhat lower. Although the cohort consisted of mixed term and preterm born children, gestational age did not seem to explain this difference. However, the threshold was based on the cut-off between the lowest and second lowest quartile, which was not verified due to a small sample size. Based on the scatterplot for the association between PIQ and cerebral saturation levels, we speculate that the threshold towards abnormal outcome may also be slightly higher. With the interpretation of the NIRS values, one needs to keep in mind, the type of NIRS sensor used, with its unique algorithms.57
As demonstrated in chapter 5, fetal brain-sparing may be an important cause of postnatal cerebral hyperoxia with saturation levels at 80-90%, which may be particularly harmful at a lower gestational age. Hyperoxia is known to cause downregulation of VEGFA.58 VEGFA has strong implications for retinal and (motor-)neuronal development due to its angiogenic and neurotrophic effects and downregulation has been implicated in the pathogenesis of both ROP and motor neuron disease.59,60 As PIQ has shown to predict and thus relate to motor
outcome, this may also explain why hyperoxia was associated with a poorer PIQ.61 Indeed, Bekkering et al. found that fetal brain-sparing was associated with placental hypermethylation at the hypoxia-response element located within the promoter region of the VEGFA in FGR neonates.62 Now we demonstrate that
within the buccal DNA of the same, slightly smaller cohort a trend towards hypermethylation at the same hypoxia-response element is still visible at 4 years of age. Moreover, hypermethylation of this region was strongly associated with poorer PIQ. Downregulation of VEGFA expression by prolonged hyperoxia during
early brain development may therefore not only disturb retinal vascularization and cell survival but possibly also neuronal development important for PIQ and motor function. The role of fetal brain-sparing in the downregulation of VEGF may also explain why in particular small-for-gestational age infants are at an increased risk of ROP.63 However, as our cohort was small, we may only speak in terms of associations between fetal brain-sparing, high cerebral oxygen levels, and hypermethylation of VEGFA, which not necessarily implies causal relation. As mentioned earlier, PE is associated with an anti-angiogenic state, which involves decreased expression of VEGF and increased circulating levels of soluble VEGF-receptor 1 (also known as soluble fms-like tyrosine kinase-1) neutralizing free VEGFA, demonstrated for placental tissue and both maternal and neonatal blood.64-67 Although PE (n=4) did not reduce the trend association between fetal
brain-sparing (n=8) and hypermethylation of VEGFA (chapter 6), this may also have been due to small sample size. It may therefore be very likely that PE may significantly confound the association between fetal brain-sparing and hypermethylation of VEGFA. Similarly, gestational age may have influenced our findings as children with fetal brain-sparing were born at lower gestational ages. Although VEGF expression seems to be constant throughout gestation, higher cerebral oxygen levels associated fetal brain-sparing may be less preferred at lower gestational ages and more quickly result in downregulation of VEGF.68
Nevertheless, VEGF has an important neurotrophic role and downregulation is likely to be associated with adverse neurodevelopmental outcome. Whether induced by (gestational age-dependent) hyperoxia or PE, downregulation of VEGF through epigenetic modification may harm the tissue by reducing levels of nitric oxide, which normally has the ability to neutralize reactive oxygen species (ROS).67,69,70 As ROS may be especially harmful to motor neurons and myelin
GD
hyperoxia and PIQ (chapter 5) and between methylation levels of VEGFA and PIQ(chapter6).71,72 Possibly, downregulated VEGF may thereby potentiate harmful effects of postnatal hyperoxia, even though the two may not be causally related. Of note, postnatal hyperoxia and not fetal brain-sparing was significantly associated with poorer PIQ in chapter 5. Interestingly, VEGF levels have also shown to be reduced in patients with autism spectrum disorder (ASD).73 ASD has been associated with placental dysfunction and fetal growth restriction and in our small cohort was reported in 15% of children.74 However, ASD patients frequently
display a PIQ > VIQ discrepancy, which seems conflicting with our findings regarding VEGF and PIQ.75,76 Since in our cohort, ASD was more common in participants without fetal brain-sparing, who were also born more closely to term, it may be possible that these children struggled with perinatal hypo- rather than hyperoxia, supporting an amplifying pathogenic link between reduced VEGF and cerebral hyperoxia as proposed above. This is, however, pure speculation and requires further investigation.
The effect of perinatal cerebral hypoxia on neurodevelopment
Chapter 5 demonstrated the absence of fetal brain-sparing and postnatal cerebral oxygen saturations below 72-77% to be related to abnormal neurodevelopmental outcome in FGR. This was mainly true for behavior and executive functioning, but due to small sample size we cannot exclude a negative effect on cognition (in particular VIQ) as well. Although fetal brain-sparing has frequently been associated with adverse perinatal and neurodevelopmental outcome, there have been few studies to assess the association between neurodevelopmental outcome and brain-sparing in FGR infants. Despite a possible negative effect on VEGFA and related brain functions, our results suggest that the presence of fetal brain-sparing in FGR benefits behavior and executive functioning at 4 years of age.
Although this partially related to a lower gestational age in these infants and thus potentially less susceptibility to cerebral hypoxia, gestational age did not appear to explain all our findings. Moreover, the examined cohort did not show any echographically evident lesions suggestive for more ischemia in FGR children without fetal brain-sparing, wherefore we suggested epigenetic modifications of neurotrophic factors to play a role. In fact, in chapter 6 we found increased methylation close to two CREB binding sites of the promoter region of BDNF exon 4 in infants with fetal brain-sparing. These binding sites are important for the activation of BDNF through erythropoietin or through neuronal activity-mediated NMDA-receptor stimulation and calcium influx.77-79 This could explain reduced methylation of these sites at low cerebral oxygen saturations, which may in particular occur in the absence of fetal brain-sparing in FGR, as hypoxia stimulates erythropoietin expression and may reduce neuronal activity. Paradoxically, hypermethylation was associated with better executive functioning although this would result in downregulation of BDNF, while one would rather expect increased BDNF activation and thus hypomethylation to be associated with better outcome, since BDNF stimulates synaptic plasticity, which is important for learning and memory.80 However, we suggested that this may relate to a positive feedback loop leading to excessive NMDA-receptor activation by BDNF and excitotoxic cell death, initiated by excess glutamate and intracellular calcium in hypoxia.81 Fetal
brain-sparing could therefore benefit neurodevelopmental outcome not by upregulation of BDNF and stimulation of synaptic plasticity, but rather through downregulation of BDNF and excitotoxic injury by reducing cerebral hypoxia. Studies reporting on increased BDNF levels following stroke or autistic patients may support our newly established associations.82,83
GD
Reducing the burden of postnatal cerebral hypo- and hyperoxia
Whether through hypoxia-induced excitotoxicity or hyperoxia-potentiated of ROS injury, our data and the results of Verhagen et al. imply that both perinatal hypo- but also hyperoxia negatively affect neurodevelopmental outcome. Moreover, our data suggests that the acceptable range of cerebral oxygen saturation levels for the brain among preterm infants below 32 weeks gestational age and FGR infants may lie within 70-85% as measured with the neonatal INVOS sensor, depending on gestational and postnatal age. Since preterm and FGR infants frequently present with cerebral saturation levels outside this range, reduction of the time spent outside these ranges seems worthwhile.
The multicenter SafeBoosC trial (Safeguarding the Brains of our smallest Children) has been the only randomized controlled trial so far studying the feasibility and outcomes of reducing the burden of cerebral hypo- and hyperoxia using NIRS in preterm infants below 28 weeks gestational age. In this trial, cerebral hypoxia was defined a cerebral rSO2 below 55% and cerebral hyperoxia as a cerebral rSO2
above 85% as measured with adult sensors.84 Their treatment guideline aimed at optimizing cerebral oxygen supply and consumption. To reduce cerebral hypoxia, evidence-based interventions were proposed to improve I) circulation (patent ductus arteriosus closure or blood pressure-increasing interventions with medication, saline, or through a reduction of mean airway pressure), II) oxygen transport (erythrocyte transfusion), or III) respiration (by increasing fractional inspired O2 (FiO2) or mean airway pressure to increase SaO2 or by decreasing
minute ventilation to increase pCO2).84 Cerebral hyperoxia was theoretically
addressed by decreasing SaO2 (by reducing FiO2 or mean airway pressure),
lowering pCO2 (by increasing minute ventilation), or increasing glucose intake in
significantly reduced the burden of hypoxia within the first 72 hours after birth by almost 40% with a trend towards improved survival and less severe brain injury as assessed by ultrasound compared to infants in whom the cerebral rSO2 was also
measured but not visible.85 It demonstrated that reducing cerebral hypoxia using NIRS was feasible, although only 25% of all alarms outside the range resulted in actions by clinicians.86 The most adopted treatment guideline was FiO
2
adjustment, while treatment of a patent ductus arteriosus and increase of glucose intake were less commonly implemented.86
Irrespective of the treatment group, the incidence of severe IVH, abnormal electroencephalography (EEG) patterns, and mortality were significantly higher in infants spending more time at cerebral hypoxic levels (4th quartile of burden of
hypoxia).87 However, they did not find a significant reduction (nor increase) of
abnormal EEG patterns, blood biomarkers of brain injury, severe brain injury as assessed by cranial ultrasound and MRI, or neurodevelopmental outcome at 2 years in the intervention group.88-90 They suggested that this may also relate to insufficient power of the trial. Also, sensor repositioning in the intervention group, which accounted for a substantial amount of reduction of hypoxia, may have overestimated the efficacy of the treatment guideline.85
The SafeBoosC trial was unable to reduce the burden of hyperoxia within the first 72 hours after birth.85 This may relate to the fact that treatment of hypoxia
may have resulted in hyperoxia instead, undetected by the predefined hyperoxic threshold, as suggested by a trend towards an increased risk of ROP and BPD in the intervention group.85 Moreover, increasing glucose intake was a rarely implemented intervention upon hyperoxia.86 It may, however, also relate to the possible causes of hyperoxia, which appeared difficult to address. As the findings reported in chapter 4 and 5 of this thesis suggest, FGR-associated brain-sparing and prematurity-related low cerebral metabolism may be important causes of
GD
high postnatal cerebral oxygen saturations. Hence, prevention of FGR andprematurity may be paramount and could possibly contribute more to a reduction of cerebral hyperoxia than available treatment guidelines.
The range of accepted cerebral rSO2 range chosen by the SafeBoosC trial was
based on published reference ranges for preterm neonates and the assumption that values outside the 95% confidence interval of these reference ranges would relate to adverse outcome.84,91 Both reference ranges and alarm limits were
determined using the adult sensor, which implements a different algorithm than the neonatal sensor based on skull thickness.57 At our unit, where a neonatal
sensor is used, this would translate to cerebral rSO2 levels of 65-95%. The results
of this thesis, however, suggest that normoxia may be found at a much narrower range. Nevertheless, the SafeBoosC trial presents highly relevant data for the practical application of NIRS in the prevention of brain damage in the postnatal period. Combining their findings with ours, it becomes clear that the (patho-) physiology behind postnatal cerebral hypo- but especially hyperoxia and associated neurodevelopmental outcomes in these infants may partially relate to intrinsic factors or prenatal events attributed to preterm birth and FGR, which are difficult to address.
Limitations of this thesis
This thesis acknowledges several limitations. First, small sample sizes in the majority of our studies limit the generalizability of our results. Second, exploratory multiple testing may have increased the chance of false positive findings. Third, there was heterogeneity concerning the causes and onset of FGR in chapter 5 and 6. Moreover, study populations in chapters 1, 2, and 4 consisted of both small- and appropriate-for-gestational age preterm infants. Fourth, only chapter 3 was part of a randomized placebo-controlled trial providing sufficient evidence about
causation behind the established associations. All together these limitations hinder straightforward analyses and interpretation of the data and largely prevent conclusions about causal relationships.
Conclusion and implications for clinical care
This thesis aimed to better understand antenatal predictors and neurodevelopmental outcomes of perinatal tissue oxygenation of preterm and FGR neonates. This is important as this patient population is particularly vulnerable to fetal and postnatal hemodynamic disturbances, which can adversely affect ongoing retinal and cerebral development. Moreover, optimal cerebral oxygenation levels are poorly defined and more knowledge regarding the causes of cerebral hypo- and hyperoxia is needed in order to reduce their burden with a positive effect on neurodevelopmental outcome. Since placental dysfunction and the prenatal exposure to antihypertensive medications are common in preterm and FGR infants, we hypothesized that they may be major predictors of postnatal cerebral and systemic hemodynamics requiring further investigation.
From this thesis we conclude that I) postnatal cerebral oxygen saturation levels between 70% and (80-)85% as measured with the neonatal INVOS sensor are related to better neurodevelopmental outcome and less severe ROP. II) compensatory fetal hemodynamic brain-sparing redistribution as a response to placental insufficiency appears to be a major contributor of increased cerebral perfusion and moderate to high perinatal cerebral oxygen saturations, which is why it may be regarded as a neuroprotective mechanism. Accordingly, we found that III) FGR with brain-sparing before birth was associated with better behavior and executive functioning at 4 years than FGR without brain-sparing, which IV) may in part be mediated by reduced hypoxia-induced excitotoxic brain injury and epigenetic modification of neurodevelopmentally important genes. Although
GD
better neurologic outcome in FGR children with brain-sparing may also relate to anon-placental etiology of FGR in children without brain-sparing, the latter were also born closer to term with potentially higher unmet cerebral oxygen demands and/or complete loss of vasoreactivity. This demands further investigation, but also better follow-up of FGR infants without fetal brain-sparing. Conversely, our and previous data suggest that fetal brain-sparing may also contribute to (postnatal) cerebral hyperoxia, which may explain an increased risk of ROP in small-for-gestational age infants. Supporting this assumption is that V) fetal brain-sparing in FGR is associated with hypermethylation of the hypoxia-response element in the promoter region of buccal VEGFA and thus possibly epigenetic downregulation of VEGFA expression. Downregulation of VEGFA may result in disturbances in angiogenesis, neurogenesis, and protection from radical oxygen species and could explain why VI) postnatal cerebral hyperoxia was associated with an increased risk of ROP and injury to brain functions related to performance IQ.
VII) PE and in particular fetal brain-sparing were associated with impaired cerebral autoregulation, which has been shown to result in pressure-passive cerebral perfusion. Although it remains unknown whether fetal brain-sparing is the cause of impaired vasoreactivity or rather the result, the association between the two is an important finding to be considered by the neonatologist. Moreover, pressure-passive cerebral perfusion and oxygenation may be particularly detrimental if other factors, such as maternal medications, cause cerebral and systemic vasodilation and thereby affect neonatal systemic blood pressure and cerebral blood flow. While our data demonstrate VIII) an association between prenatal labetalol use and an increased cerebral but decreased intestinal oxygenation, its hemodynamic impact may be rather small in comparison to the hemodynamic effects of underlying placental disease and fetal brain-sparing.
However, its anti-adrenergic effects may add to cerebral vasodilation and impaired cerebral autoregulation of infants experiencing fetal cerebrovascular remodeling due to chronic hypoxia. Conversely, we can conclude from our data that IX) MgSO4 is unlikely to affect cerebral blood flow or autoregulation in
preterm infants, if it is used for non-tocolytic purposes and does not affect blood pressure. Instead, MgSO4 increases regional oxygen availability by reducing
cerebral metabolism and oxygen consumption. This could be a pure reflection of its anticonvulsant effects counterbalancing ischemia-relating excitotoxicity but may additionally lower susceptibility to cerebral hypoxia.
Despite all attempts and progress to improve cerebral oxygenation levels and reduce ischemic and radical oxygen injury, it is obvious that cerebral hypo- and hyperoxia may best be avoided by the prevention of preterm birth and FGR. Although antenatal sildenafil citrate therapy in FGR seemed promising, this thesis demonstrates that X) sildenafil does not significantly alter the neonatal cerebrorenal oxygenation ratio and may therefore offer only little improvement of the fetoplacental perfusion. It seems to have, however, vasodilating effects on the systemic and renal vascular system and sudden withdrawal with birth can induce a powerful vasoconstrictive rebound, which may be responsible for an increased risk of PPHN in these infants. Nevertheless, little interference with cerebral autoregulation and oxygenation in the face of potentially improved renal perfusion in these infants may also be reassuring for any future perspectives of this drug.
Future perspectives
This thesis sheds more light on the pathophysiology and optimal ranges of cerebral oxygenation in preterm and FGR infants. Even so, our findings need to be reproduced and supported by larger trials to create generalizability. Moreover,
GD
this thesis raises many new questions, which provide potential for furtherresearch. These may first involve more thorough investigation into the pathophysiologic mechanisms behind fetal brain-sparing and perinatal cerebral hypo- and hyperoxia. This should also include further (epi)genetic research involving methylation and expression analysis. Potential genetic candidates may be genes contributing to ASD or white matter injury and myelination. Additionally, given the importance of VEGF in normal neurodevelopment, efforts should be made to further investigate the causal contributions of fetal brain-sparing, postnatal hyperoxia, and PE to hypermethylation of the oxygen-dependent regulatory element of VEGF and whether this results in reduced (cerebral) expression of VEGF at later age. Second, it seems worthwhile to further elucidate the neuroprotective potential of MgSO4 and to outweigh its supposedly positive
effects on cerebral oxygen availability against the negative effects of impaired cerebral autoregulation and cerebral hypoxia, such as seen in infants with hypotension or infants born following severe FGR and/or asphyxia. Last but not least, follow-up studies should investigate the long-term effects of sildenafil on neurodevelopment, renal and pulmonary function, in utero effects on the pulmonary vasculature, the exact mechanism behind the supposed postnatal rebound, and whether its inability to improve placental function and fetal growth may relate to dosage. Moreover, the feasibility of continuing sildenafil after birth with step-wise reduction needs to be evaluated as it may prevent vasoconstrictive rebound and serious adverse side effects associated with withdrawal.
References
1. Haraldsson A, Geven W. Half-life of maternal labetalol in a premature infant. Pharm Weekbl Sci. 1989;11(6):229-231. 2. Mol BW, Roberts CT, Thangaratinam S,
Magee LA, De Groot CJ, Hofmeyr GJ.
Pre-eclampsia. The Lancet.
2016;387(10022):999-1011.
3. Lafayette R. The kidney in preeclampsia. Kidney Int. 2005;67(3):1194-1203.
4. Paauw ND, Lely AT. Cardiovascular sequels during and after preeclampsia. In: Sex-specific analysis of cardiovascular function. Springer; 2018:455-470.
5. Magee LA, Namouz-Haddad S, Cao V, Koren G, von Dadelszen P. Labetalol for hypertension in pregnancy. Expert opinion on drug safety. 2015;14(3):453-461. 6. Heida KY, Zeeman GG, Van Veen TR,
Hulzebos CV. Neonatal side effects of maternal labetalol treatment in severe
preeclampsia. Early Hum Dev.
2012;88(7):503-507.
7. Verhagen EA, Kooi EM, van den Berg PP, Bos AF. Maternal antihypertensive drugs may influence cerebral oxygen extraction in preterm infants during the first days after birth. J Matern Fetal Neonatal Med. 2013;26(9):871-876.
8. Thewissen L, Pistorius L, Baerts W, Naulaers G, Van Bel F, Lemmers P. Neonatal haemodynamic effects following foetal exposure to labetalol in hypertensive disorders of pregnancy. The Journal of Maternal-Fetal & Neonatal Medicine. 2017;30(13):1533-1538.
9. Caicedo A, Thewissen L, Naulaers G, Lemmers P, Van Bel F, Van Huffel S. Effect
of maternal use of labetalol on the cerebral autoregulation in premature infants. In: Oxygen transport to tissue XXXV. Springer; 2013:105-111.
10. Van Beek AH, Claassen JA, Rikkert MGO, Jansen RW. Cerebral autoregulation: An overview of current concepts and methodology with special focus on the elderly. Journal of Cerebral Blood Flow & Metabolism. 2008;28(6):1071-1085. 11. Claassen JA, Meel-van den Abeelen, Aisha
SS, Simpson DM, Panerai RB, International Cerebral Autoregulation Research Network (CARNet). Transfer function analysis of dynamic cerebral autoregulation: A white paper from the international cerebral autoregulation research network. Journal of Cerebral Blood Flow & Metabolism. 2016;36(4):665-680.
12. Kooi EM, Verhagen EA, Elting JWJ, et al. Measuring cerebrovascular autoregulation in preterm infants using near-infrared spectroscopy: An overview of the
literature. Expert review of
neurotherapeutics. 2017;17(8):801-818. 13. Caicedo A, Varon C, Thewissen L, et al.
Influence of the maternal use of labetalol on the neurogenic mechanism for cerebral autoregulation assessed by means of NIRS. Adv Exp Med Biol. 2014;812:173-179. 14. Fisher SJ. Why is placentation abnormal in
preeclampsia? Am J Obstet Gynecol. 2015;213(4):S115-S122.
15. Ainslie PN, Brassard P. Why is the neural control of cerebral autoregulation so controversial? F1000Prime Rep. 2014;6:14. doi: 10.12703/P6-14.
GD
16. Frösen J, Joutel A. Smooth muscle cells of intracranial vessels: From development to disease. Cardiovasc Res. 2018;114(4):501-512.
17. Longo LD, Pearce WJ. Fetal cerebrovascular acclimatization responses to high-altitude, long-term hypoxia: A model for prenatal programming of adult disease? American Journal of Physiology-Regulatory, Integrative and Comparative Physiology. 2005;288(1):R16-R24.
18. Guclu S, Gol M, Saygili U, Demir N, Sezer O, Baschat AA. Nifedipine therapy for preterm labor: Effects on placental, fetal cerebral and atrioventricular doppler parameters in the first 48 hours. Ultrasound Obstet Gynecol. 2006;27(4):403-408.
19. De Heus R, Mulder EJ, Derks JB, Visser GH. The effects of the tocolytics atosiban and nifedipine on fetal movements, heart rate and blood flow. The Journal of Maternal-Fetal & Neonatal Medicine. 2009;22(6):485-490.
20. Grzesiak M, Ahmed RB, Wilczynski J. 48-hours administration of nifedipine in spontaneous preterm labor - doppler blood flow assessment of placental and fetal
circulation. Neuro Endocrinol Lett.
2013;34(7):687-692.
21. Namazov A, Grin L, Karakus R, Uludogan M, Ayvaci H. An effect of maternal nifedipine therapy on fetoplacental blood flow: A prospective study. Arch Gynecol Obstet. 2018;298(4):685-688.
22. Flenady V, Wojcieszek AM, Papatsonis DN, et al. Calcium channel blockers for inhibiting preterm labour and birth.
Cochrane Databse Syst Rev.
2014(6):CD002255. doi: 10.1002/14651858.CD002255.pub2.
23. Van Vliet EO, Nijman TA, Schuit E, et al. Nifedipine versus atosiban for threatened preterm birth (APOSTEL III): A multicentre, randomised controlled trial. The Lancet. 2016;387(10033):2117-2124.
24. Nijman T, Goedhart M, Naaktgeboren C, et al. Effect of nifedipine and atosiban on perinatal brain injury: Secondary analysis of the APOSTEL-III trial. Ultrasound in Obstetrics & Gynecology. 2018;51(6):806-812.
25. Hawkins J, Wells C, Casey B, McIntire D, Leveno K. A randomized double-blinded placebo-controlled trial of nifedipine for acute tocolysis of preterm labor. Obstet Gynecol. 2019;220(1):S59. doi: 10.1016/j.ajog.2018.11.081.
26. Clark SM, Dunn HE, Hankins GD. A review of oral labetalol and nifedipine in mild to moderate hypertension in pregnancy. Semin Perinatol. 2015;39(7):548-555. 27. de Heus R, Mulder EJ, Visser GH.
Management of preterm labor: Atosiban or
nifedipine? Int J Womens Health.
2010;2:137-142.
28. Euser AG, Cipolla MJ. Magnesium sulfate for the treatment of eclampsia: A brief review. Stroke. 2009;40(4):1169-1175. 29. Mittendorf R, Dammann O, Lee KS. Brain
lesions in newborns exposed to high-dose magnesium sulfate during preterm labor. J Perinatol. 2006;26(1):57-63.
30. Doyle LW, Crowther CA, Middleton P, Marret S, Rouse D. Magnesium sulphate for women at risk of preterm birth for neuroprotection of the fetus. Cochrane Database Syst Rev. 2009(1):CD004661. doi: 10.1002/14651858.CD004661.pub3.
31. Crowther CA, Middleton PF, Voysey M, et al. Assessing the neuroprotective benefits for babies of antenatal magnesium sulphate: An individual participant data
meta-analysis. PLoS medicine.
2017;14(10):e1002398.
32. Bachnas MA, Akbar MIA, Dachlan EG, Dekker G. The role of magnesium sulfate (MgSO4) in fetal neuroprotection. J Matern Fetal Neonatal Med. 2019:1-13.
33. Stark MJ, Hodyl NA, Andersen CC. Effects of antenatal magnesium sulfate treatment for neonatal neuro-protection on cerebral oxygen kinetics. Pediatr Res. 2015;78(3):310-314.
34. Cipolla MJ. Perivascular innervation. In: Granger D, Granger J, eds. The cerebral circulation. 2nd ed. San Rafael, CA: Morgan & Claypool Life Sciences; 2009:15-18.
35. Matei D, Luca C, Andrițoi D, Sărdaru D,
Corciovă C. The relationship between lower serum magnesium levels and heart rate variability indices. Balneo Research Journal. 2018;9(4):426-432.
36. Terstappen F, Paauw ND, Alderliesten T, et al. Elevated renal tissue oxygenation in premature fetal growth restricted neonates: An observational study. PLoS
One. 2018;13(9). doi: 10.1371/journal.pone.0204268.
37. Cohen E, Baerts W, Caicedo Dorado A, Naulaers G, van Bel F, Lemmers PMA. Cerebrovascular autoregulation in preterm fetal growth restricted neonates. Arch Dis Child Fetal Neonatal Ed. 2018;104(5):F467-F472.
38. Tanis JC, Boelen MR, Schmitz DM, et al. Correlation between doppler flow patterns in growth-restricted fetuses and neonatal
circulation. Ultrasound Obstet Gynecol. 2016;48(2):210-216.
39. Polavarapu SR, Fitzgerald GD, Contag S, Hoffman SB. Utility of prenatal doppler ultrasound to predict neonatal impaired cerebral autoregulation. Journal of Perinatology. 2018;38(5):474-481.
40. Verburg BO, Jaddoe VW, Wladimiroff JW, Hofman A, Witteman JC, Steegers EA. Fetal hemodynamic adaptive changes related to intrauterine growth: The generation R study. Circulation. 2008;117(5):649-659. 41. Giussani DA. The fetal brain sparing
response to hypoxia: Physiological
mechanisms. J Physiol (Lond ).
2016;594(5):1215-1230.
42. Cohen E, Baerts W, van Bel F. Brain-sparing in intrauterine growth restriction: Considerations for the neonatologist. Neonatology. 2015;108(4):269-276.
43. Fletcher AJ, Gardner DS, Edwards CMB, Fowden AL, Giussani DA. Development of the ovine fetal cardiovascular defense to hypoxemia towards full term. American Journal of Physiology-Heart and Circulatory Physiology. 2006;291(6):H3023-H3034.
44. Pels A, Beune IM, van Wassenaer‐Leemhuis
AG, Limpens J, Ganzevoort W. Early‐onset fetal growth restriction: A systematic review on mortality and morbidity. Acta Obstet Gynecol Scand. 2019.
45. Panda S, Das A, Md Nowroz H. Sildenafil citrate in fetal growth restriction. J Reprod Infertil. 2014;15(3):168-169.
46. Paauw ND, Terstappen F, Ganzevoort W, Joles JA, Gremmels H, Lely AT. Sildenafil during pregnancy: A preclinical meta-analysis on fetal growth and maternal
GD
blood pressure. Hypertension. 2017;70(5):998-1006.
47. Pels A, Kenny LC, Alfirevic Z, et al. STRIDER (sildenafil TheRapy in dismal prognosis early onset fetal growth restriction): An international consortium of randomised placebo-controlled trials. BMC pregnancy and childbirth. 2017;17(1). doi: 10.1186/s12884-017-1594-z.
48. Lin C. Tissue expression, distribution, and regulation of PDE5. Int J Impotence Res. 2004;16:S8-S10.
49. Lee S, Walker SP. The role of ultrasound in the diagnosis and management of the growth restricted fetus. Australas J Ultrasound Med. 2010;13(3):31-36.
50. Libório AB, Branco KM, Torres de Melo Bezerra C. Acute kidney injury in neonates: From urine output to new biomarkers. Biomed Res Int. 2014;2014. doi: 10.1155/2014/601568.
51. Heckman P, Wouters C, Prickaerts J. Phosphodiesterase inhibitors as a target for cognition enhancement in aging and alzheimer’s disease: A translational overview. Curr Pharm Des. 2015;21(3):317-331.
52. Sharp A, Cornforth C, Jackson R, et al. Maternal sildenafil for severe fetal growth restriction (STRIDER): A multicentre, randomised, placebo-controlled, double-blind trial. The Lancet Child & Adolescent Health. 2018;2(2):93-102.
53. Groom KM, McCowan LM, Mackay LK, et
al. STRIDER NZAus: A multicentre
randomised controlled trial of sildenafil therapy in early-onset fetal growth restriction. BJOG. 2019;126(8):997-1006.
54. Fathollahipour S, Patil PS, Leipzig ND. Oxygen regulation in development: Lessons from embryogenesis towards tissue engineering. Cells Tissues Organs. 2018;205(5-6):350-371.
55. Gu X, Samuel S, El-Shabrawey M, et al. Effects of sustained hyperoxia on revascularization in experimental retinopathy of prematurity. Invest Ophthalmol Vis Sci. 2002;43(2):496-502. 56. Verhagen EA, Van Braeckel KN, van der
Veere CN, et al. Cerebral oxygenation is associated with neurodevelopmental outcome of preterm children at age 2 to 3
years. Dev Med Child Neurol.
2015;57(5):449-455.
57. Dix LM, van Bel F, Baerts W, Lemmers PM. Comparing near-infrared spectroscopy devices and their sensors for monitoring regional cerebral oxygen saturation in the neonate. Pediatr Res. 2013;74(5):557-563. 58. Hosford GE, Olson DM. Effects of hyperoxia
on VEGF, its receptors, and HIF-2α in the
newborn rat lung. American Journal of Physiology-Lung Cellular and Molecular Physiology. 2003;285(1):L161-L168.
59. Carmeliet P, de Almodovar CR. VEGF ligands and receptors: Implications in
neurodevelopment and neurodegeneration. Cellular and Molecular
Life Sciences. 2013;70(10):1763-1778. 60. Cavallaro G, Filippi L, Bagnoli P, et al. The
pathophysiology of retinopathy of prematurity: An update of previous and recent knowledge. Acta Ophthalmol. 2014;92(1):2-20.
61. Kopp S, Beckung E, Gillberg C. Developmental coordination disorder and other motor control problems in girls with
autism spectrum disorder and/or attention-deficit/hyperactivity disorder. Res Dev Disabil. 2010;31(2):350-361.
62. Bekkering I, Leeuwerke M, Tanis JC, et al. Differential placental DNA methylation of VEGFA and LEP in small-for-gestational age fetuses with an abnormal cerebroplacental
ratio. PLoS One. 2019;14(8). doi:
10.1371/journal.pone.0221972.
63. Dhaliwal CA, Fleck BW, Wright E, Graham C, McIntosh N. Retinopathy of prematurity in small-for-gestational age infants compared with those of appropriate size for gestational age. Arch Dis Child Fetal Neonatal Ed. 2009;94(3):F193-F195. 64. Purwosunu Y, Sekizawa A, Yoshimura S, et
al. Expression of angiogenesis-related genes in the cellular component of the blood of preeclamptic women. Reproductive sciences. 2009;16(9):857-864. 65. Andraweera P, Dekker G, Laurence J,
Roberts C. Placental expression of VEGF family mRNA in adverse pregnancy outcomes. Placenta. 2012;33(6):467-472. 66. Hentges CR, Silveira RC, Procianoy RS.
Angiogenic and antiangiogenic factors in preterm neonates born to mothers with and without preeclampsia. Am J Perinatol. 2015;32(12):1185-1190.
67. Jim B, Karumanchi SA. Preeclampsia: Pathogenesis, prevention, and long-term complications. . 2017;37(4):386-397. 68. Hoffmann P, Feige J, Alfaidy N. Placental
expression of EG-VEGF and its receptors PKR1 (prokineticin receptor-1) and PKR2 throughout mouse gestation. Placenta. 2007;28(10):1049-1058.
69. Wink DA, Miranda KM, Espey MG, et al. Mechanisms of the antioxidant effects of
nitric oxide. Antioxidants and redox signaling. 2001;3(2):203-213.
70. Hou H, Hammock BD, Su K, et al. N-terminal domain of soluble epoxide hydrolase negatively regulates the VEGF-mediated activation of endothelial nitric oxide synthase. Cardiovasc Res. 2011;93(1):120-129.
71. Rao SD, Yin HZ, Weiss JH. Disruption of glial glutamate transport by reactive oxygen species produced in motor neurons. J Neurosci. 2003;23(7):2627-2633.
72. Ravera S, Bartolucci M, Cuccarolo P, et al. Oxidative stress in myelin sheath: The other face of the extramitochondrial oxidative phosphorylation ability. Free Radic Res. 2015;49(9):1156-1164.
73. Emanuele E, Orsi P, Barale F, di Nemi SU, Bertona M, Politi P. Serum levels of vascular endothelial growth factor and its receptors in patients with severe autism. Clin Biochem. 2010;43(3):317-319.
74. Walker CK, Krakowiak P, Baker A, Hansen RL, Ozonoff S, Hertz-Picciotto I. Preeclampsia, placental insufficiency, and autism spectrum disorder or developmental delay. JAMA pediatrics. 2015;169(2):154-162.
75. Barnea-Goraly N, Kwon H, Menon V, Eliez S, Lotspeich L, Reiss AL. White matter structure in autism: Preliminary evidence from diffusion tensor imaging. Biol Psychiatry. 2004;55(3):323-326.
76. Charman T, Jones CR, Pickles A, Simonoff E, Baird G, Happé F. Defining the cognitive phenotype of autism. Brain Res. 2011;1380:10-21.
77. Fang H, Chartier J, Sodja C, et al. Transcriptional activation of the human
