Nine de Planque

A RADIOLOGIC INVESTIGATION NINE DE PLANQUE CRANIOSYNOSTOSIS: PRIMARY AND SECONDARY BRAIN ANOMALIES

Craniosynostosis: primary and secondary brain anomalies A radiologic investigation C.A. de Planque

Colofon The studies described in this academic dissertation were performed at the Craniofacial Research Unit of the Department of Plastic and Reconstructive Surgery, Erasmus University Medical Center, Rotterdam, The Netherlands. The research has been supported by a grant of the Sophia Stichting Wetenschappelijk Onderzoek (project number: B-16-03a). The production of this book has been financially supported by Landelijke Patienten- en Oudervereniging voor Schedel- en/of Aangezichtsaandoeningen (LAPOSA) and the Traumaplatform Foundation. Provided by thesis specialist Ridderprint, ridderprint.nl Author: C.A. de Planque Printing: Ridderprint Layout and design: Eduard Boxem, persoonlijkproefschrift.nl Book Cover: Robin Edelkoort Cover Lay Out: Robin Edelkoort ISBN: 978-94-6458-727-2

Craniosynostosis: primary and secondary brain anomalies A radiologic investigation Proefschrift Craniosynostose: intrinsieke en extrinsieke brein afwijkingen Een radiologisch onderzoek Ter verkrijging van de graad van doctor aan de Erasmus Universiteit Rotterdam op gezag van de rector magnificus prof.dr. A.L. Bredenoord en volgens besluit van het College voor Promoties De openbare verdediging zal plaatsvinden op 9 december 2022 door Catherine Annemiek de Planque geboren te Rotterdam

PROMOTIECOMMISSIE Promotor: Prof. Dr. I.M.J. Mathijssen Overige leden: Prof. Dr. M. de Bruijne Prof. Dr. C.M.F. Dirven Prof. Dr. R.C. Tasker Copromotor: Dr. M.L.C. van Veelen

INDEX Chapter 1 General introduction 10 Part 1 Brain imaging in non-operated infants with isolated and syndromic craniosynostosis Chapter 2 Using perfusion contrast for spatial normalization of ASL MRI images in a pediatric craniosynostosis population 32 Chapter 3 Cerebral blood flow of the frontal lobe in untreated children with trigonocephaly vs healthy controls - an arterial spin labeling study 50 Chapter 4 Reply: discussion cerebral blood flow of the frontal lobe in untreated children with trigonocephaly versus healthy controls: an arterial spin labeling study 64 Chapter 5 A diffusion tensor imaging analysis of frontal lobe white matter microstructure in trigonocephaly patients 70 Chapter 6 A diffusion tensor imaging analysis of white matter microstructures in non-operated craniosynostosis patients 92 Part 2 Brain imaging during follow-up of operated children with isolated and syndromic craniosynostosis Chapter 7 Cortical thickness in Crouzon-Pfeiffer syndrome: findings in relation to primary cranial vault expansion 110 Chapter 8 The course and interaction of ventriculomegaly and cerebellar tonsillar herniation in Crouzon syndrome over time 130 Chapter 9 Clinical signs, interventions, and outcomes of three different treatment protocols in patients with Crouzon syndrome and acanthosis nigricans 150 Chapter 10 General discussion 166 Chapter 11 Summary 180 Chapter 12 Nederlandse samenvatting 186 Appendices List of publications 193 PhD portfolio 194 Curriculum vitae 197 Dankwoord 198

Aan mijn ouders.

1 CHAPTER GENERAL INTRODUCTION

10 Chapter 1 INTRODUCTION This thesis focuses on brain anomalies in children with isolated or syndromic craniosynostosis that arise either as a primary brain development disorder or secondary to disturbed skull growth. Affected children have a higher prevalence of neurocognitive impairment and are at risk to develop intracranial hypertension (ICH), which could further impair cognitive development, cause behavioural issues, and could cause visual loss by damaging the optic nerves.1-3 4 The most important aim of surgical treatment is to strive for best possible neurocognitive outcome by reducing the risk of developing ICH. The pathogenesis of ICH and its interconnected problems, such as hydrocephalus and herniation of the cerebellar tonsils through the foramen magnum, is multifactorial and still ill understood. This thesis seeks to differentiate between primary and secondary brain anomalies in children with craniosynostosis in a search to determine best treatment strategy. Identification of primary, inborn disorders of the brain can prevent overtreatment, as it is unlikely that surgery will be of benefit. On the other hand, the identification of secondary brain disorders and their impact on outcome can guide us to a better screening policy with earlier treatment to prevent these sequelae. Better understanding of the consequences of ICH on brain development will be the scientific basis of treatment protocols, that strive to improve the quality of life of the patients. This thesis will be focusing on answering the following questions: - To what extent do primary brain abnormalities exist in non-operated isolated and syndromic craniosynostosis, looking at intracerebral blood flow and brain microstructures? (Part I) - Are there any secondary effects of ICH and treatment on the brain of operated syndromic craniosynostosis patients, focusing on ventriculomegaly, Chiari and cortical thickness? (Part II) In this introduction, craniosynostosis, intracranial hypertension, cognitive outcome and the used types of imaging are set out, finalized by an outline of each study. Craniosynostosis A newborns’ cranial vault consists of seven bones, which are separated by the metopic, two coronal, the sagittal and two lambdoid sutures.5 These open sutures allow for transient skull distortion during birth and facilitate future growth of the brain. In the first years of life, brain growth is the main incentive of skull growth. Craniosynostosis refers to this premature closure of the skull sutures, which mainly occurs around the 15th-19th week of gestation resulting in an abnormal skull shape.6, 7 The prevalence of

11 Introduction craniosynostosis is 7.2 per 10.000 live born children in the Netherlands.8 There are three types of craniosynostosis: isolated, unisutural craniosynostosis, multisutural/complex craniosynostosis and syndromic craniosynostosis. The majority of craniosynostosis is isolated, defined as the fusion of a single suture without other congenital abnormalities (Figure 1). Figure 1. Types of single suture craniosynostosis 1

12 Chapter 1 Examples of single suture synostosis include scaphocephaly (sagittal suture synostosis), trigonocephaly (metopic suture synostosis) and posterior plagiocephaly (lambdoid suture synostosis). When two of more skull sutures are fused without any other congenital abnormalities, it is defined as multisutural/complex craniosynostosis. In 24% of the cases, craniosynostosis is syndromic, which means single (usually the coronal suture) or multiple skull sutures are fused in combination with additional congenital abnormalities, such as facial anomalies and malformations of the hands and feet.9 Syndromic craniosynostosis is caused by mutations in various genes.9 The most distinct craniosynostosis syndromes are: Apert, Crouzon-Pfeiffer, Muenke and Saethre-Chotzen syndromes. The diagnosis of craniosynostosis is primarily made based on the anamnesis and physical examination and confirmed by a 3D-CT scan.10, 11 Intracranial pressure According to the Monro-Kellie doctrine, the cranial vault is a closed box and contains the sum of the 3 main components of intracranial volume: brain tissue, intracerebral blood, and cerebrospinal fluid (CSF) (Figure 2a).12 The intracranial pressure (ICP) represents the net effect of intracranial volume and content, brain compliance, plus blood- and cerebrospinal fluid dynamics in this closed box.13 14 If there is a conflict between growth in brain volume and skull restriction, it has its effect on the brain, cerebral blood volume (CBV), CSF and in consequence on the ICP.15-17 As the brain grows in a fused fault, brain growth and arterial blood supply will be maintained as much as possible, depending on compensatory mechanisms like relocating CSF to the spinal compartment and faster discharge of blood to the extra cranial compartment by collaterals (Figure 2b). In patients with craniosynostosis, skull growth may be reduced while both compensatory mechanisms may fail. Figure 2a.) Craniosynostosis Brain Figure 2b.) Craniosynostosis brain during growth Intracranial volume Looking at the skull with its intracranial volume, literature shows that initially the intracranial volume of sCS patients is normal and enlarged in Apert patients, which

13 Introduction makes a small ICV less likely to be associated with ICH.18-22 However, restricted growth with deviation in the skull OFC growth curve demonstrates to be an important clinical sign that indicates the impending or presence of ICH.23 21Also studies on the base of the skull have been undertaken, finding that the foramen magnum is smaller in sCS patients than in controls. This smaller foramen magnum may play a role in the development of ventriculomegaly and ICH. 24-29 30 27, 29 Brain tissue Focusing on brain tissue, De Jong et al. showed that brain volumes of sCS patients were not different compared to normative data at ages 1, 4, 8 and 12 years.31 Also cerebellar volume has been found to be the same in sCS patients compared to control subjects.20, 27 Intracranial blood The cerebral blood volume is a dynamic intracranial compartment, which is under physiological control by autoregulation, but also carbon dioxide level and venous outflow play a role. Vasodilatation effect of OSA Obstructive sleep apnea (OSA) is also noticeable contributor to increased cerebral blood volume in children with craniosynostosis.16 32 OSA leads to carbon dioxide retention during obstructive episodes, which causes vasodilatation and subsequently raised ICP.16 In children with multisuture or syndromic craniosynostosis, there may be a high prevalence of OSA, around 70%. The prevalence and severity are highest in patients with Apert, Crouzon and Pfeiffer syndrome.33-38 This moderate-severe OSA occurs mostly in Apert and Crouzon patients, due to midface hypoplasia with/without mandibular hypoplasia. Impaired venous outflow The jugular foramina are skull apertures fromwhich blood outflows from the intracranial compartment. The jugular foramina of sCS patients were shown to be smaller than in controls, which could cause impaired intracranial venous outflow and venous vascular collateral adaptation. Apart from Muenke syndrome, this adaption of emissary veins is seen more often in sCS patients than in normal controls. 27, 39, 40 Because patients with Muenke syndrome rarely have ICH, the relation between emissary veins and ICH implies to be more presumable. The mutation in the fibroblast growth factor receptor (FGFR) gene might also contribute to impaired venous outflow, by affecting the premature endothelial proliferation and subsequent differentiation of the sigmoid and jugular sinuses. This results in a narrowed vascular lumen of these sinuses.41, 42 1

14 Chapter 1 Cerebrospinal fluid As the third component of the intracranial brain volume, we look into CSF. CSF accumulation can occur as the result of more secretion, increased resistance to CSF circulation or reduced resorption. Most evidence of alterations of CSF volume in sCS patients focus on obstruction of outflow of CSF and reduced resorption by elevated venous pressure in the sinus.43 Both causes can occur at the same time, where several pathogenetic mechanisms can play a role. Obstruction of CSF CSF produced in the choroid plexus flows from the intracranial compartment through the foramen magnum around the spinal cord and up to the granulations of Pacchioni back into the sagittal sinus. Potential obstructions of CSF in craniosynostosis patients could be a smaller foramen magnum, compromised CSF spaces of the posterior fossa, a small fourth ventricle and tonsillar herniation. 43-46 In Rijken et al. the smaller foramen magnum was related to premature closure of the anterior and posterior intraoccipital synchondroses.28 Crowding of the posterior fossa induces a reduction in volume of the CSF cisterns.47, 48 Because of the compromised fourth ventricle in CT scan studies some authors believe in a aqueduct stenosis, while others dispute this argument based on MRI studies.49 Tonsillar herniation, which is commonly observed in Crouzon-Pfeiffer patients, may affect the CSF flow at the level of the craniocervical junction and/or cisterna magna. 43 44 Whether tonsillar herniation develops as a result of elevated ICP or causes the ICP is an ongoing debate, but a higher cerebellar volume / posterior fossa volume ratio has been found to be a predisposing factor for the development of tonsillar herniation. 20, 29 At last, some studies have shown that premature closure of the lambdoid sutures is associated with development of tonsillar herniation.44, 50 The theory of mechanical CSF outflow obstruction remains challenged by the fact that posterior fossa decompressions could fail to sufficiently restore normal CSF circulation and, in several cases, tonsillar herniation arises without ventriculomegaly.45, 46 Reduced resorption of CSF Venous hypertension leads directly to ICH via increased cerebral blood volume and indirectly via venous hypertension-induced diminution of CSF absorption and an increase in brain interstitial fluid volume.14, 43 Normally the CSF pressure is slightly higher than the venous capillary bed pressure to maintain a gradient to drive the reabsorption of CSF. Increased venous pressure impairs this reabsorption of CSF.51 Impaired CSF absorption due to venous hypertension typically causes general dilatation of the inner and outer CSF spaces and progressive head enlargement if the skull is still capable of expanding. At the same time compensatory mechanisms as venous collaterals will develop or expand. In patients with craniosynostosis, the fused sutures do not allow

15 Introduction any ventricular dilatation, what could result in high CSF pressure with normal or even small ventricles. 14 If the skull then is opened by a vault expansion, a ‘pseudo-tumour’ /hydrocephalus of venous origin could occur postoperatively. Prevalence of Intracranial hypertension Isolated craniosynostosis Based on invasive ICP measurement or the presence of papilledema, the prevalence of ICH in isolated synostosis has been described.52-59 The prevalence of preoperatively increased ICP in sagittal suture synostosis probably ranges from 2.5 to 14% and increases with age. The prevalence of preoperatively increased ICP in metopic suture synostosis probably ranges from 2 to 8%, and for isolated coronal suture synostosis around 16%.8, 52-58, 60 The prevalence of increased ICP during follow-up after cranial correction in isolated craniosynostosis probably varies from 2 to 9% for sagittal suture synostosis, and around 1.5% for metopic suture synostosis.57, 58, 60-62 Corresponding figures for isolated coronal suture synostosis are not known. Syndromic craniosynostosis For syndromic craniosynostosis the following studies defined ICH by invasive preoperative ICP measurement, the presence of papilledema, an abnormal VEP, a deflecting head circumference curve or extensive endocortical erosion on CT images. 53, 63-68 One study uses a set of symptoms to determine increased ICP, namely papilledema, an abnormal VEP scan, a tense fontanel, progressive ventriculomegaly and invasive ICP measurement. 69 The prevalence of increased ICP in syndromic craniosynostosis before cranial surgery is likely to be 9 to 83% for Apert, 53 to 64% for Crouzon, 19 to 43% for Saethre-Chotzen and 0 to 4% for Muenke syndrome.64, 65, 67-69 The prevalence of increased ICP in syndromic craniosynostosis after cranial surgery is likely to be 35 to 45% for Apert, 20 to 47% for Crouzon, 17 to 42% for Saethre-Chotzen and 0 to 5% for Muenke syndrome, 58 to 67% for multisuture craniosynostosis and around 31% for bicoronal synostosis.53, 63-65, 68, 69 Signs and screening on Intracranial Hypertension Because ICH could impair cognitive development, cause behavioural issues, and visual loss by damaging the optic nerves, it is important to recognize signs of ICH as early as possible.1-3 4 Raised ICP is defined as increased pressure on invasive measurement, but to avoid invasive measurement a set of indicators is used to assess the risk of ICH, namely hydrocephalus, papilledema, and indirect signs on radiological images.70 At the Erasmus MC depending on suture involvement and syndromic/nonsyndromic synostosis, head circumference, fundoscopy or optical coherence tomography (OCT) are used to screen for ICH. 1

16 Chapter 1 On radiologic images intrinsic pressure from the brain could be seen by resorption of the bone, resulting in scalloping of the inner cortex and ultimately leading to full thickness defects. Also, herniation of the cerebellar tonsils through the foramen magnum and prominent venous collaterals could arise. At last, a decrease in central and peripheral CSF spaces could be seen as a sign of intrinsic pressure. Figure 3 shows an radiologic images of each of these examples. Figure 3. a.) scalloping skull b.) tonsillar herniation c.) venous collaterals. D) decrease of CSF central and peripheral Treatment To prevent/treat ICH is the main goal of surgical intervention in craniosynostosis.53, 71, 72 The timing of surgery is therefore diagnosis dependent. At the Erasmus MC, patients are surgically treated within their first year of life. Within our clinical protocol patients with Apert and Crouzon-Pfeiffer will be surgically treated by a vault expansion at the age of 6 months. An occipital expansion with springs is the primary choice because it allows to maximally increase the ICV and preserves the facial profile which facilitates a monobloc advancement, Le Fort III or facial bipartition in a later stage.73 In patients with severe exorbitism, a monobloc advancement with distraction is considered as first cranial vault expansion. In Saethre Chotzen and Muenke patients, retrusion of the orbital bar is a main clinical feature and therefore a fronto-orbital advancement will be performed in their first year of life: Saethre-Chotzen between 6-9 months of age, and Muenke between 9-12 months of age. Muenke patients have a low risk of developing ICH and therefore a better esthetical outcome of the facial appearance will be when operated at a later age.74-76 Occasionally Saethre-Chotzen patients need an additional vault expansion such as an occipital expansion or biparietal widening. In patients with complex craniosynostosis the choice of the procedure depends on the sutures that are involved. If coronal sutures have closed prematurely, a fronto-orbital advancement is the procedure of first choice, while in patients with involvement of

17 Introduction lambdoid sutures, an occipital vault expansion is typically performed. Nevertheless, in case of severe exorbitism, visual loss and/or severe OSA, a monobloc advancement with distraction should be performed as first procedure. If there is proof of elevated intracranial pressure, surgery will be performed at an earlier stage in all patients. Treatment focuses on the cause which appears to be clinically the major contributor to the increase in intracranial pressure. As described above, causes for syndromic patients could be a too small intracranial volume, moderate to severe obstructive sleep apnea, hydrocephalus and venous intracranial hypertension or a combination of these factors.4, 23, 77 Cognitive outcome sCs patients are at risk of developing intellectual disabilities and problems in behavioral and emotional function. Most patients with Crouzon-Pfeiffer and Saethre-Chotzen have a long-term intellectual outcome within the normal limits, while patients with Apert Syndrome have typically an IQ below the norm.68 In addition, a large group of patients within all 3 syndromes has an IQ that is 2SD or more below the normal limits. This means that a large proportion of patients cannot work and live independently. Trigonocephaly patients are at risk of developing mental deficiencies/disorders, behavioral problems, delays in speech and language.78, 79 However, preoperatively less than 2% of the trigonocephaly patients have papilledema as a sign of intracranial hypertension.59 Whether these derangements are caused by a primary disturbances in brain development or arise secondarily due to the synostosis is unknown.5 Is this neurocognitive disorder a consequence of mechanical distortion of the brain due to abnormal shape, ventriculomegaly and/or cerebellar tonsillar herniation, or does this finding reflect an intrinsic inborn brain problem? 80-85 While the most important aim of surgical treatment is to strive for best possible neurocognitive outcome by treating and reducing the risk of developing ICH, it remains unknown what the added value of surgery is in particularly trigonocephaly with respect to future brain development. To answer the questions in this thesis regarding to the intrinsic brain abnormalities and the potential effect of ICH, MRI brain imaging techniques have been used in both non-operated as operated craniosynostosis patients to investigate cerebral blood flow, microstructures, and cortical thickness. In the next chapter the used MRI brain imaging techniques are set out. MRI physics used in the upcoming studies Magnetic resonance imaging (MRI) is an imaging technique which can be used as a tool to image and investigate body regions or tissues, including the brain. MRI scanners 1

18 Chapter 1 use strong magnetic fields and radio waves to generate images, which in this thesis will be used to investigate images of the brain in children with craniosynostosis. T1 weighted image A T1-weighted (T1W) image is a basic MRI pulse sequence and depicts differences in signal based upon intrinsic T1 relaxation time of various tissues (Figure 4). Clinically, T1-weighted images are used for depicting the anatomy of the brain, where tissue fluid characteristics are presented in grey scale differences. In the upcoming studies we will use T1w images to measure anatomical structures i.e. the size of ventricles, the herniation of the cerebellar tonsils and the thickness of the cortex. Arterial spin labeling Two studies will use Arterial Spin Labeling (ASL). ASL is a MRI technique that provides injection-free measurements of the absolute brain perfusion by magnetically labelling water protons in the blood vessels of the patient.11 ASL is used for several pediatric imaging applications, for example in vascular diseases, tumors, epilepsy and seizures, to detect cortical hyper perfusion or hypoperfusion.86 The main advantage of the ASL technique is that it measures absolute brain perfusion and that it is does not require administration of an exogenous contrast agent.87 ASL magnetically labels arterial blood water when it flows through the neck perpendicular to the cervical arteries. By a radiofrequency pulse the water protons in the blood will get inverted; the blood will get a negative magnetization.88, 89 After a post-labeling delay (PLD) in which the labelled blood flows into brain tissue, an image of the brain will be acquired (Figure 5).90 As depicted in Figure 6, by making two images a third image can be provided with the weighted perfusion.91 The control image is the image without labelling of the blood. The second image contains both the normal stationary blood as well as the negative magnetization from the inflowing labeled blood.92 Therefore, an ASL image is weighted for the cerebral blood flow (CBF), i.e. the volume of blood that flows through 100 grams of brain tissue each minute (mL/100g/min). Figure 4. T1-weighted image.

19 Introduction Figure 5. Principle for arterial spin labeling acquisition. Figure 6. In ASL, a perfusion-weighted (ΔM) image is obtained by subtraction of a labeled from a control image. 1

20 Chapter 1 Diffusion Tensor Imaging To investigate the microstructure of the white matter of the brain, we will use Diffusion Tensor Imaging (DTI). DTI is a MRI technique that uses movement of water molecules in the brain to produce neural tract images. By evaluating and quantifying diffusion restriction and limited movement of water in white matter of the brain the pattern of neural networks in the brain could be revealed (Figure 7). In this way, the microarchitecture of the white matter tracts in the brain can be investigated. Figure 7. Tractography of bundles of axons in the brain. The main direction of diffusion is encoded in the colour of tracks with blue representing tracts coursing up and down, green representing tracts coursing from front-to-back and left representing tracts coursing left to right. To understand the technique behind DTI, we have to zoom into diffusion. In general, diffusion movement in all directions (for example in a glass of pure water) is called isotropy. Diffusion in tissues varies with direction, which is called anisotropy. In white matter of the brain, the diffusion anisotropy is primarily caused by cellular membranes, with some contribution frommyelination and the packing of axons. Anisotropic diffusion can indicate the underlying tissue orientation (Figure 8).

21 Introduction Figure 8. Representation of A.) diffusion as an isotropic shape of the corresponding diffusion lengths (λ1, λ2, and λ3), the eigenvalues. B.) diffusion as an anisotropic shape of the corresponding eigenvalues C.) fibre tracking by using the different voxels. The white matter metrics from DTI, voxel-by-voxel, are mathematically based on 3 mutually perpendicular eigenvectors, whose magnitude is given by 3 corresponding eigenvalues sorted in order of decreasing magnitude as ʎ1, ʎ2 and ʎ3. An ellipsoid is created by the long axis of ʎ1, and the small axes ʎ2 and ʎ3, from where the measured length of the three axes are the eigenvalues. DTI describes the magnitude, the degree and orientation of diffusion anisotropy. These eigenvalues are used to generate quantitative maps of fractional anisotropy (FA), the derivation of axial diffusivity (AD), radial diffusivity (RD) and mean diffusivity (MD). FA represents the amount of diffusional asymmetry in a voxel, which is presented from 0 (infinite isotropy) to 1 (infinite anisotropy). AD stands for the diffusivity along the neural tract: ʎ1. The diffusivity of the minor axes, ʎ2 and ʎ3, is called the perpendicular or radial diffusivity. The mean of these diffusivity ʎ1, ʎ2 and ʎ3 is known as MD. FA, MD, AD and RD are used as indirect markers of white matter microstructure of these young patients.93 To summarize, DTI maps the course of the neural axon bundles in the brain. With DTI we obtain a sufficient number of gradient directions to determine the tensor components per voxel. We determine eigenvalues and eigenvectors which indicate the major and minor directions of the diffusion movement down the axonal bundles. This measurement of diffusion of water in tissue could be used to investigate the microarchitecture of the white matter of the brain and produce white matter tracts. 1

22 Chapter 1 Aim and outline of this thesis The aim of this thesis is to get more understanding about the extent and the origin of brain abnormalities in children with craniosynostosis by using brain imaging. Specifically, the potential effect of mechanical causes due to shape and pressure are addressed and compared to intrinsic or genetic causes in isolated and syndromic synostosis before and after treatment. Part I Brain imaging in non-operated infants with isolated and syndromic craniosynostosis Arterial spin labeling (ASL) is a MRI technique which could be used to measure the absolute brain perfusion. In Chapter 2 a new approach of using ASL on trigonocephaly patients will be validated. In Chapter 3 blood perfusion of the frontal lobe in nonoperated trigonocephaly patients is analyzed. In Chapter 4 we respond on a letter to the editor focused on the ASL article. The microstructural properties of the frontal lobe of non-operated patients with trigonocephaly patients will be assessed in comparison to controls in Chapter 5. Chapter 6 the microstructural characteristics of the nonoperated syndromic craniosynostosis brain are described. By diffusion tensor imaging (DTI) fiber tractography diffusion parameters of multiple white matter tracts between craniosynostosis patients in comparison to controls are evaluated. Part II Brain imaging during follow-up of operated children with isolated and syndromic craniosynostosis Chapter 7 assesses the cortical thickness of the brain in Crouzon syndrome after surgical treatment. In Chapter 8 the interaction of ventriculomegaly and cerebellar tonsillar position over time and the association between abnormal anatomy of the skull base and cerebellar tonsillar position are investigated. Chapter 9 dives into the rare and severe course of patients with Crouzon syndrome and acanthosis nigricans. Treatment protocols of three international centers are reviewed in this cohort.

23 Introduction REFERENCES 1. Bartels MC, Vaandrager JM, de Jong TH, et al. Visual loss in syndromic craniosynostosis with papilledema but without other symptoms of intracranial hypertension. J Craniofac Surg 2004;15:1019-1022; discussion 1023-1014 2. Tay T, Martin F, Rowe N, et al. Prevalence and causes of visual impairment in craniosynostotic syndromes. Clin Exp Ophthalmol 2006;34:434-440 3. Mathijssen IMJ, Working Group Guideline C. Updated Guideline on Treatment and Management of Craniosynostosis. J Craniofac Surg 2021;32:371-450 4. Hayward R, Britto J, Dunaway D, et al. Connecting raised intracranial pressure and cognitive delay in craniosynostosis: many assumptions, little evidence. J Neurosurg Pediatr 2016;18:242250 5. Morriss-Kay GM, Wilkie AO. Growth of the normal skull vault and its alteration in craniosynostosis: insights from human genetics and experimental studies. J Anat 2005;207:637-653 6. Mathijssen IM, van Splunder J, Vermeij-Keers C, et al. Tracing craniosynostosis to its developmental stage through bone center displacement. J Craniofac Genet Dev Biol 1999;19:57-63 7. Connolly JP, Gruss J, Seto ML, et al. Progressive postnatal craniosynostosis and increased intracranial pressure. Plast Reconstr Surg 2004;113:1313-1323 8. Cornelissen M, Ottelander B, Rizopoulos D, et al. Increase of prevalence of craniosynostosis. J Craniomaxillofac Surg 2016;44:1273-1279 9. Sharma VP, Fenwick AL, Brockop MS, et al. Mutations in TCF12, encoding a basic helixloop-helix partner of TWIST1, are a frequent cause of coronal craniosynostosis. Nat Genet 2013;45:304-307 10. Bredero-Boelhouwer H, Treharne LJ, Mathijssen IM. A triage system for referrals of pediatric skull deformities. J Craniofac Surg 2009;20:242-245 11. Ridgway EB, Weiner HL. Skull deformities. Pediatr Clin North Am 2004;51:359-387 12. Neff S, Subramaniam RP. Monro-Kellie doctrine. J Neurosurg 1996;85:1195 13. Wagshul ME, Eide PK, Madsen JR. The pulsating brain: A review of experimental and clinical studies of intracranial pulsatility. Fluids Barriers CNS 2011;8:5 14. Sainte-Rose C, LaCombe J, Pierre-Kahn A, et al. Intracranial venous sinus hypertension: cause or consequence of hydrocephalus in infants? J Neurosurg 1984;60:727-736 15. Gonsalez S, Hayward R, Jones B, et al. Upper airway obstruction and raised intracranial pressure in children with craniosynostosis. EUR RESPIR J 1997;10:367-375 16. Hayward R, Gonsalez S. How low can you go? Intracranial pressure, cerebral perfusion pressure, and respiratory obstruction in children with complex craniosynostosis. J Neurosurg 2005;102 PEDIATRICS:16-22 17. Driessen C, Joosten KFM, Bannink N, et al. How does obstructive sleep apnoea evolve in syndromic craniosynostosis? A prospective cohort study. Archives of Disease in Childhood 2013;98:538-543 18. Gosain AK, McCarthy JG, Glatt P, et al. A study of intracranial volume in Apert syndrome. PLAST RECONSTR SURG 1995;95:284-295 19. Anderson PJ, Netherway DJ, Abbott AH, et al. Analysis of intracranial volume in apert syndrome genotypes. Pediatr Neurosurg 2004;40:161-164 20. Rijken BFM, Lequin MH, Van Der Lijn F, et al. The role of the posterior fossa in developing Chiari i malformation in children with craniosynostosis syndromes. J Cranio-Maxillofac Surg 2015;43:813-819 1

24 Chapter 1 21. Rijken BF, den Ottelander BK, van Veelen ML, et al. The occipitofrontal circumference: reliable prediction of the intracranial volume in children with syndromic and complex craniosynostosis. Neurosurg Focus 2015;38:E9 22. Breakey RWF, Knoops PGM, Borghi A, et al. Intracranial Volume and Head Circumference in Children with Unoperated Syndromic Craniosynostosis. Plast Reconstr Surg 2018;142:708e717e 23. Spruijt B, Joosten KF, Driessen C, et al. Algorithm for the Management of Intracranial Hypertension in Children with Syndromic Craniosynostosis. Plast Reconstr Surg 2015;136:331340 24. Coll G, Arnaud E, Selek L, et al. The growth of the foramen magnum in Crouzon syndrome. Child’s Nerv Syst 2012;28:1525-1535 25. Di Rocco F, Dubravova D, Ziyadeh J, et al. The foramen magnum in isolated and syndromic brachycephaly. Child’s Nerv Syst 2014;30:165-172 26. Assadsangabi R, Hajmomenian M, Bilaniuk LT, et al. Morphology of the foramen magnum in syndromic and non-syndromic brachycephaly. Childs Nervous System 2015;31:735-741 27. Coll G, Arnaud E, Collet C, et al. Skull base morphology in fibroblast growth factor receptor type 2-related faciocraniosynostosis: a descriptive analysis. Neurosurgery 2015;76:571-583; discussion 583 28. Rijken BF, Lequin MH, de Rooi JJ, et al. Foramen magnum size and involvement of its intraoccipital synchondroses in Crouzon syndrome. Plast Reconstr Surg 2013;132:993e-1000e 29. Rijken BFM, Lequin MH, Van Veelen MLC, et al. The formation of the foramen magnum and its role in developing ventriculomegaly and Chiari i malformation in children with craniosynostosis syndromes. J Cranio-Maxillofac Surg 2015;43:1042-1048 30. Tokumaru AM, Barkovich AJ, Ciricillo SF, et al. Skull base and calvarial deformities: Association with intracranial changes in craniofacial syndromes. AM J NEURORADIOL 1996;17:619-630 31. De Jong T, Rijken BFM, Lequin MH, et al. Brain and ventricular volume in patients with syndromic and complex craniosynostosis. Child’s Nerv Syst 2012;28:137-140 32. Spruijt B, Joosten KF, Driessen C, et al. Algorithm for the Management of Intracranial Hypertension in Children with Syndromic Craniosynostosis. Plast Reconstr Surg 2015;136:331340 33. Al-Saleh S, Riekstins A, Forrest CR, et al. Sleep-related disordered breathing in children with syndromic craniosynostosis. J Craniomaxillofac Surg 2011;39:153-157 34. Driessen C, Joosten KF, Bannink N, et al. How does obstructive sleep apnoea evolve in syndromic craniosynostosis? A prospective cohort study. Arch Dis Child 2013;98:538-543 35. Driessen C, Joosten KF, Florisson JM, et al. Sleep apnoea in syndromic craniosynostosis occurs independent of hindbrain herniation. Childs Nerv Syst 2013;29:289-296 36. Driessen C, Mathijssen IM, De Groot MR, et al. Does central sleep apnea occur in children with syndromic craniosynostosis? Respir Physiol Neurobiol 2012;181:321-325 37. Inverso G, Brustowicz KA, Katz E, et al. The prevalence of obstructive sleep apnea in symptomatic patients with syndromic craniosynostosis. Int J Oral Maxillofac Surg 2016;45:167169 38. Zandieh SO, Padwa BL, Katz ES. Adenotonsillectomy for obstructive sleep apnea in children with syndromic craniosynostosis. Plast Reconstr Surg 2013;131:847-852 39. Booth CD, Figueroa RE, Lehn A, et al. Analysis of the jugular foramen in pediatric patients with craniosynostosis. J Craniofac Surg 2011;22:285-288 40. Florisson JMG, Barmpalios G, Lequin M, et al. Venous hypertension in syndromic and complex craniosynostosis: The abnormal anatomy of the jugular foramen and collaterals. J Cranio-Maxillofac Surg 2015;43:312-318

25 Introduction 41. Hayward R. Venous hypertension and craniosynostosis. Childs Nerv Syst 2005;21:880-888 42. Gray JL, Kang SS, Zenni GC, et al. FGF-1 affixation stimulates ePTFE endothelialization without intimal hyperplasia. J Surg Res 1994;57:596-612 43. Collmann H, Sorensen N, Krauss J. Hydrocephalus in craniosynostosis: a review. Childs Nerv Syst 2005;21:902-912 44. Cinalli G, Renier D, Sebag G, et al. Chronic tonsillar herniation in Crouzon’s and Apert’s syndromes: the role of premature synostosis of the lambdoid suture. J Neurosurg 1995;83:575-582 45. Cinalli G, Sainte-Rose C, Kollar EM, et al. Hydrocephalus and craniosynostosis. J Neurosurg 1998;88:209-214 46. Cinalli G, Chumas P, Arnaud E, et al. Occipital remodeling and suboccipital decompression in severe craniosynostosis associated with tonsillar herniation. Neurosurgery 1998;42:66-71; discussion 71-63 47. Strahle J, Muraszko KM, Buchman SR, et al. Chiari malformation associated with craniosynostosis. Neurosurg Focus 2011;31:E2 48. Fearon JA, Rhodes J. Pfeiffer syndrome: a treatment evaluation. Plast Reconstr Surg 2009;123:1560-1569 49. Thompson DN, Harkness W, Jones BM, et al. Aetiology of herniation of the hindbrain in craniosynostosis. An investigation incorporating intracranial pressure monitoring and magnetic resonance imaging. Pediatr Neurosurg 1997;26:288-295 50. Fearon JA, Dimas V, Ditthakasem K. Lambdoid Craniosynostosis: The Relationship with Chiari Deformations and an Analysis of Surgical Outcomes. Plast Reconstr Surg 2016;137:946-951 51. Sainz LV, Zipfel J, Kerscher SR, et al. Cerebro-venous hypertension: a frequent cause of so-called “external hydrocephalus” in infants. Childs Nerv Syst 2019;35:251-256 52. Thompson DN, Malcolm GP, Jones BM, et al. Intracranial pressure in single-suture craniosynostosis. Pediatr Neurosurg 1995;22:235-240 53. Renier D, Lajeunie E, Arnaud E, et al. Management of craniosynostoses. Childs Nerv Syst 2000;16:645-658 54. Mathijssen I, Arnaud E, Lajeunie E, et al. Postoperative cognitive outcome for synostotic frontal plagiocephaly. J Neurosurg 2006;105:16-20 55. Eley KA, Johnson D, Wilkie AOM, et al. Raised intracranial pressure is frequent in untreated nonsyndromic unicoronal synostosis and does not correlate with severity of phenotypic features. Plast Reconstr Surg 2012;130:690e-697e 56. Wall SA, Thomas GP, Johnson D, et al. The preoperative incidence of raised intracranial pressure in nonsyndromic sagittal craniosynostosis is underestimated in the literature. J Neurosurg Pediatr 2014;14:674-681 57. van Veelen ML, Eelkman Rooda OH, de Jong T, et al. Results of early surgery for sagittal suture synostosis: long-term follow-up and the occurrence of raised intracranial pressure. Childs Nerv Syst 2013;29:997-1005 58. van Veelen ML, Mihajlovic D, Dammers R, et al. Frontobiparietal remodeling with or without a widening bridge for sagittal synostosis: comparison of 2 cohorts for aesthetic and functional outcome. J Neurosurg Pediatr 2015;16:86-93 59. Cornelissen MJ, Loudon SE, van Doorn FE, et al. Very Low Prevalence of Intracranial Hypertension in Trigonocephaly. Plast Reconstr Surg 2017;139:97e-104e 60. Cornelissen MJ, Loudon SE, van Doorn FEC, et al. Very Low Prevalence of Intracranial Hypertension in Trigonocephaly. Plast Reconstr Surg 2017;139:97e-104e 61. Christian EA, Imahiyerobo TA, Nallapa S, et al. Intracranial hypertension after surgical correction for craniosynostosis: a systematic review. Neurosurg Focus 2015;38:E6 1

26 Chapter 1 62. Thomas GP, Johnson D, Byren JC, et al. The incidence of raised intracranial pressure in nonsyndromic sagittal craniosynostosis following primary surgery. J Neurosurg Pediatr 2015;15:350-360 63. Thompson DN, Harkness W, Jones B, et al. Subdural intracranial pressure monitoring in craniosynostosis: its role in surgical management. Childs Nerv Syst 1995;11:269-275 64. Marucci DD, Dunaway DJ, Jones BM, et al. Raised intracranial pressure in Apert syndrome. Plast Reconstr Surg 2008;122:1162-1168 65. Kress W, Schropp C, Lieb G, et al. Saethre-Chotzen syndrome caused by TWIST 1 gene mutations: functional differentiation from Muenke coronal synostosis syndrome. Eur J Hum Genet 2006;14:39-48 66. Greene AK, Mulliken JB, Proctor MR, et al . Phenotypically unusual combined craniosynostoses: presentation and management. Plast Reconstr Surg 2008;122:853-862 67. Woods RH, Ul-Haq E, Wilkie AO, et al. Reoperation for intracranial hypertension in TWIST1confirmed Saethre-Chotzen syndrome: a 15-year review. Plast Reconstr Surg 2009;123:18011810 68. de Jong T, Bannink N, Bredero-Boelhouwer HH, et al. Long-term functional outcome in 167 patients with syndromic craniosynostosis; defining a syndrome-specific risk profile. J Plast Reconstr Aesthet Surg 2010;63:1635-1641 69. Abu-Sittah GS, Jeelani O, Dunaway D, et al. Raised intracranial pressure in Crouzon syndrome: incidence, causes, and management. J Neurosurg Pediatr 2016;17:469-475 70. Renier D, Sainte-Rose C, Marchac D, et al. Intracranial pressure in craniostenosis. J Neurosurg 1982;57:370-377 71. Mathijssen IM, Arnaud E. Benchmarking for craniosynostosis. J Craniofac Surg 2007;18:436442 72. Marchac D, Arnaud E, Renier D. Frontocranial remodeling without opening of frontal sinuses in a scaphocephalic adolescent: a case report. J Craniofac Surg 2002;13:698-705 73. de Jong T, van Veelen ML, Mathijssen IM. Spring-assisted posterior vault expansion in multisuture craniosynostosis. Childs Nerv Syst 2013;29:815-820 74. Arnaud E, Marchac D, Renier D. Reduction of morbidity of the frontofacial monobloc advancement in children by the use of internal distraction. Plast Reconstr Surg 2007;120:10091026 75. O’Connor EJF, Marucci DD, Jeelani NO, et al. Ocular advancement in monobloc distraction. Plast Reconstr Surg 2009;123:1570-1577 76. Honnebier MB, Cabiling DS, Hetlinger M, et al. The natural history of patients treated for FGFR3-associated (Muenke-type) craniosynostosis. Plast Reconstr Surg 2008;121:919-931 77. Deschamps-Braly J, Hettinger P, el Amm C, et al. Volumetric analysis of cranial vault distraction for cephalocranial disproportion. Pediatr Neurosurg 2011;47:396-405 78. van der Vlugt JJ, van der Meulen JJ, Creemers HE, et al. Cognitive and behavioral functioning in 82 patients with trigonocephaly. Plast Reconstr Surg 2012;130:885-893 79. Kelleher MO, Murray DJ, McGillivary A, et al. Behavioral, developmental, and educational problems in children with nonsyndromic trigonocephaly. J Neurosurg 2006;105:382-384 80. Vermeij-Keers C, Mazzola RF, Van der Meulen JC, et al. Cerebro-craniofacial and craniofacial malformations: an embryological analysis. Cleft Palate J 1983;20:128-145 81. O’Rahilly R, Gardner E. The initial appearance of ossification in staged human embryos. Am J Anat 1972;134:291-301 82. McBratney-Owen B, Iseki S, Bamforth SD, et al. Development and tissue origins of the mammalian cranial base. Dev Biol 2008;322:121-132

27 Introduction 83. Jiang X, Iseki S, Maxson RE, et al. Tissue origins and interactions in the mammalian skull vault. Dev Biol 2002;241:106-116 84. Merrill AE, Bochukova EG, Brugger SM, et al. Cell mixing at a neural crest-mesoderm boundary and deficient ephrin-Eph signaling in the pathogenesis of craniosynostosis. Hum Mol Genet 2006;15:1319-1328 85. Chai Y, Jiang X, Ito Y, et al. Fate of the mammalian cranial neural crest during tooth and mandibular morphogenesis. Development 2000;127:1671-1679 86. Keil VC, Hartkamp NS, Connolly DJA, et al. Added value of arterial spin labeling magnetic resonance imaging in pediatric neuroradiology: pitfalls and applications. Pediatr Radiol 2019;49:245-253 87. Wang J, Licht DJ, Jahng GH, et al. Pediatric perfusion imaging using pulsed arterial spin labeling. J Magn Reson Imaging 2003;18:404-413 88. Williams DS, Detre JA, Leigh JS, et al. Magnetic resonance imaging of perfusion using spin inversion of arterial water. Proc Natl Acad Sci U S A 1992;89:212-216 89. Detre JA, Leigh JS, Williams DS, et al. Perfusion imaging. Magn Reson Med 1992;23:37-45 90. Ferre JC, Bannier E, Raoult H, et al. Arterial spin labeling (ASL) perfusion: techniques and clinical use. Diagn Interv Imaging 2013;94:1211-1223 91. Golay X, Petersen ET. Arterial spin labeling: benefits and pitfalls of high magnetic field. Neuroimaging Clin N Am 2006;16:259-268, x 92. Alsop DC, Detre JA. Reduced transit-time sensitivity in noninvasive magnetic resonance imaging of human cerebral blood flow. J Cereb Blood Flow Metab 1996;16:1236-1249 93. Qiu A, Mori S, Miller MI. Diffusion tensor imaging for understanding brain development in early life. Annu Rev Psychol 2015;66:853-876 1

PART I BRAIN IMAGING IN NON-OPERATED INFANTS WITH ISOLATED AND SYNDROMIC CRANIOSYNOSTOSIS

2 CHAPTER USING PERFUSION CONTRAST FOR SPATIAL NORMALIZATION OF ASL MRI IMAGES IN A PEDIATRIC CRANIOSYNOSTOSIS POPULATION CATHERINE A. DE PLANQUE HENK J. M. M. MUTSAERTS VERA C. KEIL NICOLE S. ERLER MARJOLEIN H. G. DREMMEN IRENE M. J. MATHIJSSEN JAN PETR Frontiers in Neuroscience, 2021 Jul 19;15:698007

32 Chapter 2 ABSTRACT Spatial normalization is an important step for group image processing and evaluation of mean brain perfusion in anatomical regions using arterial spin labeling (ASL) MRI and is typically performed via high-resolution structural brain scans. However, structural segmentation and/or spatial normalization to standard space is complicated when gray/white matter contrast in structural images is low due to ongoing myelination in newborns and infants. This problem is of particularly clinical relevance for imaging infants with inborn or acquired disorders that impair normal brain development. We investigated whether the ASL MRI perfusion contrast is a viable alternative for spatial normalization, using a pseudo-continuous ASL acquired using a 1.5 T MRI unit (GE Healthcare). Four approaches have been compared: (1) using the structural image contrast, or perfusion contrast with (2) rigid, (3) affine, and (4) nonlinear transformations – in 16 healthy controls [median age 0.83 years, inter-quartile range (IQR) ± 0.56] and 36 trigonocephaly patients (median age 0.50 years, IQR ± 0.30) – a non-syndromic type of craniosynostosis. Performance was compared quantitatively using the realvalued Tanimoto coefficient (TC), visually by three blinded readers, and eventually by the impact on regional cerebral blood flow (CBF) values. For both patients and controls, nonlinear registration using perfusion contrast showed the highest TC, at 17.51 (CI 6.66–49.38) times more likely to have a higher rating and 17.45–18.88 ml/100 g/min higher CBF compared with the standard normalization. Using perfusion-based contrast improved spatial normalization compared with the use of structural images, significantly affected the regional CBF, and may open up new possibilities for future large pediatric ASL brain studies. Keywords: ASL, segmentation, registration, spatial normalization, pediatric, craniosynostosis

33 Using a new technique of ASL | Trigonocephaly INTRODUCTION Spatial normalization is an important step for brain image processing; it not only enables group analyses but is also required for automatic segmentation of tissue type and brain regions. Functional or physiological MRI acquisitions, such as arterial spin labeling (ASL) perfusion MRI, typically perform nonlinear registration via conventional structural – mostly T1-weighted (T1w) – scans for their higher resolution and structural contrast. However, in situations where the tissue contrast is low and changing, such as in early phases of myelination in newborns and infants, these structural reference scans may not help or even fail normalization.1-3 The use of other images with higher tissue-contrast could help registration. As an alternative to spatial normalization via segmentation and registration of structural images, studies use contrast from different MRI modalities. Feng et al. used Diffusion Tensor Imaging (DTI) as a substitute for T1w scans.4 In DTI images, premyelination is encountered prior to being detectable at T1w or T2w imaging.5 Similarly, Mutsaerts et al. used cerebral blood flow (CBF) and pseudo-CBF, created from a gray matter (GM) map from segmented T1w image to register individual ASL and T1w volumes instead of using the morphological images for the registration, for example, the ASL control images or M0 scans registered to T1w images in elderly subjects.6 This approach was especially important in cases where the image contrast difference between GM and white matter (WM) was low in ASL control images or in M0 scans, due to, for example, use of strong background suppression or short TR, respectively. This approach can be potentially extended to direct spatial normalization of ASL to standard space in the pediatric population as ASL studies of the brain show sufficient CBF contrast between GM and WM already in early age despite the potential lack of GM/WM contrast in T1w images.5 The problem with spatial normalization in subjects with ongoing myelination is of particular clinical relevance in imaging babies with inborn or acquired disorders impairing normal brain development. Craniosynostosis, referred to the premature fusion of the skull sutures leading to skull and brain deformations, is an example of such disease.7-10 Trigonocephaly, a non-syndromic type of craniosynostosis that presents within a sliding scale of severity in phenotype and brain imaging, is one of the key components in the decision making for surgical treatment in the first years of life. Imaging of these newborns is essential, and ASL is an MRI technique that could provide cerebral perfusion measurements on both a global and regional level. Spatial normalization is then necessary to be able to evaluate perfusion in predefined anatomical regions. However, automatic methods for spatial normalization are challenging in young children with craniosynostosis as there are issues with low GM/ WM contrast and skull deformity, as explained above.7, 11 In previous ASL studies in 2

34 Chapter 2 craniosynostosis patients, regions of interest (ROI) were therefore placed manually, which made spatial normalization escapable.12 However, this practice is time consuming, and it increases the likelihood of error in delineation and decreases the repeatability across subjects. To overcome the issues with spatial normalization of T1w images, we propose to use different contrast than from the structural images for the spatial normalization. In this study, we set to investigate if the ASL CBF image contrast can be directly used for spatial normalization in children with trigonocephaly and healthy controls under the age of 18 months. We combined technical and clinical expertise to compare the standard method that uses structural images for the normalization with three different registrations of ASL directly to MNI using rigid, affine, and nonlinear transformations. We hypothesize that direct ASL spatial normalization to the MNI space is possible, that nonlinear registration can be used in this context to improve the normalization quality in young healthy controls and trigonocephaly patients, and that this normalization will have a significant effect on the measured regional CBF. With this study we aim to facilitate the investigation of frontal lobe perfusion in trigonocephaly patients in a clinical setting to assess the value of vault surgery in these patients. MATERIALS AND METHODS The Ethics Committee approved this prospective imaging study in patients with trigonocephaly (METC-2018-124), which is part of ongoing work at the Erasmus Medical Center involving protocolized care, brain imaging, clinical assessment, data summary, and evaluation.13 To participate in this study, informed research consent has been obtained. Subjects Preoperative MRI brain scans from 36 children with metopic synostosis for whom a surgical correction was considered were included over a period of 2 years (2018– 2020). Surgery was considered only for moderate and severe presentation of metopic synostosis, mainly defined by severe narrowing and a protruding midline ridge of the forehead, hypotelorism (eyes close together), and biparietal widening.14 Children were less than 2 years of age at the time of the MRI brain study. The control group consisted of sixteen subjects undergoing MRI brain studies for clinical reasons, with the following inclusion criteria: (1) no neurological pathology of the head or neck (e.g., children with intracranial masses, prior neurosurgeries, known myelin disorders); (2) no neurological or psychological morbidity on follow-up; and (3) no residual motion artifacts in the subjects’ brain MRI data.

RkJQdWJsaXNoZXIy MTk4NDMw