CASE REPORT J Neurosurg Pediatr 20:164–169, 2017

Cerebral hemorrhage in monozygotic twins with hereditary hemorrhagic telangiectasia: case report and hemorrhagic risk evaluation

Abbas Rattani, MBe,1 Michael C. Dewan, MD,2 Vickie Hannig, MS, CGC,3 Robert P. Naftel, MD,2 John C. Wellons III, MD, MSPH,2 and Lori C. Jordan, MD, PhD4

1School of Medicine, Meharry Medical College, Nashville; 2Division of Pediatric Neurosurgery, Department of Neurological Surgery, and Divisions of 3Medical Genetics and Genomic Medicine and 4Pediatric Neurology, Department of Pediatrics, Vanderbilt University Medical Center, Monroe Carell Jr. Children’s Hospital at Vanderbilt, Nashville, Tennessee

The authors present a case of monozygotic twins with hereditary hemorrhagic telangiectasia (HHT) who experienced cerebral arteriovenous malformation (AVM) hemorrhage at a very young age. The clinical variables influencing HHT-re- lated AVM rupture are discussed, and questions surrounding the timing of screening and intervention are explored. This is only the second known case of monozygotic HHT twins published in the medical literature, and the youngest pair of first-degree relatives to experience AVM-related cerebral hemorrhage. Evidence guiding the screening and management of familial HHT is lacking, and cases such as this underscore the need for objective and validated protocols. https://thejns.org/doi/abs/10.3171/2017.3.PEDS16587 KEY WORDS cerebral arteriovenous malformation; hereditary hemorrhagic telangiectasia; monozygotic twins; neurosurgery; Osler-Weber-Rendu; rupture risk; vascular disorders

ereditary hemorrhagic telangiectasia (HHT), al- cerebral hemorrhage carries serious consequences.4,11,14,24,​ ​ ternatively referred to as Osler-Weber-Rendu syn- 30,32 Woodall et al. reported an increased incidence of cere- drome, is a multisystem blood vessel disorder with bral vascular anomalies among HHT patients; they noted anH autosomal-dominant inheritance pattern. By 40 years AVMs in 7.7% of patients, developmental venous anoma- of age, more than 90% of patients develop at least 1 sign lies in 4.3%, and cerebral in 2.4%.33 The annual of HHT.24 Prevalence rates between 1:5,000 and 1:100,000 risk of hemorrhage from cerebral vascular malformations have been reported, with higher rates in particular popu- is debated, but it has been estimated to be between 0.4% lations (e.g., Japanese, Afro-Caribbean).5,8,15,17,31 Due to its and 3.0%.14,16,24,32,34 highly variable expressivity and penetrance, prevalence Herein, we present a case of monozygotic twins with rates may be underreported.29 Characteristic findings of HHT who experienced cerebral AVM hemorrhage at a very HHT for clinical diagnosis are defined by the Curaçao young age. The clinical variables influencing HHT-related criteria: 1) diagnosis of HHT in a first-degree relative; 2) AVM rupture are discussed, and questions surrounding recurrent and spontaneous epistaxis (the most common timing of screening and intervention are explored. symptom); 3) mucocutaneous telangiectasia; and 4) vis- ceral arteriovenous malformations (AVMs).25,28 Patients Case Reports meeting 3 or more of these criteria are considered definite for HHT.6 Twin A Vascular lesions may be present in various vascular A 3-month-old boy presented to our center after an epi- beds, including pulmonary (45%), hepatic (52%), and ce- sode of seizures. His parents reported a 2-month history rebral (10%–12.8%) involvement.7,12 Cerebral AVMs are of right gaze deviation and a bulging fontanel. The patient often multiple and clinically silent, but the potential for was also noted to have right head deviation and left arm

ABBREVIATIONS AICA = anterior inferior cerebellar ; AVF = arteriovenous fistula; AVM = arteriovenous malformation; CVM = capillary vascular malformation; HHT = hereditary hemorrhagic telangiectasia; PICA = posterior inferior cerebellar artery. SUBMITTED October 17, 2016. ACCEPTED March 21, 2017. INCLUDE WHEN CITING Published online May 19, 2017; DOI: 10.3171/2017.3.PEDS16587.

164 J Neurosurg Pediatr Volume 20 • August 2017 ©AANS, 2017

Unauthenticated | Downloaded 09/26/21 04:39 PM UTC Rupture risk in monozygotic HHT twins weakness, which was initially attributed to torticollis. His head circumference was above the 90th percentile. Non- contrast CT scanning demonstrated a large right hemi- spheric hyperdensity with significant midline shift. MRI and CT studies revealed a large AVM arising from the middle cerebral artery (MCA) circulation. There was also an associated superficial 2.7-cm that appeared to be eroding through the dura mater and the inner surface of the parietal calvaria. The patient underwent a right-sided craniotomy for resection of the AVM (Fig. 1) and aneurysm. While the superficial aneurysm was encountered first, its dural and bony adherence and complex feeder anatomy precluded di- rect clip isolation. Instead, the proximal arterial supply was identified after careful decompression of the cystic AVM complex. Aneurysm clips were placed on 2 large anoma- lous arising from the MCA, which resulted in im- mediate deflation of the pathologic vessels. A large clot was then evacuated, providing circumferential visualization of the interface between normal brain and the residual AVM. After selective cauterization of the remaining small arte- rial feeders and suction removal of abnormal tissue, bipolar cautery was used to ligate and excise the single large drain- FIG. 1. Twin A. Images obtained in a monozygotic twin brother. A: Axial ing that was then sent for pathologic examination. The noncontrast CT scan demonstrating a right hemispheric hypodense wound was irrigated and closed, and the patient was sent space-occupying mass surrounding a well-circumscribed, mixed hy- to the ICU for recovery. The estimated blood loss was 100 perdense focus with scattered calcifications. Gross effacement of the ml. Pathologic specimens revealed fibrovascular-walled ipsilateral lateral ventricle is observed along with midline shift. B: Axial solid-cystic tissue with disorganized thick- and thin-walled postcontrast T1-weighted MR image demonstrating a hyperintense vessels and foci of calcification and hemosiderin. Cathe- saccular lesion extruding beyond the margins of the surrounding hypoin- ter angiography performed on the 12th postoperative day tensity and encroaching on the dura and parietal calvaria. C: Coronal T2-weighted MR image demonstrating multilobular hypointensities of the demonstrated no evidence of residual aneurysm, and the vascular lesion and the surrounding cystic structure occupying the ma- patient was discharged with improving left hemiparesis. jority of the right cerebral hemisphere. D: Postcontrast T1-weighted MR Two-month follow-up MRI demonstrated AVM oblit- image demonstrating complete resection of the vascular abnormality eration, with resolution of the cystic space and normaliza- with radiographic normalization of adjacent anatomical structures. tion of the ventricular architecture (Fig. 1). At the 6-month follow-up, the patient was noted to have made a complete recovery with no demonstrable neurological deficits. the left AICA-PICA. The patient was asymptomatic and therefore was discharged the next day; elective cerebral Twin B angiography for AVM characterization was scheduled for Following the events described above, a hereditary the following week. On the evening prior to the elective vascular phenomenon was considered. A family history angiography session, the patient presented to the emergen- of epistaxis and facial telangiectasias rendered the clini- cy department with vomiting and lethargy, and a CT scan cal diagnosis of HHT, which was confirmed when ge- without contrast revealed a posterior fossa hemorrhage netic workup identified the endoglin ENG( ) mutation in and obstructive hydrocephalus (Fig. 3C). Twin A and presumed in Twin B, the mother, maternal Following emergency placement of an external ventric- half-brother, and maternal grandmother, given symptom ular drain, a suboccipital craniectomy and C-1 laminec- constellation (Fig. 2). Given the monozygotic relationship tomy were performed. The posterior fossa contents were between twins, Twin B was presumed to harbor the mu- under significant pressure, and immediately on opening tation, and therefore underwent comprehensive screening the dura, a large blood clot spontaneously extruded from for associated pathology at 5 months of age. Indeed, MRI the cortical surface. With gentle suction and irrigation, the of his brain revealed a posterior fossa vascular malfor- remaining blood products were evacuated. On explora- mation (Fig. 3A and B), and the patient was referred to tion, the medial and lateral arterial feeders were identified the emergency department. MR angiography sequences and ligated with aneurysm clips. The large dilated vein revealed a left cerebellar vascular malformation, which draining directly into the tentorium was then ligated, and was thought to most likely represent an arteriovenous fis- the entire lesion was removed. After inspecting the entire tula (AVF) supplied by a prominent left anterior inferior hemorrhage cavity and ensuring , the dura and cerebellar artery (AICA) and posterior inferior cerebel- incision were closed, and the patient was taken to the ICU. lar artery (PICA), and drained by multiple cortical An estimated blood loss of 150 ml was noted. Patholog- emptying into an enlarged left transverse sinus. The le- ic examination demonstrated thick-walled vessels, some sion also contained what appeared to be a large bi-lobed with marked luminal hemorrhagic distention consistent feeding aneurysm arising from the distal branch vessels of with an AVM.

J Neurosurg Pediatr Volume 20 • August 2017 165

Unauthenticated | Downloaded 09/26/21 04:39 PM UTC A. Rattani et al.

FIG. 2. Pedigree. The proband (Twin B) is indicated by the arrow. The twin brother (Twin A) and half-sister were the only ones to undergo DNA testing. The ENG mutation was presumed in the proband, mother, maternal half-brother, and maternal grandmother given the severity of symptomology. All have had screening MRI of the brain with and without contrast and other screening tests per the international HHT guidelines. The squares represent males and the circles, females. The diagonal lines indicate that the individual is deceased. *Confirmed cerebral AVM; ENG† mutation.

Despite prolonged external ventricular drain weaning and ICU course, the patient was eventually discharged home without neurological deficits or the need for perma- nent CSF diversion. Like his brother, Twin B has made a full neurological recovery, remains neurologically intact, and has regained lost milestones. Discussion To our knowledge, this is only the second known case of monozygotic HHT twins published in the medical lit- erature,22 and the youngest pair of first-degree relatives to experience AVM-related cerebral hemorrhage. Evidence guiding the screening and management of familial HHT is lacking, and cases such as this underscore the need for objective and validated protocols. Through multiinstitu- tional collaboration and data aggregation, the HHT con- sortium represents the best effort yet to better quantify the rates and risk factors for HHT-related cerebral AVM hem- orrhage.2 These risk factors are explored below and related to the present cases to offer a unique window into the rela- tionship between genetic makeup and clinical manifesta- tions in HHT.

Risk Factors for HHT-Related Cerebral Hemorrhage Patient Age Infants and children are at risk for sudden, serious hem- FIG. 3. Twin B. Images obtained in the 5-month-old proband. A and orrhage if a cerebral AVM is present; therefore, if there is B: Postcontrast axial (A) and sagittal (B) T1-weighted MR images dem- a family history of HHT in a first-degree relative, evalu- onstrating a bi-lobed vascular lesion in the left cerebellar hemisphere. ation for HHT in childhood has been suggested to limit A prominent and tortuous AICA-PICA complex is seen along with an 26 engorged ipsilateral transverse sinus. C: Noncontrast axial CT scan potential catastrophic sequelae. Moreover, in a study of demonstrating an acute posterior fossa hemorrhage causing complete 50 individuals with HHT, 75 vascular lesions were noted effacement of the fourth ventricle and ventriculomegaly. D: Postoper- in the CNS, 90% of which were AVFs that were diag- ative postcontrast axial T1-weighted MR image demonstrating hemor- nosed by the age of 6 years, with 53% being infratento- rhage evacuation and complete resection of the vascular lesion.

166 J Neurosurg Pediatr Volume 20 • August 2017

Unauthenticated | Downloaded 09/26/21 04:39 PM UTC Rupture risk in monozygotic HHT twins rial.19 AVFs, generally regarded as high-risk AVM lesions, agnosis was younger than the parent, suggesting possible were noted primarily in children younger than 5 years of anticipation of a cerebral AVM.30 Such a pattern is sup- age; adolescents mainly presented with small AVMs, with ported for HHT—though hardly confirmed—by our case, micro-AVMs in younger adolescents.11,19 wherein presentation occurred at a much earlier age after skipping a generation. In a review of 9 patients, Morgan Genetics et al. reported a higher and notably severe AVM rupture A correlation between genotypic and phenotypic mani- risk among children and infants with a family history of festations is currently being explored for HHT; the evo- HHT.26 It is clear that further research is needed to clarify lution of vascular phenotype varies across patients with the relationship between first-degree relatives and AVM identical mutations. HHT pathophysiology is associated rupture risk, especially among the HHT population. with mutations in the transforming growth factor–b sig- naling cascade, and most cases of HHT are due to a lack Vascular Phenotype of sufficientENG or activin receptor–like kinase (ALK)–1 Rupture risk of cerebral AVMs increases based on le- protein function, both necessary for vascular integrity and sion type, with AVFs posing the most serious and dev- vasculogenesis.6,10,13,25 Mutations tend to be found across astating consequences. Multiple cerebral AVMs have a all coding regions; no common mutation targets have been 10%–20% incidence rate of rupture in HHT cases.6,11,16,19 associated with either ENG or ALK-1.25 Interestingly, Willemse et al. noted that while HHT pa- The ENG gene, located on the long arm of chromo- tients have a higher risk of harboring a cerebral AVM, some 9 at position 34.11 (9q34.11), encodes for a protein especially if a history of pulmonary AVM is noted, the that exists as covalently linked homodimer composed bleeding risk of the cerebral AVM in HHT may be less of inter- and intrachain bonds; thus, ENG mutations are than that in patients with sporadic AVMs.32 linked to structural instability and protein function loss.1 Cerebral AVMs can be categorized into 3 types: non- Abdalla and Letarte noted that 80% of ENG gene muta- shunting CVMs with characteristic enhancing vessel tions identified in HHT1 cases led to “stop codons and conglomeration feeding into arteries or veins, shunting truncated polypeptides,”1 further concluding that ENG “nidus-type” AVMs, and single-hole pial (brain) AVFs.18 mutations result in decreased levels of functional endog- HHT-related CVMs are usually smaller than 1 cm, su- lin on endothelial cells and peripheral blood monocytes, perficial, supratentorial lesions with a suspected low risk thus leading to the clinical manifestations of HHT1.1 The of rupture, multiplicity, and no dilated arterial feeder or lack of vascular integrity leads to a wide variation in an- draining vein; these constitute 61% of all vascular mal- gioarchitecture, thus giving rise to phenotypically differ- formations.18 Nidus-type AVMs occur with a frequency ent AVMs, most commonly AVFs, shunting “nidus-type” of 43% among aneurysms with characteristic features of AVMs, and nonshunting capillary vascular malformations early venous drainage, compact/discrete nidus, and flow- (CVMs).6 voids, occurring in settings of prior hemorrhage; these A genotype-phenotype relationship has been noted be- AVMs have a low-to-moderate risk of rupture.18 AV Fs o c- tween certain genetic mutations and resulting phenotype. cur with a frequency of 12% and are primarily found in ENG mutations are diagnostic for HHT1, with pulmonary children. Multiple vascular lesions are observed in 36%– and cerebral AVMs (some have reported a variable associ- 44% of patients.18,19 Features of AVFs include an enlarged ation of cerebral AVMs to HHT121) being more common in arterial feeder, early venous drainage, a large venous this population, while ALK-1 mutations are diagnostic for pouch, venous congestion/reflux, associated aneurysms, HHT2, with hepatic AVMs being more common.23 Oth- and a relatively higher rupture risk.18 ers have noted that while diagnostic mutations have been Both of our twins had experienced cerebral AVM rup- described for particular HHT genotypes, no definitive re- ture before angiography was performed; thus, lationship can be concluded between genotype and pheno- thorough angioarchitecture evaluation and subtype clas- typic manifestation of HHT.3,19 Krings et al. purported that sification were not performed. Nonetheless, CT angiogra- the phenotypic differences in terms of blood vessel capac- phy findings were most consistent with AVF (Figs. 1 and ity and strength observed among HHT patients were age 2), and therefore carried a higher rupture risk. Summar- dependent.19 Hypoxia, presumably over time, has been con- ily, based purely on radiographic features of the posterior sidered as a potential environmental candidate for trigger- fossa lesion, Twin B harbored a higher-risk lesion, and he ing vascular remodeling in HHT AVM development.19,27 was being evaluated for possible early (prophylactic) sur- gical intervention. Family History There seems to be an increased risk for cerebral AVMs Other Factors in individuals with HHT who have a first-degree rela- Prior history of hemorrhage has been reported as a tive with a cerebral AVM, although the risk of rupture risk factor in HHT patients harboring a cerebral AVM.14,16 in HHT patients with first-degree relatives is unclear.24 Gross and Du noted prior hemorrhage risk with a hazard In a systematic review of familial occurrence of sporad- ratio of 3.2 (95% CI 2.1–4.3).14 Kim et al. reported that ic cerebral AVMs unrelated to HHT, van Beijnum et al. HHT patients with cerebral AVMs presenting with rupture noted that 79% of familial relationships of patients with have higher rerupture rates (10.1%, 95% CI 3.3%–31.2%) cerebral AVMs were first-degree relatives.30 Additionally, than patients with unruptured cerebral AVMs at presenta- they reported that in families with at least 2 successive tion (0.4%, 95% CI 0.1%–1.7%); rerupture rates were com- generations of cerebral AVMs, the child at the time of di- parable to patients with sporadic cerebral AVMs.16 Other

J Neurosurg Pediatr Volume 20 • August 2017 167

Unauthenticated | Downloaded 09/26/21 04:39 PM UTC A. Rattani et al. major high-risk features for cerebral AVM rupture include tasia: current views on genetics and mechanisms of disease. associated aneurysms (HR 1.8, 95% CI 1.6–2.0), deep J Med Genet 43:97–110, 2006 AVM location (HR 2.4, 95% CI 1.4–3.4), and exclusively 2. Akers AL, Ball KL, Clancy M, Comi AM, Faughnan ME, 14 Gopal-Srivastava R, et al: Brain vascular malformation con- deep venous drainage (HR 2.4, 95% CI 1.1–3.8). sortium: overview, progress and future directions. J Rare Disord 1:5, 2013 Timing of Screening and/or Intervention 3. Begbie ME, Wallace GM, Shovlin CL: Hereditary haemor- Per HHT international guidelines, for children with rhagic telangiectasia (Osler-Weber-Rendu syndrome): a view possible or definite HHT, screening for cerebral vascular from the 21st century. Postgrad Med J 79:18–24, 2003 malformations with unenhanced MRI is recommended in 4. Bharatha A, Faughnan ME, Kim H, Pourmohamad T, Krings T, Bayrak-Toydemir P, et al: Brain arteriovenous malfor- the first 6 months of life or at the time of clinical diagno- 9 mation multiplicity predicts the diagnosis of hereditary sis. The timing of treatment remains an ongoing debate, as hemorrhagic telangiectasia: quantitative assessment. rupture rates are of relevant concern in terms of pursuing 43:72–78, 2012 interventions sooner than later. Brinjikji et al. noted that 5. Bideau A, Plauchu H, Brunet G, Robert J: Epidemiological despite the international guidelines, there is less consen- investigation of Rendu-Osler disease in France: its geograph- sus about screening children with HHT for cerebral AVMs ical distribution and prevalence. Popul 44:3–22, 1989 (64%) compared with the adult population (77%).6 More- 6. Brinjikji W, Iyer VN, Sorenson T, Lanzino G: Cerebrovascu- over, they argued that treatment for AVMs should be eval- lar manifestations of hereditary hemorrhagic telangiectasia. Stroke 46:3329–3337, 2015 uated on a case-by-case basis with evaluation of AVM an- 6,9 7. Brinjikji W, Iyer VN, Yamaki V, Lanzino G, Cloft HJ, gioarchitecture, progression, and related comorbidities. Thielen KR, et al: Neurovascular manifestations of heredi- In this case, the patient’s twin history of early rupture tary hemorrhagic telangiectasia: a consecutive series of combined with the high-risk fistulous appearance of the 376 patients during 15 years. AJNR Am J Neuroradiol lesion may indicate a greater risk of rupture and therefore 37:1479–1486, 2016 the need for early intervention. If early CNS hemorrhagic 8. Dakeishi M, Shioya T, Wada Y, Shindo T, Otaka K, Manabe sequelae of HHT are identified in a first-degree relative, M, et al: Genetic epidemiology of hereditary hemorrhagic we advocate for early, prophylactic surgery. A devastat- telangiectasia in a local community in the northern part of Japan. Hum Mutat 19:140–148, 2002 ing outcome was narrowly avoided in this case, but could 9. Faughnan ME, Palda VA, Garcia-Tsao G, Geisthoff UW, have been prevented altogether if swift action were taken McDonald J, Proctor DD, et al: International guidelines for as soon as the posterior fossa pathology was identified in the diagnosis and management of hereditary haemorrhagic Twin B. In such cases, even vigilant observation may be telangiectasia. J Med Genet 48:73–87, 2011 too conservative a treatment strategy. 10. Fernández-L A, Sanz-Rodriguez F, Blanco FJ, Bernabéu C, On the other hand, one may reasonably argue that the Botella LM: Hereditary hemorrhagic telangiectasia, a vas- body of literature indicates low overall rupture rates. This cular dysplasia affecting the TGF-b signaling pathway. Clin final point is of particular importance as one considers the Med Res 4:66–78, 2006 11. Fulbright RK, Chaloupka JC, Putman CM, Sze GK, Merriam risks assumed when pursuing a prophylactic craniotomy MM, Lee GK, et al: MR of hereditary hemorrhagic telangi- for AVM resection, particularly in a young, developing ectasia: prevalence and spectrum of cerebrovascular malfor- brain. This disparity between innumerable clinical vari- mations. AJNR Am J Neuroradiol 19:477–484, 1998 ables and the limited clinical evidence underscore the im- 12. Giordano P, Lenato GM, Suppressa P, Lastella P, Dicuonzo F, portance of evaluating patients and family members with Chiumarulo L, et al: Hereditary hemorrhagic telangiectasia: HHT on a case-by-case basis until better evidence exists arteriovenous malformations in children. J Pediatr 163:179– to support a definitive screening and treatment algorithm. 186, 186.e1–186.e3, 2013 As the HHT guidelines suggest, when a cerebral AVM is 13. Govani FS, Giess A, Mollet IG, Begbie ME, Jones MD, Game L, et al: Directional next-generation RNA sequencing found, individuals should be referred to a center with neu- and examination of premature termination codon mutations rovascular expertise for further evaluation and manage- in endoglin/hereditary haemorrhagic telangiectasia. Mol ment; specific pediatric neurovascular expertise is advis- Syndromol 4:184–196, 2013 able in young children.20 14. Gross BA, Du R: Natural history of cerebral arteriovenous In sum, we advocate for the following: 1) comprehen- malformations: a meta-analysis. J Neurosurg 118:437–443, sive genetic workup in patients and family members with 2013 suspected disease; 2) early imaging; 3) early and collabor- 15. Guttmacher AE, Marchuk DA, White RI Jr: Hereditary hem- ative consultation with both neurology and neurosurgery orrhagic telangiectasia. N Engl J Med 333:918–924, 1995 16. Kim H, Nelson J, Krings T, terBrugge KG, McCulloch CE, teams if neuroimaging demonstrates a malformation; and Lawton MT, et al: Hemorrhage rates from brain arteriove- 4) early neurosurgical intervention if there is evidence of nous malformation in patients with hereditary hemorrhagic cerebral hemorrhagic complications in first-degree rela- telangiectasia. Stroke 46:1362–1364, 2015 tives. 17. Kjeldsen AD, Vase P, Green A: Hereditary haemorrhagic telangiectasia: a population-based study of prevalence and mortality in Danish patients. J Intern Med 245:31–39, 1999 Acknowledgments 18. Krings T, Kim H, Power S, Nelson J, Faughnan ME, Young We acknowledge the Vanderbilt Medical Scholars Program for WL, et al: Neurovascular manifestations in hereditary hem- providing Abbas Rattani with support on this project. orrhagic telangiectasia: imaging features and genotype-phe- notype correlations. AJNR Am J Neuroradiol 36:863–870, 2015 References 19. Krings T, Ozanne A, Chng SM, Alvarez H, Rodesch G, Las- 1. Abdalla SA, Letarte M: Hereditary haemorrhagic telangiec- jaunias PL: Neurovascular phenotypes in hereditary haemor-

168 J Neurosurg Pediatr Volume 20 • August 2017

Unauthenticated | Downloaded 09/26/21 04:39 PM UTC Rupture risk in monozygotic HHT twins

rhagic telangiectasia patients according to age. Review of 50 30. van Beijnum J, van der Worp HB, Schippers HM, van Nieu- consecutive patients aged 1 day–60 years. Neuroradiology wenhuizen O, Kappelle LJ, Rinkel GJ, et al: Familial occur- 47:711–720, 2005 rence of brain arteriovenous malformations: a systematic 20. Ladner TR, Mahdi J, Attia A, Froehler MT, Le TM, Lorinc review. J Neurol Neurosurg Psychiatry 78:1213–1217, 2007 AN, et al: A multispecialty pediatric neurovascular confer- 31. Westermann CJ, Rosina AF, De Vries V, de Coteau PA: The ence: a model for interdisciplinary management of complex prevalence and manifestations of hereditary hemorrhagic disease. Pediatr Neurol 52:165–173, 2015 telangiectasia in the Afro-Caribbean population of the Neth- 21. Lesca G, Olivieri C, Burnichon N, Pagella F, Carette MF, erlands Antilles: a family screening. Am J Med Genet A Gilbert-Dussardier B, et al: Genotype-phenotype correla- 116A:324–328, 2003 tions in hereditary hemorrhagic telangiectasia: data from the 32. Willemse RB, Mager JJ, Westermann CJ, Overtoom TT, French-Italian HHT network. Genet Med 9:14–22, 2007 Mauser H, Wolbers JG: Bleeding risk of cerebrovascular 22. Lesser BA, Wendt D, Miks VM, Norum RA: Identical twins malformations in hereditary hemorrhagic telangiectasia. J with hereditary hemorrhagic telangiectasia concordant for Neurosurg 92:779–784, 2000 cerebrovascular arteriovenous malformations. Am J Med 33. Woodall MN, McGettigan M, Figueroa R, Gossage JR, Al- 81:931–934, 1986 leyne CH Jr: Cerebral vascular malformations in hereditary 23. Letteboer TG, Mager JJ, Snijder RJ, Koeleman BP, Lindhout hemorrhagic telangiectasia. J Neurosurg 120:87–92, 2014 D, Ploos van Amstel JK, et al: Genotype-phenotype relation- 34. Yen CP, Sheehan JP, Schwyzer L, Schlesinger D: Hemor- ship in hereditary haemorrhagic telangiectasia. J Med Genet rhage risk of cerebral arteriovenous malformations before 43:371–377, 2006 and during the latency period after Gamma Knife radiosur- 24. Maher CO, Piepgras DG, Brown RD Jr, Friedman JA, Pol- gery. Stroke 42:1691–1696, 2011 lock BE: Cerebrovascular manifestations in 321 cases of hereditary hemorrhagic telangiectasia. Stroke 32:877–882, 2001 25. McDonald J, Wooderchak-Donahue W, VanSant Webb C, Disclosures Whitehead K, Stevenson DA, Bayrak-Toydemir P: Hereditary The authors report no conflict of interest concerning the materi- hemorrhagic telangiectasia: genetics and molecular diagnos- als or methods used in this study or the findings specified in this tics in a new era. Front Genet 6:1, 2015 paper. 26. Morgan T, McDonald J, Anderson C, Ismail M, Miller F, Mao R, et al: Intracranial hemorrhage in infants and children Author Contributions with hereditary hemorrhagic telangiectasia (Osler-Weber- Rendu syndrome). Pediatrics 109:E12, 2002 Conception and design: Dewan, Rattani. Acquisition of data: 27. Sánchez-Elsner T, Botella LM, Velasco B, Langa C, Bern- Dewan, Rattani, Hannig, Jordan. Analysis and interpretation of abéu C: Endoglin expression is regulated by transcriptional data: all authors. Drafting the article: Rattani. Critically revising cooperation between the hypoxia and transforming growth the article: Dewan, Rattani, Naftel, Wellons, Jordan. Reviewed factor-b pathways. J Biol Chem 277:43799–43808, 2002 submitted version of manuscript: all authors. Approved the 28. Shovlin CL, Guttmacher AE, Buscarini E, Faughnan ME, final version of the manuscript on behalf of all authors: Dewan. Hyland RH, Westermann CJ, et al: Diagnostic criteria for Administrative/technical/material support: Dewan, Hannig, Jor- hereditary hemorrhagic telangiectasia (Rendu-Osler-Weber dan. Study supervision: Dewan. syndrome). Am J Med Genet 91:66–67, 2000 29. Shovlin CL, Jackson JE, Bamford KB, Jenkins IH, Benjamin Correspondence AR, Ramadan H, et al: Primary determinants of ischaemic Michael C. Dewan, Division of Pediatric Neurosurgery, Depart- stroke/brain abscess risks are independent of severity of pul- ment of Neurological Surgery, Vanderbilt University Medical monary arteriovenous malformations in hereditary haemor- Center, T-4224 Medical Center North, Nashville, TN 37232-2380. rhagic telangiectasia. Thorax 63:259–266, 2008 email: [email protected].

J Neurosurg Pediatr Volume 20 • August 2017 169

Unauthenticated | Downloaded 09/26/21 04:39 PM UTC