Hereditary haemorrhagic telangiectasia (HHT) is a dominantly inherited disorder characterized by mucocutaneous telangiectasias and arteriovenous malformations (AVM) in multiple organs, with a frequency of at least 1 in 10 000.1 The most common manifestations are epistaxis, characteristic telangiectasias, and gastrointestinal haemorrhage.2 Neurological features include pial AVM (nidal and micro [nidus <1cm] AVM), high-flow arteriovenous fistulae (AVF) in the brain and spinal cord, capillary telangiectasias,3 and also cerebral ischaemia or abscess due to paradoxical emboli arising via pulmonary vascular shunts. Cerebrovascular malformations (AVM/AVF) affect up to 14% of the HHT population.2 Recent clinical guidelines recommending neurovascular screening4 means that it is timely for awareness of HHT, and the complexities of clinical and genetic diagnosis, to be raised in the paediatric neuroscience community.

Diagnosis of this disorder remains primarily clinical (Table I); individual features carry differential weight in terms of predictive diagnostic power. Three genes have been identified: endoglin (HHT1; OMIM 187300), ACVRLK1 (HHT2; OMIM 600376; which together account for most genetically characterized cases), and SMAD4 (juvenile polyposis/hereditary haemorrhagic telangiectasia syndrome [JPHT] OMIM 175050), with many mutations described. These genes are components of the transforming growth factor beta and bone morphogenic protein signaling pathways.5,6 Two further loci have been assigned HHT3 (OMIM 601101) and HHT4 (OMIM 610655).

Table I. Curaçao criteria4
  1. ‘Definite’ HHT=3 or more criteria; ‘possible’ HHT=2 criteria; ‘unlikely’ HHT if 0 or 1 criteria. HHT, hereditary haemorrhagic telangiectasia.

TelangiectasesMultiple; lips/nose/oral cavity/fingers
Visceral lesionsGastrointestinal telangiectasia, pulmonary, hepatic, cerebral or spinal arteriovenous malformation
Family historyAffected 1st degree relative

Where the diagnosis is beyond reasonable clinical doubt, mutations are detected in 87 to 93% of affected individuals. However, this leaves a significant proportion with a ‘definite’ clinical diagnosis, and a larger number with a ‘possible’ diagnosis (Table I legend), without an identified mutation. A combination of epistaxis and family history alone is not a reliable indicator of diagnosis and characteristic skin lesions are rarely present in childhood. Penetrance is age dependent6,7 and children are therefore unlikely to show sufficient features for a clinical diagnosis. This means that, unless genetic testing is informative, all children of a parent with HHT are considered to have possible HHT, even without any phenotypic manifestations.

Some of the excess mortality in HHT, especially in young people, has been attributed to haemorrhage from brain AVMs (BAVMs)2 but data is scant, with a likely ascertainment bias.3,7 BAVMs are more commonly associated with mutations in endoglin than ACVRLK1, with frequencies of 8 to 14% in the former compared with 2 to 3%.8,9 The suggestion that BAVMs in HHT may carry a lower haemorrhage risk than sporadic BAVMs remains contentious.10 Although BAVMs are usually considered congenital,3 as vascular lesions in HHT are thought to represent an aberrant response to angiogenic stimuli2 and spontaneously regressing AVMs have been documented,11 it seems likely that there is also a dynamic component, further supported by frequent presentations during pregnancy.10 It has been suggested that early mortality associated with high-flow AVF explains the discrepancy in age of presentation between these (infancy/early childhood3) and nidal/microAVM (adolescents/young adults); however, overall, high-flow AVFs remain a rare feature of HHT.

Arguments for and against neurovascular screening for children with HHT

  1. Top of page
  2. Arguments for and against neurovascular screening for children with HHT
  3. The role of genetic testing
  4. To screen or not to screen
  5. References

Recent international clinical guidelines recommend screening children with definite and possible HHT for cerebrovascular malformations,4 using brain magnetic resonance imaging (MRI) with contrast from 6 months of age or when reviewed. This extends to all children of a parent with HHT, half of whom will be unaffected. The potential benefits and hazards of screening for an asymptomatic disease are complex, but in principle, screening is justified if there is effective presymptomatic treatment. In this context, treatment may be any combination of surgery, endovascular embolization, or irradiation, all of which carry significant morbidity and mortality. For example, Krings et al.12 achieved endovascular obliteration in 38.7% of children with HHT and high-flow pial AVFs, with 6.5% mortality and a similar rate of new neurological deficits. Whilst some BAVMs can be cured, at least 10% may only be amenable to partial treatment and this may encourage evolution to more unfavourable morphologies.13 Once a lesion is detected, this will, even if asymptomatic, cause considerable anxiety, especially if treatment options are limited.

Treatment of asymptomatic sporadic BAVMs in adults remains highly contentious and is the subject of an ongoing randomized trial.14 Most paediatric BAVMs present symptomatically,15 so the risk of haemorrhage in asymptomatic paediatric BAVMs is unknown, although this depends partially on lesion characteristics.16 However, treatment approaches to asymptomatic BAVMs in children have generally been more aggressive based on longer life expectancy relative to adults. If BAVMs in HHT really carry a lower haemorrhage risk, the role of screening is even more complex.

There was only around 65% agreement on the screening recommendation in the guidelines4 and there remain geographical discrepancies in practice, with screening being common in North America but less consistent elsewhere. Some have advocated screening higher-risk patients, for example families with a history of AVM/AVF associated with HHT,2 but phenotypic heterogeneity within families argues against this. Current concepts of the genesis of cerebrovascular malformations suggesting that lesion morphology and extent is determined by the timing and site of a ‘second hit’ in a genetically predisposed individual17 make familial heritability of BAVM less likely.

Whilst brain MRI and MR angiography will identify larger malformations, smaller lesions can only be reliably excluded on catheter angiography. The clinical relevance of micro AVMs in HHT is unclear but smaller sporadic BAVMs carry higher haemorrhage risk in both adults18–20 and children.16 catheter angiography has an appreciable risk of vascular (5%) and neurological morbidity (0.1%) as well as exposure to ionizing radiation,21 making it unattractive as a screening tool. In a young child, MRI often requires a general anaesthesia. Any identified lesions will require further characterization by catheter angiography. In addition to concerns about side effects of investigation and treatment, single time point screening may be insufficient but issues of when and how often to re-image are unresolved. Spinal AVFs are not discussed in the recommendations but similar considerations apply.

The role of genetic testing

  1. Top of page
  2. Arguments for and against neurovascular screening for children with HHT
  3. The role of genetic testing
  4. To screen or not to screen
  5. References

The guidelines suggest pursuing a molecular diagnosis in clinically definite and possible HHT – this would usually be in the parent and should be undertaken via a clinical genetics service, after appropriate clinical assessment and counseling. Currently, molecular diagnosis is achieved in around 70% of such individuals, and in these families asymptomatic family members can have mutation analysis to establish whether they are affected if they so wish. Molecular diagnosis has been shown to be economically beneficial compared with indiscriminate clinical screening of all potentially affected individuals for cerebral and pulmonary arteriovenous shunts.22,23

The place of genetic testing in children is less certain. For children to have presymptomatic testing for a genetic condition, it is accepted that there should be a clinical benefit to this testing for the child24 ( Avoidance of an invasive screening test would be an appropriate indication. However, in childhood, screening for the major treatable complication, pulmonary arteriovenous malformations (PAVMs) should usually be performed after puberty, as lesions tend to evolve under hormonal influence. Screening of asymptomatic children before puberty is often restricted to pulse oximetry to exclude detectable hypoxia. PAVM screening, does not, therefore provide a good reason for childhood testing. A molecular diagnosis in childhood might also have major financial implications for the individual when they reach adulthood, for example with access to life and health insurance. Unless, therefore, there is a reasonable case for benefit to the child, molecular testing should be deferred until they can make the decision about testing for themselves.

As the likelihood of BAVM is higher in those with endoglin compared with ACVRLK1 mutations, formal risk- and cost-benefit analyses would favour screening patients with an endoglin mutation (where approximately 10 scans will be performed for every lesion identified) than those with an ACVRLK1 mutation (where approximately 50 scans will be performed per lesion identified). These relative frequencies may also influence a family’s choice about whether or not to opt for neurovascular screening. In a family without a known mutation, it is important to be certain about the clinical diagnosis, as there are a number of other hereditary vascular dysplasias that can be confused with HHT, including capillary malformation arteriovenous malformation syndrome (CM-AVM, OMIM 608354), familial cerebral cavernous malformations (FCCM, OMIM 116860), and venous malformations, multiple cutaneous and mucosal (VMCM, OMIM 600195). Review by an experienced clinician and gene testing can be very effective in establishing the correct diagnosis.

To screen or not to screen

  1. Top of page
  2. Arguments for and against neurovascular screening for children with HHT
  3. The role of genetic testing
  4. To screen or not to screen
  5. References

It is unfortunate that a recommendation in favour of neurovascular screening has been made despite the lack of supportive evidence of efficacy; the level of agreement within the expert group suggests that there is near equipoise as to whether or not this is a valid strategy. Nonetheless, now the recommendation has been made, clinicians will need to deal with expectations from both patients and professional colleagues in practical terms.

In our view, where the issue of neurovascular screening for definite or possible HHT arises, the child and family (after a thorough genetic characterization and prior to any neuroimaging) should be referred to an experienced paediatric neurovascular team conversant with the above issues for counseling. Potential contributors to such a team might be paediatric neurologists, neurosurgeons, and interventional neuroradiologists. The issues that should be covered include: likely frequency of BAVM, lack of evidence favouring treatment of asymptomatic lesions, treatment related morbidity, and the possibility of identifying small/untreatable malformations or other incidental intracranial findings (e.g. arachnoid cysts). It should be made clear that small lesions may not be detected and that the identification of a vascular malformation may have significant wider impact, for example for future insurability.

Should the family wish to proceed, MRI of the brain and spine could be performed. Gadolinium enhancement will increase conspicuity of small vascular lesions, adding weight to a positive diagnosis, although these lesions are unlikely to be suitable for treatment. Given potential dynamic factors in development of cerebrovascular malformations in HHT, it seems reasonable (until more definitive data is available) to discuss the option of re-imaging in adult life if the index scan was during childhood. This may be especially important in females as symptomatic haemorrhages appear to cluster around child-bearing years.10

Given the current dearth of evidence, it is essential that any systematic approach to screening is evaluated in a prospective manner; this is likely to require multi-center and probably international collaboration. Such a strategy has potential to give rise to a variety of difficult clinical scenarios and the involvement of an experienced specialist team from the outset will facilitate onward management.

Perhaps most importantly, adults with HHT should be educated about the potential presentation of a cerebrovascular lesion in their children. This may include cardiac failure in newborns (tachypnea, sweating, failure to thrive), hydrocephalus in infants (enlarging head size, sunsetting), or acute focal neurological deficits from cerebral haemorrhage or venous ischaemia at any age (seizures/hemiparesis). Empowering parents with this information will mean that such presentations are recognized and acted on swiftly, both by parents and healthcare professionals.

Children presenting de novo with neurovascular malformations (and their families) should be evaluated in detail for clinical features of HHT. This includes taking an extended pedigree and enquiring about specific non-neurological symptoms, such as epistaxis or gastrointestinal haemorrhage, as well as detailed clinical (including cutaneous) evaluation. Radiological flags for potential HHT include multiple cerebral AVM25 and high-flow cerebral or spinal AVF, especially in infants.3 In this context, genetic testing may frequently be applicable and useful in management for the wider family.


  1. Top of page
  2. Arguments for and against neurovascular screening for children with HHT
  3. The role of genetic testing
  4. To screen or not to screen
  5. References
  • 1
    Westermann CJ, Rosina AF, De Vries V, de Coteau PA. The prevalence and manifestations of hereditary hemorrhagic telangiectasia in the Afro-Caribbean population of the Netherlands Antilles: a family screening. Am J Med Genet A2003; 116A: 3248.
  • 2
    Shovlin CL. Hereditary haemorrhagic telangiectasia: pathophysiology, diagnosis and treatment. Blood Rev2010; 24: 20319.
  • 3
    Krings T, Ozanne A, Chng SM, Alvarez H, Rodesch G, Lasjaunias PL. Neurovascular phenotypes in hereditary haemorrhagic telangiectasia patients according to age. Review of 50 consecutive patients aged 1 day–60 years. Neuroradiology2005; 47: 71120.
  • 4
    Faughnan ME, Palda VA, Garcia-Tsao G, et al.International guidelines for the diagnosis and management of hereditary haemorrhagic telangiectasia. J Med Genet2011; 48: 7387.
  • 5
    Ricard N, Bidart M, Mallet C, et al.Functional analysis of the BMP9 response of ALK1 mutants from HHT2 patients: a diagnostic tool for novel ACVRL1 mutations. Blood2010; 116: 160412.
  • 6
    ten Dijke P, Goumans MJ, Pardali E. Endoglin in angiogenesis and vascular diseases. Angiogenesis2008; 11: 7989.
  • 7
    Morgan T, McDonald J, Anderson C, et al.Intracranial hemorrhage in infants and children with hereditary hemorrhagic telangiectasia (Osler-Weber-Rendu syndrome). Pediatrics2002; 109: E12.
  • 8
    Berg J, Porteous M, Reinhardt D, et al.Hereditary haemorrhagic telangiectasia: a questionnaire based study to delineate the different phenotypes caused by endoglin and ALK1 mutations. J Med Genet2003; 40: 58590.
  • 9
    Letteboer TG, Mager JJ, Snijder RJ, et al.Genotype-phenotype relationship in hereditary haemorrhagic telangiectasia. J Med Genet2006; 43: 3717.
  • 10
    Easey AJ, Wallace GM, Hughes JM, Jackson JE, Taylor WJ, Shovlin CL. Should asymptomatic patients with hereditary haemorrhagic telangiectasia (HHT) be screened for cerebral vascular malformations? Data from 22,061 years of HHT patient life. J Neurol Neurosurg Psychiatry2003; 74: 7438.
  • 11
    Leung KM, Agid R, terBrugge K. Spontaneous regression of a cerebral arteriovenous malformation in a child with hereditary hemorrhagic telangiectasia. Case report. J Neurosurg2006; 105: 42831.
  • 12
    Krings T, Chng SM, Ozanne A, Alvarez H, Rodesch G, Lasjaunias PL. Hereditary haemorrhagic telangiectasia in children. Endovascular treatment of neurovascular malformations. Results in 31 patients. Interv Neuroradiol2005; 11: 1323.
  • 13
    Reitz M, Schmidt NO, Vukovic Z, et al.How to deal with incompletely treated AVMs: experience of 67 cases and review of the literature. Acta Neurochir Suppl2011; 112: 1239.
  • 14
    Mohr JP, Moskowitz AJ, Stapf C, et al.The ARUBA trial: current status, future hopes. Stroke2010; 41: e53740.
  • 15
    Fullerton HJ, Achrol AS, Johnston SC, et al.Long-term hemorrhage risk in children versus adults with brain arteriovenous malformations. Stroke2005; 36: 2099104.
  • 16
    Ellis MJ, Armstrong D, Vachhrajani S, et al.Angioarchitectural features associated with hemorrhagic presentation in pediatric cerebral arteriovenous malformations. J Neurointerv Surg2012. Mar 13 [Epub ahead of print].
  • 17
    Krings T, Geibprasert S, Terbrugge K. Classification and endovascular management of pediatric cerebral vascular malformations. Neurosurg Clin N Am2010; 21: 46382.
  • 18
    Crawford PM, West CR, Chadwick DW, Shaw MD. Arteriovenous malformations of the brain: natural history in unoperated patients. J Neurol Neurosurg Psychiatry1986; 49: 110.
  • 19
    Graf CJ, Perret GE, Torner JC. Bleeding from cerebral arteriovenous malformations as part of their natural history. J Neurosurg1983; 58: 3317.
  • 20
    Guidetti B, Delitala A. Intracranial arteriovenous malformations. Conservative and surgical treatment. J Neurosurg1980; 53: 14952.
  • 21
    Fung E, Ganesan V, Cox TS, Chong WK, Saunders DE. Complication rates of diagnostic cerebral arteriography in children. Pediatr Radiol2005; 35: 11747.
  • 22
    Bernhardt BA, Zayac C, Trerotola SO, Asch DA, Pyeritz RE. Cost savings through molecular diagnosis for hereditary hemorrhagic telangiectasia. Genet Med2012; 14: 60410.
  • 23
    Cohen JH, Faughnan ME, Letarte M, Vandezande K, Kennedy SJ, Krahn MD. Cost comparison of genetic and clinical screening in families with hereditary hemorrhagic telangiectasia. Am J Med Genet A2005; 137: 15360.
  • 24
    Borry P, Stultiens L, Nys H, Cassiman JJ, Dierickx K. Presymptomatic and predictive genetic testing in minors: a systematic review of guidelines and position papers. Clin Genet2006; 70: 37481.
  • 25
    Bharatha A, Faughnan ME, Kim H, et al.Brain arteriovenous malformation multiplicity predicts the diagnosis of hereditary hemorrhagic telangiectasia: quantitative assessment. Stroke2012; 43: 728.