Infantile Hemangiomas: An Update on Pathogenesis and ... · Twolesscommontypesof“hemangioma...

DOI: 10.1542/peds.2012-1128; originally published online December 24, 2012; 2013;131;99Pediatrics

Tina S. Chen, Lawrence F. Eichenfield and Sheila Fallon FriedlanderInfantile Hemangiomas: An Update on Pathogenesis and Therapy

http://pediatrics.aappublications.org/content/131/1/99.full.html

located on the World Wide Web at: The online version of this article, along with updated information and services, is

of Pediatrics. All rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.Boulevard, Elk Grove Village, Illinois, 60007. Copyright © 2013 by the American Academy published, and trademarked by the American Academy of Pediatrics, 141 Northwest Pointpublication, it has been published continuously since 1948. PEDIATRICS is owned, PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

at UNIV OF CHICAGO on January 30, 2013pediatrics.aappublications.orgDownloaded from


http://pediatrics.aappublications.org/

Infantile Hemangiomas: An Update on Pathogenesisand Therapy

abstractInfantile hemangiomas (IHs) are the most common vascular tumors ofchildhood, affecting ∼5% of all infants. Although most lesions pro-liferate and then involute with minimal consequence, a significantminority can be disfiguring, functionally significant, or, rarely, life-threatening. Recent discoveries concerning hemangioma pathogenesisprovide both an improved understanding and more optimal approachto workup and management. Important detrimental associations canbe seen with IH, such as significant structural anomalies associatedwith segmental IH. Standards of care have dramatically changedevaluation and management of hemangiomas. The goal of timely rec-ognition and therapy is to minimize or eliminate long-term sequelae.New modalities, such as oral propranolol, provide the caregiver withbetter therapeutic options, which can prevent or minimize medicalrisk or scarring, but the side effect profile and risk-benefit ratio ofsuch interventions must always be evaluated before instituting therapy.Pediatrics 2013;131:99–108

AUTHORS: Tina S. Chen, MD, Lawrence F. Eichenfield, MD,and Sheila Fallon Friedlander, MD

Pediatric Dermatology, Rady Children’s Hospital San Diego, SanDiego, California

KEY WORDhemangioma

ABBREVIATIONSEPC—endothelial progenitor cellIH—infantile hemangiomaPDL—pulsed dye laserPHACE—Posterior fossa brain abnormalities, Hemangiomas, Ar-terial malformations, Coarctation of the aorta and other cardiacdefects, Eye abnormalitiesVEGF—vascular endothelial growth factor

www.pediatrics.org/cgi/doi/10.1542/peds.2012-1128

doi:10.1542/peds.2012-1128

Accepted for publication Aug 29, 2012

Address correspondence to Tina S. Chen, MD, 8010 Frost St, Suite602, San Diego, CA 92122. E-mail: [email protected]

PEDIATRICS (ISSN Numbers: Print, 0031-4005; Online, 1098-4275).

Copyright © 2013 by the American Academy of Pediatrics

FINANCIAL DISCLOSURE: Drs Friedlander and Eichenfield haveserved as investigators for Pierre-Fabre, without compensation;and Dr Chen has indicated she has no financial relationshipsrelevant to this article to disclose.

FUNDING: No external funding.

PEDIATRICS Volume 131, Number 1, January 2013 99

STATE-OF-THE-ART REVIEW ARTICLE


mailto:[email protected]


Infantile hemangiomas (IHs) are themost common pediatric vasculartumors, affecting ∼5% of all infantsborn in the United States.1 A recentprospective Australian study of new-borns noted an incidence of 2.6% by 6weeks of age,2 and an American studyof similar design found that 4.5% ofinfants developed IH, all of which werepresent by 3 months of age.3 IHs arebenign tumors that are usually notpresent at birth but instead are notedwithin the first few weeks of life. Pre-cursor lesions are common but oftensubtle; findings may include telangiec-tasias, pallor, a bruiselike appearance,and, rarely, ulceration. IHs typically havean initial proliferative phase, with rapidgrowth of the tumor in the first severalmonths of life. This is followed by aninvolution stage, with slow, spontane-ous resolution spanning years. Afterinvolution of the vascular component,a residual fibrofatty mass often per-sists.

Although many of these lesions resolvespontaneously without concern, a sig-nificant proportion lead to function-threatening and cosmetically disfiguringconsequences. For functionally signi-ficant or potentially deforming lesions,timely intervention is important tominimize the possibility of a poor out-come and permanent scarring. Manyimportant and management-alteringdiscoveries have occurred regardingIH in the past decade. The followinghighlights the most important of thesefindings.

PATHOPHYSIOLOGY

IHs are vascular tumors that involve theproliferation of benign endotheliallikecells that possess histochemicalmarkers (GLUT-1, LewisYantigen, FcyRII,and merosin); these markers are alsopresentonplacental bloodvessels.4 Theimmunohistochemical profile differ-entiates IH from other vascular birth-marks or tumors.

The pathophysiology associated withthe unique natural history of theselesions, with initial rapid proliferationfollowed by gradual involution and re-gression, has not been completelyelucidated. One etiologic hypothesisspeculates that cells are “embolized”from the placenta.5 Another suggeststhat IHs result from somatic mutationsin a gene mediating endothelial cellproliferation.6 Recent data suggest anendothelial progenitor cell as thesource of origin of the tumors.7–10 Ithas been speculated that hypoxia, ei-ther systemically (eg, due to placentalinsufficiency) or in a specific “niche”area of poorly perfused tissue5,11

stimulates endothelial progenitor cellsto proliferate inappropriately. The fol-lowing summarizes evidence for thesevarious theories (see Table 1).

The placental theory is attractive be-cause it would explain the programmedlife cycle of IH. Subsequent to North’sdiscoveries regarding the histochemi-cal similarities of IH and placenta,4

Barnes et al noted that placenta and IHhave high levels of genetic similaritywhen compared with other vasculartumors and normal structures.12 Waneret al noted that IH tend to develop alongembryonic fusion lines of the facialplacodes.13 Piecing these 2 seeminglydisparate facts together, Mihm et alsuggested that IH might represent “be-nign metastases” originating from theplacenta or other cells that proliferatein areas of low oxygen tension, suchas the “end artery, vascular dead end”sites occurring in embryonic fusionplanes.5 Pittman et al were unable todetect the presence of maternal-fetalchimerism in IH tissue, but this doesnot rule out the possibility of theplacental origin of IH tissue becausethe placenta is predominantly fetalin origin.14

It has also been hypothesized that im-mature endothelial cells and pericytes,which coexist in the late stages of fetal

development, perhaps maintain per-sistent proliferative properties fora period of time postnatally, leading tothe development of IH.15 However, Boyeet al demonstrated the clonality of IHcells, making it less likely that a dispa-rate group of cells serve as the sourceof this tumor.16

Hypoxia has been proposed as a drivingfactor for the pathogenesis of vascularproliferation in general. IH proliferationmay be a homeostatic attempt to nor-malize hypoxic tissue. Epidemiologic

TABLE 1 Pathogenesis of IH: Hypotheses andSupporting Data

Placental embolizationHistochemicalGLUT1+, LeY+, FcyRII +, Merosin+4

Genetic analysesTranscriptome similarities between

placental and IH tissue12

Life historyRapid proliferation, followed by stabilization

an involutionMetastatic niche theoryIH cluster at embryonic fusion placode sites13

IH precursor cell comes from the placenta asa “benign metastasis”5

Somatic mutation or hyperreactivity of anendothelial-type cellIncidence of IH generally sporadicClonality of IH cells16

Mutation of VEGR2 receptors in IH tissue19

Endothelial progenitor cells (EPC) play anetiologic roleCirculating endothelial progenitor cells

increased in IH infants7

EPCs present in proliferative but not involutingphase of IH growth8

Human IH EPCs injected into immunodeficientmice recapitulate IH life cycle and expressGLUT110

IH growth is mediated by angiogenic peptidesUpregulated VEGF2 signaling in IH19

Insulin-like growth factor (angiogenic)upregulated in IH21

Hypoxia stimulates the release of EPCs; EPCs thenhone to hypoxic sitesEpidemiology: hypoxia-associated factors (low

birth weight, advanced maternal age)overrepresented in IH population23

IH occur in “end artery, vascular “dead-end”embryonic fusion plane sites5,13

IH associated with retinopathy ofprematurity17

GLUT1 is a sensor for hypoxia and present onIH cells18

Hypoxia induced mediators of stem-celltrafficking increased in children with IH9

100 CHEN et al at UNIV OF CHICAGO on January 30, 2013pediatrics.aappublications.orgDownloaded from


findings support this hypothesis, giventhat factors thatare thought tobe linkedto hypoxia, such as low birth weight andadvanced maternal age, are over-represented in IHpopulations.11 Anothersupportive finding is the association ofIH with retinopathy of prematurity, acondition known to be linked to ische-mia.17 GLUT-1, present on IH tissue, isa facilitative glucose transporter that isan important sensor for hypoxia.18

The growth of IH likely involves angio-genic peptides, such as vascular en-dothelial growth factor (VEGF) andbasic fibroblast growth factor, whichinduce proliferation of blood vessels.Receptors for these growth factorsare also crucial in endothelial cellregulation, and a misbalance of VEGF-receptor-1 expression with conse-quent hyperactivity of VEGF-receptor-2function has been noted in IH tissue.19

The suppressive effect of glucocorti-coidsmay bemediated through VEGF-A.20

Additionally, insulinlike growth factor-2,which stimulates angiogenesis, is up-regulated in proliferating but not in-voluting IH.21

Endothelial progenitor cells (EPCs) arevascular stem cells with the capacity tocontribute to postnatal vascular de-velopment. There is now compellingevidence that these EPCs play an etio-logic role in the development of IH. Asubset of progenitor cells isolated fromIH tissues, which possess the surfacemarkers CD34+ CD133+, are of partic-ular interest. These EPCs have beenshown to differentiate into endothelialcells in vitro22 and are increased 15-fold in IH compared with controls.7

Cultured EPCs from patients with IHstain positively for known hemangiomamarkers GLUT1, CD32, and merosin.Several mediators of EPC traffickingand vasculogenesis, such as VEGF-Aand hypoxia inducible factor-1 alpha(a transcription factor that regulatesthe formation of new blood vessels byEPCs), were found to be elevated in

blood and IH specimens taken fromchildren with proliferating IH.9

A major breakthrough occurred whenKhan et al were able to successfullyinject CD133+ EPCs from human hem-angioma tissue into immunodeficientmice. Thesemice then developed GLUT1vascular tumors, which recapitulatedthe development of human IH, providinginvestigators with the first viable IHanimalmodel.10 These studies highlightthe importance of CD133+ EPCs in thepathophysiology of IH and provide ameans of testing putative therapies inthis animal model.

EPIDEMIOLOGY AND DIAGNOSIS

Traditionally, misuse of the term “hem-angioma” to describe other vascularlesions has impeded the collection ofaccurate demographic data. In par-ticular, misdiagnosis of port winestains, venous and arterial malforma-tions, and vascular tumors such astufted angiomas, affected the accuracyof many studies performed in thepast. In 2007, a large, multicenter,prospective study was conducted bypediatric dermatologists skilled indistinguishing vascular lesions, 1058children with IH were identified. Thetumors were more commonly seen inpatients who were female, white (non-Hispanic), premature, of low birthweight, a product of multiple gestation,or born to mothers with advancedmaternal age.23 Placenta previa andpreeclampsia were also found to bemore common.

Two less common types of “hemangiomamimickers,” congenital hemangiomas,occasionally confuse practitioners butdo not possess the classic attributesof IH. Both noninvoluting congenitalhemangiomas and rapidly involutingcongenital hemangiomas are “fully for-med” at birth or can even be involutingor ulcerating. They may possess telan-giectases and a rim of pallor. In contrastto IH, they lack GLUT1 surface markers,

and thus histochemical evaluation isoften useful in distinguishing betweenthese vascular lesions.24 Rapidly invo-luting congenital hemangiomas lesionsrapidly involute, often within the firstyear, whereas noninvoluting congenitalhemangiomas persist for a prolongedperiod.

Hemangiomas are now classified into 3primary subtypes: segmental, focal(see Fig 1), and indeterminate. Waneret al found that focal hemangiomaswere 3 times more common than dif-fuse or segmental hemangiomas onthe face.13 The segmental subtype isassociated with a higher risk of com-plications, functional compromise, de-formity, and ulceration, as well asa greater need for therapy.25 Other IHthat are regarded as being higher riskand may have a greater need fortherapy are outlined in Table 2.

One of the most important complica-tions associated with segmental IH isPHACE syndrome. A novel neurocuta-neous syndrome associated with facialhemangiomas described by Friedenet al in 1996, the acronym PHACE refersto posterior fossa brain abnormalities,hemangiomas, arterial malformations,coarctation of the aorta and othercardiac defects, as well as eye abnor-malities.26 Although less well known,this syndrome may be more commonthat Sturge-Weber syndrome (facialport-wine stain with associated glau-coma and neurologic anomalies). Ina multicenter study of children withlarge facial IH (see Fig 2), 31% met diag-nostic criteria for PHACE syndrome.27,28

FIGURE 1Focal IH.


PEDIATRICS Volume 131, Number 1, January 2013 101 at UNIV OF CHICAGO on January 30, 2013pediatrics.aappublications.orgDownloaded from


Criteria for the diagnosis of this dis-order were delineated by using a stan-dard consensus method with review ofpublished data and a multidisciplinaryteam of experts.29,30 Current criteriaare listed in Table 2, as are those forpossible PHACE (see Table 3). The syn-drome has a 9:1 female-to-male pre-dominance.27 PHACES syndrome alsoincludes sternal malformation andsupraumbilical raphe.31

Ninety-eight percent of patients withPHACEsyndromehavea largesegmentalIHpresenton the faceorhead.Rarely, thedisordercanoccur inassociationwith IHlocated elsewhere or even in the ab-sence of IH; in such circumstances,consensus criteria would term this“probable PHACE syndrome.”28,31,32,33

Brain and cerebral vascular anomaliesare the most common extracutaneousfeatures of PHACE syndrome; neuro-

logicandcognitive impairmentsremainthe greatest source of morbidity inthese patients.34 Central nervous sys-tem arterial anomalies are the mostcommon vascular abnormalities seenin PHACE. Cerebrovascular accidents inIH patients with PHACE syndrome havebeen reported, with no previous spe-cific anomaly; these patients tend tostabilize neurologically over time.35 Inaddition, structural abnormalities in-volving the posterior fossa and cere-bellum have also been associated withPHACE syndrome.

A number of cardiac abnormalities canoccur in this syndrome, the mostcommon of which is coarctation of theaorta. The overall incidence of eye ab-normalities is small and includesmicrophthalmia and optic nerve hypo-plasia.29,36,37

Facial hemangiomas that are$5 cm indiameter should prompt evaluation forPHACE syndrome including MRI andmagnetic resonance arteriogram ofthe brain, cardiovascular imaging, andan ophthalmologic examination. Thevessels of the upper chest and neckshould also be evaluated.

Large, segmental IH in the anogenitalregion also carry a risk for associatedunderlying anomalies (see Fig 3). Sev-eral pneumonics have been coined(PELVIS, LUMBAR, SACRAL) to emphasizemajor features, which include a lum-bosacral or perineal IH in associationwith spinal cord, anogenital, and renalanomalies.38–40

Infants with multiple classic focal IHmay have extracutaneous involvement

(see Fig 4). Although significant mor-tality was historically attributed tomultifocal hemangiomas with organinvolvement, termed “diffuse neonatalhemangiomatosis,” it is now appreci-ated that multifocal hemangiomas of-ten do not involve extracutaneous sitesand that medical consequences arevariable when they do.41 The commonlyinvolved extracutaneous site is theliver, and hepatic ultrasound should beperformed in patients with $5 cuta-neous IH.42 Consumptive hypothyroid-ism has been associated with IH andother large vascular lesions of the liver.This is a result of excessive iodothyr-onine deiodinase expression, and suchpatients require monitoring and ag-gressive thyroid hormone supplemen-tation.43

MANAGEMENT

Although most IHs proliferate and in-volute without functional impairment,a significant minority requires someform of intervention.44 It is important toconsider the psychological as well asmedical impact of IH, particularly whenlocated on the face. Many central faciallesions leave residual scars or struc-tural deformities, which may have life-long effects. In the past, treatmentoptions for IH were limited and theirpotential side effects considerable. Al-though most IHs do not pose significantrisks, and careful observation is still theappropriate management option formany lesions, the introduction of rela-tively safer topical and systemic agentsnow allows earlier and easier inter-vention in appropriate cases (Table 4).However, a Cochrane analysis of inter-ventions for IH noted that a lack of well-designed clinical trials and the absenceof US Food and Drug Administration–approved medications for IH limits theability to clearly identify the single besttreatment option.25

Until recently, intralesionalandsystemiccorticosteroids were the mainstay of

TABLE 2 High-Risk IH Lesions

Location Type Growth Phase

Periorificiala (eyes, nose, mouth) Segmentalb Maximal proliferationphase (usually 3–6 mo)

Central Facial MultipleLumbosacralb Rapidly proliferatingb

Genitalb,c

a Function-threatening.b High risk for ulceration.c Risk of associated structural anomalies.

FIGURE 2Large, segmental IH.



therapy for problematic IH. Intralesionalinjection can be used for localizedlesions.Prednisone,administeredorallyat doses of 1 to 3 mg/kg per day, is aneffective therapy for the majority ofpatients.45 In a quantitative systematicliterature review, Bennett et al foundthat systemic corticosteroids had a84% response rate with 36% rebound ininfants with problematic IH.46 However,corticosteroid therapy has significantside effects, including increased riskfor systemic infection, hypertension,increased appetite, stomach irritation,growth suppression, and cardiomyop-athy. In addition, some IHs are resistantto corticosteroid therapy. On theother hand, corticosteroids remainuseful in certain situations, particu-larly in those who cannot tolerateother therapeutic options.

Other systemic therapeuticoptionshaveincluded interferon and vincristine.47

Adverse effects have limited the utilityof both of these drugs. In particular,reports of serious side effects with in-terferon, including blood abnormalitiesand spastic diplegia in up to 20% ofpatients are cause for concern.48 Simi-larly, vincristine, a vinca alkaloid used

widely in cancer chemotherapy, is effi-cacious for life-threatening IH but alsohas limited use due to the strong vesi-cant qualities of the drug, with need forcentral line access for chronic admin-istration as well as potential peripheralmixed sensory-motor neurotoxicity.49 Ofnote, some of the data for interferonand vincristine may be complicated bythe misdiagnosis of Kaposiform hem-angioendotheliomas or other vasculartumors as IH.

In recent years, propranolol therapyhasbecome increasinglymoreuseful inthe management of IHs that requireintervention. Leaute-Labreze et al firstfortuitously discovered the efficacy ofb-blockers for the treatment of IH in2008.50 An infant developed cardiomy-opathy after systemic corticosteroidtherapy for a large facial IH. Whenpropranolol therapy was initiated forthis complication, a remarkable flat-tening and fading of the IH occurred.The investigators subsequently docu-mented the drug’s efficacy as a first-line, as well as second-line (after steroiduse), therapy. All 11 patients werenoted to have softening and change incolor from red to purple of their IHswithin 24 hours of starting propranolol.Subsequently, .170 reports and stud-ies have substantiated the usefulnessof this drug.51–53 A prospective studycomparing propranol to placebo docu-mented drug efficacy.54 In a recentmulticenter retrospective chart review,propranolol therapy was found to bemore clinically efficacious and cost-effective than oral corticosteroidsin the treatment of IH.55 In addition,propranolol therapy led to fewer sur-gical interventions and had fewer sideeffects compared with oral cortico-steroids. Most recently, a response rateof 98% was noted in a systemic reviewof 41 studies of.1200 children treatedwith propranolol. The mean dose was2.1 mg/kg per day, and mean treatmentduration was 6.4 months. Serious side

TABLE 3 Anomalies Reported in PHACE Syndrome

Category Abnormality

Structural brain Posterior fossa: Dandy-Walker complex, cerebellar hypoplasia/atrophy,subependymal and arachnoid cysts

Hypoplasia or agenesis of cerebrum, corpus callosum, septum pellucidum,vermis

Polymicrogyria: microcephaly, heterotopia, absent pituitary or partiallyempty sella turcica

Cerebral vascular Dysplasia of the large cerebral arteriesAbsence or moderate to severe hypoplasia of the large cerebral arteriesAberrant origin or course of the large cerebral arteriesSaccular aneurysmsPersistent embryonic arteries (predominantly trigeminal)Pial enhancementCerebral sinus malformationsSinus pericraniiDural arteriovenous malformations/pial malformationsIntracranial hemangiomaArterial stenosis or occlusion with or without Moyamoya collateralsAbsent foramen lacerumAcute arterial stroke

Reprinted with permission from Metry DW, Heyer G, Hess C, et al. Consensus statement on diagnostic criteria for PHACEsyndrome. Pediatrics. 2009;124(5):1447-1456.

FIGURE 3Segmental IH in the sacral region.

FIGURE 4Multifocal infantile hemangioma.




effects were rare, occurring in,1% ofpatients.56 Unfortunately, direct com-paritive studies of these 2 agents havenot been performed.

Propranolol is a systemic nonselectiveb-blocker that has been used for dec-ades in pediatric patients with cardio-vascular disease with standard dosingof 0.5 to 4mg/kg per day. Currently,many uncertainties exist regarding theappropriate and optimal use in chil-dren with IH, including optimal dosing,frequency of dosing, duration of ther-apy, age of therapy initiation, and timingand method of tapering to minimize thechance of rebound. An ongoing in-

ternational multicenter study may shedsignificant light on these issues.57

Theories regarding propranolol’smechanism of action include an ini-

tial capillary vasoconstrictive effect,

suppression/blockade of growth fac-

tors with induction of apoptosis of

endothelial cells,51,58 and blockade of

GLUT1 receptors.59 CD34+ endothelialprogenitor cells in IH express factors in-fluencing the renin-angiotensin system.Because the renin-angiotensin systemcan stimulate angiogenesis, Itinteanget al suggested that propranol’s in-hibitory effect on the renin-angiotensin

system might account for propranolol-induced involution.60

Although systemic propranolol isclearly efficacious, rare side effects,a few of which may be life-threatening,are cause for concern. These includesymptomatic hypoglycemia, hypoten-sion, bronchial hyperreactivity, seizure,restless sleep, constipation, and coldextremities.59 Hypoglycemia has beenthe most commonly noted serious ef-fect. Careful monitoring, particularlyof medically fragile infants (eg small,premature, failure to thrive) is re-quired. Those patients with PHACE syn-drome should be evaluated for cerebralvascular anomalies because poor ce-rebral perfusion secondary to abnor-mal vessels can place them at higherrisk for hypotensive stroke. Epinephrineinjections for emergency treatment ofallergic reactions may be ineffective inchildren on chronic b-blocker therapy.Therefore, the risk-benefit ratiomust beconsidered when deciding to use pro-pranolol in children with a history ofanaphylactic allergic reactions thatmight require epinephrine therapy.Collaborative care is often optimal forappropriate propranolol treatment ofIH, including a pediatrician, dermatolo-gist, ophthalmologist, cardiologist, andother specialists, if needed.

Topical agents are appropriate therapyin some situations (eg, small, thinlesions), and their use is sometimesdriven by parental anxieties or the de-sire for active therapy. The risks of ad-verseeffects are lesswith topical ratherthan systemic agents, but there is lim-ited data on efficacy. Topical clobetasolhas been used with some success,particularly in small periorbital lesions.However, concerns regarding atrophy,glaucoma, and cataracts exist with thisdrug. Investigators have also consid-ered the utility of topical, rather thansystemic, b-blockers. Timolol maleateis a nonselective topical b-blocker thathas been approved for the treatment of

TABLE 4 Management Options for IH

Specific Approach Dosage Safety Concerns

Activenonintervention

Adjust scheduled visits based on rapidityof growth and family concern

Monitor for function- or life-threateningcomplications during proliferativephase

SystemicPropranolol 1–3 mg/kg/day after mealsa Hypotension, hypoglycemia, bronchial

hyperreactivity, seizure, restlesssleep, constipation, cold extremities

Corticosteroids 1–3 mg/kg/day High blood pressure, increasedappetite, stomach irritation,cardiomyopathy, growthsuppression, increased risk ofsystemic infection, aseptic necrosisof bones

Vincristineb Recommended dosages vary Strong vesicant; irritates vesselsMixed sensory-motor neurotoxicityLoss of deep tendon reflexesConstipation

Interferonb Recommended dosages vary Flulike reactions, transammonitis,neutropenia, skin necrosis; spasticdiplegia (up to 20%)

TopicalTimolol .5% gel 1 drop twice daily Theoretical risks similar to those for

propranolol but not yet welldocumented in patients treatedwith timolol for IH

Laser Standard: pulsed dyelaser 585–595 nm

May lead to ulceration in some patients,particularly in lip area

Use every 2–4 weeks as needed ND:Yag, alexandrite, and CO2 lasers—higher risk of scarring

Particularly useful for ulcerating lesions,residual erythema, telangiectasia

Refractory lesions: ND:Yag oralexandrite laser

Residual scar: FractionatedCO2 laser

Topicalcorticosteroids

Applied twice daily Cutaneous atrophy and telangiectasia

N/A, not applicable; ND:Yag, neodymium-doped yttrium aluminum garnet.a Dosing guidelines have not been fully elucidated.b Dosing recommendations vary; consult those with experience in administering these agents.



ocular glaucoma and hypertension inchildren and infants by the Food andDrug Administration.61 The ophthalmo-logic literature suggests that the sideeffects of timolol might be similar tothose found in oral propranolol; how-ever, in studies of ophthalmic prepara-tions of timolol used topically for thetreatment of IH, adverse effects repor-ted to date are limited to a single epi-sode of severe sleep disturbance.Nonetheless, experts urge caution withthe drug and recommend using nomore than 1 drop twice a day to af-fected lesions.62

Recent studies showing efficacy of thistopical agent for IH are promising, butmost published works consist of indi-vidual cases or small pilot studies.63–67

The largest study to date found that thegreatest indicators of optimal thera-peutic response were increased con-centration and duration of drugtreatment, as well as superficial natureof the lesion.67

Other therapeutic options have beenconsidered, but risk-benefit issueshavethus far limitedextensive investigationsinto these therapies. Angiogenesisinhibitors are theoretically a naturalchoice for the treatment of hemangio-mas. Sirolimus (also known as rapa-mycin), an inhibitor of mTOR, negativelyaffects cell proliferation and metabo-lismaswell as angiogenesis. In vivo andin vitro studies have demonstratedsuppression of IH growth in a mousemodel. The drug appears to limit stemcell replicative capabilities and acts in amechanism distinct from that of corti-costeroids.68 Although this is a poten-tially attractive therapeutic option,possible risks related to inhibition ofangiogenesis, particularly in a growingorganism raise concerns. At this time,it is most appropriate to restrict use toclinical trials until better safety dataare available.

Procedural and multimodal therapieshave been used to treat IH, but con-

troversy exists regarding their appro-priate role, and few evidence-based,controlled studies are available thatevaluate pediatric laser therapy. Battaet al published a controlled trial inwhich the author noted no significantdifference in outcome when pulsed dyelaser (PDL; 585–595 nm) therapy wasused in the treatment of IH; however,the suboptimal fluences used, lack ofappropriate cooling during treatment,and positive evidence that laser-treated lesions healed faster weak-ened the validity of their assessment.69

Most experts believe that PDL therapycan diminish pain and hasten healingin ulcerating lesions, particularly inthose located in the perineal area thathave not responded to topical or sys-temic therapeutic measures. Treat-ment can also decrease redness andresidual telangiectasias.

The ideal form of laser therapy is ina state of evolution, but PDL is the mostcommonly used modality. This therapycan safely diminish and sometimeseliminate superficial lesions with min-imal requirements for anesthesia andonly rare scarring. It penetrates toa depth of ∼1 mm and is thereforemost useful for superficial lesions orulcers. Longer wavelength alexandriteor neodymium-doped yttrium alumi-num garnet laser therapy is some-times used for recalcitrant lesions butcarries a higher risk for scarring.Fractionated CO2 laser holds promisefor diminishing textural changes andscars that can develop in affectedchildren. Multimodal therapy, usingsystemic agents in conjunction with ju-dicious use of laser or other proceduraltherapy, can optimize therapy for se-lected patients (see Figs 5, 6, and 7).Some experts use PDL as adjunctive or“mop-up” therapy to treat residual tel-angiectasia or erythema, whereasothers believe in early utilization whenlesions are flatter and more amenableto therapy.70

Occasionally, surgical excision is theoptimal therapeutic intervention. Ap-propriate lesions are often large pe-dunculated IH located in a site whereasurgical scarwill be lessnoticeable. Insome instances, surgical intervention isinevitable because permanent, baggyresidual scars develop. In such in-stances, health care providers may optto remove the lesion early in its lifecycle, given that a surgical interventionand scar are inevitable.

FIGURE 5Segmental IH before propranolol and PDL ther-apy.

FIGURE 6Segmental IH during propranolol and PDL ther-apy.

FIGURE 7Segmental IH after propranolol and PDL therapy.




SUMMARY

Recent discoveries have led to an im-proved understanding of the patho-genesis and clinical behavior of IH. Thebest scientific evidence to date sup-ports the hypothesis that IHs originatefrom a subset of endothelial progenitorcells (CD133+) that are stimulatedand proliferate under hypoxic con-ditions. These cells theoretically “hone”to areas of relative hypoxia, such asembryonic fusion planes. Perturbationof angiogenic factors may also playa role in the inappropriate prolif-eration of these cells. The clinicianmust be aware of potential anomaliesthat can occur in association withlarge segmental IH. Facial lesions raise

concern for PHACES, whereas lumbo-sacral and perineal lesions shouldraise suspicion for associated spinalcord, renal, and genital anomalies.When diagnosis is uncertain and inpatients with high-risk lesions (Table2), referral to a multidisciplinary vas-cular anomaly center with experiencedsubspecialists is optimal for patientcare.

Newer treatment options for IH maywell pose less risk for the patient,allowing the practitioner to intervene ina relatively safe, and more timelymanner. Propranolol is now first-linetherapy for many practitioners, and itis hoped that future studieswill confirmits efficacy and safety. Timolol, a topical

b-blocker, may have a particular role inthe treatment of superficial lesions.Pulsed dye and other laser modalitiesmay be useful as adjunctive or “mop-up” therapy. Other antiangiogenicagents may prove to be more effectivein the future. However, it is importantto proceed cautiously when imple-menting new therapies and evaluateappropriately for both short and long-term safety issues.71 The risk-benefitratio of any therapy must be scruti-nized, keeping in mind that “watchfulwaiting” may often be appropriate, buttimely intervention is sometimes cru-cial in minimizing long-term sequelaesuch as functional deformity or per-manent scars.

REFERENCES

1. Kilcline C, Frieden IJ. Infantile hemangio-mas: how common are they? A systematicreview of the medical literature. PediatrDermatol. 2008;25(2):168–173

2. Dickison P, Christou E, Wargon O. A pro-spective study of infantile hemangiomaswith a focus on incidence and risk factors.Pediatr Dermatol. 2011;28(6):663–669

3. Kanada KN, Merin MR, Munden A, FriedlanderSF. A prospective study of cutaneous findingsin newborns in the United States: correlationwith race, ethnicity, and gestational statususing updated classification and nomencla-ture. J Pediatr. 2012;161(2):240–245

4. North PE, Waner M, Mizeracki A, Mihm MCJr. GLUT1: a newly discovered immunohis-tochemical marker for juvenile hemangio-mas. Hum Pathol. 2000;31(1):11–22

5. Mihm MC Jr, Nelson JS. Hypothesis: themetastatic niche theory can elucidate in-fantile hemangioma development. J CutanPathol. 2010;37(suppl 1):83–87

6. Walter JW, North PE, Waner M, et al. So-matic mutation of vascular endothelialgrowth factor receptors in juvenile hem-angioma. Genes Chromosomes Cancer.2002;33(3):295–303

7. Kleinman ME, Tepper OM, Capla JM, et al.Increased circulating AC133+ CD34+ endo-thelial progenitor cells in children withhemangioma. Lymphat Res Biol. 2003;1(4):301–307

8. Yu Y, Flint AF, Mulliken JB, Wu JK, Bischoff J.Endothelial progenitor cells in infantile

hemangioma. Blood. 2004;103(4):1373–1375

9. Kleinman ME, Greives MR, Churgin SS, et al.Hypoxia-induced mediators of stem/progenitorcell trafficking are increased in childrenwith hemangioma. Arterioscler Thromb VascBiol. 2007;27(12):2664–2670

10. Khan ZA, Boscolo E, Picard A, et al. Multi-potential stem cells recapitulate humaninfantile hemangioma in immunodeficientmice. J Clin Invest. 2008;118(7):2592–2599

11. Drolet BA, Frieden IJ. Characteristics of in-fantile hemangiomas as clues to patho-genesis: does hypoxia connect the dots?Arch Dermatol. 2010;146(11):1295–1299

12. Barnés CM, Huang S, Kaipainen A, et al.Evidence by molecular profiling for a pla-cental origin of infantile hemangioma. ProcNatl Acad Sci U S A. 2005;102(52):19097–19102

13. Waner M, North PE, Scherer KA, Frieden IJ,Waner A, Mihm MC Jr. The nonrandomdistribution of facial hemangiomas. ArchDermatol. 2003;139(7):869–875

14. Pittman KM, Losken HW, Kleinman ME, et al.No evidence for maternal-fetal micro-chimerism in infantile hemangioma: a molec-ular genetic investigation. J Invest Dermatol.2006;126(11):2533–2538

15. Takahashi K, Mulliken JB, Kozakewich HP,Rogers RA, Folkman J, Ezekowitz RA. Cellu-lar markers that distinguish the phases ofhemangioma during infancy and childhood.J Clin Invest. 1994;93(6):2357–2364

16. Boye E, Yu Y, Paranya G, Mulliken JB, OlsenBR, Bischoff J. Clonality and altered be-havior of endothelial cells from hemangio-mas. J Clin Invest. 2001;107(6):745–752

17. Praveen V, Vidavalur R, Rosenkrantz TS,Hussain N. Infantile hemangiomas and reti-nopathy of prematurity: possible association.Pediatrics. 2009;123(3). Available at: www.pediatrics.org/cgi/content/full/123/3/e484

18. Mobasheri A, Richardson S, Mobasheri R,Shakibaei M, Hoyland JA. Hypoxia induciblefactor-1 and facilitative glucose trans-porters GLUT1 and GLUT3: putative molec-ular components of the oxygen and glucosesensing apparatus in articular chondrocytes.Histol Histopathol. 2005;20(4):1327–1338

19. Jinnin M, Medici D, Park L, et al. Sup-pressed NFAT-dependent VEGFR1 expres-sion and constitutive VEGFR2 signaling ininfantile hemangioma. Nat Med. 2008;14(11):1236–1246

20. Greenberger S, Boscolo E, Adini I, MullikenJB, Bischoff J. Corticosteroid suppressionof VEGF-A in infantile hemangioma-derivedstem cells. N Engl J Med. 2010;362(11):1005–1013

21. Ritter MR, Dorrell MI, Edmonds J, Fried-lander SF, Friedlander M. Insulin-like growthfactor 2 and potential regulators of hem-angioma growth and involution identified bylarge-scale expression analysis. Proc NatlAcad Sci USA. 2002;99(11):7455–7460

22. Peichev M, Naiyer AJ, Pereira D, et al. Ex-pression of VEGFR-2 and AC133 by



circulating human CD34(+) cells identifiesa population of functional endothelial pre-cursors. Blood. 2000;95(3):952–958

23. Haggstrom AN, Drolet BA, Baselga E, et al;Hemangioma Investigator Group. Prospectivestudy of infantile hemangiomas: demo-graphic, prenatal, and perinatal character-istics. J Pediatr. 2007;150(3):291–294

24. Enjolras O, Mulliken JB, Boon LM, Wassef M,Kozakewich HP, Burrows PE. Noninvolutingcongenital hemangioma: a rare cutaneousvascular anomaly. Plast Reconstr Surg.2001;107(7):1647–1654

25. Haggstrom AN, Drolet BA, Baselga E, et al.Prospective study of infantile hemangio-mas: clinical characteristics predictingcomplications and treatment. Pediatrics.2006;118(3):882–887

26. Frieden IJ, Reese V, Cohen D. PHACE syn-drome. The association of posterior fossabrain malformations, hemangiomas, arte-rial anomalies, coarctation of the aortaand cardiac defects, and eye abnormalities.Arch Dermatol. 1996;132(3):307–311

27. Haggstrom AN, Garzon MC, Baselga E, et al.Risk for PHACE syndrome in infants withlarge facial hemangiomas. Pediatrics. 2010;126(2). Available at: www.pediatrics.org/cgi/content/full/126/2/e418

28. Metry DW, Haggstrom AN, Drolet BA, et al. Aprospective study of PHACE syndrome ininfantile hemangiomas: demographic fea-tures, clinical findings, and complications.Am J Med Genet A. 2006;140(9):975–986

29. Metry DW, Heyer G, Hess C, et al; PHACESyndrome Research Conference. Consen-sus statement on diagnostic criteria forPHACe syndrome. Pediatrics. 2009;124(5):1447–1456

30. Metry DW, Garzon MC, Drolet BA, et al.PHACE syndrome: current knowledge, fu-ture directions. Pediatr Dermatol. 2009;26(4):381–398

31. Chan YC, Eichenfield LF, Malchiodi J, Fried-lander SF. Small facial haemangioma andsupraumbilical raphe—a forme fruste ofPHACES syndrome? Br J Dermatol. 2005;153(5):1053–1057

32. Torer B, Gulcan H, Kilicdag H, Derbent M.PHACES syndrome with small, late-onsethemangiomas. Eur J Pediatr. 2007;166(12):1293–1295

33. Haggstrom AN, Lammer EJ, Schneider RA,Marcucio R, Frieden IJ. Patterns of infantilehemangiomas: new clues to hemangiomapathogenesis and embryonic facial de-velopment. Pediatrics. 2006;117(3):698–703

34. Metry DW, Dowd CF, Barkovich AJ, FriedenIJ. The many faces of PHACE syndrome. JPediatr. 2001;139(1):117–123

35. Drolet BA, Dohil M, Golomb MR, et al. Earlystroke and cerebral vasculopathy in chil-dren with facial hemangiomas and PHACEassociation. Pediatrics. 2006;117(3):959–964

36. Cartwright MS, Hickling WH, Roach ES. Is-chemic stroke in an adolescent with arte-rial tortuosity syndrome. Neurology. 2006;67(2):360–361

37. Layde PM, Dooley K, Erickson JD, EdmondsLD. Is there an epidemic of ventricularseptal defects in the U.S.A.? Lancet. 1980;1(8165):407–408

38. Girard C, Bigorre M, Guillot B, Bessis D.PELVIS syndrome. Arch Dermatol. 2006;142(7):884–888

39. Iacobas I, Burrows PE, Frieden IJ, et al.LUMBAR: association between cutaneousinfantile hemangiomas of the lower bodyand regional congenital anomalies. JPediatr. 2010;157(5):795–801, e1–e7

40. Drolet B, Garzon M. SACRAL syndrome.Dermatology. 2007;215(4):360–, author re-ply 360–361

41. Glick ZR, Frieden IJ, Garzon MC, Mully TW,Drolet BA. Diffuse neonatal hemangioma-tosis: An evidence-based review of casereports in the literature. J Am Acad Der-matol. 2012; [epub ahead of print]

42. Horii KA, Drolet BA, Frieden IJ, et al; Hem-angioma Investigator Group. Prospectivestudy of the frequency of hepatic heman-giomas in infants with multiple cutaneousinfantile hemangiomas. Pediatr Dermatol.2011;28(3):245–253

43. Huang SA, Tu HM, Harney JW, et al. Severehypothyroidism caused by type 3 iodo-thyronine deiodinase in infantile heman-giomas. N Engl J Med. 2000;343(3):185–189

44. Leonardi-Bee J, Batta K, O’Brien C, Bath-Hextall FJ. Interventions for infantile hae-mangiomas (strawberry birthmarks) of theskin. Cochrane Database Syst Rev. 2011;(5):CD006545

45. Bennett ML, Fleischer AB Jr, Chamlin SL,Frieden IJ. Oral corticosteroid use is ef-fective for cutaneous hemangiomas: anevidence-based evaluation. Arch Dermatol.2001;137(9):1208–1213

46. Rössler J, Wehl G, Niemeyer CM. Evaluatingsystemic prednisone therapy for pro-liferating haemangioma in infancy. Eur JPediatr. 2008;167(7):813–815

47. Ezekowitz RA, Mulliken JB, Folkman J. In-terferon alfa-2a therapy for life-threateninghemangiomas of infancy. N Engl J Med.1992;326(22):1456–1463

48. Dubois J, Hershon L, Carmant L, Bélanger S,Leclerc JM, David M. Toxicity profile of in-terferon alfa-2b in children: A prospectiveevaluation. J Pediatr. 1999;135(6):782–785

49. Pérez-Valle S, Peinador M, Herraiz P, SaénzP, Montoliu G, Vento M. Vincristine, an effi-cacious alternative for diffuse neonatalhaemangiomatosis. Acta Paediatr. 2010;99(2):311–315

50. Léauté-Labrèze C, Dumas de la Roque E,Hubiche T, Boralevi F, Thambo JB, Taïeb A.Propranolol for severe hemangiomas ofinfancy. N Engl J Med. 2008;358(24):2649–2651

51. Sans V, de la Roque ED, Berge J, et al.Propranolol for severe infantile hemangio-mas: follow-up report. Pediatrics. 2009;124(3). Available at: www.pediatrics.org/cgi/content/full/124/3/e423

52. Buckmiller L, Dyamenahalli U, Richter GT.Propranolol for airway hemangiomas: casereport of novel treatment. Laryngoscope.2009;119(10):2051–2054

53. Marsciani A, Pericoli R, Alaggio R, BrisigottiM, Vergine G. Massive response of severeinfantile hepatic hemangioma to prop-anolol. Pediatr Blood Cancer. 2010;54(1):176

54. Hogeling M, Adams S, Wargon O. A ran-domized controlled trial of propranolol forinfantile hemangiomas. Pediatrics. 2011;128(2). Available at: www.pediatrics.org/cgi/content/full/128/2/e259

55. Price CJ, Lattouf C, Baum B, et al. Pro-pranolol vs corticosteroids for infantilehemangiomas: a multicenter retrospectiveanalysis. Arch Dermatol. 2011;147(12):1371–1376

56. Marqueling AL, Oza V, Frieden IJ, PuttgenKB. Propranolol for infantile hemangiomas:A meta analysis. Poster at The Society forPediatric Dermatology annual meeting; July2012; Monterey, CA

57. A randomized, controlled, multidose, multi-centre, adaptive phase II/III study in infantswith proliferating infantile hemangiomas re-quiring systemic therapy to compare fourregimens of propranolol (1 or 3 mg/kg/dayfor 3 or 6 months) to placebo (double blind).Ongoing (active) but not recruiting. Clinicaltrials.gov identifier: NCT01056341. Available at:http://clinicaltrials.gov/show/NCT01056341.Accessed November 7, 2012

58. Manunza F, Syed S, Laguda B, et al. Pro-pranolol for complicated infantile hae-mangiomas: a case series of 30 infants. BrJ Dermatol. 2010;162(2):466–468

59. Lawley LP, Siegfried E, Todd JL. Propranololtreatment for hemangioma of infancy: risksand recommendations. Pediatr Dermatol.2009;26(5):610–614

60. Itinteang T, Brasch HD, Tan ST, Day DJ.Expression of components of the renin-angiotensin system in proliferating in-fantile haemangioma may account for the




propranolol-induced accelerated involution.J Plast Reconstr Aesthet Surg. 2011;64(6):759–765

61. Coppens G, Stalmans I, Zeyen T, Casteels I.The safety and efficacy of glaucoma medi-cation in the pediatric population. J PediatrOphthalmol Strabismus. 2009;46(1):12–18

62. McMahon P, Oza V, Frieden IJ. Topical ti-molol for infantile hemangiomas: puttinga note of caution in “cautiously optimistic.”Pediatr Dermatol. 2012;29(1):127–130

63. Ni N, Guo S, Langer P. Current concepts inthe management of periocular infantile(capillary) hemangioma. Curr Opin Oph-thalmol. 2011;22(5):419–425

64. Khunger N, Pahwa M. Dramatic response totopical timolol lotion of a large hemifacial

infantile haemangioma associated withPHACE syndrome. Br J Dermatol. 2011;164(4):886–888

65. Pope E, Chakkittakandiyil A. Topical timololgel for infantile hemangiomas: a pilotstudy. Arch Dermatol. 2010;146(5):564–565

66. Guo S, Ni N. Topical treatment for capillaryhemangioma of the eyelid using beta-blocker solution. Arch Ophthalmol. 2010;128(2):255–258

67. Chakkittakandiyil A, Phillips R, Frieden IJ,et al. Timolol maleate 0.5% or 0.1% gel-forming solution for infantile hemangio-mas: a retrospective, multicenter, cohortstudy. Pediatr Dermatol. 2012;29(1):28–31

68. Greenberger S, Yuan S, Walsh LA, et al.Rapamycin suppresses self-renewal and

vasculogenic potential of stem cells iso-lated from infantile hemangioma. J InvestDermatol. 2011;131(12):2467–2476

69. Batta K, Goodyear HM, Moss C, Williams HC,Hiller L, Waters R. Randomised controlledstudy of early pulsed dye laser treatmentof uncomplicated childhood haemangio-mas: results of a 1-year analysis. Lancet.2002;360(9332):521–527

70. Admani S, Krakowski AC, Nelson JS,Eichenfield LF, Friedlander SF. Beneficialeffects of early pulsed dye laser therapy inpatients with infantile hemangiomas. Der-matol Surg. 2012;38(10):1739–1740

71. Frieden IJ. Infantile hemangioma research:looking backward and forward. J InvestDermatol. 2011;131(12):2345–2348

HEAD BANGING: A few weeks ago, I saw two girls collide while trying to ‘head’a ball during a high school soccer game. One fell to the ground holding her handto her face. I immediately thought that she had fractured her nose. However, thelonger she stayed on the ground the more worried I became. Eventually, she gotup and unsteadily made her way to the sideline. A few questions confirmed mysuspicions: she had a concussion. It did not seem much of a knock so I was a bitsurprised, especially since onmy farm, I often see our ram hitting everything withhis head and emerging unfazed. According to an article in The New York Times(Science: October 1, 2012), animals that repetitively bang their heads (e.g.woodpeckers and antlered mammals) have developed specific adaptations toprevent brain damage. Their brains tend to be a bit smaller and have a smoothsurface compared to the richly folded and textured human brain. Also, the skulltends to be quite thick and there is little fluid between the brain and the skull, withthe result that the brain does not jostle back and forth after impact—a keymechanism of injury in humans. Lastly, the animals minimize side-to-side torsionon their brains by banging their heads only along a single plane. Gannets, largeseabirds that dive for fish, have an even more difficult problem. They dive from100 feet—hitting the water at 60 mph—and continue chasing fish underwater,using their wings to propel them. A supersized skull would be heavy. Instead, theskull is narrow, laced with air pockets along the face, and designed to displacethe energy of impact to the side. The human skull, while remarkable, is notdesigned to withstand the same sort of impact that rams, woodpeckers, andgannets routinely withstand. The implications tome are clear: alwayswear a bikehelmet and be a bit careful whipping your head around for headers in a soccergame.

Noted by WVR, MD



DOI: 10.1542/peds.2012-1128; originally published online December 24, 2012; 2013;131;99Pediatrics

Tina S. Chen, Lawrence F. Eichenfield and Sheila Fallon FriedlanderInfantile Hemangiomas: An Update on Pathogenesis and Therapy

ServicesUpdated Information &

mlhttp://pediatrics.aappublications.org/content/131/1/99.full.htincluding high resolution figures, can be found at:

References

ml#ref-list-1http://pediatrics.aappublications.org/content/131/1/99.full.htat:This article cites 68 articles, 14 of which can be accessed free

Rs)3Peer Reviews (PPost-Publication

http://pediatrics.aappublications.org/cgi/eletters/131/1/99

R has been posted to this article: 3One P

Subspecialty Collections

d_dermatologyhttp://pediatrics.aappublications.org/cgi/collection/allergy_anAllergy & Dermatologythe following collection(s):This article, along with others on similar topics, appears in

Permissions & Licensing

tmlhttp://pediatrics.aappublications.org/site/misc/Permissions.xhtables) or in its entirety can be found online at: Information about reproducing this article in parts (figures,

Reprints http://pediatrics.aappublications.org/site/misc/reprints.xhtml

Information about ordering reprints can be found online:

rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.Grove Village, Illinois, 60007. Copyright © 2013 by the American Academy of Pediatrics. All and trademarked by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elkpublication, it has been published continuously since 1948. PEDIATRICS is owned, published, PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly



http://pediatrics.aappublications.org/content/131/1/99.full.html#ref-list-1

http://pediatrics.aappublications.org/cgi/eletters/131/1/99

http://pediatrics.aappublications.org/cgi/collection/allergy_and_dermatology

http://pediatrics.aappublications.org/site/misc/Permissions.xhtml

http://pediatrics.aappublications.org/site/misc/reprints.xhtml


Infantile Hemangiomas: An Update on Pathogenesis and ... · Twolesscommontypesof“hemangioma...

Documents

Transcript of Infantile Hemangiomas: An Update on Pathogenesis and ... · Twolesscommontypesof“hemangioma...