diagnosis and management of infantile hemangioma

CLINICAL REPORT Guidance for the Clinician in Rendering Pediatric Care

Diagnosis and Management of InfantileHemangiomaDavid H. Darrow, MD, DDS, Arin K. Greene, MD, Anthony J. Mancini, MD, Amy J. Nopper, MD,the SECTION ON DERMATOLOGY, SECTION ON OTOLARYNGOLOGY–HEAD AND NECK SURGERY, and SECTION ON PLASTIC SURGERY

abstract Infantile hemangiomas (IHs) are the most common tumors of childhood. Unlikeother tumors, they have the unique ability to involute after proliferation, oftenleading primary care providers to assume they will resolve withoutintervention or consequence. Unfortunately, a subset of IHs rapidly developcomplications, resulting in pain, functional impairment, or permanentdisfigurement. As a result, the primary clinician has the task of determiningwhich lesions require early consultation with a specialist. Although severalrecent reviews have been published, this clinical report is the first based oninput from individuals representing the many specialties involved in thetreatment of IH. Its purpose is to update the pediatric community regardingrecent discoveries in IH pathogenesis, treatment, and clinical associations andto provide a basis for clinical decision-making in the management of IH.

NOMENCLATURE

The nomenclature and classification of vascular tumors andmalformations have evolved from clinical descriptions (“strawberrybirthmark,” “salmon patch,” “cavernous hemangioma,” and “port winestain”) to terminology based on their cellular features, natural history, andclinical behavior. Originally described by Mulliken and Glowacki in 1982,the most current and widely accepted classification of vascular anomaliesis that adopted by the International Society for the Study of VascularAnomalies (Table 1).1 This system includes infantile hemangioma (IH)among the vascular neoplasms, which are lesions characterized byabnormal proliferation of endothelial cells and aberrant blood vesselarchitecture. In contrast, vascular malformations are structural anomaliesand inborn errors of vascular morphogenesis.

Although IH is the most common neoplasm, this group also includes suchtumors as congenital hemangiomas, pyogenic granulomas, tuftedangiomas (TAs), and several types of hemangioendothelioma. Congenitalhemangiomas are biologically and behaviorally distinct from IH. Asreflected in the name, congenital hemangiomas are present and fullyformed at birth; they do not exhibit the postnatal proliferative phase

This document is copyrighted and is property of the AmericanAcademy of Pediatrics and its Board of Directors. All authors have filedconflict of interest statements with the American Academy ofPediatrics. Any conflicts have been resolved through a processapproved by the Board of Directors. The American Academy ofPediatrics has neither solicited nor accepted any commercialinvolvement in the development of the content of this publication.

Clinical reports from the American Academy of Pediatrics benefit fromexpertise and resources of liaisons and internal (American Academyof Pediatrics) and external reviewers. However, clinical reports fromthe American Academy of Pediatrics may not reflect the views of theliaisons or the organizations or government agencies that theyrepresent.

The guidance in this report does not indicate an exclusive course oftreatment or serve as a standard of medical care. Variations, takinginto account individual circumstances, may be appropriate.

All clinical reports from the American Academy of Pediatricsautomatically expire 5 years after publication unless reaffirmed,revised, or retired at or before that time.

www.pediatrics.org/cgi/doi/10.1542/peds.2015-2485

DOI: 10.1542/peds.2015-2485

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

Copyright © 2015 by the American Academy of Pediatrics

FINANCIAL DISCLOSURE: The authors have indicated they do not havea financial relationship relevant to this article to disclose.

POTENTIAL CONFLICT OF INTEREST: The authors have indicated theyhave no potential conflicts of interest to disclose.

FROM THE AMERICAN ACADEMY OF PEDIATRICS PEDIATRICS Volume 136, number 4, October 2015 at Miami Children's Hospital on September 28, 2015pediatrics.aappublications.orgDownloaded from

http://pediatrics.aappublications.org/

characteristic of IH. The 2 variants arethe noninvoluting congenitalhemangioma (NICH), which remainsstable without growth orinvolution,2,3 and the rapidlyinvoluting congenital hemangioma(RICH), which undergoes a rapidinvolution phase beginning in the firstyear of life (Fig 1).4 RICHs, in somecases, have been associated withthrombocytopenia but with milderand more transient coagulopathythan that seen in Kasabach-Merrittphenomenon (KMP; see discussionthat follows); rarely, they can beassociated with congestive heartfailure.5,6 Some RICHs showincomplete involution, and it ispossible that RICH and NICH lie atopposite ends of the same clinicalspectrum.7,8 Both subtypes ofcongenital hemangioma were initiallybelieved to be variants of IH thatexhibited prenatal growth until Northet al9 showed that, unlike IH, neitherlesion expresses glucose transporterprotein isoform 1 (GLUT1).

Pyogenic granuloma, also known aslobular capillary hemangioma, isneither pyogenic nor granulomatous.

It is a reactive proliferating vascularlesion that is classified as a vascularneoplasm (Table 1). This commonacquired vascular lesion of the skinand mucous membranes primarilyaffects infants and children and isfrequently misdiagnosed as IH.Approximately 12% occur in infancy,and 42% present during the first 5years of life.10 Pyogenic granulomasare most commonly located on thehead and neck, rapidly enlarge toa median size of 6.5 mm, frequentlydevelop a pedunculated base, and,with erosion, are prone to bleedingthat is difficult to control (Fig 2).10

Pyogenic granulomas are seen withhigher frequency within the skincontaining capillary malformations.

Two other distinct benign vascularneoplasms, kaposiformhemangioendothelioma (KHE) andTA, have been confused with IH. KHEpresents primarily in infancy but witha far wider age range than IH, whichis usually apparent in the first monthof life. KHE is considered a locallyaggressive neoplasm that typicallyappears as a deep, soft tissue mass.This lesion has been associated withKMP,11 a potentially life-threateningconsumptive coagulopathycharacterized by severe platelettrapping. Before KHE was describedin the early 1990s, KMP waserroneously thought to occur inassociation with IH.Histopathologically, KHE showsinfiltrating sheets of slender, GLUT1-negative endothelial cells liningslitlike capillaries.12 TAs are benignvascular tumors that occur in infants,children, or young adults and areusually located on the neck or theupper part of the thorax.13 Theirclinical appearance is variable andincludes erythematous to violaceouspatches, plaques, and nodules.Histopathologically, TA shows well-defined tufts of capillaries in thedermis that lack cellular atypia orGLUT1 positivity and, like KHE, isassociated with increased lymphaticvessels and a predisposition to KMP.Both tumors behave unpredictably

and may grow slowly over the courseof months to years, grow rapidly,spontaneously regress, or remaindormant for years.14–16 Unlike KHEand TA, IHs are not associated withthrombocytopenia or coagulopathy.

Vascular malformations arecongenital lesions, but some maybecome clinically apparent only laterin life, presumably because of slowlyprogressive ectasia resulting fromintraluminal flow. They exhibita normal rate of endothelial cellturnover throughout their naturalhistory but expand as the patientgrows. Vascular malformations do notinvolute, and their growth may beinfluenced by trauma, infection, andhormonal changes. Classification isbased on the predominant vesseltype: capillary or venulocapillary,venous, lymphatic, arterial, ormixed.17 As with vascular neoplasms,the nomenclature of vascularmalformations has led to greatconfusion. Capillary orvenulocapillary malformations havehad numerous alternativedesignations, the most common being“port wine stain” and “nevusflammeus.” Venous malformationshave often been mistaken for IH,

TABLE 1 Classification of CutaneousVascular Anomalies, 2014

Vascular malformationsVenous malformationsLymphatic malformationsCapillary malformationsArteriovenous malformations and fistulaeMixed (combined) malformations

Vascular tumorsBenignInfantile hemangioma (IH)Congenital hemangioma (rapidly involuting

[RICH]; non-involuting [NICH])Lobulated capillary hemangiomas (LCH)

(pyogenic granuloma)*Tufted angioma (TA)OthersLocally aggressiveKaposiform hemangioendothelioma (KHE)Kaposi sarcomaOthersMalignantAngiosarcomaOthers

Adapted from the International Society for the Study ofVascular Anomalies, 2014, ref 1 (issva.org/classification).*Reactive proliferating vascular lesion

FIGURE 1RICH is fully formed at birth (A) and theninvolutes, mostly during the first year of life. B,The same lesion seen at 8 months of age.

PEDIATRICS Volume 136, number 4, October 2015 e1061 at Miami Children's Hospital on September 28, 2015pediatrics.aappublications.orgDownloaded from

https://s3.amazonaws.com/ClubExpressClubFiles/298433/documents/issva_classification_2014_final_trial.pdf?AWSAccessKeyId=AKIAIB6I23VLJX7E4J7Q&Expires=1416946137&response-content-disposition=inline%3B%20filename%3Dissva_classification_2014_final_trial.pdf&Signature=WbXqfspyYqa2LbA2hgACG2WNxnM%3D


termed “cavernous hemangiomas”and “venous hemangiomas” in theliterature (Fig 3A). Lymphaticmalformations, which are subdividedinto microcystic and macrocysticvarieties on the basis of predominantlacuna size, may also be mistaken forIH when there is bleeding intovesicles at the surface of the skin ormucosa (Fig 3B). These lesions havetraditionally been referred to as“cystic hygromas” or“lymphangiomas,” designations thatinaccurately presume proliferativepotential, thereby perpetuating thediagnostic confusion.

The use of various names for IH hasresulted in immense diagnosticconfusion. For instance, the terms“capillary hemangioma” and“capillary angioma” have been used torefer to an IH that is located primarilyin the dermis and is bright red incolor. In contrast, the designations“cavernous” or “venous” haveinappropriately been used to definean IH that, because of its depth belowthe dermis, may impart a blue tinge tothe skin surface. In addition, deepvenous and lymphatic malformationsas well as arteriovenousmalformations have been incorrectlydiagnosed as deep IH. Finally, byvirtue of its sheer prevalence, theterm “hemangioma,” without theadjectival descriptor “infantile” orwith the descriptor “juvenile,” hasbeen used in reference to IH for many

years, especially predating thedistinction between IH and thecongenital hemangiomas.“Hemangioma” has also beeninappropriately used to describe, ingeneral terms, varieties of othernoninfantile hemangiomas andvascular malformations.

EPIDEMIOLOGY

Studies of the incidence of IH,including a prospective study anda review incorporating 1retrospective study and 2 cross-sectional cohorts, suggest that 4% to5% of infants are affected.18,19 Otherstudies suggest that IH is observed in1% to 3% of newborn infants20,21

and 2.6% to 9.9% of olderchildren,22,23 but methodologicshortcomings may have influencedthese findings. IHs are more common

among female infants; however,although older data suggest female-to-male ratios ranging from 3:1 to 5:1,more recent studies suggest a rangeof 1.4:1 to 3:1.23,24 The genderdiscrepancy appears to be increasedamong children with PHACEsyndrome (Posterior fossa defects,Hemangiomas, cerebrovascularArterial anomalies, Cardiovascularanomalies including coarctation ofthe aorta, and Eye anomalies), inwhich studies have found a 9:1female-to-male ratio.25 There is nota definitive explanation for thisgender difference.

Most studies report a significantlyhigher incidence in whiteinfants.23,24,26 On the basis of thesuccess of IH treatment usingb-blocker therapy, it has beenproposed that black infants mayexhibit some form of “endogenousbeta blockade,” and there aremolecular biological data to supportthis notion.27

The incidence of IH is increasedamong preterm infants, affecting 22%to 30% of infants weighing less than1 kg.24,28 Multivariate analysis hasrevealed that low birth weight (LBW)is the major contributor to this risk;there is a 25% increase in risk ofdeveloping an IH with every 500-greduction in birth weight.29 Prenatalfactors have also been investigatedfor their role in IH. Studies differregarding an increased risk resultingfrom maternal chorionic villussampling24,30 or amniocentesis,30,31

and any increased risk attributable tochorionic villus sampling appears tobe limited to procedures performedtranscervically.31 Other possibleprenatal factors include oldermaternal age, multiple gestationpregnancy, placenta previa, andpreeclampsia.24 Placental anomalies,such as retroplacental hematoma,infarction, and dilated vascularcommunications, have also beenassociated with IH development.32 Itis theorized that the common thread

FIGURE 2Pyogenic granulomas have some clinical andhistologic features similar to IHs, but they aregenerally smaller, pedunculated, and morelikely to bleed.

Highlights of This Section

• Infantile hemangioma (IH) isthe currently accepted termi-nology for the lesions that arethe focus of this clinical report.

• Congenital hemangiomas arebiologically and behaviorallydistinct from IH.

• Pyogenic granuloma is a re-active proliferating vascularlesion that is classified asa vascular neoplasm andthat may occasionally bemisdiagnosed as IH.

• Lesions diagnosed as “cav-ernous hemangiomas” areusually, in fact, deep IHs orvenous malformations.

• Kasabach-Merritt phenome-non or KMP (a consumptivecoagulopathy) is not associ-ated with IH but rather with2 other vascular neoplasms,kaposiform hemangioendo-thelioma (KHE) and tuftedangioma (TA).

e1062 FROM THE AMERICAN ACADEMY OF PEDIATRICS at Miami Children's Hospital on September 28, 2015pediatrics.aappublications.orgDownloaded from


in these associations is placentalhypoxia.32,33

Although often suggested as a riskfactor, a family history of IH isreported in only 12% of cases24;however, familial clustering has beenreported.34,35 Associations are alsoreported with maternal use of fertilitydrugs,36 use of erythropoetin,37 levelof maternal education,36 breechpresentation,23 and being the firstborn.23

PATHOGENESIS AND HISTOPATHOLOGY

Pathogenesis

The pathogenesis of IH, despiteintensive study, has not beencompletely elucidated. Lines ofevidence support a cellular originfrom either intrinsic endothelialprogenitor cells (EPCs) or angioblastsof placental origin, but intrinsic andextrinsic factors are also thought tocontribute to their development.38

Intrinsic factors include the influenceof angiogenic and vasculogenicfactors within the IH. External factorsinclude tissue hypoxia anddevelopmental field disturbances.

The EPC theory holds that IHsdevelop from clonal expansion ofcirculating EPCs, resulting in

vasculogenesis, or the de novoformation of new blood vessels.39,40

This theory is supported by studiesshowing increased numbers ofcirculating EPCs in blood samplesfrom children with IH.41 Additionalevidence comes from studies in whichmultipotential stem cells derivedfrom IH specimens (HemSCs) haveshown the ability to recapitulatehuman IH in immunodeficient mice.42

These HemSCs and cord blood EPCsbehave similarly to each other inseveral in vitro assays, suggestingthat circulating EPCs could be theorigin of IH endothelial cells.42 Theconcept that IHs originate fromcirculating multipotent progenitorcells could explain some of thefeatures they share with placentalblood vessels, because dysregulatedcirculating EPCs have also beenimplicated in many of the associatedmaternal and fetal comorbidconditions (preeclampsia, retinopathyof prematurity, etc). HemSCs havealso been shown to have anadipogenic potential,43 which mayexplain the presence of adipocytesnoted during involution. The stimulusfor division of EPCs is unknown butmay be a somatic mutation orabnormal signals from local tissues.

The theory of placental originsuggests that fetal progenitor cellsarise from the disruption of theplacenta during gestation or birth.

This concept developed fromresearch showing that molecularmarkers characteristic of placentaltissue, including GLUT1, Lewis Yantigen, merosin, Fc-g receptor-IIb,indoleamine 2,3-deoxygenase, andtype III iodothyronine deiodinase,were also present in IHs.3,9 Clinicalevidence for this theory is suggestedby those studies showing anincreased incidence of IH inassociation with chorionic villussampling, placenta previa, andpreeclampsia.24,30,31

A unifying theory suggests that IHresults from aberrant proliferationand differentiation of a hemogenicendothelium with a neural crestphenotype and a capacity forendothelial, hematopoetic,mesenchymal, and neuronaldifferentiation. It is hypothesized thatplacental chorionic villusmesenchymal core cells embolize tothe developing fetus and that thetiming of this embolization in relationto the migration of neural crest cellsalong their somitic routes determinesthe morphology of the IH (segmentalversus localized [focal]; see sectionentitled “Clinical Appearance”).44

The cytokine niche within the IH,including vascular endothelial growthfactors (VEGFs), insulin-like growthfactors, the tumor necrosisfactor–related apoptosis-inducingligand-osteoprotegerin (TRAIL-OPG)


• The incidence of in the gen-eral population is approxi-mately 5%.

• Risk factors for IH includebeing white, being female,and having a low birthweight.

• Associations are alsoreported with older mater-nal age, multiple gestationpregnancy, placenta previa,preeclampsia, use of fertilitydrugs or erythropoietin,breech presentation, andbeing the first born.

FIGURE 3A, The venous blood contained within a venous malformation imparts a bluish hue that may lead tomisdiagnosis as a deep IH. B, Bleeding into surface vesicles of a lymphatic malformation may lead tomisdiagnosis as an IH.



pathway, and the renin-angiotensinsystem, subsequently regulatesgrowth of the IH and its response topharmacologic therapies.44 Otherauthors have also embraced the“niche” concept, suggesting thatcirculating EPCs find their way tocertain locations that provideconditions favorable for growth intoplacentalike tissues.44 In tissues suchas the skin and liver, progenitor cellsmay encounter the cellular signalsand local tissue factors required tostimulate their development.

On the basis of the rapid proliferationof endothelial cells, earlierinvestigations of IH origin focused onangiogenesis, the sprouting ofendothelial cells from existing bloodvessels. Such studies have shown anincreased concentration of angiogenicfactors in IH, such as basic fibroblastgrowth factor (bFGF), VEGF-A,insulin-like growth factor, and matrixmetalloprotease (MMP) 9 within thelesion during proliferation.45 Alsoin this phase, investigatorshave identified indoleamine2,3-deoxygenase, a protein thought toslow the involution of IH by inhibitingcytotoxic T-lymphocyte response.46

During involution, endothelial cellapoptosis is accompanied bydownregulation of angiogenic factors,whereas inhibitors of angiogenesissuch as interferon-b and markers ofcell maturation such as intercellularadhesion molecule 1 areupregulated.47 It has also been shownthat involuting IHs exhibit decreasedproduction of nitric oxide,a potentiator of the VEGF pathway, asmeasured by reduced levels ofendothelial nitric oxide synthase.48

It has been hypothesized that hypoxiatriggers a vascular response ininfants. As discussed above, LBW isa significant risk factor for IH, and inutero hypoxia is a common cause ofLBW. Not surprisingly, there ismounting evidence of the role ofhypoxia in the development of IH.GLUT1, a facilitative glucosetransporter used as a marker for IH,

is an important sensor of hypoxia.9

GLUT1 has been shown to beupregulated in hypoxic zones ofmesenchymal tumors and inumbilical cord–derived humanmesenchymal stem cells underhypoxic conditions.49,50 Hypoxia-induced factors produced byendothelial cells appear to play animportant role in trafficking ofprogenitor cells to ischemic tissue.These factors have been shown to beupregulated in the blood (VEGF-A,MMP-9) and in IH tissue (stromalcell–derived factor 1a, MMP-9, VEGF-A,and hypoxia-inducible factor 1a)from children with proliferating IH.41

In addition, it has been shown thatthe use of erythropoietin in preterminfants increases the risk ofdeveloping an IH.37 Thus, tissueischemia resulting inneovascularization from circulatingEPCs has been proposed as thestimulus leading to the developmentof IH.41 Clinically, an area of pallor ordecreased blood flow in the skin hasbeen noted to precede thedevelopment of IH, furthersupporting this hypothesis.51

Histopathology

Grossly proliferative and earlyinvolutive IHs are well-circumscribed,unencapsulated masses with red-to-tan cut surfaces. Later involutivelesions are fibrofatty in consistencyand less defined. The histologicfeatures of IH change dramatically asthey proceed through their naturalcourse of neonatal presentation, rapidgrowth, and subsequent involution,requiring interpretation within theproper clinical context.52–54 There isno sharp dividing line betweenproliferation and involution, andfeatures of involution typically coexistwith features of proliferation duringmuch of the process. Early proliferativephase IHs are composed of well-defined, unencapsulated masses ofcapillaries lined by plump endothelialcells rimmed by plump pericytesembedded within a multilaminatedbasement membrane withoutassociated smooth muscle cells (Fig 4).Lesions at this stage may at leastfocally resemble other rapidly growingvascular proliferations such as earlypyogenic granulomas. The proliferatingcapillaries are arranged in lobules,separated by delicate fibrous septae orby normal intervening tissue. Theselesional capillaries, depending on tissuelocation, intermingle nondestructivelywith superficial skeletal muscle fibers,peripheral nerves, salivary glands, andadipocytes. Endothelial cells andpericytes show variably enlargednuclei and abundant clear cytoplasm.Normally configured mitotic figures arerelatively numerous (Fig 1B); andwidespread expression of cellproliferation markers, such as Ki-67,confirm that both pericytes andendothelial cells are actively dividing.Because proliferative phase IHs arehigh-flow lesions, although typicallywithout significant arteriovenousshunting, they often contain enlargeddraining veins with thick, asymmetricwalls.

Involuting IHs present differentdiagnostic challenges. Mitotic figureswane, and apoptotic bodies and masts


• may develop either from in-trinsic endothelial progenitorcells (EPCs) or from angio-blasts of placental origin.

• IH growth is affected by in-trinsic influences, such asangiogenic and vasculogenicfactors within the IH, and byexternal factors such as tis-sue hypoxia and de-velopmental fielddisturbances.

• A unifying theory proposesthat circulating EPCs migrateto locations in which con-ditions are favorable forgrowth into placentaliketissues.



cells increase in number during earlyinvolution.48,55 Lesional capillariesbegin to disappear. There is noevidence of thrombosis, andinflammation is not prominent. Asinvolution proceeds, lesional capillarybasement membranes become thickand hyalinized and contain specks ofapoptotic debris (Fig 5). Eventuallyall that remains in an end-stage lesionis loose fibrous or fibrofatty stroma,containing a few residual “ghost”vessels composed of residual,thickened rinds of basementmembrane material containingapoptotic debris and without intact

cellular linings. Epidermal atrophyand underlying fibrous scar tissuemay be present if the lesion ulceratedin the proliferative phase. Largearteries and veins modeled during thehigh-flow proliferative phase do notcompletely regress when the capillarybed drops out and thus are oftenpresent in involuting IH. Thisphenomenon, paired with loss ofendothelial mitotic activity, may leadto mistaken histologic diagnosis asa vascular malformation.Misdiagnosis can usually be avoidedby considering overall histologicappearance and clinical history.Ultimately, as discussed below, theissue can be resolved by GLUT1immunoreaction, because involutinginfantile IHs, but not malformations,will show GLUT1 immunopositivity inresidual lesion-type capillaries.9,56

Histologic examination, accompaniedby routine immunohistochemicalstudies, shows that proliferativephase infantile IHs are complexcellular mixtures with largecomplements of endothelial cells,pericytes, mast cells, and interstitialdendritic cells. Electron microscopyreveals plump endothelial cells liningsmall lumina and resting ona multilaminated basementmembrane that envelops a cuff ofpericytes. The endothelial cells of IHhave been reported to immunoreactpositively for “normal” endothelialmarkers of the blood vasculature,such as CD31, CD34, factorVIII–related antigen (von Willebrandfactor), and others.57 Currently, themost useful and widely usedimmunohistochemical marker for thediagnosis of IH is GLUT1.9,57 GLUT1 isstrongly expressed by endothelialcells of IHs at all stages of theirevolution and is not expressed byother benign vascular anomalies andreactive proliferations. GLUT1immunohistochemistry is frequentlyused to distinguish IHs from othervascular neoplasms and providesconvincing evidence that IHs areindeed as biologically distinctive asthey are clinically distinctive.

CLINICAL PRESENTATION,COMPLICATIONS, AND ASSOCIATIONS

Phases of Growth

IHs exhibit a characteristic life cycle.Clinical observations have suggestedthat there are at least 2 dynamicevolutionary phases, namely,proliferation and involution.Proliferation occurs during earlyinfancy; gradual spontaneousinvolution or regression starts by1 year of age.58–65 An intermediateperiod between proliferation andinvolution during mid-to-late infancy,often referred to as the “plateau”phase, more likely representsa period of temporary balancebetween individual cells that areproliferating and those undergoinginvolution and apoptosis.59–66 Theprocess of involution takes severalyears and varies in duration.

Proliferative Phase (Up to 12 Months ofAge)

Premonitory findings in the skinduring early infancy may includelocalized blanching or localizedmacular telangiectatic erythema.As endothelial cell proliferationcontinues, the IH enlarges, becomesmore elevated, and developsa rubbery consistency. During thisperiod, IHs often show surroundingpallor and dilatation of surroundingveins. During rapid growth periods,ulceration may arise, leading to painand eventual scarring.

IHs typically have their clinical onsetbefore 4 weeks of age.66,67 Theyproliferate for variable periods of

FIGURE 4Proliferative phase IH. Well-circumscribed lobulesof closely packed capillaries composed ofplump endothelial cells and pericytes are sep-arated by normal-appearing dermal stromalelements (hematoxylin and eosin stain; originalmagnification 3100; photo courtesy of PaulaNorth, MD.)

FIGURE 5Involutive phase IH. Lesional capillaries are setwithin loose fibro-adipose tissue and are lessdensely packed than in the proliferative phase.Note the thickened and hyalinized basementmembranes studded with apoptotic debris,reflective of the involutive process. Residuallining endothelial cells are mitotically inactive.(Photo courtesy of Paula North, MD)


• Proliferating IHs are wellcircumscribed and lacka capsule.

• Involuting IHs are fibrofattyand less defined.

• GLUT1 is a commonly usedimmunochemical marker forIH.



time, depending in part on theirmorphology and configuration.However, most IH growth appears tooccur between 1 and 2 months ofage.68 A large prospective study hasindicated that 80% of IH size isgenerally reached by 3 months, andmost growth is completed by around5 months of age.66 Deep IHs appearsomewhat later and grow somewhatlonger than their superficialcounterparts.66

Involution Phase

For most infants with IH, involutionbegins between 6 and 12 months ofage. Although the process continuesover years, the majority of tumorregression occurs before age 4.48,69,70

As IHs involute, most lesions flattenand shrink from the center outward.For those with a superficial component,this is accompanied by “centralclearing” or graying of the surface.

Although IHs generally undergospontaneous regression, observationsof “maximal involution” do notnecessarily imply completeresolution. Indeed, approximately50% to 70% of IHs resolve, leavingbehind residual skin changes,including telangiectasia, fibrofattytissue, redundant skin, anetoderma,dyspigmentation, or scar.71

Clinical Appearance

During the proliferative phase, IHscan be classified on the basis of their

soft tissue depth.61,62,72 SuperficialIHs (Fig 6A) are those in which thesurface of the tumor appears red andthere is little to no discerniblesubcutaneous component;historically, these IHs have beendescribed as being of the“strawberry” type. Deep IHs (Fig 6B)are those in which the tumor residesdeep to the skin surface, and theirsubcutaneous location results ina bluish surface hue or no evidentsurface changes; historically, thesehave been referred to as “cavernous,”an imprecise term that is no longercommonly used. Combined, mixed, orcompound IHs (Fig 6C) are those inwhich both superficial and deepcomponents coexist.

Superficial IHs tend to appear earlierand begin to involute sooner thantheir deep counterparts, which,by contrast, tend to arise laterand grow for longer periodsof time before involuting (onaverage, approximately 1 monthmore).62,64,66 Investigations intothese differences confirm that thesetimelines represent characteristicgrowth patterns for these IHs ratherthan arising out of observationalbias.66 As might be expected, thoseIHs with a mixed morphology havea growth pattern that is intermediatebetween those associated with

superficial and deep IHs. Theseobservations indicate that deep IHsrequire a longer period ofmonitoring than those withsuperficial morphology.

A specific subtype of superficial IHhas been variably referred to as anabortive, nonproliferative, arrested-growth, minimal-growth, nascent,reticular, or telangiectatic IH.73–76

This type of IH presents as a macular,telangiectatic patch that may beaccompanied by blanching of theinvolved skin (Fig 7). Unlike most IHs,abortive IHs lack an obvioussignificant proliferative phase.Approximately two-thirds of theselesions are situated on the lowerextremities. Many are accompaniedby localized, small papular regions ofvascular tissue growth, often aroundthe periphery.77 Abortive IHs sharewith more typical IHs characteristicsurface markers (eg, GLUT1),confirming that they are true IHs;however, their growth phase may bearrested. Many of these telangiectaticIHs also involute more rapidly,sometimes before 1 year of age.78

Nevertheless, complications such asulceration may occur. These IHs mayalso be segmental and occasionallyhave syndromic associations (seesection entitled “IH Syndromes andAssociations”).79

FIGURE 6Cutaneous IHs may be classified on the basis of their depth. A, Superficial IHs are visible only at theskin surface and may be focal (as shown) or segmental. B, Deep IHs have no surface involvement. C,Mixed, or compound, IHs have both superficial and deep components.


• IHs usually make their initialappearance before 4 weeksof age and complete most oftheir growth by 5 months ofage.

• Involution of IHs begins asthe child approaches 12months of age. In most cases,the majority of involution iscompleted by age 4.



IHs may also be classified on the basisof their anatomic configuration aseither localized (focal), segmental,indeterminate, or multifocal.26,80,81

Localized (focal) IHs are discretelesions that seem to arise froma single focal point, whereassegmental lesions cover a territorythat is presumed to be determined byembryonic neuroectodermalplacodes.82 Segmental IHs tend toinvolve a larger surface area of skin.Segmental IHs of the face have beenobserved to conform to uniquedevelopmental units, which havebeen mapped into 4 distinct patterns:frontotemporal, maxillary,mandibular, and frontonasal(Fig 8).82 Lesions that are not

definitively focal or segmental areconsidered indeterminate. Multifocallesions are focal lesions occurring atmore than 1 anatomic site. One largestudy found that most IHs (67.5%) arelocalized, whereas the remainder weresegmental (13%), indeterminate(16.5%), or multifocal (3.6%).83

The presence of a large, facialsegmental IH is a hallmark sign ofPHACE syndrome,25 whereas largesegmental IHs of the anogenital andlumbosacral areas may be associatedwith genitourinary system andspinal cord anomalies as part ofother syndromes84–86 (see sectionentitled “IH Syndromes andAssociations”). More recently, it hasbeen recognized that extracutaneousmanifestations may also arise inassociation with segmental IHsinvolving other anatomic sites, aspart of the so-called PHACE-without-face phenomenon.87 These patientsmay have segmental IHs of the upperchest, shoulder, or arm in theabsence of facial IH involvement andin conjunction with structural heartdisease, aortic or other major vesselanomalies, central nervous systemand sternal defects, or eyeanomalies.

Multifocal cutaneous IHs arefrequently isolated to the skin butmay also serve as markers forunderlying hepatic involvement(Fig 9).88–91 Previous retrospectivereports26,81 suggested that thepresence of a large or segmental(.5 cm) cutaneous IH might provea useful marker for hepatic IHs.However, results from a largeprospective study suggest that it isthe number of cutaneous IHs ratherthan their size that is the morepredictive factor.92 When 5 or moreIHs are present on cutaneousexamination, ultrasonography maybe helpful in assessing potentialhepatic involvement.84,93

Hepatomegaly and congestive heartfailure also suggest the presence ofliver IH.

FIGURE 7Abortive IHs are macular, telangiectatic patchesthat have failed to fully proliferate.

FIGURE 8(A) Patterns of segmental IH of the face extracted from image analysisdefined. Seg1 (fronto-temporal), Seg2 (maxillary), Seg3 (mandibular), and Seg4 (frontonasal). (B) An ulcerated segmentalIH in the maxillary distribution.

FIGURE 9Multifocal cutaneous IHs in a child with IH ofthe liver.


• IHs are characterized as su-perficial, deep, or mixed andas focal, multifocal, orsegmental.

• Superficial IHs appear ear-lier and begin involutionsooner than their deepercounterparts.

• Segmental IHs are morecommonly involved inPHACE (see text for defini-tion) and other IH syn-dromes and associations.

• The presence of more than 5focal IHs suggests a higherrisk of hepatic involvement.



Complications

Although most IHs do not requireurgent treatment, a minority maydevelop function-threatening or life-threatening complications,necessitating therapeuticintervention. One study determinedthat approximately 24% of patientswith IH who were referred to a groupof tertiary care dermatology practicesexperienced some complicationrelated to their IH.84 It is thereforeprudent for pediatric providers toremain vigilant of possiblecomplications and of risk factors thatmay herald future complications.

Ulceration accounts for the majorityof IH complications; others includebleeding, visual impairment,auditory impairment, congestiveheart failure, and airwayobstruction.84 Gastrointestinalbleeding has been reported asa complication of segmentalintestinal hemangiomatosis, in whichthe IH is typically situated in thedistribution of the mesentericarterial system.94

The single best predictor ofcomplications and the need fortherapeutic intervention for IH ismorphologic subtype.84 Focal IHshave the potential to causecomplications primarily by virtue oftheir location on or near vitalstructures, such as the eye(amblyopia, astigmatism), nose(anatomic distortion andcartilaginous destruction), ears(anatomic distortion andcartilaginous destruction), lips(anatomic distortion and ulceration),airway (obstruction), or anogenitalregion (ulceration). On the face, focallesions are 3 times more commonthan segmental IHs.81 Segmental IHsare more frequently complicated byulceration.84 A prospective cohortstudy in 1058 patients undertaken toidentify clinical characteristicspredicting complications and need fortreatment found, after controlling forsize, that segmental IHs were 11times more likely than localized IHs

to develop complications and 8 timesmore likely to receive treatment.84

Segmental lesions tend to have longerproliferative phases, some withsignificantly prolonged duration ofgrowth as long as 10 to 44 months,and may therefore requiresignificantly longer treatmentdurations.94

The size of the IH was also animportant predictor of the need fortreatment in the aforementionedcohort study, although this analysisdid not appear to control foranatomic subtype.84 The mean size ofcomplicated IHs was 37.3 cm2,compared with 19.1 cm2 foruncomplicated IHs. In addition, IHsthat received treatment of any typehad a mean size of 30.4 cm2, whichwas 11.1 cm2 larger than those thatdid not receive treatment. However,the mean size of segmental IHs isapproximately 10 times that oflocalized IHs,66 suggesting thatmorphology may indeed be a moreimportant indicator of potentialcomplications.

Anatomic location was alsoa predictor of complications due toIH.84 Facial IHs were complicated 1.7times more frequently than nonfacialIHs; they were also 3.3 times morelikely than their nonfacialcounterparts to receive some form oftherapy, likely because of concernsfor cosmesis. Periocular IHs andthose in the “beard” distribution arealso more likely to requireintervention, as described below. In1 study, perineal IHs were the mostlikely to ulcerate.95

Ulceration

Ulceration, or breakdown of the IHskin surface, occurs with anestimated incidence of 5% to 21%.96

Ulceration was the most commoncomplication in a large prospectivecohort of children with IHs,occurring in 16% of the studypopulation.97 Ulceration can lead tosignificant pain, bleeding, andsecondary infection. Ulceration also

usually results in scarring, with therisk of permanent disfigurement. Asa result, prompt initiation of therapyis essential in the management ofulcerating IHs.

The specific mechanisms resulting inIH ulceration are poorly understood.It has been hypothesized thatulceration may develop secondary toincreased tissue hypoxia, which leadsto the development of dermal fibrosisand then progresses to surfacebreakdown.98 In such cases, earlywhite discoloration of an IH, possiblyrepresenting superficial dermalfibrosis, may be a premonitory sign ofimpending ulceration.99 Otherproposed mechanisms includeoutgrowing of the blood supply orrapid expansion exceeding the elasticcapabilities of the skin.99,100

Several studies have shown thatcertain subsets of patients with IHsare at higher risk of ulceration. Asdiscussed previously, superficial andsegmental IHs have been found tobe at higher risk ofulcerating.96–98,101 In addition,specific locations at higher risk ofulceration include the head, neck,perioral, and perineal/perianalregions and intertriginous sites(Fig 10).96–98,102 The neck andanogenital regions sustainmaceration and friction, which maycontribute to the development ofulceration. Ulceration has also beennoted to occur more frequently ininfants younger than 4 months,a period of time during which the IH

FIGURE 10Ulcerated segmental IH of the perineal/perianalregion.



is actively proliferating.96–98 See thesection entitled “Management ofUlcerated IH” for a discussion ingreater detail.

Bleeding

Although concern for potentialbleeding in IH is common amongcaregivers and providers, it occursrarely and almost exclusively inulcerated lesions. The majority ofbleeding that occurs in nonulceratedIHs is minor and easily controllablewith pressure. The most commonsuch scenario is an IH that hassustained minor surface trauma (ie,from friction or a fingernail), bledminimally, stopped bleedingspontaneously or with minimalsustained pressure, and subsequentlypresents with surface hemorrhagiccrusting.

In a large prospective study ofulcerated IHs, bleeding occurred in41% of lesions but was clinicallysignificant in only 2% of thesecases.96,98 Significant bleedingrequiring blood transfusion or otherintervention is infrequentlyreported.98 Rare instances of life-threatening bleeding have beenobserved, including 1 report ofulceration of a segmental neck IH intoarterial vessels, the bleeding fromwhich necessitated transfusions,systemic and topical treatment of theIH, embolization, and surgicalexcision.102

Feeding Impairment

Feeding impairment can occur ininfants with IHs involving either theperioral region or the airway. Infantswith ulcerated lip IHs may be unableto latch onto a nipple secondary tosevere pain, which can lead toimpaired feeding.103 Obstructiveairway IHs may complicatebreathing and swallowing, alsoleading to impaired feeding.104 Ina small case series in infants withcomplicated facial IHs, several withulcerated perioral lesions or airway

IHs had feeding and oral sensoryproblems that resulted in failure tothrive.105

Airway Involvement and Obstruction

Airway IHs can occur in the presenceor absence of skin findings.Symptomatic obstructive airway IHs,including supra- and subglottic IHs,usually present with progressivebiphasic inspiratory and expiratorystridor during the first 6 to 12 weeksof age as the lesion is proliferating.105

Affected infants may also rapidly

develop noisy breathing or a hoarsecry.106

The cutaneous findings associatedwith underlying airway involvement,when present, help to identify thosepatients at greatest risk of airwayIHs. Cutaneous IHs in a “beard”distribution, defined as involving thepreauricular regions, chin, anteriorneck, or lower lip (Fig 11), have beenassociated with airwayinvolvement.106,107 Infants with IHswithin this distribution bilaterallyappear to be at an even higher risk of

FIGURE 11A, The presence of multiple IHs in the “beard” distribution is associated with a higher likelihood ofairway involvement (reproduced with permission from J Pediatr. 1997;131(4):643–646 ©Elsevier).106

B and C, Patient with airway involvement requiring tracheotomy is shown with “beard” involvementat the lip and chin (B) as well as the parotid area and neck (C).



having associated airwayinvolvement; in a recent series in17 infants with airway IHs, bilateralinvolvement of the lower facialsegment was present in 13(76%).106,108 Early referral tootolaryngology of infants with severestridor and a cutaneous IH in the“beard” distribution is advisable,because airway involvement can belife-threatening if diagnosis andtreatment are delayed. In lesssymptomatic children, a highkilovoltage radiograph of the airwaymay be useful in identifyingsubglottic IH. Airway IHs arediscussed in greater detail in thesubsection under “IHs With SpecialAnatomic Concerns” entitled“Airway.”

Visual Impairment and Other OcularComplications

IHs occurring within the orbit have thepotential to cause mechanical ptosis,strabismus, anisometropia, orastigmatism, which can quickly lead tothe development of amblyopia.108

Studies have identified specificcharacteristics of periocular IHs, whichplace the child at higher risk ofamblyopia. These include periocularIHs that are larger than 1 cm indiameter, nasal location of the IH,associated ptosis, eyelid marginchange, or displacement of theglobe.109–111 Orbital IHs are discussedin greater detail in the subsectionentitled “Eye and Orbit” under “IHsWith Special Anatomic Concerns.”

Congestive Heart Failure andHypothyroidism

Although rare, high-output congestiveheart failure can occur in infants withlarge IHs as a result of arteriovenousshunting of a large blood volumethrough the lesion. This complicationhas been reported in infants withlarge cutaneous IHs and RICHs and inthose with diffuse or multifocalhepatic IHs.91,92,112,113 Symptomaticinfants may present with difficultyfeeding, poor growth, heart murmur,or hepatomegaly. The cardiac

compromise usually improves withtreatment of both the heart failureand the IH.

Diffuse lesions of the liver may alsobe associated with severeconsumptive hypothyroidism causedby excess production of type 3iodothyronine deiodinase. Liver IHsare discussed in greater detail in thesubsection entitled “Liver” of “IHsWith Special Anatomic Concerns.”

IH Syndromes and Associations

A small subset of children with IH willexhibit associated congenital

anomalies. The best-known suchassociation is PHACE (OnlineMendelian Inheritance in Man606519).114 The disorder is alsoreferred to as PHACES to includepotential ventral midline defects,specifically Sternal cleft and/ orSupraumbilical raphe. Originallydescribed as a “syndrome,” PHACE ismore appropriately termed anassociation, although there are recentdata suggesting that chromosomalregion 7q33 may provide a geneticsusceptibility to exhibit the PHACEphenotype.115

The spectrum of anomalies in PHACEsyndrome and the ipsilateralrelationship between such anomaliesand cutaneous IH strongly suggesta “developmental field defect,”whereby an insult at a critical time inembryogenesis gives rise to similardevelopmental outcomes.116 Theprecise timing of such an insult inPHACE syndrome is speculative, butboth the anatomic IH patterns andseveral of the associated structuralabnormalities point to changes earlyduring the first trimester, probablywithin the first 3 to 12 weeks ofgestation before or during earlyvasculogenesis.117 PHACE syndromeis now understood to bepredominantly a congenitalvasculopathy. In fact, many of itsfeatures can be explained asdownstream events of arteriopathywith resultant ischemia, and it hasbeen hypothesized that vasculardysplasia may be a key or evenprimary event in the pathogenesis ofPHACE syndrome.118

Consensus criteria were recentlydeveloped for the diagnosis of PHACEsyndrome (Tables 2 and 3).25 Clinicalexamination of the skin and eyes aswell as detailed imaging of the head,neck, and chest are required to makethe diagnosis. More than 90% ofinfants with PHACE syndrome exhibitmore than 1 extracutaneous anomaly,although very few manifest thecomplete spectrum.119 In contrast tononsyndromic IH, PHACE syndrome


• Segmental IHs are far morelikely than focal IHs to resultin a complication, usuallyulceration.

• Focal IHs cause complica-tions primarily by virtue oftheir location on or near vi-tal structures.

• Facial IHs cause complica-tions more frequently thannonfacial IHs and are severaltimes more likely to receivesome form of therapy.

• Minor bleeding from anulcerated IH is common, butrarely of clinical significance;bleeding from a non-ulcerated IH is rare.

• Patients with an extensiveIH in the “beard” distribu-tion are more likely to haveinvolvement of the airway.

• High-risk periocular IHs arethose that are that are largerthan 1 cm in diameter, lo-cated near the nose, associ-ated with ptosis or eyelidmargin change, or displacingthe globe.

• Diffuse IH of the liver maybe associated with severeconsumptivehypothyroidism.



is more common in full-termsingleton infants of normal birthweight, although females are stillmore commonly affected.120

The hallmark of PHACE syndrome isa large, segmental, often superficialIH, characteristically located on theface, scalp, and/or neck (Fig 12).PHACE syndrome–associated IHsmost commonly affect facial segments1 and/or 3, which also confers aparticularly high risk of associatedcentral nervous systeminvolvement.84,119

PHACE syndrome is not exceedinglyrare, and probably even morecommon than Sturge-Webersyndrome.121 The segmental IHassociated with PHACE issometimes confused with the portwine stain associated with Sturge-Weber syndrome, especially in thenewborn period before significantIH proliferation or in cases of“minimal growth” IH in which thereis an absence of significantproliferation. The risk of PHACEsyndrome in an infant presentingwith a large, segmental IH ($22 cm2)of the head or neck is approximatelyone-third.120

Cerebrovascular anomalies, presentin more than 90% of patients, are themost common extracutaneous featureof PHACE syndrome, followed bycardiac anomalies (67%) andstructural brain anomalies (52%).84

The most common arterialabnormality in PHACE syndrome isdysgenesis of the anterior circulation,particularly within the internalcarotid artery.119 The neuroanatomicand cerebrovascular anomaliesobserved in PHACE may lead toa number of neurologic sequelae,including motor and speech delays,seizures, migraine-like headaches,and rarely, arterial ischemicstroke.121 Hearing loss (conductive orsensorineural) has also been reportedin PHACE syndrome, particularlywhen the IH involves the ear andperiauricular scalp, which can berelated to the presence of ipsilateral

TABLE 2 Consensus Algorithm for the Diagnosis of PHACE Syndrome

PHACE Syndrome Possible PHACE Syndrome

Facial hemangioma.5 cm in diameterplus 1 major criterionor 2 minor criteria

Facial hemangioma.5 cm in diameterplus 1 minorcriterion

Hemangioma of theneck or uppertorso plus 1 majorcriterion or 2minor criteria

No hemangioma plus2 major criteria

Adapted from ref 25.

TABLE 3 Consensus Diagnostic Criteria for PHACE Syndrome

Organ System Major Criteria Minor Criteria

Cerebrovascular Anomaly of major cerebral arteriesDysplasiaa of the large

cerebral arteriesb

Arterial stenosis or occlusionwith or without moyamoyacollaterals

Persistent embryonic artery otherthan trigeminal arteryProatlantal intersegmentalartery (types 1 and 2)

Primitive hypoglossal arteryPrimitive otic artery

Absence or moderate-severehypoplasia of the largecerebral arteries

Aberrant origin or course ofthe large cerebral arteriesb

Persistent trigeminal arterySaccular aneurysms of anycerebral arteries

Structural brain Posterior fossa anomalyDandy-Walker complex or

unilateral/bilateral cerebellarhypoplasia/dysplasia

Enhancing extraaxial lesion withfeatures consistent withintracranial hemangioma

Midline anomalyc

Neuronal migration disorderd

Cardiovascular Aortic arch anomaly Ventricular septal defectCoarctation of aorta Right aortic arch (double aortic arch)Dysplasiaa

AneurysmAberrant origin of the subclavianartery with or without avascular ring

Ocular Posterior segment abnormality Anterior segment abnormalityPersistent hyperplasticprimary vitreous

MicrophthalmiaSclerocornea

Persistent fetal vasculature ColobomaRetinal vascular anomalies CataractsMorning glory disc anomalyOptic nerve hypoplasiaColobomaPeripapillary staphyloma

Ventral or midline Sternal defect HypopituitarismSternal cleft Ectopic thyroidSupraumbilical rapheSternal defects

Adapted from ref 25.a Includes kinking, looping, tortuosity, and/or dolichoectasia.b Internal carotid artery, middle cerebral artery, anterior cerebral artery, posterior cerebral artery, or vertebrobasilarsystemc Callosal agenesis or dysgenesis, septum pellucidum agenesis, pituitary malformation, or pituitary ectopia.d Polymicrogyria, cortical dysplasia, or gray matter heterotopia.



intracranial IH involving auditorystructures. It has been suggested thatchildren with PHACE syndrome andperiauricular IH be evaluated withboth MRI and audiometric testing.122

Cardiovascular anomalies are thesecond most common extracutaneousmanifestation of PHACE syndrome. Theaortic coarctation observed differs fromclassic coarctation in that it occurs ina more proximal location, ofteninvolves the arteries feeding the upperextremities, and affects longersegments, which may precludedetection based on a blood pressuregradient between the upper and lowerextremities. The aortic anomalies inPHACE syndrome are often noted to beparticularly unusual and severe, oftenrequiring surgical repair; thus, detailedimaging of the arch is essential.123

Even in asymptomatic infants, MRI ormagnetic resonance angiography(MRA) of the head and neck isindicated, especially given the knownpotential for progressive vasculopathyand resultant ischemic events in a smallsubset of severely affected patients.124

Arterial ischemic stroke, a rare butdevastating complication, appears to bemore likely in patients with PHACEwho exhibit significant narrowing ornonvisualization of large cerebralarteries, especially when more than 1vessel is involved and/or if there areassociated cardiovascular morbiditiessuch as coarctation of the aorta.125

Through serial neuroimaging of high-

risk infants, progressivecerebrovascular changes may beidentified early, and neurosurgicalrevascularization procedures can beperformed to potentially reduce arterialischemic stroke–related morbidity andmortality.125 The presence of severecerebrovascular and/or cardiovasculararterial anomalies in patients withPHACE syndrome may preclude the useof propranolol for treatment of IH inthis population, or require dosemodification. (see section entitled“Medical Therapy for IH”).

LUMBAR syndrome (Lower body IHand other cutaneous defects, Urogenitalanomalies and ulceration, Myelopathy,Bony deformities, Anorectalmalformations and arterial anomalies,and Renal anomalies) may be bestconsidered the “lower half of the body”variant of PHACE syndrome.85

LUMBAR has also been previouslydescribed under the competingacronyms SACRAL87 (Spinaldysraphism, Anogenital anomalies,Cutaneous anomalies, Renal andurologic anomalies, associated withAngioma of Lumbosacral localization)and PELVIS86 (Perineal hemangioma,External genitalia malformations,Lipomyelomeningocele, Vesicorenalabnormalities, Imperforate anus, Skintag). The IHs are usually segmentallumbosacral or anogenital lesions. In ananalysis of 24 new patients anda review of 29 published cases, IHs inassociation with LUMBAR were noted

to be segmental and often “minimalgrowth” in morphology.85 Such IHswere often extensive (eg, involving theentire leg) and showed additionalpotential for ulceration and, rarely,underdevelopment of the affected limb.Like PHACE syndrome, the cutaneousIHs and underlying anomalies showedregional correlation. Myelopathies,particularly spinal dysraphism, werethe most common extracutaneousanomaly. Interestingly, arterialanomalies were noted in a minority ofpatients who were specifically studiedfor such anomalies, although the trueincidence and long-term risks areunknown. Imaging is region-specific;MRI is useful in delineating the extentof lumbosacral involvement andpotential myelopathy, whereasadditional imaging with MRA may bewarranted for infants with extensivelower limb involvement to assess forarterial anomalies.

FIGURE 12A, Frontotemporal segmental IH typical of PHACE syndrome. B, Sternal clefting characteristic ofPHACE syndrome (scar is congenital, not surgical).


• PHACE syndrome includesfeatures of Posterior fossadefects, Hemangiomas, cere-brovascular Arterial anoma-lies, Cardiovascularanomalies including co-arctation of the aorta, andEye anomalies.

• The hallmark of PHACEsyndrome is a large, seg-mental IH, characteristicallylocated on the face, scalp,and/or neck.

• The most common extra-cutaneous features of PHACEsyndrome are cerebrovascu-lar anomalies, followed bycardiac anomalies andstructural brain anomalies.

• LUMBAR syndrome may bebest considered the “lowerhalf of the body” variant ofPHACE syndrome and maybe associated with urogeni-tal, anal, skeletal, and spinalcord anomalies.



IMAGING OF IH

The diagnosis of IH is generally madeon the basis of history and clinicalappearance. Occasionally, imaging ofthe lesion may be required when thediagnosis is uncertain, whenevaluation of extent is necessary, orwhen response to therapy needs to bemonitored. IH-associated anomalies,such as spinal dysraphism,anogenitourinary anomalies, andPHACE syndrome, are also bestimaged with MRI. However, the risk ofearly exposure to anesthesia isa consideration in determining theurgency and type of imaging for IH.

Ultrasonography is a reasonableinitial imaging modality fordiagnosing IH, because it isinexpensive and does not requiresedation. The sonogram generallyreveals a well-defined high-flowparenchymal mass with possibleshunting. During the involutionphase, areas of increased echogenicity(fat replacement) can be seen withinthe lesions. Gray scale and colorDoppler ultrasonography have alsodemonstrated utility in monitoringthe response of IH to medicaltherapy.127 Ultrasonography is alsoa good first-line modality to screenpatients with multifocal IHs for liveror visceral involvement, although MRIis preferable to assess complicated orextensive visceral lesions.128

The extent of the lesion and thesurrounding anatomy are bettershown on MRI. The most usefulsequences include T1-weightedimages with and without fatsaturation, T1-weighted images withfat saturation post–gadoliniumadministration, T2-weighted imageswith fat saturation, and flow-sensitivesequences such as gradient echo orMRA.129 Proliferating IHs typicallyappear as well-defined masses withfeatures of high flow and intermediatesignal intensity on T1-weightedimages and high signal intensity onT2-weighted images.130,131 Flow voidsmay be apparent on T2-weighted andflow-sensitive sequences, along with

high-flow feeding arteries anddraining veins. With administration ofintravenous gadolinium, lesionenhancement is usually early, withintense and uniform enhancement ondelayed images; however, diffuse ormultifocal hepatic lesions may showearly enhancement peripherally anddelayed filling centrally (centripetalenhancement).91 Nonenhancingregions may represent thrombosis ornecrosis. During the involution phase,as deposits of fat replace the lesion,foci of increased signal can be seen onT1-weighted images, and postcontrastimages reveal less avid enhancement.

Computed tomography (CT) ismentioned only because it isoccasionally ordered whena diagnosis of IH is not expected. CTfindings are similar to those on MRI,including well-defined and avidlyenhancing lesions during theproliferating phase and lesspronounced enhancement during theinvolution phase. Although thesestudies are shorter in duration thanMRI (reducing the likelihood thatgeneral anesthesia will be necessary),CT carries the disadvantage ofexposing young children to ionizingradiation and its attendant risks.

CLINICAL APPROACH TO IH

The traditional clinical approach toIH has been one of “benignneglect.”60 The observation that, asa neoplasm, IH has the uniqueability to undergo involution hasled many practitioners to believethat the best management is to“leave it alone and it will go away.”However, 1 study from a group ofreferral pediatric dermatologypractices suggests that more thanone-third of patients with IHrequire some sort ofintervention.84 Hence, althoughthis number may reflect referralbias, it is clear that some IHs areassociated with a high risk ofcomplications or permanentdisfigurement. In many such cases,early intervention may be justifiedto potentially arrest the growth ofthe lesion, reduce associatedcomplications, and avoid years ofpsychosocial concerns.

The first consideration in themanagement of IH is whetherintervention is necessary. Theindications for intervention includethe following: (1) emergencytreatment of potentially life-threatening complications; (2)urgent treatment of existing orimminent functional impairment,pain, or bleeding; (3) evaluation toidentify structural anomaliespotentially associated with IH; and(4) elective treatment to reduce thelikelihood of long-term orpermanent disfigurement. Life-threatening lesions includeobstructing IHs of the airway aswell as liver IHs associated withhigh-output congestive heartfailure and severe hypothyroidism.Pain and bleeding are examples ofurgent sequelae that occur asa result of ulceration; affectedchildren may have failure to thriveas well. Other examples offunctional impairment includeocular impairment (loss of visualaxis leading to deprivationamblyopia, astigmatism,


• Imaging of IH is not usuallynecessary.

• When imaging of IH is per-formed, ultrasound is gen-erally the preferred modalityfor diagnosis, whereas MRIis better to assess extent ofthe lesion.

• Imaging may be requiredwhen the diagnosis is un-certain, when evaluation ofextent is necessary, whenthe IH is a possible markerof PHACE or LUMBAR syn-drome, or when response totherapy needs to bemonitored.



strabismus, visual field cuts),impaired feeding because ofinvolvement of the lips or mouth,and reduced mobility because ofcomplicated involvement of theextremities. Structural anomalies ofconcern include spinal dysraphismassociated with lumbosacral IHs aswell as anomalies associated withPHACE and other IH “syndromes.”Long-term and permanentdisfigurement is a concern with IHsof certain anatomic sites (eyelid,nasal tip, lip, breast, genitalia) andwith severe ulceration, becausethese will ultimately leave a scar ofsimilar size.

Most IHs are uncomplicated andare not likely to fall into any ofthese categories (Fig 13); thepractice of initial observation or“watchful waiting” is reasonablefor such lesions. However,because the clinical presentationof IH can change within days,it is prudent for pediatricproviders to reexamine frequently,as often as weekly, thosechildren with lesions at high riskof causing functional orcosmetically critical changes,because many uncomplicatedlesions may become complicated

ones in early infancy.132 For lessconcerning IHs, providers maywish to counsel the familyregarding changes of importance,such as rapid growth or ulceration,and to establish with the familya means to see the child on shortnotice if such changes areobserved.

Central to the decision of whetherto intervene is a discussion of therisks, benefits, and alternativesassociated with each of thesechoices and with each potentialintervention. Proper informedconsent from the family isparamount in achievinga successful and satisfyingphysician-patient relationship aswell as a good therapeutic result.

Once a decision has been made tointervene, the second consideration iswhich therapeutic modalities are mostappropriate. There is no formula oralgorithm that easily addresses all ofthe factors in this decision; as a result,the treatment plan is customized foreach patient. Relevant factors includeage and medical condition of thepatient; growth phase, location,and size of the lesion or lesions;degree of skin involvement; severity of

complication and urgency ofintervention; potential for adversepsychosocial consequences; parentalpreference; and clinician experience.A Cochrane review found a dearth ofwell-designed clinical trials on whichappropriate interventions for IH couldbe determined.133

Whether IH affects the psychosocialwell-being of affected patients, and theage at which it may do so, is uncertain.Some authors suggest that “childrenfirst represent and reflect onthemselves as independent, objectiveentities” in the latter half of the secondyear of life134 and that the emotionalresponses of others may affecta child’s mood even earlier than 12months of age.135 Thus, there may besome effect on the child even beforeentering preschool. In contrast,a review of the existing literature onpsychosocial ramifications of IH wasless concerning.136 Among the 7studies cited, questionnaires that werevalidated but not specific for IHrevealed few or no signs ofpsychosocial impact in children withIHs. The authors did note that thestudies were conducted in smallgroups of parents, and all were flawedin one way or another. In a small study,quality of life among children 5 to 8

FIGURE 13Uncomplicated IHs. These lesions generally do not require medical or surgical intervention.



years of age who had IHs of the headand neck measuring 2 cm or greaterwas compared with that of normalcontrols. Although the questionnaireswere not disease-specific, the authorsfound no differences in quality-of-lifeindices or in self-perception scores.137

Further complicating this subject isthe fact that quality-of-lifeinvestigations in young childrenrequire proxy reporting that oftenreflects quality of life as perceived bythe proxy rather than the child. Thus,no definitive statement can be maderegarding psychosocial ramificationsof IH; however, many clinicians whospecialize in the field will considertreatment of those IHs projected to becosmetically significant beyond 4years of age.70

Details related to IH stage have beenaddressed in a study examininggrowth characteristics of IHs.66 Thelargest increase in IH size occurred ata mean age of 3 months, and by 5months of age both segmental andlocalized IHs had reached 80% oftheir final size. As a result, therapyinitiated after the early proliferativephase is less likely to be effective incontrolling the growth of the lesion orin prevention of complications. Ageand stage are also factors in theeffectiveness of pharmacotherapy forIH. Early experience with systemicsteroids for IH suggested thatyounger children had a betterresponse to therapy,138,139 andmost studies of their mechanism ofaction suggest they inhibitcomponents of the proliferativeprocess.140 Similar data forpropranolol therapy are not yetavailable, but proposed mechanismsof its action also suggest primaryactivity during proliferation.139,141

Conversely, recent retrospectivestudies indicate that b-blockertherapy for IH can be effectivebeyond the proliferative phase aswell.142,143

Age and growth phase are alsoconsiderations when surgery for IH isbeing considered. For example, in

patients who have undergone earlysurgery and in whom residual IHintentionally or inadvertently remainspostoperatively, there is potential foradditional growth of the lesion afterthe procedure. In addition, becauseIHs are benign lesions with thepotential to involute, surgeons do notalways endeavor to obtain disease-freemargins, instead allowing involutionto assist in achieving the final result.On occasion, in surgery for well-involuted IHs, tissue may even be leftbehind intentionally, using thefibrofatty remnant to serve as filler topreserve normal tissue contours.

Lesion-related factors, such aslocation, size, and degree of skininvolvement or ulceration, oftendictate the feasibility of a giventreatment modality. For example,a pedunculated eyelid lesioncausing ptosis or ectropion ora small, ulcerated lesion that iscertain to scar may lend itselfbetter to early surgical excisionrather than medical therapy.Conversely, an extensivelyulcerated segmental IH or a lesionof the genitalia is moreappropriately addressed withmedical therapy.

Parental preference is anothercritical factor in treatmentselection, more so in elective casesthan in those that are emergenciesor urgent cases. Parents will oftenhave strong feelings, particularlyabout surgical intervention orsystemic medical therapy, that willhave a major effect on treatmentplanning for the child. Similarly, thechoice of treatment may beinfluenced by the experience of thetreating provider and his or herfamiliarity with the diversity ofpotential interventions. For thesereasons, it may be preferable forcomplicated IHs to be managed bya practitioner or team who hasexperience with all of the availabletherapeutic options.

MANAGEMENT OF ULCERATED IHS

The management of ulcerated IHsincludes attention to wound care,pain, and IH growth.96 Unfortunately,there are no high-quality studies toguide care of the ulcer, and thereforetreatment preferences are oftenbased on case experience.132


• The indications for in-tervention for IH include thefollowing:1. emergency treatment of

potentially life-threateningcomplications;

2. urgent treatment ofexisting or imminentfunctional impairment,pain, or bleeding;

3. evaluation to identifystructural anomalies po-tentially associated withIH; and

4. elective treatment to re-duce the likelihood oflong-term or permanentdisfigurement.

• There is no algorithm to de-termine the most appropri-ate intervention for IH.Factors affecting this choiceinclude the following:1. age of the patient,

2. growth phase of the lesion,

3. location and size of thelesion,

4. degree of skininvolvement,

5. severity of complicationand urgency ofintervention,

6. potential for adversepsychosocialconsequences,

7. parental preference, and

8. physician experience.



Approaches to ulcerated IHs aresummarized in Table 4.

Some clinicians liken themanagement of ulceration to that ofsuperficial burns and suggest cultureand diligent wound care. In 1 study,16% of ulcerated IHs wereconsidered to be infected on clinicalgrounds, and cultures revealedpathogens in half of these cases.98

Reepithelialization may be facilitatedby debriding crusts with the use ofwarm compresses. The ulcer is thencovered with a barrier to preventexcessive drying, control pain, reducethe risk of trauma and potentialbleeding, and reduce the risk ofbacterial colonization or infection.Such treatments may consist oftopical antibiotics or anesthetics,wound dressings, barrier creams, orall of the above. Most ulcerationsimprove with this conservativeapproach to wound care.

Because ulceration is usuallyassociated with proliferation of theIH, therapies to curb its growth areoften used. Several case series havereported successful treatment of IHulceration with propranololtherapy.144–146 Topical timolol hasbeen reported to be successful forulceration,147 but its absorption isunpredictable in this setting. Systemicsteroids may also be a reasonablealternative.

In refractory cases, pulsed-dye laser(PDL) therapy may also be effective inmanaging ulcerated IHs.96,148–150 Ina prospective study in 78 children,91% of the patients responded tolaser therapy with a mean number of2.0 treatments.148 Thus, PDL therapyhas been used as both monotherapyand as adjunctive therapy inmanaging ulcerated IHs; however, ithas been recommended that lasertherapy be used with caution inpatients with proliferating IHs becauseof the risks of atrophic scarring and/orulceration.151 Surgical excision mayalso be a consideration for smallulcerations that are poorly responsiveto medical therapy.

Pain control is a significant issue ininfants with ulcerated IH. Pain can besevere and can disrupt sleep as wellas interfere with daily activities and/or function. For example, ulcerationlocated on the lips or oral mucosamay affect oral intake or feeding,whereas interference with urinationor stooling may be seen in the settingof perineal ulcerations. Oralacetaminophen and cautious use oftopical 2.5% lidocaine ointment maybe effective in managing the pain ofulceration.96 With more severeulceration, the use of narcotics maybe indicated for inadequatelycontrolled pain. Collaboration withexperts in pain management may beuseful in this high-risk group ofinfants.

MEDICAL THERAPY FOR IH

Background

Medical therapy for IH includesboth topical and systemicadministration of medications.Topical agents may bea consideration for smaller, moresuperficial IHs or those for whichsystemic therapy is contraindicated.Systemic therapy is usually initiatedfor large IHs, those with a high riskof functional impairment ordisfigurement, and those refractoryto other initial therapies.

Beginning in the 1960s, systemic andintralesional steroids were thecornerstone of medical therapy forIH. Shrinkage of IH with systemiccorticosteroid therapy was firstobserved serendipitously amongpatients withhemangioendotheliomas (KMP)treated for thrombocytopenia in thelate 1950s and early 1960s.152 In1967, Zarem and Edgerton153 treated7 consecutive children with oralprednisolone for enlarging IHs. All 7experienced cessation of lesionalgrowth and no rebound growth aftertreatment. On the basis of this result,Fost and Esterly138 treated 6 childrenwith oral prednisone for extensiveIHs. All but 1 had dramatic regressionafter only 2 weeks of therapy.Subsequent studies have continued toshow efficacy, although physicianshave raised concerns about potential


• Management of ulcerated IHconsists primarily of thefollowing:(1) barrier dressings,

(2) pain control, and

(3) control of IH growth.

• Adjuvant therapies may in-clude the following:1. topical agents, including

antibiotics, anesthetics, orwound dressings, and

2. pulsed dye laser.

TABLE 4 Treatment Options in the Management of Ulcerated IH

Wound Care Adjuvant Therapies

Dressings AntimicrobialsWhite petrolatum–impregnated gauze Metronidazole gelNonadherent dressings (eg, Mepitel [Mölnlycke

Health Care; Gothenburg, Sweden], Telfa[Covidien/Medtronic; Minneapolis, MN])

Mupirocin, gentamicin, bacitracin ointmentPain control

TopicalHydrocolloid dressings (eg, DuoDERM

[ConvaTec; Luxembourg])Anesthetics (eg, lidocaine, benzocaine)

OralTopical agents Acetaminophen with or without narcoticsWhite petrolatum, Aquaphor [Beiersdorf Inc.;

Hamburg, Germany], Silversulfadiazine (Silvadene; MonarchPharmaceuticals; Bristol, TN)

OtherBecaplermin gelTopical timololPDLEarly excisionOral propranolol or steroids




adverse effects of steroidsadministered in large doses for longperiods of time.

In the late 1980s and early 1990s,interferon-a showed some promise inthe treatment of steroid-resistantIHs. This drug is a cytokineproduced by leukocytes that playa role in the innate immuneresponse against viruses. Whensynthetic interferon-a2a was used totreat patients with HIV, animprovement in their Kaposisarcoma lesions was noted.154

Because interferon-induced genesare upregulated during involution ofIH, there was a theoretical basis forits mechanism of action in IH,152 andanecdotal reports as well as clinicaltrials subsequently documented itsefficacy in vascular lesions, includingIHs refractory to corticosteroidtherapy.156,157 However, it is nowclear that significant neurologictoxicities, including impairment ofhigher cortical and motor function,can occur and generally preclude itsuse as a first-line therapy.

In 2008, Léauté-Labrèze et al158

reported their serendipitousobservation that oral propranolol,a nonselective blocker ofb-adrenergic receptors used fordecades to treat cardiac disorders inchildren, is effective and welltolerated in the management of IH.A year later, they reported theirexperience with 32 infants withsevere IH who were treated withpropranolol at 2 to 3 mg/kg per dayin 2 to 3 divided doses.159 Theseinfants responded well with a rapid,consistent, therapeutic effect andminimal adverse effects. Since thattime, there have been numerousadditional reports of the safe andeffective use of propranolol for thetreatment of medically complex aswell as cosmetically significantIHs.160

b-Adrenergic Blockers

For most clinicians treatingcomplicated IHs, propranolol has

become the first-line medical therapy;however, optimal dosing, treatmenttiming and duration, and risk ofcomplications have not yet beenestablished in randomized trials, andrecommendations for monitoring arestill evolving.161 An oral formulationfree of alcohol, sugar, and parabendeveloped for use in children(Hemangeol; Pierre Fabre, Castres,France) received approval from theUS Food and Drug Administration inMarch 2014. There is a paucity ofdata regarding other b-adrenergicblockers.162

Mode of Action in IH

The mode of action of propranolol inthe treatment of IH is unknown.Proposed mechanisms includevasoconstriction, inhibition ofangiogenesis (via suppression ofVEGF-A and downregulation of MMPsand interleukin [IL] 6), regulation ofthe renin-angiotensin system, andinhibition of nitric oxide production;propranolol’s ability to stimulateapoptosis isequivocal.44,49,140,141,159,161,163–167

Investigators have shown the presenceof b2-adrenergic receptors oncapillary endothelial cells inproliferating IH, and vascularendothelial cell growth factors, whichare elevated in rapidly growing IHs,are suppressed in the presence ofb-adrenergic receptorblockade.158–160,168 It has also beensuggested that propranolol mayprevent the differentiation of IH stemcells into endothelial cells orpericytes,169 reduce contractility ofpericytes,170 and/or promoteadipogenesis.140,165,171

Efficacy

In a randomized controlled trial oforal propranolol in 460 infants aged 1to 5 months with IH, patientsadministered a dose of 3.4 mg/kg perday exhibited a 60% rate ofsuccessful treatment (complete ornearly complete resolution of the

target hemangioma), compared witha 4% rate among those treated withplacebo.172 Another randomized trialin 40 patients found a markedimprovement in IH volume, redness,and elevation among those takingpropranolol compared with thosetaking placebo.173 In a 2011comprehensive review of theliterature, response to propranololtherapy was evaluated in 79 articlesbut quantified in only 6.160 Positiveresponses in all treated patientswere reported in 86% ofpublications; the remaining 14%discussed at least some treatmentfailures. In total, 19 of 1175 patientsin these publications were reportedas treatment failures, suggestinga 1.6% treatment failure rate.160

Moreover, lightening of the color andsoftening of the tumor was noted inmost children within hours to days ofthe initial dose ofpropranolol.158,159,173,174 Afterinitiation of propranolol therapy,progressive improvement has beennoted for at least 3 months in mostpatients.158,159,173,174 Like systemiccorticosteroids, propranolol appearsto stabilize IHs in their growth phase;however, it may also be effective afterproliferation has ended. In 1 studythat assigned visual analog scores toIHs in children treated withpropranolol at ages 7 to 120 months,more than half of the patientsachieved a greater rate ofimprovement in their scores afterstarting the medication.143 Otherauthors have also reported similarobservations.142,175

Pretreatment Assessment,Contraindications, and Risks of Therapy

A complete history and physicalexamination, with special attention tothe cardiac and pulmonary systems,aid in assessing a child’s candidacyfor propranolol initiation.Electrocardiography is often orderedas well, particularly in youngerinfants, those with a low heart rate,and those with an examination or



family history consistent withcongenital heart disease.161 Someclinicians also prefer to havea cardiology consultation beforestarting the medication. However,pretreatment cardiac screeningappears to be of limited value inpatients with an unremarkablecardiac history andexamination.176–179 Relativecontraindications to the use ofpropranolol for IH includecardiogenic shock, sinus bradycardia,hypotension, heart block greater thanthe first degree, heart failure,bronchial asthma, and knownhypersensitivity to the drug(Table 5).161 Special precautions havebeen suggested for childrendiagnosed with PHACE syndrome andsignificant intracranial vascularanomalies because of the theoreticallyincreased risk of acute ischemicstroke.161

Experience in the management ofhundreds of infants with IH hasshown propranolol to have anexcellent safety profile and hightolerability. The most commonlyreported adverse effects ofpropranolol are sleep disturbanceand coolness and mottling of thedistal extremities. The use ofb-blockers can be also be associatedwith adverse cardiac effects,including bradycardia andhypotension, both of which are

generally asymptomatic and do notrequire intervention.177 Less commoncomplications include bronchospasmand hypoglycemia, the latter of whichhas the potential to induceseizures.159,161,180,181 In a systematicreview of propranolol treatment ofIH, there were 371 total adverseeffects reported in 1189 patients.182

Those most commonly reported weresleep disturbance (136 patients),acrocyanosis (61 patients),hypotension (39 patients, including 5considered “symptomatic”),bradycardia (8 patients, including 1considered symptomatic), andrespiratory events includinginfections, wheezing, and stridor (35patients).

Initiation of Therapy and Dosing

Although the optimal setting for theinitiation of propranolol has yet tobe established, a consensus grouphas suggested that inpatienthospitalization be considered forinfants 8 weeks of age or younger,preterm infants less than 48 weeks’postconceptional age, those withpoor social support, and those withcardiac or pulmonary riskfactors.161 The grouprecommended initiating therapy ata dose of 1 mg/kg per day, withescalation to a target dose of 1 to3 mg/kg per day, although thisrecommendation was made prior tothe FDA approval of Hemangeol,which is dosed maximally at 3.4 mg/kgper day. The optimal dose formaintenance has yet to beestablished. The group’srecommended dosing frequency was3 times daily; however, the drug hasalso been dosed twice daily andshowed both safety andefficacy.172,183,184 Because the peakeffect of oral propranolol on heartrate and blood pressure is 1 to 3hours after administration, thegroup suggested that thesemeasurements be taken atbaseline, 1 and 2 hours after thefirst dose, and 1 and 2 hours after

each dosage increase of $0.5 mg/kgper day. Heart rates or bloodpressure measurements lower than2 SDs from the mean suggest theneed for cardiologic evaluation. Itshould be noted that, where therewas controversy, the grouprecommended the mostconservative approach topropranolol initiation.

The risk of hypoglycemia may bereduced by administeringpropranolol and feeding childrenat intervals not to exceed 8 hours(or 6 hours in younger infants).161

Children with any acute illness,especially one interfering withnormal oral intake or oneassociated with vomiting ordiarrhea, will require closemonitoring and often a temporarydecrease in dosing or cessation oftherapy.

Duration of Therapy

The most dramatic improvementusing propranolol for IH occurswithin 3 to 4 months of initiation oftherapy. However, manyinvestigators continue therapy untilpatients reach an age when IH wouldnormally begin to regress withouttreatment. Hence, treatment is oftencontinued until at least 8 to 12months of age, which, in moststudies, equated to 3 to 12 months oftherapy.160,172 For discontinuation oftherapy, most practitioners taperpropranolol gradually over a periodof 1 to 3 weeks, primarily in aneffort to prevent rebound sinustachycardia. Rebound growth ofIH has been observed in 6% to25% of children, often well aftertheir first birthday, leading someclinicians to wean propranolol overweeks to months.185–189 Reboundgrowth may be more likely inpatients whose IH exhibited a longproliferative stage and a largesubcutaneous component.185 In suchcases, reinitiation of therapy forvariable periods of time may benecessary.

TABLE 5 Contraindications and PotentialComplications Associated WithPropranolol Therapy

Contraindications Complications

Sinus bradycardia Sinus bradycardiaHypotension HypotensionGreater than

first-degreeheart block

Diarrhea

Heart failure

Cool extremitiesSleep disturbanceReactive airways

Cardiogenic shock Hypoglycemia/seizuresReactive airwaysHypoglycemiaHypersensitivity

to propranololhydrochloride




Topical b-Adrenergic Blockers

Several investigators havereported success using topicalb-blockers in the treatment of IH.Timolol maleate is a nonselectiveb-adrenergic receptor inhibitoravailable in a concentration of0.25% and 0.5%, which has beenused by pediatricophthalmologists in the UnitedStates for more than 30 years asa first-line therapy in children withglaucoma. In recent years, anextended-release gel-formingsolution has become available inconcentrations of 0.25% and 0.5%.Systemic absorption of the gel-forming solution is significantlylower than that of the solution, andabsorption through intact skin islikely much less than that throughthe conjunctivae and lacrimalduct.190

Case reports and case series haveshown a good response of IH totwice-daily topical application oftimolol.191–195 In a randomizedcontrolled trial, timolol was moreeffective than placebo in reducingthe size and color intensity of smallsuperficial IHs.196 Laboratorystudies were not monitored in themajority of studies, and only 1infant in 1 large series developeda transient sleep disturbance.Responses were best in patientswho had superficial IHs, used the0.5% gel-forming solution, andapplied the medication for morethan 3 months.195 Manyexperts now consider topicaltimolol gel-forming solutiona reasonable consideration foruncomplicated, superficial IHs forwhich treatment is desired but therisk-to-benefit ratio is too great tojustify systemic b-blocker therapy.However, there are valid concernsregarding the bioavailability of thedrug when used topically inneonates and infants, especially inthe treatment of larger or ulceratedlesions and those on or nearmucous membranes.190

Corticosteroid Therapy

The precise mechanism of action ofglucocorticoids in the treatment of IHsremains largely unknown. Evidencesuggests that corticosteroid therapyhas several effects on IH, involvingboth vasculogenesis and adipogenesis.

The diversity of these effects mayaccount for the variability in response,particularly with the stage of thetreated lesion. Steroids inhibitneovessel growth in cultured humanIH biopsies197 and IL-6–mediatedneovascularization in a rat cornealmodel.198 Corticosteroids also inhibitthe expression of proangiogenicproteins, including VEGF-A, urokinaseplasminogen activator receptor,monocyte chemoattractant protein-1,IL-6, and MMP-1, from human IH stemcells in a murine model.40,197 Inaddition, glucocorticoids inhibit theantiadipocytic differentiation effect ofpreadipocyte factor 1199,200 andpromote adipogenesis by increasingthe expression of peroxisomeproliferator activated receptor.201,202

This activity is thought to explain thedevelopment of the fibrofattyresiduum during involution of thevascular components of IH.

Systemic Corticosteroids

Systemic therapy with corticosteroidsfor large and complicated IHs has, inmany centers, been supplanted bysystemic b-blockers. Nevertheless,steroids have played a significant rolein IH management over the past fewdecades, and properly dosed andmonitored, they remain an effectivemodality in the management of IH,especially in patients in whomb-blocker therapy is risky orcontraindicated.203 One report in 60children with IHs treated with either 3or 5 mg/kg per day of oral prednisonefound an excellent response in 68%and a good response in 25%; therapyfailed in 7%.204 A systematic literaturereview showed an 84% response rateat an average dose of 2.9 mg/kg perday of oral prednisone.205 Anotherrecent article reported the response tosystemic corticosteroids wassignificant in 30% to 53% of cases,equivocal in 35% to 40%, andnegligible in the remainder.206 Ina prospective, randomized,investigator-blinded trial comparingprednisolone and propranolol dosedat 2.0 mg/kg per day, the drugs


• Propranolol, administeredorally at a dose of 1 to3.4 mg/kg per day, is effica-cious in reducing the sizeand color intensity of IH.

• The mechanism of propran-olol’s effect on IH likelyinvolves several processes,including vasoconstriction,inhibition of angiogenesis,and stimulation ofapoptosis.

• Common side effects of pro-pranolol include sleep dis-turbance and discolorationwith cooling of the handsand feet.

• Contraindications to the useof propranolol for IH includecardiogenic shock, sinusbradycardia, hypotension,heart block greater thanfirst-degree, heart failure,bronchial asthma, andknown hypersensitivity tothe drug.

• A consensus report suggeststhat heart rate and bloodpressure be determined atbaseline, 1 and 2 hours afterthe first dose of propranolol,and 1 and 2 hours aftereach dosage increaseof $0.5 mg/kg per day.

• Administration of pro-pranolol with feedings, andholding doses if oral intakeis compromised, reduces thelikelihood of hypoglycemia.

• Topical application of timololhas shown efficacy in themanagement of superficial IHs.



showed similar efficacy for reducingthe area of symptomatic IH; however,although prednisolone showeda somewhat faster response rate,propranolol was better tolerated withsignificantly fewer severe adverseeffects.207 Rebound growth occurs in14% to 37% during dose tapering,occasionally requiring the resumptionof steroid therapy.205 This wide rangein rebound rates likely reflects thevaried duration of corticosteroidtherapy reported in the literature.

Optimal dosing of systemiccorticosteroids remains somewhatcontroversial. However, althoughrecommendations for prednisolonedosing have ranged from 2 to 5 mg/kg per day,132,203,204,208,209 optimaldosing appears to be 2 to 3 mg/kgper day. The duration of therapydepends on response rate as well asthe age of the patient and phase of IHgrowth but generally ranges from 4 to12 weeks at full dose, followed bytapering over several months andcompletion of treatment by 9 to12 months of age.132,205

Intralesional Corticosteroids

The effectiveness of intralesionalcorticosteroid therapy forproblematic IHs was first described in1967 by Zarem and Edgerton,153 inthe same article in which theyreported their success treating IHwith oral corticosteroids.Subsequently, numerous studies havesuggested that intralesionalcorticosteroid injection is a safe andeffective treatment of IH.210–219

In general, corticosteroid injection isreserved for small, bulky, well-localized IH lesions. Large or diffuseIHs are more difficult to manage withintralesional corticosteroids becauseof the following: (1) a large volume ofinjectable steroid is more likely tocause systemic adverse effects220 and(2) it is difficult to evenly distributethe corticosteroid throughout a largetumor. In lesions that are relatively flator superficial, intralesional steroidinjection carries an increased risk of

localized complications involving theoverlying skin or underlying tissues.However, in appropriately selectedlesions, many authors considerintralesional corticosteroid injectionan effective intervention, given itseffectiveness and the relativelylow frequency of reported systemicadverse effects at low doses (#2–3mg/kg).210–219

In most studies, patients were injectedwith either triamcinolone alone ora mixture of triamcinolone andbetamethasone, at total equivalentdoses of triamcinolone doses of,3 mg/kg, by using a 27- or 30-gaugeneedle.218 The interval betweeninjections varied from 1 to 6 weeks.208

After corticosteroid injection, largestudies have reported acceleratedregression in 77% to 100% of patientswith IH and cessation of growth in16% to 23%.210–219 The effects of thesteroid last approximately 3 to 4weeks216 and thus patients mayrequire additional treatments duringthe proliferating phase for reboundgrowth.

Local complications of intralesionalcorticosteroids include fat and/ordermal atrophy and/orhypopigmentation (0%–3%).213–215

Systemic adverse effects, includingcushingoid features (0%–3%)213–217

and adrenal suppression,220 canoccur when very large doses ofintralesional steroids are given ($5mg/kg). A more serious complicationof intralesional corticosteroid therapy

occurs in lesions of the upper eyelid,with 3 cases of retinal embolizationhaving been reported after aninjection of corticosteroids into IHs inthis region.221–223 This complicationlikely results from a combination ofhigh injection pressures (causingretrograde flow of the drug from theeyelid toward the apex of the orbit)and excessive injection volume.224

However, in several large series ofintralesional corticosteroids forperiorbital lesions, this complicationwas not reported. Avoidance of thiscomplication is discussed further inthe subsection “Eye and Orbit” under“IHs With Special AnatomicConcerns.”

Topical Corticosteroids

The use of high-potency topicalcorticosteroids in IH is usually limitedto thin, superficial lesions. In theinitial reports in the 1990s, topicalclobetasol propionate was used forperiocular IHs with good efficacy andno significant adverse effects.225,226 Asubsequent retrospective chartreview of 34 infants withproliferating IHs who had beentreated with high-potency topicalsteroids found that 35% of the infantshad good response, whereas 38% hada partial response.227 A more recentcomparison of topical mometasonefuroate versus intralesionaltriamcinolone acetonide in superficialIHs less than 5 cm in diametershowed that 86.5% (50% excellent,36.5% good) of patients in the topicalgroup and 95.7% (63.8% excellent,31.9% good) in the intralesionalgroup responded to the therapy.228

Adverse Effects of CorticosteroidTherapy

Potential systemic adverse effects ofcorticosteroids used in the treatmentof IH are presented in Table 6 and arethe most common reason cited forusing propranolol as first-linetherapy. It should be noted, however,that a few physicians have favored thesafety profile of corticosteroids overthat of propranolol.229

TABLE 6 Potential Adverse Effects ofSystemic Corticosteroids

HPA axis suppressionCushingoid featuresGrowth decelerationWeight gain/increased appetiteHypertensionGastric irritationIrritabilityInsomniaImmune suppressionCardiomyopathySteroid myopathyOsteopeniaOcular adverse effects (glaucoma, cataracts)

HPA, hypothalamic-pituitary-adrenal.



Suppression of the hypothalamic-pituitary-adrenal axis has beenobserved during therapy with bothintralesional215–217,220,230,231 andsystemic132,232–236 steroids.However, incidence estimates forhypothalamic-pituitary-adrenal axissuppression vary widely, from1.7%222 to 87%.234 Furthermore,although many patients experienceabnormal morning cortisol levels,nearly all appear to normalize ina few months.220,231

Temporary growth deceleration hasalso been reported with bothintralesional218,220,232 andsystemic132,236,237 steroid therapy ofIHs. Almost all children experience“catch up” growth after completion oftherapy.236 Gastric irritation is seen in21%236 to 32%234 of patients takingoral corticosteroids. This adverseeffect can be ameliorated byconcomitant use of H2-receptorantagonists.132,215,218 Mild behavioralchanges have been seen in up to 29%of infants receiving systemiccorticosteroid for IH therapy.236 Theseinclude irritability, fussiness, insomnia,and personality changes.230,234,236,237

Osteopenia is a known adverse effectof long-term systemic corticosteroidtherapy but is rarely observed inchildren with IH, which presumably isrelated to the relatively short durationand nonrepetitive nature of therapytypically used for the treatment of IHs.Hypertension is also a risk of systemiccorticosteroid therapy, but thepercentage of affected individuals isunknown and is likely dose dependent.

Immunosuppressive effects ofsystemic corticosteroid therapy arewell known. These include increasedinfection risk, reduced B- andT-lymphocyte counts, and poorresponse to vaccines.132,218,236,237

Rare cases of pneumonia attributableto Pneumocystis carinii infection havebeen reported in infants takingcorticosteroids for IH, andprophylaxis of these patients withtrimethoprim-sulfamethoxazole hasbeen advocated by some

experts.238–240 Furthermore, it hasbeen suggested that infants notreceive live vaccines during long-termcorticosteroid therapy and thatclinicians consider checking vaccinetiters on completion of corticosteroidtherapy to assess the adequacy ofresponse.237

Ocular adverse effects of long-termsystemic corticosteroid therapyinclude cataracts241,242 and increasedintraocular pressure,242–245 althoughneither has been frequently reportedamong children in general or amongthose being treated for IH inparticular. The most serious ocularadverse effect is that of vision losscaused by embolic occlusion of thecentral retinal artery afterintralesional injection.221–223

However, the actual risk is thought tobe quite low and is primarily relatedto high injection pressure.224,246,247

This complication is discussed ingreater detail in the subsection “Eyeand Orbit” under “IHs With SpecialAnatomic Concerns.”

Cutaneous adverse effects of steroidsare most often associated withintralesional and topical therapy. Themost common risks are atrophy andhypopigmentation, although theformer is often attributable to the IHitself, whereas the latter is usuallytransient.227,233 Other potential butunusual risks include acne,periorificial dermatitis, striaedistensae, and hypertrichosis. Intreating bulky IHs with intralesionaltherapy, cutaneous complications canbe avoided by keeping the injectionwell below the dermis.

Given the many potential adverseeffects of glucocorticoid therapy, manyphysicians will periodically reevaluatethose infants receiving systemiccorticosteroids (oral or intralesional),with specific attention to growthvariables and blood pressure. Somephysicians will also reassess adrenalfunction at the end of therapy anddetermine the need for stress doses ofsteroids on cessation of therapy.

Other Medical Therapies

Before the discovery of the therapeuticefficacy of propranolol for IH, severalother agents were used in an attemptto optimize efficacy and safety. Thissection will focus on 3 agents that havedocumented utility in the treatment ofIH: vincristine, interferon-a, andimiquimod. Unfortunately, the adverseeffect profiles of these agents limittheir usefulness, and they are generallyreserved as treatments only forrecalcitrant lesions. In addition, thepotential usefulness of newerangiogenesis inhibitors will bediscussed.


• Despite their efficacy, sys-temic corticosteroids are nolonger considered by mostclinicians to be first-linetherapy for IH due to theassociated risk of adverseeffects.

• Corticosteroids, adminis-tered orally at a dose of 2 to3 mg/kg per day, are effica-cious in reducing size anddiscoloration of IH.

• The mechanism of IH growthinhibition by corticosteroidslikely involves reduced vas-culogenesis and enhancedadipogenesis.

• Corticosteroids adminis-tered intralesionally andtopically also appear to beeffective in certain subsetsof patients with more local-ized IH, but their dosing andsafety profile are not wellstudied.

• Periodic reexamination ofchildren receiving cortico-steroid therapy for IH hasbeen suggested for monitor-ing of growth and bloodpressure as well as changesin the lesion(s) beingtreated.



Vincristine

Vincristine is a plant-derived vincaalkaloid that impairs mitosis viamicrotubule formation. It hastraditionally been used asa chemotherapeutic agent andpossesses multiple antiangiogenicqualities. It induces endothelial cellapoptosis and is also a potentinhibitor of endothelial cell growth,migration, and in vitro capillary-liketube formation.248 Given thatendothelial cells also possess a hightubulin content, a biological rationalefor IH sensitivity to vincristineexists.249 Most reports on the efficacyof vincristine address the treatmentof patients with vascular lesions thatwere not true IHs but rather KHE orTAs associated with KMP.250

However, vincristine has also beenused successfully in the managementof function-threatening or life-threatening IHs (airway, orbital, orhepatic).251,252 The drug isadministered weekly througha central catheter because of itsextreme vesicant and irritativepotential. Adverse effects includeirritation, neurotoxicity, loss of deeptendon reflexes, constipation, cranialnerve palsies, and bone pain.Alopecia, rash, and myelosuppressionare also possible. Reported adverseeffects were transient.253 This drugappears to be particularly useful inpatients with corticosteroid-resistantKMP, but it is not a first-line therapyfor IH.

Interferon-a

Interferon-a 2a and 2b have bothbeen used successfully for IH inchildren.156,157 Interferon-a is givensubcutaneously with an initial dose of1 million IU/m2, increasing to 3million U daily over the first month oftherapy while monitoring neurologicstatus, white blood cell count, andliver function status.206 Most patientshave required between 2 and12 months of therapy. Adverse effectsare significant and include flulikereactions, rash, gastrointestinalsymptoms, transaminitis,

neutropenia, and spastic diplegia.Although some have reportedresponse rates of up to 90% insteroid-resistant lesions, the effect isgradual in onset, and rebound canoccur on discontinuation.254

Up to 20% of children treated withinterferon-a appear to developspastic diplegia.255 This complicationtends to occur later in the treatmentcourse and may be irreversible.256

Some practitioners initially theorizedthat only interferon-a 2a, or perhapsthe preservative or vehicle, were thecause of these symptoms; however,similar toxicities have also beenreported with interferon-a 2b.257

Given these concerns, interferon-a isgenerally considered a “last resort”treatment, and most physicians preferto use propranolol, systemiccorticosteroids, or vincristine beforetreating with this agent.

Imiquimod (Imidazoquinoline 5%)

This topical immune-responsemodifier stimulates the innateimmune system by augmenting theproduction of cytokines, includinginterferons (a, b, and g); IL-10, IL-12,and IL-18; and tumor necrosis factor.These agents enhance cell-mediatedimmunity and induce apoptosis.However, it may well be thatimiquimod’s therapeutic effect on IHresults from the inhibition ofangiogenesis by these cytokines. Inaddition, imiquimod downregulatesproangiogenic factors such as bFGFand MMP-9 and upregulates otherendogenous angiogenesis inhibitors,including interferon-inducible protein10, tissue inhibitor of MMPs, andthrombospondins.258 Topicalapplication of imiquimod has beenshown to markedly inhibit tumorcell–induced angiogenesis in a humankeratinocyte model.259,260

In 2002, the successful use ofimiquimod was reported in thetreatment of scalp IHs.261

Subsequently, a retrospective studyreported the efficacy of topicalimiquimod in 18 children with IH.262

The drug was used 3 times weekly in10 patients and 5 times weekly in 8patients, with a mean duration oftherapy of 17 weeks. All superficialIHs improved, but little or no changeoccurred in mixed and deep IHs.Irritation and crusting were themost common adverse effects. Thisstudy was criticized because of thelack of a control group.263

A subsequent phase II, open-labelstudy followed 16 children withmixed results.264

Although some proponents ofimiquimod continue to use it for thetreatment of superficial IH, irritation,crusting, and occasionally significantulceration noted with treatmentseriously limit its utility. Imiquimoddoes not have a role in the treatmentof deep IHs.

Antiangiogenic Agents

Although all of the above treatmentshave antiangiogenic effects, a fewnewer therapies have specificantiangiogenic mechanisms of actionthat make them theoretically attractivetherapeutic options. Antagonists tovascular growth factors or receptors,such as inhibitors of VEGF or bFGF,have been successfully used to bluntangiogenesis in tumor models and astherapies for advanced neoplasms;however, the simultaneousadministration of cytotoxic agentsappears to be required to achievesignificant response rates.265 Anincidental decrease in the size ofa liver IH was noted in a patienttreated with bevacizumab, a VEGFinhibitor.266 Most recently, theantivasculogenic effect of rapamycinhas been shown in a mousehemangioma model; the drugdiminished the self-renewal capacityof the IH stem cells whilesimultaneously inducing otherantiangiogenic effects on IHendothelial cells.267 Rapamycin isa macrolide known to have bothimmunosuppressant andantiangiogenic actions, and thereforethe risk-benefit ratio of this agent willneed to be clearly established before it



can be considered a safe alternative tothe agents currently available for thetreatment of IH.

LASER THERAPY FOR IH

Before the discovery of the efficacyof propranolol in the treatment of IH,PDL therapy was frequentlya component of the treatmentstrategy for IHs.268–277 However,given their limited depth ofpenetration of less than 2 mm, theselasers proved useful primarily forsuperficial lesions and for deep andcompound lesions in which salvageof the superficial skin was desired.Although many such lesions are nowtreated medically, laser therapy maystill have a role in IH management,particularly when used as a part ofmultimodal therapy or in ulceratinglesions refractory to other therapies.

In the 1980s, technological advancesimparted to lasers the capability ofselective photothermolysis, a processby which blood vessels are selectivelydestroyed while causing minimalcollateral damage to surroundingtissue.278 For laser light to beoptimized for this purpose, it neededto be of an appropriate wavelength

for absorption by hemoglobin and ofan appropriate exposure time toavoid the generation of excessthermal energy in adjacent tissues.As indicated in these reports, PDLachieves these parameters and hasbecome the “workhorse” laser in thetreatment of superficial vascularlesions.

PDL was developed primarily for thetreatment of port wine stains. IHsdiffer greatly in many respects.Although port wine stains aremalformations made up of low-flow,thin-walled ectatic postcapillaryvenules, IHs are tumors made up ofsmall capillaries lined with plumpendothelial cells and have a higherrate of blood flow. IHs are also ofmore varied thickness. As a result,improvement in IH with the use ofPDL is somewhat less predictablethan that for port wine stains. Themechanism for laser destruction ofIHs has not been completelyelucidated.

PDL has undergone multipleimprovements since it was firstintroduced. Current models usea wavelength of 595 nm and largerspot sizes (up to 10 mm) with higherfluences, allowing the laser topenetrate deeper.279 Longer pulsedurations facilitate the treatment oflarger vessels.279 In addition, theintroduction of dynamic coolingdelivered to the skin before the laserpulse has made treatment safer andless painful.

In the 1990s, many laser proponentsembraced the theory that early lasertreatment could eradicate superficialproliferating lesions and, at the veryleast, reduce the discoloration of thesuperficial component of a mixed IH.As a result, children often receivedmultiple laser treatments during theirfirst year of life. However, a 2002study in 120 children randomlyassigned to laser treatment orobservation found that the completeclearance or presence of minimumresidual IH at 1 year was notsignificantly different in the PDL-

treated and observation groups.280

Although these findings have beendisputed,281 another article authoredby leading laser authorities suggestedthat infants were particularlysusceptible to complications fromPDL treatment.151 The controversyover laser use was recently rekindledby a study that showed an advantageto early laser management.282

Although small IHs may be treatedwithout sedation, children with largerlesions often require generalanesthesia. Recent data suggest thatearly exposure to general anesthesiamay have a negative effect onlearning and behavior.283 As a result,repeated anesthetics required to treatsuch lesions may be less desirablewhen the likelihood of improvementis low and other available treatmentoptions are available.

Proposed uses for PDL in IHmanagement include the following:(1) early superficial facial IHs, (2)treatment of compound IHs in whichsacrifice of the overlying skin isundesirable, (3) refractory ulceration,and (4) significant residualtelangiectasia or flat IH persistingafter involution.

Early Superficial IHs

Although not all superficial lesionswarrant intervention, early treatmentin cosmetically critical areas canreduce or eradicate a superficialdermal IH, allowing the return ofnormal dermis and preventing theatrophic scarring commonly observedafter involution.268–277 Althoughtopical b-blocker therapy is also anoption in managing small superficialIHs, there may be greaterbioavailability of the drug whentreating infants, particularly thosewith IHs that have ulcerated or arelocated near mucous membranes.189

PDL is a treatment option in suchcases.

Treatment of Critical Skin

In certain anatomic locations, thesacrifice of skin that is atrophic or


• Medications other thanb-blockers and cortico-steroids may have efficacy intreating IH, but their utility islimited by their safety profile.

• Vincristine is used forlesions associated with KMP;however, such lesions areKHEs and TAs rather thanIHs and are associated withpotential risks of irritationand neurotoxicity.

• Interferon-a and imiquimod,although effective in IHtreatment, are associatedwith an undesirable rate ofcomplications.



still contains IH tissue is undesirable.The face, and the nasal tip inparticular, are areas in which theremoval of affected skin may leaveundesirable scars or a poor colormatch with grafted skin. In suchcases, early laser treatment maypreserve the overlying skin, allowingit to be later lifted as a flap toprovide access for excision of thedeeper IH (demonstrated in Fig 17later in report).

Treatment of Ulcerated IH

The potential benefit of, andevidence for, laser treatment ofulcerated IH has previously beendiscussed (see the previous sectionentitled “Management of UlceratedIHs”).96,148–150 In most cases, suchintervention is a consideration onlyafter medical management hasfailed.

Persisting Telangiectasia or IH AfterInvolution

After treatment or involution of IH,superficial vascular ectasiasfrequently remain. These areeffectively treated with PDL.284 PDLmay also be used to treat flat areas ofresidual IH tissue.

Complications of Laser Therapy

Complications of laser treatmentinclude atrophic scarring andhypopigmentation, particularly inindividuals of darker complexion.151

Lasers are also capable of inducingulceration, although this is rare andseen more commonly in rapidlyproliferating IHs and segmental IHs,which, if left untreated, also havea higher risk of ulceratingspontaneously.81,84,151 Scarring isseen when the dermis between thevessels is coagulated.285 In IHs inwhich the dermis is largely replacedby vessels, efficientphotocoagulation of these vesselswill lead to scarring. Although thecomplication rate of PDL use for IHhas not been studied, it is less than1% in the treatment of port winestains.286

SURGICAL THERAPY FOR IH

Timing of Intervention

Elective resection of an IH during theproliferative phase is usually notnecessary and occasionally ill advised.Given their young age and thevascularity of the tumor, affectedpatients are at greater risk of anestheticmorbidity, blood loss, and iatrogenicinjury.287,288 In addition, in manylocations such as the lip and nasal tip,the final cosmetic result is superiorwhen growth of the lesion has ceasedand the number of surgicalinterventions can be kept toa minimum. However, certain factorslower the threshold for resection ofa cosmetically or functionallyproblematic lesion during early infancy,including the following: (1)contraindication to pharmacotherapy,(2) failure of pharmacotherapy toameliorate the problem, (3) focalinvolvement in an anatomicallyfavorable area, and (4) a high likelihoodthat resection will be necessary in thefuture (due to bulk or ulceration) andthe scar will be the same.287–289

During the involution phase of IH,children are reevaluated periodically ifit is likely they might require excisionof residual fibrofatty tissue, resection ofscarred/excess skin, or reconstructionof damaged structures. A residual scarwill always be present after ulcerationof an IH and may therefore lower thethreshold for surgery in some cases.Similarly, those lesions likely to leavelarge fibrofatty residua may beaddressed early. Surgery is rarely anoption for extensive IHs fromwhich theresulting defect is too large to closeprimarily or for those cases in whichsurgery will result in significantfunctional impairment or anunacceptable scar. Conversely,resection of small, complicated IHs incosmetically favorable locations mayoccasionally be preferable to months ofobservation and/or medical therapy.

As discussed earlier, the timing ofintervention is based on the age ofthe patient, the degree of deformity,and whether the tumor is stillregressing. If a child has a minordeformity, postponing interventionuntil maximal involution hasoccurred may obviate the need fora procedure. Following the patientfor as long as possible may also beindicated for lesions in problematiclocations (eg, lip or nasal tip),because maximal involution mayfacilitate reconstruction and reducethe number of requiredinterventions.

Once it is obvious that a child willrequire operative intervention, surgeryis usually preferable in early childhood(#4 years of age), before the child hasmuch awareness of thelesion.287,288,290–292 However, bypostponing intervention until at least 3years of age, the tumor has had time toinvolute, often facilitating the procedureand improving the final cosmeticresult.287,288

Location Considerations

For an IH that causes a deformity thatcannot be easily concealed (eg, on theface), surgical intervention may be


• Laser treatment of IHs maybe useful in early, non-proliferating, superficiallesions; management ofcritical skin; treatment ofulcerating lesions; “multi-modal” therapy; and man-agement of persistingpostinvolutiontelangiectasia.

• Pulsed dye laser (PDL) isused most commonly be-cause its light is preferentiallyabsorbed by hemoglobin.

• Use of laser on proliferatingand superficial IHs may leadto ulceration.

• Atrophic scarring and hypo-pigmentation are also po-tential complications of laseruse in IH.



considered in early childhood toprevent psychosocial morbidity. If thetumor is hidden by clothing and is notbothersome to the child, waiting formaximal involution to occur isacceptable. When consideringresection, it is important to weigh thepostoperative scar after removal ofthe IH against the preoperativeappearance of the lesion. Linear scarsare ideally placed along the relaxedskin tension lines in the anatomicarea where the IH is located tofacilitate the best possible cosmeticoutcome.

Certain anatomic locations presentspecific challenges to the surgeon.IHs of the auricular helix and nasaltip involve skin that does not movewell over the underlying cartilageand is difficult to replace. Lesionsof the lip that extend outsidethe vermilion may requirereestablishment of the vermilion-cutaneous border as well as thenatural labial contours. Surgeryinvolving the eyelids, oralcommissure, and genitalia alsocarries a risk of functionalimpairment from the procedureitself.

Technical Considerations

Because an IH itself acts as a tissueexpander, there is usually adequateskin to allow primary, linear closure ofthe wound; skin grafts and local flapsare rarely needed. Because the tumoris benign, the entire lesion does notneed to be removed; the goal is toimprove the appearance of the childand subtotal excision is oftenperformed. For circular lesionslocated in visible areas, the length ofthe scar and distortion ofsurrounding structures can beminimized by circular excision andpurse-string closure.293 Althoughlenticular excision of such a lesion willresult in a scar 2 to 3 times thediameter of the lesion, purse-stringclosure, followed by second-stagelenticular excision severalmonths later, will leave a scarapproximately the same length as the

diameter of the original IH.293 Thistechnique can also be used to altera vertical forehead scar to ahorizontal orientation. In thescalp, lenticular excision and linearclosure is preferred to circularexcision/purse-string closure becausea long linear scar is camouflagedby hair, whereas a circular scar mayleave an area of visible alopecia.288

Fibrofatty residuum from thedeep component of an IH canbe removed by using suction-assistedlipectomy; similar lesions on thecheek can occasionally be approachedintraorally to avoid a majorcutaneous scar.

IHS WITH SPECIAL ANATOMICCONCERNS

Eye and Orbit

IHs of the orbit, eyelid, andconjunctiva, also known as periocularIHs, have the potential to causea unique set of vision-relatedcomplications because of theiranatomic location. Compression ofthe globe, obstruction of the visualaxis, and extension into theretrobulbar space have the potentialto cause refractive errors, strabismus,and amblyopia. Although evaluationand management follow principlescommon to all IHs, additionalconsiderations in these cases stronglyinfluence clinical decision-making.

The importance of earlyophthalmologic assessment ofpatients with periocular IH cannot beoveremphasized. Ophthalmologicconsultation is often necessary todetermine the urgency ofintervention. In addition, the fulldepth of these lesions within the orbitis often underappreciated on routinephysical examination, as are visualfield cuts and changes in refraction orextraocular motion.

Amblyopia is the most common andmost serious ophthalmiccomplication of periocular IHs. Thisdisorder results when, because ofimproper stimulation of the involvedeye, the portion of the brain servingthat eye does not develop properly.Amblyopia occurs in 43% to 60% of

FIGURE 14IH of the left eye causing visual field cut andastigmatism. Untreated, the lesion could pro-liferate, potentially resulting in deprivationamblyopia.


• Indications for surgery forIH during infancy are limitedto the following:1. failure of, or contraindi-

cation to,pharmacotherapy;

2. focal involvement in anarea anatomically favor-able for resection; and

3. a high likelihood that re-section will ultimately benecessary and the scarwill be the same regard-less of timing.

• During involution, surgerymay be indicated for excisionof residual fibrofatty tissue,resection of scarred/excessskin, and/or reconstructionof damaged structures.

• Timing of surgery is basedon the age of the patient, thelocation and degree of de-formity, and whether thetumor is still regressing.

• Elective surgical interventionfor IH is reasonable after age4 years because, by this age,self-esteem and long-termmemory begin to form andthe tumor has completedmost of its involution.



children with untreated periocularIHs, usually as a result of visualdeprivation or refractiveerrors.294–296

Deprivation amblyopia occurs whena bulky IH, usually in the uppereyelid, completely obstructs visualinput into the involved eye. The lackof input causes a maldevelopment ofvisual pathways and may result inirreversible loss of vision.297

Refractive errors are due toastigmatism or anisometropia.Astigmatism is the production ofa blurred image on the retina due toaltered curvature of the cornea andoccurs in 20% to 46% of patientswith periocular IHs.298 IHs causingthis disorder usually involve theupper lid294 (Fig 14) but may occur inthe lower lid as well. The astigmatismcan be reversed with earlyintervention, preferably before 9months of age. Beyond 13 months ofage, astigmatism typically persistsdespite involution of the IH;295,296

however, some cases of improvementduring involution have also beenreported.299–301 Anisometropia isa difference in refractive errorbetween the eyes that results ina relatively clear retinal image in theeye with the smaller refractive errorand a relatively blurred retinal imagein the eye with the larger refractiveerror. Although children withrefractive errors attributable to IHcan be treated with contralateralpatching regimens, many still developpermanent spectacle dependence; inmany cases, the brain may ignore theblurry image in the involved eye,resulting in amblyopia.

Strabismus, or misalignment of theeyes, occurs in approximately one-third of children with untreatedperiocular IHs.294 Strabismus mayresult from deprivation amblyopia,mechanical obstruction of extraocularmuscle movements, or directextraocular muscle invasion. Medialrectus involvement is most commonand most obvious, producingesotropia. Superior oblique

involvement is common in typicalsuperonasal eyelid and orbit cases,but the strabismus is subtle andrequires forced ductions or othertesting to diagnose.

Permanent eyelid deformity resultsfrom direct invasion, vascular steal, orprolonged pressure of adjacentstructures, including the levatorpalpebrae superioris, tarsus, eyelashfollicles, and lamina papyracea. Thelevator muscle can be salvaged withearly treatment, but prolongedinvasion produces a fatty, atrophicmuscle akin to true congenital ptosis.Tarsus, lash follicles, and the bones ofthe orbit also respond well to earlyintervention because the ongoinganatomic destruction or deformitycan be arrested at a very early stage.Preservation of the tarsus ensureseyelid margin stability, whereasmaintenance of the bony socketprevents enophthalmos and facialasymmetry.

Proptosis is the forward displacementof the globe from an intraorbital IH.Occurring in approximately one-thirdof children with orbital IHs, proptosiscan result in impaired approximationof the eyelids and corneal exposure.Optic neuropathy may also resultfrom compression or stretching of theoptic nerve.294

Patients with PHACE syndrome maypresent with a unique set ofophthalmologic abnormalities,including increased retinalvascularity, microphthalmia, opticnerve hypoplasia, exophthalmos,choroidal hemangiomas, strabismus,colobomas, cataracts, and glaucoma.Twenty percent of affected patientswill have at least 1 of these findings,often on the side contralateral to theIH.118

Alternatives available to treatperiocular IHs mirror generaltreatment options, but specialregional factors may influence thefinal treatment plan. Urgency, lasersafety, and the luxuriant vascularity of

the region all influence the treatmentmethod selected.

As with most IHs, treatment withpropranolol has become the mainstayof systemic therapy for periocularlesions. Numerous case seriessuggest success not only incontrolling the growth and size of thelesion but also in improvement ofastigmatism.302–304 Unfortunately,perceived successes with propranololtherapy may, in some cases, lead todelayed ophthalmologic referral formore subtle sequelae, resulting inirreversible changes and limitingtreatment alternatives, underscoringthe need for early ophthalmologicevaluation.

Before the advent of propranolol,intralesional steroid injection wasa popular intervention for themanagement of bulky periocularIHs.299 With the use of a combinationof triamcinolone and betamethasone,a response within 2 weeks could beanticipated in 60% to 80% ofpatients.305 However, intralesionalsteroids have been associated witha number of complications. Mostfeared is embolism of the centralretinal artery.221–223 Thiscomplication is thought to resultfrom several factors, including highinjection pressures (causingretrograde flow of the drug fromthe eyelid toward the apex of theorbit), excessive injection volumes,and direct intravascularinjection.224,246,247 Although it hasbeen argued that high pressures canbe avoided by using a large-capacitysyringe and small bore cannula,306

studies suggest that even theseprecautions may be insufficient inpreventing embolization.224 Otherreported complications includehypopigmentation, atrophy ofsubcutaneous fat, and full-thicknesseyelid necrosis.307–310 Withimprovements in systemicmedications and surgical techniques,many now believe that there arebetter options for the management ofperiocular IHs.



Topical use of timolol has shownefficacy in the management of

superficial periocular IHs, as well asIHs involving the conjunctiva and

iris.194,311 This drug has replacedother topical therapies such as

imiquimod, which causes significantirritation of the skin, conjunctiva,and cornea, and topical steroids,which carry the risk of glaucomaand cataract formation. Lasertreatment (PDL) can also be veryeffective in treating superficialperiocular IHs but usually requirescorneal protection and generalanesthesia.

Periocular IHs are often diffuse andare therefore difficult to excise.Surgery can also cause hemorrhagethat complicates the surgery and cancause postsurgical changes, such asptosis and ectropion, which may bedifficult to correct. However, earlysurgical removal, in selected lesions,eliminates the risk of amblyopia,decreases amblyopia treatment times,and dramatically improves thechances for extraocular muscle,

eyelid, and orbit preservation.312,313

For periocular IHs that areconsidered for surgical resection,preoperative imaging aids inestablishing the extent of the lesion.Those that are best suited for surgeryare located outside the bony orbit andwell circumscribed andnoninfiltrative on MRI.313

Airway

IHs can involve any part of theairway.314 The exact incidence ofairway IH is unknown; however, overa 4-year period, 33 freestanding USchildren’s hospitals discharged anaverage of 2.5 patients with thisdiagnosis who had undergone anairway procedure.315

Symptoms of airway IH usuallymanifest during the period of rapidgrowth during the first few monthsof life. A total of 80% to 90% ofaffected infants will present withinthe first 6 months of life, witha mean age of 3.6 months atdiagnosis.316,317 Most developbiphasic stridor and barky cough asthe IH enlarges in the subglottis, thenarrowest portion of the pediatricairway (Fig 15). Voice andswallowing are generally normal.Often, the symptoms are mistakenfor those of infectious orinflammatory croup or reactiveairway disease, especially when thesymptoms worsen in the presence ofupper respiratory illness. Asa result, symptoms are often presentfor several weeks before a definitivediagnosis is made.

Approximately half of infants inwhom an airway IH is diagnosedalso will have a cutaneous IH, butonly 1% to 2% of children withcutaneous IHs also have airwayIHs.318,319 Symptomatic airway IHscan be associated with lower facialcutaneous or oral/pharyngeal mucosalIHs in approximately 50% ofcases.106,107

Most infants suspected of having anairway IH on the basis of historicaland clinical presentation undergooperative endoscopy.320 Imaging ofthe head and neck may be useful insome cases to confirm the diagnosis ofIH, to define the extent of the airwaylesion, or to detect any associatedvascular, brain, or chest anomalies thatcould affect treatment.321,322 Becausethese airway lesions havea characteristic clinical and endoscopic

FIGURE 15Airway IH extending from the vocal folds inferiorly into the subglottic space, the narrowest region ofthe pediatric airway. (Photo courtesy of Jonathan Perkins, DO.)


• IHs of the periocular area havethe potential to cause com-pression of the globe, obstruc-tion of the visual axis, andextension into the retrobulbarspace, resulting in refractiveerrors, strabismus, and ambly-opia, leading to vision loss.

• The permanence of ophthal-mic complications due to IHis often related to their se-verity and duration, under-scoring the need for earlyophthalmologic evaluation.



appearance, biopsy of airway IH is notusually necessary, but whenperformed, the histologic appearanceand cellular markers are the same asthose of cutaneous IH.323,324 As incutaneous IH, airway IH can belocalized (focal) or extensive(segmental), and many subglotticlesions show transglottic orparatracheal extension.108,314 Themore extensive the lesion, the moresignificant the airway compromise,and aggressive therapy is usuallynecessary to prevent airwayobstruction.

The need for and type of interventionfor airway IH is determined byseveral factors, including the degreeof airway obstruction, the extent ofextralaryngeal IH, the location of thepatient at the time of diagnosis, theexperience of the treating physician,and the preferences of the parents/caregivers. Because involution is theultimate fate of virtually all IHs,“watchful waiting” is reasonable incases involving minimalsymptomatology. In rare symptomaticcases for which observation is stillpreferred or pediatric airwayexpertise is not readily available,tracheotomy is occasionallynecessary.325 However, in most cases,some alternative intervention for theIH is more desirable.

Before the advent of propranololtherapy, endoscopic laser ablationor surgical excision was oftenrecommended in the managementof airway IH, because the onlymedical therapy was long-termcorticosteroids.317,326–330 However,many clinicians have now reportedsuccess and low complication ratesusing propranolol in the managementof airway IHs.326,331–336 Althoughthe optimal dose and appropriateduration of therapy remain uncertain,most clinicians are dosing the drugas they would for cutaneous IH.337

Some authors have suggestedthat treatment with propranololshould become the standard for

initial management of all airwayIHs.338

Corticosteroids may be helpful inrefractory cases. Initial doses ofprednisolone at 2 to 4 mg/kg per dayare generally necessary to controlgrowth. Maintenance doses ofapproximately 1 to 2 mg/kg per dayare often considered before weaningthe medication, and treatmentduration is based on clinical response.Response rates reported in theliterature vary between 30% and93%, although there is littleconsistency among dosingregimens.339

Intralesional steroids area consideration for patients whoseIHs have necessitated endoscopy orendoscopic resection. Althoughrepeated injections are usuallynecessary as single-modalitytherapy,327 these medications may beeffective adjuvant therapy forpatients whose lesions are beingobserved, treated pharmacologically,or partially resected. Successfulmanagement has been achieved in77% to 87% of cases with the use ofintralesional steroids.317,327

Airway IHs causing focal obstructionmay be addressed surgically bya subtotal endoscopic approach byusing a microscope or telescope orby total excision through an openapproach. Subtotal approachesgenerally use a laser317 or rotary-powered instrument.340 However,subtotal resection carries the risk ofgrowth of the residual lesion duringthe proliferative phase, and lasertreatment carries a 5% to 25% risk ofsubglottic stenosis that is greatestwith deeper resections and in cases ofbilateral or circumferentialdisease.326,341,342

Open surgical excision of a focallyobstructing airway IH became popularin the 1990s, after complications oflaser therapy became increasinglyapparent.329,343–345 The procedure isuseful in cases refractory to medicaltherapy and may be preferable to laser

in patients with bilateral orcircumferential lesions who may be atrisk of postoperative stenosis orrequire tracheotomy. However, opensurgical excision may be more difficultin cases involving significant extensionoutside the larynx, and the proceduremay potentially result in some degreeof dysphonia.

Nose

Nasal IHs deserve special attention forseveral reasons. First, the prominenceand central location of the nose makeit a critical facial feature. As a result,even a small IH of the nose can havea greater effect on appearance thana lesion of the same size locatedelsewhere. In addition, damaged nasaltip skin is exceedingly difficult toexcise or replace without considerablecosmetic consequence. Functionally,a nasal IH may also cause collapse ofthe nasal introitus and intranasalobstruction.

Focal nasal IHs account for 15% to20% of all focal IHs of the face; ofthese, approximately one-thirdinvolve the nasal tip.81 These lesionsappear to have their origin in theintercartilaginous ligament of thelower lateral nasal cartilages.346 Asthe IH grows, it displaces the


• Most patients with IHs of theairway have subglottic in-volvement causing biphasicstridor and barky cough, of-ten mistaken as croup. Voiceand swallowing are gener-ally normal.

• Diagnosis of airway IHs isusually made by endoscopyin the operating room.

• In most cases, IHs of theairway may be managedmedically; in cases of severeobstruction, surgical re-duction or excision may beentertained.



cartilages outward, rotating them inan “open book” fashion (Fig 16). Thenet effect is a bulbous, distorted nasaltip, which, even with complete

involution of the IH, will remaindisfigured. The deformity has beenreferred to as the “Cyrano” nose, afterCyrano de Bergerac, the French

playwright known as much for hisprodigious proboscis as for hisdramatic works.290,347 Segmental IHsinvolving the nose typically affectmore nasal subunits than their focalcounterparts, and they are morelikely to ulcerate, resulting indestruction of critical areas such asthe columella and ala.

Conservative management of nasalIHs is associated with a poor outcomein most cases.129,209,348 As a result,early management of nasal IHs hasbeen advocated to avoidcomplications and improve thelikelihood of a favorableoutcome,290,346,349,350 although theassumed benefit is yet unproven. Theapproach to nasal IHs is oftenmultimodal, varying with the location,the stage, and the depth of theIH.290,350,351

Medical therapy may commence oncethe diagnosis has been made and it isclear that the IH is growing. Theduration of treatment will depend onthe type of lesion and its response.Focal IHs that respond to propranololare generally treated until at least 9to 10 months of age, at which timegrowth of these lesions is likely tocease. Rebound growth may beaddressed by restarting themedication for a month at a time untilthere is no further growth. Becausesegmental IHs may proliferate forlonger periods of time, weaning isoften delayed until 18 months of age;rebound may be addressed withadditional treatment in 3-monthintervals.

Propranolol-treated IHs of the nasaltip are less likely than untreatedlesions to undergo surgery or lasertherapy, but many lesions stillrequire such interventions.352

Extensive skin involvement in focallesions may respond to topicalb-blockers or judicious use of laser(PDL) during proliferation (Fig 17).This treatment will salvage the skinby reducing the number ofintracutaneous vessels and will alsodiminish the risk of venous stasis in

FIGURE 17A, Compound IH of the nasal tip with significant surface involvement. B, Nasal tip after treatmentwith PDL to salvage tip skin before surgical resection.

FIGURE 16Open rhinoplasty approach to nasal tip IH. These lesions (H) originate within intercartilaginousligament of the lower lateral nasal cartilages (arrows), rotating them outward.



the deep component after surgicalresection. It also helps to maintainmore normal collagen and skin afterproliferation has ceased. Similarly,for nasal lesions within segmentalIHs that have failed to respond tomedical therapy, it may be best todelay surgery until the overlying skinhas been adequately treated.350 It isgenerally preferable to delay surgicalintervention for such lesions untilthere is no further proliferation. PDLmay also be used to treat residualinvolvement of the overlying skinafter surgery.

Nasal IHs that fail to respond topropranolol or leave significantresidual tissue after treatment areusually addressed surgically.Although some authors haveadvocated surgery for focal IHs asearly as 10 to 12 months of age,346

most physicians will operate at 1 to3 years of age.353 This approachallows for complete cessation ofgrowth and adequate time forinvolution of small lesions that mayultimately cause no significantdistortion of the tip.

Several publications have dealt withthe surgical approach to nasalIHs.346,347,350,353–357 All of thesepublications predate the treatmentof IH with propranolol, which hasclearly changed the managementparadigm. The major issues insurgery for IHs of the nasal tip relateto adequate access and the ability todispose of the excess skin once thelesion has been removed. In general,smaller lesions may be addressed ina single procedure through anexternal rhinoplasty approach,leaving a scar only on the columellaof the nose.290,350,353,356 Larger IHsmay require external incisions onthe ala, also described as a “modifiedsubunit approach,” which simplifiesexcision and redraping of theskin.346 Surgery is also useful inrestoring the normal position of thelower lateral cartilages and othercomponents of the cartilaginousframework.

Lips

The lips deserve special considerationin the management of IH due to theircritical role in cosmesis and function.Distortion of the lips from IH iscommon, and restoring normal lipcontour is one of the greatestchallenges in reconstructive surgery.Furthermore, the lips (and the lowerlip, in particular) are at increased riskof ulceration,96,98 resulting in painand bleeding in the short term and inincreased scarring and disfigurementin the long term.

During proliferation, the goals ofmanaging IH of the lips are tominimize distortion and to controlulceration (Fig 18). Once ulceration

occurs, some infants will havedifficulty latching on to a breast orbottle nipple without discomfort,occasionally leading to failure tothrive. Furthermore, the vermilion ofthe lip is a unique tissue that cannotbe replaced if permanently damagedby ulceration,358 and reconstructionof the normal lip contours afterulceration can be challenging.Proactive management of IH of the lip,often systemic, is vital if ulceration isto be avoided. However, in the lowerlip, some 30% of IH lesions ultimatelyulcerate.98 Once ulceration hasoccurred, occlusive dressings areimpractical, and therapies such astopical anesthetics and petroleum-based products carry the risk ofaccidental oral ingestion. Occasionally,children with IHs on the lip benefitfrom laser treatment of the ulcer, butworsening of the ulceration isa risk.151 Early surgical resection isa consideration, but only for smallulcers in cosmetically favorableareas. Otherwise, attention is bestdirected to systemic therapies toreduce the likelihood of furtherulceration and of excessive lengtheningof the lip.

Reconstruction of a lip that is scarredand disfigured because of IH is bestperformed only after growth of theIH has definitively ceased, because

FIGURE 18Ulcerated IH of the lower lip has resulted in distortion of the soft tissue and obliteration of thevermilion-cutaneous border.


• Early management of nasaltip IH reduces the likelihoodof poor cosmesis resultingfrom skin excision and/orreplacement and effects onthe underlying cartilage.

• Goals of surgery for nasal tipIHs include complete IH ex-cision, reconstruction of thecartilaginous framework,and judicious skin excisionand redraping.



recurrence is certain to interferewith recontouring. Furthermore,the mucosa of the mandibular oralvestibule is most useful forreconstruction of the vermilion whenit is free of the IH.

Lesions located exclusively on thevermilion can be removed by usinga transverse mucosal incision to hidethe scar at the junction of thevermilion and vestibular mucosa358;lesions traversing both tissues mayrequire a vertical incision. Bulkierlesions that cause lengthening of thelip and those that cross the vermilion-cutaneous border are best addressedby using a wedge excision,359 withsome authors advocating a 2-stageprocedure to improve scarcamouflage.358 In such cases,incisions may be placed along themucocutaneous junction or philtralcolumns. Debulking lip IH whilepreserving vermilion can beperformed through a mucosalincision; however, it is oftenexceedingly difficult to separate IHfrom orbicularis oris muscle.359

Eversion of the lower lip can becorrected by excision of a mucosalstrip, and correction of inversion mayrequire a lyophilized dermal implantor dermal graft.359 Setting the “whiteroll” (ridge at the vermilion-cutaneous border) of the lower lipand restoring normal sublabialconcavity may be particularlychallenging.360

Perineum

Although perineal area IHs occur ina nonconspicuous area, they are highlyprone to ulceration. In case series fromtertiary care dermatology practices,perineal IHs account for approximatelyone-third of all ulcerating lesions, andapproximately 50% of IHs occurring inthis area ultimately ulcerate.96,98

Perineal ulceration can, at leasttransiently, result in significant painduring defecation and urination andduring diaper changes.

Because surgical interventioninvolving structures in this area may

be fraught with complications,management strategies generallyfocus on those topical and systemicmedical remedies previouslydiscussed (see “Management of AcuteIH Ulceration”). In rare cases,colostomy has been performed tofacilitate wound care.361

Liver

The liver is the most common locationfor visceral IH. Research from the pastdecade has contributed greatly to theclassification of hepatic tumors and toclarify their natural history.91,362

Previously viewed homogenously, andoften treated inappropriately, it is nowbecoming clear which patients are atrisk and what treatment options aresensible.

Patients at risk of hepatic and otherextracutaneous IHs are those withmultiple or multifocal cutaneous IHs.Because studies have documentedthat infants with multiple cutaneousIHs are at increased risk of havinghepatic involvement, screening forhepatic lesions with abdominalultrasonography has been suggestedfor infants with 5 or more cutaneousIHs.89,92,93 Other nonhepatic visceralorgan involvement is relatively rare ininfants with multiple cutaneous IHs;however, performing a thoroughreview of systems and physicalexamination on any infant withmultiple cutaneous hemangiomas isadvisable.

Infants with hepatic IHs may remainclinically asymptomatic or canpresent with life-threateningsymptoms of congestive heart

failure, hepatomegaly, or abdominaldistension.84,91,92 Hepatic IHs havebeen characterized as occurring in 3patterns: focal, multifocal, anddiffuse.362 Focal singular lesions areusually detected on antenatalimaging or as an abdominal mass inthe newborn infant. It is now clearthat these lesions are not true IHs;rather, they are the hepaticmanifestation of a RICH, whichexplains why they are fully grown atbirth.362,363 They spontaneouslyinvolute 90% volumetrically by 13months of age, and involution is notlikely to be hastened bypharmacologic agents.362 Diagnosismay be made by ultrasonography onthe basis of the presence ofarteriovenous shunting or on CT orMRI, which reveal hyperintenseperipheral contrast enhancementand central sparing. Some hepaticRICHs have associatedmacrovascular shunts (usuallyhepatic artery to hepatic vein) thatcan cause high-output cardiacfailure. If shunts are absent, a hepaticRICH can be observed with serialultrasonography to determinewhether its behavior is typical.Differential diagnosis can includehepatoblastoma and mesenchymalhamartoma. If imaging andinvolution are not diagnostic,percutaneous biopsy may beindicated. If shunts are present andcausing high-output failure, selectiveembolization can ameliorate cardiacfailure, and the lesion can be allowedto involute. There is a very small riskof rupture and hemorrhage withextremely large tumors. Surgicalresection or transplantation is rarelynecessary.

Multifocal and diffuse hepatic IHsexist on a spectrum. They are trueIHs and often coexist with cutaneouslesions. In a prospective study in 151infants, 16% of infants with 5 ormore cutaneous IHs were found tohave hepatic lesions on screeningultrasonography.364 Multifocallesions have ample normal hepaticparenchyma between them.362 They


• IHs involving the lips andperineum have a tendency toulcerate, and these regionsare difficult to reconstruct.Such lesions are appropri-ately managed aggressivelywith medical therapy.



are often asymptomatic; however,some patients will have macrovascularshunting causing high-flow andpossible high-output cardiac failure.These patients are best treatedpharmacologically with propranolol orcorticosteroids. Shunts will usuallyclose with involution of the IH.Selective shunt embolization isa consideration in those rare instancesin which rapid cardiac failure does notallow sufficient time for response topharmacotherapy.

Diffuse lesions have little hepaticparenchyma apparent betweendensely packed nodular IHsthroughout the liver.362 Patientspresent with hepatomegaly, which canbecome massive. They may developabdominal compartment syndromewith compromised ventilation, renalfailure attributable to renal veincompression, or poor inferior venacaval blood return to the heart andmay progress to death.92 Virtually alldiffuse hepatic IHs cause acquiredhypothyroidism attributable to theinactivation of thyroid hormones bytype 3 iodothyronine deiodinaseconstituent in the lesions.365

Hypothyroidism can be very profoundand can require massive replacementhormone dosing. Multifocal lesionsalso suggest the need for promptthyroid screening, because they maycollectively contain enough tumormass to overwhelm endogenousthyroid production.91,92,365

Infants with diffuse hepatic IH,particularly those with congestiveheart failure, are at greatest risk ofmortality.363 Aggressivepharmacologic therapy and thyroidhormone replacement are indicatedfor infants with such lesions. EffectiveIH treatment will result in a gradualreduction in the requirement forthyroid replacement and eventualreturn to the euthyroid state.366,367

High-flow shunts are uncommon indiffuse lesions. There is no role forembolization in the absence ofmacrovascular shunts and a high-output state. An infant who presents

with massive hepatomegaly andabdominal compartment syndromeoccasionally has disease that cannotwait for drug-induced involution; inrare cases, such a child may bea candidate for hepatictransplantation.

CONCLUSIONS

The management of IHs has evolvedconsiderably in the past decade. Theserendipitous discovery of theresponse of IH to systemic b-blockershas expanded therapeutic optionsfor these tumors. Timolol, a topicalb-blocker, has shown promise asa potential therapy for superficiallesions. A greater understanding ofthe indications and limitations oflaser therapy also has emerged.Concurrently, bench research hasprovided a deeper understanding ofthe origins of IH and patterns of IHgrowth. It is anticipated thatcontinued research will further clarifythe etiology of IH, hopefully leading topathogenesis-directed therapeuticoptions.

Although many IHs can be observedwithout treatment, others will clearlybenefit from medical or surgicalintervention. It is important forpediatricians to keep abreast ofadvances in IH management, becausethe types of intervention and thethreshold for their use are likely toevolve. When complications are likelyor the threshold for interventionis uncertain, referral to anexperienced specialist ora multidisciplinary vascularanomalies center may beadvantageous.

LEAD AUTHORS

David H. Darrow, MD, DDS, FAAPArin K. Greene, MD, FAAPAnthony J. Mancini, MD, FAAPAmy J. Nopper, MD, FAAP

CONTRIBUTING AUTHORS

Richard J. Antaya, MD, FAAP (Dermatology)Bernard Cohen, MD, FAAP (Dermatology)Beth A. Drolet, MD (Dermatology)Aaron Fay, MD (Ophthalmology)Steven J. Fishman, MD, FAAP (PediatricSurgery)Sheila F. Friedlander, MD, FAAP (Dermatology)Fred E. Ghali, MD, FAAP (Dermatology)Kimberly A. Horii, MD, FAAP (Dermatology)Manish N. Patel, DO (Radiology, InterventionalRadiology)Denise W. Metry, MD (Dermatology)Paula E. North, MD (Pediatric Pathology)Teresa M. O, MD (Otolaryngology)


• Hepatic IHs have been char-acterized as occurring in 3patterns: focal, multifocal,and diffuse.

• Focal hepatic IHs are thehepatic manifestation ofRICH; they are fully grown atbirth, and involution is al-most complete by 1 year ofage.

• Multifocal and diffuse he-patic IHs are true IHs andoften coexist with cutaneouslesions.

• Multifocal hepatic IHs havenormal hepatic parenchymabetween them. Manypatients are asymptomatic;however, those with high-flow and/or high-outputcardiac failure require phar-macologic therapy with pro-pranolol or corticosteroids.

• Patients with diffuse hepaticIHs present with hepato-megaly that can lead tocompromised ventilation,renal failure attributable torenal vein compression, poorinferior vena caval bloodreturn to the heart, anddeath.

• Diffuse hepatic IHs maycause acquiredhypothyroidism.

• Most hepatic IHs are man-aged medically; rarely, em-bolization, surgicalresection, and trans-plantation have beennecessary.



Jonathan A. Perkins, DO (Otolaryngology)Michael L. Smith, MD, FAAP (Dermatology)Patricia A. Treadwell, MD, FAAP (Dermatology)Milton Waner, MD (Otolaryngology, Facial PlasticSurgery)Albert C. Yan, MD, FAAP (Dermatology)

SECTION ON DERMATOLOGY EXECUTIVECOMMITTEE, 2014–2015

Bernard A. Cohen, MD, FAAP, ChairpersonRichard J. Antaya, MD, FAAD, FAAPAnna L. Bruckner, MD, FAAPKim Horii, MD, FAAPNanette B. Silverberg, MD, FAAPTeresa S. Wright, MD, FAAPAlbert C. Yan, MD, FAAD, FAAP, Chairperson-ElectMichael L. Smith, MD, FAAP, Ex Officio

FORMER SECTION ON DERMATOLOGYEXECUTIVE COMMITTEE MEMBER

Sheila F. Friedlander, MD, FAAP

STAFF

Lynn Colegrove, MA

SECTION ON OTOLARYNGOLOGY–HEAD ANDNECK SURGERY EXECUTIVE COMMITTEE,2014–2015

Charles Bower, MD, FAAP, ChairpersonChristina Baldassari, MD, FAAPGerman Paul Digoy, MD, FAAPAndrew Hotaling, MD, FAAPStacey Ishman, MD, MPH, FAAPJohn McClay, MD, FAAPDiego Preciado, MD, PhD, FAAPKristina Rosbe, MD, FAAPScott Schoem, MD, FAAP, Past ChairpersonJeffrey Simons, MD, FAAPSteven Sobol, MD, FAAPDavid Walner, MD, FAAP

STAFF

Vivian Thorne

SECTION ON PLASTIC SURGERY EXECUTIVECOMMITTEE, 2014–2015

Peter J. Taub, MD, FAAP, ChairpersonStephen B. Baker, MD, FAAPArin K. Greene, MD, FAAPTimothy W. King, MD, MPH, FAAPDonald R. Mackay, MD, FAAP, Immediate PastChairpersonDelora L. Mount, MD, FAAP, Chairperson-ElectJordon Philip Steinberg, MD, FAAPMark M. Urata, MD, DDS, FAAP,Vice Chairperson

STAFF

Kathleen Kuk Ozmeral

ABBREVIATIONS

bFGF: basic fibroblast growthfactor

CT: computed tomographyEPC: endothelial progenitor cellGLUT1: glucose transporter

protein isoform 1HemSC: multipotential stem cell

derived from IHspecimens

IH: infantile hemangiomaIL: interleukinKHE: kaposiform

hemangioendotheliomaKMP: Kasabach-Merritt

phenomenonLBW: low birth weightLUMBAR: Lower body IH and

other cutaneous defects,Urogenital anomaliesand ulceration,Myelopathy, Bonydeformities, Anorectalmalformations andarterial anomalies andRenal anomalies

MMP: matrix metalloproteasePDL: pulsed-dye laserPHACE: Posterior fossa defects,

Hemangiomas,cerebrovascular Arterialanomalies, Cardiovascularanomalies includingcoarctation of the aorta,and Eye anomalies

MRA: magnetic resonanceangiography

NICH: noninvoluting congenitalhemangioma

RICH: rapidly involuting congenitalhemangioma

TA: tufted angiomaVEGF: vascular endothelial growth

factor

REFERENCES

1. International Society for the Study ofVascular Anomalies. ISSVA classificationof vascular anomalies. 2014. Availableat: issva.org/classification. AccessedApril 2015

2. Enjolras O, Mulliken JB, Boon LM,Wassef M, Kozakewich HP, Burrows PE.

Noninvoluting congenital hemangioma:a rare cutaneous vascular anomaly.Plast Reconstr Surg. 2001;107(7):1647–1654

3. North PE, Waner M, James CA, MizerackiA, Frieden IJ, Mihm MC Jr. Congenitalnonprogressive hemangioma: a distinctclinicopathologic entity unlike infantilehemangioma. Arch Dermatol. 2001;137(12):1607–1620

4. Boon LM, Enjolras O, Mulliken JB.Congenital hemangioma: evidence ofaccelerated involution. J Pediatr. 1996;128(3):329–335

5. Baselga E, Cordisco MR, Garzon M, LeeMT, Alomar A, Blei F. Rapidly involutingcongenital haemangioma associatedwith transient thrombocytopenia andcoagulopathy: a case series. Br JDermatol. 2008;158(6):1363–1370

6. Rangwala S, Wysong A, Tollefson MM,Khuu P, Benjamin LT, Bruckner AL.Rapidly involuting congenitalhemangioma associated with profound,transient thrombocytopenia. PediatrDermatol. 2014;31(3):402–404

7. Berenguer B, Mulliken JB, Enjolras O,et al. Rapidly involuting congenitalhemangioma: clinical andhistopathologic features. Pediatr DevPathol. 2003;6(6):495–510

8. Mulliken JB, Enjolras O. Congenitalhemangiomas and infantilehemangioma: missing links. J Am AcadDermatol. 2004;50(6):875–882

9. North PE, Waner M, Mizeracki A, MihmMC Jr. GLUT1: a newly discoveredimmunohistochemical marker forjuvenile hemangiomas. Hum Pathol.2000;31(1):11–22

10. Patrice SJ, Wiss K, Mulliken JB.Pyogenic granuloma (lobular capillaryhemangioma): a clinicopathologic studyof 178 cases. Pediatr Dermatol. 1991;8(4):267–276

11. Enjolras O, Wassef M, Mazoyer E, et al.Infants with Kasabach-Merritt syndromedo not have “true” hemangiomas.J Pediatr. 1997;130(4):631–640

12. Zukerberg LR, Nickoloff BJ, Weiss SW.Kaposiform hemangioendothelioma ofinfancy and childhood: an aggressiveneoplasm associated with Kasabach-Merritt syndrome andlymphangiomatosis. Am J Surg Pathol.1993;17(4):321–328


http://issva.org/classification.


13. Jones EW, Orkin M. Tufted angioma(angioblastoma): a benign progressiveangioma, not to be confused withKaposi’s sarcoma or low-gradeangiosarcoma. J Am Acad Dermatol.1989;20(2 pt 1):214–225

14. Osio A, Fraitag S, Hadj-Rabia S, BodemerC, de Prost Y, Hamel-Teillac D. Clinicalspectrum of tufted angiomas inchildhood: a report of 13 cases anda review of the literature. ArchDermatol. 2010;146(7):758–763

15. Lyons LL, North PE, Mac-Moune Lai F,Stoler MH, Folpe AL, Weiss SW.Kaposiform hemangioendothelioma:a study of 33 cases emphasizing itspathologic, immunophenotypic, andbiologic uniqueness from juvenilehemangioma. Am J Surg Pathol. 2004;28(5):559–568

16. Enjolras O, Mulliken JB, Wassef M, et al.Residual lesions after Kasabach-Merrittphenomenon in 41 patients. J Am AcadDermatol. 2000;42(2 pt 1):225–235

17. Requena L, Sangueza OP. Cutaneousvascular anomalies. Part I.Hamartomas, malformations, anddilation of preexisting vessels. J AmAcad Dermatol. 1997;37(4):523–549;quiz: 549–552

18. Kanada KN, Merin MR, Munden A,Friedlander SF. A prospective study ofcutaneous findings in newborns in theUnited States: correlation with race,ethnicity, and gestational status usingupdated classification and nomenclature.J Pediatr. 2012;161(2):240–245

19. Kilcline C, Frieden IJ. Infantilehemangiomas: how common are they?A systematic review of the medicalliterature. Pediatr Dermatol. 2008;25(2):168–173

20. Pratt AG. Birthmarks in infants. AMAArch Derm Syphilol. 1953;67(3):302–305

21. Jacobs AH, Walton RG. The incidence ofbirthmarks in the neonate. Pediatrics.1976;58(2):218–222

22. Dickison P, Christou E, Wargon O.A prospective study of infantilehemangiomas with a focus onincidence and risk factors. PediatrDermatol. 2011;28(6):663–669

23. Hoornweg MJ, Smeulders MJ, UbbinkDT, van der Horst CM. The prevalenceand risk factors of infantilehaemangiomas: a case-control study in

the Dutch population. Paediatr PerinatEpidemiol. 2012;26(2):156–162

24. Haggstrom AN, Drolet BA, Baselga E,et al; Hemangioma Investigator Group.Prospective study of infantilehemangiomas: demographic, prenatal,and perinatal characteristics. J Pediatr.2007;150(3):291–294

25. Metry D, Heyer G, Hess C, et al; PHACESyndrome Research Conference.Consensus statement on diagnosticcriteria for PHACE syndrome.Pediatrics. 2009;124(5):1447–1456

26. Chiller KG, Passaro D, Frieden IJ.Hemangiomas of infancy: clinicalcharacteristics, morphologic subtypes,and their relationship to race, ethnicity,and sex. Arch Dermatol. 2002;138(12):1567–1576

27. Liggett SB, Cresci S, Kelly RJ, et al. AGRK5 polymorphism that inhibits beta-adrenergic receptor signaling isprotective in heart failure. Nat Med.2008;14(5):510–517

28. Amir J, Metzker A, Krikler R, Reisner SH.Strawberry hemangioma in preterminfants. Pediatr Dermatol. 1986;3(4):331–332

29. Drolet BA, Swanson EA, Frieden IJ;Hemangioma Investigator Group.Infantile hemangiomas: an emerginghealth issue linked to an increased rateof low birth weight infants. J Pediatr.2008;153(5):712–715; 715.e1

30. Bauland CG, Smit JM, Bartelink LR,Zondervan HA, Spauwen PH.Hemangioma in the newborn: increasedincidence after chorionic villussampling. Prenat Diagn. 2010;30(10):913–917

31. Bauland CG, Smit JM, Scheffers SM,et al. Similar risk for hemangiomasafter amniocentesis andtransabdominal chorionic villussampling. J Obstet Gynaecol Res. 2012;38(2):371–375

32. Colonna V, Resta L, Napoli A, Bonifazi E.Placental hypoxia and neonatalhaemangioma: clinical and histologicalobservations. Br J Dermatol. 2010;162(1):208–209

33. Drolet BA, Frieden IJ. Characteristics ofinfantile hemangiomas as clues topathogenesis: does hypoxia connect thedots? Arch Dermatol. 2010;146(11):1295–1299

34. Walter JW, Blei F, Anderson JL, Orlow SJ,Speer MC, Marchuk DA. Geneticmapping of a novel familial form ofinfantile hemangioma. Am J Med Genet.1999;82(1):77–83

35. Grimmer JF, Williams MS, Pimentel R,et al. Familial clustering ofhemangiomas. Arch Otolaryngol HeadNeck Surg. 2011;137(8):757–760

36. Li J, Chen X, Zhao S, et al. Demographicand clinical characteristics and riskfactors for infantile hemangioma:a Chinese case-control study. ArchDermatol. 2011;147(9):1049–1056

37. Doege C, Pritsch M, Frühwald MC, BauerJ. An association between infantilehaemangiomas and erythropoietintreatment in preterm infants. Arch DisChild Fetal Neonatal Ed. 2012;97(1):F45–F49

38. Greenberger S, Adini I, Boscolo E,Mulliken JB, Bischoff J. Targeting NF-kBin infantile hemangioma-derived stemcells reduces VEGF-A expression.Angiogenesis. 2010;13(4):327–335

39. Yu Y, Flint AF, Mulliken JB, Wu JK,Bischoff J. Endothelial progenitor cellsin infantile hemangioma. Blood. 2004;103(4):1373–1375

40. Greenberger S, Boscolo E, Adini I,Mulliken JB, Bischoff J. Corticosteroidsuppression of VEGF-A in infantilehemangioma-derived stem cells. N EnglJ Med. 2010;362(11):1005–1013

41. Kleinman ME, Blei F, Gurtner GC.Circulating endothelial progenitor cellsand vascular anomalies. Lymphat ResBiol. 2005;3(4):234–239

42. Khan ZA, Boscolo E, Picard A, et al.Multipotential stem cells recapitulatehuman infantile hemangioma inimmunodeficient mice. J Clin Invest.2008;118(7):2592–2599

43. Yu Y, Fuhr J, Boye E, et al. Mesenchymalstem cells and adipogenesis inhemangioma involution. Stem Cells.2006;24(6):1605–1612

44. Itinteang T, Withers AH, Davis PF, Tan ST.Biology of infantile hemangioma. FrontSurg. 2014;1(38):38

45. Mihm MC Jr, Nelson JS. Hypothesis:the metastatic niche theory canelucidate infantile hemangiomadevelopment. J Cutan Pathol. 2010;37(suppl 1):83–87



46. Chang J, Most D, Bresnick S, et al.Proliferative hemangiomas: analysis ofcytokine gene expression andangiogenesis. Plast Reconstr Surg.1999;103(1):1–9; discussion: 10

47. Léauté-Labrèze C, Prey S, Ezzedine K.Infantile haemangioma: part I.Pathophysiology, epidemiology, clinicalfeatures, life cycle and associatedstructural abnormalities. J Eur AcadDermatol Venereol. 2011;25(11):1245–1253

48. Razon MJ, Kräling BM, Mulliken JB,Bischoff J. Increased apoptosiscoincides with onset of involution ininfantile hemangioma. Microcirculation.1998;5(2–3):189–195

49. Dai Y, Hou F, Buckmiller L, et al.Decreased eNOS protein expression ininvoluting and propranolol-treatedhemangiomas. Arch Otolaryngol HeadNeck Surg. 2012;138(2):177–182

50. Ahrens WA, Ridenour RV III, Caron BL,Miller DV, Folpe AL. GLUT-1 expression inmesenchymal tumors: animmunohistochemical study of 247 softtissue and bone neoplasms. HumPathol. 2008;39(10):1519–1526

51. Lavrentieva A, Majore I, Kasper C, HassR. Effects of hypoxic culture conditionson umbilical cord-derived humanmesenchymal stem cells. Cell CommunSignal. 2010;8:18

52. Chisti M, Banka N, Alfadley A. Pallorsign: an indicator of hemangioma inevolution. J Cutan Med Surg. 2012;16(6):451–452

53. North PE, Mihm MC Jr.Histopathological diagnosis of infantilehemangiomas and vascularmalformations. Facial Plast Surg ClinNorth Am. 2001;9(4):505–524

54. North PE, Waner M, Buckmiller L, JamesCA, Mihm MC Jr. Vascular tumors ofinfancy and childhood: beyond capillaryhemangioma. Cardiovasc Pathol. 2006;15(6):303–317

55. North PE. Pediatric vascular tumorsand malformations. Surg Pathol Clin.2010;3(3):455–494

56. Tan ST, Wallis RA, He Y, Davis PF. Mastcells and hemangioma. Plast ReconstrSurg. 2004;113(3):999–1011

57. North PE, Waner M, Mizeracki A, et al.A unique microvascular phenotypeshared by juvenile hemangiomas and

human placenta. Arch Dermatol. 2001;137(5):559–570

58. Li Q, Yu Y, Bischoff J, Mulliken JB, OlsenBR. Differential expression of CD146 intissues and endothelial cells derivedfrom infantile haemangioma andnormal human skin. J Pathol. 2003;201(2):296–302

59. Lister WA. Natural history of strawberrynevi. Lancet. 1938;231(5991):1429–1430

60. Bivings L. Spontaneous regression ofangiomas in children; twenty-two years’observation covering 236 cases.J Pediatr. 1954;45(6):643–647

61. Margileth AM, Museles M. Currentconcepts in diagnosis and managementof congenital cutaneous hemangiomas.Pediatrics. 1965;36(3):410–416

62. Moroz B. Long-term follow-up ofhemangiomas in children. In: WilliamsHB, ed. Symposium on VascularMalformations and Melanotic Lesions.St Louis, MO: CV Mosby; 1982:27–35

63. Mulliken JB. Diagnosis and naturalhistory of hemangiomas. In: MullikenJB, Young AE, eds. Vascular Birthmarks:Hemangiomas and Malformations.Philadelphia, PA: WB Saunders; 1988:41–62

64. Esterly NB. Cutaneous hemangiomas,vascular stains and malformations, andassociated syndromes. Curr ProblPediatr. 1996;26(1):3–39

65. Bowers RE, Graham EA, Tomlinson KM.The natural history of the strawberrynevus. Arch Dermatol. 1960;82(5):667–670

66. Chang LC, Haggstrom AN, Drolet BA,et al; Hemangioma Investigator Group.Growth characteristics of infantilehemangiomas: implications formanagement. Pediatrics. 2008;122(2):360–367

67. Jacobs AH. Strawberry hemangiomas;the natural history of the untreatedlesion. Calif Med. 1957;86(1):8–10

68. Takahashi K, Mulliken JB, KozakewichHPW, Rogers RA, Folkman J, EzekowitzRAB. Cellular markers that distinguishthe phases of hemangioma duringinfancy and childhood. J Clin Invest.1994;93(6):2357–2364

69. Tollefson MM, Frieden IJ. Early growthof infantile hemangiomas: whatparents’ photographs tell us.

Pediatrics. 2012;130(2). Available at:www.pediatrics.org/cgi/content/full/130/2/e31

70. Couto RA, Maclellan RA, Zurakowski D,Greene AK. Infantile hemangioma:clinical assessment of the involutingphase and implications formanagement. Plast Reconstr Surg.2012;130(3):619–624

71. Bauland CG, Lüning TH, Smit JM,Zeebregts CJ, Spauwen PH. Untreatedhemangiomas: growth pattern andresidual lesions. Plast Reconstr Surg.2011;127(4):1643–1648

72. Jackson R. The natural history ofstrawberry naevi. J Cutan Med Surg.1998;2(3):187–189

73. Drolet BA, Esterly NB, Frieden IJ.Hemangiomas in children. N Engl JMed. 1999;341(3):173–181

74. Martinez-Perez D, Fein NA, Boon LM,Mulliken JB. Not all hemangiomas looklike strawberries: uncommonpresentations of the most commontumor of infancy. Pediatr Dermatol.1995;12(1):1–6

75. Corella F, Garcia-Navarro X, Ribe A,Alomar A, Baselga E. Abortive orminimal-growth hemangiomas:immunohistochemical evidence thatthey represent true infantilehemangiomas. J Am Acad Dermatol.2008;58(4):685–690

76. Toledo-Alberola F, Betlloch-Mas I,Cuesta-Montero L, et al. Abortivehemangiomas: description of clinicaland pathological findings with specialemphasis on dermoscopy. Eur JDermatol. 2010;20(4):497–500

77. Suh KY, Frieden IJ. Infantilehemangiomas with minimal orarrested growth: a retrospective caseseries. Arch Dermatol. 2010;146(9):971–976

78. Przewratil P, Sitkiewicz A, AndrzejewskaE. Serum levels of basic fibroblasticgrowth factor (bFGF) in children withvascular anomalies: another insightinto endothelial growth. Clin Biochem.2010;43(10–11):863–867

79. Przewratil P, Sitkiewicz A, AndrzejewskaE. Local serum levels of vascularendothelial growth factor in infantilehemangioma: intriguing mechanism ofendothelial growth. Cytokine. 2010;49(2):141–147


http://www.pediatrics.org/cgi/content/full/130/2/e31



80. Mulliken JB, Marler JJ, Burrows PE,Kozakewich HP. Reticular infantilehemangioma of the limb can beassociated with ventral-caudalanomalies, refractory ulceration, andcardiac overload. Pediatr Dermatol.2007;24(4):356–362

81. Waner M, North PE, Scherer KA, FriedenIJ, Waner A, Mihm MC Jr. Thenonrandom distribution of facialhemangiomas. Arch Dermatol. 2003;139(7):869–875

82. Haggstrom AN, Lammer EJ, SchneiderRA, Marcucio R, Frieden IJ. Patterns ofinfantile hemangiomas: new clues tohemangioma pathogenesis andembryonic facial development.Pediatrics. 2006;117(3):698–703

83. Jinnin M, Medici D, Park L, et al.Suppressed NFAT-dependent VEGFR1expression and constitutive VEGFR2signaling in infantile hemangioma. NatMed. 2008;14(11):1236–1246

84. Haggstrom AN, Drolet BA, Baselga E,et al. Prospective study of infantilehemangiomas: clinical characteristicspredicting complications andtreatment. Pediatrics. 2006;118(3):882–887

85. Iacobas I, Burrows PE, Frieden IJ, et al.LUMBAR: association betweencutaneous infantile hemangiomas ofthe lower body and regional congenitalanomalies. J Pediatr. 2010;157(5):795–801.e1, 7

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

87. Stockman A, Boralevi F, Taïeb A, Léauté-Labrèze C. SACRAL syndrome: spinaldysraphism, anogenital, cutaneous,renal and urologic anomalies,associated with an angioma oflumbosacral localization. Dermatology.2007;214(1):40–45

88. Nabatian AS, Milgraum SS, Hess CP,Mancini AJ, Krol A, Frieden IJ. PHACEwithout face? Infantile hemangiomas ofthe upper body region with minimal orabsent facial hemangiomas andassociated structural malformations.Pediatr Dermatol. 2011;28(3):235–241

89. Hughes JA, Hill V, Patel K, Syed S,Harper J, De Bruyn R. Cutaneoushaemangioma: prevalence andsonographic characteristics of

associated hepatic haemangioma. ClinRadiol. 2004;59(3):273–280

90. Metry DW, Hawrot A, Altman C, FriedenIJ. Association of solitary, segmentalhemangiomas of the skin with visceralhemangiomatosis. Arch Dermatol. 2004;140(5):591–596

91. Christison-Lagay ER, Burrows PE,Alomari A, et al. Hepatic hemangiomas:subtype classification and developmentof a clinical practice algorithm andregistry. J Pediatr Surg. 2007;42(1):62–67; discussion: 67–68

92. Dickie B, Dasgupta R, Nair R, et al.Spectrum of hepatic hemangiomas:management and outcome. J PediatrSurg. 2009;44(1):125–133

93. Horii KA, Drolet BA, Baselga E, et al;Hemangioma Investigator Group. Riskof hepatic hemangiomas in infants withlarge hemangiomas. Arch Dermatol.2010;146(2):201–203

94. Drolet BA, Pope E, Juern AM, et al.Gastrointestinal bleeding in infantilehemangioma: a complication ofsegmental, rather than multifocal,infantile hemangiomas. J Pediatr. 2012;160(6):1021–6.e3

95. Brandling-Bennett HA, Metry DW,Baselga E, et al. Infantile hemangiomaswith unusually prolonged growthphase: a case series. Arch Dermatol.2008;144(12):1632–1637

96. Kim HJ, Colombo M, Frieden IJ.Ulcerated hemangiomas: clinicalcharacteristics and response totherapy. J Am Acad Dermatol. 2001;44(6):962–972

97. Shin HT, Orlow SJ, Chang MW. Ulceratedhaemangioma of infancy:a retrospective review of 47 patients.Br J Dermatol. 2007;156(5):1050–1052

98. Chamlin SL, Haggstrom AN, Drolet BA,et al. Multicenter prospective study ofulcerated hemangiomas. J Pediatr.2007;151(6):684–689; 689.e1

99. Maguiness SM, Hoffman WY, McCalmontTH, Frieden IJ. Early white discoloratonof infantile hemangioma: a sign ofimpending ulceration. Arch Dermatol.2010;146(11):1235–1239

100. Thomas RF, Hornung RL, Manning SC,Perkins JA. Hemangiomas of infancy:treatment of ulceration in the head andneck. Arch Facial Plast Surg. 2005;7(5):312–315

101. Waner M, Suen JY. The natural historyof hemangiomas. In: Waner M, Suen JY,eds. Hemangiomas and VascularMalformations of the Head and Neck.New York, NY: Wiley-Liss; 1999:13–45

102. Hermans DJ, Boezeman JB, Van deKerkhof PC, Rieu PN, Van der Vleuten CJ.Differences between ulcerated and non-ulcerated hemangiomas:a retrospective study of 465 cases. EurJ Dermatol. 2009;19(2):152–156

103. Connelly EA, Viera M, Price C, Waner M.Segmental hemangioma of infancycomplicated by life-threatening arterialbleed. Pediatr Dermatol. 2009;26(4):469–472

104. Yan AC. Pain management for ulceratedhemangiomas. Pediatr Dermatol. 2008;25(6):586–589

105. Thomas MW, Burkhart CN, Vaghani SP,Morrell DS, Wagner AM. Failure to thrivein infants with complicated facialhemangiomas. Pediatr Dermatol. 2012;29(1):49–52

106. Orlow SJ, Isakoff MS, Blei F. Increasedrisk of symptomatic hemangiomas ofthe airway in association withcutaneous hemangiomas in a “beard”distribution. J Pediatr. 1997;131(4):643–646

107. Haggstrom AN, Skillman S, Garzon MC,et al. Clinical spectrum and risk ofPHACE syndrome in cutaneous andairway hemangiomas. Arch OtolaryngolHead Neck Surg. 2011;137(7):680–687

108. O TM, Alexander RE, Lando T, et al.Segmental hemangiomas of the upperairway. Laryngoscope. 2009;119(11):2242–2247

109. Jockin YM, Friedlander SF. Periocularinfantile hemangioma. Int OphthalmolClin. 2010;50(4):15–25

110. Schwartz SR, Blei F, Ceisler E, Steele M,Furlan L, Kodsi S. Risk factors foramblyopia in children with capillaryhemangiomas of the eyelids and orbit.J AAPOS. 2006;10(3):262–268

111. Dubois J, Milot J, Jaeger BI, McCuaig C,Rousseau E, Powell J. Orbit and eyelidhemangiomas: is there a relationshipbetween location and ocular problems?J Am Acad Dermatol. 2006;55(4):614–619

112. Frank RC, Cowan BJ, Harrop AR, AstleWF, McPhalen DF. Visual developmentin infants: visual complications of



periocular haemangiomas. J PlastReconstr Aesthet Surg. 2010;63(1):1–8

113. Howell DM, Gumbiner CH, Martin GE.Congestive heart failure due to giantcutaneous cavernous hemangioma. ClinPediatr (Phila). 1984;23(9):504–506

114. Weber TR, West KW, Cohen M, GrosfeldJL. Massive hemangioma in infants:therapeutic considerations. J VascSurg. 1984;1(3):423–428

115. Frieden IJ, Reese V, Cohen D. PHACEsyndrome: the association of posteriorfossa brain malformations,hemangiomas, arterial anomalies,coarctation of the aorta and cardiacdefects, and eye abnormalities. ArchDermatol. 1996;132(3):307–311

116. Mitchell S, Siegel DH, Shieh JT, et al.Candidate locus analysis for PHACEsyndrome. Am J Med Genet A. 2012;158A(6):1363–1367

117. Opitz JM, Gilbert EF. CNS anomalies andthe midline as a “developmental field”.Am J Med Genet. 1982;12(4):443–455

118. Metry DW, Dowd CF, Barkovich AJ,Frieden IJ. The many faces of PHACEsyndrome. J Pediatr. 2001;139(1):117–123

119. Hess CP, Fullerton HJ, Metry DW, et al.Cervical and intracranial arterialanomalies in 70 patients with PHACEsyndrome. AJNR Am J Neuroradiol.2010;31(10):1980–1986

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

121. Metry DW, Haggstrom AN, Drolet BA,et al. A prospective study of PHACEsyndrome in infantile hemangiomas:demographic features, clinical findings,and complications. Am J Med Genet A.2006;140(9):975–986

122. Tangtiphaiboontana J, Hess CP, Bayer M,et al. Neurodevelopmentalabnormalities in children with PHACEsyndrome. J Child Neurol. 2013;28(5):608–614

123. Duffy KJ, Runge-Samuelson C, Bayer ML,et al. Association of hearing loss withPHACE syndrome. Arch Dermatol. 2010;146(12):1391–1396

124. Rao RP, Drolet BA, Holland KE, FrommeltPC. PHACES association:a vasculocutaneous syndrome. PediatrCardiol. 2008;29(4):793–799

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

126. Siegel DH, Tefft KA, Johnson C, et al.Stroke in children with posterior fossabrain malformations, hemangiomas,arterial anomalies, coarctation of theaorta and cardiac defects, and eyeabnormalities (PHACE) syndrome:a systematic review of the literature.Stroke. 2012;43(6):1672–1674

127. Bingham MM, Saltzman B, Vo NJ,Perkins JA. Propranolol reducesinfantile hemangioma volume andvessel density. Otolaryngol Head NeckSurg. 2012;147(2):338–344

128. Kassarjian A, Zurakowski D, Dubois J,Paltiel HJ, Fishman SJ, Burrows PE.Infantile hepatic hemangiomas: clinicaland imaging findings and theircorrelation with therapy. AJR Am JRoentgenol. 2004;182(3):785–795

129. Frieden IJ, Haggstrom AN, Drolet BA,et al. Infantile hemangiomas: currentknowledge, future directions.Proceedings of a research workshop oninfantile hemangiomas, April 7–9, 2005,Bethesda, Maryland. Pediatr Dermatol.2005;22(5):383–406

130. Meyer JS, Hoffer FA, Barnes PD,Mulliken JB. Biological classification ofsoft-tissue vascular anomalies: MRcorrelation. AJR Am J Roentgenol. 1991;157(3):559–564

131. Flors L, Leiva-Salinas C, Maged IM, et al.MR imaging of soft-tissue vascularmalformations: diagnosis,classification, and therapy follow-up.Radiographics. 2011;31(5):1321–1340;discussion: 1340–1341

132. Maguiness SM, Frieden IJ. Currentmanagement of infantile hemangiomas.Semin Cutan Med Surg. 2010;29(2):106–114

133. Leonardi-Bee J, Batta K, O’Brien C, Bath-Hextall FJ. Interventions for infantilehaemangiomas (strawberrybirthmarks) of the skin. CochraneDatabase Syst Rev. 2011;(5):CD006545

134. Brownell CA, Zerwas S, Ramani GB. “Sobig”: the development of body self-

awareness in toddlers. Child Dev. 2007;78(5):1426–1440

135. Brogdon J. Psychosocial impact ofcongenital vascular lesions. In: WanerM, Suen JY, eds. Hemangiomas andVascular Malformations of the Headand Neck. New York, NY: Wiley-Liss;1999:13–45

136. Zweegers J, van der Vleuten CJ. Thepsychosocial impact of an infantilehaemangioma on children and theirparents. Arch Dis Child. 2012;97(10):922–926

137. Cohen-Barak E, Rozenman D, Shani AdirA. Infantile haemangiomas and qualityof life. Arch Dis Child. 2013;98(9):676–679

138. Fost NC, Esterly NB. Successfultreatment of juvenile hemangiomaswith prednisone. J Pediatr. 1968;72(3):351–357

139. Cohen SR, Wang CI. Steroid treatment ofhemangioma of the head and neck inchildren. Ann Otol Rhinol Laryngol.1972;81(4):584–590

140. Greenberger S, Bischoff J. Infantilehemangioma-mechanism(s) of drugaction on a vascular tumor. Cold SpringHarb Perspect Med. 2011;1(1):a006460

141. Storch CH, Hoeger PH. Propranolol forinfantile haemangiomas: insights intothe molecular mechanisms of action. BrJ Dermatol. 2010;163(2):269–274

142. Vivas-Colmenares GV, Bernabeu-Wittel J,Alonso-Arroyo V, Matute de CardenasJA, Fernandez-Pineda I. Effectiveness ofpropranolol in the treatment of infantilehemangioma beyond the proliferationphase. Pediatr Dermatol. 2015;32(3):348–352

143. Zvulunov A, McCuaig C, Frieden IJ, et al.Oral propranolol therapy for infantilehemangiomas beyond the proliferationphase: a multicenter retrospectivestudy. Pediatr Dermatol. 2011;28(2):94–98

144. Hermans DJ, van Beynum IM, SchultzeKool LJ, van de Kerkhof PC, Wijnen MH,van der Vleuten CJ. Propranolol, a verypromising treatment for ulceration ininfantile hemangiomas: a study of 20cases with matched historical controls.J Am Acad Dermatol. 2011;64(5):833–838

145. Saint-Jean M, Léauté-Labrèze C,Mazereeuw-Hautier J, et al; Groupe de





Recherche Clinique en DermatologiePédiatrique. Propranolol for treatmentof ulcerated infantile hemangiomas.J Am Acad Dermatol. 2011;64(5):827–832

146. Hong E, Fischer G. Propranolol forrecalcitrant ulcerated hemangioma ofinfancy. Pediatr Dermatol. 2012;29(1):64–67

147. Cante V, Pham-Ledard A, Imbert E,Ezzedine K, Léauté-Labrèze C. Firstreport of topical timolol treatment inprimarily ulcerated perinealhaemangioma. Arch Dis Child FetalNeonatal Ed. 2012;97(2):F155–F156

148. David LR, Malek MM, Argenta LC.Efficacy of pulse dye laser therapy forthe treatment of ulceratedhaemangiomas: a review of 78 patients.Br J Plast Surg. 2003;56(4):317–327

149. Lacour M, Syed S, Linward J, Harper JI.Role of the pulsed dye laser in themanagement of ulcerated capillaryhaemangiomas. Arch Dis Child. 1996;74(2):161–163

150. Morelli JG, Tan OT, Weston WL.Treatment of ulcerated hemangiomaswith the pulsed tunable dye laser. Am JDis Child. 1991;145(9):1062–1064

151. Witman PM, Wagner AM, Scherer K,Waner M, Frieden IJ. Complicationsfollowing pulsed dye laser treatment ofsuperficial hemangiomas. Lasers SurgMed. 2006;38(2):116–123

152. Katz HP, Askin J. Multiple hemangiomatawith thrombopenia: an unusual casewith comments on steroid therapy. AmJ Dis Child. 1968;115(3):351–357

153. Zarem HA, Edgerton MT. Inducedresolution of cavernous hemangiomasfollowing prednisolone therapy. PlastReconstr Surg. 1967;39(1):76–83

154. Groopman JE, Gottlieb MS, Goodman J,et al. Recombinant alpha-2 interferontherapy for Kaposi’s sarcomaassociated with the acquiredimmunodeficiency syndrome. AnnIntern Med. 1984;100(5):671–676

155. Ritter MR, Dorrell MI, Edmonds J,Friedlander SF, Friedlander M. Insulin-like growth factor 2 and potentialregulators of hemangioma growthand involution identified by large-scale expression analysis. ProcNatl Acad Sci USA. 2002;99(11):7455–7460

156. White CW, Sondheimer HM, Crouch EC,Wilson H, Fan LL. Treatment ofpulmonary hemangiomatosis withrecombinant interferon alfa-2a. N Engl JMed. 1989;320(18):1197–1200

157. Ezekowitz RA, Mulliken JB, Folkman J.Interferon alfa-2a therapy for life-threatening hemangiomas of infancy.N Engl J Med. 1992;326(22):1456–1463

158. Léauté-Labrèze C, Dumas de la Roque E,Hubiche T, Boralevi F, Thambo JB, TaïebA. Propranolol for severe hemangiomasof infancy. N Engl J Med. 2008;358(24):2649–2651

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

160. Starkey E, Shahidullah H. Propranololfor infantile haemangiomas: a review.Arch Dis Child. 2011;96(9):890–893

161. Drolet BA, Frommelt PC, Chamlin SL,et al. Initiation and use of propranololfor infantile hemangioma: report ofa consensus conference. Pediatrics.2013;131(1):128–140

162. Pope E, Chakkittakandiyil A, Lara-Corrales I, Maki E, Weinstein M.Expanding the therapeutic repertoire ofinfantile haemangiomas: cohort-blindedstudy of oral nadolol compared withpropranolol. Br J Dermatol. 2013;168(1):222–224

163. Pan WK, Li P, Guo ZT, Huang Q, Gao Y.Propranolol induces regression ofhemangioma cells via the down-regulation of the PI3K/Akt/eNOS/VEGFpathway. Pediatr Blood Cancer. 2015;62(8):1414–1420

164. Sharifpanah F, Saliu F, Bekhite MM,Wartenberg M, Sauer H. b-Adrenergicreceptor antagonists inhibitvasculogenesis of embryonic stem cellsby downregulation of nitric oxidegeneration and interference with VEGFsignalling. Cell Tissue Res. 2014;358(2):443–452

165. Ma X, Zhao T, Ouyang T, Xin S, Ma Y,Chang M. Propranolol enhancedadipogenesis instead of induction ofapoptosis of hemangiomas stem cells.Int J Clin Exp Pathol. 2014;7(7):3809–3817

166. Kum JJ, Khan ZA. Propranolol inhibitsgrowth of hemangioma-initiating cellsbut does not induce apoptosis. PediatrRes. 2014;75(3):381–388

167. Ji Y, Chen S, Xu C, Li L, Xiang B. The useof propranolol in the treatment ofinfantile haemangiomas: an update onpotential mechanisms of action. Br JDermatol. 2015;172(1):24–32

168. Lamy S, Lachambre MP, Lord-Dufour S,Béliveau R. Propranolol suppressesangiogenesis in vitro: inhibition ofproliferation, migration, anddifferentiation of endothelial cells.Vascul Pharmacol. 2010;53(5-6):200–208

169. Frieden IJ, Drolet BA. Propranolol forinfantile hemangiomas: promise, peril,pathogenesis. Pediatr Dermatol. 2009;26(5):642–644

170. Lee D, Boscolo E, Durham JT, MullikenJB, Herman IM, Bischoff J. Propranololtargets the contractility of infantilehaemangioma-derived pericytes. Br JDermatol. 2014;171(5):1129–1137

171. England RW, Hardy KL, Kitajewski AM,et al. Propranolol promotes acceleratedand dysregulated adipogenesis inhemangioma stem cells. Ann PlastSurg. 2014;73(suppl 1):S119–S124

172. Léauté-Labrèze C, Hoeger P, Mazereeuw-Hautier J, et al. A randomized,controlled trial of oral propranolol ininfantile hemangioma. N Engl J Med.2015;372(8):735–746

173. Hogeling M, Adams S, Wargon O.A randomized controlled trial ofpropranolol for infantile hemangiomas.Pediatrics. 2011;128(2). Available at:www.pediatrics.org/cgi/content/full/128/2/e259

174. Schiestl C, Neuhaus K, Zoller S, et al.Efficacy and safety of propranolol asfirst-line treatment for infantilehemangiomas. Eur J Pediatr. 2011;170(4):493–501

175. de Graaf M, Breur JM, Raphaël MF, VosM, Breugem CC, Pasmans SG. Adverseeffects of propranolol when used in thetreatment of hemangiomas: a caseseries of 28 infants. J Am AcadDermatol. 2011;65(2):320–327

176. Raphael MF, Breugem CC, Vlasveld FA,et al. Is cardiovascular evaluationnecessary prior to and during beta-blocker therapy for infantile







hemangiomas? A cohort study. J AmAcad Dermatol. 2015;72(3):465–472

177. Blei F, McElhinney DB, Guarini A, PrestiS. Cardiac screening in infants withinfantile hemangiomas beforepropranolol treatment. PediatrDermatol. 2014;31(4):465–470

178. Puttgen KB, Summerer B, Schneider J,Cohen BA, Boss EF, Bauman NM.Cardiovascular and blood glucoseparameters in infants duringpropranolol initiation for treatment ofsymptomatic infantile hemangiomas.Ann Otol Rhinol Laryngol. 2013;122(9):550–554

179. McSwiney E, Murray D, Murphy M.Propranolol therapy for cutaneousinfantile haemangiomas initiated safelyas a day-case procedure. Eur J Pediatr.2014;173(1):63–68

180. Holland KE, Frieden IJ, Frommelt PC,Mancini AJ, Wyatt D, Drolet BA.Hypoglycemia in children takingpropranolol for the treatment ofinfantile hemangioma. Arch Dermatol.2010;146(7):775–778

181. Fusilli G, Merico G, Gurrado R, Rosa T,Acquafredda A, Cavallo L. Propranololfor infantile haemangiomas andneuroglycopenic seizures. ActaPaediatr. 2010;99(12):1756

182. Marqueling AL, Oza V, Frieden IJ,Puttgen KB. Propranolol and infantilehemangiomas four years later:a systematic review. Pediatr Dermatol.2013;30(2):182–191

183. Léauté-Labrèze C, Frieden I, Vabres P,Prey S, Voisard J. Propranolol ininfantile hemangiomas (IH): resultsfrom an international randomized,placebo controlled, multidose, adaptivephase 2/3 study. In: 20th Workshop ofthe International Society for the Studyof Vascular Anomalies; April 2–4, 2014;Melbourne, Australia. Abstract

184. Prey S, Léauté-Labrèze C, Frieden I,Delarue A, Voisard J. Safety of oralpropranolol for the treatment ofinfantile hemangioma: 2-years resultsof a phase II/III randomized, double-blind, placebo-controlled multicentertrial. In: 20th Workshop of theInternational Society for the Study ofVascular Anomalies; April 2–4, 2014;Melbourne, Australia. Abstract

185. Braqazquoitia L, Hernandez-Martin A,Torrelo A. Recurrence of infantile

hemangiomas treated withpropranolol. Pediatr Dermatol. 2011;28(6):658–662

186. Shehata N, Powell J, Dubois J, et al. Laterebound of infantile hemangioma aftercessation of oral propranolol. PediatrDermatol. 2013;30(5):587–591

187. Shah S, Frieden I, Baselga E, McCuaig C,Pope E. Rebound after discontinuationof propranolol in the therapy ofinfantile hemangiomas: a retrospectivestudy. In: 20th Workshop of theInternational Society for the Study ofVascular Anomalies; April 2-4, 2014;Melbourne, Australia. Abstract

188. Solman L, Murabit A, Gnarra M, HarperJI, Syed SB, Glover M. Propranolol forinfantile haemangiomas: single centreexperience of 250 cases and proposedtherapeutic protocol. Arch Dis Child.2014;99(12):1132–1136

189. Giachetti A, Garcia-Monaco R, Sojo M,et al. Long-term treatment with oralpropranolol reduces relapses ofinfantile hemangiomas. PediatrDermatol. 2014;31(1):14–20

190. McMahon P, Oza V, Frieden IJ. Topicaltimolol for infantile hemangiomas:putting a note of caution in “cautiouslyoptimistic”. Pediatr Dermatol. 2012;29(1):127–130

191. Semkova K, Kazandjieva J. Topicaltimolol maleate for treatment ofinfantile haemangiomas: preliminaryresults of a prospective study. Clin ExpDermatol. 2013;38(2):143–146

192. Moehrle M, Léauté-Labrèze C, SchmidtV, Röcken M, Poets CF, Goelz R. Topicaltimolol for small hemangiomas ofinfancy. Pediatr Dermatol. 2013;30(2):245–249

193. Oranje AP, Janmohamed SR, Madern GC,de Laat PC. Treatment of smallsuperficial haemangioma with timolol0.5% ophthalmic solution: a series of 20cases. Dermatology. 2011;223(4):330–334

194. Ni N, Langer P, Wagner R, Guo S. Topicaltimolol for periocular hemangioma:report of further study. ArchOphthalmol. 2011;129(3):377–379

195. Chakkittakandiyii A, Phillips R, FriedenIJ, et al. Timolol maleate 0.5% or 0.1%gel-forming solution for infantilehemangiomas: a retrospective,

multicenter, cohort study. PediatrDermatol. 2012;29(3):28–31

196. Chan H, McKay C, Adams S, Wargon O.RCT of timolol maleate gel forsuperficial infantile hemangiomas in 5-to 24-week-olds. Pediatrics. 2013;131(6).Available at: www.pediatrics.org/cgi/content/full/131/6/e1739

197. Hasan Q, Tan ST, Xu B, Davis PF. Effectsof five commonly used glucocorticoidson haemangioma in vitro. Clin ExpPharmacol Physiol. 2003;30(3):140–144

198. Ebrahem Q, Minamoto A, Hoppe G,Anand-Apte B, Sears JE. Triamcinoloneacetonide inhibits IL-6- and VEGF-induced angiogenesis downstream ofthe IL-6 and VEGF receptors. InvestOphthalmol Vis Sci. 2006;47(11):4935–4941

199. Itinteang T, Vishvanath A, Day DJ, Tan ST.Mesenchymal stem cells in infantilehaemangioma. J Clin Pathol. 2011;64(3):232–236

200. Pantoja C, Huff JT, Yamamoto KR.Glucocorticoid signaling defines a novelcommitment state during adipogenesisin vitro. Mol Biol Cell. 2008;19(10):4032–4041

201. Ameshima S, Golpon H, Cool CD, et al.Peroxisome proliferator-activatedreceptor gamma (PPARgamma)expression is decreased in pulmonaryhypertension and affects endothelialcell growth. Circ Res. 2003;92(10):1162–1169

202. Wu Z, Bucher NL, Farmer SR. Inductionof peroxisome proliferator-activatedreceptor gamma during the conversionof 3T3 fibroblasts into adipocytes ismediated by C/EBPbeta, C/EBPdelta, andglucocorticoids. Mol Cell Biol. 1996;16(8):4128–4136

203. Greene AK, Couto RA. Oral prednisolonefor infantile hemangioma: efficacy andsafety using a standardized treatmentprotocol. Plast Reconstr Surg. 2011;128(3):743–752

204. Sadan N, Wolach B. Treatment ofhemangiomas of infants with highdoses of prednisone. J Pediatr. 1996;128(1):141–146

205. Bennett ML, Fleischer AB Jr, Chamlin SL,Frieden IJ. Oral corticosteroid use iseffective for cutaneous hemangiomas:an evidence-based evaluation. ArchDermatol. 2001;137(9):1208–1213





206. Itinteang T, Withers AHJ, Leadbitter P,Day DJ, Tan ST. Pharmacologictherapies for infantile hemangioma: isthere a rational basis? Plast ReconstrSurg. 2011;128(2):499–507

207. Bauman NM, McCarter RJ, Guzzetta PC,et al. Propranolol vs prednisolone forsymptomatic proliferating infantilehemangiomas: a randomized clinicaltrial. JAMA Otolaryngol Head Neck Surg.2014;140(4):323–330

208. Grantzow R, Schmittenbecher P, CremerH, et al. Hemangiomas in infancy andchildhood. S 2k Guideline of the GermanSociety of Dermatology with theworking group Pediatric Dermatologytogether with the German Society forPediatric Surgery and the GermanSociety for Pediatric Medicine. J DtschDermatol Ges. 2008;6(4):324–329

209. Frieden IJ, Eichenfield LF, Esterly NB,Geronemus R, Mallory SB; AmericanAcademy of Dermatology Guidelines/Outcomes Committee. Guidelines ofcare for hemangiomas of infancy. J AmAcad Dermatol. 1997;37(4):631–637

210. Azzolini A, Nouvenne R. Nuoveprospettive nella terapia degli angiomiimmaturi dell infanzia, 115 lesion;trattate can infiltraziona; intralesionalidi triamcinolone acetonide. Acta BiolMed (Gdansk). 1970;41(suppl 1):41–51

211. Mazzola RF. Treatment ofhaemangiomas in children byintralesional injection of steroids. ChirPlast (Berl). 1978;4:161–171

212. Kushner BJ. The treatment ofperiorbital infantile hemangioma withintralesional corticosteroid. PlastReconstr Surg. 1985;76(4):517–526

213. Sloan GM, Reinisch JF, Nichter LS, SaberWL, Lew K, Morwood DT. Intralesionalcorticosteroid therapy for infantilehemangiomas. Plast Reconstr Surg.1989;83(3):459–467

214. Chowdri NA, Darzi MA, Fazili Z, Iqbal S.Intralesional corticosteroid therapy forchildhood cutaneous hemangiomas.Ann Plast Surg. 1994;33(1):46–51

215. Chen MT, Yeong EK, Horng SY.Intralesional corticosteroid therapy inproliferating head and neckhemangiomas: a review of 155 cases.J Pediatr Surg. 2000;35(3):420–423

216. Buckmiller LM, Francis CL, Glade RS.Intralesional steroid injection for

proliferative parotid hemangiomas. IntJ Pediatr Otorhinolaryngol. 2008;72(1):81–87

217. Chantharatanapiboon W. Intralesionalcorticosteroid therapy inhemangiomas: clinical outcome in 160cases. J Med Assoc Thai. 2008;91(suppl3):S90–S96

218. Prasetyono TOH, Djoenaedi I. Efficacy ofintralesional steroid injection in headand neck hemangioma: a systematicreview. Ann Plast Surg. 2011;66(1):98–106

219. Samimi DB, Alabiad CR, Tse DT. Ananatomically based approach tointralesional corticosteroid injectionfor eyelid capillary hemangiomas.Ophthalmic Surg Lasers Imaging. 2012;43(3):190–195

220. Goyal R, Watts P, Lane CM, Beck L,Gregory JW. Adrenal suppression andfailure to thrive after steroid injectionsfor periocular hemangioma.Ophthalmology. 2004;111(2):389–395

221. Shorr N, Seiff SR. Central retinal arteryocclusion associated with periocularcorticosteroid injection for juvenilehemangioma. Ophthalmic Surg. 1986;17(4):229–231

222. Ruttum MS, Abrams GW, Harris GJ, EllisMK. Bilateral retinal embolizationassociated with intralesionalcorticosteroid injection for capillaryhemangioma of infancy. J PediatrOphthalmol Strabismus. 1993;30(1):4–7

223. Egbert JE, Schwartz GS, Walsh AW.Diagnosis and treatment of anophthalmic artery occlusion during anintralesional injection of corticosteroidinto an eyelid capillary hemangioma.Am J Ophthalmol. 1996;121(6):638–642

224. Egbert JE, Paul S, Engel WK, SummersCG. High injection pressure duringintralesional injection ofcorticosteroids into capillaryhemangiomas. Arch Ophthalmol. 2001;119(5):677–683

225. Elsas FJ, Lewis AR. Topical treatment ofperiocular capillary hemangioma.J Pediatr Ophthalmol Strabismus. 1994;31(3):153–156

226. Cruz OA, Zarnegar SR, Myers SE.Treatment of periocular capillaryhemangioma with topical clobetasolpropionate. Ophthalmology. 1995;102(12):2012–2015

227. Garzon MC, Lucky AW, Hawrot A, FriedenIJ. Ultrapotent topical corticosteroidtreatment of hemangiomas of infancy.J Am Acad Dermatol. 2005;52(2):281–286

228. Pandey A, Gangopadhyay AN, SharmaSP, Kumar V, Gupta DK, Gopal SC.Evaluation of topical steroids in thetreatment of superficial hemangioma.Skinmed. 2010;8(1):9–11

229. Greene AK. Current concepts ofvascular anomalies. J Craniofac Surg.2012;23(1):220–224

230. Tan BH, Leadbitter PH, Aburn NH, Tan ST.Steroid therapy for problematicproliferating haemangioma. N Z Med J.2011;124(1329):57–65

231. Morkane C, Gregory JW, Watts P, WarnerJT. Adrenal suppression followingintralesional corticosteroids forperiocular haemangiomas. Arch DisChild. 2011;96(6):587–589

232. Ranchod TM, Frieden IJ, Fredrick DR.Corticosteroid treatment of periorbitalhaemangioma of infancy: a review ofthe evidence. Br J Ophthalmol. 2005;89(9):1134–1138

233. Pope E, Krafchik BR, Macarthur C, et al.Oral versus high-dose pulsecorticosteroids for problematicinfantile hemangiomas: a randomized,controlled trial. Pediatrics. 2007;119(6).Available at: www.pediatrics.org/cgi/content/full/119/6/e1239

234. George ME, Sharma V, Jacobson J,Simon S, Nopper AJ. Adverse effects ofsystemic glucocorticosteroid therapy ininfants with hemangiomas. ArchDermatol. 2004;140(8):963–969

235. Lomenick JP, Reifschneider KL, LuckyAW, et al. Prevalence of adrenalinsufficiency following systemicglucocorticoid therapy in infants withhemangiomas. Arch Dermatol. 2009;145(3):262–266

236. Boon LM, MacDonald DM, Mulliken JB.Complications of systemiccorticosteroid therapy for problematichemangioma. Plast Reconstr Surg.1999;104(6):1616–1623

237. Kelly ME, Juern AM, Grossman WJ,Schauer DW, Drolet BA.Immunosuppressive effects in infantstreated with corticosteroids forinfantile hemangiomas. Arch Dermatol.2010;146(7):767–774





238. Aviles R, Boyce TG, Thompson DM.Pneumocystis carinii pneumonia in a3-month-old infant receiving high-dosecorticosteroid therapy for airwayhemangiomas. Mayo Clin Proc. 2004;79(2):243–245

239. Maronn ML, Corden T, Drolet BA.Pneumocystis carinii pneumonia ininfant treated with oral steroids forhemangioma. Arch Dermatol. 2007;143(9):1224–1225

240. Pokorny JJ, Roth F, Balfour I, Rinehart G.An unusual complication of thetreatment of a hemangioma. Ann PlastSurg. 2002;48(1):83–87

241. Toogood JH, Markov AE, Baskerville J,Dyson C. Association of ocular cataractswith inhaled and oral steroid therapyduring long-term treatment of asthma.J Allergy Clin Immunol. 1993;91(2):571–579

242. Carnahan MC, Goldstein DA. Ocularcomplications of topical, peri-ocular,and systemic corticosteroids. CurrOpin Ophthalmol. 2000;11(6):478–483

243. Razeghinejad MR, Katz LJ. Steroid-induced iatrogenic glaucoma.Ophthalmic Res. 2012;47(2):66–80

244. Yamashita T, Kodama Y, Tanaka M,Yamakiri K, Kawano Y, Sakamoto T.Steroid-induced glaucoma in childrenwith acute lymphoblastic leukemia:a possible complication. J Glaucoma.2010;19(3):188–190

245. Al-Shahwan S, Khan AO. Buphthalmosfollowing systemic steroid treatment.J Pediatr Ophthalmol Strabismus. 2006;43(5):311–312

246. Kushner BJ, Lemke BN. Bilateral retinalembolization associated withintralesional corticosteroid injectionfor capillary hemangioma of infancy.J Pediatr Ophthalmol Strabismus. 1993;30(6):397–399

247. Kushner BJ. Hemangiomas. ArchOphthalmol. 2000;118(6):835–836

248. Mabeta P, Pepper MS. A comparativestudy on the anti-angiogenic effects ofDNA-damaging and cytoskeletal-disrupting agents. Angiogenesis. 2009;12(1):81–90

249. Ghadially FN. Ultrastructural Pathologyof the Cell and Matrix. 3rd ed.London, United Kingdom: Butterworths;1988

250. Perez Payarols J, Pardo Masferrer J,Gomez Bellvert C. Treatment of life-threatening infantile hemangiomaswith vincristine. N Engl J Med. 1995;333(1):69

251. Enjolras O, Brevière GM, Roger G, et al.Vincristine treatment for function- andlife-threatening infantile hemangioma.Arch Pediatr. 2004;11(2):99–107

252. Pérez-Valle S, Peinador M, Herraiz P,Saénz P, Montoliu G, Vento M.Vincristine, an efficacious alternativefor diffuse neonatal haemangiomatosis[in French]. Acta Paediatr. 2010;99(2):311–315

253. Mabeta P, Pepper MS. Hemangiomas—current therapeutic strategies. Int J DevBiol. 2011;55(4-5):431–437

254. Greinwald JH Jr, Burke DK, Bonthius DJ,Bauman NM, Smith RJ. An update on thetreatment of hemangiomas in childrenwith interferon alfa-2a. ArchOtolaryngol Head Neck Surg. 1999;125(1):21–27

255. Barlow CF, Priebe CJ, Mulliken JB, et al.Spastic diplegia as a complication ofinterferon alfa-2a treatment ofhemangiomas of infancy. J Pediatr.1998;132(3 pt 1):527–530

256. Wörle H, Maass E, Köhler B, Treuner J.Interferon alpha-2a therapy inhaemangiomas of infancy: spasticdiplegia as a severe complication. Eur JPediatr. 1999;158(4):344

257. Dubois J, Hershon L, Carmant L,Bélanger S, Leclerc JM, David M. Toxicityprofile of interferon alfa-2b in children:a prospective evaluation. J Pediatr.1999;135(6):782–785

258. Li VW, Li WW, Talcott KE, Zhai AW.Imiquimod as an antiangiogenic agent.J Drugs Dermatol. 2005;4(6):708–717

259. Majewski S, Marczak M, Mlynarczyk B,Benninghoff B, Jablonska S. Imiquimodis a strong inhibitor of tumor cell-induced angiogenesis. Int J Dermatol.2005;44(1):14–19

260. Sidbury R, Neuschler N, Neuschler E,et al. Topically applied imiquimodinhibits vascular tumor growth in vivo.J Invest Dermatol. 2003;121(5):1205–1209

261. Martinez MI, Sanchez-Carpintero I,North PE, Mihm MC Jr. Infantilehemangioma: clinical resolution with

5% imiquimod cream. Arch Dermatol.2002;138(7):881–884; discussion: 884

262. Ho NT, Lansang P, Pope E. Topicalimiquimod in the treatment of infantilehemangiomas: a retrospective study.J Am Acad Dermatol. 2007;56(1):63–68

263. Czernik A, Bystryn JC. Does imiquimodwork in infantile hemangiomas? J AmAcad Dermatol. 2007;57(3):535;– authorreply: 536

264. McCuaig CC, Dubois J, Powell J, et al.A phase II, open-label study of theefficacy and safety of imiquimod in thetreatment of superficial and mixedinfantile hemangioma. PediatrDermatol. 2009;26(2):203–212

265. Jayson GC, Hicklin DJ, Ellis LM.Antiangiogenic therapy—evolving viewbased on clinical trial results. Nat RevClin Oncol. 2012;9(5):297–303

266. Mahajan D, Miller C, Hirose K,McCullough A, Yerian L. Incidentalreduction in the size of liverhemangioma following use of VEGFinhibitor bevacizumab. J Hepatol. 2008;49(5):867–870

267. Greenberger S, Yuan S, Walsh LA, et al.Rapamycin suppresses self-renewaland vasculogenic potential of stem cellsisolated from infantile hemangioma.J Invest Dermatol. 2011;131(12):2467–2476

268. Ashinoff R, Geronemus RG. Capillaryhemangiomas and treatment with theflash lamp-pumped pulsed dye laser.Arch Dermatol. 1991;127(2):202–205

269. Barlow RJ, Walker NP, Markey AC.Treatment of proliferativehaemangiomas with the 585 nm pulseddye laser. Br J Dermatol. 1996;134(4):700–704

270. Garden JM, Bakus AD, Paller AS.Treatment of cutaneous hemangiomasby the flashlamp-pumped pulsed dyelaser: prospective analysis. J Pediatr.1992;120(4 pt 1):555–560

271. Glassberg E, Lask G, Rabinowitz LG,Tunnessen WW Jr. Capillaryhemangiomas: case study of a novellaser treatment and a review oftherapeutic options. J Dermatol SurgOncol. 1989;15(11):1214–1223

272. Haywood RM, Monk BE, Mahaffey PJ.The treatment of early cutaneouscapillary haemangiomata (strawberry



naevi) with the tunable dye laser. Br JPlast Surg. 2000;53(4):302–307

273. Landthaler M, Hohenleutner U,el-Raheem TA. Laser therapy ofchildhood haemangiomas. Br JDermatol. 1995;133(2):275–281

274. Poetke M, Philipp C, Berlien HP.Flashlamp-pumped pulsed dye laser forhemangiomas in infancy: treatment ofsuperficial vs mixed hemangiomas.Arch Dermatol. 2000;136(5):628–632

275. Scheepers JH, Quaba AA. Does thepulsed tunable dye laser have a role inthe management of infantilehemangiomas? Observations based on3 years’ experience. Plast ReconstrSurg. 1995;95(2):305–312

276. Waner M, Suen JY, Dinehart S.Treatment of hemangiomas of the headand neck. Laryngoscope. 1992;102(10):1123–1132

277. Waner M, Suen JY, Dinehart S, MallorySB. Laser photocoagulation ofsuperficial proliferating hemangiomas.J Dermatol Surg Oncol. 1994;20(1):43–46

278. Anderson RR, Parrish JA. Selectivephotothermolysis: precisemicrosurgery by selective absorption ofpulsed radiation. Science. 1983;220(4596):524–527

279. Alora MB, Anderson RR. Recentdevelopments in cutaneous lasers.Lasers Surg Med. 2000;26(2):108–118

280. Batta K, Goodyear HM, Moss C, WilliamsHC, Hiller L, Waters R. Randomisedcontrolled study of early pulsed dyelaser treatment of uncomplicatedchildhood haemangiomas: results ofa 1-year analysis. Lancet. 2002;360(9332):521–527

281. Kolde G. Early pulsed-dye lasertreatment of childhood haemangiomas.Lancet. 2003;361(9354):348–349; authorreply: 349

282. Rizzo C, Brightman L, Chapas AM, et al.Outcomes of childhood hemangiomastreated with the pulsed-dye laser withdynamic cooling: a retrospective chartanalysis. Dermatol Surg. 2009;35(12):1947–1954

283. Flick RP, Katusic SK, Colligan RC, et al.Cognitive and behavioral outcomesafter early exposure to anesthesia andsurgery. Pediatrics. 2011;128(5).

Available at: www.pediatrics.org/cgi/content/full/128/5/e1053

284. Smit JM, Bauland CG, Wijnberg DS,Spauwen PH. Pulsed dye lasertreatment, a review of indications andoutcome based on published trials. Br JPlast Surg. 2005;58(7):981–987

285. Anderson RR. Infant hemangiomas:a controversy worth solving. LasersSurg Med. 2006;38(2):92–93

286. Wareham WJ, Cole RP, Royston SL,Wright PA. Adverse effects reported inpulsed dye laser treatment for portwine stains. Lasers Med Sci. 2009;24(2):241–246

287. Mulliken JB, Fishman SJ, Burrows PE.Vascular anomalies. Curr Probl Surg.2000;37(8):517–584

288. Greene AK. Management ofhemangiomas and other vasculartumors. Clin Plast Surg. 2011;38(1):45–63

289. Arneja JS, Mulliken JB. Resection ofamblyogenic periocular hemangiomas:indications and outcomes. PlastReconstr Surg. 2010;125(1):274–281

290. Arneja JS, Chim H, Drolet BA, Gosain AK.The Cyrano nose: refinements insurgical technique and treatmentapproach to hemangiomas of the nasaltip. Plast Reconstr Surg. 2010;126(4):1291–1299

291. Soltani AM, Reinisch JF. Algorithmicapproach to the management ofhemangiomas. J Craniofac Surg. 2011;22(2):585–588

292. Li WY, Chaudhry O, Reinisch JF. Guide toearly surgical management of liphemangiomas based on our experienceof 214 cases. Plast Reconstr Surg. 2011;128(5):1117–1124

293. Mulliken JB, Rogers GF, Marler JJ.Circular excision of hemangioma andpurse-string closure: the smallestpossible scar. Plast Reconstr Surg.2002;109(5):1544–1554; discussion:1555

294. Haik BG, Jakobiec FA, Ellsworth RM,Jones IS. Capillary hemangioma of thelids and orbit: an analysis of the clinicalfeatures and therapeutic results in 101cases. Ophthalmology. 1979;86(5):760–792

295. Robb RM. Refractive errors associatedwith hemangiomas of the eyelids and

orbit in infancy. Am J Ophthalmol. 1977;83(1):52–58

296. Stigmar G, Crawford JS, Ward CM,Thomson HG. Ophthalmic sequelae ofinfantile hemangiomas of the eyelidsand orbit. Am J Ophthalmol. 1978;85(6):806–813

297. von Noorden GK. Application of basicresearch data to clinical amblyopia.Ophthalmology. 1978;85(5):496–504

298. Schwartz SR, Kodsi SR, Blei F, Ceisler E,Steele M, Furlan L. Treatment ofcapillary hemangiomas causingrefractive and occlusional amblyopia.J AAPOS. 2007;11(6):577–583

299. Kushner BJ. Intralesional corticosteroidinjection for infantile adnexalhemangioma. Am J Ophthalmol. 1982;93(4):496–506

300. Morrell AJ, Willshaw HE. Normalisationof refractive error after steroidinjection for adnexal haemangiomas. BrJ Ophthalmol. 1991;75(5):301–305

301. Plager DA, Snyder SK. Resolution ofastigmatism after surgical resection ofcapillary hemangiomas in infants.Ophthalmology. 1997;104(7):1102–1106

302. Ni N, Guo S, Langer P. Current conceptsin the management of periocularinfantile (capillary) hemangioma. CurrOpin Ophthalmol. 2011;22(5):419–425

303. Claerhout I, Buijsrogge M, Delbeke P,et al. The use of propranolol in thetreatment of periocular infantilehaemangiomas: a review. Br JOphthalmol. 2011;95(9):1199–1202

304. Al Dhaybi R, Superstein R, Milet A, et al.Treatment of periocular infantilehemangiomas with propranolol: caseseries of 18 children. Ophthalmology.2011;118(6):1184–1188

305. Ceisler EJ, Santos L, Blei F. Periocularhemangiomas: what every physicianshould know. Pediatr Dermatol. 2004;21(1):1–9

306. Bullock JD, Warwar RE, Green WR.Ocular explosion during cataractsurgery: a clinical, histopathological,experimental, and biophysical study.Trans Am Ophthalmol Soc. 1998;96(12):243–276; discussion: 276–281

307. Cogen MS, Elsas FJ. Eyeliddepigmentation followingcorticosteroid injection for infantileocular adnexal hemangioma. J Pediatr





Ophthalmol Strabismus. 1989;26(1):35–38

308. Vazquez-Botet R, Reyes BA, Vazquez-Botet M. Sclerodermiform linearatrophy after the use of intralesionalsteroids for periorbital hemangiomas:a review of complications. J PediatrOphthalmol Strabismus. 1989;26(3):124–127

309. Droste PJ, Ellis FD, Sondhi N, HelvestonEM. Linear subcutaneous fat atrophyafter corticosteroid injection ofperiocular hemangiomas. Am JOphthalmol. 1988;105(1):65–69

310. Sutula FC, Glover AT. Eyelid necrosisfollowing intralesional corticosteroidinjection for capillary hemangioma.Ophthalmic Surg. 1987;18(2):103–105

311. Chambers CB, Katowitz WR, Katowitz JA,Binenbaum G. A controlled study oftopical 0.25% timolol maleate gel forthe treatment of cutaneous infantilecapillary hemangiomas. Ophthal PlastReconstr Surg. 2012;28(2):103–106

312. Schneider D, Lee MS, Harrison AR,Sidman J. Excision of periorbitalhemangiomas to correct visualabnormalities. Arch Facial Plast Surg.2011;13(3):195–198

313. Deans RM, Harris GJ, Kivlin JD. Surgicaldissection of capillary hemangiomas:an alternative to intralesionalcorticosteroids. Arch Ophthalmol. 1992;110(12):1743–1747

314. Perkins JA, Duke W, Chen E, Manning S.Emerging concepts in airway infantilehemangioma assessment andmanagement. Otolaryngol Head NeckSurg. 2009;141(2):207–212

315. Perkins JA, Oliaei S, Garrison MM,Manning SC, Christakis DA. Airwayprocedures and hemangiomas:treatment patterns and outcome in U.S.pediatric hospitals. Int J PediatrOtorhinolaryngol. 2009;73(9):1302–1307

316. Shikhani AH, Jones MM, Marsh BR,Holliday MJ. Infantile subglottichemangiomas. An update. Ann OtolRhinol Laryngol. 1986;95(4 pt 1):336–347

317. Bitar MA, Moukarbel RV, Zalzal GH.Management of congenital subglottichemangioma: trends and success overthe past 17 years. Otolaryngol HeadNeck Surg. 2005;132(2):226–231

318. Sherrington CA, Sim DK, Freezer NJ,Robertson CF. Subglottic haemangioma.Arch Dis Child. 1997;76(5):458–459

319. Sie KC, McGill T, Healy GB. Subglottichemangioma: ten years’ experiencewith the carbon dioxide laser. Ann OtolRhinol Laryngol. 1994;103(3):167–172

320. Longstreet B, Bhama PK, Inglis AF Jr,Saltzman B, Perkins JA. Improvedairway visualization during directlaryngoscopy using self-retaininglaryngeal retractors: a quantitativestudy. Otolaryngol Head Neck Surg.2011;145(2):270–275

321. Parhizkar N, Manning SC, Inglis AF Jr,Finn LS, Chen EY, Perkins JA. Howairway venous malformations differfrom airway infantile hemangiomas.Arch Otolaryngol Head Neck Surg. 2011;137(4):352–357

322. Drolet BA, Dohil M, Golomb MR, et al.Early stroke and cerebral vasculopathyin children with facial hemangiomasand PHACE association. Pediatrics.2006;117(3):959–964

323. Badi AN, Kerschner JE, North PE, DroletBA, Messner A, Perkins JA.Histopathologic and immunophenotypicprofile of subglottic hemangioma:multicenter study. Int J PediatrOtorhinolaryngol. 2009;73(9):1187–1191

324. Zur KB, Wood RE, Elluru RG. Pediatricpostcricoid vascular malformation:a diagnostic and treatment challenge.Int J Pediatr Otorhinolaryngol. 2005;69(12):1697–1701

325. Feuerstein SS. Subglottic hemangiomain infants. Laryngoscope. 1973;83(4):466–475

326. Truong MT, Perkins JA, Messner AH,Chang KW. Propranolol for thetreatment of airway hemangiomas:a case series and treatment algorithm.Int J Pediatr Otorhinolaryngol. 2010;74(9):1043–1048

327. Hoeve LJ, Küppers GL, Verwoerd CD.Management of infantile subglottichemangioma: laser vaporization,submucous resection, intubation, orintralesional steroids? Int J PediatrOtorhinolaryngol. 1997;42(2):179–186

328. Sierpina DI, Chaudhary HM, Walner DL,Aljadeff G, Dubrow IW. An infantilebronchial hemangioma unresponsive topropranolol therapy: case report and

literature review. Arch OtolaryngolHead Neck Surg. 2011;137(5):517–521

329. Vijayasekaran S, White DR, Hartley BE,Rutter MJ, Elluru RG, Cotton RT. Openexcision of subglottic hemangiomas toavoid tracheostomy. Arch OtolaryngolHead Neck Surg. 2006;132(2):159–163

330. Rahbar R, Nicollas R, Roger G, et al. Thebiology and management of subglottichemangioma: past, present, future.Laryngoscope. 2004;114(11):1880–1891

331. Mistry N, Tzifa K. Use of propranolol totreat multicentric airwayhaemangioma. J Laryngol Otol. 2010;124(12):1329–1332

332. Maturo S, Hartnick C. Initial experienceusing propranolol as the sole treatmentfor infantile airway hemangiomas. Int JPediatr Otorhinolaryngol. 2010;74(3):323–325

333. Buckmiller LM, Munson PD,Dyamenahalli U, Dai Y, Richter GT.Propranolol for infantile hemangiomas:early experience at a tertiary vascularanomalies center. Laryngoscope. 2010;120(4):676–681

334. Manunza F, Syed S, Laguda B, et al.Propranolol for complicated infantilehaemangiomas: a case series of 30infants. Br J Dermatol. 2010;162(2):466–468

335. Jephson CG, Manunza F, Syed S, MillsNA, Harper J, Hartley BE. Successfultreatment of isolated subglottichaemangioma with propranolol alone.Int J Pediatr Otorhinolaryngol. 2009;73(12):1821–1823

336. Denoyelle F, Leboulanger N, Enjolras O,Harris R, Roger G, Garabedian EN. Roleof propranolol in the therapeuticstrategy of infantile laryngotrachealhemangioma. Int J PediatrOtorhinolaryngol. 2009;73(8):1168–1172

337. Parikh SR, Darrow DH, Grimmer JF,Manning SC, Richter GT, Perkins JA.Propranolol use for infantilehemangiomas: American Society ofPediatric Otolaryngology VascularAnomalies Task Force practice patterns.JAMA Otolaryngol Head Neck Surg.2013;139(2):153–156

338. Denoyelle F, Garabédian EN. Propranololmay become first-line treatment inobstructive subglottic infantilehemangiomas. Otolaryngol Head NeckSurg. 2010;142(3):463–464



339. Waner M, Suen JY. Treatment optionsfor the management of hemangiomas.In: Waner M, Suen JY, eds.Hemangiomas and VascularMalformations of the Head and Neck.New York, NY: Wiley-Liss Inc; 1999:233–262

340. Pransky SM, Canto C. Management ofsubglottic hemangioma. Curr OpinOtolaryngol Head Neck Surg. 2004;12(6):509–512

341. Cotton RT, Tewfik TL. Laryngeal stenosisfollowing carbon dioxide laser insubglottic hemangioma: report of threecases. Ann Otol Rhinol Laryngol. 1985;94(5 pt 1):494–497

342. Chatrath P, Black M, Jani P, Albert DM,Bailey CM. A review of the currentmanagement of infantile subglottichaemangioma, including a comparisonof CO2 laser therapy versustracheostomy. Int J PediatrOtorhinolaryngol. 2002;64(2):143–157

343. Froehlich P, Stamm D, Floret D, MorgonA. Management of subglottichaemangioma. Clin Otolaryngol AlliedSci. 1995;20(4):336–339

344. Bajaj Y, Hartley BEJ, Wyatt ME, AlbertDM, Bailey CM. Subglottic haemangiomain children: experience with opensurgical excision. J Laryngol Otol. 2006;120(12):1033–1037

345. O-Lee TJ, Messner A. Open excision ofsubglottic hemangioma withmicroscopic dissection. Int J PediatrOtorhinolaryngol. 2007;71(9):1371–1376

346. Waner M, Kastenbaum J, Scherer K.Hemangiomas of the nose: surgicalmanagement using a modified subunitapproach. Arch Facial Plast Surg. 2008;10(5):329–334

347. Thomson HG, Lanigan M. The Cyranonose: a clinical review of hemangiomasof the nasal tip. Plast Reconstr Surg.1979;63(2):155–160

348. Koopmann CF Jr. The “Pinocchio” nasaldeformity—hemangioma vs.angiolipoma: esthetic correction andetiology. J Otolaryngol. 1988;17(4):169–172

349. Burgos L, López Gutiérrez JC, AndrésAM, et al. Early surgical treatment innasal tip hemangiomas: 36 casesreview [in Spanish]. Cir Pediatr. 2007;20(2):83–86

350. Hochman M, Mascareno A. Managementof nasal hemangiomas. Arch FacialPlast Surg. 2005;7(5):295–300

351. Perkins JA, Chen BS, Saltzman B,Manning SC, Parikh SR. Propranololtherapy for reducing the number ofnasal infantile hemangioma invasiveprocedures. JAMA Otolaryngol HeadNeck Surg. 2014;140(3):220–227

352. Eivazi B, Cremer HJ, Mangold C,Teymoortash A, Wiegand S, Werner JA.Hemangiomas of the nasal tip: anapproach to a therapeutic challenge. IntJ Pediatr Otorhinolaryngol. 2011;75(3):368–375

353. Faguer K, Dompmartin A, Labbé D,Barrellier MT, Leroy D, Theron J. Earlysurgical treatment of Cyrano-nosehaemangiomas with Rethi incision. Br JPlast Surg. 2002;55(6):498–503

354. Pitanguy I, Machado BH, Radwanski HN,Amorim NF. Surgical treatment ofhemangiomas of the nose. Ann Plast Surg.1996;36(6):586–592; discussion: 592–593

355. Demiri EC, Pelissier P, Genin-EtcheberryT, Tsakoniatis N, Martin D, Baudet J.Treatment of facial haemangiomas: thepresent status of surgery. Br J PlastSurg. 2001;54(8):665–674

356. McCarthy JG, Borud LJ, Schreiber JS.Hemangiomas of the nasal tip. PlastReconstr Surg. 2002;109(1):31–40

357. Jackson IT, Sosa J. Excision of nasal tiphemangioma via open rhinoplasty:a skin sparing technique. Eur J PlastSurg. 1998;21(5):265–268

358. Chang CS, Wong A, Rohde CH,Ascherman JA, Wu JK. Management oflip hemangiomas: minimizing peri-oralscars. J Plast Reconstr Aesthet Surg.2012;65(2):163–168

359. Waner M, Suen JY. The treatment ofhemangiomas: hemangiomas requiringspecial consideration. In: Waner M,

Suen JY, eds. Hemangiomas andVascular Malformations of the Headand Neck. New York, NY: Wiley-Liss;1999:293–307

360. Zide BM, Glat PM, Stile FL, Longaker MT.Vascular lip enlargement: part I.Hemangiomas—tenets of therapy. PlastReconstr Surg. 1997;100(7):1664–1673

361. Bouchard S, Yazbeck S, Lallier M.Perineal hemangioma, anorectalmalformation, and genital anomaly:a new association? J Pediatr Surg.1999;34(7):1133–1135

362. Kulungowski AM, Alomari AI, Chawla A,Christison-Lagay ER, Fishman SJ.Lessons from a liver hemangiomaregistry: subtype classification.J Pediatr Surg. 2012;47(1):165–170

363. Rialon KL, Murillo R, Fevurly RD,Kulungowski AM, Christison-Lagay ER,Zurakowski D, et al. Risk factors formortality in patients with multifocaland diffuse hepatic hemangiomas. JPediatr Surg. 2015;50(5):837–41.

364. Horii KA, Drolet BA, Frieden IJ, et al;Hemangioma Investigator Group.Prospective study of the frequency ofhepatic hemangiomas in infants withmultiple cutaneous infantilehemangiomas. Pediatr Dermatol. 2011;28(3):245–253

365. Huang SA, Tu HM, Harney JW, et al.Severe hypothyroidism caused by type3 iodothyronine deiodinase in infantilehemangiomas. N Engl J Med. 2000;343(3):185–189

366. Mhanna A, Franklin WH, Mancini AJ.Hepatic infantile hemangiomas treatedwith oral propranolol—a case series.Pediatr Dermatol. 2011;28(1):39–45

367. Mazereeuw-Hautier J, Hoeger PH,Benlahrech S, et al. Efficacy ofpropranolol in hepatic infantilehemangiomas with diffuse neonatalhemangiomatosis. J Pediatr. 2010;157(2):340–342

368. Luu M, Frieden IJ. Haemangioma:clinical course, complications andmanagement. Br J Dermatol. 2013;169(1):20–30



DOI: 10.1542/peds.2015-2485; originally published online September 28, 2015;Pediatrics

AND NECK SURGERY, and SECTION ON PLASTIC SURGERYHEAD−SECTION ON DERMATOLOGY, SECTION ON OTOLARYNGOLOGY

David H. Darrow, Arin K. Greene, Anthony J. Mancini, Amy J. Nopper and theDiagnosis and Management of Infantile Hemangioma

ServicesUpdated Information &

/peds.2015-2485http://pediatrics.aappublications.org/content/early/2015/09/22including high resolution figures, can be found at:

Subspecialty Collections

gy_subhttp://pediatrics.aappublications.org/cgi/collection/dermatoloDermatologythe 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 © 2015 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

at Miami Children's Hospital on September 28, 2015pediatrics.aappublications.orgDownloaded from

http://pediatrics.aappublications.org/content/early/2015/09/22/peds.2015-2485


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

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

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

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

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


DOI: 10.1542/peds.2015-2485; originally published online September 28, 2015;Pediatrics

AND NECK SURGERY, and SECTION ON PLASTIC SURGERYHEAD−SECTION ON DERMATOLOGY, SECTION ON OTOLARYNGOLOGY

David H. Darrow, Arin K. Greene, Anthony J. Mancini, Amy J. Nopper and theDiagnosis and Management of Infantile Hemangioma


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 © 2015 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 Miami Children's Hospital on September 28, 2015pediatrics.aappublications.orgDownloaded from



diagnosis and management of infantile hemangioma

Documents