State of the Literature

We identified 148 unique studies (15 randomized controlled trials [RCTs], 5 prospective and 19 retrospective cohort studies, 2 diagnostic accuracy studies, 1 study comparing pairs of treated and untreated infantile hemangioma (IH), and 106 case series) addressing our Key Questions. Forty-two comparative studies reported effectiveness outcomes (6 good quality, 22 fair quality, and 14 poor quality). One-hundred and forty-four studies (comparative studies and case series) reported harms/adverse events data (14 good quality for harms reporting, 3 fair quality, 127 poor quality). Eighty-one studies addressed beta-blockers (13 of which compared a beta-blocker to another category of intervention such as corticosteroids or laser); 26 addressed lasers; 24 addressed steroids; 15 addressed surgical approaches; and 2 addressed diagnostic modalities.

The literature on pharmacologic and surgical approaches for the treatment of IH is heterogeneous in terms of populations, interventions, comparators, and outcomes. Comparative studies included individuals with ages of less than one month to over 40 years (though the mean age in this study was below 3 years), and lesion types and locations varied across studies. Most studies included children with IH in multiple anatomic locations and of multiple types (e.g., deep, superficial) without stratifying outcomes on these characteristics. Studies typically did not clearly describe diagnostic criteria, and few clearly noted whether prior treatment had been administered (n=11/42 comparative studies).

Studies assessed varied pharmacologic agents (corticosteroids, beta-blockers, immunomodulators) administered through various routes (topical, intralesional, intravenous, oral) at multiple doses and durations as well as varied forms of laser and surgical treatment (e.g., pulsed dye laser [PDL], argon laser, neodymium yttrium aluminum garnet [Nd:YAG] laser, cryotherapy) using varied regimens. Few (n=2) comparative studies addressed surgical treatment aside from laser modalities. Comparators also varied across studies and included placebo, observation, historical control groups, and other active interventions. Outcome measures similarly differed. While studies generally assessed change in lesion size or appearance, scales and methods varied and included visual analog scales, assessment of percentage size change, and more subjective assessments of good, fair, or poor response.

Summary of Key Findings

Key Findings From Key (Comparative Effectiveness) Questions

Until fairly recently, corticosteroids were the treatment of choice for IH. As reported in this review, corticosteroids demonstrate some effectiveness but are associated with clinically significant side effects. More recently, beta-blockers, and propranolol specifically, have been studied and recommended for use. Studies of propranolol have compared its effectiveness to placebo/observation, to corticosteroids and other modalities, and to other beta-blockers. Relative to observation or placebo arms, oral propranolol has been consistently shown to be superior in individual studies and in our network meta-analysis. Relative to other modalities, including steroids and bleomycin, we find that propranolol is generally superior with the exception of no significant differences in reducing lesion size in two studies comparing it to steroids. Finally, given that propranolol has been demonstrated to be associated with positive outcomes, the question of whether effectiveness is associated with propranolol specifically or beta-blockers in general has been studied. Although there are only three small studies available, early results are as positive as those noted for propranolol, and we believe that they suggest that these and potentially other beta-blockers may also be effective, potentially with fewer side effects. These findings, however, are preliminary. Studies of the beta-blocker timolol, used as a topical gel or solution typically to treat superficial IH, also reported greater effectiveness for timolol compared with placebo/observation in reducing IH lesion size, no differences in effects in one study comparing ophthalmic timolol and imiquimod; no differences in average overall improvement in another study comparing timolol and laser modalities; and greater response to timolol in superficial IH with greater response of mixed IH to timolol plus laser in a fourth study.

In our network meta-analysis specifically, the expected efficacy of control arms was estimated to be 6 percent (95% Bayesian credible interval [BCI]=1% to 11%). All non-control treatments were estimated to have a larger expected clearance than control arms. The largest mean estimate of expected clearance was for oral propranolol (95%, 95% BCI: 88% to 99%), followed by timolol (62%, 95% BCI: 39% to 83%) and intralesional triamcinolone (58%, 95% BCI: 22% to 93%), albeit with wider confidence bounds. Oral steroids had a clearance rate of 43 percent (95% BCI: 21% to 66%). The preponderance of available evidence used in the network meta-analysis was derived from studies of propranolol and corticosteroids.

In terms of surgical interventions, only laser has been adequately studied. Most studies focused on PDL and generally it was found to be more effective than other types of laser, but effects remain unclear as studies were heterogeneous, and the role of laser vis-a-vis beta-blockers is not clearly described in the literature. Data are inadequate to address the role of imaging in guiding treatment.

We review specific findings and strength of evidence (SOE) by Key Question and provide more detailed results from our network meta-analysis below.

KQ1. Effectiveness and Harms of Imaging

Two poor quality diagnostic accuracy studies addressed imaging modalities.^68,70 Studies assessed IH in different anatomic locations and reported differing findings for the sensitivity of ultrasound and effectiveness of imaging modalities depending on location or subtype. Studies were limited by the size of cohorts, lack of standard processes, and lack of direct comparison at the same time point using the various imaging modalities.

We considered the SOE for all imaging modalities to be insufficient given single, small studies addressing different approaches (Table 34) using weaker study designs and precluding a meta-analysis. The studies did not address harms.

Table 34

Strength of evidence for effectiveness of imaging modalities.

KQ2. Effectiveness and Harms of Corticosteroids and Beta-Blockers

Effectiveness and Harms of Corticosteroids

We identified 24 studies (3 RCTs, 1 cohort study, and 20 case series) reporting outcomes and/or harms following corticosteroid use in children with IH.^{40,107–129,133} In addition, seven studies (described in the section on beta-blockers) compared beta-blockers and steroids.^{96–98,100,130–133} Steroids studied varied in dose, type, and route of administration, and the ages of children included in comparative studies ranged widely from 1 to 72 months. Children in treatment arms typically had improvement in lesion size. Of the 219 children who received steroids in three studies^108,122,252 reporting lesion change data, 140 had a “good” or “fair” response to steroids. In our network meta-analysis, oral steroids had a mean estimated expected clearance rate of 43 percent (95% BCI: 21% to 66%). Intralesional triamcinolone had a rate of 58 percent but with wide confidence bounds (95% BCI: 22% to 93%).

Thus, there is adequate evidence to support a moderate strength of evidence for oral steroids to have a modest effect on clearance rates and low SOE for intralesional steroids to have a modest (albeit larger) effect relative to control with wide confidence bounds.

However, steroids were consistently associated with clinically important harms including Cushingoid appearance, infection, growth retardation, hypertension, and mood changes that may be important in making treatment decisions. The SOE is moderate for the association of steroids with these clinically important harms (Table 35).

Table 35

Strength of evidence for effectiveness and harms of steroids.

Effectiveness and Harms of Beta-Blockers

Eighty-one studies (25 comparative studies and 56 case series) evaluated propranolol (oral, topical, intralesional), oral nadolol, oral atenolol, or timolol (topical gel or ophthalmic solution). Beta-blockers typically demonstrated significantly greater effects on reducing lesion size or volume than did control or other active comparators.

Compared with a mean estimated expected clearance rate of 6 percent (95% BCI: 1% to 11%) in placebo or observation arms, oral propranolol had a rate of 95 percent (95% BCI: 88% to 99%). With adequate data and precision, we considered the SOE to be high for the effect of oral propranolol on lesion size relative to observation or placebo arms. Individual studies assessed qualitatively typically also demonstrated greater effectiveness for propranolol compared with other active treatments.

Other oral beta-blockers have demonstrated promising effectiveness; we considered the SOE to be low for no difference in response of propranolol and nadolol or atenolol based on three small studies. We considered SOE to be low for greater effectiveness of topical timolol compared with observation or placebo (Table 36); SOE was insufficient for studies comparing timolol to other modalities including laser and imiquimod. Most studies of timolol included children with superficial lesions.

Table 36

Strength of evidence for effectiveness of beta-blockers.

Harms most frequently reported with beta-blockers included hypotension, hypoglycemia, bradycardia, sleep disturbances, cold extremities, gastrointestinal symptoms, and bronchial irritation (classified as hyperreactivity, bronchospasm, bronchiolitis, cold induced wheezing). Harms generally were not severe enough to cause treatment discontinuation (n=75/4872 children receiving beta-blockers [1.5%]) in case series and comparative studies). We considered the SOE to be moderate for the association of propranolol with clinically important and minor harms (Table 37).

Table 37

Strength of evidence for harms of beta-blockers.

KQ3. Effectiveness and Harms of Second-Line Drugs

We did not identify any studies addressing this question.

KQ4. Effectiveness and Harms of Surgical Interventions

Effectiveness and Harms of Laser and Surgical Treatment

Eleven comparative studies (three RCTs,^210–212 seven retrospective cohort studies,^213–219 and one study that compared cryotherapy-treated and untreated IH pairs in individual children²²⁰) addressed surgical approaches. In addition, one RCT and one cohort study (described in KQ2 above) compared topical timolol and laser modalities,^14,106 and 28 case series addressed surgical approaches.^{221–228,230–249} Most comparative studies were small (≤55 participants), but one RCT and three retrospective cohort studies included more than 120 children. Lasers varied across studies in type, pulse width, or cooling materials. Most studies assessed variations of PDL (n=7) and examined heterogeneous endpoints. Most studies reported on treatment of cutaneous lesions.

Overall, longer pulse PDL with epidermal cooling was the most commonly used laser for cutaneous lesions and Nd:YAG was the most commonly used intralesionally. Most studies reported a higher success rate with longer pulse PDL compared to observation in managing the size of IH, although the magnitude of effect differed substantially. CO₂ laser was used for subglottic IH in a single study, and was noted to have a higher success rate and lower complication rate than both Nd:YAG and observation.

Two comparative studies addressed surgical approaches (cryotherapy, intense pulsed light photothermolysis, sclerosis) and reported some positive effects in reducing IH size or improving appearance, but their smaller size and low quality preclude conclusions (insufficient SOE). Strength of evidence for outcomes after surgical treatments ranged from insufficient to low for effectiveness outcomes. The evidence was limited by low sample size, lack of comparisons of the same modalities, and variations in the laser settings used including wavelength and cooling protocols. For Nd:YAG and CO₂ lasers, cryotherapy, and intense pulsed light photothermolysis, all studies were severely limited by sample size, and SOE was determined to be insufficient in all outcome parameters (Table 38).

Table 38

Strength of evidence for effectiveness of laser modalities.

For harms, a moderate strength of evidence was noted for pigmentation changes with PDL, which was most frequently hypopigmentation. Low SOE was noted for bleeding in the immediate postoperative period. Due to low sample size and limitations in reporting, pain and scarring were found to have insufficient SOE. For Nd:YAG lasers, evaluation for scarring was most frequently reported, and there was low SOE to support no difference in scarring between Nd:YAG and observation. Evidence was deemed insufficient to comment on pigmentation changes and bleeding for children treated with Nd:YAG and scarring after cryotherapy.

Most surgical case series (n=13) were retrospective and included a total of 838 children. We considered all to be poor quality for harms reporting. Frequently reported harms included scarring and wound dehiscence. SOE was insufficient for the association of surgical approaches with harms given the small numbers of harms reported (Table 39).

Table 39

Strength of evidence for harms of laser modalities.

Findings in Relation to What is Already Known

We identified ten recent (2010-present) systematic review or meta-analyses assessing interventions for IH.^91,253–261 Most reviews addressed propranolol or beta-blockers: three addressed propranolol generally;^258,259,261 two examined effectiveness specifically for airway IH;^91,260 one for periocular IH;²⁶² and two compared beta-blockers and steroids.^253,254 One Cochrane review assessed multiple interventions,²⁵⁷ and two additional reviews examined intralesional steroids²⁵⁶ and laser treatment.²⁵⁵

Across reviews, investigators commented on small sample sizes, disparate outcome measures, and typically low to moderate quality studies. Most reviews noted the promise of propranolol for reducing IH lesion size but also a need for additional, larger studies with longer term followup. Overall, our findings related to the effectiveness of propranolol in most children and limited effectiveness of steroids for cutaneous IH align with findings in prior reviews. One review and meta-analysis of 10 comparative studies (six considered high quality, four of moderate quality) of children with cutaneous IH meta-analyzed data related to adverse events and reported no differences in the rate of adverse events between propranolol and corticosteroids (18 events in propranolol studies and 19 in steroid, p=0.73, 95% CI: 0.56 to 1.50).²⁵³

Only one prior review addressed laser treatments (two IH studies) and concluded that, despite favorable results, the evidence is weak to support the use of lasers in IH treatment (level 3b on the Oxford Centre of Evidence-based Medicine scale).²⁵⁵

Applicability

We set inclusion criteria intended to identify studies with applicability to children with IH between the ages of 0 and 18 years. Studies differed in terms of study population and outcome measures. Most studies included children with IH in multiple anatomic locations and did not report effectiveness by lesion site or type. Most studies were non-comparative, and lack of direct comparisons of treatment options and few studies addressing the same interventions and comparators further hinder our ability to understand what findings will best extrapolate to children at specific ages, with specific lesion types, or in specific anatomic locations. Further, most comparative studies were conducted in larger medical centers or referral centers, which is in line with typical treatment as most children with IH are referred to specialists from general practitioners.

Overall the available data on the effectiveness and harms of beta-blockers and corticosteroids are largely applicable to the general population of children with IH. Most studies included a majority of females, in line with the female predominance of IH, and ages in comparative studies generally ranged from 1 month to 9 years. One cohort study included individuals between 1 month and 43 years of age, with a mean age of 2 years and 11 months.²¹⁶

Few studies addressed imaging modalities, and those that did evaluated modalities to assess hepatic or intraspinal IH. Studies compared ultrasound, magnetic resonance imaging, computed tomography, and angiography. Imaging was sometimes not conducted at the same time, which limits comparability, and potentially the applicability of findings. Studies were also completed prior to 2010, so imaging techniques and practices may have changed.

Studies addressing steroids compared various routes of steroid administration (oral, topical, and intralesional) and various agents (methylprednisolone, triamcinolone, mometasone furoate) in children with ages ranging from less than 1 to 72 months. Studies likely included children with IH in the proliferative and involution phase, which may limit applicability to younger or older children. One comparative study was conducted in Canada and the others in Turkey, Pakistan, and India. Applicability may be limited given differences in the systems of care in these lower resource countries. Comparative studies were also published between 2001 and 2014 and may not fully represent evolutions in standards of care.

Studies of beta-blockers typically included infants of both sexes ages 1 to 12 months of age (range: 1 month – 9 years) with superficial, deep, and mixed lesions primarily involving the head and neck and occurring as focal or segmental lesions. Studies of topical or ophthalmic timolol typically included children with superficial lesions, though two of six comparative studies included children with superficial and deep lesions. Children were treated with a variety of beta-blockers including propranolol at various doses and administrations (oral, intralesional, or topical), timolol (topical or ophthalmic), atenolol (oral), or nadolol (oral), most commonly for up to 6 months duration. These agents and dosage forms are typically easily available in the United States and not universally available. Dosage amounts ranged from 1 to 4 mg/kg/day. Doses over 2 mg/kg/day are not typically administered and may limit applicability of findings of two studies of propranolol.^92,97

Surgical studies, conducted in the United States, the United Kingdom, the Netherlands, Germany, Greece, Japan and Singapore, included infants of both sexes with a preponderance of females (age range: 1 week to 43 years of age) with superficial and cutaneous infantile hemangiomas in varied locations. One study reported laser use for subglottic IH and one evaluated photothermolysis with intense pulsed light and cryosurgery in children of maxillary IH. Most comparative studies evaluated laser treatments including short-pulse and longer pulse PDL, Nd:YAG, and argon. Two studies evaluated cryotherapy, one of which compared it to photothermolysis with intense pulsed light with or without concomitant sclerosis. Applicability of many of these studies is limited by historical changes in care and technology.

Newer lasers and adjunctive features such as dynamic cooling have resulted in older lasers being out of date, thus limiting the applicability of studies conducted with those models. Most laser studies evaluated lasers as first-line treatment, which is currently less common in practice since the advent of beta-blocker treatment in countries, like the United States, where such treatments are readily available, as beta-blockers have generally superseded other treatments as first-line management of IH. Additionally, most comparative literature evaluated PDL, which is typically used only for the treatment of superficial lesions. Appendix G contains full applicability tables.

Implications for Clinical and Policy Decisionmaking

This review provides evidence for use in clinical care of children who present with IH. It particularly demonstrates that there are moderate benefits with steroid treatment and significantly greater improvements with beta-blockers, with propranolol being the agent most commonly studied. When a decision to treat is made, our review provides qualitative and quantitative evidence that beta-blockers are associated with substantial improvement in IH size/volume (mean expected clearance rates of 95% for oral propranolol [95% BCI: 88% to 99%] and 62% for topical timolol [95% BCI: 39% to 83%], compared with 6% for observation/placebo arms [95% BCI: 1% to 11%]).

Steroids were associated with mean expected clearance rates of 43 percent (95% BCI: 21% to 66%) for oral steroids and 58 percent (95% BCI: 22% to 99%) for intralesional triamcinolone in our network meta-analysis, but side effects are clinically significant, and clinicians and families will need to weigh the benefits and harms.

It is important for clinicians to know that the literature summarized here typically examines children with problematic or complicated IH and thus may not apply to all children, particularly those with minor IH. In one large trial evaluating active treatment with propranolol for children without problematic IH, propranolol was associated with complete resolution or near complete resolution in 60 percent of cases (vs. 4% in placebo arm).⁹² In addition, studies typically reported outcomes only in the short term (generally ≤12 months followup); thus, our understanding of the longer term effects of these medications is lacking. Further, though the literature demonstrates a strong shift towards beta-blocker therapy, uncertainty still remains about the most effective agent, dosage, and duration of treatment, and the need for pre-treatment evaluation and monitoring while on beta-blockers.

Limited research is available to guide decision-making about the use of lasers as the initial intervention. Historically, lasers provided a fair benefit in primary management of IH, which was comparable in many cases series to steroid treatment, and generally was superior to observation. The advent of propranolol, however, has largely relegated laser treatment to secondary management. There is little comparative data between lasers and beta-blockers, but the success rates for complete or near complete resolution in historical laser studies are notably lower than those in more recent propranolol studies. Under current treatment paradigms, PDL with epidermal cooling is most often used for residual cutaneous changes after the completion of the proliferative growth phase and with incomplete resolution after pharmacologic management, while Nd:YAG laser is most often used intralesionally for medically refractory lesions. A variety of other lasers are used for intralesional treatment or resection, though no conclusions can be drawn regarding the superiority of any of these modalities over any other.

The literature identified to answer contextual questions describes a broader range of indications for referral of patients with IH and suggests that indications for referral include large size; segmental type; risk for complications including bleeding, ulceration, and pain; involvement of critical structures; and risk factors for occult lesions (numerous cutaneous lesions, beard distribution). Further, the potential for psychosocial concerns may support referral for patients with uncomplicated lesions in highly visible areas on a case-by-case basis.

Given the lack of long-term data on harms of interventions, clinicians and families must balance the potential of both short- and long-term harms with the benefits of potential resolution or size reduction of lesions.

Limitations of the Comparative Effectiveness Review Process

We included studies published in English only and did not seek or include unpublished data. In our scan of the non-English language literature published since 1982 and located via our MEDLINE search, we determined that the majority would not meet our review criteria. Given the high percentage of non-eligible items in this scan, we feel that excluding non-English studies did not introduce significant bias into the review.

We also required that studies reporting on “second-line” treatments such as imiquimod, bleomycin, or alpha interferon address such treatments after a trial of beta-blockers or corticosteroids, and we did not identify any such studies. While this undoubtedly means that some treatment outcomes are not included in this review, these drugs are not frequently used since the advent of beta-blocker treatment for IH in the opinion of our clinical experts.

We also used only comparative studies to address questions of effectiveness and case series with at least 25 participants to provide harms data. These requirements eliminated some smaller case series reporting on rarer presentations of IH (e.g., liver IH). We were also dependent upon the characterization of IH as presented in each study. Given changes in nomenclature and variations in the way IH are described, it may be that some studies included non-IH lesions. However, our clinical experts carefully reviewed studies to attempt to ascertain that included studies were reporting on true IH. We also note that other approaches to meta-analysis could be used, but that our estimates of a high anticipated response to propranolol largely align with those in other reviews of propranolol.^{254,258,259,261}

Limitations of the Evidence Base

The evidence base for IH treatment is limited by a small number of comparative studies including a limited number of participants. While cohort studies compared at least two different interventions, few presented truly comparative data. A number of studies reported only absolute differences in resolution or other outcomes, with no statistical comparison, in part likely due to their small sample sizes. Similarly, few studies reported baseline characteristics of the lesion, so understanding the magnitude of change reported is challenging. Most studies included children with problematic IH, so change was likely substantial, and parents and children may value any lessening of lesion size or change in color or texture.

A growing number of studies address beta-blockers, but current studies are limited by a general lack of long-term followup and analyses to explore differences in response among subgroups. Studies may also have used compounded forms of beta-blockers, which may add to the complexity of interpreting dosage amounts. Few comparative studies addressed steroids, and indications for steroid treatment compared with beta-blockers are unclear. Few comparative studies addressed surgical approaches besides laser modalities, and those addressing lasers used different interventions and comparators, limiting comparisons across studies. Technological advances have also changed the indications for treatment, and a historical trend towards treating smaller, less severe lesions, similarly make analyses difficult because of changing indications for and expectations of treatment.

Studies are also limited by the use of multiple and variable outcome measures to assess resolution of lesions. As no objective lab value or other measures exist to determine size changes, investigators have developed multiple techniques, and studies did not always report scales or other approaches clearly. The variety of scales (e.g., percentage change, mean change, VAS, HAS) makes combining outcomes challenging. Similarly, studies typically included multiple lesion types in multiple locations, which complicates determining potential differences in response, and treatment approaches varied across studies (e.g., doses and dosage forms, level of patient monitoring, timing of treatment and followup).

The most important deficiency in the reported outcomes across studies is the tendency for the reporting of discretized outcomes, when the underlying outcome is a continuous variable. Specifically, though outcomes are likely recorded as a continuous measure (i.e., the proportion of an existing lesion that is cleared or reduced in size following treatment), authors often chose an arbitrary cutoff proportion (or a small number of bins) and reported only the numbers in each of the resulting categories. This results in an immediate and unrecoverable loss in power for any quantitative meta-analyses. Researchers should be encouraged to report outcome variables as they were recorded, without transforming them in such a way that information is lost. In addition, methods for measurement of outcomes such as rebound growth are not clearly reported; thus, our understanding of the magnitude of regrowth is limited.

Research Gaps and Areas for Future Research

While a growing number of comparative studies address treatments for IH, a number of research gaps exist. These gaps include a lack of information on:

• Indications, optimal timing, and optimal modalities for imaging and diagnostic approaches. Few studies in the literature we reviewed reported imaging or diagnostic techniques, and data on optimal approaches for each are lacking in the current research base. In general, imaging is infrequently used to differentiate accurately an IH from other vascular lesions. When a diagnosis is in question, a tissue biopsy is the most accurate method to determine the diagnosis. Future studies should use imaging modalities at the same point in the IH course to allow direct comparison. Studies should also report adverse effects of imaging, which are not addressed in the literature meeting criteria for this review.
• Indications for treatment and treatment referral. While it is likely that non-placebo-controlled studies reviewed here included mostly children with problematic IH (e.g., lesions that are vision-threatening or disfiguring, ulcerated lesions, airway/life-threatening lesions), studies did not always clearly report indications for treatment or referral for treatment. Children may be referred for life-, functional-, or vision-threatening reasons, but in the beta-blocker era, potential disfigurement is likely a cause for referral.
• Appropriate dosing for propranolol and timing of treatment. The largest RCT to date⁹² used doses of either 1 mg/kg or 3 mg/kg, but other studies typically used doses of 2–2.5 mg/kg, and ages of children and number, severity, and type of lesions varied among study populations. Existing studies do not provide data to determine optimal dosing. Similarly, few studies reported on resolution outcomes by phase (i.e., proliferative, involution). Studies likely included mostly children in the proliferative phase, but the effectiveness of propranolol during the involution phase is not clear. Similarly, because proliferation may occur up to and after 12 months of age, the effectiveness of starting beta-blockers in older children is not clear.
• Optimal duration of beta-blocker use. Duration of propranolol treatment ranged from 3 to 13 months in comparative studies, but the optimal duration of treatment is not clear. Studies generally treated children for 6 months, potentially so that effects observed were likely drug-related and not the result of natural involution. However, current studies have not addressed the question of optimal timing to achieve maximal benefit.
• Long-term outcomes and harms of beta-blockers. While harms reported in studies of beta-blockers were typically not severe, only one comparative study¹⁴⁴ had greater than 6 months followup after the end of treatment. Longer term effects on cardiovascular and metabolic parameters known to be affected by beta-blocker use as well as effects on cognition, memory, and the central nervous system are not well-understood in the population of very young children receiving beta-blockers for IH.²⁶³
• Treatment choice for specific lesion types and locations. Characteristics, such as lesion size, location, and persistence, as well as modifiers such as patient age, functional impact, and IH subtype influence whether children are treated with pharmacologic agents or surgically. Lesion characteristics also influence the choice of specific pharmacologic agents. Most studies included multiple lesion types and in multiple locations, and few included specific modifier analyses or reported outcomes by lesion characteristics. Research to improve understanding of which lesions are likely to respond best to specific agents is critical, especially as understanding of the effectiveness of beta-blockers in the involution phase is limited. Optimal treatment in the proliferative phase may be key to maximal resolution of IH.
• Assessment of methods for assessing rebound growth. A number of studies reported regrowth of lesions but typically did not indicate what constituted rebound growth. Greater clarity in reporting this outcome would help to clarify our understanding of effectiveness.
• Characteristics that may influence response to beta-blockers. Studies of beta-blockers were typically not powered to provide information on subgroups, but a percentage of children did not respond or responded minimally to propranolol. In 10 comparative studies of beta-blockers reporting these data,^{17,93,94,98,103,104,130,144,146,147,150} 20 percent of children (n=63/314) had a limited or no response to the agent. We lack data to assess whether improvement in lesions or promotion of involution is affected by child age or number, severity, type, or anatomic location of lesions. Similarly, understanding the mechanisms of growth of IH will promote our understanding of response to treatments and treatment safety.
• Use of beta-blockers other than propranolol. Small cohort studies of oral atenolol and nadolol and topical or ophthalmic timolol showed positive effects on IH resolution with few side effects. Additional RCTs of these agents, with clear reporting of lesion parameters and child characteristics, would increase our understanding of their effectiveness and comparative effectiveness versus propranolol.
• Treatments for hepatic IH. Few treatment studies explicitly reported if children had hepatic IH. Most studies included children with IH in multiple locations, so children could have had hepatic IH as well; however, the applicability of findings to children with visceral IH is not clear.
• Use of steroids and laser treatments in the beta-blocker era. Clinical practice in the United States is moving toward use of a beta-blocker as the first-line treatment for IH;¹⁵ however, a number of recent studies report use of steroids and laser treatments in younger children with lesions in the proliferative stage. Given the side effect profile of steroids, understanding of whether or when to use such agents in the absence of life-threatening lesions or contraindications to beta-blockers is needed. Current literature does not provide sufficient data to address these questions.
• Interventions to follow beta-blockers or corticosteroids if such treatments fail. We did not identify any studies that clearly reported data on this question. While most children receiving beta-blockers in the studies reviewed here responded to the medication, some had no or minimal response.
• Standardization of scoring tools to assess change in IH. IH outcomes are necessarily assessed using subjective measures, and investigators typically reported grading scales used to assess change in IH size or appearance. Few studies, however, commented on interrater reliability of instruments. Research to improve standardization among tools and the development of uniform scoring systems and measurements would improve our ability to combine outcomes across studies.
• Standardization of nomenclature. Data extraction and comparisons in the review were limited by inconsistent naming conventions. Agreement and adherence to a standard classification of lesions would improve the ability of researchers to focus on individual lesion types and determine optimal treatment regimens for specific lesions.