Interventions for Infantile Haemangiomas of the Skin

Abridged Cochrane Systematic Review and Grade Assessments

M. Novoa; E. Baselga; S. Beltran; L. Giraldo; A. Shahbaz; H. Pardo-Hernandez ID; I. Arevalo-Rodriguez ID

Disclosures

The British Journal of Dermatology. 2019;180(3):527-533. 

In This Article

Results

The search results are shown in Figure 1. We included 28 studies reported in 37 references enrolling 1728 children.[2,8,10,11,13,19–48] We assessed seven studies as awaiting classification because only partial information was available for these references (Appendix S2; see Supporting Information).[49–55] Of the 45 studies excluded after full-text assessment, six were excluded owing to an ineligible population or other diseases, and one for being developed in animals. The remaining 38 excluded studies were not randomized trials (Appendix S2; see Supporting Information).

Figure 1.

Study flow diagram. RCT, randomized controlled trial.

Sample sizes ranged from 12 to 460 children. Different subtypes of IH were assessed, including children with mixed or deep IH, ulcerated or problematic IH, or high-risk haemangiomas. Six studies exclusively assessed superficial haemangiomas, whereas one trial evaluated mixed haemangiomas only. Regarding outcomes, clearance was assessed in 14 studies, a subjective measurement of improvement in one trial and adverse events in 20 trials. For the secondary outcomes, 16 trials reported other measures of resolution, and one trial reported the proportion of parents who still consider that their child has a problem, requirement of surgical correction or aesthetic appearance.

Regarding the methodological quality of trials, we found several cases of unclear risk of bias (Figure 2). Blinding of outcomes assessment was fully reported for 13 trials.[2,10,11,13,20,26,27,31,32,36,38–40] We considered only three studies at low risk of bias for the three blinding items.[10,26,32] In addition, most of the comparisons were based on a single trial with a small number of recruited children; only four trials recruited more than 100 children.[23,28,34,56]

Figure 2.

Risk-of-bias graph. Review authors' judgements about each risk-of-bias item, presented as percentages across all included studies.

We discuss the results from the three key comparisons, which we chose on the basis of current use: oral propranolol vs. placebo; topical timolol maleate vs. placebo; and oral propranolol vs. topical timolol maleate. Additional details about the remaining 10 interventions and 27 comparisons assessed can be consulted in the full Cochrane review.1

Oral Propranolol Versus Placebo

For this comparison, we included information from three trials and 312 children.[2,10,32] One of the included trials enrolled 40 children between the age of 9 weeks and 5 years with facial IH or IH in sites with potential for disfigurement. Administration was initiated at a dosage of 1 mg kg−1 daily divided three times daily and then increased to 2 mg kg−1 daily divided three times daily from weeks 2 to 24. The children were followed up at weeks 0, 4, 8, 12, 16, 20 and 24. Duration of treatment and follow-up was 24 weeks in both cases. A second trial enrolled 14 children younger than 16 weeks of age with one or more nonthreatening IH of > 1 cm diameter, without vital or functional impairment. Children were randomly assigned to receive placebo or propranolol 3 mg kg−1 daily for 15 days, then 4 mg kg−1 daily for 15 additional days. Duration of treatment and follow-up was 1 month in both cases. The remaining trial enrolled 456 infants between 35 and 150 days old with proliferating IH requiring systemic therapy. Children were randomly assigned to receive propranolol 1 mg kg−1 per day, 3 mg kg−1 per day or placebo for 3 or 6 months. We reported data for the groups followed for 6 months in this analysis.

Regarding clearance of lesions, 50 of 102 (49·0%) children in the oral propranolol group (1 mg kg−1 daily) and two of 55 (4%) in the placebo group achieved clearance of lesions. The risk of clearance after oral propranolol 1 mg kg−1 day was 13·48 times that after placebo (RR 13·48, 95% CI 3·41–53·30). Similarly, 61 of 101 (60·4%) children in the oral propranolol group (3 mg kg−1 day) and two of 55 (4%) in the placebo group reached clearance of lesions. The risk of clearance after oral propranolol 3 mg kg−1 day was 16·61 times that after placebo (RR 16·61, 95% CI 4·22–65·34).

In addition, three trials reported information about 26 cases of serious adverse events in general, but this was based only on information from a single trial (5·1% of cases),[2] as the other two trials had zero events in both arms.[10,32] Twenty-three children of 427 (5·4%) in the oral propranolol group and three of 82 (4%) in the placebo group had serious adverse events. This study found no differences in terms of adverse events when comparing oral propranolol with placebo (RR 1·05, 95% CI 0·33–3·39). Likewise, we did not find a significant difference between oral propranolol and placebo at any doses, in terms of serious cardiovascular adverse events and other adverse events, including bronchospasm and hypoglycaemia.

Only one study provided information about other measures of resolution, including percentage change in volume at 24 weeks, and redness.[32] For change in volume, the authors reported a reduction of mean haemangioma volume at 24 weeks of 45·9% (95% CI 11·60–80·20%) comparing oral propranolol with placebo. The authors of this trial also found an improvement in redness at week 24 in four of 20 (20%) children in the oral propranolol group and in no (n = 20; 0%) children in the placebo group. However, this study found no clear difference in terms of improvement in redness when comparing oral propranolol with placebo due to imprecision (RR 9·00, 95% CI 0·52–156·9).

The quality of the evidence for these outcomes ranged from moderate to low (Table S1; see Supporting Information).

Topical Timolol Maleate Versus Placebo

For this comparison, we included the information from one trial and 41 children aged from 5 weeks to 24 weeks with small, focal superficial IH.[26] Children were randomly assigned to receive placebo or timolol maleate 0·5% gel twice a day. Duration of treatment and follow-up was 6 months in both cases.

The authors of this trial provided information on the absence of redness at the end of follow-up, reporting eight cases (41 children; 19·5% with absence of redness). Seven of 19 (37%) children in the topical timolol maleate group and one of 22 (4%) in the placebo group achieved clearance of lesions. The absence of redness after topical timolol maleate was 8·11 times that after placebo (RR 8·11, 95% CI 1·09–60·09). This study also provided information about cases of volume reduction (≥ 5%), reporting 11 cases (41 children; percentage of volume reduction 26·8%). Nine of 19 (47%) children in the topical timolol maleate group and two of 22 (9%) in the placebo group achieved clearance of lesions. Volume reduction after topical timolol maleate was 5·21 times that after placebo (RR 5·21, 95% CI 1·28–21·21). Regarding adverse events, this trial provided information about serious cardiovascular events, reporting no bradycardia or hypotension (41 children).

The quality of the evidence for these outcomes was low (Table S2; see Supporting Information).

Oral Propranolol Versus Topical Timolol Maleate

For this comparison, we included information from one trial with 26 children.[31] Children with superficial IH were randomized to receive oral propranolol (via a tablet taken once daily, at a dose of 1·0 mg kg−1), topical timolol maleate (0·5% eye drops applied twice daily) or a combination of these treatments. Treatment efficacy was evaluated based on clinical photographs taken at the onset of treatment, during treatment and at the end of treatment. Treatment ended if lesions had regressed or after 6 months without improvement; children were followed from 3 to 12 months.

This trial provided information about a size reduction of ≥ 50% after treatment (change in size before vs. after treatment, based on clinical photographs), reporting a total of 17 cases (26 children; 65%). Nine of 13 (69%) children in the oral propranolol group and eight of 13 (61%) in the topical timolol maleate group achieved a size reduction of ≥ 50%. This study found no differences in terms of this measurement when comparing oral propranolol with topical timolol maleate (RR 1·13, 95% CI 0·64–1·97). The study provided information about adverse events in general, such as severe diarrhoea, lethargy, and loss of appetite, reporting a total of three cases (26 children; 11%). Three of 13 (23%) children in the oral propranolol group and no (n = 13; 0%) children in the topical timolol maleate group reported adverse events. This study found no clear difference in terms of adverse events when comparing oral propranolol with topical timolol maleate (RR 7·00, 95% CI 0·40–123·35).

The quality of the evidence for these outcomes ranged from low to very low (Table S3; see Supporting Information).

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....