The Spanish Academy of Dermatology and Venereology (AEDV) began a nationwide prospective cohort of IHs in June 2016, and the study period extended to October 2019. Patients were recruited from 12 Spanish hospitals with paediatric dermatology departments. A 5-year follow-up study was initially planned and is still ongoing. This paper describes the baseline data of the cohort.

New-borns younger than 9 months with one or more IHs diagnosed by their characteristic clinical features or, in case of doubt, by immunohistochemical GLUT-1 staining were recruited. All consecutive patients attending one of the participating hospitals were included regardless of the proposed therapy (clinical observation or treatment). As this is a clinic-based observational study, the therapeutic approach was decided during routine clinical practice.

The following patients were excluded: patients whose family was planning a short-term move that would make follow-up difficult; patients who had already started any kind of treatment for IHs at the moment of the baseline visit; and patients who refused to participate in the study.

Maternal obstetric history, demographic and biometric characteristics of the patients, IH locations and characteristics, such as morphologic and clinical subtypes, size, height or the kind of surface and edge, the presence of complications or likelihood of complications and treatment initiation and the reasons for starting therapy were recorded at baseline and constitute the data used for this paper. IHs were classified according to their distribution (morphologic subtypes: localized, indeterminate, segmental and multifocal) and to their deepness and edge (clinical subtypes: superficial, mixed-sessile, mixed-pedunculated, deep and abortive). Complications were defined as the presence of any of the following: ocular, airway or visceral involvement, ulcerations or functional impairment (visual, nasal, feeding, airway, torticollis, external auditory canal obstruction or any other function considered important by the physician). Definitions were recorded in the study protocol, and a photographic atlas was created to ensure homogeneity in the collection of data on IH characteristics.

Data were collected by previously trained physicians (the dermatologists responsible for the paediatric dermatology departments at each participating hospital) using an online data collection system (Openclinica 3.1). Quality control consisted of continuous monthly online monitoring and annual on-site monitoring from a simple random sample of 10 patients from each of the hospitals (approx. 20% of the included patients per year).

Descriptive statistics were used to explore the characteristics of the sample. The Shapiro–Wilk test and histograms were used to explore the normality of the variables. Continuous variables are expressed as the mean and standard deviation (SD) or median (25th–75th percentiles). For qualitative variables, absolute and relative frequencies were estimated. Differences between qualitative variables were assessed by means of the χ2 test or Fisher’s exact test, when necessary. Differences between continuous variables were assessed by means of Student’s t-test or the Mann–Whitney U test. Statistical analyses were performed using STATA statistical software (version 16.0; Statacorp, College Station, TX, USA).

The study protocol was approved by the institutional review board of Hospital Santa Creu i Sant Pau (Barcelona, 16/079). All patients’ parents or caregivers signed written informed consent at the time of enrolment.

The baseline characteristics of the sample are shown in Table 1. The mean age was approximately 4 months; distinguishing between clinical IH subtypes, the mean age of deep IH patients was 4.5 months, while in patients with IH with superficial components, it was 3.8 months (p = 0.0075). There was a higher proportion of females (ratio 2:1). The mean weight and height at the baseline visit were normal for age, as was the weight at birth. The majority of new-borns were full-term. More than 60% of mothers did not have any abnormal maternal obstetric history. Among those who had obstetric antecedents, twin pregnancy, preeclampsia, gestational diabetes, and treatment with progesterone were the most prevalent. Patients whose mothers were treated with progesterone showed no differences in complications, median size, or proportion of segmental IH from those whose mothers did not receive progesterone.

The characteristics and locations of the IHs are shown in Table 2. Most IHs were localized and superficial. The median size (maximum × minimum diameter) was 17 mm × 12 mm. The most frequent locations were the trunk and face. Among haemangiomas with superficial components, superficial IHs more frequently had a steep edge, while mixed IHs were more likely to have a progressive edge (p < 0.05). Most of the IHs had a smooth surface. The height of the superficial component was greater in mixed IHs (p < 0.0001).

Colour was evaluated as bright red in 460 (78%) IHs, pale pink in 84 (14%), barely perceptible in 32 (6%) and imperceptible in 11 (2%).

Only 16 (3%) patients presented with complications at the baseline visit, mostly ulcerations, and 15 (3%) presented with functional impairment, mainly visual impairment (Table 3). Forty-nine (8%) patients presented with a risk of functional impairment, with vision being again the most compromised function. Five (1%) patients presented with PHACE syndrome, and 1 (0%) patient presented with LUMBAR syndrome.

The onset of and reasons for treatment are shown in Table 4. Slightly more than half of the patients started treatment at the baseline visit. Of these patients, approximately 3/4 started with topical timolol (5 mg/ml applied twice a day; data were available for 80%, but since this is the posology recommended in previous research21, it was expected that this posology would be used for most patients), and approximately 1/4 started with oral propranolol (median initial dosage: 1 mg/kg/day, as recommended by the European Medicine Agency22; data on starting dose was available for 76%, but it is expected that most patients started this dosage). Only 2 patients started laser therapy, and there were no other treatments.

In terms of sociodemographic characteristics, a higher proportion of females started treatment with propranolol (Table 1). Compared with that of the pivotal trial, age was similar, and the female-to-male ratio was higher; the other data were not reported in the trial.

Regarding IH characteristics, patients with segmental IHs, mixed and deep IHs, IHs with a larger diameter, a higher height, a progressive border, a smooth surface, and IHs located on the face were more prone to start treatment with propranolol. Compared with the trial population, higher proportions of indeterminate and segmental IHs treated with propranolol and similar proportions of superficial IHs and facial IHs. Other data about IH characteristics were not reported in the trial (Table 2).

Patients who suffered complications, functional impairment, or risk of functional impairment and those with PHACE syndrome started treatment with propranolol more frequently (Table 3). In the trial, data about complications were not reported.

The main reason for starting treatment, both with propranolol (79%) and with timolol (52%), was aesthetic impairment (Table 4). Only 47 (66%) patients treated with propranolol and 260 (50%) patients not treated with propranolol had an IH diameter larger than 15 mm (which was an inclusion criterion for the trial). Patients who started treatment because of aesthetic impairment were more likely to have received timolol (140, 70%) than propranolol (57, 29%). Information about the causes of treatment onset was not reported in the clinical trial.

Age at baseline visit in our cohort was similar to several studies published since the first use of propranolol18,23, which was first reported as a treatment for IHs in 200824, but lower than the age reported in studies carried out previously19,25. This finding might reflect the increased training and education of primary care physicians, who have greater accuracy in diagnosis and faster consultation with dermatologists. However, patients may still be seen late by dermatologists, given that most IHs appear at approximately 2 weeks of age and that the fastest growing stage occurs during the second month of life1. Other demographic characteristics, such as sex ratio, weight at birth or gestational age, were similar to those reported in previous studies18,26,27.

The most frequent morphologic subtype was localized, as previously reported28,29. Superficial and mixed IHs, as well as IHs with steep edges, have been related to a higher incidence of sequelae and worse aesthetic outcomes14,15,18. In comparison with these studies, in our cohort, we observed a higher proportion of superficial IHs and a lower proportion of mixed IHs. The prevalence of a cobblestone surface, which has been associated with anetoderma15, was similar between superficial and mixed IHs.

The median size of the IHs in our sample was smaller than that reported by Chamlin et al. in 200712, which could also reflect the earlier detection of and dermatological consultation for IHs. Regarding location, face and trunk were the areas most frequently affected.

The proportion of IH patients in our sample who suffered ulceration or other complications was lower than that reported previously for the use of propranolol12,26. These differences may be due to the effectiveness of the treatment but also to the exclusion of patients who had already started treatment by the time of the baseline visit in our research. Vision was the more frequently impaired function, since a periorbital localization is relatively common16.

More than 50% of patients started treatment at the baseline visit. The rapidly increased use of beta blockers as a treatment for IHs has resulted in a greater proportion of patients who receive treatment than previously reported27. The main reason for starting treatment was aesthetic impairment. Timolol was the most frequent treatment. This drug has a response rate similar to propranolol and higher than laser and observation according to a recent meta-analysis21. However, there are no clinical trials currently assessing its effectiveness, and it is used off-label. The presence of IH complications was an important reason for starting treatment with propranolol, although the main reason, as with timolol, was aesthetic impairment. Female patients and those with segmental IHs (which are associated with ulceration and complications7,12,28), with mixed or deep IHs, with IHs with a larger height and size or with IHs located on the face were more likely to receive propranolol. However, superficial IHs, IHs with a steep edge and IHs with a cobblestone surface, factors related to worse aesthetic outcomes14,15,18, were treated with propranolol less often. A remarkable fact is that 4 of the 5 patients with PHACE syndrome started propranolol, since it is currently considered a safe treatment in the absence of severe arteriopathies30,31. The two patients who did not receive beta-blocker treatment had ulcerated IHs and instead underwent laser treatment.

The use of propranolol is based principally on the outcomes of one clinical trial17, but the restrictive eligibility criteria may limit the generalizability of its results. Eligible patients were 35–150 days old with a proliferating IH requiring systemic therapy with a largest diameter of at least 15 mm. The exclusion criteria comprised, among others, IHs on the diaper area, function-threatening IHs, ulcerated IHs with a lack of response to simple wound care measures, patients who had been previously treated for IH, IHs with clinically uncertain diagnosis (particularly sub-dermal lesions) or patients born preterm who had not yet reached their term-equivalent age.

In our investigation, the age of patients who started propranolol at the baseline visit was similar to that of the patients in the clinical trial, while the female-to-male ratio was higher (3.5:1 vs 2.5:1). Other demographic characteristics were not reported in the trial. Compared with those of the clinical trial, the proportion of localized IHs was lower, that of indeterminate and segmental IHs was higher and that of superficial IHs was similar. Size was not reported in the trial, although the minimum size to meet the inclusion criteria was 15 mm, as mentioned above. In our cohort, only 47 (66%) patients treated with propranolol had an IH diameter larger than 15 mm. The proportion of facial/non-facial involvement was similar. In the clinical trial, the superficial component was only assessed as flat or elevated, and the last category was subdivided into slight, moderate and marked, making it difficult to compare with our results. Data about other IH characteristics, complications and causes of treatment onset were not reported in the clinical trial, which limits data comparison. Therefore, although 1/3 of our patients would not have been included in the trial because of the size of the IH and the proportion of segmental IH (related to worse outcomes) is higher, it seems that propranolol is generally being used in a similar population as that of the trial, but this statement cannot be definitely confirmed due to the lack of data.

The main strength of this study is that it is one of the largest prospective cohorts of patients with IH following the commercialization of propranolol and is very likely to be representative of the current characteristics and therapy of IHs in the clinic. The main limitation of our study is the exclusion of patients already undergoing treatment at the time of the baseline visit, which could have led to a selection bias. However, as the main outcome measure of the research will be the blindly aesthetic comparison of standardized photographs obtained at the baseline visit prior to treatment onset (or clinical observation) with those obtained at the end of the 5-year follow-up period, the availability of a pre-treatment photograph was considered necessary for inclusion.