• infantile haemangioma
• preterm infants
• cryocontact therapy
• controlled study

Despite the high prevalence of infantile haemangioma (IH) of preterm infants, methods for treating it have not yet been evaluated in controlled trials.1 Therefore, we conducted a prospective controlled study, comparing nitrogen-cooled cryocontact therapy (NCCT,−196°C) for early treatment of IH in preterm infants with no treatment.

From 1 August 2004 to 31 December 2008, we allocated IH of infants ≤34 weeks gestational age with at least two IH into either a cryocontact or a control group, thus using intraindividual controls to exclude interpersonal differences potentially affecting growth. If IH appeared simultaneously, they were numbered from head to foot; even numbers were treated, odd numbers served as controls. In sequentially erupting IH, always the second one was treated, and in critical locations (face, skull, joints and anogenital region) IHs were always primarily treated. NCCT was achieved by using a metal cylinder of variable diameter (max 10 mm) cooled by liquid nitrogen to cover and compress just the maximum IH diameter.2 We hypothesised that NCCT shows more successful resolutions than no therapy. The primary endpoint, at 1 or 2 years corrected age, was intact skin with mild or no pigmentation, without haemangioma or scar formation.

A sample size of 25 pairs of treated and untreated IH was calculated.3 However, the study was terminated in December 2008 because of a much slower recruitment of appropriate IH pairs than anticipated. To ensure analysis of independent pairs, we recalculated our data including only one pair of IH per infant using the unsuccessful result for calculation if two or more pairs with different results were found in the same infant, using SAS V.9.1.3 for Windows with Fisher’s Exact Test for differences in proportions. The study was approved by the ethics committee of Tuebingen University Hospital, and registered at ClinicalTrials.gov, Identifier NCT01059045. Parents gave written informed consent.

We recruited 13 infants with 17 pairs of IH (table 1) from 1 August 2004 to 31 December 2008. Thirteen out of 17 IH (76%) treated with cryocontact therapy, and 2 of 17 (12%) untreated controls met the primary endpoint (p<0.001) (table 2). The difference remained statistically significant even if only one pair of IH per infant was included, using the unsuccessful result if applicable (9/13 treated IH vs 2/13 controls). The time of NCCT application ranged from 2 to 6 s. Side effect of cryocontact therapy was scarring in 4/17 IH, and of controls persisting IH in 14/17 IH, and one large scar after surgical removal (1/17).

In conclusion, in our study group, cryocontact therapy promoted fast regression of IH with very good cosmetic results, preventing the unpredictable complications of certain lesion if left untreated. Additionally, it is fast, easy, well tolerated and less expensive than, for example, laser therapy. However, limitations of our study, for example, its premature termination, a short duration of follow-up and scarring as a side effect of NCCT, have to be considered.