Comparison of Infantile Hemangiomas in Preterm and Term Infants: A Prospective Study FREE

Infantile hemangiomas (IH) occur more frequently in premature infants.^{1 - 2} To our knowledge, studies comparing the clinical features of IH in term and preterm infants have not been reported. Results of the demographic and clinical characteristics of our prospective cohort study of children with IH have previously been reported.^{3 - 4} The current study compares the characteristics of IH in preterm and term infants within this cohort.

Data on gestational age (GA) was available for 1047 patients who were enrolled at the sites in the United States and formed the basis of the study. Preterm infants were defined as those with a GA of less than 37 weeks, and term infants were those with a GA of 37 weeks and greater.

Demographic and clinical characteristics are summarized in the Table. Two hundred and fourteen subjects were preterm (20%), and 73 of these had a gestational age of 32 weeks or less (7%). Factors associated with preterm birth included a higher reported incidence of preeclampsia, placental anomalies, and use of infertility treatments than in term counterparts (P < .001). The female to male ratio was less pronounced in premature infants (1.85) than in term infants (2.62) (P = .04). No differences were noted in the age when IH was first noted or age at the time of presentation to a specialist. The mean (SD) number of hemangiomas was inversely related to GA: 1.37 (0.78) in term infants; 1.60 (1.15) in preterm infants with a GA of 33 to 36 weeks; and 1.83 (1.17) in preterm infants with a GA of 32 weeks or less (P < .001).

Forty percent of preterm infants had 2 to 5 lesions vs 24.5% of term infants (P < .001), and 7.5% of preterm infants had more than 5 lesions compared with 3% for term infants (P < .001) (Figure). Analysis of birth weight association alone showed that lower birth weight also correlated with increased numbers of IH (P < .001). Localized hemangiomas were more common than either segmental or indeterminate subtypes in both preterm and term groups; however, there was no significant difference in the incidence of segmental hemangiomas in preterm infants (14.6%) and term infants (18.1%) (P = .24). Anatomic location differed, with facial involvement being more common in term infants (P = .005). There was no difference in the incidence of complications (eg, visual compromise, ulceration, or cardiac or airway compromise), and need for treatment did not differ, nor did the treatment techniques used in those infants requiring treatment.

While previous studies have confirmed the higher incidence of IH in premature infants, to our knowledge, this is the first study to address whether characteristics of IH differ between term and preterm infants. The most notable differences found were the increased number of hemangiomas in preterm infants and the decreased female to male ratio (Figure and Table). The recent suggestion for a role of endothelial progenitor cells in IH development could help explain both the increase in frequency and numbers of hemangiomas in preterm infants because these progenitor cells might be more likely to be present and in higher numbers in an immature fetus than in a mature one.⁵ This hypothesis might also help explain the decrease in female to male ratio noted, with prematurity offsetting whatever factors (as yet unknown) cause a female predominance.

Because premature infants have greater numbers of hemangiomas, a higher percentage overall are of the localized subtype, but there is no significant difference in the per-patient incidence of segmental hemangiomas between term and preterm infants. Segmental hemangiomas appear to arise as an error in neural crest development as early as 4 to 8 weeks GA, and their occurrence is thus less likely to be affected by premature delivery.⁶

The higher incidence of multiple-gestation pregnancies, placental anomalies, and preeclampsia in the preterm group is predictable because these factors are associated with preterm birth regardless of the presence of IH. However, a more causal relationship cannot be completely discounted. These pregnancy-related morbidities could hypothetically lead to an imbalance toward proangiogenic factors as responses to a hypoxic environment, either via the expression of growth factors, such as insulinlike growth factor 2,⁷ or via hypoxia-inducible factor 1α, which is upregulated in proliferating hemangiomas.⁵

Limitations of the study include an ascertainment bias due to the nature of the study sites (pediatric dermatology practices) and the reliance on parental recall vs obstetrical data to determine perinatal history. Nevertheless, several findings should provide assistance to clinicians in managing preterm infants with IH. Hemangiomas did not differ in the time they were first noted or when they were first seen by specialists, suggesting that actual rather than adjusted GA can be used to predict growth characteristics. And although hemangiomas are more common in preterm infants, they do not appear to behave either more or less aggressively than in term infants: nor were differences noted in complication rates or need for treatment.

Correspondence: Dr Garzon, Department of Dermatology, Columbia University, 161 Fort Washington Ave, 12th Floor, New York, NY 10032 (mcg2@columbia.edu).

Author Contributions: Drs Garzon, Drolet, and Frieden had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Garzon, Drolet, Chamlin, Haggstrom, Mancini, Metry, and Frieden. Acquisition of data: Garzon, Drolet, Baselga, Chamlin, Haggstrom, Horii, Lucky, Mancini, Metry, Newell, Nopper, and Frieden. Analysis and interpretation of data: Garzon, Drolet, Mancini, and Frieden. Drafting of the manuscript: Garzon, Drolet, Mancini, and Frieden. Critical revision of the manuscript for important intellectual content: Garzon, Drolet, Baselga, Chamlin, Haggstrom, Horii, Lucky, Mancini, Metry, Newell, and Nopper. Statistical analysis: Drolet. Obtained funding: Drolet and Frieden. Administrative, technical, and material support: Garzon, Drolet, and Horii. Study supervision: Garzon, Drolet, and Frieden.

Group Information: Drs Garzon, Drolet, Baselga, Chamlin, Haggstrom, Horii, Lucky, Mancini, Metry, Newell, Nopper, and Frieden make up the entirety of the Hemangioma Investigator Group.

Financial Disclosure: None reported.

Funding/Support: This study was supported in part by Children's Research Institute, Milwaukee, Wisconsin; the Dermatology Foundation, Evanston, Illinois; and the American Skin Association, New York, New York.

Role of the Sponsors: The sponsors had no role in the design or conduct of the study; in the collection, analysis, or interpretation of data; or in the preparation, review, or approval of the manuscript.

Previous Presentation: The preliminary data from this study were presented in abstract form at the 16th International Workshop on Vascular Anomalies of the International Society for the Study of Vascular Anomalies; June 15, 2006; Milan, Italy.

Additional Contributions: We thank our colleagues and staff at the participating institutions. Alan Bostrom, PhD, provided statistical expertise.