Analysis of Risk Factors for Kasabach Merritt Phenomenom in Children With Kaposiform Hemangioendothelioma.

Background: This study generalized and analyzed the clinical attributes observed in patients afflicted with Kaposiform hemangioendothelioma (KHE) with the aim of elucidating the risk factors contributing to the manifestation of Kasabach-Merritt phenomenon (KMP).

Methods: We retrospectively analyzed 96 pediatric cases diagnosed with KHE at the Children's Hospital of Fudan University from January 2013 to December 2021. Among them, 62 patients (65%) showed KMP (KHE + KMP group), while 34 patients (35%) did not (KHE-KMP group). The risk factors for KMP associated with KHE were analyzed using univariate analysis and binary logistic regression analysis, comparing the differences between KHE + KMP group and KHE-KMP group.

Results: Univariate analysis indicated no statistically significant differences between the two groups in gender, prematurity, family history, or color of involved skin. However, statistically significant differences were observed in age of onset, lesion site, and lesion depth. Multivariate analysis revealed significant associations: children with onset age ≤1 month had a 51-fold increased risk of KMP compared to those with onset age >1 month (95% CI 5.238-501.663); non-extremity lesion sites exhibited a 21-fold higher risk of KMP compared to extremity sites (95% CI 3.970-105.958); deeper lesions conferred a 5-fold higher risk of KMP compared to superficial lesions (95% CI 1.073-21.005); lesions >60 mm carried a 17-fold higher risk of KMP compared to lesions ≤60 mm (95% CI 2.999-96.157). A comprehensive predictive model was developed using the fitting formula: Logit (P) = 3.937∗(age at onset) + 1.558∗(lesion depth) + 3.021∗(lesion site) + 2.832∗(lesion size), demonstrating an accuracy of 82.9%. Furthermore, a scoring system was established to assess the likelihood of KMP occurrence. Children diagnosed with KHE were likely to have KMP if their score exceeded 72.5, as determined by Receiver Operating Characteristic (ROC) curve analysis.

Conclusion: Age of onset ≤1 month, deeper lesions, non-extremity sites, and lesions ＞60 mm are independent risk factors for KMP in children with KHE. The cumulative presence of these factors escalates the likelihood of KMP development. Additionally, the identification of these factors allows for the early recognition of potential KMP cases among children with KHE, facilitating prompt therapeutic intervention.

Category of the manuscript: Clinical Research article.

Level of evidence: LEVEL Ⅲ.