Clinical features and management of kaposiform hemangioendothelioma and tufted angioma: Similarities and differences

To the Editor: To some extent, kaposiform hemangioendothelioma (KHE) and tufted hemangioma (TA) have overlapping clinical and histopathologic features, and both the tumors may be associated with the development of the Kasabach-Merritt phenomenon (KMP). At present, most investigators consider KHE and TA to be within the same disease spectrum rather than distinct entities, although the aggressiveness and risks of KMP development are different between KHE and TA.1Wassef M. Blei F. Adams D. et al.Vascular anomalies classification: recommendations from the International Society for the Study of Vascular Anomalies.Pediatrics. 2015; 136: e203-e214Crossref PubMed Scopus (675) Google Scholar KHE and TA may have different clinical features and treatment strategies. Unfortunately, relevant data are currently lacking. After ethics approval, we retrospectively analyzed the clinical characteristics and management of 255 patients with KHE and TA diagnosed by histopathology, with the aim of providing useful information that may aid in clinical practice. The main characteristics are summarized in Table I. There were no significant differences between the sexes in the 2 groups (P = .312). No differences in the mean age at disease onset (P = .606) or the mean age at diagnosis of KHE or TA (P = .055) were observed between the 2 groups. A marked difference was found in the size and depth of lesions between the KHE and TA groups. The frequencies of KMP, musculoskeletal disorders, and compression of vital structures were higher in the KHE group than in the TA group.Table IComparisons of clinical features between KHE and TA groupsVariablesKHE (n = 215)TA (n = 40)Total (n = 255)P values∗Subjects with KHE compared with subjects with TA.Sex, n (%).312†P value was calculated using the χ2 test. Female89 (41.4)20 (50.0)109 (42.7) Male126 (58.6)20 (50.0)146 (57.3)Age at onset (mo) Mean ± SD12.6 ± 39.617.2 ± 54.913.4 ± 42.3.606§P value was calculated using the Mann-Whitney U test. Median (range)2.0 (0.0-432.0)3.0 (0.0-288.0)2.0 (0.0-432.0)Age at diagnosis (mo) Mean ± SD18.1 ± 47.233.3 ± 76.920.5 ± 53.1.055§P value was calculated using the Mann-Whitney U test. Median (range)4.0 (0.3-480.0)8.0 (1.0-336.0)5.0 (0.3-480.0)Location, n (%).250†P value was calculated using the χ2 test. Head-face-neck42 (19.5)4 (10.0)46 (18.0).150†P value was calculated using the χ2 test. Extremities104 (48.4)26 (65.0)130 (51.0).053†P value was calculated using the χ2 test. Trunk39 (18.1)6 (15.0)45 (17.7).632†P value was calculated using the χ2 test. Multiple anatomic sites30 (14.0)4 (10.0)34 (13.3).499†P value was calculated using the χ2 test.Maximum diameter (cm) Mean ± SD5.9 ± 3.45.0 ± 3.85.8 ± 3.5.015§P value was calculated using the Mann-Whitney U test. Median (range)5.0 (1.2-25.0)4.1 (0.7-20.0)4.9 (0.7-25.0)Depth of invasion, n (%)‡The depth of infiltration was divided into 3 subtypes: superficial (involving the dermis, subcutis and deep fascia), deep (involving muscle, bones, joints, intrathoracic tissue, the retroperitoneum, and internal organs, without cutaneous lesions), and mixed (features of both superficial and deep subtypes).<.001†P value was calculated using the χ2 test. Superficial54 (25.1)35 (87.5)89 (34.9)<.001†P value was calculated using the χ2 test. Mixed123 (57.2)5 (12.5)128 (50.2)<.001†P value was calculated using the χ2 test. Deep38 (17.7)0 (0)38 (14.9).004†P value was calculated using the χ2 test.Complications, n (%) KMP99 (46.0)0 (0)99 (38.8)<.001†P value was calculated using the χ2 test. Musculoskeletal disorders‖Musculoskeletal disorders were characterized by tumor-induced pain or the limitation of movement.135 (62.8)6 (15.0)141 (55.3)<.001†P value was calculated using the χ2 test. Compression of vital structures26 (12.1)0 (0)26 (10.2).042†P value was calculated using the χ2 test. Lymphedema17 (7.9)0 (0)17 (6.7).135†P value was calculated using the χ2 test.KHE, Kaposiform hemangioendothelioma; KMP, Kasabach-Merritt phenomenon; SD, standard deviation; TA, tufted angioma.∗ Subjects with KHE compared with subjects with TA.† P value was calculated using the χ2 test.‡ The depth of infiltration was divided into 3 subtypes: superficial (involving the dermis, subcutis and deep fascia), deep (involving muscle, bones, joints, intrathoracic tissue, the retroperitoneum, and internal organs, without cutaneous lesions), and mixed (features of both superficial and deep subtypes).§ P value was calculated using the Mann-Whitney U test.‖ Musculoskeletal disorders were characterized by tumor-induced pain or the limitation of movement. Open table in a new tab KHE, Kaposiform hemangioendothelioma; KMP, Kasabach-Merritt phenomenon; SD, standard deviation; TA, tufted angioma. The proportions of patients requiring intervention were significantly different between the KHE and TA groups (P < .001). Systemic pharmacotherapy was administered to 11 of 40 and 182 of 215 patients with TA and KHE, respectively, to control the disease (P < .001). Overall, the response rate in the TA group was better than that in the KHE group (P = .016). Majority of the patients with KHE (56.3%) received sirolimus with or without short-term corticosteroids as a first-line treatment (Table II).Table IIManagement of KHE and TAManagementKHE (n = 215)TA (n = 40)P values†Subjects with KHE compared with subjects with TA.No. of patientsEffective, n (%)∗Efficacy was defined as >20% reduction in target lesions, with an improvement in complications and/or the restoration of hematologic parameters during treatment or expectant management.No. of patientsEffective, n (%)Expectant management157 (46.7)1110 (90.9).036‡P value was calculated using the Fisher's exact test.Topical tacrolimus76 (85.7)1717 (100).292‡P value was calculated using the Fisher's exact test.Surgical excision2916 (55.2)32 (66.7)1.000‡P value was calculated using the Fisher's exact test.Systemic pharmacotherapy Corticosteroids124 (33.3)0NANA Propranolol103 (30.0)32 (66.7).510‡P value was calculated using the Fisher's exact test. Propranolol with corticosteroids125 (41.7)0NANA Vincristine96 (66.7)0NANA Sirolimus7773 (94.8)66 (100)1.000‡P value was calculated using the Fisher's exact test. Sirolimus with corticosteroids4442 (95.5)0NANATotal215162 (75.3)4037 (92.5).016§P value was calculated using the χ2 test.KHE, Kaposiform hemangioendothelioma; NA, not applicable; TA, tufted angioma.∗ Efficacy was defined as >20% reduction in target lesions, with an improvement in complications and/or the restoration of hematologic parameters during treatment or expectant management.† Subjects with KHE compared with subjects with TA.‡ P value was calculated using the Fisher's exact test.§ P value was calculated using the χ2 test. Open table in a new tab KHE, Kaposiform hemangioendothelioma; NA, not applicable; TA, tufted angioma. In the present study, effective intervention with sirolimus, timely referral for high-risk patients, and smaller lesion size in patients with TA than in patients with KHE might be related to the differences in the onset of the KMP. A previous study found that younger age, larger lesion size, and mixed lesions are independent risk factors for KMP development.2Ji Y. Yang K. Peng S. et al.Kaposiform haemangioendothelioma: clinical features, complications and risk factors for Kasabach–Merritt phenomenon.Br J Dermatol. 2018; 179: 457-463PubMed Google Scholar Furthermore, musculoskeletal disorders and the compression of vital structures were more common in patients with KHE than in those with TA; moreover, lymphedema occurred only in the KHE group, which further supports the concept that TA is milder than KHE. Clinically, treatment practices and regimens for these rare vascular tumors should be tailored to individual patients and guided by specific clinical circumstances.3Drolet B.A. Trenor C.C. Brandão L.R. et al.Consensus-derived practice standards plan for complicated Kaposiform hemangioendothelioma.J Pediatr. 2013; 163: 285-291Abstract Full Text Full Text PDF PubMed Scopus (148) Google Scholar Previously, corticosteroids and vincristine were considered the first-line treatments for KHE and TA.3Drolet B.A. Trenor C.C. Brandão L.R. et al.Consensus-derived practice standards plan for complicated Kaposiform hemangioendothelioma.J Pediatr. 2013; 163: 285-291Abstract Full Text Full Text PDF PubMed Scopus (148) Google Scholar Sirolimus has been shown to be one of the most effective treatments for KHE and TA.4Ji Y. Chen S. Xiang B. et al.Sirolimus for the treatment of progressive kaposiform hemangioendothelioma: a multicenter retrospective study.Int J Cancer. 2017; 141: 848-855Crossref PubMed Scopus (70) Google Scholar Consistent with a previous study,5Browning J. Frieden I. Baselga E. Wagner A. Metry D. Congenital, self-regressing tufted angioma.Arch Dermatol. 2006; 142: 749-751Crossref PubMed Scopus (48) Google Scholar we found that wait-and-see and topical tacrolimus may be effective strategies for patients with TA because of the nature of spontaneous regression, superficial invasion, and fewer complications. In contrast, systemic sirolimus with or without corticosteroids is needed for most patients with KHE. We also demonstrated that the overall response rate in the TA group was better than that in the KHE group, regardless of management strategies, which has implications for clinical decision making. None disclosed.