PURPOSE: To determine whether an epidemiologic association exists between glomus tumors and neurofibromatosis. METHODS: Using a pathology database, we established a study cohort consisting of all patients who had undergone excision of a glomus tumor of the hand between 1995 and 2010. We created a control cohort by randomly selecting 200 patients who had undergone excision of a ganglion cyst over the same period. We reviewed medical records for each cohort to identify patients with a diagnosis of neurofibromatosis. We calculated the odds ratio was calculated and performed Fisher's exact test to determine the significance of the association. RESULTS: We identified 21 patients with glomus tumors of the hand. Six of these patients carried the diagnosis of neurofibromatosis (29%). In contrast, no patients in the control group carried the diagnosis of neurofibromatosis. The odds ratio for a diagnosis of neurofibromatosis in association with a glomus tumor compared with controls was 168:1. CONCLUSIONS: This study provides evidence of a strong epidemiologic association between glomus tumors and neurofibromatosis. Glomus tumor should be included in the differential diagnosis in neurofibromatosis patients who present with a painful lesion of the hand or finger. TYPE OF STUDY/LEVEL OF EVIDENCE: Diagnostic III.
PURPOSE: To determine whether an epidemiologic association exists between glomus tumors and neurofibromatosis. METHODS: Using a pathology database, we established a study cohort consisting of all patients who had undergone excision of a glomus tumor of the hand between 1995 and 2010. We created a control cohort by randomly selecting 200 patients who had undergone excision of a ganglion cyst over the same period. We reviewed medical records for each cohort to identify patients with a diagnosis of neurofibromatosis. We calculated the odds ratio was calculated and performed Fisher's exact test to determine the significance of the association. RESULTS: We identified 21 patients with glomus tumors of the hand. Six of these patients carried the diagnosis of neurofibromatosis (29%). In contrast, no patients in the control group carried the diagnosis of neurofibromatosis. The odds ratio for a diagnosis of neurofibromatosis in association with a glomus tumor compared with controls was 168:1. CONCLUSIONS: This study provides evidence of a strong epidemiologic association between glomus tumors and neurofibromatosis. Glomus tumor should be included in the differential diagnosis in neurofibromatosispatients who present with a painful lesion of the hand or finger. TYPE OF STUDY/LEVEL OF EVIDENCE: Diagnostic III.