PURPOSE: To ascertain the clinical spectrum of severity of dermoid cysts within the paediatric population, and to contribute to the understanding of which lesions are at higher risk of intracranial extension and warrant more thorough pre-operative work up. METHODS: We conducted a single-centre, consecutive, non-randomized comparative case series, reviewing the case notes of all patients treated surgically for craniofacial dermoids at the Royal Children’s Hospital in Melbourne, Australia. The study period was between August 2013 and May 2017, and a total of 221 patients were identified using coding for “dermoid cyst” procedures. Of these, 147 were identified as having craniofacial dermoids and included in the study. Three surgical units were involved in care (Plastic and Reconstructive Surgery, Neurosurgery and Paediatric Surgery). The data collected included demographic information, pre-operative assessment, pre-operative imaging, date of surgery, age at surgery, surgical details, pathology reports and post-operative follow up. RESULTS: 147 patients are presented within this series, with a total of 149 dermoid cysts removed. The age at surgery ranged from 2 months to 15 years, with an average age of 20 months. 88 (60%) children presented with lesions of the lateral brow. The next most common locations in descending order were midline nasal with 17 (11.5%), 11 occipital (7.5%), 9 temporal (6%), 6 midline frontal (4%), 5 medial brow/medial canthus (3.5%), 3 glabella (2%), 3 anterior fontanelle (2%), 3 posterior auricular (2%), and 2 parietal (1.5%). 53 of the 147 patients identified had preoperative imaging (36%). 21 patients underwent CT scan alone (40%), 18 underwent MRI alone (34%), 8 patients underwent combined CT and MRI scans (15%), 4 patients underwent ultrasound alone (7.5%), 1 patient had a plain radiograph (2%) and 1 patient underwent combined MRI and ultrasound (2%). In 4 cases (7.5%) the reported imaging was discordant with the operative findings. 3 of these cases failed to report attachment of the lesion to the underlying dura, and 1 failed to report a small breach of the inner table. To classify the depth of the lesions, we modified the system used by Hartley et al1 to describe midline nasal lesions and applied it to lesions in all locations within the head and neck. We divided the lesions into superficial (type 1), bony remodelling/stalk to suture (type 2), intraosseous (type 3), intracranial extradural (type 4) and intracranial intradural (type 5). In this study, 96 lesions were classified as superficial, 31 showed a significant stalk to a suture or bony remodelling, 8 lesions were interosseous, 10 lesions were intracranial extradural and 2 lesions were intracranial intradural. Average age at surgery was highest in group 2 and lowest in group 5. CONCLUSION: Based on our data, we propose a system using specific classification of anatomical location to assist in the prompt identification of high risk lesions and facilitate sound pre-operative planning.