Complex Palpebral Odontogenic Choristoma: A Reappraisal of the Origin of Teeth-bearing Periocular Lesions FREDERICK A. JAKOBIEC, JOHN NGUYEN, KENNETH MANDELL, AND AARON FAY ● PURPOSE: To describe the findings observed in a congenital ectopic tooth of the lower inner eyelid and explain its origin. ● DESIGN: Retrospective and interventional clinicopath- ologic study. ● METHODS: Histopathologic evaluation of sequentially excised specimens over 4 years and radiographic study of any associated dental abnormalities, compared with crit- ical reevaluation of similar reported cases in the litera- ture. ● RESULTS: The ectopic eyelid tooth clinically created a dome-shaped mass and microscopically displayed ad- vanced cytodifferentiation but incomplete morphogenesis with clearly identifiable ameloblasts, enamel, odonto- blasts, and dentin, surrounded by an intact dental sac. Swollen oral mucosal epithelium with associated seba- ceous Fordyce nodules, residual ciliated odontogenic epithelium, embryonic epithelial rests, and accessory mucinous and sero-mucinous accessory salivary glands were discovered in place of normal conjunctiva. The tarsus with meibomian glands was absent but islands of cartilage were present. ● CONCLUSIONS: Because of the variety of the ectopic tissues that were exhibited, a causative displacement of oral embryonic epithelium with extensive differentiating potentialities is posited. Two previously reported palpe- bral dental choristomas also had identical locations in the inferomedial eyelid. This is the site where the phakoma- tous choristoma may be encountered, but this unencap- sulated mass creates an eyelid thickening without a globoid configuration and infiltrates the inferior orbit. After a careful re-appraisal of published reports of orbital cysts bearing a tooth that were formerly diagnosed as “teratomas,” we have concluded that they developed from similar embryonic oral (buccal) epithelial displace- ments rather than from aberrant germ cells, based on the cysts’ linings and the associated accessory salivary glands in their walls. This view is further supported by the absence of heterologous teratomatous elements of endodermal or neuroectodermal derivation. A clinical and surgical approach to management and a new diagnos- tic schema for odontogenic choristomas are proposed. (Am J Ophthalmol 2009;147:531–543. © 2009 by Elsevier Inc. All rights reserved.) C ONGENITAL TUMORS OF THE EYELIDS ARE RARE, and for most ophthalmologists, the first condition to come to mind is the dermoid cyst of the upper-outer eyelid. While typically presenting in the eye- lid, this lesion is characteristically intraosseous in origin, developing and expanding behind the eyelid septum. We describe a choristoma indisputably developing in the eyelid and featuring a tooth; this lesion was more complex than those previously reported. 1,2 It mimicked in its location, but not in its clinical appearance or histopathological characteristics, a phakomatomas choristoma. 3–6 A displacement of stomodeal/embryonic oral epithelium with odontogenic potential into the presumptive infero- medial eyelid is offered as the explanation for this most unusual lesion. 7–9 Furthermore, a critical review of 6 earlier, adequately documented reports of “cystic terato- mas” of the orbit containing teeth 10 –15 leads us to a revised interpretation that these conditions more likely arose from embryonic oral epithelial ectopias rather than from aber- rant germ cells, as has been traditionally proposed. This conclusion is validated by the absence of all 3 germ layers (ectoderm, mesoderm, and endoderm) in such masses, and the to-date universal absence of fully formed teeth in rigorously defined primary orbital teratomas. METHODS A PATIENT SEEN ON REFERRAL IN AN ACADEMIC, INSTITU- tion-based ophthalmic plastic and orbital practice was evaluated for a congenital eyelid tumor. An initial surgical excision was followed by multiple eyelid excisions for reconstructions over a 4-year period. A thorough review of the medical records was performed and a panoramic dental x-ray study was obtained. Retrospective histopathologic evaluation of the initial tissue specimens was limited to hematoxylin and eosin-stained, paraffin-embedded sec- tions, attributable to failure to locate the tissue blocks. On Accepted for publication Sep 26, 2008. From the Department of Ophthalmology, the David G. Cogan Ophthal- mic Pathology Laboratory (F.A.J., K.M.), and the Division of Oculoplastic and Reconstructive Surgery, Harvard Medical School, Massachusetts Eye and Ear Infirmary (J.N., A.F.), Boston, Massachusetts. Inquiries to Frederick A. Jakobiec, David G. Cogan Ophthalmic Pathology Laboratory, Massachusetts Eye and Ear Infirmary, 243 Charles Street, Room 321, Boston, MA 02114; e-mail: Fred_Jakobiec@meei. harvard.edu © 2009 BY ELSEVIER INC.ALL RIGHTS RESERVED. 0002-9394/09/$36.00 531 doi:10.1016/j.ajo.2008.09.033