Subperiosteal Hematoma of the Orbit With Osteoneogenesis

Although orbital surgeons work extensively with the bony structures of the orbit, the mechanism of bone healing has been a relatively neglected topic in the ophthalmic literature. A detailed understanding of the basic science of bone healing may pave the way for future innovations in surgical and medical management of problems of the orbital bones. One factor long thought to be important in bone healing is the hematoma overlying the fracture site.^1,2 Some authors have proposed that it acts as a mechanical and biochemical bridge for the migration of cells that will eventually form a callus.³ Others, however, have suggested that it does not play an important role in the healing of bone fractures.⁴ In this report we describe a case of a traumatic orbital subperiosteal hematoma with no obvious underlying fracture and bone formation within the hematoma. This stage in bone healing has rarely been captured histologically in a clinical setting, and appears to support the importance of the role of the hematoma in osteoneogenesis.

A 9-year-old boy was struck in the right periocular region by another boy's head while participating in a sporting event. There was no immediate change in vision or onset of symptoms other than a dull periorbital ache. The following morning the vision was subjectively blurred with increased periocular edema and ecchymosis. Five days following the injury the right eye became proptotic and the patient complained of vertical diplopia. Medical attention was sought. The patient's medical history was unremarkable; the only medication he was taking was acetaminophen for the ache in the right eye. Ophthalmologic examination revealed a best-corrected visual acuity of 20/40 OD and 20/20 OS. Pupils were symmetrically round and reactive without a relative afferent pupillary defect. Ocular motility was full in the left eye but significantly limited in supraduction and mildly limited in horizontal gaze in the right eye with negative forced ductions. External examination results were unremarkable, and fundus examination revealed a bulge in the supertemporal quadrant without any chorioretinal folds. Hertel exophthalmometry measured 19 mm OD and 15 mm OS with a base of 83 mm. There was 3 mm of hypoglobus in the right eye, and soft tissue could be palpated through the right upper eyelid. No resistance to retropulsion was appreciated.

Orbital computed tomography revealed a large right superior orbital mass contiguous with the orbital roof (Figure 1), suggesting a subperiosteal location with inferior displacement of the extraocular muscles and globe. A diagnosis of right superior orbital subperiosteal hematoma was made.

An urgent anterior orbitotomy through an upper eyelid crease incision was performed to improve vision and relieve the mass effect. Approximately 10 mm posterior to the superior orbital rim, a bluish-tinged mass was identified beneath a thin, bulging periosteum. A 6 × 5-mm patch of this periosteum was removed and submitted for histopathologic examination, with care being taken to avoid the underlying bone and the medial portion of the orbit to preserve the trochlea and neurovascular bundles. This resulted in expulsion of a large volume of liquefied and coagulated blood, which was completely irrigated. No fracture of the underlying bone was identified. On follow-up examination 2 weeks postoperatively, the patient's vision had returned to 20/25 OD, and he was orthophoric in all positions of gaze. Hertel exophthalmometry readings were 14 mm OD and 15 mm OS.

The gross specimen consisted of a triangular piece of black tissue measuring 6 × 5 × 4 mm. Light microscopic examination of routinely stained sections (Figure 2) revealed red blood cells and numerous polymorphonuclear and mononuclear inflammatory cells within a highly vascular connective tissue matrix, consistent with granulation tissue. In the granulation tissue was an island of osteoblasts within the lacunae of an osseous matrix. A diagnosis of subperiosteal hematoma with granulation tissue and foci of osteoneogenesis was made.

While the etiology, epidemiology, and histopathology of orbital hematomas have been described in the ophthalmic literature,^5-9 the relationship of these hematomas to bone repair has not been described. The sequence of events leading to osteoneogenesis after fractures has been a subject of interest in the nonophthalmic literature.^1,10,11 However, the exact biochemical and cellular mechanisms remain unclear and are a subject of controversy among clinicians and basic scientists. The healing process at the site of a fracture has conventionally been divided into 3 overlapping morphologic stages: (1) an initial inflammatory stage, (2) a reparative stage, and (3) a remodeling stage.¹ The duration of each stage is variable and depends on factors such as the size and severity of the bony defect, as well as mechanical and biochemical influences at the site of fracture. Our patient exhibited features of both stages 1 and 2. Stage 3 is a long-term process in which local forces and stress act to remodel and establish the final shape and contour of the healed fracture and surrounding bone, and was not considered in this case.

The first stage of bone fracture repair, the inflammatory stage, is characterized by hematoma formation and inflammatory exudate from ruptured blood vessels in bone, periosteum, and/or surrounding tissues.¹¹ Within hours, platelets aggregate at the site of the injury, releasing cytokines, which cause a marked inflammatory response and vasodilation. The ends of the ruptured blood vessels soon clot off, and the loss of nutrition results in necrosis of local tissues and osteocyte degeneration within 24 hours. Monocytes, multinuclear phagocytes, and osteoclasts soon engulf and digest this necrotic debris, including acellular bone.

The second stage, the reparative stage, is characterized by the formation of a fracture callus and its subsequent transformation to mature bone. A proliferation of blood vessels and loose connective tissue creates a bed of granulation tissue within the periosteal tissues and marrow. The sources of the pluripotent mesenchymal cells that transform into new bone are marrow, endosteum, periosteum, endothelial cells, circulating cells, and surrounding muscle.^12-18 These mesenchymal cells differentiate into osteoblasts, chondrocytes, and fibroblasts, each laying down its own respective extracellular matrix. The factors from fractured bone that attract and trigger the differentiation of these osteoprogenitor cells remain elusive, but data obtained to date indicate that these factors are numerous and have very complex interactions.¹⁹ The proteins most extensively studied have been the bone morphogenetic proteins^11,20,21 of the transforming growth factor β supergene family.

The process of differentiation of these primitive osteoprogenitor cells begins within 2 to 3 days and is most marked by 1 week. By the fourth day, nests of cartilage cells are apparent, and are soon replaced by bone tissue.^15,16 By the end of the first week, the tissue has matured to form a callus, which is of firmer consistency than a hematoma and provides a natural internal fixation for the fracture. Subsequently, mineralization and ossification of this new osteoid tissue progress, and new bone tissue becomes visible radiographically as flecks of radiodense material within approximately 1 to 3 weeks.

A hematoma occurring at the site of a bony fracture has long been suggested to play a critical role in bone healing,² and the absence of a fracture hematoma, whether due to surgical drainage or anticoagulation by heparin, has been shown to result in a decrease in callus production. In an interesting study using a rat animal model, Mizumo et al²² demonstrated that the hematoma surrounding a bony fracture has inherent osteogenic potential, and that the osteogenic factor arises from bone marrow.

Figure 2 shows an osseous fleck developing within a subperiosteal hematoma. The specimen was collected 5 days after the development of the hematoma. The surrounding chronic inflammatory exudate shows that this tissue is predominantly in the latter stages of the stage I inflammatory process. In our patient, osteoneogenesis occurred without any radiographic or clinical evidence of an underlying fracture. The absence of an underlying fracture raises questions about the hypothesis developed by Mizumo et al, that the osteogenic factor lies within the marrow. With further understanding of these processes, it may be possible in the future to devise less invasive and more satisfactory interventions in the management of orbital fractures.

We thank Daniel Aeschlimann, PhD, from the Department of Orthopedic Surgery at the University of Wisconsin, Madison, for his critical review of the manuscript.

Corresponding author and reprints: Daniel M. Albert, MD, MS, Department of Ophthalmology, F4/334 CSC, 600 Highland Ave, Madison, WI 53792-3220 (e-mail: albert@eyesee.ophth.wisc.edu).