Orbital wall fractures: Evaluation and management J. DAVID OSGUTHORPE, MD, Charleston, South Carolina
Over a 49-month period, 121 orbital wall fractures were treated and 92 were followed for a median of 6.5 months (minimum, 3 months). Associated injuries included a 17% incidence of serious globe or optic nerve injuries and 13% incidence of lacrimal drainage disruption. Diplopia occurred in 23% and dystopia in 11%. Management was by observation alone in 14% and exploration in the remainder, with layered gelfilm for defects smaller than 4 cm2,alloplastic sheeting for defects to 6 cm2,and outer cortex of parietal bone for larger dehiscences. There were no decrements in vision from operation, dystopias were corrected to within 2 mm of normal, and diplopia persisted only in those with extraocular muscle paresis. There was no benefit to exploration of orbital wall defects smaller than 2.5 cm2or with reduction of other midfacial fractures (e.g., malar) when neither dystopia nor entrapment was present, because defects not obturated in such CaSeS had no SeqUelae. (OTOLARYNGOL HEAD NECK SURG 1991:105:702.)
T h e bony orbit can be divided into three sections, with the anterior third being a thick rim, a middle third of four thin walls (lamina papyracea, orbital roof, floor, and lateral wall), and a posterior third or apex of the moderately thick sphenoid wings that enclose the optic canal and superior orbital fissure. Only 10% of orbital fractures of the middle third of the orbit are “pure” (the remainder, “impure”, are associated with zygomatic or other facial fractures), and a preponderance are comminuted rather than the discrete fracture lines common with rim Two theories on the origin of “pure,” or isolated, fractures of the middle third of the orbit are: (1) soft tissue “hydraulic” pressure transmitted from blunt trauma to the lids and the globe, and (2) a “buckling” from pressure on the orbital rim.4,5The orbital floor is weakest and involved in 85% of the fractures, the medial wall (lamina papyracea) in 49%, the roof in 18%, and the lateral wall in 8%.’,6,7 Of the floor fractures, most occur in the medial to mid-floor, with 20% affecting the thicker bone lateral to the infraorbital nerve.
From the Department of Otolaryngology and Communicative Sciences, Medical University of South Carolina. Presented at the Annual Meeting of the American Acadamy of Otolaryngology-Head and Neck Surgery, San Diego, Calif., Sept. 9-13, 1990. Received for publication Nov. 30, 1990; revision received Feb. 12, 1991; accepted Feb. 21, 1991. Reprint requests: J. David Osguthorpe, MD, 171 Ashley Ave., Charleston, SC 29425. 23 I 1I29225
702
Fractures of the posterior third are unusual, as the midorbit absorbs (by fracturing) force transmitted from the orbital rims. The commonest facial fractures associated with orbital wall disruptions are, in order: of the zygoma, of the nasoethmoid complex, of the midface (LeFort), and of the skull. Concomitant serious globe or adnexal injuries (hyphema, retinal detachment, globe perforation, optic nerve damage, lacrimal drainage disruption) occur in 2.7% to 18%, increasing to 18% to 69% if minor injuries are On clincial examination, small to medium fractures of the orbital walls may be difficult to differentiate from posttraumatic edema and/ or hemorrhage. Even substantial orbital floor fractures do not commonly manifest as enophthalmos until edema resolves. Early post-injury vertical diplopia can result not only from inferior rectus muscle entrapment, but also from soft tissue herniation into the antrum, extraocular muscle contusion, or oculomotor paresis. Most medial floor fractures are asymptomatic, as enophthalmos or medial rectus entrapment are infrequent, though some manifest emphysema from air entering from the nasal cavity. Roof fractures are commonly associated with a bruise or laceration over the brow and disproportionate upper eyelid swelling, occasionally with inferior and anterior displacement of the globe. Numbness in the supraorbital nerve distribution is common, and loss of vision from a fracture line or concussion extending to the optic canal, occurring in 3% to 4.9% of orbital fractures, is most frequent in roof fractures. Lateral wall fractures are usually asymptomatic unless the superior orbital fissure or optic canal
Volume 105 Number 5 November 1991
is encroached, in which case the signs emanate from the cranial nerve(s) involved. The indications for and the timing of orbital fracture repair are controversial. In the 1950’s and 1960’s, the dictum was early exploration of all but linear, nondisplaced wall fractures, and obturation of any defect encountered. Advocates Converse and Smith reported a 40% incidence of persisting diplopia and 22% incidence of enophthalmos with uncorrected blow-out fractures, and emphasized the substantial difficulties with late correction of these sequelae as a result of scarring and bony distortion. 1.4.10.11 In the 1970’s, Putterman and others urged observation alone for 4 to 6 months after “pure” blow-out fractures, citing the potential complications of surgical exploration and less than 5% persistence of diplopia or e n ~ p h t h a l m o s . ~ . ~ ~ - ’ ~ Many surgical approaches to orbital fractures have been promulgated, with most involving the orbital floor. Floor fractures can be approached through a transconjunctival (inferior fornix) incision with a lateral canthotomy extension for wider exposure, an infraciliary skin/ muscle blepharoplasty route, or a skin incision directly over the inferior orbital rim.2.9,11.15 Medial wall fractures are accessed through a Lynch-type incision. Roof and lateral wall fractures can be approached through appropriately placed incisions below the eyebrow, though postoperative lid edema and suboptimal scarring can be a problem with long incisions and most use a coronal incision and forehead degloving approach that can also access the medial orbital wall.16-18 Many materials have been suggested for orbital wall defects, and partly depend on the defect size. Most use autogenous material for the primary or secondary repair of large defects. The ideal material for medium to small defects can be divided into organic and nonorganic materials, 1.2.1 I . 12.18-20 In the organic group, autogenous bone (rib, iliac crest, outer table of parietal bone), cartilage (rib, septal, auricular), and fascia (lata, temporalis) have been used, with the latter only for small defects. Homologous irradiated cartilage, decalcified bone, lyophilized human dura, and porcine dermis or denatured collagen have been tried. Recent reports disclosed unexpected resorption of the former two, whereas in the latter three resorption after 1 to 3 months is expetted. 11.19.20,22 Alloplastic materials include tantalum, Teflon, polyethylene, methylmethacrylate, silicone rubber, supramid, marlex, and hydroxyapatite. Proponents of autogenous materials note that as vascularization develops, new bone is usually formed, extrusion is negligible and there is resistance to infection if exposed to a sinus or nasal cavity.1.20,21Advocates of alloplastic materials point to a low infection and extrusion rate
Orbital wall fractures: Evaluation and management
703
Table 1. Wall orbital exploration lndicatlons
Extraocular muscle entrapment Dystopia of 2 mm or more “Large” bony defect (? 2.5 to 3.0 square cm) Reduction of other midfacial fractures involving the orbit
(less than 3%), lack of donor site morbidity, and the ease of obtaining and shaping such materials to a defect.7.9.11.12.15 CASE REPORTS
Over a 49-month period, 121 orbital wall fractures were treated and 92 were followed in 76 patients by physical examination for a minimum of 3 months, with a median of 6.5 months. The median patient age was 32 years, with 14% having “pure” fractures and the remainder associated within order of frequency-malar (47%), LeFort (22%), and nasoethmoid and skull fractures. Evaluation included ophthalmologic and radiographic examinations consisting of plain films for suspected “pure” blowout fractures (9%) and finecut coronal computed tomography for the remainder, with sagittal reconstructions to estimate the size of complex defects. Pertinent findings included 17% with a decrement in vision from serious globe (hyphema, retinal detachment, laceration) or optic nerve damage, with 5% sustaining permanent impairment and another having 5% loss of vision. Eyelid lacerations occured in 35% and lacrimal drainage disruption necessitating reanastomosis and/or stenting in 13% (4% required secondary repairs for persisting epiphora). Twentythree percent of fractures caused diplopia- 17% of these were from muscle entrapment and the remainder of neuropraxic origin. In 11% of patients, 3mm or more of globe dystopia developed, usually ptosis and enophthalmos from a bony floor defect, within 2 weeks of injury as edema/hematoma resolved. Approximately two thirds of our study was retrospective, and we did not change the surgical indications while accumulating the final third. Management consisted of observation alone in 14% and surgical exploration in the remainder, primarily though a subciliary skin muscle flap (77%) or transconjunctival access (12%). The wall defect was obturated in 29%. Layered gelfilm was usually selected for defects less than 4 cm2, silicone rubber or Teflon sheeting (fixed to bone with a permanent suture or wire) for defects to 6 cm, and outer cortex of parietal bone for larger dehiscences. Per ophthalmologic examination, no case operated on experienced a further decrement of vision, and the diplopia that persisted in only 2 of 16 cases resulted from extraocular muscle pareses. All globe dystopias were corrected to within 2 mm of normal, regardless of the size of the orbital wall defect(s), and no implant has extruded to date, though inflammation around two alloplastic and a bone implant required second courses of antibiotics after the initial perioperative course.
704 OSGUTHORPE
OtolaryngologyHead and Neck Surgery
Flg. I. Coronal computed tomograms of identical bony floor fractures between the infraorbital nerve canal and the lamina papyraceo.with A showing periorbital disruption and a ”trapdoor“ displacement of the bony fragment with substantial fat herniation, and B showing intact periorbita with a minimally displaced bony fragment adherent to it [maxillary sinus opacified with blood].
DISCUSSION
In the past few years, the diverse viewpoints on orbital fracture management have moved closer to a consensus. Most that have advocated an early. aggressive approach agree that a 10 to 14 day delay in exploration
does not compromise repair because the bony fragments are not fused or scarring/contracture establi~hed.’.‘“’.~”~~.”.’’ Many patients with orbital wall fractures have diplopia and/or dystopia from edema and hematoma, which substantially resolve in the first 2
Volume 105 Number 5 November 1991
Orbital wall fractures: Evaluation and management 705
Fig. 2. Note depressed and hypertrophic scar (on left] from orbital floor exploration via an inferior rim incision, despite use of a prominent natural skin crease [see normal right side].
weeks. Rather than prolonged observation. Putternian and Millman" now favor exploration if diplopia or enophthalmos persist at 3 weeks. though librous bony union and soft tissue adhesions can be substantial by that time, notably in children.".'" There were no serious consequences of surgical treatment in this series, yet the benefit to some patients is questionable. If one subtracts the 14% without surgery, assumes the 29% receiving implants needed them. and the "periorbital mobilization" or "realignment of bony fragments" recorded in another 22% of operative notes were beneficial, 35% of patients had orbital exploration with no obvious benefit. Our general surgery colleagues accept a 30% rate of normal appendixes found at laparotoniy for acute appendicitis. but a siniilar guideline for orbital wall fractures requiring open reduction or an implant is not available. The common indications for exploration (and. heretofore, mine) are enumerated in Table I .-'.".""3.".'3 Extraocular muscle entrapment and/or globe dystopia of 3 mm or more mandate intervention unless contraindicated by unstable vision or pertinent medical factors. Though 3 nim of dystopia is the commonly accepted threshold for cosmetic deformity, 2 nim within 2 weeks of trauma, with residue edema and scarring yet to occur. is a reasonable indication for an implant. In the absence
of the aforementioned or unequivocal entrapment, delaying surgery for I to 2 weeks also allows any muscle
hematonia mimicking entrapment to subside and the ophthalmologist a better opportunity to detect neuropraxia. Many authors advocate early exploration of "large" wall defects. without specifying size. The rationale is that herniation of soft tissue will eventually cause signilicant dystopia as edema resolves and scarring occurs. Conservative proponents counter that removal of the entire orbital Hoor in decompression for Graves' disease only causes 3 m m of enophthalmos."' The analogy has two Haws-namely that the mucopolysaccharide deposition and fibrosis in Graves' disease makes the orbital soft tissues stiff. and post-traumatic enophthalnos is a combination of the orbital wall defect, traumatic fat atrophy. and scar retraction.'.'" Paskert et al.' stated that orbital floor defects of more than 2.5 to 3 cni' are much more likely to evoke enophthalnios than smaller gaps."," In my modest cumulation. no enophthalnios of 2. mm or more resulted from unobturated defects smaller than 3 cm-' as estimated by coronal computed toniography with sagittal reconstruction (recent 3-D reconstruction software packages make the calculation of defects straightforward), or as measured intraoperatively.
OtolaryngologyHead and Neck Surgery
706 OSGUTHORPE
Defects of the medial, lateral, or superior orbital walls must be much larger than floor defects to cause dystopia, because the latter is aggravated by gravity. For orbital roof dehiscences, the potential problem is proptosis from an encephalocele (one case in this series) rather than enophthalmos. As an alternative to computed tomography, some use canine fossa antroscopy with sinus endoscopes to inspect a floor fracture documented by plain films and to assess herniation of soft tissue.2.22Periorbital disruption with fat herniation is more apt to cause enophthalmos than a similarly sized bony defect with intact periorbita (Fig. 1, A and B ) . It is common practice to explore an orbital fracture documented by radiography at the time of reducing a zygomatic, LeFort II/III, or nasoethmoid fracture, because most fractures compress and their disimpaction tends to increase an orbital wall defect. Though a forced duction test should be performed after reduction of midfacial fractures to verify that extraocular muscle motility was not compromised, orbital exploration in such circumstances without entrapment or increased dystopia was of no apparant benefit in this series because no bony defect sufficiently large to warrant an implant was encountered. Though floor exploration at the time of stabilizing an infraorbital rim in a zygomatic fracture has little further potential for complication, the recent trend to reduce malar fractures with a plate at the frontozygomatic suture-often with a wire at the zygomaticomaxillary buttress-requires a separate incision for inspection of the orbital f l o ~ r . ’ , ~ ’ , ~ ~ Of the three incisions for accessing floor fractures, I favor the transconjunctival approach, using a lateral canthotomy for visualization of large defects or accessing inferior rim fracture^.^.^.'^ The incidence of ectropion has been reported as less than 1% vs. 8.9% with a subciliary incision, with little possibility of cutaneous scarring. I5 The transcutaneous inferior rim route was abandoned because of depressed and1or easily visible scars in 2 of 5 cases despite “stairstepping” the incision between the skin and orbicularis muscle (Fig. 2). In this series, diverse types of implants were used and, as with all other reports to date, the alloplasts were not followed long enough. I have removed a silicone rubber implant extruding through the lower lid that had bee-i placed 8 years earlier. There were no late sequela to obturating orbital floor or medial wall defects of less than 4.0 cm2 (estimated at surgery) with 2 to 3 layers of gelfilm. Lyophilized dura, porcine dermis, and other resorbing materials may be satisfactory alternatives. Parkin et aLL9and Iro2* have advocated an antral balloon to temporarily support the pliable gelfilm im-
plant in medium to large orbital defects, but substantial case experience is lacking. The largest floor defect I have treated this way (10 days of antral balloon support) was 5 cm’. With the discomfort of the antral balloon, and the potential for premature deflation or infection, firmer materials are chosen for defects larger than 4 cm2. For 4- to 6-cm2defects, cartilage (concha1 bowl, septum) and unfractured anterior antral wall are autologous alternatives to the commonly used silicone rubber, Teflon, or marlex sheets. If any of the latter are used, migration is minimized by fixing the implant to bone with a permanent non-metallic suture. It is essential to delineate all edges of the fracture, which may be difficult posteriorly, to support the implant and avoid herniation of orbital fat. Because the sinuses and possibly the dehiscence of a fracture will enlarge, orbital fractures in children are obturated with absorbing or autogenous materials. Membranous bone undergoes the least resorption, and the outer table of parietal bone from the nondominant hemisphere is widely used. Cranial bone is brittle, however, and must be left attached to periosteum for sculpting to curved defects, or the more pliable rib grafts used. Large orbital defects, whether primary or with reconstruction, are best bridged with autogenous material. 1 ~ 7 ~ ’ 7 * 1 8 ~ 2 5 REFERENCES 1. Antonyshyn 0, Gruss JS, Galbraith DJ, Hunvitz JJ. Complex
orbital fractures: a critical analysis of immediate bone graft reconstruction. Ann Plast Surg 1989;22:220-33. 2. Camuzard JF, Raspaldo H , Santini J , Vaille G, Demond F. Les fractures du plancher de l’orbite. Rev Stomatol Chir Maxillofac 1988;89:204-9. 3. Funk GF, Stanley RB, Becker TS. Reversible visual loss due to impacted lateral orbital wall fractures. Head Neck 1989;11:295300. 4. Paskert JP, Manson PN, Iliff NT. Nasoethmoidal and orbital fractures. Clin Plast Surg 1988;15:209-23. 5. Straker CA, Hill JC. Management of orbital blow-out fractures. S Afr Med J 1989;76:535-7. 6. Ioannides C, Treffers W, Rutten M, Noverraz P. Ocular injuries associated with fractures involving the orbit. J Craniomaxillofac Surg 1988; 16: 157-9. 7. Dufresne CR, Manson PN, Iliff NT. Early and late complications of orbital fractures. Clin Plast Surg 1988;15:239-53. 8. Arts HA, Eisele DW, Duckert LG. Intraocular pressure as an index of ocular injury in orbital fractures. Arch Otolaryngol Head Neck Surg 1989;115:213-6. 9. Holt GR, Holt JE. Management of orbital trauma and foreign bodies. Otolaryngol Clin North Am 1988;21:35-52. 10. Luce EA, Tubb TD, Moore AM. Review of 1000 major facial fractures and associated injuries. Plast Reconstr Surg 1979;63:26-30. 1 I . Scapini DA, Mathog RH. Repair of orbital floor fractures with marlex mesh. Laryngoscope 1989;99:697-701. 12. Putterman AM, Millman AL. Custom orbital implant in the
Volume 105 Number 5 November 1991
13. 14.
15. 16.
17. 18.
19.
repair of late posttraumatic enophthalmos. Am J Ophthalmol 1989;108:153-9. Gilbard SV, Mafee MF, Logouros PA, Langer BG. Orbital blowout fractures. Ophthalmology 1985;92:1523-8. Kersten RC, Kulwin DR. Selective approach to surgery for delayed enophthalmos (letter to the editor). Arch Otolaryngol Head Neck Surg 1989;115:634. Maniglia AJ. Conjunctival approach for the repair of pure orbital blowout fractures. Otolaryngol Clin N Am 1983;16:575-83. Raveh J, Vuillemin T. The surgical one-stage management of combined cranio-maxillo-facial and frontobasal fractures. J Craniomaxillofac Surg 1988;16: 160-72. Jackson IT. Classification and treatment of orbitozygomatic and orbitoethmoid fractures. Clin Plast Surg 1989;16:77-91. Stanley RB. The temporal approach to impacted lateral orbital wall fractures. Arch Otolaryngol Head Neck Surg 1988;114: 550-3. Parkin JL, Stevens MH, Stringham JC. Absorbable gelatin film
Orbital wall fractures: Evaluation and management 707
20. 2I . 22. 23.
24.
25.
verse silicone rubber sheeting in orbital fracture treatment. Laryngoscope 1987;97:1-3. Webster K. Orbital floor repair with lyophilized porcine dermis. Oral Surg Oral Med Oral Pathol 1988;65:161-4. Wolfe SA. Treatment of post-traumatic orbital deformities. Clin Plast Surg 1988;15:225-38. Iro H. Funktionelle Ergebnisse nach operativer versorgung von isolierten orbitabodenfrakturen. HNO 1989;37:292-4. Jordan DR, White GL, Anderson RL, Thiese SM. Orbital emphysema: a potentially blinding complication following orbital fractures. Ann Emerg Med 1988;17:853-5. Manson PN, Clifford CM, Su CT, Iliff NT, Morgan R . Mechanisms of global support and posttraumatic enophthalmos. Plast Reconstr Surg 1986;77:193-202. Mathog RH, Hillstrom RP, Nesi FA. Surgical correction of enophthalmos and diplopia. Arch Otolaryngol Head Neck Surg 1989;115: 169-78.
Over a 49-month period, 121 orbital wall fractures were treated and 92 were followed for a median of 6.5 months (minimum, 3 months). Associated injuries included a 17% incidence of serious globe or optic nerve injuries and 13% incidence of lacrimal drainage disruption. Diplopia occurred in 23% and dystopia in 11%. Management was by observation alone in 14% and exploration in the remainder, with layered gelfilm for defects smaller than 4 cm2,alloplastic sheeting for defects to 6 cm2,and outer cortex of parietal bone for larger dehiscences. There were no decrements in vision from operation, dystopias were corrected to within 2 mm of normal, and diplopia persisted only in those with extraocular muscle paresis. There was no benefit to exploration of orbital wall defects smaller than 2.5 cm2or with reduction of other midfacial fractures (e.g., malar) when neither dystopia nor entrapment was present, because defects not obturated in such CaSeS had no SeqUelae. (OTOLARYNGOL HEAD NECK SURG 1991:105:702.)
T h e bony orbit can be divided into three sections, with the anterior third being a thick rim, a middle third of four thin walls (lamina papyracea, orbital roof, floor, and lateral wall), and a posterior third or apex of the moderately thick sphenoid wings that enclose the optic canal and superior orbital fissure. Only 10% of orbital fractures of the middle third of the orbit are “pure” (the remainder, “impure”, are associated with zygomatic or other facial fractures), and a preponderance are comminuted rather than the discrete fracture lines common with rim Two theories on the origin of “pure,” or isolated, fractures of the middle third of the orbit are: (1) soft tissue “hydraulic” pressure transmitted from blunt trauma to the lids and the globe, and (2) a “buckling” from pressure on the orbital rim.4,5The orbital floor is weakest and involved in 85% of the fractures, the medial wall (lamina papyracea) in 49%, the roof in 18%, and the lateral wall in 8%.’,6,7 Of the floor fractures, most occur in the medial to mid-floor, with 20% affecting the thicker bone lateral to the infraorbital nerve.
From the Department of Otolaryngology and Communicative Sciences, Medical University of South Carolina. Presented at the Annual Meeting of the American Acadamy of Otolaryngology-Head and Neck Surgery, San Diego, Calif., Sept. 9-13, 1990. Received for publication Nov. 30, 1990; revision received Feb. 12, 1991; accepted Feb. 21, 1991. Reprint requests: J. David Osguthorpe, MD, 171 Ashley Ave., Charleston, SC 29425. 23 I 1I29225
702
Fractures of the posterior third are unusual, as the midorbit absorbs (by fracturing) force transmitted from the orbital rims. The commonest facial fractures associated with orbital wall disruptions are, in order: of the zygoma, of the nasoethmoid complex, of the midface (LeFort), and of the skull. Concomitant serious globe or adnexal injuries (hyphema, retinal detachment, globe perforation, optic nerve damage, lacrimal drainage disruption) occur in 2.7% to 18%, increasing to 18% to 69% if minor injuries are On clincial examination, small to medium fractures of the orbital walls may be difficult to differentiate from posttraumatic edema and/ or hemorrhage. Even substantial orbital floor fractures do not commonly manifest as enophthalmos until edema resolves. Early post-injury vertical diplopia can result not only from inferior rectus muscle entrapment, but also from soft tissue herniation into the antrum, extraocular muscle contusion, or oculomotor paresis. Most medial floor fractures are asymptomatic, as enophthalmos or medial rectus entrapment are infrequent, though some manifest emphysema from air entering from the nasal cavity. Roof fractures are commonly associated with a bruise or laceration over the brow and disproportionate upper eyelid swelling, occasionally with inferior and anterior displacement of the globe. Numbness in the supraorbital nerve distribution is common, and loss of vision from a fracture line or concussion extending to the optic canal, occurring in 3% to 4.9% of orbital fractures, is most frequent in roof fractures. Lateral wall fractures are usually asymptomatic unless the superior orbital fissure or optic canal
Volume 105 Number 5 November 1991
is encroached, in which case the signs emanate from the cranial nerve(s) involved. The indications for and the timing of orbital fracture repair are controversial. In the 1950’s and 1960’s, the dictum was early exploration of all but linear, nondisplaced wall fractures, and obturation of any defect encountered. Advocates Converse and Smith reported a 40% incidence of persisting diplopia and 22% incidence of enophthalmos with uncorrected blow-out fractures, and emphasized the substantial difficulties with late correction of these sequelae as a result of scarring and bony distortion. 1.4.10.11 In the 1970’s, Putterman and others urged observation alone for 4 to 6 months after “pure” blow-out fractures, citing the potential complications of surgical exploration and less than 5% persistence of diplopia or e n ~ p h t h a l m o s . ~ . ~ ~ - ’ ~ Many surgical approaches to orbital fractures have been promulgated, with most involving the orbital floor. Floor fractures can be approached through a transconjunctival (inferior fornix) incision with a lateral canthotomy extension for wider exposure, an infraciliary skin/ muscle blepharoplasty route, or a skin incision directly over the inferior orbital rim.2.9,11.15 Medial wall fractures are accessed through a Lynch-type incision. Roof and lateral wall fractures can be approached through appropriately placed incisions below the eyebrow, though postoperative lid edema and suboptimal scarring can be a problem with long incisions and most use a coronal incision and forehead degloving approach that can also access the medial orbital wall.16-18 Many materials have been suggested for orbital wall defects, and partly depend on the defect size. Most use autogenous material for the primary or secondary repair of large defects. The ideal material for medium to small defects can be divided into organic and nonorganic materials, 1.2.1 I . 12.18-20 In the organic group, autogenous bone (rib, iliac crest, outer table of parietal bone), cartilage (rib, septal, auricular), and fascia (lata, temporalis) have been used, with the latter only for small defects. Homologous irradiated cartilage, decalcified bone, lyophilized human dura, and porcine dermis or denatured collagen have been tried. Recent reports disclosed unexpected resorption of the former two, whereas in the latter three resorption after 1 to 3 months is expetted. 11.19.20,22 Alloplastic materials include tantalum, Teflon, polyethylene, methylmethacrylate, silicone rubber, supramid, marlex, and hydroxyapatite. Proponents of autogenous materials note that as vascularization develops, new bone is usually formed, extrusion is negligible and there is resistance to infection if exposed to a sinus or nasal cavity.1.20,21Advocates of alloplastic materials point to a low infection and extrusion rate
Orbital wall fractures: Evaluation and management
703
Table 1. Wall orbital exploration lndicatlons
Extraocular muscle entrapment Dystopia of 2 mm or more “Large” bony defect (? 2.5 to 3.0 square cm) Reduction of other midfacial fractures involving the orbit
(less than 3%), lack of donor site morbidity, and the ease of obtaining and shaping such materials to a defect.7.9.11.12.15 CASE REPORTS
Over a 49-month period, 121 orbital wall fractures were treated and 92 were followed in 76 patients by physical examination for a minimum of 3 months, with a median of 6.5 months. The median patient age was 32 years, with 14% having “pure” fractures and the remainder associated within order of frequency-malar (47%), LeFort (22%), and nasoethmoid and skull fractures. Evaluation included ophthalmologic and radiographic examinations consisting of plain films for suspected “pure” blowout fractures (9%) and finecut coronal computed tomography for the remainder, with sagittal reconstructions to estimate the size of complex defects. Pertinent findings included 17% with a decrement in vision from serious globe (hyphema, retinal detachment, laceration) or optic nerve damage, with 5% sustaining permanent impairment and another having 5% loss of vision. Eyelid lacerations occured in 35% and lacrimal drainage disruption necessitating reanastomosis and/or stenting in 13% (4% required secondary repairs for persisting epiphora). Twentythree percent of fractures caused diplopia- 17% of these were from muscle entrapment and the remainder of neuropraxic origin. In 11% of patients, 3mm or more of globe dystopia developed, usually ptosis and enophthalmos from a bony floor defect, within 2 weeks of injury as edema/hematoma resolved. Approximately two thirds of our study was retrospective, and we did not change the surgical indications while accumulating the final third. Management consisted of observation alone in 14% and surgical exploration in the remainder, primarily though a subciliary skin muscle flap (77%) or transconjunctival access (12%). The wall defect was obturated in 29%. Layered gelfilm was usually selected for defects less than 4 cm2, silicone rubber or Teflon sheeting (fixed to bone with a permanent suture or wire) for defects to 6 cm, and outer cortex of parietal bone for larger dehiscences. Per ophthalmologic examination, no case operated on experienced a further decrement of vision, and the diplopia that persisted in only 2 of 16 cases resulted from extraocular muscle pareses. All globe dystopias were corrected to within 2 mm of normal, regardless of the size of the orbital wall defect(s), and no implant has extruded to date, though inflammation around two alloplastic and a bone implant required second courses of antibiotics after the initial perioperative course.
704 OSGUTHORPE
OtolaryngologyHead and Neck Surgery
Flg. I. Coronal computed tomograms of identical bony floor fractures between the infraorbital nerve canal and the lamina papyraceo.with A showing periorbital disruption and a ”trapdoor“ displacement of the bony fragment with substantial fat herniation, and B showing intact periorbita with a minimally displaced bony fragment adherent to it [maxillary sinus opacified with blood].
DISCUSSION
In the past few years, the diverse viewpoints on orbital fracture management have moved closer to a consensus. Most that have advocated an early. aggressive approach agree that a 10 to 14 day delay in exploration
does not compromise repair because the bony fragments are not fused or scarring/contracture establi~hed.’.‘“’.~”~~.”.’’ Many patients with orbital wall fractures have diplopia and/or dystopia from edema and hematoma, which substantially resolve in the first 2
Volume 105 Number 5 November 1991
Orbital wall fractures: Evaluation and management 705
Fig. 2. Note depressed and hypertrophic scar (on left] from orbital floor exploration via an inferior rim incision, despite use of a prominent natural skin crease [see normal right side].
weeks. Rather than prolonged observation. Putternian and Millman" now favor exploration if diplopia or enophthalmos persist at 3 weeks. though librous bony union and soft tissue adhesions can be substantial by that time, notably in children.".'" There were no serious consequences of surgical treatment in this series, yet the benefit to some patients is questionable. If one subtracts the 14% without surgery, assumes the 29% receiving implants needed them. and the "periorbital mobilization" or "realignment of bony fragments" recorded in another 22% of operative notes were beneficial, 35% of patients had orbital exploration with no obvious benefit. Our general surgery colleagues accept a 30% rate of normal appendixes found at laparotoniy for acute appendicitis. but a siniilar guideline for orbital wall fractures requiring open reduction or an implant is not available. The common indications for exploration (and. heretofore, mine) are enumerated in Table I .-'.".""3.".'3 Extraocular muscle entrapment and/or globe dystopia of 3 mm or more mandate intervention unless contraindicated by unstable vision or pertinent medical factors. Though 3 nim of dystopia is the commonly accepted threshold for cosmetic deformity, 2 nim within 2 weeks of trauma, with residue edema and scarring yet to occur. is a reasonable indication for an implant. In the absence
of the aforementioned or unequivocal entrapment, delaying surgery for I to 2 weeks also allows any muscle
hematonia mimicking entrapment to subside and the ophthalmologist a better opportunity to detect neuropraxia. Many authors advocate early exploration of "large" wall defects. without specifying size. The rationale is that herniation of soft tissue will eventually cause signilicant dystopia as edema resolves and scarring occurs. Conservative proponents counter that removal of the entire orbital Hoor in decompression for Graves' disease only causes 3 m m of enophthalmos."' The analogy has two Haws-namely that the mucopolysaccharide deposition and fibrosis in Graves' disease makes the orbital soft tissues stiff. and post-traumatic enophthalnos is a combination of the orbital wall defect, traumatic fat atrophy. and scar retraction.'.'" Paskert et al.' stated that orbital floor defects of more than 2.5 to 3 cni' are much more likely to evoke enophthalnios than smaller gaps."," In my modest cumulation. no enophthalnios of 2. mm or more resulted from unobturated defects smaller than 3 cm-' as estimated by coronal computed toniography with sagittal reconstruction (recent 3-D reconstruction software packages make the calculation of defects straightforward), or as measured intraoperatively.
OtolaryngologyHead and Neck Surgery
706 OSGUTHORPE
Defects of the medial, lateral, or superior orbital walls must be much larger than floor defects to cause dystopia, because the latter is aggravated by gravity. For orbital roof dehiscences, the potential problem is proptosis from an encephalocele (one case in this series) rather than enophthalmos. As an alternative to computed tomography, some use canine fossa antroscopy with sinus endoscopes to inspect a floor fracture documented by plain films and to assess herniation of soft tissue.2.22Periorbital disruption with fat herniation is more apt to cause enophthalmos than a similarly sized bony defect with intact periorbita (Fig. 1, A and B ) . It is common practice to explore an orbital fracture documented by radiography at the time of reducing a zygomatic, LeFort II/III, or nasoethmoid fracture, because most fractures compress and their disimpaction tends to increase an orbital wall defect. Though a forced duction test should be performed after reduction of midfacial fractures to verify that extraocular muscle motility was not compromised, orbital exploration in such circumstances without entrapment or increased dystopia was of no apparant benefit in this series because no bony defect sufficiently large to warrant an implant was encountered. Though floor exploration at the time of stabilizing an infraorbital rim in a zygomatic fracture has little further potential for complication, the recent trend to reduce malar fractures with a plate at the frontozygomatic suture-often with a wire at the zygomaticomaxillary buttress-requires a separate incision for inspection of the orbital f l o ~ r . ’ , ~ ’ , ~ ~ Of the three incisions for accessing floor fractures, I favor the transconjunctival approach, using a lateral canthotomy for visualization of large defects or accessing inferior rim fracture^.^.^.'^ The incidence of ectropion has been reported as less than 1% vs. 8.9% with a subciliary incision, with little possibility of cutaneous scarring. I5 The transcutaneous inferior rim route was abandoned because of depressed and1or easily visible scars in 2 of 5 cases despite “stairstepping” the incision between the skin and orbicularis muscle (Fig. 2). In this series, diverse types of implants were used and, as with all other reports to date, the alloplasts were not followed long enough. I have removed a silicone rubber implant extruding through the lower lid that had bee-i placed 8 years earlier. There were no late sequela to obturating orbital floor or medial wall defects of less than 4.0 cm2 (estimated at surgery) with 2 to 3 layers of gelfilm. Lyophilized dura, porcine dermis, and other resorbing materials may be satisfactory alternatives. Parkin et aLL9and Iro2* have advocated an antral balloon to temporarily support the pliable gelfilm im-
plant in medium to large orbital defects, but substantial case experience is lacking. The largest floor defect I have treated this way (10 days of antral balloon support) was 5 cm’. With the discomfort of the antral balloon, and the potential for premature deflation or infection, firmer materials are chosen for defects larger than 4 cm2. For 4- to 6-cm2defects, cartilage (concha1 bowl, septum) and unfractured anterior antral wall are autologous alternatives to the commonly used silicone rubber, Teflon, or marlex sheets. If any of the latter are used, migration is minimized by fixing the implant to bone with a permanent non-metallic suture. It is essential to delineate all edges of the fracture, which may be difficult posteriorly, to support the implant and avoid herniation of orbital fat. Because the sinuses and possibly the dehiscence of a fracture will enlarge, orbital fractures in children are obturated with absorbing or autogenous materials. Membranous bone undergoes the least resorption, and the outer table of parietal bone from the nondominant hemisphere is widely used. Cranial bone is brittle, however, and must be left attached to periosteum for sculpting to curved defects, or the more pliable rib grafts used. Large orbital defects, whether primary or with reconstruction, are best bridged with autogenous material. 1 ~ 7 ~ ’ 7 * 1 8 ~ 2 5 REFERENCES 1. Antonyshyn 0, Gruss JS, Galbraith DJ, Hunvitz JJ. Complex
orbital fractures: a critical analysis of immediate bone graft reconstruction. Ann Plast Surg 1989;22:220-33. 2. Camuzard JF, Raspaldo H , Santini J , Vaille G, Demond F. Les fractures du plancher de l’orbite. Rev Stomatol Chir Maxillofac 1988;89:204-9. 3. Funk GF, Stanley RB, Becker TS. Reversible visual loss due to impacted lateral orbital wall fractures. Head Neck 1989;11:295300. 4. Paskert JP, Manson PN, Iliff NT. Nasoethmoidal and orbital fractures. Clin Plast Surg 1988;15:209-23. 5. Straker CA, Hill JC. Management of orbital blow-out fractures. S Afr Med J 1989;76:535-7. 6. Ioannides C, Treffers W, Rutten M, Noverraz P. Ocular injuries associated with fractures involving the orbit. J Craniomaxillofac Surg 1988; 16: 157-9. 7. Dufresne CR, Manson PN, Iliff NT. Early and late complications of orbital fractures. Clin Plast Surg 1988;15:239-53. 8. Arts HA, Eisele DW, Duckert LG. Intraocular pressure as an index of ocular injury in orbital fractures. Arch Otolaryngol Head Neck Surg 1989;115:213-6. 9. Holt GR, Holt JE. Management of orbital trauma and foreign bodies. Otolaryngol Clin North Am 1988;21:35-52. 10. Luce EA, Tubb TD, Moore AM. Review of 1000 major facial fractures and associated injuries. Plast Reconstr Surg 1979;63:26-30. 1 I . Scapini DA, Mathog RH. Repair of orbital floor fractures with marlex mesh. Laryngoscope 1989;99:697-701. 12. Putterman AM, Millman AL. Custom orbital implant in the
Volume 105 Number 5 November 1991
13. 14.
15. 16.
17. 18.
19.
repair of late posttraumatic enophthalmos. Am J Ophthalmol 1989;108:153-9. Gilbard SV, Mafee MF, Logouros PA, Langer BG. Orbital blowout fractures. Ophthalmology 1985;92:1523-8. Kersten RC, Kulwin DR. Selective approach to surgery for delayed enophthalmos (letter to the editor). Arch Otolaryngol Head Neck Surg 1989;115:634. Maniglia AJ. Conjunctival approach for the repair of pure orbital blowout fractures. Otolaryngol Clin N Am 1983;16:575-83. Raveh J, Vuillemin T. The surgical one-stage management of combined cranio-maxillo-facial and frontobasal fractures. J Craniomaxillofac Surg 1988;16: 160-72. Jackson IT. Classification and treatment of orbitozygomatic and orbitoethmoid fractures. Clin Plast Surg 1989;16:77-91. Stanley RB. The temporal approach to impacted lateral orbital wall fractures. Arch Otolaryngol Head Neck Surg 1988;114: 550-3. Parkin JL, Stevens MH, Stringham JC. Absorbable gelatin film
Orbital wall fractures: Evaluation and management 707
20. 2I . 22. 23.
24.
25.
verse silicone rubber sheeting in orbital fracture treatment. Laryngoscope 1987;97:1-3. Webster K. Orbital floor repair with lyophilized porcine dermis. Oral Surg Oral Med Oral Pathol 1988;65:161-4. Wolfe SA. Treatment of post-traumatic orbital deformities. Clin Plast Surg 1988;15:225-38. Iro H. Funktionelle Ergebnisse nach operativer versorgung von isolierten orbitabodenfrakturen. HNO 1989;37:292-4. Jordan DR, White GL, Anderson RL, Thiese SM. Orbital emphysema: a potentially blinding complication following orbital fractures. Ann Emerg Med 1988;17:853-5. Manson PN, Clifford CM, Su CT, Iliff NT, Morgan R . Mechanisms of global support and posttraumatic enophthalmos. Plast Reconstr Surg 1986;77:193-202. Mathog RH, Hillstrom RP, Nesi FA. Surgical correction of enophthalmos and diplopia. Arch Otolaryngol Head Neck Surg 1989;115: 169-78.
