Fifty-one patients with pure blowout fracture of the inferior orbital wall from January 2000 to December 2011, at the Department of Otolaryngology-Head and Neck Surgery of Maryknoll Medical Center were included in the study. All enrolled patients underwent reduction surgery using folded silicon tube placement via the transantral approach. Clinical indications for fracture repair included persistent diplopia (>2 weeks), extraocular muscle entrapment on face computed tomography (CT) scan, a large posterior fracture (>50% of the orbital floor), or early enophthalmos [581213]. Exclusion criteria were as follows: (1) patients with orbital floor fractures, who did not undergo reduction surgery; (2) patients with accompanying ethmoidal, maxillary, or zygomatic bone fractures; (3) insufficient medical data (preoperative and postoperative CT scans, operation, and medical records); (4) patients, who underwent surgery with other implanted materials; (5) patients, who did not follow up for more than three months despite persistent orbital symptoms; (6) patients, who underwent reduction surgery via both transantral and transorbital approaches; (7) patients, who underwent simultaneous reduction surgery for both orbital floor and medial wall fractures. All protocols were approved by the Kosin University Institutional Review Board (IRB No. 12-104).

All patients received a complete, preoperative and postoperative ophthalmic examination for visual acuity, diplopia, extraocular muscle limitation, and enophthalmos. Diplopia in the cardinal fields of gaze was considered significant. Enophthalmos was regarded as significant if a 2-mm discrepancy between eyeballs existed. A Hertel exophthalmometer was used for measurements a day presurgery and 4 weeks postsurgery. Facial CT was performed in all patients, preoperatively and postoperatively. The mean displacement between fractured bone chip and intact orbital floor was based on anatomic positioning on CT scans. Displacement was classified as: grade I, an exact anatomic position of reduced orbital floor or minimal displacement; grade II, discrepancy of displacement below 2 mm; and grade III, a discrepancy of displacement above 2 mm. Complications and management strategies were evaluated by chart review.

Surgical procedures were performed under general anesthesia. Typically, 1% lidocaine with 1:100,000 dilution of epinephrine was injected into the upper gingivobuccal sulcus. A 3- to 4-cm horizontal incision was made just superior to the sulcus. The periosteum and overlying soft tissue were gently elevated from the underlying maxillary bone to the infraorbital foramen using a periosteal elevator. An antral window (2 cm×1.5 cm) was made into the maxillary sinus with an electric saw. The bone fragment was preserved in saline solution for repositioning at the end of surgery. The maxillary sinus and prolapsed orbital contents were visualized employing a 0-degree endoscope, through the antral window. The herniated orbital contents were reduced by digital manipulation or by using surgical instruments like an elevator, without removing the mucosa and fractured bone chip. A silastic sheet (1 mm), slightly larger than the fractured area, was inserted into the maxillary sinus and placed below the orbital floor. A silicon tubing (1 mm in adult and 0.5 mm in pediatric cases) was bent one or two times, and inserted into the maxillary sinus under the silastic sheet (Fig. 1). When placed into the maxillary sinus, the tube recoiled automatically and adjusted to the sinus wall to firmly support the reduced orbital floor with its elasticity (Fig. 2). Care was taken not to obstruct the natural ostium of the maxillary sinus (Fig. 3). A forced duction test was performed to confirm correct positioning of the orbital floor and to avoid entrapment of the orbital contents before closing. The antral wall fragment, stored in saline, was fixed in its original position using a plate and screws. The gingivobuccal incision was closed, and postoperative antibiotics were continued for 14 days. The silastic sheet and tube were removed 6 to 8 weeks postsurgery under local anesthesia through the same antral maxillary window, created during the first surgery (Fig. 4). The second surgery was performed under general anesthesia in all children and some adults.

Twenty-five patients reported diplopia before surgery, of which postoperative improvement was seen in 18 patients. In seven patients, diplopia persisted at extreme upper gaze postsurgery, but disappeared within the 3-month follow-up period. Thirteen out of 15 patients had improved extraocular muscle limitation postoperatively. Enophthalmos resolved postoperatively in four of five patients (Table 1).

Preoperatively, there were three cases of grade II and 48 cases of grade III displacement. All three preoperative grade II cases were restored to grade I postoperatively. Among 48 patients with a preoperative status of grade III, 16, 25, and seven cases had grades I, II, III, respectively, postoperatively (Table 2).

Inflammatory changes of maxillary sinus mucosa were observed in 21 patients upon silicone tube removal. The inflamed mucosa was either shaved using a microdebrider after removal of the implants, or left undisturbed. One patient complained of severe diplopia, ocular pain, and aggravated extraocular muscle limitation after surgery. The symptoms were attributed to overcorrection of the fractured bony fragments, and were alleviated after immediate revision surgery. A patient, 3 days postsurgery, presented extrusion of the silicon tube through the maxillary sinus window. Corrective implant repositioning was achieved by revision surgery. Another patient developed maxillary sinusitis at 4 weeks postsurgery, with symptoms of cheek pain, swelling, and nasal discharge. The sinusitis symptoms resolved on removal of the implant. These results are summarized in Table 3.