Surgical Approach to Orbital Floor Fractures: Comparing Complication Rates Between Subciliary and Subconjunctival Approaches

Abstract

Background: There are multiple approaches to repairing orbital floor fractures. This study compares the postoperative complications of the subciliary and transconjunctival approaches. Methods: The electronic medical records from 2 hospitals were screened for CPT codes associated with orbital floor reconstruction. A total of 184 patients were identified and included in the study. Patient characteristics and complications were compared by surgical approach. Results: Of the 184 patients, 82 (44.6%) were in the subciliary group and 102 (55.4%) were in the transconjunctival group. The overall postoperative complication rate was 25.5%. The most common of these were diplopia (11.4%), corneal injury (7.1%), proptosis (5.4%), and enopthalmos (4.9%). The complication rate was not statistically significant between the 2 groups. Conclusion: Subciliary and transconjunctival approaches to orbital floor repair are equally safe. This study is limited by a smaller sample size, and a larger study will likely be necessary to fully address this question.

Background

The orbital floor is the weakest part of the orbit, making orbital floor fractures a common result of facial trauma.[1] Depending on the nature of injury, both the maxilla and the zygomatic bone can be affected. To expose the orbital floor for localized reconstruction, a surgeon must employ one of the common surgical approaches.

The surgical approaches to the orbital floor can be categorized as either transconjunctival or transcutaneous (subciliary, subtarsal, and infraorbital). The transconjunctival incision is made through the fornix of conjunctiva, which is believed to reduce the visible cutaneous scarring.[2,3] The short incision and narrow lower conjunctival fornix requires the transconjunctival approach to be supplemented with a lateral canthotomy to increase exposure in 56% to 83% of the procedures, thus precluding the potential of an invisible scar.[4,5,6,7] This decreased exposure is also believed to inhibit the intraoperative management of any complications that may arise.[6] Alternatively, the subciliary approach is achieved by creating an incision 2 mm inferior to the lower lid.[2] This results in consistently adequate access to the orbital floor while leaving a visible scar at the incision site.[4] Additionally, the subtarsal approach is made 5 to 7 mm inferior to the lower lid margin, in one of the subtarsal creases, and extends laterally into one of the resting skin tension lines located along the lateral aspect of the orbit. The approach maintains a band of pretarsal orbicularis muscle as well as its innervation on the tarsus.[8] Lastly, in the infraorbital incision, the skin, orbicularis oculi muscle, and periosteum are incised together. This approach is the quickest and most direct route to the orbital rim and floor.[8] For the purposes of this article, we will focus on the transconjunctival and subciliary approaches.

Postoperative complications of these procedures have long been used as a metric to debate the utility of each approach. In previous case series, the transconjunctival approach has been found to increase the rate of entropion, canthal malposition, and longer operating times, while the subciliary approach has been found to result in more lagophthalmos and a higher rate of ectropion and lower lid retraction.[1,2,5,6,9,10,11]

The purpose of this study is to retrospectively evaluate the surgical results from 2005 to 2016 of this large urban trauma center’s management of orbital floor fractures to assess the degree to which transconjunctival or subciliary approach specifically contributes to postoperational complications. Surgeons at our institution routinely use both approaches; however, based on our experience, we hypothesized that the overall risk of complications is higher when using the transconjunctival approach.

Methods

The Electronic Medical Records of Loyola University Medical Center and Gottlieb Memorial hospital from 2005 through 2016 were reviewed following the institutional review board approval. Patient charts were screened via CPT coding and were included if they underwent a reconstructive procedure using either a transconjunctival or a subciliary approach for the diagnosis of an orbital floor fracture, zygomaticomaxillary complex fracture, nasoorbitoethmoid fracture, or any LeFort fracture excluding isolated mandible fracture. Patients were excluded from this study if their fracture was traumatically exposed, if they had a history of a previous surgical procedure involving the lower eyelid, if they failed to follow-up at the institution within 6 months of the procedure, or if documentation was insufficient for the requirements of this study (cover documentation that was insufficient).

Demographic data collected included patient’s age, sex, race, smoking history, and presence of a medical comorbidity prior to surgery. Specific surgical data collected included preoperative diagnosis, associated injuries prior to surgery, surgical approach used, surgical complications within 6 months of surgery, and the need for revision or readmission to the hospital within 30 days of the procedure.

Statistical Methods

Patient characteristics and complications were calculated and compared by surgical approach for facial fracture reconstruction, and the statistical significance of differences was determined with a t test for age and chi-square or Fisher exact test as appropriate for all other comparisons. Univariable and multivariable proportional odds models were used to determine the association of patient characteristics and surgical approach with the number of complications (0, 1, and ≥2).

Results

A total of 393 patients were identified by chart review who underwent surgical repair for orbital floor fractures between 2005 and 2016. Of these, 209 patients were excluded due to use of an approach other than subciliary or transconjunctival, presence of an open wound involving the lower eyelid, a history of prior lower eyelid surgery, or insufficient documentation. Of the 184 facial fracture reconstruction procedures included in our analysis, 82 (44.6%) utilized the subciliary approach and 102 (55.4%) utilized a transconjunctival approach. The mean age at the time of surgery was 35 years (standard deviation =12). In all, 17.9% (n =33) of patients were female, and 48.5% (n = 89) were non-white.

No statistically significant differences in the rates of any specific postoperative complication or the number of complications were detected in univariable analysis (

Table 1. Patient Characteristics and Complications by Surgical Approach for Facial Fracture Reconstruction.^a.

). Overall postoperative complication rate at our institution was 25.5% (n =47). The most common of these was diplopia occurring in 11.4% (n =21) of patients, followed by corneal injury in 7.1% (n =13), proptosis in 5.4% (n =10), and enopthalmos in 4.9% (n =9). Complications involving the lower lid were rare, with entropion (n = 5) and ectropion (n = 4) reported in less than 3% of patients each. The postoperative complication rate was slightly greater for transconjunctival repairs compared to the subciliary approach after multivariable adjustment, though this was not statistically significant (odds ratio: 1.26, 95% confidence interval: 0.64-2.48;

Table 2. Association Between Surgical Approach and Any Postoperative Complications.^a.

).

Race, gender, age, smoking status, and presence of glaucoma or diabetes mellitus prior to injury were all determined to have no significant effect on the rate of postoperative complications after multivariable adjustment.

Discussion

The patients presenting to our trauma center with facial fractures were middle aged (mean =35 years), male (82.1%), and ethnically diverse with 51.6% white, 22.8% black, and 21.7% Hispanic (Table 1). There were no gender or ethnicity-based disparities observed in the rate of postoperative complications in this population (Table 2).

Consistent with the previous literature, the transconjunctival approach was associated with an increased number of reports of entropion and enopthalmos, though not statistically significant. While we did have 5.4% rate of proptosis, this was primarily transient and did not require surgical correction. There were no cases of late proptosis. Similarly, patients treated with a transconjunctival approach exhibited a slightly higher rate of surgical site infections and brow ptosis, also not statistically significant. All other specific complications occurred at roughly equivalent rates. Multivariable analysis determined that in addition to both surgical approaches having a roughly equal rate of overall complications, there was no increased risk of postoperative surgical complications in patients with medical comorbidities that would affect wound healing and vascularity, such as glaucoma, diabetes, or tobacco use.

This study is limited by the data available to retrospective chart review. Inconsistency in surgical dictation and provider documentation resulted in a number of patients being excluded from the study, as the operative approach utilized during their procedure was unclear. The smaller sample size in this review was insufficient to reveal major differences in complication rates, as each complication occurred in less than 15 of the patients studied. Future analysis should seek to incorporate large sets of pooled data, such as that found in administrative databases, to address this issue on a sufficiently broad scale. Unfortunately, while surgical procedure types are included in these databases, difference in surgical technique and approach is often not captured outside of individual progress notes. Since these databases do not include this detailed information, a prospective, randomized study will likely be necessary to fully address this question.

Conclusions

Based on this institutional series, there is no significant outcome difference in postoperative complications (entropion, ectropion, enopthalmos, lagopthalmos, proptosis, keratoconjunctivitis sicca, corneal injury, diplopia, ptosis, decreased visual acuity, limited extraocular motility, and surgical site infection) between subciliary and transconjunctival access for the repair of orbital floor fractures. Both methods are equally safe and efficacious, and choice should be individualized based on each surgeon’s individual experience and the presentation of the injury.