- Email: [email protected]
Facial Nerve Injury during External Dacryocystorhinostomy
Facial Nerve Injury during External Dacryocystorhinostomy M. Reza Vagefi, MD,1 Bryan J. Winn, MD,2 Chun Cheng Lin, MD,3 Bryan S. Sires, MD, PhD,2 Steven J. LauKaitis, MD,2 Richard L. Anderson, MD,3 John D. McCann, MD, PhD3 Objective: To describe weakness of the orbicularis oculi muscle after external dacryocystorhinostomy (DCR) and propose an anatomic explanation for the complication. Design: Retrospective, observational study. Participants: Sixteen patients (13 female, 3 male) with a mean age of 60 years (median, 61 years; range, 34 – 85 years). Methods: A retrospective chart review was performed of consecutive patients who had nasolacrimal duct obstruction repair by external DCR. Patients were identified who developed postoperative orbicularis oculi muscle weakness that manifested as hypometric blink or lagophthalmos with or without punctate keratopathy on the operated side. Patient parameters collected included demographic data, type of incision, incision length, use of lacrimal stent, length of follow-up, intraoperative and postoperative complications, and time to resolution of clinical findings. Statistical analysis was performed using a 2-tailed Fisher exact test with clinical significance designated at ␣ ⫽ 0.05. Main Outcome Measures: Identification of patients with orbicularis oculi muscle weakness after external DCR, documentation of incision type, clinical findings, and recovery of function. Results: Among 215 patients and 247 surgeries, 16 individuals (7.4%) were identified who demonstrated abnormalities of eyelid closure in the postoperative period after external DCR. Of these, 13 patients had lagophthalmos with or without hypometric blink and 3 patients had hypometric blink alone. Eleven patients underwent surgery through a nasojugal incision, 4 patients underwent surgery through a vertical incision, and 1 patient underwent surgery through an eyelid margin incision. The degree of postoperative lagophthalmos was on average 1.5 mm. Four patients developed punctate keratopathy. Follow-up ranged from 3 to 50 weeks (mean, 20 weeks). Resolution of lagophthalmos was seen on average by 14 weeks with the longest time to resolution of 32 weeks. Three individuals continued to have residual hypometric blink at the time of last follow-up. Conclusions: Damage to peripheral fibers of the zygomatic and buccal branches of the facial nerve as they course through the medial canthal area to innervate the upper eyelid orbicularis oculi muscle may occur during external DCR surgery. Such injury may be responsible for orbicularis oculi muscle weakness manifesting as postoperative abnormal eyelid closure and lagophthalmos. In our cohort of patients, these findings were temporary and typically resolved in several months. Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article. Ophthalmology 2009;116:585–590 © 2009 by the American Academy of Ophthalmology.
External dacryocystorhinostomy (DCR) is a commonly performed procedure associated with a success rate of greater than 90%.1–7 Since its first modern description by Toti in 1904, advancements in surgical technique have improved outcomes and decreased morbidity.8,9 Nevertheless, intraoperative and postoperative complications after DCR surgery occur.10 Bleeding is the most common intraoperative complication because of the close relation of the surgical site to the angular vessels and the nasal mucosa. Orbital hemorrhage from disruption of the anterior ethmoidal artery or cerebrospinal fluid leak from violation of the cribriform plate may occur.11,12 Other uncommon intracranial complications have been reported, including meningitis and pneumocephalus.13,14 A range of postoperative complications can be encountered.10,15 Postoperative bleeding is usually self-limited and rarely significant in nature. Surgical failure is the most © 2009 by the American Academy of Ophthalmology Published by Elsevier Inc.
frequent complication and is typically caused by scarring at the rhinostomy site or secondary stenosis of the canalicular system. Operative trauma and postoperative inflammation can contribute to unsuccessful surgery. The use of silicone stents may, on occasion, lead to acute and chronic allergic reactions, as well as canalicular erosion.16 –18 Poor incision healing can lead to scar formation, and the semi-sterile nature of nasal surgery lends to the possibility of wound infection.19 Although less commonly seen, incision necrosis can occur from aggressive cauterization during surgery and lipogranuloma formation from reaction to antibiotic-coated intranasal gauze.20,21 The most common external DCR incision is 1.0 to 1.5 cm in length and is placed tangentially to the inferonasal rim of the orbit vertically along the nose approximately 1.0 cm medial to the insertion of the medial canthal tendon (Fig 1).10,22 VariaISSN 0161-6420/09/$–see front matter doi:10.1016/j.ophtha.2008.09.050
585
Ophthalmology Volume 116, Number 3, March 2009 fied as incomplete voluntary closure of the eyelid equal to or greater than 1 mm. Punctate keratopathy related to abnormal eyelid closure was characterized as being present if there was an inferior corneal fluorescein staining pattern. Records were reviewed to assess possible contributing factors and outcomes. Patient parameters collected include age, gender, indication for surgery, laterality, type of incision, incision length, use of lacrimal stent, length of follow-up, intra- and postoperative complications, and time to resolution of eyelid closure abnormalities. Statistical Analysis Software 9.1 (SAS Inc., Cary, NC) was used to perform a 2-tailed Fisher exact test with clinical significance designated at ␣ ⫽ 0.05.
Results Figure 1. Incision types for external dacryocystorhinostomy (DCR). The vertical incision is placed 1 cm medial to the medial canthus, extending 1 to 1.5 cm in length (black line). The nasojugal incision is placed along the shallow curvilinear groove that lies infralateral to the medial canthus and extends 1 to 1.5 cm in length (white line). The eyelid margin incision is placed along the medial canthal angle predominantly along the lower eyelid margin and is approximately 1 cm in total length (dotted line).
tions that aim for improved scar concealment are also used, with incision placement along the nasojugal skin fold or the eyelid margin.23–25 The surgeon must dissect through skin, orbicularis oculi muscle, and periosteum to reach the anterior lacrimal crest while avoiding the angular vessels. In traditional teaching, injury to the facial nerve is not a concern during external DCR surgery because the motor innervation of the periocular muscles is believed to be well away from the incision site.26,27 It is commonly accepted that branches of the temporal, zygomatic, and buccal divisions of the facial nerve travel from lateral to medial across the zygoma and maxilla to send terminal nerve fibers to the posterior surface of the orbicularis oculi muscle, providing its innervation.26,27 Damage to facial nerve fibers of the upper or lower eyelid during ocular procedures, such as blepharoplasty or ptosis correction, can result in orbicularis oculi weakness, incomplete or delayed eyelid closure, and lagophthalmos.28 In this review, we propose that facial nerve injury is possible during external DCR surgery, resulting in orbicularis oculi muscle weakness, and propose an anatomic explanation for this observation.
Materials and Methods A retrospective chart review was performed of 4 oculofacial plastic surgeons’ practices (BSS, SJL, RLA, JDM) from 2005 to 2007 of consecutive patients who had nasolacrimal duct obstruction repair by external DCR. Approval for the study was obtained from the Western Institutional Review Board (Olympia, WA). The study was compliant with the Health Insurance Portability and Accountability Act and in accord with the principles outlined in the Declaration of Helsinki. Patients were identified who developed clinical signs of postoperative orbicularis oculi muscle weakness in the forms of hypometric blink or lagophthalmos with or without punctate keratopathy on the operated side. Hypometric blink was defined as delayed or incomplete blink of the upper eyelid on the operated side compared with the contralateral side. Lagophthalmos was identi-
586
There were 16 patients (7.4%) identified among 215 individuals and 247 surgeries who demonstrated abnormalities of eyelid closure in the postoperative period after external DCR. Of these, 13 patients had lagophthalmos with or without hypometric blink and 3 patients had purely hypometric blink (Table 1). Indications for surgery included acquired nasolacrimal duct obstruction in 13 patients and concurrent canalicular stenosis and nasolacrimal duct obstruction in 3 patients. In 2 of these patients, repeat external DCR was performed because of primary surgical failure. Concurrent lower lid ectropion surgery via a tarsal strip procedure was performed in 2 patients. The group was composed of 13 female and 3 male patients with a mean age of 60 years (median 61 years; range, 34 – 85 years). The left side was operated in 69% of cases. The Fisher exact test demonstrated no statistically significant differences in either the gender or laterality distributions between patients who underwent DCR surgery with eyelid closure abnormalities and those without (␣ ⫽ 0.77 and 0.20, respectively). We did not find any correlation between the indication for surgery and orbicularis oculi weakness. Of the 247 DCRs performed, 121 were through a nasojugal incision, 92 were through a vertical incision, and 34 were through an eyelid margin incision. When analyzed by incision type, 9.1% (11) of DCRs with a nasojugal incision demonstrated eyelid closure abnormalities, whereas those with vertical and eyelid margin incisions were associated with abnormalities 4.3% (4) and 2.9% (1) of the time, respectively. The Fisher exact test comparing the rates of abnormalities across the 3 distinct incision types did not reveal a statistically significant difference (␣ ⫽ 0.37). Incision length was on average 13.5 mm. Ten of the 16 patients had placement of a lacrimal stent. One surgeon (SJL) did not routinely place stents. The stent remained in place for an average of 10 weeks. Follow-up ranged from 3 to 50 weeks (mean, 20 weeks; standard deviation, ⫾13 weeks). The degree of postoperative lagophthalmos was on average 1.5 mm (range, 1–2 mm) (Fig 2). Four patients had inferior punctate keratopathy demonstrated by slit-lamp examination. Resolution of lagophthalmos was seen on average by 14 weeks (standard deviation, ⫾9 weeks). The longest time to resolution was 32 weeks. Three individuals (patients 1, 3, and 7) had residual hypometric blink after resolution of lagophthalmos, and one individual (patient 13) was lost to follow-up in the early postoperative period at 3 weeks before resolution could be documented.
Discussion Iatrogenic injury to the seventh cranial nerve is often a concern during facial plastic surgery, including lifting procedures for the forehead, mid-face, and lower face.29 Damage to temporal and zygomatic branches during the aforemen-
Vagefi et al 䡠 Facial Nerve Injury during External DCR Table 1. Characteristics of Patients with Facial Nerve Injury during External Dacryocystorhinostomy Age Patient (y) Gender Side 1 2 3 4
42 70 34 56
F F F M
R L L L
5 6 7
76 61 85
M F F
L R L
8
41
F
L
9
84
F
L
10
44
F
L
11 12
39 60
F F
R R
13 14 15 16
68 59 72 74
M F F F
L R L L
Indication Acquired NLDO Acquired NLDO Acquired NLDO Acquired NLDO/ canalicular stenosis Acquired NLDO Acquired NLDO Recurrent Acquired NLDO Acquired NLDO/ chronic dacryocystitis Acquired NLDO/ chronic dacryocystitis Recurrent acquired NLDO/canalicular stenosis Acquired NLDO Acquired NLDO/ canalicular stenosis Acquired NLDO Acquired NLDO Acquired NLDO Acquired NLDO
Time to Prior Type of Incision/ Punctate Hypometric Resolution of Resolution DCR Stent Placement Lagophthalmos Keratopathy Blink Symptoms (wk) No No No No
Vertical/Yes Vertical/Yes Vertical/Yes Vertical/Yes
2 mm None 1 mm 2 mm
None None None Yes
Yes Yes Yes Yes
Continued HB Yes Continued HB Yes
28 32 14 10
No Eyelid Margin/Yes No Nasojugal/Yes Yes Nasojugal/Yes
None 2 mm 1 mm
None None Yes
Yes Yes Yes
Yes Yes Continued HB
27 12 9
No
Nasojugal/No
None
Yes
Yes
Yes
10
No
Nasojugal/Yes
1 mm
Yes
Yes
Yes
8
Yes
Nasojugal/Yes
1 mm
None
Yes
Yes
12
No No
Nasojugal/Yes Nasojugal/No
2 mm 1 mm
None None
Yes Not noted
Yes Yes
3 10
No No No No
Nasojugal/No Nasojugal/No Nasojugal/No Nasojugal/No
1 mm 2 mm 2 mm 1 mm
None None None None
Yes Not noted Not noted Not noted
Lost to F/U Yes Yes Yes
– 12 8 6
DCR ⫽ dacryocystorhinostomy; F/U ⫽ follow-up; HB ⫽ hypometric blink; NLDO ⫽ nasolacrimal duct obstruction.
tioned procedures and during cosmetic and reconstructive eyelid surgery can result in orbicularis oculi muscle weakness. As a consequence, poor eyelid closure and lagophthalmos may occur from injury to fibers innervating the upper eyelid and scleral show and ectropion from injury to fibers to the lower eyelid.28,29 Traditionally, innervation of the eyelids is believed to proceed from lateral to medial with the temporal branches providing terminal fibers to the upper eyelid and the zygomatic and buccal branches supplying fibers to the lower eyelid.25,26 These fibers initially travel in the superficial musculoaponeurotic system and become more superficial as they approach the lateral orbit, eventually assuming a vertical orientation in the post-orbicularis oculi fascial plane and penetrating the muscle on its posterior surface.28 Facial nerve injury has not been described after external DCR surgery. We have observed in a small portion of patients a spectrum of findings consistent with orbicularis oculi muscle weakness of the upper eyelid, ranging from delayed or incomplete blink to clinically significant lagophthalmos with corneal exposure. We speculate that these are secondary to intraoperative trauma to the seventh cranial nerve. We were only able to identify a single report of 2 cases of abnormal eyelid closure after external DCR, and these were not attributed to nerve injury.30 Recent anatomic studies have provided a more detailed understanding of the facial nerve distribution of the eyelids and periorbital region.31–33 Nemoto and colleagues31 demonstrated through microscopic dissections that superficial
buccal branches of the facial nerve not only supply the levator labii superioris, the lower portion of the orbicularis oculi and the levator labii superioris alaeque nasi, but also course superiorly over the medial palpebral ligament with the angular artery to innervate the upper part of the orbicularis oculi, procerus, and corrugator supercilii (Fig 3). In 42% of specimens, they observed that the nasal superficial buccal branches innervate the upper orbicularis oculi muscle via a supramedial approach. Similarly, other studies have found the zygomatic and buccal branches coalesce to form what some authors have termed the angular nerve, which tracks superiorly to supply the procerus, corrugator, and presumably the medial orbicularis oculi muscles.32,33 The external DCR incision and dissection can be situated in the path of these superior and medial coursing branches of the zygomatic and buccal nerve. We propose that in a minority of patients, disruption of these fibers, whether from surgical transection or thermal injury from cautery, results in abnormal eyelid closure. The majority of patients who undergo external DCR surgery do not experience eyelid closure abnormalities. We postulate several reasons for this finding. First, there exists sufficient anatomic variation in the distribution of the peripheral branches of the zygomatic and buccal nerve that these fibers could be preserved and not disrupted by the incision, dissection, or cautery as they course superiorly through the medial canthal region.31 Furthermore, it is possible that despite injury to the facial nerve in the medial canthal area, certain individuals have redundant innervation
587
Ophthalmology Volume 116, Number 3, March 2009
Figure 2. Postoperative clinical appearance of orbicularis oculi weakness after external dacryocystorhinostomy (DCR). A, Patient 3, a 34-year-old woman, presented 4 weeks after left-sided external DCR with 1 mm of lagophthalmos and hypometric blink of the left eye. B, Patient 4, a 56-yearold man, was reviewed at 4 weeks after left-sided external DCR and found to have 2 mm of lagophthalmos, punctuate keratitis, and hypometric blink of the left eye. C, Patient 1, a 42-year-old woman, presented 6 weeks after right-sided external DCR with 2 mm of lagophthalmos and hypometric blink of the right eye.
of the upper eyelid orbicularis muscle provided by the temporal branch of the facial nerve as it courses lateral to medial across the upper eyelid.34 In addition, most surgeons typically use blunt dissection techniques and cautious use of cautery after the skin incision, which likely reduce the chance for nerve injury. Other iatrogenic causes for the constellation of findings should be considered. Myotoxic effects of local anesthetics may result in muscle degeneration; however, studies have demonstrated that muscle hypertrophy is induced after a brief period of paralysis, resulting in a strengthening effect by 3 weeks.35 No cases of human orbicularis oculi toxicity from lidocaine or bupivacaine have been reported in the literature. Bupivacaine-associated orbicularis oculi myotoxicity has been demonstrated in a rabbit model; however, the toxicity usually required high doses of the anesthetic with concurrent use of hyaluronidase.36 In our study, none of the surgeons used hyaluronidase and one of the surgeons never used bupivacaine, yet all experienced this complication. Alternatively, a contusion of the orbicularis oculi muscle from local anesthesia injection or surgical dissection may cause muscle dysfunction. Orbicularis oculi weakness may occur as a consequence of muscle necrosis from aggressive
588
cautery. Surgical dissection can possibly result in disinsertion of the orbicularis oculi muscle or medial canthal tendon at the time of periosteal cleavage. Nevertheless, these mechanisms would not explain the weakness of the superior orbicularis oculi muscle, because surgery is performed below the medial canthal tendon, or the gradual resolution of symptoms over the course of months, because disinsertion of the orbicularis oculi muscle or medial canthal tendon would not be expected to spontaneously improve. Furthermore, no telecanthus was observed in any patients. Because the study was retrospective in nature, no quantitative direct measure of facial nerve function (e.g., electromyelography) was performed. However, we believe our clinical observations are consistent with trauma to one or more peripheral branches of the facial nerve in the inferior eyelid that continue to course superiorly through the medial canthal area to supply the upper eyelid orbicularis oculi muscle from a medial approach. Perhaps the most supportive observation is that of delayed blink. A clinical sign of acute to subacute peripheral nerve injury affecting the function of an extraocular muscle is the observation of a delayed saccade.37 If a blink is analogous to a saccade, it follows that the observation of a delay in addition to an incomplete blink suggests a neurologic rather than a purely mechanical cause. In the present study, signs of orbicularis oculi muscle weakness were found in 7.4% of patients after external DCR surgery. The true incidence of orbicularis oculi weakness is likely higher. As we have gained greater awareness of this problem, subtle abnormalities of eyelid closure have been more frequently noted. In our cohort, a larger number of patients operated through a nasojugal approach demonstrated orbicularis oculi weakness. Theoretically, this incision is more likely to cross the path of the superficial buccal branch of the facial nerve and thus probably carries a higher risk of injury. Nevertheless, a statistically significant difference between the 3 incision types was not found. This, however, may be due to a small sample size. The finding of orbicularis oculi weakness with all incision types provides evidence of the anatomic variability of the superficial fibers of the facial nerve as they course to the medial upper eyelid. We believe this could also explain the spectrum of muscle weakness noted. Depending on the variable distribution of these medial fibers, which and how many of these fibers were affected during surgery, and the redundant innervation provided by the temporal branches of the facial nerve to the upper eyelid, a range of clinical findings could be observed. We suspect that with a superficial incision followed by blunt dissection and the judicious use of cautery, the likelihood of injury could be reduced. Normal function is typically regained by 3 months, although in one case weakness persisted to 8 months. Supportive care by means of artificial tears and lubricating ointment should be used in the interim. Furthermore, in cases of early surgical failure or persistent tearing after external DCR, lacrimal pump dysfunction secondary to orbicularis oculi weakness should be considered and repeat surgery avoided until recovery of nerve function. We propose that injury of the facial nerve during external DCR surgery can result in orbicularis oculi muscle weakness that manifests as abnormal eyelid closure and lagoph-
Vagefi et al 䡠 Facial Nerve Injury during External DCR
Figure 3. Left superficial buccal branches of the facial nerve. A, Anterior view (left) and in diagrammatic form (right). Superficial buccal branches spread into terminal rami to surrounding muscles. Also demonstrated is angular artery. B, Posterior view (left) and in diagrammatic form (right). Lower forceps turn over levator labii superioris, and upper forceps pull upper end of levator labii superioris alaeque nasi nasally, to show course of superficial buccal branches. Terminal rami supply procerus, corrugator supercilii, and orbicularis oculi. Reprinted with permission from Nemoto Y, Sekino Y, Kaneko H. Facial nerve anatomy in eyelids and periorbit. Jpn J Ophthalmol 2001;45:445–52. Aa ⫽ angular artery; CS ⫽ corrugator supercilii; OO ⫽ orbicularis oculi; Pr ⫽ procerus; SBBr ⫽ superficial buccal branch.
thalmos. This complication is likely caused by an insult to peripheral fibers of the zygomatic and buccal branches of the facial nerve as they course in the medial canthal area and provide innervation to the upper eyelid orbicularis muscle in a subset of individuals. In our cohort, the signs and symptoms of nerve injury were temporary and resolved in several months as neural function recovered. Patients should be appropriately informed of this operative risk and reassured that symptoms typically resolve should such a complication occur.
References 1. Tarbet KJ, Custer PL. External dacryocystorhinostomy: surgical success, patient satisfaction, and economic cost. Ophthalmology 1995;102:1065–70. 2. Dolman PJ. Comparison of external dacryocystorhinostomy with nonlaser endonasal dacryocystorhinostomy. Ophthalmology 2003;110:78 – 84. 3. Mirza S, Al-Barmani A, Douglas SA, et al. A retrospective comparison of endonasal KTP laser dacryocystorhinostomy
589
Ophthalmology Volume 116, Number 3, March 2009
4. 5. 6. 7.
8. 9. 10.
11. 12. 13. 14. 15. 16. 17. 18. 19.
20.
versus external dacryocystorhinostomy. Clin Otolaryngol Allied Sci 2002;27:347–51. Tsirbas A, Davis G, Wormald PJ. Mechanical endonasal dacryocystorhinostomy versus external dacryocystorhinostomy. Ophthal Plast Reconstr Surg 2004;20:50 – 6. Walland MJ, Rose GE. Factors affecting the success rate of open lacrimal surgery. Br J Ophthalmol 1994;78:888 –91. Becker BB. Dacryocystorhinostomy without flaps. Ophthalmic Surg 1988;19:419 –27. Cokkeser Y, Evereklioglu C, Er H. Comparative external versus endoscopic dacryocystorhinostomy: results in 115 patients (130 eyes). Otolaryngol Head Neck Surg 2000;123: 488 –91. Albert DM. Ophthalmic plastic surgery. In: Albert DM, Edwards DD, eds. The History of Ophthalmology. Cambridge, MA: Blackwell Science; 1996:235–53. Duffy MT. Advances in lacrimal surgery. Curr Opin Ophthalmol 2000;11:352– 6. Gonnering RS. Dacryocystorhinostomy and conjunctivodacryocystorhinostomy. In Dortzbach RK, ed. Ophthalmic Plastic Surgery: Prevention and Management of Complications. New York: Raven Press; 1994:237–50. Hurwitz JJ, Eplett CJ, Fliss D, Freeman JL. Orbital hemorrhage during dacryocystorhinostomy. Can J Ophthalmol 1992; 27:139 – 42. Neuhaus RW, Baylis HI. Cerebrospinal fluid leakage after dacryocystorhinostomy. Ophthalmology 1983;90:1091–5. Beiran I, Pikkel J, Gilboa M, Miller B. Meningitis as a complication of dacryocystorhinostomy. Br J Ophthalmol 1994;78:417– 8. Usul H, Kuzeyli K, Cakir E, et al. Meningitis and pneumocephalus: a rare complication of external dacryocystorhinostomy. J Clin Neurosci 2004;11:901–2. Besharati MR, Rastegar A. Results and complications of external dacryocystorhinostomy surgery at a teaching hospital in Iran. Saudi Med J 2005;26:1940 – 4. Jordan DR, Nerad JA. An acute inflammatory reaction to silicone stents. Ophthal Plast Reconstr Surg 1987;3:147–50. Anderson RL, Edwards JJ. Indications, complications and results with silicone stents. Ophthalmology 1979;86:1474 – 87. Jordan DR, Nerad JA, Tse DT. Complete canalicular erosion associated with silicone stents. Am J Ophthalmol 1986;101: 382–3. Hanna IT, Powrie S, Rose GE. Open lacrimal surgery: a comparison of admission outcome and complications after planned day case or inpatient management. Br J Ophthalmol 1998;82:392– 6. Salour H, Montazerin N. Incision site tissue necrosis after dacryocystorhinostomy. Ophthal Plast Reconstr Surg 1998;14: 146 – 8.
21. Kerendian J, Conn H. Lipogranuloma: a preventable complication of dacryocystorhinostomy. Ophthalmic Surg Lasers 1996;27:713–5. 22. Nerad JA. Oculoplastic surgery. In Krachmer JH, ed. The Requisites in Ophthalmology. St. Louis, MO: Mosby; 2001: 215–53. 23. Harris GJ, Sakol PJ, Beatty RL. Relaxed skin tension line incision for dacryocystorhinostomy. Am J Ophthalmol 1989; 108:742–3. 24. Dortzbach R, Woog JJ. Small-incision techniques in ophthalmic plastic surgery. Ophthalmic Surg 1990;21:615–22. 25. Putterman AM. Eyelid incision approach to dacryocystorhinostomy facilitated with a mechanical retraction system. Am J Ophthalmol 1994;118:672– 4. 26. Dutton JJ. Orbital Nerves. In: Dutton JJ, ed. Atlas of Clinical and Surgical Orbital Anatomy. Philadelphia, PA: Saunders; 1994: 33– 64. 27. May M. Anatomy for the clinician. In: May M, Schaitkin BM, eds. The Facial Nerve: May’s Second Edition. New York: Thieme; 2000:19 –56. 28. Jordan DR, Anderson RL. The facial nerve in eyelid surgery [letter]. Arch Ophthalmol 1989;107:1114 –5. 29. Seckel BR. Facial Danger Zones: Avoiding Nerve Injury in Facial Plastic Surgery. St. Louis, MO: Quality Medical Pub; 1994:1–52. 30. Fayet B, Bernard JA, Ritleng P, et al. Internal transient palpebral inclusion following dacryocystorhinostomy [in French]. J Fr Ophtalmol 1994;17:195–9. 31. Nemoto Y, Sekino Y, Kaneko H. Facial nerve anatomy in eyelids and periorbit. Jpn J Ophthalmol 2001;45:445–52. 32. Caminer DM, Newman MI, Boyd JB. Angular nerve: new insights on innervation of the corrugator supercilii and procerus muscles. J Plast Reconstr Aesthet Surg 2006;59:366 –72. 33. Hwang K, Jin S, Park JH, Chung IH. Innervation of the procerus muscle. J Craniofac Surg 2006;17:484 – 6. 34. Ouattara D, Vacher C, de Vasconcellos JJ, et al. Anatomical study of the variations in innervation of the orbicularis oculi by the facial nerve. Surg Radiol Anat 2004;26:51–3. 35. Scott AB, Alexander DE, Miller JM. Bupivacaine injection of eye muscles to treat strabismus. Br J Ophthalmol 2007;91: 146 – 8. 36. McLoon LK, Wirtschafter J. Regional differences in the subacute response of rabbit orbicularis oculi to bupivacaineinduced myotoxicity as quantified with a neural cell adhesion molecule immunohistochemical marker. Invest Ophthalmol Vis Sci 1993;34:3450 – 8. 37. Wong AM, McReelis K, Sharpe JA. Saccade dynamics in peripheral vs central sixth nerve palsies. Neurology 2006;66: 1390 – 8.
Footnotes and Financial Disclosures Originally received: April 20, 2008. Final revision: September 19, 2008. Accepted: September 26, 2008. Available online: December 16, 2008.
3
Manuscript no. 2008-491.
1
Scheie Eye Institute, University of Pennsylvania, Philadelphia, Pennsylvania.
2
Allure Facial Laser Center and Medispa, Kirkland, Washington.
590
Center for Facial Appearances, Salt Lake City, Utah. Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article. Correspondence: John D. McCann, MD, PhD, 1002 E. South Temple, Suite 308, Salt Lake City, UT 84102. E-mail: [email protected].
External dacryocystorhinostomy (DCR) is a commonly performed procedure associated with a success rate of greater than 90%.1–7 Since its first modern description by Toti in 1904, advancements in surgical technique have improved outcomes and decreased morbidity.8,9 Nevertheless, intraoperative and postoperative complications after DCR surgery occur.10 Bleeding is the most common intraoperative complication because of the close relation of the surgical site to the angular vessels and the nasal mucosa. Orbital hemorrhage from disruption of the anterior ethmoidal artery or cerebrospinal fluid leak from violation of the cribriform plate may occur.11,12 Other uncommon intracranial complications have been reported, including meningitis and pneumocephalus.13,14 A range of postoperative complications can be encountered.10,15 Postoperative bleeding is usually self-limited and rarely significant in nature. Surgical failure is the most © 2009 by the American Academy of Ophthalmology Published by Elsevier Inc.
frequent complication and is typically caused by scarring at the rhinostomy site or secondary stenosis of the canalicular system. Operative trauma and postoperative inflammation can contribute to unsuccessful surgery. The use of silicone stents may, on occasion, lead to acute and chronic allergic reactions, as well as canalicular erosion.16 –18 Poor incision healing can lead to scar formation, and the semi-sterile nature of nasal surgery lends to the possibility of wound infection.19 Although less commonly seen, incision necrosis can occur from aggressive cauterization during surgery and lipogranuloma formation from reaction to antibiotic-coated intranasal gauze.20,21 The most common external DCR incision is 1.0 to 1.5 cm in length and is placed tangentially to the inferonasal rim of the orbit vertically along the nose approximately 1.0 cm medial to the insertion of the medial canthal tendon (Fig 1).10,22 VariaISSN 0161-6420/09/$–see front matter doi:10.1016/j.ophtha.2008.09.050
585
Ophthalmology Volume 116, Number 3, March 2009 fied as incomplete voluntary closure of the eyelid equal to or greater than 1 mm. Punctate keratopathy related to abnormal eyelid closure was characterized as being present if there was an inferior corneal fluorescein staining pattern. Records were reviewed to assess possible contributing factors and outcomes. Patient parameters collected include age, gender, indication for surgery, laterality, type of incision, incision length, use of lacrimal stent, length of follow-up, intra- and postoperative complications, and time to resolution of eyelid closure abnormalities. Statistical Analysis Software 9.1 (SAS Inc., Cary, NC) was used to perform a 2-tailed Fisher exact test with clinical significance designated at ␣ ⫽ 0.05.
Results Figure 1. Incision types for external dacryocystorhinostomy (DCR). The vertical incision is placed 1 cm medial to the medial canthus, extending 1 to 1.5 cm in length (black line). The nasojugal incision is placed along the shallow curvilinear groove that lies infralateral to the medial canthus and extends 1 to 1.5 cm in length (white line). The eyelid margin incision is placed along the medial canthal angle predominantly along the lower eyelid margin and is approximately 1 cm in total length (dotted line).
tions that aim for improved scar concealment are also used, with incision placement along the nasojugal skin fold or the eyelid margin.23–25 The surgeon must dissect through skin, orbicularis oculi muscle, and periosteum to reach the anterior lacrimal crest while avoiding the angular vessels. In traditional teaching, injury to the facial nerve is not a concern during external DCR surgery because the motor innervation of the periocular muscles is believed to be well away from the incision site.26,27 It is commonly accepted that branches of the temporal, zygomatic, and buccal divisions of the facial nerve travel from lateral to medial across the zygoma and maxilla to send terminal nerve fibers to the posterior surface of the orbicularis oculi muscle, providing its innervation.26,27 Damage to facial nerve fibers of the upper or lower eyelid during ocular procedures, such as blepharoplasty or ptosis correction, can result in orbicularis oculi weakness, incomplete or delayed eyelid closure, and lagophthalmos.28 In this review, we propose that facial nerve injury is possible during external DCR surgery, resulting in orbicularis oculi muscle weakness, and propose an anatomic explanation for this observation.
Materials and Methods A retrospective chart review was performed of 4 oculofacial plastic surgeons’ practices (BSS, SJL, RLA, JDM) from 2005 to 2007 of consecutive patients who had nasolacrimal duct obstruction repair by external DCR. Approval for the study was obtained from the Western Institutional Review Board (Olympia, WA). The study was compliant with the Health Insurance Portability and Accountability Act and in accord with the principles outlined in the Declaration of Helsinki. Patients were identified who developed clinical signs of postoperative orbicularis oculi muscle weakness in the forms of hypometric blink or lagophthalmos with or without punctate keratopathy on the operated side. Hypometric blink was defined as delayed or incomplete blink of the upper eyelid on the operated side compared with the contralateral side. Lagophthalmos was identi-
586
There were 16 patients (7.4%) identified among 215 individuals and 247 surgeries who demonstrated abnormalities of eyelid closure in the postoperative period after external DCR. Of these, 13 patients had lagophthalmos with or without hypometric blink and 3 patients had purely hypometric blink (Table 1). Indications for surgery included acquired nasolacrimal duct obstruction in 13 patients and concurrent canalicular stenosis and nasolacrimal duct obstruction in 3 patients. In 2 of these patients, repeat external DCR was performed because of primary surgical failure. Concurrent lower lid ectropion surgery via a tarsal strip procedure was performed in 2 patients. The group was composed of 13 female and 3 male patients with a mean age of 60 years (median 61 years; range, 34 – 85 years). The left side was operated in 69% of cases. The Fisher exact test demonstrated no statistically significant differences in either the gender or laterality distributions between patients who underwent DCR surgery with eyelid closure abnormalities and those without (␣ ⫽ 0.77 and 0.20, respectively). We did not find any correlation between the indication for surgery and orbicularis oculi weakness. Of the 247 DCRs performed, 121 were through a nasojugal incision, 92 were through a vertical incision, and 34 were through an eyelid margin incision. When analyzed by incision type, 9.1% (11) of DCRs with a nasojugal incision demonstrated eyelid closure abnormalities, whereas those with vertical and eyelid margin incisions were associated with abnormalities 4.3% (4) and 2.9% (1) of the time, respectively. The Fisher exact test comparing the rates of abnormalities across the 3 distinct incision types did not reveal a statistically significant difference (␣ ⫽ 0.37). Incision length was on average 13.5 mm. Ten of the 16 patients had placement of a lacrimal stent. One surgeon (SJL) did not routinely place stents. The stent remained in place for an average of 10 weeks. Follow-up ranged from 3 to 50 weeks (mean, 20 weeks; standard deviation, ⫾13 weeks). The degree of postoperative lagophthalmos was on average 1.5 mm (range, 1–2 mm) (Fig 2). Four patients had inferior punctate keratopathy demonstrated by slit-lamp examination. Resolution of lagophthalmos was seen on average by 14 weeks (standard deviation, ⫾9 weeks). The longest time to resolution was 32 weeks. Three individuals (patients 1, 3, and 7) had residual hypometric blink after resolution of lagophthalmos, and one individual (patient 13) was lost to follow-up in the early postoperative period at 3 weeks before resolution could be documented.
Discussion Iatrogenic injury to the seventh cranial nerve is often a concern during facial plastic surgery, including lifting procedures for the forehead, mid-face, and lower face.29 Damage to temporal and zygomatic branches during the aforemen-
Vagefi et al 䡠 Facial Nerve Injury during External DCR Table 1. Characteristics of Patients with Facial Nerve Injury during External Dacryocystorhinostomy Age Patient (y) Gender Side 1 2 3 4
42 70 34 56
F F F M
R L L L
5 6 7
76 61 85
M F F
L R L
8
41
F
L
9
84
F
L
10
44
F
L
11 12
39 60
F F
R R
13 14 15 16
68 59 72 74
M F F F
L R L L
Indication Acquired NLDO Acquired NLDO Acquired NLDO Acquired NLDO/ canalicular stenosis Acquired NLDO Acquired NLDO Recurrent Acquired NLDO Acquired NLDO/ chronic dacryocystitis Acquired NLDO/ chronic dacryocystitis Recurrent acquired NLDO/canalicular stenosis Acquired NLDO Acquired NLDO/ canalicular stenosis Acquired NLDO Acquired NLDO Acquired NLDO Acquired NLDO
Time to Prior Type of Incision/ Punctate Hypometric Resolution of Resolution DCR Stent Placement Lagophthalmos Keratopathy Blink Symptoms (wk) No No No No
Vertical/Yes Vertical/Yes Vertical/Yes Vertical/Yes
2 mm None 1 mm 2 mm
None None None Yes
Yes Yes Yes Yes
Continued HB Yes Continued HB Yes
28 32 14 10
No Eyelid Margin/Yes No Nasojugal/Yes Yes Nasojugal/Yes
None 2 mm 1 mm
None None Yes
Yes Yes Yes
Yes Yes Continued HB
27 12 9
No
Nasojugal/No
None
Yes
Yes
Yes
10
No
Nasojugal/Yes
1 mm
Yes
Yes
Yes
8
Yes
Nasojugal/Yes
1 mm
None
Yes
Yes
12
No No
Nasojugal/Yes Nasojugal/No
2 mm 1 mm
None None
Yes Not noted
Yes Yes
3 10
No No No No
Nasojugal/No Nasojugal/No Nasojugal/No Nasojugal/No
1 mm 2 mm 2 mm 1 mm
None None None None
Yes Not noted Not noted Not noted
Lost to F/U Yes Yes Yes
– 12 8 6
DCR ⫽ dacryocystorhinostomy; F/U ⫽ follow-up; HB ⫽ hypometric blink; NLDO ⫽ nasolacrimal duct obstruction.
tioned procedures and during cosmetic and reconstructive eyelid surgery can result in orbicularis oculi muscle weakness. As a consequence, poor eyelid closure and lagophthalmos may occur from injury to fibers innervating the upper eyelid and scleral show and ectropion from injury to fibers to the lower eyelid.28,29 Traditionally, innervation of the eyelids is believed to proceed from lateral to medial with the temporal branches providing terminal fibers to the upper eyelid and the zygomatic and buccal branches supplying fibers to the lower eyelid.25,26 These fibers initially travel in the superficial musculoaponeurotic system and become more superficial as they approach the lateral orbit, eventually assuming a vertical orientation in the post-orbicularis oculi fascial plane and penetrating the muscle on its posterior surface.28 Facial nerve injury has not been described after external DCR surgery. We have observed in a small portion of patients a spectrum of findings consistent with orbicularis oculi muscle weakness of the upper eyelid, ranging from delayed or incomplete blink to clinically significant lagophthalmos with corneal exposure. We speculate that these are secondary to intraoperative trauma to the seventh cranial nerve. We were only able to identify a single report of 2 cases of abnormal eyelid closure after external DCR, and these were not attributed to nerve injury.30 Recent anatomic studies have provided a more detailed understanding of the facial nerve distribution of the eyelids and periorbital region.31–33 Nemoto and colleagues31 demonstrated through microscopic dissections that superficial
buccal branches of the facial nerve not only supply the levator labii superioris, the lower portion of the orbicularis oculi and the levator labii superioris alaeque nasi, but also course superiorly over the medial palpebral ligament with the angular artery to innervate the upper part of the orbicularis oculi, procerus, and corrugator supercilii (Fig 3). In 42% of specimens, they observed that the nasal superficial buccal branches innervate the upper orbicularis oculi muscle via a supramedial approach. Similarly, other studies have found the zygomatic and buccal branches coalesce to form what some authors have termed the angular nerve, which tracks superiorly to supply the procerus, corrugator, and presumably the medial orbicularis oculi muscles.32,33 The external DCR incision and dissection can be situated in the path of these superior and medial coursing branches of the zygomatic and buccal nerve. We propose that in a minority of patients, disruption of these fibers, whether from surgical transection or thermal injury from cautery, results in abnormal eyelid closure. The majority of patients who undergo external DCR surgery do not experience eyelid closure abnormalities. We postulate several reasons for this finding. First, there exists sufficient anatomic variation in the distribution of the peripheral branches of the zygomatic and buccal nerve that these fibers could be preserved and not disrupted by the incision, dissection, or cautery as they course superiorly through the medial canthal region.31 Furthermore, it is possible that despite injury to the facial nerve in the medial canthal area, certain individuals have redundant innervation
587
Ophthalmology Volume 116, Number 3, March 2009
Figure 2. Postoperative clinical appearance of orbicularis oculi weakness after external dacryocystorhinostomy (DCR). A, Patient 3, a 34-year-old woman, presented 4 weeks after left-sided external DCR with 1 mm of lagophthalmos and hypometric blink of the left eye. B, Patient 4, a 56-yearold man, was reviewed at 4 weeks after left-sided external DCR and found to have 2 mm of lagophthalmos, punctuate keratitis, and hypometric blink of the left eye. C, Patient 1, a 42-year-old woman, presented 6 weeks after right-sided external DCR with 2 mm of lagophthalmos and hypometric blink of the right eye.
of the upper eyelid orbicularis muscle provided by the temporal branch of the facial nerve as it courses lateral to medial across the upper eyelid.34 In addition, most surgeons typically use blunt dissection techniques and cautious use of cautery after the skin incision, which likely reduce the chance for nerve injury. Other iatrogenic causes for the constellation of findings should be considered. Myotoxic effects of local anesthetics may result in muscle degeneration; however, studies have demonstrated that muscle hypertrophy is induced after a brief period of paralysis, resulting in a strengthening effect by 3 weeks.35 No cases of human orbicularis oculi toxicity from lidocaine or bupivacaine have been reported in the literature. Bupivacaine-associated orbicularis oculi myotoxicity has been demonstrated in a rabbit model; however, the toxicity usually required high doses of the anesthetic with concurrent use of hyaluronidase.36 In our study, none of the surgeons used hyaluronidase and one of the surgeons never used bupivacaine, yet all experienced this complication. Alternatively, a contusion of the orbicularis oculi muscle from local anesthesia injection or surgical dissection may cause muscle dysfunction. Orbicularis oculi weakness may occur as a consequence of muscle necrosis from aggressive
588
cautery. Surgical dissection can possibly result in disinsertion of the orbicularis oculi muscle or medial canthal tendon at the time of periosteal cleavage. Nevertheless, these mechanisms would not explain the weakness of the superior orbicularis oculi muscle, because surgery is performed below the medial canthal tendon, or the gradual resolution of symptoms over the course of months, because disinsertion of the orbicularis oculi muscle or medial canthal tendon would not be expected to spontaneously improve. Furthermore, no telecanthus was observed in any patients. Because the study was retrospective in nature, no quantitative direct measure of facial nerve function (e.g., electromyelography) was performed. However, we believe our clinical observations are consistent with trauma to one or more peripheral branches of the facial nerve in the inferior eyelid that continue to course superiorly through the medial canthal area to supply the upper eyelid orbicularis oculi muscle from a medial approach. Perhaps the most supportive observation is that of delayed blink. A clinical sign of acute to subacute peripheral nerve injury affecting the function of an extraocular muscle is the observation of a delayed saccade.37 If a blink is analogous to a saccade, it follows that the observation of a delay in addition to an incomplete blink suggests a neurologic rather than a purely mechanical cause. In the present study, signs of orbicularis oculi muscle weakness were found in 7.4% of patients after external DCR surgery. The true incidence of orbicularis oculi weakness is likely higher. As we have gained greater awareness of this problem, subtle abnormalities of eyelid closure have been more frequently noted. In our cohort, a larger number of patients operated through a nasojugal approach demonstrated orbicularis oculi weakness. Theoretically, this incision is more likely to cross the path of the superficial buccal branch of the facial nerve and thus probably carries a higher risk of injury. Nevertheless, a statistically significant difference between the 3 incision types was not found. This, however, may be due to a small sample size. The finding of orbicularis oculi weakness with all incision types provides evidence of the anatomic variability of the superficial fibers of the facial nerve as they course to the medial upper eyelid. We believe this could also explain the spectrum of muscle weakness noted. Depending on the variable distribution of these medial fibers, which and how many of these fibers were affected during surgery, and the redundant innervation provided by the temporal branches of the facial nerve to the upper eyelid, a range of clinical findings could be observed. We suspect that with a superficial incision followed by blunt dissection and the judicious use of cautery, the likelihood of injury could be reduced. Normal function is typically regained by 3 months, although in one case weakness persisted to 8 months. Supportive care by means of artificial tears and lubricating ointment should be used in the interim. Furthermore, in cases of early surgical failure or persistent tearing after external DCR, lacrimal pump dysfunction secondary to orbicularis oculi weakness should be considered and repeat surgery avoided until recovery of nerve function. We propose that injury of the facial nerve during external DCR surgery can result in orbicularis oculi muscle weakness that manifests as abnormal eyelid closure and lagoph-
Vagefi et al 䡠 Facial Nerve Injury during External DCR
Figure 3. Left superficial buccal branches of the facial nerve. A, Anterior view (left) and in diagrammatic form (right). Superficial buccal branches spread into terminal rami to surrounding muscles. Also demonstrated is angular artery. B, Posterior view (left) and in diagrammatic form (right). Lower forceps turn over levator labii superioris, and upper forceps pull upper end of levator labii superioris alaeque nasi nasally, to show course of superficial buccal branches. Terminal rami supply procerus, corrugator supercilii, and orbicularis oculi. Reprinted with permission from Nemoto Y, Sekino Y, Kaneko H. Facial nerve anatomy in eyelids and periorbit. Jpn J Ophthalmol 2001;45:445–52. Aa ⫽ angular artery; CS ⫽ corrugator supercilii; OO ⫽ orbicularis oculi; Pr ⫽ procerus; SBBr ⫽ superficial buccal branch.
thalmos. This complication is likely caused by an insult to peripheral fibers of the zygomatic and buccal branches of the facial nerve as they course in the medial canthal area and provide innervation to the upper eyelid orbicularis muscle in a subset of individuals. In our cohort, the signs and symptoms of nerve injury were temporary and resolved in several months as neural function recovered. Patients should be appropriately informed of this operative risk and reassured that symptoms typically resolve should such a complication occur.
References 1. Tarbet KJ, Custer PL. External dacryocystorhinostomy: surgical success, patient satisfaction, and economic cost. Ophthalmology 1995;102:1065–70. 2. Dolman PJ. Comparison of external dacryocystorhinostomy with nonlaser endonasal dacryocystorhinostomy. Ophthalmology 2003;110:78 – 84. 3. Mirza S, Al-Barmani A, Douglas SA, et al. A retrospective comparison of endonasal KTP laser dacryocystorhinostomy
589
Ophthalmology Volume 116, Number 3, March 2009
4. 5. 6. 7.
8. 9. 10.
11. 12. 13. 14. 15. 16. 17. 18. 19.
20.
versus external dacryocystorhinostomy. Clin Otolaryngol Allied Sci 2002;27:347–51. Tsirbas A, Davis G, Wormald PJ. Mechanical endonasal dacryocystorhinostomy versus external dacryocystorhinostomy. Ophthal Plast Reconstr Surg 2004;20:50 – 6. Walland MJ, Rose GE. Factors affecting the success rate of open lacrimal surgery. Br J Ophthalmol 1994;78:888 –91. Becker BB. Dacryocystorhinostomy without flaps. Ophthalmic Surg 1988;19:419 –27. Cokkeser Y, Evereklioglu C, Er H. Comparative external versus endoscopic dacryocystorhinostomy: results in 115 patients (130 eyes). Otolaryngol Head Neck Surg 2000;123: 488 –91. Albert DM. Ophthalmic plastic surgery. In: Albert DM, Edwards DD, eds. The History of Ophthalmology. Cambridge, MA: Blackwell Science; 1996:235–53. Duffy MT. Advances in lacrimal surgery. Curr Opin Ophthalmol 2000;11:352– 6. Gonnering RS. Dacryocystorhinostomy and conjunctivodacryocystorhinostomy. In Dortzbach RK, ed. Ophthalmic Plastic Surgery: Prevention and Management of Complications. New York: Raven Press; 1994:237–50. Hurwitz JJ, Eplett CJ, Fliss D, Freeman JL. Orbital hemorrhage during dacryocystorhinostomy. Can J Ophthalmol 1992; 27:139 – 42. Neuhaus RW, Baylis HI. Cerebrospinal fluid leakage after dacryocystorhinostomy. Ophthalmology 1983;90:1091–5. Beiran I, Pikkel J, Gilboa M, Miller B. Meningitis as a complication of dacryocystorhinostomy. Br J Ophthalmol 1994;78:417– 8. Usul H, Kuzeyli K, Cakir E, et al. Meningitis and pneumocephalus: a rare complication of external dacryocystorhinostomy. J Clin Neurosci 2004;11:901–2. Besharati MR, Rastegar A. Results and complications of external dacryocystorhinostomy surgery at a teaching hospital in Iran. Saudi Med J 2005;26:1940 – 4. Jordan DR, Nerad JA. An acute inflammatory reaction to silicone stents. Ophthal Plast Reconstr Surg 1987;3:147–50. Anderson RL, Edwards JJ. Indications, complications and results with silicone stents. Ophthalmology 1979;86:1474 – 87. Jordan DR, Nerad JA, Tse DT. Complete canalicular erosion associated with silicone stents. Am J Ophthalmol 1986;101: 382–3. Hanna IT, Powrie S, Rose GE. Open lacrimal surgery: a comparison of admission outcome and complications after planned day case or inpatient management. Br J Ophthalmol 1998;82:392– 6. Salour H, Montazerin N. Incision site tissue necrosis after dacryocystorhinostomy. Ophthal Plast Reconstr Surg 1998;14: 146 – 8.
21. Kerendian J, Conn H. Lipogranuloma: a preventable complication of dacryocystorhinostomy. Ophthalmic Surg Lasers 1996;27:713–5. 22. Nerad JA. Oculoplastic surgery. In Krachmer JH, ed. The Requisites in Ophthalmology. St. Louis, MO: Mosby; 2001: 215–53. 23. Harris GJ, Sakol PJ, Beatty RL. Relaxed skin tension line incision for dacryocystorhinostomy. Am J Ophthalmol 1989; 108:742–3. 24. Dortzbach R, Woog JJ. Small-incision techniques in ophthalmic plastic surgery. Ophthalmic Surg 1990;21:615–22. 25. Putterman AM. Eyelid incision approach to dacryocystorhinostomy facilitated with a mechanical retraction system. Am J Ophthalmol 1994;118:672– 4. 26. Dutton JJ. Orbital Nerves. In: Dutton JJ, ed. Atlas of Clinical and Surgical Orbital Anatomy. Philadelphia, PA: Saunders; 1994: 33– 64. 27. May M. Anatomy for the clinician. In: May M, Schaitkin BM, eds. The Facial Nerve: May’s Second Edition. New York: Thieme; 2000:19 –56. 28. Jordan DR, Anderson RL. The facial nerve in eyelid surgery [letter]. Arch Ophthalmol 1989;107:1114 –5. 29. Seckel BR. Facial Danger Zones: Avoiding Nerve Injury in Facial Plastic Surgery. St. Louis, MO: Quality Medical Pub; 1994:1–52. 30. Fayet B, Bernard JA, Ritleng P, et al. Internal transient palpebral inclusion following dacryocystorhinostomy [in French]. J Fr Ophtalmol 1994;17:195–9. 31. Nemoto Y, Sekino Y, Kaneko H. Facial nerve anatomy in eyelids and periorbit. Jpn J Ophthalmol 2001;45:445–52. 32. Caminer DM, Newman MI, Boyd JB. Angular nerve: new insights on innervation of the corrugator supercilii and procerus muscles. J Plast Reconstr Aesthet Surg 2006;59:366 –72. 33. Hwang K, Jin S, Park JH, Chung IH. Innervation of the procerus muscle. J Craniofac Surg 2006;17:484 – 6. 34. Ouattara D, Vacher C, de Vasconcellos JJ, et al. Anatomical study of the variations in innervation of the orbicularis oculi by the facial nerve. Surg Radiol Anat 2004;26:51–3. 35. Scott AB, Alexander DE, Miller JM. Bupivacaine injection of eye muscles to treat strabismus. Br J Ophthalmol 2007;91: 146 – 8. 36. McLoon LK, Wirtschafter J. Regional differences in the subacute response of rabbit orbicularis oculi to bupivacaineinduced myotoxicity as quantified with a neural cell adhesion molecule immunohistochemical marker. Invest Ophthalmol Vis Sci 1993;34:3450 – 8. 37. Wong AM, McReelis K, Sharpe JA. Saccade dynamics in peripheral vs central sixth nerve palsies. Neurology 2006;66: 1390 – 8.
Footnotes and Financial Disclosures Originally received: April 20, 2008. Final revision: September 19, 2008. Accepted: September 26, 2008. Available online: December 16, 2008.
3
Manuscript no. 2008-491.
1
Scheie Eye Institute, University of Pennsylvania, Philadelphia, Pennsylvania.
2
Allure Facial Laser Center and Medispa, Kirkland, Washington.
590
Center for Facial Appearances, Salt Lake City, Utah. Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article. Correspondence: John D. McCann, MD, PhD, 1002 E. South Temple, Suite 308, Salt Lake City, UT 84102. E-mail: [email protected].