Ординатура / Офтальмология / Английские материалы / Strabismus Surgery and Its Complications_Coats, Olitsky_2007

Fig. 18.6. The deviation in prism diopters produced by a 40-prism- diopter plastic prism depending on the angle of incidence (degrees) when deviation is measured at a fixation distance of infinity. (Reprinted from Ophthalmology, volume 92, Thompson JT, Guyton DL, Ophthalmic prisms. Deviant behavior at near, page 684-690, 1985, with permission from American Academy of Ophthalmology [9])

(Δt true deviation, m measured deviation, D diopters) [7]. Table 18.4 provides representative examples offered by Scattergood and coworkers [7] of the impact of various spectacle lens powers on the measurement of the true deviation. They suggest that mental interpolation between these points is usually sufficient for clinical purposes.

18.2.4.2 Unrecognized Prism

Patients often do not know or recall that prism has been prescribed in their glasses. Routine assessment of spectacles in an ophthalmologist’s office may not detect the presence of prism (>Fig. 18.7). Undetected prism can lead to surgical undercorrection, which may go unrecognized until the patient obtains new glasses without prism following surgery. It is good practice to assess spectacles for the presence of induced prism prior to proceeding with strabismus surgery since identification of prism in the spectacles usually results in changes to the surgical plan. The portion of the spectacle lenses overlying the papillary axis should be marked prior to measuring the lens, and this mark centered in the lensometer when the lens reading is taken (>Fig. 18.8).

Fig. 18.7a,b. Prism can easily go undetected in spectacles if a the optical center of the lens is centered in the lensometer, rather than b the portion of the lens overlying the pupillary axis

18.2.5 Duction Limitation Errors

Patients with significant unilateral or asymmetric duction limitations can present interesting measurement issues, particularly when measurements are performed in the diagnostic positions of gaze. A significant duction limitation may prevent the patient from utilizing both eyes in the gaze position that the examiner is attempting to test. The patient may not recognize his/her role in the measurement process and may not realize that their inability to see the target with one eye represents a problem. This can result in marked underestimation of the

Fig. 18.8. The portion of the lens overlying the papillary axis should be marked and this mark centered in the lensometer to detect induced prism in the patient’s spectacles

deviation. The ophthalmologist must be aware of this potential problem and ensure that the patient actually alternates fixation during alternate prism and cover testing. Though this sounds simple and intuitive, it can present quite a problem in some patients. If a duction limitation precludes accurate measurements in the desired position, two approaches can be utilized:

(1) estimation of the deviation in the standard gaze position using the Krimsky light reflection test, or (2) measurement of the deviation in a reduced position of gaze that allows the patient to fixate the target with either eye.

18.2.6 Poor Fixation

Patients, young and old, do not necessarily understand the objectives of alternate prism and cover testing. Patients often do not see the fixation target with one eye, and sometimes both eyes, and may not inform the ophthalmologist that the target is not seen, believing that the doctor knows what he/she is doing without their help. If the patient does not alter fixation during alternate cover prism and testing, so that the uncovered eye always takes up fixation on the visual target, the angle of strabismus can be dramatically underestimated. Causes of poor fixation in a child include both fear and lack of interest. Fixation issues can even be a problem in estimating strabismus in adult patients. To ensure that a patient routinely fixates on the visual target, the examiner can have the patient continue to read a string of letters thereby assuring continued fixation.

When using larger prisms, those above 25–30 prism diopters, the image can be deviated so far toward the apex of the prism that it may not be possible for the patient to readily see the target even when they are trying to do so. When using large

Fig. 18.9a,b. Adjustment in the position of a large prism required to allow the patient to visualize the fixation target. a Patient unable to fixate target, and b patient able to fixate target after prism is moved to the right

prisms, we always ask whether the patient can see the target through the large prism, and adjust the position of the prism as needed until the patient can visualize the target before beginning testing (>Fig. 18.9).

18.2.7 Poor Cooperation

Diagnosis of a patient with strabismus can be difficult, and sometimes a definitive diagnosis is not possible. Preoperative diagnostic inaccuracy should be included in the differential diagnosis when postoperative alignment is not as expected. For example, we recently treated a newborn who had what we believed to be congenital left sixth nerve palsy. The child fixated with the right eye and had amblyopia of the involved left eye. The child had an esotropia of 50 prism diopters and we could never demonstrate abduction past the midline in his left eye despite repeated attempts. After a period of observation and following conclusion of amblyopia treatment, surgery was recommended. The surgical plans called for a transposition of the superior and inferior rectus muscles to the lateral rectus

muscle insertion with posterior fixation suture augmentation to treat his sixth nerve palsy.

This would have been the plan carried out except that during the preoperative assessment 1 week before surgery the child’s mother mentioned that she had occasionally seen the left eye turn outward past the midline. With much effort, we did eventually demonstrate that the child had almost full abduction of the left eye, greatly altering our surgical plans. A recess/resect operation was performed with an excellent postoperative result. Such diagnostic errors do not reflect poor performance on the part of the ophthalmologist, but rather reflect the complexities and difficulties associated with evaluation and management of strabismus, especially in a young child. The most common problems to be misdiagnosed prior to surgery in our experience include sixth nerve palsy, Duane syndrome, and strabismus with an unrecognized accommodative component. In situations where we feel that patient cooperation significantly limits our ability to make an accurate diagnosis, we recommend deferring surgical intervention until the child is older and more cooperative.

18.3 Incomplete Diagnosis

We once operated on a 4-year-old girl with a right superior oblique palsy who had marked right inferior oblique overaction and a left head tilt. A right inferior oblique recession was performed without complications. The patient returned 1 week later at which time her parents proclaimed that the surgery had been a complete failure and that her inferior oblique overaction and head tilt were completely unchanged. Compared with preoperative photographs and records, which demonstrated a left head tilt and right inferior oblique overaction, the patient now had a right head tilt and left inferior oblique overaction, prompting a diagnosis of bilateral superior oblique palsy that was not recognized prior to surgery. The patient underwent an inferior oblique muscle weakening procedure on the contralateral eye with good results. Obviously, it would have been better to diagnose the bilateral nature of the problem prior to surgery. Unlike an adult patient, where the nine diagnostic positions of gaze and head tilt test can be performed with good patient cooperation and usually with good accuracy, this is not always possible with a child, where surgical decisions must often be made based upon primary position measurements and evaluation of ductions alone. Undiagnosed Duane syndrome, accommodative esotropia, dissociated strabismus complex, and mild cranial neuropathies are also occasionally associated with unexpected postoperative alignment problems.

18.4Management Errors at the Time of Surgery

Once the patient has been properly examined, the strabismic deviation accurately measured and decisions regarding the surgical plan finalized, the basic plan should be understood

Chapter 18

by the patient and the surgeon must accurately carry out the plan, unless intraoperative findings call for an alteration of the original plans. The value of making sure that the patient understands the basic surgical plan and that the patient understands the possibility of an intraoperative change in plans cannot be overemphasized. It is much easier to explain a decision to perform an operation that may not be easily understood by the patient prior to surgery than attempting to do so after surgery. This is especially true if the outcome is less than what was anticipated. For example, many patients and/or families may not understand the recommendation to operate on both eyes when the patient has only “bad one,” the eye which the patient and family see deviating. An explanation in the examination room before surgery can usually resolve this confusion. This is best done prior to surgery and attempts to explain the decision about the surgical plan can be difficult after the surgery has been completed.

Opinion varies as to the optimal time to obtain surgical consent. Some feel that it may be preferable for the operating surgeon to obtain surgical consent just prior to surgery, giving the patient the opportunity to have any remaining questions answered prior to surgery and it also helps reinforce to the surgical team which eye is to be operated and what procedure is to be performed. Others believe that it is optimal to obtain surgical consent in the more relaxed setting of the office days or weeks prior to surgery, followed by a brief review just prior to surgery.

18.4.1 Marking the Surgical Site

Once the surgical consent has been obtained and the patient has been taken to the operating room, strict adherence to fundamental operating room safety protocols can help ensure that the procedure for which consent was obtained is the actual procedure that is performed and that the procedure is performed on the correct patient. Errors involving the performance of the wrong surgical procedure or at the wrong surgical site should, in theory, never occur. Errors in performing the consented procedure represent a breakdown in operating room protocol, and often involve errors at multiple steps in the process. Many hospitals require the surgeon to mark the operative site prior to the patient being brought into the operating room, especially when unilateral surgery is planned. Marking the surgical site should be done prior to taking the patient to the operating room after confirming the surgical site with the patient and/or parent, as appropriate, in the preoperative holding area. Even if this is not a requirement in a particular hospital, this safety precaution is easy to implement and provides reassurance to the patient and surgical team alike.

When marking the surgical site, marking with an “X” is not recommended, but rather the surgeon’s initials should be utilized. This mark should be placed in an area that will still be visible after the surgical draping has been placed and the marking pen should contain ink that is resistant to removal during surgical preparation (Chap. 5). Additionally, some surgeons also prefer to mark the type of deviation (i.e., XT, ET,

etc.) for an added measure of safety. Prior to the first incision being made, a “timeout” should take place when the operating surgeon, anesthesiologist and other relevant operating room personnel all confirm that the surgical plan matches the operative consent, and the surgical team should confirm the identity of the patient through at least two patient identifiers (name, date of birth, and medical record number). This should be done even for patients who are undergoing surgery under local anesthesia. Unfailing and meticulous attention to these protocols, even when the site seems obvious or routine, can help to reduce the potential for a wrong site or wrong procedure error.

Strabismus surgery, like all surgeries, has known associated risks. Some risks may be unavoidable even in the best of circumstances. With thoughtful and accurate preoperative management, the strabismus surgeon can help to minimize most of these risks and starting with an accurate surgical plan and surgical process of care are important first steps.

References

1.Kushner BJ (1991) Pseudo inferior oblique overaction associated with Y and V patterns. Ophthalmology 98:1500–1505

2.Kushner BJ (1997) Restriction of elevation in abduction after inferior oblique anteriorization. J AAPOS 1:55–62

3.Capó H, Mallette RA, Guyton DL (1988) Overacting oblique muscles in exotropia: a mechanical explanation. J Pediatr Ophthalmol Strabismus 25:281–285

4.Choi RY, Kushner BJ (1998) The accuracy of experienced strabismologists using the Hirschberg and Krimsky tests. Ophthalmology 105:1301–1306

5.Repka MX, Arnoldi KA (1991) Lateral incomitance in exotropia: fact or artifact? J Pediatr Ophthalmol Strabismus 28:125–128; discussion 129–130

6.Thompson JT, Guyton DL (1983) Ophthalmic prisms. Measurement errors and how to minimize them. Ophthalmology 90:204–210

7.Scattergood KD, Brown MH, Guyton DL (1983) Artifacts introduced by spectacle lenses in the measurement of strabismic deviations. Am J Ophthalmol 96:439–448

8.Hirayama K, Satou M, Gotou H, Watanabe K, Yamamoto T (1995) [Cerebral diplopia and triplopia – a proposal for responsible lesion and mechanism.] Rinsho Shinkeigaku 35:744–750

9.Thompson JT, Guyton DL (1985) Ophthalmic prisms. Deviant behavior at near. Ophthalmology 92(5):684–690

19

Complications involving the anterior segment and ocular surface can vary from mild, inconvenient, and self-limiting to serious and vision threatening. This chapter will highlight common and uncommon problems that can be encountered during and after strabismus surgery and will review methods to both prevent complications and treat complications that do occur. The chapter is organized anatomically, with a section on corneal complications, conjunctival complications, scleral complications, and intraocular complications. Anterior segment ischemia is discussed separately in Chap. 20.

19.1 Corneal Complications

Pedersen [1] reported corneal abnormalities in more than half of patients who underwent horizontal rectus muscle surgery. Among 44 patients she studied, dellen developed in four patents while the remaining had less significant abnormalities, including defects in the precorneal tear film, and fluorescein and/or rose bengal staining. Sterile corneal infiltrates were noted in five patients.

19.1.1 Dellen

The term delle (plural dellen) means low ground or pit. Dellen are characterized as shallow, clearly defined excavations at the margin of the cornea. They typically develop within the first 2 weeks after surgery and are generally 1.5–2 mm in diameter and occur following localized evaporation and dehydration of the cornea (>Fig. 19.1). Disruption of the tear film and localized evaporation result in increasing compactness of the corneal stromal lamellae. The base of dellen typically appears hazy and dry and is rarely transparent. Fuchs [2] described the histological aspects of a delle. He found thinning of the epithelium peripherally and observed that the epithelium in the center of the lesion showed irregular thinning of the external stromal layer and Bowman’s membrane and noted the presence of leukocytes beneath Bowman’s membrane.

Dellen occur commonly after strabismus surgery, with a reported incidence of between 0.3% and 22.45% [3]. Tessler and Urist [4] retrospectively studied 170 cases of horizontal rectus

muscle surgery. Dellen formation occurred in 6.5% of patients operated with a limbal approach compared with 2.2% for those operated using a nonlimbal approach. Dellen probably often go undiagnosed because subjective symptoms may be absent and clinical findings may be subtle. Additionally, strabismus surgery is often performed on small children who cannot be readily examined at a slit lamp. Mai and Yang [3] reported the occurrence of dellen after 22.45% of strabismus surgeries in patients who had undergone a rectus muscle recession or resection. The occurrence of dellen was much more common after rectus muscle resection procedures (47.75%) compared to rectus muscle recession procedures in which this complication occurred in only 5.13% of eyes. These investigators performed rectus muscle recession surgery through a limbal conjunctival incision with recession of the conjunctival flap to the insertion of the rectus muscle. The surgical approach used for rectus muscle resections was not clearly delineated in the report. It has been our experience that dellen are much more likely to occur following the very large resections, while small resections are not likely to be associated with dellen formation.

Mai and Yang [3] related the formation of dellen after strabismus surgery to changes in the tear film breakup time before and after surgery. The average preoperative tear film breakup time was 28.75 seconds (10.96–91.80 s). The breakup time was reduced in all patients postoperatively, regardless of the procedure performed and whether or not a delle formed.

Fig. 19.1. Corneal delle following strabismus surgery

The tear film breakup time in the group that experienced dellen formation was 23.22 s preoperatively and was reduced to 8.61 s postoperatively. In comparison, in the group that did not experience dellen formation, the tear film breakup time for those who had undergone a rectus muscle recession was 30.22 s preoperatively compared to 22.25 s postoperatively. For those without dellen formation who had undergone a rectus muscle resection, the breakup time was 29.71 s preoperatively and 15.83 s postoperatively. They concluded that local corneal dehydration and ultimately dellen formation was caused by reduced tear film breakup time. It has also been our anecdotal experience that dellen formation is more likely following rectus muscle resection surgery compared with rectus muscle recession surgery. Additionally, dellen formation appears to be more common in our experience following strabismus surgery performed through a limbal incision, though we have not formally studied this association. Common to almost all cases of dellen formation is elevation of the bulbar conjunctiva near the limbus adjacent to the lesion. It seems logical that limbal incisions and rectus muscle resection procedures, both of which tend to produce a greater degree of bulbar conjunctival edema and elevation near the limbus, would be associated with an increase in the formation of dellen. While we agree that the relationship between dellen formation and tear film breakup is interesting, we do not advocate the need for assessment of the breakup time prior to strabismus surgery.

We have seen corneal dellen formation following horizontal rectus muscle surgery, but have never seen dellen formation after vertical rectus muscle or oblique muscle surgery. Presumably this is because tear film evaporation and corneal drying are more likely to occur in the middle of the open palpebral fissure compared to the more protected areas superiorly and inferiorly. Though most patients with dellen are asymptomatic, some experience excessive lacrimation. Patients may also complain of mild ocular discomfort, but it has been our impression that dellen rarely produce pain. While usually mild and self-limiting, serious complications have been reported following formation of dellen. Insler and co-workers [5] reported the formation of a descemetocele in a patient who developed a long-lasting corneal delle following vitrectomy surgery. The patient required a patch graft to the cornea to prevent corneal perforation. Zehl and Snell [6] reported corneal ulceration with dellen-like formation “in a region where there was considerable conjunctival thickening adjacent to the limbus,” in a patient with paralysis of cranial nerves 5, 6, and 7 who underwent strabismus surgery.

Treatment of dellen involves corneal rehydration and measures to reduce limbal conjunctival elevation. We generally prescribe a lubricating ophthalmic ointment three to four times per day to the affected eye. Dellen usually resolve within a few days to a week, accompanied by spontaneous reduction in bulbar conjunctival swelling as postoperative healing progresses. We have occasionally seen corneal dellen and bulbar conjunctival swelling persist for longer periods of time, prompting empirical use of topical steroids three to four times a day in an effort to hasten healing of the conjunctiva, which will translate into improvement of the associated corneal pathology. While we have not experienced serious complications associated with corneal dellen formation, the possibility of a serious compli-

Chapter 19

cation exists and patients with corneal dellen are monitored more closely than the standard postoperative patient. Patients with an underlying tear film deficiency may be at highest risk and may require closer follow-up.

19.1.2 Corneal Abrasions

Corneal abrasions can occur due to unintentional corneal trauma caused by needles, sutures, and other instrumentation during surgery. Spontaneous corneal epithelial defects may be seen during strabismus surgery in patients with epithelial basement membrane disturbances. Hydration of the cornea during surgery may reduce the occurrence of spontaneous corneal erosion in patients with basement membrane disease. In general, however, we favor minimized application of balanced salt solution and other hydrating solutions during strabismus surgery, because these solutions tend to cause edema of exposed Tenon’s fascia in the operative site. Hydration of Tenon’s fascia can complicate surgery and can make closure more difficult. A corneal abrasion occurring during strabismus surgery should be managed using ordinary measures. If the abrasion is small, we generally do not recommend an eye patch but we do advise patients and/or parents of the presence of the corneal epithelial defect and we follow the patient more closely after surgery until the epithelial defect heals.

19.1.3 Corneal Ulcer

Both infectious and noninfectious corneal ulceration have been reported following strabismus surgery. We are aware of one unpublished case of bilateral pseudomonas microbial keratitis that resulted following strabismus surgery in which the surgeon utilized corneal bridle sutures to position the eyes for surgery. The use of corneal bridle sutures imposes an unnecessary risk of serious ocular infection by disruption of the corneal epithelium and stroma and should be avoided. Positioning of the eye for surgery can be accomplished through other means including placement of bridle sutures through the conjunctiva or use of positioning forceps.

Zehl and Snell [6] reported a case of serious corneal complication following strabismus surgery to realign the eyes of a child with a left sixth nerve palsy. The child had a history of a left cerebellar hemisphere astrocytoma and concurrent fifth and seventh nerve palsies. A left lateral rectus muscle resection and left medial rectus muscle free tenotomy were performed. Two weeks after surgery the authors reported that “corneal ulceration had supervened with dellen-like formation temporally, in a region where there was considerable conjunctival thickening adjacent to the limbus.” Additional infiltrative lesions developed 2 weeks later. Supportive therapy with a therapeutic soft contact lens and antimicrobial therapy in the form of topical antibiotics resulted in healing of the corneal lesions and 6 months following surgery the cornea remained stabile. The authors believed that extraocular muscle surgery in patients with complete loss of fifth and seventh nerve function

is associated with an increased, yet not unacceptable risk of complications.

Wintle and coworkers [7] reported the development of an inferior corneal ulcer in a patient with Pendred syndrome. This autosomal recessive condition is associated with congenital deafness and thyroid goiter. The child underwent a transposition procedure in both eyes to treat a bilateral sixth nerve palsy. Corneal ulceration of the left eye developed 2 months following surgery, without significant ocular discomfort. Corneal sensation testing revealed normal sensation of the right eye but significantly reduced sensation of the left cornea. The ulceration responded to treatment with topical antibiotics, ocular lubricants, and a glue tarsorrhaphy. The author stressed the importance of determining corneal sensation in patients with atypical disease. While this may be good advice for a rare and unusual patient, the routine testing of corneal sensation prior to strabismus surgery is unnecessary.

Several ocular diseases can cause reduced corneal sensation, including a herpes simplex keratitis, postoperative anterior segment ischemia, intraoperative damage to the long posterior ciliary nerves or ciliary ganglion, and congenital absence of corneal sensation. Caution should obviously be exercised when performing any surgery, including strabismus surgery, on patients who are at increased risk for development of corneal complications due to exposure keratopathy and/or reduced corneal sensation. The presence of one of these conditions, however, is not a contraindication to strabismus surgery, but affected patients should be made reasonably aware of the potential risks associated with surgery and should not undergo surgery unless a significant clinical benefit is anticipated. Detailed preoperative discussion with patients and/or families is important and such patients should be followed carefully after surgery. Frequent topical lubrication is helpful and such patients should be advised of the warning signs of a serious corneal complication.

The most common condition that the strabismus surgeon is likely to encounter in which there is a substantial risk of a corneal complications is the surgical management of patients with a third cranial nerve palsy. Such patients present a complex treatment matrix and the risk-to-benefit profile must be evaluated on a case-by-case basis. Affected patients desire an eye that is surgically aligned in the primary position and an upper eyelid that is surgically elevated to match their unaffected eye. Absence of a Bell’s phenomenon and lagophthalmos resulting from ptosis repair combine to place the ocular surface in a compromised situation.

Our typical surgical paradigm for adult patients with a third nerve palsy is to manage their strabismus and ptosis in separate operations. Ptosis repair is not considered until ocular alignment in the primary position has been optimized and the patient has demonstrated ability to comfortably fuse when the lid is elevated with a reasonably comfortable head posture or demonstrates that they are not significantly bothered by diplopia from residual ocular misalignment. Ptosis surgery is offered to complete the procedure after a detailed discussion with the patient. It is very important to discuss the goals of ptosis repair with patients and families, which for us is undercorrection (>Fig. 19.2). The goal of ptosis undercorrection is much easier to accept when discussed prior to surgery

and patients and parents may be rather disappointed if they do not become fully aware of the surgical goals until after surgery. Patients who are unable to achieve comfortable single vision or who only achieved a very small field of single vision following strabismus surgery may not be good candidates for aggressive ptosis repair. We prefer the use of a frontalis suspension technique for correction of ptosis in this setting, because we believe the procedure results in a more predictable lid position and that the surgery is easier to reverse, if needed, should a corneal complication related to exposure develop. Fascia lata suspensions should be avoided because they are more difficult to reverse. We will generally use silicon rods or suture to accomplish frontalis suspension in patients with a third nerve palsy and always aim to undercorrect the ptosis to minimize the risk of corneal exposure complications. Silicone rods may be especially useful in this setting as they may allow better closure in patients with good orbicularis function.

We have treated several patients with corneal exposure following surgery to treat a third cranial nerve palsy. These complications have generally been mild, consisting of corneal drying, punctuate erosions, and occasionally large epithelial erosions. Aggressive lubrication, occasionally accompanied by a temporary tarsorrhaphy, has generally resulted in healing with a good outcome. We had one child who presented 6 months following surgery with microbial keratitis. Fortified

Fig. 19.2. Intentional undercorrection of ptosis following strabismus surgery in a child with a bilateral third nerve palsy

topical antibiotics resulted in healing of the corneal ulcer, but persistent exposure problems ultimately forced us to reverse his ptosis repair. The resulting corneal scar was also associated with development of severe deprivational amblyopia.

19.1.4 Filamentary Keratitis

Mild filamentary keratitis is not infrequently seen following strabismus surgery particularly in patients who have concurrent ptosis or who develop temporary mechanical ptosis following strabismus surgery. The condition is generally mild, asymptomatic, and self-limited. Pons and Rosenberg [8] reported a patient who developed filamentary keratitis following a recess/resect operation to treat a third nerve palsy (>Fig. 19.3). The patient responded favorably to standard treatment including debridement, ocular lubrication, punctal plugs, and topical steroids. Her symptoms were controlled with long-term use of an extended wear soft contact lens.

19.1.5 Reduced Endothelial Cell Count

Müller and coworkers [9] reported an unexpectedly high loss of endothelial cell density in children following strabismus surgery, and suggested that inflammatory stress and differences in eye growth associated with strabismus surgery in children might be the cause. The implications of this findings for children undergoing strabismus surgery is unclear and probably not clinically important.

Chapter 19

19.1.6 Corneal Toxicity

Hamed and co-workers [10] reported a case of presumed accidental corneal exposure to Hibiclens® (chlorhexidine 4% and detergent) in two patients during craniofacial surgery and cataract surgery, respectively. Both patients developed severe and permanent corneal ectasia and opacification. They subsequently confirmed toxicity of this agent to the cornea in rabbit eyes. Hibiclens®, sometimes utilized for surgical preparation of operative sites, should not be used in preparation of patients for ophthalmologic surgery. We utilize 5% povidone iodine for preoperative preparation (Chap 5) prior to strabismus surgery both because of its lack of significant toxicity to the ocular surface and its excellent role in reducing the normal bacterial flora of the ocular surface [11].

19.2 Conjunctival Complications

Complications involving the conjunctiva are probably the most common problems that occur during or following strabismus surgery. Complications vary from anatomical defects related to closure of the conjunctiva to infection, and can range in severity from trivial to vision threatening. Conjunctival closure problems occur most commonly following strabismus surgery performed through a limbal incision, but can occur with a fornix incision also. Careful analysis of conjunctival anatomy while in the operating room prior to starting surgery is key to accurate closure of the conjunctiva at the end of surgery.

Fig. 19.3. Filamentary keratitis of a patient’s left eye 1 week following a recess/recess operation to treat a third cranial nerve palsy. (Reprinted from [8] Journal of AAPOS, Volume 8, Pons ME, Rosenberg SE, Filamentary keratitis occurring after strabismus surgery, pp 190–191, 2004, with permission from American Association for Pediatric Ophthalmology and Strabismus)

19.2.1Inadvertent Advancement

of the Plica Semilunaris Conjunctivae

The term plica refers to an anatomical structure in which there is folding over of the parts. The plica semilunaris conjunctivae (referred to as plica here) represents a fold in the palpebral conjunctiva of the medial angle of the conjunctiva. It serves no particular function in the human eye but when surgically malpositioned it can present a serious cosmetic defect and may cause restriction of abduction.

Inadvertent advancement of the plica typically results from the suturing of plica to the conjunctiva adjacent to the limbus. This complication may occur following strabismus surgery using a limbal incision. During standard limbal surgery, the anterior aspect of the limbal conjunctival flap can become folded beneath the plica. At the conclusion of the surgery, the surgeon attempts to identify the anterior edges of the conjunctival flap. The plica may have become hydrated and swollen during surgery and the surgeon is lulled into believing that he/she has grasped the anterior corners of the conjunctival flap, when in fact the edges of the plica have been inadvertently grasped. Figure 19.4 demonstrates how easily this can occur. In the figure, the surgeon is intentionally holding the plica against the limbus at the conclusion of a case. Hydration of the plica dur-

ing surgery produces the casual appearance that the surgeon has accurately grasped the anterior corners of the limbal conjunctival flap. Only by careful study of the anatomy will the surgeon recognize that the medial angle of the eye appears abnormal and that the swollen plica has been inadvertently advanced forward. Once the mistake is recognized, key landmarks can be easily identified.

This complication is most likely to occur in the hands of a surgeon who performs strabismus surgery infrequently. In our experience, disorientation during closure appears to occur most commonly with resident and fellow surgeons in the early part of their training. Advancement of the plica may be more likely following prolonged surgery and is most prone to occur after an extensive reoperation and in older patients with extremely thin conjunctiva. The use of excessive hydration during surgery may result in greater distortion of the anatomy by increasing edema of tissues in the operative site and, thus, we do not use a

large amount of hydrating fluids during strabismus surgery in part to avoid obscuring important surgical landmarks.

Prevention of inadvertent advancement of the plica should begin before surgery is initiated. The surgeon should examine the anatomy of the medial aspect of the eye and take care to restore the anatomy to its preoperative state at the time of closure. In cases where we anticipate a difficult closure we will often place 6-0 Vicryl marker sutures (>Fig. 19.5) at the anterior corners of the conjunctival flap before proceeding with surgery. At the conclusion of the case, these corner markers are easy to locate and they help facilitate accurate conjunctival closure. To facilitate closure in routine cases for the surgeon who performs strabismus surgery infrequently, we often recommend using a sterile methylene blue skin marking pen to outline the conjunctiva prior to making the incision, a process that facilitates both accurate creation of the conjunctival incision and closure at the end of the case (>Fig. 19.6).

Fig. 19.4a,b. Inadvertent advancement of the plica semilunaris during closure of a limbal incision: a appearance of the medial angle of the conjunctiva when the plica is held against the limbus, demonstrating how the surgeon can be fooled into believing that the edge of the

conjunctival flap has been identified, and b appearance of the same eye after release of the plica and acquisition of the edges of the conjunctival flap

Fig. 19.5. Placement of Vicryl suture tags on the corners of the conjunctival flap during the initial stages of surgery to facilitate identification of the flap at the conclusion of surgery

Fig. 19.6. Use of a sterile methylene blue skin-marking pen to outline the conjunctiva prior to incision facilitates later closure of the conjunctiva