Ocular Surface Disease: Surgical Management

Published on 08/03/2015 by admin

Filed under Opthalmology

Last modified 22/04/2025

Print this page

This article have been viewed 1930 times

Ocular Surface Disease

Surgical Management

David S. Rootman, Judy Y.F. Ku and Sonia N. Yeung

Introduction

Ocular surface dysfunction is the final common pathway that occurs as a result of an imbalance of ocular surface protective mechanisms. Each protective mechanism has its specific role, be it mechanical and/or physiological. Externally, the eyelids act as a physical barrier for protection of the ocular surface, and with each blink it distributes the tear film. Meanwhile, the corneal and conjunctival epithelia provide the biodefense system against microorganisms and proteolytic enzymes. The tear film is crucial, since it lubricates, protects and nourishes, as well as provides a smooth optical refractive surface. As in an orchestra, each component must function in concert to create a harmonious (and healthy) whole. For example, poor epithelial adherence to the basement membrane can lead to recurrent corneal erosion; lagophthalmos can lead to exposure keratopathy; these can all be exacerbated by the loss of corneal sensation and dry eyes.

When conventional medical therapy fails and/or secondary complications (such as persistent epithelial defect, scarring) occur, surgical intervention needs to be considered. The primary goals are to increase lubrication, to assist healing and epithelial adherence, to remove visually significant opacities and to restore sight with minimal side effects. This chapter reviews the current knowledge on various surgical procedures that assist in the protection and stabilization of the ocular surface in a variety of diseases.

In 1986, McLean and MacRae observed that recurrent corneal erosion (RCE) often followed superficial corneal trauma but was not seen after deep stromal laceration.¹ In light of this, they described the technique of anterior stromal puncture (ASP) for the treatment of RCE, which employed a 20G needle at the slit lamp to make multiple punctures through loose epithelium and Bowman’s layer into the anterior stroma. This created a more secure bonding of the epithelium to the underlying basement membrane (BM), Bowman’s layer, and stroma. This treatment has been shown to be effective, particularly in post-traumatic corneal erosions.

Indication

The therapeutic aim of ASP is to enhance epithelial adhesion to the BM by microscopic, superficial scar formation. In RCE, the cause is thought to be failure of the wounded epithelial cells to adhere to the underlying stroma. This could be secondary to weak hemidesmosomal attachment, reduplication of the BM, the action of metalloproteinases,² and/or disruption of type VII collagen fibrils.³

ASP is a well-accepted treatment for RCE.1,4,5 ASP induces reactive subepithelial fibrosis or production of extracellular matrix proteins, both of which may be responsible for increased adhesion of the epithelium.⁶ Success rates of up to 80% have been reported in recalcitrant RCE.^1,4,5

While phototherapeutic keratectomy (PTK) has also been shown to be effective, there are advantages of ASP over PTK. ASP can be easily performed in the office or outpatient setting with simple, inexpensive equipment and causes minimal discomfort. Moreover, there is a low risk of inducing visually significant scarring and changes in refraction.

Recently, ASP was reported to be used for RCE after laser in situ keratomileusis (LASIK), which helped in resolving secondary diffuse lamellar keratitis.⁴ However, performing ASP immediately after LASIK may cause flap displacement and must be used judiciously.

Procedure

Anterior stromal puncture (ASP) is performed under topical anesthesia. Multiple superficial punctures are placed less than 0.5 mm apart in the affected area with a bent 25G needle attached to a 1- or 3-mL syringe. The needle can be bent at the tip with a needle holder to allow only superficial penetration into the cornea. Fluorescein can be instilled to define the affected area. Treatment is extended at least 1 mm into the normal epithelium bordering the lesion. Following the procedure, a bandage contact lens (BCL) is placed, and antibiotics drops are given until complete re-epithelialization. Usually, when ASP is begun, a much larger area of poorly adherent epithelium becomes visible.

ASP can also be performed with a short-pulsed Nd:YAG laser.⁷ The Nd:YAG laser (1.8–2.2 mJ) is focused at the BM zone after epithelial debridement. Spots are placed in rows approximately 0.20–0.25 mm apart. The advantage of laser over needle puncture is that the laser puncture is more uniform, shallow and translucent. There may also be less corneal scarring, so the procedure can be repeated and can be used closer to the visual axis.⁷

Punctal Occlusion

History

Occlusion of the nasolacrimal outflow tract was first described by Beetham in 1935.⁸ He reported resolution of dry eye in eight eyes with filamentary keratitis with electrocautery or diathermy to occlude the puncta. In 1975, Freeman⁹ proposed the use of punctal plugs to provide a reversible blockade of the nasolacrimal tract at the punctum. Semi-permanent and permanent punctal plugs are now widely used and are available in a range of sizes to ensure a good fit.

Indication

The principle behind punctal occlusion is the increase of the aqueous component of the tear film by blocking tear outflow and retention of both natural and artificial tears on the ocular surface. It has been shown to improve symptoms of lacrimal insufficiency. Tamponade is the most popular method, because surgery is not required and it is reversible.

Patients with severe keratoconjunctivitis sicca with or without underlying systemic collagen vascular disease often require permanent punctal occlusion. Punctal occlusion also plays a beneficial role in aqueous tear deficiencies secondary to ocular surface conditions, such as ocular cicatricial pemphigoid, Stevens–Johnson syndrome, and Sjögren syndrome.10,11 Dry eye secondary to reduced reflex tearing found in neurotrophic keratitis can also be effectively managed with this procedure. Finally, punctal occlusion may be beneficial in patients with dry eyes, due to increased evaporation from an exposed ocular surface. This may occur with lagophthalmos, exophthalmos seen with thyroid conditions, and following blepharoplasties. Other than dry eye syndromes, punctal occlusion has been shown to improve contact lens comfort and also may be helpful in the management of superior limbic keratoconjunctivitis.¹²

Punctal occlusion can dramatically improve the quality of life in many patients with moderate cases of dry eyes and can prevent visual loss in patients with severe cases of dry eyes.

Procedure

Temporary Procedures

Collagen Implants

Collagen implants provide temporary (usually less than 2 weeks) occlusion of the lacrimal drainage system. They are threaded into the canaliculus with jewelers’ forceps. They can sometimes act as a test to see if more permanent methods would be beneficial or cause symptoms, such as epiphora.

Silicone Plugs

In this technique, the eye is anesthetized with a topical anesthetic. At the slit lamp, the size of the punctum is measured with a special gauge, which determines the size of plug to be used. The plug is inserted into the punctum until the domed head lies flush with the punctum. The inserter is then released and withdrawn. The plug can easily be removed with jeweler’s forceps if its placement rubs against the conjunctiva or if epiphora results.

Permanent Procedures

Thermal Methods

Thermal methods involve occlusion of the punctum by shrinking the canalicular walls with argon laser, cautery, or diathermy.13,14 Even with thorough cauterization, the canaliculus may reopen in time. One of the most common techniques to accomplish permanent punctual closure is electrocautery. The eyelid adjacent to the punctum is infiltrated with local anesthetic (LA). A topical anesthetic is instilled in the cul-de-sac. The electrocautery instrument (Hyfrecator, Birtcher Corp., Los Angeles, CA) with a fine-needle tip is threaded into the punctum and along the canaliculus. The instrument is engaged while withdrawing the instrument slowly. The canaliculus is thermally de-epithelialized. Additional cautery may be performed at the punctal opening.

Thermal cauterization has been reported to be efficient in achieving punctal occlusion, with a reported recanalization rate of 26.1%.¹⁰ Recently, a recanalization rate of 1.4% was reported with a high-heat-energy-releasing cautery device.¹⁴ Thermal cauterization is performed similarly to electrocautery. Following LA, the loop tip of a disposable cautery is pinched together with sterile forceps to create a needle-tipped probe. This tip is resterilized at any time by turning the cautery on. If the probe does not insert into the canaliculus, the punctum is dilated with a dilator. The tip is threaded into the punctum and along the canaliculus. The instrument is then engaged while withdrawing the instrument slowly.

Contraindication/Cautions

Punctal occlusion is not indicated for all patients who have symptoms of dry eye. Many mild to moderate cases can be managed with the use of topical lubricating drops. Dry eye symptoms may manifest as a result of improperly fitted contact lenses; in these cases, a proper fitting will usually resolve the symptoms. Many commonly used oral medications with antihistaminic or anticholinergic properties can cause reversible ocular dryness. As such, punctal occlusion is not indicated when ocular dryness is a temporary condition.

Key Surgical Points

Although permanent punctal occlusion can significantly reduce the symptoms of dry eye, epiphora can result with improper patient selection.

Blepharitis and dry eyes frequently coexist and treatment of both is necessary.

Patients with severe chronic dry eye or those who benefit from punctal plugs but experience repeated plug loss should be considered for permanent punctal occlusion.

As occlusion of upper and lower puncta at the same time may increase the risk of epiphora, the procedures should be performed one at a time.

Even with cautery, permanent punctal occlusion may not be achieved and the procedure may need to be repeated.

If granulation around a silicone plug is noted, it should resolve with removal of the punctual plug.

Complications

Punctal plugs can cause ocular discomfort and complications,15,16 such as punctum enlargement, loss or migration or extrusion of the implant, epiphora, pyogenic granuloma, local inflammatory reaction to silicone, dacryocystitis, and canaliculitis.^16,17

Granuloma formation is thought to be caused by local stimulation by the plug. Intracanalicular plugs are associated with higher rate of granulation tissue formation in the lacrimal tract when, compared with other forms of punctal plugs. Cases have been reported in which plug migration inside the lacrimal passage resulted in peripheral inflammations and infections, requiring surgical removal.¹⁸

Epithelial irritation caused by the exposed portion of the plug coming into contact with the cornea and conjunctiva may occur. Removing the plug or placing the plug with one of a different configuration may help.

Phototherapeutic Keratectomy

History

The laser–tissue phenomenon of photoablation was first demonstrated in 1983 by Trokel and Srinivasan,¹⁹ who were working on the ultraviolet 193-nm excimer laser. Its potential role in refractive and therapeutic corneal surgery was quickly recognized. The excimer laser underwent extensive preclinical trials before it was applied to human eyes, and it is now being used for photorefractive keratectomy (PRK), LASIK and PTK.

Excimer laser PTK utilizes 193-nm wavelength ultraviolet light to break the molecular bonds in corneal tissue, in turn ablating the anterior stroma in a highly predictable fashion. It was approved by the FDA in 1995 for the treatment of visually significant anterior corneal pathologies, namely superficial corneal dystrophies, epithelial basement membrane dystrophy (EBMD) with irregular corneal surfaces, corneal scars and opacities.²⁰

Advantages

The main advantages of PTK over mechanical superficial keratectomy (SK) include precise tissue ablation with minimal damage to the surrounding tissues, a smooth residual bed for corneal re-epithelialization, and repeatability of the treatment. Compared to lamellar or penetrating keratoplasty (PKP), it is also less invasive, with more rapid visual recovery. PTK can be performed before or after PKP and does not impair the prognosis of PKP.²¹

Disadvantages

Phototherapeutic keratectomy may ablate normal tissue at a different rate from scar tissue, which may result in an irregular surface. Hence, masking agents are often used to fill in any irregular troughs to achieve a smoother surface. In contrast, a surgical plane between normal and abnormal tissue is more readily identified during mechanical SK, which provides a better surgical dissection.

Indications

Phototherapeutic keratectomy is best utilized for corneal pathologies affecting the epithelium or anterior 10–20% of the corneal stroma. For safety reasons, the residual corneal bed thickness must be greater than 300 µm at the end of the procedure. The indications for PTK can be separated into four broad categories,²⁰ although they often overlap:

superficial opacities, e.g. granular dystrophy (Fig. 36.1), scars from trauma or keratitis, post-PRK haze (Fig. 36.2)

Figure 36.1 Granular dystrophy treated with PTK and PRK. (A) A 46-year-old woman with decreased best-corrected visual acuity (BCVA) of 20/40 (+0.50/−2.75 × 13) from granular dystrophy in the left eye. (B) After transepithelial PTK combined with PRK treatment, at 1 month postoperatively, there was a significant reduction in density of opacities. The BCVA improved to 20/30 with significant reduction in refractive error (−0.50/−0.25 × 141).

Figure 36.2 A 41-year-old man with previous IntraLase™ enabled keratoplasty and post PRK haze in the left eye. (A) Subepithelial and superficial stromal haze centrally, in a honeycomb reticular pattern. This was visually significant (BCVA 20/40−1). (B) Post PTK with MMC. At day 4 the central graft was clear and the epithelium has healed. His BCVA improved to 20/25−2.

elevated lesions or irregularities of the corneal surface, e.g. Salzmann’s nodular degeneration (SND), keratoconus nodules

EBMD, e.g. RCE

Others

painful bullous keratopathy

band keratopathy

post-LASIK complications, e.g. flap striae.

Contraindications

It is important to identify patients who have conditions that predispose them to delayed epithelial healing and who would not be suitable candidates for PTK. They include immunocompromised individuals and patients with anesthetic corneas, severe dry eyes or uncontrolled uveitis.²⁰ Others may require additional procedures prior to PTK, e.g. a patient with lagophthalmos secondary to paralytic ectropion may require ectropion repair first. Any deep corneal pathology requiring removal of more than 20–30% of corneal thickness may be more suitable for anterior lamellar keratoplasty rather than PTK. If there is significant thinning in the treatment zone, PTK may predispose them to ectasia and hence should be avoided.

Preoperative Assessment

All patients should be assessed carefully to determine suitability for PTK. There are four main areas that the surgeon should consider: (1) whether the corneal pathology is amenable to PTK treatment anatomically, (2) the patient’s healing ability after surgery, (3) whether the goal of the PTK treatment is achievable, and (4) tailoring a surgical approach to the pathology. As with all surgeries, it is important to establish realistic expectations with the patient. For example, for a patient with granular dystrophy, the goal of PTK is often not a crystal-clear cornea, but rather improvement in vision so that one could delay or even avoid keratoplasty.

Patient History

Once the diagnosis is established, one should clarify the chief complaints. For example, in a patient with RCE, how frequent and how long are the painful episodes? Were they amenable to medical therapy? This will give some guidance as to whether the symptoms will likely be alleviated with PTK. Past ocular and medical history are important to determine whether the patient has potential corneal healing abnormalities, such as neurotrophic corneas (e.g. herpes simplex or herpes zoster keratitis), previous corneal grafts, exposure keratopathy, collagen vascular conditions (e.g. rheumatoid arthritis) and diabetes mellitus. Certain systemic immunosuppressive agents may also prevent healing after PTK. In addition, those with a history of herpes simplex keratitis (HSK) are at risk of recurrence after PTK.

Examination

A thorough examination should be performed to: (1) confirm the diagnosis, (2) ensure that the symptoms and signs correlate (e.g. that the level of visual loss corresponds to the severity of the corneal opacity), (3) detect other co-morbidities that may be responsible for poor vision (e.g. glaucoma, maculopathy), and (4) evaluate the corneal abnormality in detail.

Using slit lamp biomicroscopy, one should determine the size, depth, location, and density of the corneal abnormality, as well as any corneal thinning. In general, PTK is most suited to patients with superficial stromal opacities without significant irregularity and thinning, and those with small, central, elevated corneal lesions not amenable to SK.²² If the lesion is deep enough that the residual corneal thickness approaches 300 µm, one should exercise caution. In this scenario, either lamellar or penetrating keratoplasty may be the preferred treatment.

Imaging

Recent advancements in anterior segment imaging technology can be useful in the surgical planning of these cases. Subtle surface irregularities can be difficult to detect clinically. Corneal topography can highlight any irregularities of the cornea and document any irregular astigmatism. Pentacam® tomography (OCULUS Optikgerate GmbH, Watzlar, Germany) also provides Scheimflug imaging, which can illustrate the depth of the corneal lesions, although the resolution is limited. High-frequency ultrasound biomicroscopy may be of more use for large lesions. High-resolution anterior segment optical coherence tomography is a quick, non-contact imaging modality that can provide pachymetric mapping of corneal opacities.²³ It could potentially result in more accurate resection of tissues during PTK. Sometimes, accurate depth is difficult to determine with these modalities as they all are subject to posterior shadowing which may overestimate the depth of lesions.

Procedure

For PTK, the surgical technique varies depending on the characteristics of the pathology. This includes the size, shape, location, density and depth of the lesions.

The general principles are:

Maintain a smooth surface if possible, e.g. a patient with granular dystrophy with a smooth epithelium could be treated with transepithelial PTK, in which case the epithelium is used as the masking agent.

In those with loose epithelium (e.g. RCE) or elevated lesions (e.g. SND), mechanical debridement (or with adjunctive 20–50% alcohol for 5–10 seconds) with a blade should be utilized prior to PTK.

Remove the least amount of tissue to achieve the desired results by frequently stopping and checking the results under the microscope or the slit lamp before proceeding with further laser treatments.

Maintain centration of the treatment zone to avoid inducing irregular astigmatism.

Use a controlled amount of modulating or masking agent to achieve a uniform corneal surface (e.g. artificial tears or balanced saline solution).

For superficial opacities, the goal is to clear as much of the opacity centrally as possible while resisting the temptation to ablate deeper tissues (which can result in excessive induced hyperopia).

For RCE, only aim to remove 5–6 µm of the Bowman’s membrane.

For focal corneal scars, one may partially treat any induced refractive error, or use an opaque mask (e.g. use a specifically shaped weckcell sponge) if the treatment will remove tissue in an area where it is not desirable.

In general, the treatment zone should be tailored to the abnormality. For example, granular dystrophy could be treated with a larger treatment zone of around 6–7 mm diameter, while a 2-mm Salzmann’s nodule centrally should be treated with a 2–3-mm spot size. After laser ablation has been completed, topical antibiotics and steroids are applied along with a soft bandage contact lens (BCL). However, some advocate the use of a pressure patch rather than a BCL if there is a history of HSK or in a neurotrophic cornea. The patient should be reviewed frequently until re-epithelialization is complete. For RCE patients, after removal of the BCL, the use of hyperosmotic 5% sodium chloride ointment at bedtime may reduce the recurrence of erosions.

Adjunct Therapy

In recent years, mitomycin C (MMC) has been shown to be effective in reducing the occurrence of corneal haze following excimer laser ablation.²⁴ Corneas requiring PTK are inherently at higher risk of haze, due to the presence of stimulated fibroblasts, surface irregularities (e.g. SND) and potential suboptimal healing ability (e.g. RCE). The ideal concentration and duration of MMC exposure remains controversial. Depending on the surgeon’s preference and the pathology, the dosage may range from 0.001% to 0.04%, with a duration of application between 12 seconds to 2 minutes.²⁴ For high-risk patients, such as PTK post-corneal transplant or repeat treatments, the authors recommend 0.02% with a minimum application of 30 seconds.

Hyperopic shifts are common after PTK, due to central flattening, particularly in those with deeper ablations.²⁵ This could be managed post-PTK with contact lenses or sequential hyperopic PRK treatment. If the patient has pre-existing hyperopia, one could combine PTK and hyperopic PRK treatment at the same time. With deeper ablations, one could perform a simple anti-hyperopic ablation concurrent with the PTK treatment. The authors recommend setting the laser with a standard +1.00 D treatment (ablation zone 9 mm, correction diameter 5 mm) to smooth out the mid periphery and increase the peripheral zone treated and the curvature. Others recommend using the joystick to maneuver a 2 mm diameter circular spot around the periphery of the central ablation (between 5 and 6 mm) to smooth out the edge of the ablation zone, thus counteracting the induced hyperopia.²⁶

Outcomes

The outcome of PTK in RCE is related to the etiology, with traumatic etiology (success rate 74.4–80.0%) being better than EBMD (success rate 53.8%).²⁷ In general, several studies have reported good results in anterior corneal dystrophies of the Bowman’s and stroma.^28,29 However, recurrences are not uncommon, especially in Reis–Buckler and granular dystrophy.²⁸ Fortunately, recurrences are often superficial and amenable to repeat PTK with good results. One must distinguish the difference between early haze from laser ablation (homogenous opacity over ablation zone) and recurrences (more heterogeneous with patterns resembling the underlying dystrophy).²⁸ There are also reports of differential prognosis depending on the genotypes of the corneal dystrophy, with Gly623Arg mutation responding better to PTK.³⁰ Others have shown success in treating lattice dystrophy,²⁸ SND,³¹ keratoconus nodules,³² LASIK flap striae³³ and symptomatic bullous keratopathy.

Key Surgical Points

Careful selection of patients who are good candidates for PTK avoids any disappointments or unwanted complications.

PTK is best for corneal pathologies affecting the epithelium or anterior 10–20% of the stroma.

For patient safety, the residual corneal bed thickness must be greater than 300 µm.

Try to maintain a smooth surface during PTK; either perform a transepithelial PTK if able, or use masking agents.

Remove the least amount of tissue required to achieve the intended goal.

Be aware of hyperopic shift and try and correct it within the same setting if possible.

Complications

In the immediate postoperative period, delayed epithelial healing can present a challenge. The time to re-epithelialization ranges between 2 and 6 days.³⁴ Poor healing can be associated with more severe consequences, such as scarring, irregular astigmatism and infection. Induced hyperopia is the most common refractive change.³⁵ However, induced myopia and irregular astigmatism have also been reported. Combined symptomatic and morphologic recurrence for RCE related to EBMD occurs in 14% of patients.³⁶ Other side effects post-PTK include haze, graft rejection and reactivation of HSK.

Superficial Keratectomy

History

Treatment of RCE with superficial debridement of loose epithelium was reported as early as 1900.³⁷ This, combined with chemical cautery, was advocated by Chandler in 1944, as he recognized that the pathology is one of epithelial adherence.³⁷ However, this resulted in clouding of the cornea. Few decades later, Buxton and Fox reported their series of SK in patients with EBMD in 1983.³⁸ They performed a total superficial epithelial keratectomy from limbus to limbus with a blade and then applied diamond drill to areas of persistent abnormalities. Eleven out of 13 patients were relieved of their symptoms. However, given the potential limbal stem cell damage from the large debridement, others advocate sparing of the peripheral corneal epithelium.³⁹

Indications

Superficial keratectomy involves mechanical excision of the epithelium, Bowman’s membrane and superficial stroma with a sharp blade or a rotating diamond burr. This procedure is inexpensive, requires commonly used instruments and does not tend to induce significant refractive changes. It works well in a selected group of patients with ocular surface diseases not amenable to medical treatment. The advantage of mechanical debridement is that the surgical planes are clearer, potentially creating a smoother surface, especially for superficial elevated lesions. The primary indications include superficial corneal scars, SND and RCE.

Procedure

This is performed under topical or local anesthesia. It can be performed at the slit lamp or under an operating microscope in a minor operating room. For RCE, use a sharp instrument (e.g. a 64 Beaver blade) to debride the loose epithelium in a sheet, combined with gentle wiping with a cellulose sponge. Care should be taken to leave a 1 2-mm rim of peripheral corneal epithelium to avoid stem cell damage. This can be combined with the use of a fine diamond burr to polish in a gentle circular motion in the affected areas, which is shown to reduce recurrence rates with similar efficacy to PTK.40–42 For elevated lesions, such as a Salzmann’s nodule, grasp the nodule with a toothed forceps, e.g. Colibri, and dissect with a crescent or a 64 beaver blade. Scraping with the blade at 90 degrees to the surface of the cornea will often reveal a clear plane of dissection at the level of the Bowman’s membrane, leaving a smooth bed (Fig. 36.3. Some also promote the use of MMC intraoperatively to prevent recurrence and haze.⁴² A soft BCL is then applied and the patient is commenced on topical antibiotics for 1 week and topical steroids on a tapering regimen over 2–3 weeks. Frequent review is required until the epithelium is healed.

Figure 36.3 Superficial keratectomy for a Salzmann’s nodule. (A) Scrape with a blade to reveal a clear plane of dissection (B) before advancing forward. Place a bandage contact lens at the end and leave it until epithelialization occurs.

Key Surgical Points

Care must be taken to leave a peripheral 1–2-mm rim of corneal epithelium to avoid stem cell damage.

When using the diamond burr, apply in gentle small circular motions to avoid inducing irregular astigmatism.

For those with SND, one might consider using MMC to prevent recurrent and haze.

Outcomes

In general, SK with or without diamond burr polishing is thought to be a safe and cost-effective treatment option. In RCE associated with EBMD, diamond burr polishing has been shown to be as efficacious as PTK, and it has a lower recurrence rate of 11.1% compare to 26.7% in the PTK group.⁴² Diamond burr SK is also thought to be more effective, compared to epithelial debridement alone, with less recurrences and the need for repeated surgical interventions.⁴¹ In a retrospective study of patients with SND treated with SK combined with MMC, there were no recurrences and most had improvement in their symptoms.⁴³

Complications

Similar to PTK, early postoperative complications include delayed epithelialization and infection. Care needs to be taken to avoid induced astigmatism after diamond burr SK by using only gentle small circular motions.⁴⁴ Postoperative subepithelial haze was reported in around 23–26% of patients, which can occur at any time after surgery for RCE.^45,46

Tarsorrhaphy

Tarsorrhaphy is closure of the eyelids via an adhesion between the two upper and lower lid margins, which reduces corneal exposure and evaporation of the tear film while minimizing friction between the eyelid and ocular surface during blinking. It is, therefore, an effective procedure for patients requiring prolonged surface protection for persistent corneal epithelial defects, secondary to neurotrophic keratopathy following herpes simplex or herpes zoster disease, exposure keratopathy (e.g. paralytic ectropion in seventh nerve palsy), keratoconjunctivitis sicca and corneal melt. It should be used in conjunction with the appropriate medical therapy.

Tarsorrhaphy can be either temporary or permanent, depending on the expectant recovery phase of the underlying condition. For example, a patient with Bell’s palsy, who is expected to have some recovery of the facial nerve function over the ensuring few weeks, requires a temporary tarsorrhaphy. Meanwhile, a patient with an anesthetic cornea and a large neurotrophic ulcer may require a permanent tarsorrhaphy.

Temporary Suture Tarsorrhaphy

Temporary suture tarsorrhaphy was first described by Koenig and Harris ⁴⁷ in 1991, with various modifications published since then, including without the bolsters⁴⁸ and a drawstring technique.⁴⁹ The most common type is a partial lateral tarsorrhaphy, which permits examination of the ocular surface and preserves the patient’s visual axis. If additional protection is required, one could also add a medial tarsorrhaphy. Often, complete temporary closure may be needed to achieve epithelial healing and should not be withheld.

Procedure

1. Assess the amount of tarsorrhaphy that is required (usually one-third to one-quarter length of the eyelid).

2. Infiltrate the eyelids with 2–3 mL of LA with 1 : 100 000 epinephrine (for hemostasis). Flip over the lid and inject above the superior tarsus and below the inferior tarsus to achieve complete anesthesia.

3. Prepare two 1–1.5-cm silicone or rubber bolsters. These prevent erosion of the sutures through the eyelids and aid eversion of the lids to prevent trichiasis.

4. Use a double-armed 5-0 nylon or 6-0 silk suture (the authors prefer nylon as it is easier to remove later on), pass them through one of the bolsters completely at one end, then through the skin 3–4 mm from the upper eyelid margin, into the tarsus, and exits at the gray line of the upper lid. The needle is then passed through the lower lid gray line and exits 2–3 mm inferior to the lid margin. Repeat the same maneuver with the other end of the needle, but displace it laterally. Tie the sutures externally over the bolsters. Ensure that there are no sutures internally that could cause further irritation of the ocular surface.

5. One modification is the drawstring method ⁴⁹ which allows complete closure but also easy access to the ocular surface. This technique uses three bolsters, two 2 cm and one 1 cm sections. The first step is as described above, but using the two 2-cm bolsters. Then the two suture arms below the lower lid margin are passed through the 1-cm bolster. This is tied, but 2–3 cm of slack is left in the suture to allow the surgeon to loosen the bolsters (move the smaller bolster away from the larger bolster) and examine the ocular surface as required. In the interim, the loop can be taped to the skin with steri-strips.

Permanent Tarsorrhaphy

Permanent tarsorrhaphy involves the removal of the mucocutaneous margins of the upper and lower lids to create an inter-marginal lid adhesion. Although it is more permanent, it is reversible.

Procedure

1. As with temporary tarsorrhaphy, infiltrate the eyelids with local anesthetic with epinephrine.

2. Use a sharp blade (e.g. No.11 stab blade) and split the eyelids down the gray line, approximately 2–3 mm deep. Then make two relieving incisions perpendicular to the gray line incision, directed posteriorly.

3. Use Westcott scissors to excise a strip of the posterior eyelid margin in both the upper and the lower lids.

4. Appose the denuded posterior lid margin using multiple interrupted 5-0 or 6-0 Vicryl mattress sutures through the tarsus and tie the knots anteriorly away from the cornea.

5. Close the anterior lamellar with multiple interrupted 6-0 vicryl or 6-0 double-armed nylon through a bolster (as for temporary tarsorrhaphy) as it everts the lid during healing. The nylon and bolster can be removed after 2 weeks.

Postoperatively, topical antibiotic ointment is applied to the eyelids for 2 weeks. The lid margin will be fused laterally, but this can be reversed at any stage by dividing across the intermarginal adhesion.

Botox Tarsorrhaphy

Botulinum toxin was first used for ocular disorders for the treatment of strabismus. In 1987, Adams and Kirkness ⁵⁰ identified that temporary induced ptosis from injection of botulinum toxin into the levator palpebrae superioris (LPS) could act as a protector of a compromised ocular surface. For patients unwilling or medically unable to undergo a surgical procedure, such as suture tarsorrhaphy, this may be an alternative. It provides a complete ptosis and the ocular surface is easily accessible to the examiner and for instilling drops. However, it has the expense of the neurotoxin, it is mildly invasive, with more obvious appearance asymmetry. It also commits the patient to monovision for 6–9 weeks until the ptosis recovers.

There are several types of botulinum toxin, but the majority of studies were based on Dysport (Ipsen, Slough, United Kingdom) and Type A Botox (Allergan, Irvine, California). These are not bioequivalent; hence, one must be careful when reconstituting the drug and should also be aware of the difference in concentrations. The suggested relative bioequivalence or per-unit strength ratio of Dysport compared to Botox is 3 : 1.⁵¹ The earlier reports described administration of the botulinum toxin with a 25-mm needle, through the eyelid, aiming at the levator palpebrae superioris muscle belly.^50,52,53 The toxin doses were 0.0625 ng for Dysport^50,52 and 2.5–5 units for Botox.⁵³ The mean duration of ptosis ranged between 6.5 and 8.5 weeks. However, this was associated with 24–80% superior rectus underaction,^50,52,53 which could exacerbate exposure keratopathy. A recent report suggested a more anterior chemo-denervation of the LPS, which was associated with no superior rectus dysfunction.⁵⁴ In this approach, the clinician used a tuberculin syringe and a half-inch 26G or 30G needle and placed the needle tip near the anterior orbital roof just behind the superior orbital rim in the mid-pupillary axis before injection. The mean effective dose was 12.5 units and the mean ptosis duration was 9.2 weeks. In general, the time from injection to complete ptosis is 3.6–4 days.⁵³ The patient should be continued on appropriate topical medications as prescribed by the surgeon.

Other Modalities

Other types of tarsorrhaphies that have been described which may be useful in selected patients. Cyanoacrylate adhesion of the eyelids is an easy and effective method of closing the palpebral aperture.⁵⁵ However, it does not permit easy examination of the eye and its duration is unpredictable (range 1–15 days). This may be useful as a short-term solution in those who have contraindications or are unwilling to have other surgical procedures (e.g. on anticoagulants). Mulhern and Rootman described the use of the Stamler temporary lid splints,⁵⁶ which are inexpensive, non-invasive and easy to apply. However, these only last for a short period of time (average 3–7 days) and require frequent re-application. Their role is limited to application in those patients requiring a total therapeutic ptosis for a short period of time (less than 2 weeks) and to those who are able to return, or to be taught how to reapply the splint.

Key Surgical Points

The decision on the type of tarsorrhaphy depends on the diagnosis and the expected recovery time.

For temporary and permanent suture tarsorrhaphy, avoid leaving any internal sutures that may inadvertently rub against the ocular surface.

Using the drawstring technique can make it easier to access and examine the ocular surface.

For botulinum toxin ptosis, one must be mindful of which serotype and product, as they are not all bioequivalent.

Outcomes

In general, temporary and permanent suture tarsorrhaphies are successful in achieving re-epithelialization. A series evaluating the outcome of suture tarsorrhaphies in a cohort of cornea and external eye disease patients showed that 90.9% of the epithelial defects completely resolved. The mean time to healing after tarsorrhaphy was 18 days.⁵⁷ When comparing lateral tarsorrhaphy to patching in persistent post-keratoplasty epithelial defects, the epithelial healing was significantly faster (7.61 veersus 12.6 days) and the patients were more comfortable in the lateral tarsorrhaphy group.

Botox ptosis is effective in inducing complete ptosis in 75% of cases,⁵² although in some cases the patient may require a second injection to achieve the desired result.

Complications

Complications of suture temporary tarsorrhaphy include premature separation of tarsorrhaphy, trichiasis and cellulitis. When reversing permanent tarsorrhaphy, the lid margins can scar and deform with possible secondary cicatricial entropion. Beware of doing a lateral tarsorrhaphy in patients with a sixth nerve palsy as it may exacerbate the exposure. In patients with central melting, it may be better to completely close the lid fissure, as a lateral tarsorrhaphy may exacerbate the exposure and melting. Superior rectus underaction is the most commonly reported side effect (68–80%) following Botox ptosis, leading to hypotropia and reduced Bell’s phenomenon.50,52 This could exacerbate exposure keratopathy. The mean recovery time of superior rectus underaction was reported to be 6 weeks, although 16% persisted beyond resolution of ptosis.⁵² Diplopia occurs in 16–24%, with some persisting beyond the resolution of the ptosis.

Conjunctival Flaps

History

Conjunctival flaps have been used in the treatment of corneal diseases since the 1800s. Described in the German literature in the 1877, Gundersen 58,59 reported the largest series and described his technique for thin conjunctival flaps in 1958. He advocated their use for corneal ulcerations and thinning disorders, such as neuroparalytic keratitis, marginal ulcerations, relapsing erosions, and herpetic ulcers. In the early 1900s this procedure was used to reinforce cataract wounds.⁵⁸

Indication

The treatment of recalcitrant corneal surface disease can be a challenge. Although local and systemic treatment is often successful, there are situations in which medical or surgical therapy may fail, with resultant recurrent epithelial breakdown and stromal ulceration. In specific situations, the conjunctival flap is effective and definitive treatment for persistent ocular surface disease.59–62 Patients are relieved of pain, frequent regimen of topical medication, as well as more invasive surgery or enucleation.

The purpose of a conjunctival flap is to restore the integrity of a chronically compromised corneal surface,63,64 and provide metabolic and mechanical support for corneal healing.^63,65 They do not add tectonic support and should not be used in very thin or perforated corneas. Conjunctival flaps act as biologic patches, conferring a trophic effect because of the nutritional and immunologic supply by its vascular connective tissue. A conjunctival flap will often provide comfort, reduce the ocular inflammation, and promote healing in these patients. This is particularly helpful in the setting of a blind painful eye where the flap can provide a new ocular surface for the placement of a cosmetic scleral shell (Fig. 36.4).

Figure 36.4 A total conjunctival flap covering the ocular surface.

The most common indication for the use of a conjunctival flap is in the management of a persistent sterile corneal ulcer. Quite often, this chronic breach of corneal surface integrity has not responded to extensive lubrication, patching, a bandage contact lens, or a tarsorrhaphy. This may be a result of sensory denervation of the cornea (i.e. neurotrophic keratitis, paralysis of the seventh cranial nerve leading to exposure keratitis, corneal anesthesia following herpes zoster ophthalmicus, or metaherpetic ulceration following chronic HSK) or limbal stem cell deficiency. Ectatic thinning of the cornea near the limbus may also be managed with a conjunctival flap as long as the cornea is not excessively thinned.

A conjunctival flap is rarely used in the management of microbial keratitis. Marginal fungal ulcers unresponsive to antimycotic therapy have been successfully treated with conjunctival flaps.66,67 This treatment is not appropriate in the setting of an active suppurative infection that is in danger of perforation.⁶⁷ A repeat graft is generally preferable once the inflammation associated with graft infections has resolved. In cases where medical treatment has been exhausted, a conjunctival flap may offer a more effective means to resolve the infection.⁶⁸ A repeat graft can then be safely performed at a later date.

In the technique described by Gunderson,58,59 coverage of the entire cornea with the conjunctiva obstructed any view of the anterior chamber^58,69 and precluded monitoring of corneal disease progression. In addition, this procedure was challenging in patients with short fornices and had the potential to cause ptosis.⁵⁸ It also required a more careful and extensive surgical procedure in the operating room. Any buttonholes or traction could ultimately lead to flap failure. In cases where only partial coverage of the cornea is required, a selective partial conjunctival flap tailored to cover the desired part of the diseased cornea provides an alternative to a total conjunctival flap.⁵⁸ This is particularly helpful if the patient has short fonices. Visualization of the anterior segment is not limited with this technique, and therefore, progression of corneal disease can be monitored. Intraocular pressure can also be accurately measured. Following stabilization of the ocular surface with either a total or partial conjunctival flap, PKP may be considered.

Procedure

Types of Conjunctival Flaps

Total Conjunctival Flap

The most commonly employed technique for a total conjunctival flap is described by Gundersen ⁵⁸ or modifications thereof.^60,70 Anesthesia may be local or general, but retrobulbar anesthesia is adequate in mos casest. Before the procedure, it is important to evaluate the availability and mobility of the conjunctiva. Conjunctival scarring in the area to be mobilized may preclude the success of this procedure.

After placement of the lid speculum, a 7-0 silk traction suture is placed through the peripheral cornea at the superior limbus (Fig. 36.5A). This allows the surgeon to control the globe and rotate it downward to expose the entire upper bulbar conjunctiva. Removal of the corneal epithelium can be performed after mobilization of the conjunctival flap. However, if done first, it is completed in a bloodless field and ensures it is not forgotten.

Figure 36.5 (A) Placement of a traction suture at the 12 o’clock limbus to infraduct the eye. This allows greater exposure of the superior bulbar conjunctiva and fornix. (B) 360° peritomy. (C) Subconjunctival injection of anesthetic. (D,E) Blunt dissection of the superior conjunctiva. (F) Placement of conjunctival flap over the debrided corneal surface and individual sutures. (G) Multiple sutures in place. (Reproduced with permission from Krachmer JH, Mannis MJ, Holland EJ. Cornea. 3rd ed. Elsevier Inc; 2011. Figure 145.2.)

Next, the conjunctival flap is mobilized. A caliper is used to measure at least 14 mm from and concentric to the superior limbus into the fornix, the area necessary to cover the cornea. The incision is carried from medial to lateral canthal area superiorly in an arc (Fig. 36.5B). The conjunctiva is then ballooned up with a subconjunctival injection of 1% lidocaine with epinephrine (Fig. 36.5C). Some surgeons prefer not to inject subconjunctivally as it swells the Tenon’s capsule and makes dissection more difficult. The injection site should not reside within the area to be used to cover the cornea. With the globe rotated downward, an incision in the previously marked superior fornix is made, avoiding the underlying Tenon’s capsule (Fig. 36.5D). A very thin flap of conjunctiva is dissected downward towards the limbus without creating buttonholes (Fig. 36.5E). When the limbus is reached, the resultant flap is freed with a 360-degree peritomy. Bleeding in the dissection bed is controlled with cautery.

The traction suture is released and the conjunctival flap is pulled down over the cornea (Fig. 36.5F). Relaxing incisions may help to release tension. Finally, the flap is sutured into place with Vicryl or nylon sutures placed in the superficial sclera just outside the limbus (Fig. 36.5G).

Although most authors agree that the procedure of choice is performing thin conjunctival flaps, without Tenon’s capsule,⁶⁰ there are special situations in which a keratectomy is necessary and where Tenon’s capsule may be useful to occupy the space.⁷¹ Sanitato et al.⁶⁶ reported on a thick conjunctival flap with Tenon’s capsule for the treatment of peripheral corneal mycotic abscesses. The thicker flaps do thin out over time, thereby improving the overall appearance.

Bridge Flap

A bridge flap is used to cover focal central or paracentral corneal defects (Fig. 36.6A). This technique is similar to a Gundersen flap except that the width of the flap to be dissected is measured to be large enough to cover the corneal lesion (typically 20–30% larger). The corneal epithelium is removed in the area to be covered. Following appropriate measurements and marking in the superior bulbar conjunctiva, a thin flap is dissected from periphery to limbus, and released with a limited peritomy (Fig. 36.6B). The flap is then mobilized to cover the corneal lesion and fixed into place with Vicryl or nylon sutures (Fig. 36.6C).