INTRODUCTION

Tarsorrhaphy is the surgical fusion of the upper and lower eyelids to partially close the eye. The main goals of tarsorrhaphy are to protect the cornea and ocular surface and to allow/promote corneal healing (Rajak, Rajak, and Selva 2015).

Goals of tarsorrhaphy

The main goals or indications for tarsorrhaphy are protection of the cornea and ocular surface, and promotion of corneal healing. Other indications include conjunctival exposure after ocular surgery, and retention of a conformer or other ocular devices, for example in children with anophthalmia or adults after evisceration or enucleation (Rajak, Rajak, and Selva 2015).

Tarsorrhaphy may be performed as a standalone procedure, or it can be performed in combination with other strategies to improve corneal and ocular surface health and integrity, including amniotic membrane transplantation (Dua et al. 2018). The next paragraphs will describe the indications of the technique.

Protection of the cornea and ocular surface

1. Inadequate eyelid closure, such as patients with facial nerve palsy.

2. Active cicatricial conjunctivitis.

3. Conjunctival scars and damage to eyelid skin. In patients with keratinized eyelid margins, tarsorrhaphy prevents epithelial damage by the friction caused by eyelid movements (Dua et al. 2018).

4. Corneal hypesthesia or anaesthesia (neurotrophic keratitis, NK) with risk of loss of corneal integrity and infection: this is considered by some authors as the gold standard in the treatment of NK, as closure of the lids protects the cornea from the environment (Dua et al. 2018; Dua and Said 2015).

5. Severe proptosis with risk of exposure keratopathy, as in patients with craniosynostosis syndromes or patients with thyroid eye disease;

6. Scleral show post-blepharoplasty surgery;

7. Poor or infrequent blinking, including patients with severe neurologic diseases and patients in intensive care units.

Promotion of corneal healing

1. Difficult-to-treat, long standing corneal ulceration associated with infectious keratitis;

2. Non-healing epithelial abrasions and defects;

3. Severe dry eye;

4. Post-operatively in cases of failing or exposed corneal transplant, including keratoplasties in which delayed epithelial healing is present or suspected. Some corneal surgeons perform temporary or even permanent tarsorrhaphy at the time of corneal transplantation, particularly in high-risk patients. (Cosar et al. 2001; Tzelikis et al. 2005).

In two large studies of tarsorrhaphy, neurotrophic keratopathy, exposure keratopathy, and post-keratoplasty eyes were the most frequent indications for tarsorrhaphy; other less common but important indications included severe dry eye or cicatricial conjunctivitis, and chemical burns (Cosar et al. 2001; Tzelikis et al. 2005).

Classification

Tarsorrhaphy procedures may be classified in several ways:

1. According to the position of the adhesion of the palpebral fissure:

   • medial,

   • central,

   • lateral,

2. According to the estimated time required to heal the lesions:

   • temporary,

   • permanent,

3. According to the surgical technique used:

   • temporary bolster tarsorrhaphy,

   • permanent tarsorrhaphy,

   • pillar tarsorrhaphy,

   • botulinum toxin temporary tarsorrhaphy,

   • cyanoacrylate temporary tarsorrhaphy,

   • modified lacorrhaphy.

Preoperative Evaluation

An adequate preoperative evaluation before tarsorrhaphy is essential to recognize factors with prognostic significance, as well as to properly estimate the type, location and extent of the procedure and the time estimated for ocular surface healing and protection. The following aspects are important in the preoperative evaluation of a candidate to tarsorrhaphy (Panda, Pushker, and Bageshwar 1999a; Korn 2021):

1. Indication for surgery: paralytic vs non-paralytic indications, post-keratoplasty or high-risk corneal transplantation;

2. Previous medical and surgical history, including ocular and eyelid surgery, trauma, head and neck cancer treatments, neurologic disease, herpetic or zoster eye disease, and thyroid eye disease;

3. Eyelid closure and degree of lagophthalmos or corneal exposure;

4. Presence of Bell’s phenomenon;

5. Blinking and orbicularis muscle strength;

6. Ocular surface examination, including corneal staining;

7. Assessment of corneal sensation, as the presence of corneal anaesthesia places the patient at significant risk of rapid corneal decompensation.

General aspects of tarsorrhaphy – Surgical technique

Some general aspects are common to all tarsorrhaphy techniques:

• Start by determining the position and extent of tarsorrhaphy to be performed, depending on the degree of corneal exposure or compromise;

• Tarsorrhaphy can be performed using local anaesthesia in most cases, for instance using lidocaine 1-2% or bupivacaine 0.5%;

• The eyelids should be disinfected (e.g. using 5% povidone iodine) before the procedure.

Temporary bolster tarsorrhaphy

Temporary tarsorrhaphy is the partial occlusion of the eyelids without entailing intramarginal adhesions. As the name implies, it is a temporary measure, indicated in situations where the desired eyelid closure is under 8 weeks (Rajak, Rajak, and Selva 2015). This may be an alternative to assess the utility of partial eyelid occlusion before performing a permanent tarsorrhaphy in some cases.

The use of bolsters is common in temporary tarsorrhaphy. Bolsters can be made of plastic, fashioned from sterile plastic tubing, or of small cotton-wool balls. They are used to prevent the eyelid skin from tight sutures, and allow better suture tension control to close the eyelids the desired amount. Drawstring temporary bolster tarsorrhaphy is an alternative technique (Rajak, Rajak, and Selva 2015; Kitchens, Kinder, and Oetting 2002). However, bolsterless temporary tarsorrhaphy is also a viable option, with good results (McInnes et al. 2006).

The surgical steps of a temporary bolster tarsorrhaphy are the following (Figure 1):

1. Line up the bolster in the upper eyelid and pass a non-absorbable suture (for example Prolene 5-0) through the bolster,

2. Pass the suture through the upper eyelid skin 3-4 mm above the margin, and through the tarsal plate,

3. Exiting the upper lid at the grey line,

4. Pass the suture through the grey line of the lower eyelid, then into the tarsal plate and out of the skin 2-3 mm below the lower eyelid margin,

5. Pass the suture through the second bolster in the lower eyelid,

6. Turn the needle backwards to engage the inferior bolster first, then the inferior eyelid, then the upper eyelid, and finally the bolster in the upper eyelid,

7. Finally, the suture is tied to complete the tarsorrhaphy.

Figure 1.Schematic representation of a temporary bolster tarsorrhaphy.

Permanent tarsorrhaphy

Permanent tarsorrhaphy involves the intramarginal adhesion of the upper and lower eyelids. It provides horizontal and vertical narrowing, which protects the cornea (Panda, Pushker, and Bageshwar 1999a).

Permanent tarsorrhaphy should be considered in the following cases:

1. When maximal therapy has failed to treat corneal exposure,

2. In cases where prolonged eyelid closure is desired, or if the length of time required is uncertain.

Importantly, even permanent tarsorrhaphy can be reopened or enlarged few weeks after healing. This might be useful in patients requiring additional ocular surgery after tarsorrhaphy. However, opening a tarsorrhaphy prematurely may result in recurrence of corneal epithelial breakdown, especially in cases of persistent corneal anaesthesia (Dua et al. 2018).

The surgical steps of a permanent tarsorrhaphy are the following (Figure 2):

1. Start by determining and marking the location and amount of tarsorrhaphy to be performed. It is important to spare a segment of epithelized eyelid margin at the lateral canthus,

2. After eyelid marking, local anaesthesia, and local disinfection, clamping the operative eyelid can be performed to improve haemostasis,

3. Incise the lower eyelid at the grey line using a #11 or a #15 blade, to separate the anterior and the posterior lamellae for about 2 mm,

4. Further dissect the anterior lamella from the posterior lamella using scissors, to a depth of about 4-5 mm from the tarsal plate,

5. Attention is then directed to the upper lid, where steps 3 and 4 are performed.

6. Excise the mucocutaneous junction from the posterior lamella, along the length of the proposed tarsorrhaphy,

7. Suture the posterior lamellae of both eyelids together, making partial thickness tarsal plate passes using absorbable sutures such as Vicryl 5-0 or 6-0; spatulated needles are preferable in order to minimize damage to the tarsal plate with each pass.

8. The anterior lamellae of both eyelids are then sutured together, using interrupted 7-0 Vicryl sutures.

Figure 2.Schematic representation of a lateral permanent tarsorrhaphy.

Figure 3.Clinical photograph of a lateral permanent tarsorrhaphy in a patient with a persistent epithelial defect post-keratoplasty.

Pillar Tarsorrhaphy

Although less frequently performed, conjunctival pillar tarsorrhaphy may be a particularly useful technique in cases of refractory medial corneal exposure, including cases of impending failure or recurrent PKP grafts (Korn 2021). The conjunctival pillar continually lubricates the corneal graft with blinking and routine eye movements. Main advantages include rapidly reversibility in the clinical setting and superior cosmesis compared with permanent lateral tarsorrhaphy (Korn 2021).

In this technique, a pillar tarso-conjunctival flap from the superior eyelid is sutured through the lower eyelid. The size and location of the pillar can be customized for the degree of corneal exposure. The size and location of the tarso-conjunctival pillar can be customized for the degree of corneal exposure.

The surgical steps of pillar tarsorrhaphy are the following (Korn 2021):

1. Mark the lower eye 4 mm below the lid margin, to correlate with the inferior border of the tarsal plate,

2. Determine the location and size of the conjunctival pillar(s), and mark the width on the upper tarsus. A 3-4 mm wide pillar is usually sufficient, and a 2 mm tall tarsal flap is sufficient to anchor the conjunctival flap inferiorly,

3. Everting the upper eyelid with a Desmarres retractor, perform a full-thickness horizontal tarsal incision using a #15 blade; avoid incision of the underlying orbicularis muscle to prevent bleeding,

4. Perform the vertical tarsal incisions, dissecting the levator aponeurosis free,

5. Raise and rotate the tarsoconjunctival flap inferiorly, and remove the Müller’s muscle fibres adherent to the conjunctiva,

6. Place a large chalazion clamp over the lower eyelid,

7. Perform a full thickness inferior blepharotomy at the marked line using a #15 blade, creating the inferior conjunctival button hole,

8. Open the inferior palpebral conjunctiva medially and laterally,

9. Advance the tarsoconjunctival flap through the inferior blepharotomy incision,

10. Suture the inferior tarsal plate to the cut edge of the lower eyelid retractors, using interrupted absorbable sutures (e.g. 6-0 Vicryl),

11. Skin closure of the lower eyelid.

Botulinum toxin temporary tarsorrhaphy

Botulinum toxin type A (Dysport 0.0625 ng; Botox 2.5-15 U) can be used to produce a temporary superior blepharoptosis. Anterior transcutaneous chemodenervation of the levator palpebrae superioris has shown to be an effective and safe procedure to induce temporary ptosis which aids in corneal healing (Ellis and Daniell 2001; Naik et al. 2008; Kasaee et al. 2010).

Cyanoacrylate glue tarsorrhaphy

Cyanoacrylate glue is a readily available substance that can be used to temporarily occlude the eyelids, which has been termed “superglue tarsorrhaphy” by some authors. The main advantages of this procedure are that it is a simple procedure and it can be done in the office setting and at the patient’s bedside. This procedure is a good alternative for patients with severe epithelial defects and unsuitable for more invasive procedures (Donnenfeld, Perry, and Nelson 1991; Ehrenhaus and D’Arienzo 2003; Moin et al. 2009; Wong et al. 2005).

Modified Lacorrhaphy

A new surgical technique for the treatment of medial paralytic lagophthalmos has been recently described by Matayoshi et al. (Matayoshi et al. 2012).

In modified lacorrhaphy, the lacrimal portion of the eyelid margin is used to create an area of superior and inferior canalicular adhesion, where the anterior canalicular wall is sutured with the orbicularis muscle and skin. This technique does not address the upper eyelid immobility, and so far, there is very scarce data on the postoperative outcomes.

METHODS

Retrospective analysis of primary permanent tarsorrhaphies performed at IMO Barcelona between January 2011 and May 2021. We analysed the electronic medical records (EMRs) of these patients. Data collected and analysed included indications for tarsorrhaphy, systemic and ocular comorbidities, previous ocular surgeries, surgical technique, and outcome of primary tarsorrhaphy. Success of the primary tarsorrhaphy was defined as healing of the ocular surface without the need for additional procedures after tarsorrhaphy; cases requiring additional surgery after tarsorrhaphy or redo tarsorrhaphies were considered as surgical failure.

RESULTS

Forty-five eyes of 41 patients (mean patient age = 59.2 ± 21.6 years; 65% male patients) underwent primary permanent tarsorrhaphy, the main indications were epithelial defects post-keratoplasty (22%), exposure keratopathy (24%), lagophthalmos (13%), active cicatricial conjunctivitis (11%, all of which were under immunosuppressive therapy), and chemical burn (9%). Of the four eyes with a baseline diagnosis of chemical burn, two of them were in the sub-acute phase, having undergone concomitant penetrating keratoplasty (PK) and permanent tarsorrhaphy (one case underwent a lateral tarsorrhaphy, and the other case a medial tarsorrhaphy); the other two eyes were cases where the baseline diagnosis in the EMR was chemical burn, but the patients had already been submitted to previous PK surgery which had delayed epithelial healing. Of note, 25 eyes (55%) had corneal grafts at the time tarsorrhaphy was performed; six of these cases had previous multiple corneal grafts. Twenty-seven percent of cases had undergone some procedure to improve ocular surface health before undergoing tarsorrhaphy, including amniotic membrane transplantation and limbal stem cell transplant. Tarsorrhaphy was performed as a standalone procedure in 25 eyes (55% of cases), 29% of cases underwent tarsorrhaphy at the end of keratoplasty (patients in whom a high risk for post-keratoplasty inadequate healing was suspected), and in 9% of cases amniotic membrane transplantation was performed in the same operative time of tarsorrhaphy.

According to anatomical location of the permanent tarsorrhaphy, 76% of eyes had lateral tarsorrhaphies, 16% medial tarsorrhaphies, 4% central tarsorrhaphies, and 4% had combined medial + lateral tarsorrhaphies.

Eighty-two-point-two percent (82.2%) of eyes were considered successful cases, after primary tarsorrhaphy. The remaining cases - eight eyes of seven patients – had failure of primary tarsorrhaphy. All cases of failed primary tarsorrhaphy had history of previous ocular surgery, and 71.4% had underlying systemic diseases. The surgical success rate after redo tarsorrhaphy was 50% (4 cases), whereas the remaining cases required multiple additional surgeries to restore ocular surface integrity (Table 1). Other than surgical failure, no other postoperative complications were observed.

DISCUSSION

In our study, we have found that permanent tarsorrhaphy is highly effective, in line with previous evidence supporting tarsorrhaphy as one of the safest and most effective procedures in difficult-to-treat corneal lesions, with reported success rates ranging between 83.3% and 100% (Cosar et al. 2001; Tzelikis et al. 2005). In addition, tarsorrhaphy can be combined with other techniques to promote further ocular surface healing and protection. There are some clear benefits of tarsorrhaphy compared with eye patching, which result from having the eye unoccluded (Panda, Pushker, and Bageshwar 1999b). Firstly, it allows visual function. Secondly, it promotes increased oxygenation to corneal epithelium, which has a high oxygen demand for metabolism and regeneration. The benefit of increased ambient oxygenation is three-fold: a) it may promote faster epithelial metabolism and healing; b) the reduced production of lactic acid reduces stromal oedema; and c) prevention of an anaerobic atmosphere for microbial growth. Finally, it allows for examination of the cornea and administration of topical therapy.

Remarkably, however, its high effectiveness contrasts with the delay in deciding for surgery, with mean times between onset of signs and symptoms ranging from 89.8-98.7 days, and in some reported cases as long as 635 days (Cosar et al. 2001; Tzelikis et al. 2005). Time-to-healing after tarsorrhaphy may be prolonged, but usually rapid corneal healing can be expected; reported mean times-to-improvement range from 9.8 to 53.2 days, although it may take up to 90 days in some cases (Cosar et al. 2001; Tzelikis et al. 2005; Dhillon, Bahadur, and Raj 2017). Our study is not without limitations, mainly those inherent to its retrospective design and the limitations of EMRs. Many patients were referred to our institution from other centres, and thus accurate determination of time-to-healing was not possible.

The complication rates following surgical tarsorrhaphy include the following (Cosar et al. 2001; Tzelikis et al. 2005; Panda, Pushker, and Bageshwar 1999a; Korn 2021):

1. Failure to protect ocular surface or corneal epithelization (16.7%),

2. Dehiscence with early tarsorrhaphy aperture (11.1%),

3. Trichiasis,

4. Pyogenic granuloma,

5. Corneal abrasion,

6. Haemorrhage,

7. Infection.

Our failure rate after primary tarsorrhaphy is in agreement with reported rates of failure (Cosar et al. 2001; Tzelikis et al. 2005). Notably, all failure cases in our study had previous ocular surgeries, and a high proportion of these cases had systemic underlying diseases. It is likely that patients with systemic diseases, including diabetes mellitus (due to corneal neuropathic disease and defective corneal epithelial healing), connective tissue disorders and malignancy have a higher risk for failure of tarsorrhaphy (Table 2). The surgical success rate of redo tarsorrhaphy is inferior to that of primary tarsorrhaphy (50% vs 82.2% in our study). Although we observed an acceptable rate of surgical success after redo tarsorrhaphy, in many cases the prognosis is poor, as most eyes will fail redo tarsorrhaphy, and a relatively high proportion of cases from the series of Cosar et al. and Tzelikis et al. required evisceration or enucleation (Cosar et al. 2001; Tzelikis et al. 2005).

Tarsorrhaphy is a very underutilized technique in Ophthalmology, and ought to be considered sooner rather than later in complicated cases. In patients unsuitable for more invasive procedures or before considering permanent tarsorrhaphy, a number of techniques may be considered. In a high number of cases, these temporary techniques may lead to significant corneal/ocular surface healing, sufficient to avoid permanent tarsorrhaphy. Induced ptosis onset usually takes place 4-7 days after botulinum toxin injection, with a mean duration of action (ptosis) lasting for 6.5-12 weeks (Ellis and Daniell 2001; Naik et al. 2008; Kasaee et al. 2010). This procedure may be sufficient to allow adequate healing of the underlying pathologic process that would prevent the need for a surgical tarsorrhaphy in at least 75% of cases (Ellis and Daniell 2001; Naik et al. 2008; Kasaee et al. 2010). Cyanoacrylate temporary tarsorrhaphy has also shown a good efficacy profile, with a high rate of success, as well as low rates of reapplication of glue or need for surgical tarsorrhaphy (Donnenfeld, Perry, and Nelson 1991; Ehrenhaus and D’Arienzo 2003; Moin et al. 2009; Wong et al. 2005).

Lastly, the anatomical, functional, and patient outcomes following permanent tarsorrhaphy rely on establishing the degree of eyelid closure that provides the best balance between the intended protection of the ocular surface, visual field preservation, and cosmetic effects. Patient counselling regarding the therapeutic benefits is essential.

Conflicts of Interest

The authors have no conflicts to disclose.

Funding

No funding was received for this work.