Clinical and surgical outcomes of tectonic corneal patch grafts : insights from a tertiary referral hospital

Abstract

Purpose

This study aimed to report the clinical characteristics, surgical indications, outcomes and long-term complications of full-thickness or lamellar corneal patch grafts for various indications.

Material-method

This retrospective study included 48 eyes of 47 patients who underwent full-thickness or lamellar corneal patch keratoplasty for treatment of corneal perforation, melting and descemetocele. Patient demographics, preoperative features, best-corrected visual acuity before and after surgery, mean follow-up time, corneal graft type, long-term complications, need for additional surgery, and anatomical success were analyzed.

Results

Of the 48 eyes, 21 had corneal perforation, 21 had severe corneal melting, and six had descemetocele. Sjögren’s syndrome (eight eyes, 16.7%), trauma (eight eyes, 16.7%), and limbal stem cell deficiency (eight eyes, 16.7%) were the most common indications for corneal patch grafting. The most common complications after surgery were graft melting (11 eyes, 22.9%) and glaucoma (5 eyes, 10.4%). One eye developed phthisis bulbi. Penetrating keratoplasty was performed in 8 of the 11 eyes with recurrent graft melting. All descemetocele cases were central and corneal melts were more frequently located paracentrally (p = 0.0001). Anatomical success was achieved in 39 eyes (81.2%).

Conclusion

Corneal patch grafting is a safe and effective method for maintaining ocular integrity in corneal melting and perforation that can not be closed using conventional methods. With a success rate of over 80%, corneal patch grafts can delay a larger full-thickness graft, especially in eyes with acute inflammation, until systemic therapy is effective.

Introduction

Corneal perforation, stromal melting, and descemetocele are ophthalmic emergencies that threaten ocular integrity [1]. In addition to preserving the visual function, proper attempts are needed to prevent the development of secondary glaucoma and endophthalmitis [2]. Rheumatological diseases, trauma, persistent epithelial defects, and infectious keratitis are among the most common causes of corneal stromal melting [2, 3]. Medical and surgical interventions may vary depending on the lesion size and location. While small corneal perforations and melts can be treated with bandage contact lenses, tissue adhesives, and amniotic membranes, larger defects may require tectonic corneal transplants [3,4,5,6,7]. Corneal patch grafts are essential for ensuring ocular integrity, particularly in eyes with acute inflammation. In addition, corneal patch grafts provide a critical window of opportunity for systemic immunosuppressive and anticollagenolytic treatment to take effect before more definitive penetrating surgery is performed [8].

This study aimed to present the clinical characteristics, surgical results, and long-term complications of full-thickness and lamellar corneal patch keratoplasty for different indications.

Material-method

This retrospective study evaluated 48 eyes of 47 patients who underwent full-thickness or lamellar corneal patch grafting between January 2020 and November 2023. The study was conducted in accordance with the Declaration of Helsinki and approved by the University’s Ethics Committee (IRB number: 2-24-63). Tectonic corneal patch grafting was performed in the eyes with corneal perforation, severe corneal stromal melting, and descemetocele. Cases of perforation in which bandage contact lenses, tissue adhesives, or amniotic membranes failed to restore the integrity were also included. Defects involving the central 4 mm of the cornea were classified as central and paracentral if they were located between the central 4 mm and the limbus. Demographics, medical history, mean follow-up, surgical indications, preoperative and postoperative best-corrected visual acuity (BCVA), graft type (full-thickness or lamellar), long-term complications, need for additional surgery, and anatomical success were analyzed. Anatomical success was defined as grafts that did not require additional intervention to ensure global integrity after primary corneal patch grafting until the last follow-up.

Surgical technique

Corneal patch keratoplasty was performed under general anesthesia in all patients. In cases where corneal perforations exceeded 3 mm in diameter and infectious keratitis was identified, a full-thickness graft was prepared. On the other hand, in cases of severe corneal stromal melting and descemetocele, and in cases where the perforation area was less than 3 mm, a lamellar corneal stromal graft was used. A side port was created and the anterior chamber was reformed with viscoelastics in the presence of ocular hypotony. After removal of necrotic tissue in the area of the corneal defect, the dimensions of the perforated/thinned corneal area were measured, and the edges were deepened with a 45 °knife. If infectious keratitis was present, the area to be dissected was calculated to be approximately 0.5 mm larger than the area of infectious keratitis, and the recipient bed was created accordingly. Partial or full-thickness trephination was then initiated. The host cornea was dissected freehand by using a crescent knife. The donor corneal button was prepared 0.25 mm larger than the host cornea using an appropriate trephine with a diameter between 3.00 and 6.00 mm. The donor corneal stroma was thinned with a crescent knife when a partial-thickness graft was used. Circular or crescent-shaped corneal grafts were prepared according to the defect in the host tissue. The graft was secured to the site of the corneal defect using interrupted 10 − 0 monofilament sutures. After ensuring a watertight globe, amniotic membrane transplantation was performed at the end of each procedure (Figs. 1 and 2).

Fig. 1

Central corneal perforation in rheumatoid arthritis

Fig. 2

Postoperative image of corneal patch graft with sutures

Postoperative treatment and follow-up

The postoperative treatment regime is summarized in Table 1. In cases of microbial keratitis, appropriate microbiological testing was performed on the sample taken from the infiltrate, and appropriate medical treatment was initiated. The topical steroid doses were adjusted according to the causative agents. In cases of herpetic etiology, antiviral therapy (acyclovir 3 × 800 mg orally) was initiated. Corneal perforation and stromal melting in immune-mediated diseases were evaluated by a rheumatologist, and systemic immunosuppressants were initiated in the eligible patients.

Table 1 Postoperative treatment regime

In the postoperative period, patients were followed up weekly for the first month and monthly thereafter. Suture removal was indicated in cases of loose, infected or vascularized sutures.

Statistical analysis

Continuous data were evaluated for normality of distribution and are described as mean ± SD for parametric data. The suitability of continuous data for a normal distribution was assessed using the Shapiro-Wilk test. Categorical data are expressed as proportions(%). The Chi-square test was used for cross-tabulation of nominal variables for group comparisons. The IBM SPSS version 20 (Chicago, IL, USA) program was employed for analyses, with p < 0.05 accepted as the limit for statistical significance.

Results

The mean age of the patients was 59.18 ± 19.69 years. The demographic and preoperative characteristics of the patients are shown in Table 2. Sjögren’s syndrome (eight eyes, 16.7%), trauma (eight eyes, 16.7%), and limbal stem cell deficiency (LSCD) (eight eyes, 16.7%) were the most common indications for corneal patch grafts (Fig. 3). We performed amniotic membrane transplantation in 19 eyes, and tissue adhesives were used in 2 eyes prior to patch keratoplasty. Nine eyes underwent full-thickness transplantation and 39 required lamellar corneal patch grafts. There was no significant difference in the BCVA before and after surgery. During the follow-up period, the most common complication was graft melting (11 eyes, 22.9%), followed by glaucoma (5 eyes, 10.4%), cataract (4 eyes, 8.3%), and keratitis (4 eyes, 8.3%). One eye developed phthisis bulbi (Fig. 4). Penetrating keratoplasty was performed in 8 of the 11 eyes with recurrent graft melting, and Ahmed glaucoma valve implantation was performed in 3 of the 5 eyes that developed glaucoma. Four eyes underwent cataract extraction (2 phacoemulsification and 2 pars plana lensectomy), and 2 eyes underwent pars plana vitrectomy during the follow-up period. While 23 eyes had central corneal pathology and 25 eyes had paracentral corneal defects, all descemetocele cases were central, and corneal melts were more frequently located paracentrally (p = 0.0001) (Table 3). Anatomical success was achieved in 39 eyes (81.2%). An example of an anterior segment optic coherence tomography image of a patient with corneal perforation at the 6th postoperative month is shown in Fig. 5.

Table 2 Demographics and preoperative features of the patients

Fig. 3

Indications for corneal patch graft surgery (number of patients)

Fig. 4

Complications in the long-term follow-up (number of patients)

Table 3 A comparison of the corneal defect localization with surgical indications and outcomes

Fig. 5

Anterior segment optic coherence tomography image of a corneal perforation case after corneal patch grafting

Discussion

A step-wise approach is often preferred in cases of corneal perforation, stromal melting, and descemetocele. Bandage contact lenses, tissue adhesives (cyanoacrylate or fibrin glue), and amniotic membrane transplantation are the first-line treatments for patients with corneal defects less than 2 mm [3, 9]. However, if corneal melting/perforation can not be sealed with these methods, or if the defect is larger than 2 mm, tectonic corneal grafts should be considered [3,4,5,6,7,8,9]. Loya-Garcia reported that 35 of 49 eyes (71.4%) in which tissue adhesives were the initial treatment required additional surgery to achieve anatomical success [10]. In our study, cyanoacrylate application failed in two eyes. Amniotic membrane transplantation was unsuccessful in 19 eyes to repair corneal melting before corneal patch grafting was required.

LSCD, trauma, and Sjögren’s syndrome were the primary indications in almost half of the 48 patients with tectonic corneal patch grafts. While two patients presented with corneal perforation and were diagnosed with Sjögren’s syndrome after systemic work-up, the dosage of immunosuppressive therapy was increased in six eyes with pre-existing Sjögren’s syndrome.

Patients with ocular cicatricial pemphigoid (OCP) and Stevens–Johnson syndrome (SJS) were included in the LSCD group in our study. Resistant inflammation in OCP, SJS and Sjögren’s disease can lead to non-healing epithelial defects, corneal stromal thinning, melting, and eventual perforation [3,4,5,6,7,8,9,10,11]. Steroids and immunosuppressive agents are required to slow this collagenolytic process [12]. Corneal melting may still occur with this treatment. In such cases, corneal patch grafting is an excellent intervention to overcome this period of active inflammation and ensure the viability of large-diameter grafts that may be required in the future. Corneal patch grafts are useful in removing necrotic stroma and epithelium, which are sources of pro-inflammatory enzymes, including collagenase [13].

Another point to evaluate the patch graft indications in our study was the 6 descemetocele cases (12.5%). A descemetocele can be defined as the protrusion of the descemet membrane due to advanced stromal thinning, and infectious keratitis has been reported as the main cause [14]. However, in our study, 3 out of 6 cases were post-traumatic, and all 6 cases were centrally located (p < 0.0001). None of the descemetocele cases required additional intervention for additional intervention for tectonic support.

We performed nine full-thickness and 39 lamellar corneal grafts in our study. After the endothelium of the full-thickness grafts was reserved for Descemet Membrane Endothelial Keratoplasty, the remaining stroma was used for lamellar patch grafting. In this manner, two corneal defects were repaired from one donor cornea. It is also noteworthy that two eyes had corneal perforation during pterygium surgery and were referred to our clinic. Both eyes required full-thickness transplantation.

Several studies have reported that microbial keratitis is the most common cause of patch grafting [6, 7, 15]. Infectious keratitis was the fourth indication in our study, after Sjögren’s syndrome, trauma, and LSCD. Of the five patients with infectious keratitis, only one eye had recurrent fungal keratitis (Candida Albicans) after patch grafting.

Soong et al. reported that 19% of patients who underwent corneal patch surgery experienced graft melting, and this rate was 12.2% in another study by Vanathi et al. [5, 6]. In our study, graft melting was the main complication (22.9%). Among these 11 eyes, tectonic penetrating keratoplasty was performed in 8 eyes. In 3 eyes, graft melting was controlled by medical therapy and amniotic membrane transplantation. One patient who did not return for regular follow-up after surgery developed phthisis bulbi in the long term. Although inflammatory mediators are removed after surgery, graft melting may recur if the underlying disease is not adequately controlled systemically. This ongoing inflammation is also considered as the underlying cause of cataracts and glaucoma that could develop in the post-operative period [5, 6]. Four eyes developed keratitis in the postoperative period after patch grafting. Three of these cases represented fungal keratitis starting from the edges. In one instance, herpetic recurrence was observed on the graft.

In addition to steroids and lubricant therapy used in the postoperative period after patch grafting, doxycycline plays an active role in suppressing pro-inflammatory mediators including matrix metalloproteinases [16]. We routinely use doxycycline after corneal patch grafting to take advantage of its anti-collagenolytic effect. In our practice, doxycycline is administered at a dosage of 100 mg twice a day for four weeks following patch grafting (a treatment duration of ten weeks for ocular rosacea).

In this study, we achieved tectonic integrity, defined as anatomical success, in 39 eyes (81.2%). Sharma et al. [15] reported a 63.2% success rate with penetrating and lamellar patch grafting, and Calli et al. found that 22 of 23 eyes achieved ocular integrity with lamellar grafts [17].

The main limitation of this study was its retrospective nature. However, to the best of our knowledge, this is also the largest corneal patch graft series from Türkiye, including rare indications, such as post-pterygium corneal perforation and ocular rosacea.

In conclusion, corneal patch grafting is a safe and effective method for ensuring ocular integrity in corneal melting and perforation that can not be closed with tissue adhesives or amniotic membranes. With a success rate of > 80%, we believe that corneal patch grafting is an excellent surgical procedure that can delay a larger full-thickness graft, especially in acutely inflamed eyes with corneal melting.