Introduction
Corneal epithelial defect which fails to heal within fourteen days of corneal injury, even with conservative medical management, is known as Persistent Epithelial Defect (PED).1 It is a clinical condition which is infrequently encountered in ophthalmology. Usually it responds to conservative management, but sometimes it doesn’t. Then the other modalities come into play. It has been observed that tarsorrhaphy is now being forgotten as compared to bandage contact lens, amniotic membrane transplantation and other emerging modalities. Tarsorrhaphy is the most economical procedure with easy surgical technique. In this study, we have tried to evaluate role of tarsorrhaphy in healing of PED as first line of modality.
Materials and Methods
It was a prospective interventional study conducted at Regional Institute of Ophthalmology, Prayagraj. Twenty one patients were selected over a period of 24 months (January 2018 to December 2019) by convenient sampling method.
Inclusion criteria
Patients with PED, who failed to respond to conservative treatment in the study period were included in this study.The various etiologies of PED presenting to our hospital during this study period were - exposure keratopathy, post penetrating keratoplasty, neurotrophic ulcers, post-infection, surgeries other than penetrating keratoplasty, chemical injuries and trauma. All these patients were advised standard treatment in the form of lubricants and patching. Patients refractory to this standard therapy were taken up for tarsorrhaphy.
Exclusion criteria
Patients not fulfilling inclusion criteria were excluded from this study. Other causes of exclusion were – acute infections, adnexal pathology, raised digital IOP and patients on long term systemic medications except for diabetes. Permission was taken from Institutional ethical committee. Written informed consent was taken from patients. Demographic parameters were recorded. A detailed history of ocular symptoms and treatment received was taken. Pain, photophobia, lacrimation and foreign body sensation was asked.
Anterior segment was evaluated with slit lamp. Half of length and breadth of epithelial defect was multiplied to the area of defect. IOP was recorded with non-contact tonometer. It was within normal range in all patients.
Table 1
| S.No. | Symptoms | Number of patients | Percentage |
| 1 | Photophobia | 18 | 85.71 |
| 2 | Lacrimation | 15 | 71.43 |
| 3 | Pain | 12 | 57.14 |
| 4 | Foreign body sensation | 10 | 47.62 |
Tarsorrhaphy was performed in eligible candidates. Type of tarsorrhaphy was permanent and lateral in all patients. For performing the procedure 2 – local anesthesia was injected in lateral 1/3rd of both eyelids. Two incisions were given on the lid margin – 1st on the gray line and 2nd incision was placed 0.5 mm behind the 1st incision and the part of post lamella was excised. Now the respective lamellas were sutured together, with 5-0 non-absorbable sutures using bolsters. Bolsters and sutures were kept in place for 10 to 15 days. Upon their removal all tarsorrhaphies were found to be successful. Conservative management in the form of drops were continued.
Observation
The present study included 21 eyes of 21 patients over the study period. In the present study, following causes were found to be responsible for PED –
Table 2
Neurotrophic and exposure keratopathy were the commonest cause for PED. Four penetrating keratoplasty patients had PED in the graft area which failed to heal in two weeks' time. Two patients of PED had surgeries other than PK – i.e. one developed it 3 days after cataract surgery whereas other patient developed it immediately after VR surgery. Chemical injuries were responsible for PED in two patients of chemical injuries. Both of them had grade III injuries, one due to acid injury and another due to alkali injury.
Table 3
| Mean age in years(±SD) | 59.3±16.4 |
| Gender M:F | 9:12 (42.85:57.15%) |
| Residence Rural: Urban | 15:6 (71.43:28.57%) |
| Diabetics | 2 (9.52%) |
| Smoker | 1 (4.76%) |
Mean age of these patients were 59.3±16.4 years. Male: female ratio in the study group was 9:12. Seventy-one% of the patients were from the rural background. In our study population, two patients (9.52%) were diabetics and one patient (4.76%) patient was smoker. All these patients underwent permanent lateral tarsorrhaphy.
Epithelial defect was recorded preoperatively, at 1 week, 2 week and at 4 week. It was as follows:
Table 4
| Preoperative | 28.99 ± 16.22 mm2 |
| 1 week | 19.62 ± 11.57 mm2 |
| 2 week | 7.85 ± 4.29 mm2 |
| 4 week | 3.41 ± 3.64 mm2 |
Three patients failed to heal even at fourth week after tarsorrhaphy. Two of them had chemical injuries while the third patient had neurotrophic keratopathy. Add on treatment in the form of cryopreserved amniotic membrane was applied on the cornea following which all three PED resolved in 10.2 ± 4.9 days.
Time of removal of tarsorrhaphy was decided by the aetiology of PED. In four patients of exposure keratopathy, tarsorrhaphy was kept for >1 year; and in two patients of chemical injuries, it was removed after 6 months. In rest twelve patients, tarsorrhaphy was removed 4 weeks after healing of PED.
Discussion
PED is sometimes an innocuous entity, healing by medications but sometimes it fails to heal by medications alone, then other interventions come into play. Tarsorrhaphy has always been advised in such situations. But it is a common observation that tarsorrhaphy is now being forgotten. Hence the role of tarsorrhaphy alone in patients with PED in our cornea clinic was evaluated.
In our study, most common cause of PED was neurotrophic keratopathy in five patients (23.81%), followed by exposure keratopathy and Post penetrating keratoplasty patients of four each (19.05%). Chemical injuries causing PED was seen in three patients(14.29%).Two eyes had PED due to surgeries other than PK – one had VR surgery, while other had cataract surgery in a controlled diabetic patient. Trauma was the etiology in only one patient (4.76%) while two patients (9.52%) cause could not be determined. Previous studies by Prabhasawat et al 3 and Cosar et al 4 have shown that the predominant etiology of PED was neurotrophic keratitis. Hamza 5 et al reported ocular surface disorders as predominant cause. Blanco et al 6 studied PED associated with stromal thinning or corneal ulcers. PED seen after penetrating keratoplasty is a specific entity, which was studied by Fu et al, 7 Raj et al 8 and Seitz et al. 9
The mean area of epithelial defect preoperatively was 28.99±16.22 mm² in our study. Area of elliptical defect was calculated in the method as described below.
Largest dimension was measured and then dimension perpendicular to it was measured. They are known as major and minor axis. They were halved to obtain semi-major and semi-minor axis. Suppose they are a and b. Now the area of elliptical defect is calculated as πab in mm². The mean area of epithelial defect was reduced to 19.62 ± 11.57 mm² at week one, 7.85 ± 4.29 mm² at week two. At fourth week, the mean area of defect was 3.41 ± 1.64 mm², due to three patients of PED which failed to heal even at fourth week.
Mean area of epithelial defect was 30.74 ± 27.34 mm² in study by Dhillon et al, 10 which was almost similar to the mean area of PED in our study. Whereas mean area (±SD) of PED in a study by Seitz et al 9 ranged from 13.86 ± 4.37 in erosions to 10.5±3.74 mm² in ulcers. It was quite smaller in comparison to our study. Prabhasawat et al 3 measured the largest dimension and not the total area. It was 3.8 mm on average, hence quite smaller PED were studied. In our study, the area of PED gradually reduced from preoperative values to each postoperative visit, and finally healed in eighteen patients. In remaining three patients though the size gradually decreased in size but the PED still remained at 4 weeks. Hence an additional procedure, in the form of amniotic membrane transplantation was performed in all three cases. Following the procedure these PED’s healed in 10.2 ± 4.9 days. Cryopreserved amniotic membrane not only acts as a scaffold but also provides various growth factors which contribute in healing of PED.8
Tarsorrhaphy partially closes the eyelids, so that interpalpebral area and hence the exposure of cornea to external environment decreases. It avoids drying of cornea. After tarsorrhaphy, the eye drops remain in cul-de-sac for longer time and their duration of contact with cornea increases, increasing the absorption of drugs. Most importantly, blinking is reduced in amplitude hence it’s windshield wiper like effect on regenerating corneal epithelium is avoided. All these factors together lead to healing of PED.8
Other treatment modalities have some issues associated with them. Bandage contact lens is one of the treatment modalities, but it is well known fact that bandage contact lens per se predisposes to infection beside causing foreign body sensation. Added cost is another factor. Amniotic membrane is beneficial in the view that it provides a basement membrane where proliferating epithelial cells migrate and get firmly adhered. It also provides various growth factors which help in healing of PED. But this benefit is seen only with cryopreserved amniotic membrane, not with dried amniotic membrane which is commercially available. Cryopreserved amniotic membrane is limited in availability, reducing its practical usage.
Conclusion
Tarsorrhaphy is quite useful modality for healing of PED. Though many other methods have come in vogue, due to which tarsorrhaphy is quite forgotten nowadays. But still, it is superior to other modalities, in terms of ease of affordability and surgical technique. Eighty five percent of our patients responded to this therapy alone, hence it should be considered as the first intervention in patients with PED.
Article DOI
