Anti-fibrotic, anti-VEGF or radiotherapy treatments as adjuvants for pterygium excision: a systematic review and network meta-analysis

Pterygium is an uncontrolled fibrovascular tissue overgrowth of the conjunctiva that overlays the sclera and involves the corneal surface [1]. In addition to cosmetic problems, it could restrict ocular motility, impair visual function, and lead to redness and irritation [2, 3]. Surgical removal may be required when patients want to resolve cosmetic problems, relieve discomfort, improve visual acuity, or plan other ophthalmic surgeries (e.g., for cataracts).

Recurrence after pterygium treatment is a major challenge for both surgeons and patients, as recurrences indicate a greater likelihood of additional recurrences and shorter intervals between recurrences [4]. Surgical removal is an effective technique for pterygium treatment. The bare sclera technique (BST) is the oldest and most basic approach for pterygium; however, the greater recurrence rate after BST, which could be up to 80% [5], has encouraged surgeons to search for more efficient surgical approaches and adjuvant therapies. Tissue grafting, such as with the conjunctival autograft (CA), conjunctival flap, or amniotic membrane transplants (AMTs), can significantly reduce recurrences compared to those occurring with BST alone and have been shown to be superior to BST [6‐8]. The combination of the surgical procedure with toxic agents, such as 5-fluorouracil (5-FU), mitomycin C (MMC), and radiotherapy, has also been investigated and has been shown to be effective in preventing recurrences [9‐14]. In conjunction with BST, MMC treatment can significantly reduce recurrence compared to that following the use of a placebo [5, 15, 16]. Other studies have also shown a lower recurrence rate following the use of MMC with CAs, conjunctival flap, and AMT, with acceptable complications [17, 18].

More recently, the efficacy and toxicity of anti-VEGF treatment agents such as bevacizumab (avastin) and ranibizumab (lucentis) have been evaluated in primary and recurrent pterygium. Subconjunctival ranibizumab injection in conjunction with pterygium surgery was reported to be well tolerated [19]. Subconjunctival bevacizumab injection without surgery for primary pterygium was shown to be effective for reducing the pterygium size and improving visual function [20, 21]. However, the efficacy of the administration of bevacizumab combined with surgical excision is more controversial [22‐27]. Castañeda et al. believed that subconjunctival bevacizumab injection could prevent pterygium recurrences in conjunction with surgery [23]. Ozgurhan et al. found that topical bevacizumab therapy was effective to prevent neovascularization and had lower recurrence rate [27]. However, other studies disagreed with it [24‐26]. Treatment efficacy of such antifibrotic or anti-VEGF medications is not currently clear because direct comparisons are rare. Therefore, the goal of this study was to comprehensively review the literature and perform a meta-analysis to determine the relative efficacy of these medications combined with surgical techniques for the treatment of pterygium.

Recurrence after BST, which is associated with more ocular morbidity and the risk of recurrence, has been a major challenge for pterygium treatment. Tissue grafting and adjuvant treatments have been developed to prevent recurrence and are currently commonly used in pterygium excision. Studies have indicated that recurrence was decreased when combining adjuvants with tissues grafting [9‐14]. The use of anti-VEGF medications, such as bevacizumab, with improvement of recurrence, deserves a fresh assessment of published evidence for the adjuvants for pterygium surgery. For pterygium, the RCTs are rare, relatively small, and lacking all of the simultaneous intervention comparisons. Network meta-analysis is thus the optimal tool for evaluating the efficacy of these adjuvant agents in pterygium, considering its advantage in pooling data from both direct and indirect comparisons.

To our knowledge, the present study is the first network meta-analysis in this field. Results from the present study found out that, in both pairwise direct comparisons and Bayesian network meta-analyses, MMC, β-RT, and bevacizumab, but not 5-FU, were significantly more effective than was placebo for reducing recurrence following pterygium excision with or without tissue grafting. MMC had the highest probability of having the most efficacy in reducing recurrence. Although bevacizumab ranked higher than β-RT for efficacy, this difference was not statistically significant. Results from subgroup analyses, including primary pterygium and pterygium excision with BST or CA, were consistent with these data. Data for recurrent pterygium and pterygium excision with AMT were insufficient, and no such analyses were conducted in this study.

Despite the valuable data gathered on this subject, several aspects remain to be considered in future studies. For instance, the use of adjuvants made the surgical procedures more complicated and added economic burden, especially bevacizumab. Analogously, side effects associated with adjuvant administration also challenges the benefit-risk ratio. Such issues might decrease the acceptability of adjuvants. Therefore, the selection of suitable candidates who will benefit the most from adjuvant treatments may be an important issue. Thus, it is important to determine risk factors for recurrence. Some studies reported that age, morphology, and increased inflammation after surgery might contribute to a greater risk of recurrence after surgery [58‐64]. However, recurrence risk assessment tools for routine clinical practice have been neglected in pterygium.

In addition, the optimal dosage, duration, and administration for adjuvants are unclear. The dose of β-RT is 25 Gy with BST [52], which is decreased to 10 Gy when it is combined with CA [51]. The dosage and administration of 5-FU were consistent among trials, but not the duration, which varied from 3 to 5 min. For bevacizumab, the dosage varied from 1.25 mg to 7.5 mg, and the administration approach can be a subconjunctival injection or eye drops. The administration procedure for MMC is more complicated, with different dosages, durations, and approaches. Many studies have addressed this need for standardization, but unfortunately, no routine practice has yet been recommended.

Finally, our findings must be taken in light of several limitations. First, the number of recurrent events in our study was small (472 recurrences), leading to an imprecise point assessment, which might increase the possibility of a Type II statistical error. Second, despite the effort taken to ensure a full search of the literature, missing reports might have affected the calculated estimates of the treatment effects, especially as trials in this field were rare and small in size. Finally, the follow-up duration in some studies included in the present review might not have been sufficient. Reports showed that 50% of recurrences might occur with in 4 months, and 97% might occur within 12 months [62, 63]. Thus, follow-up of at least 1 year is appropriate. The mean follow-up duration in 13 studies was less than 1 year, which might have affected the calculated estimates of the treatment effects. Finally, the recurrence definition varied among studies. According to a 4-point scale [2], termed Tseng’s criteria, a true corneal recurrence is defined when a fibrovascular tissue invades the cornea, termed as Grade 4. The recurrence definition for trials included in the present study reached Grade 4 recurrence but more intensive: any, >1.0 mm and >1.5 mm fibrovascular overgrowth on the cornea. A standardized definition for pterygium recurrence is warranted when RCTs are conducted.

Collectively, with RCT-based evidence, our study results suggest that bevacizumab, MMC, and β-RT are likely to reduce recurrence compared with placebo, and MMC might have the lowest risk of recurrence as an adjuvant agent for pterygium after surgery. However, the efficacy and acceptability of these adjuvants still needs to be determined.