Pterygium, a prevalent ophthalmic Cinderella disease, has consequences for the eye, patient, and community that are often underestimated. 1 Sometimes trivialized as a minor ailment, advances in our understanding of the basic pathophysiology of this condition as well as refinement of surgical techniques have resulted in improved outcomes. 2,3,4 Central to this pterygium surgery renaissance has been the use of a reconstructive rather than a destructive surgical approach through the use of conjunctival (CAG) and limbal-conjunctival autografting (LCAG). 2 This chapter will trace the history of the evolution of these procedures and examine the reasons for their success. Further refinement of these techniques will also be explored.
7.2 Background
The mission in pterygium surgery is to restore ocular surface anatomy, function, and cosmesis both in the short and long term and in so doing, alleviate all associated symptoms, painlessly and cost effectively. The plethora of surgical techniques utilized 2 to achieve these ends is evidence of both lack of deep understanding of the pathophysiology of this disease 2,3,4,5 and until recent times, lack of evidence of superiority of a particular intervention.
Pterygium surgery has evolved through a number of eras with the possibility of
Minimal recurrence.
Ocular surface rehabilitation.
Excellent cosmesis.
Minimal refractive error.
This has contributed to a shift toward earlier intervention with an aim of keeping induced corneal astigmatism to less than 0.5D, a level at which there is minimal effect on visual acuity and later in life, 6 to allow the maximal benefit of high quality intraocular lenses. This may be of particular importance since there is an association between the presence of pterygium and cataract. 7
Pterygium impacts quality of life, affecting the psyche (redness and dryness), blunting sight via several mechanisms and can result in blindness via transcorneal extension. 2 It can threaten the eye (largely as a consequence of adjunctive measures, such as mitomycin C [MMC]) with subsequent infection and can threaten life if a diagnosis of malignancy, such as mucoepidermoid or squamous spindle cell carcinoma, is missed. 8,9
CAG/LCAGs address these issues. Thin, Tenon’s free, adequately sized, tension-free grafts tend not to retract and are associated with good outcomes. 2,10 They allow wide pterygium excision so that chronically inflamed pterygium tissue (a source of inflammatory mediators) is removed, reducing the risk of a “pseudo” dry eye state seen in pterygium patients. 11 Wide excision also results in a better cosmetic result 2,12,13 and also reduces the risk of leaving behind ocular surface squamous neoplasia (OSSN) that can occur in ~ 10 to 30% of pterygia 14,15 and atypical epithelial and melanocytic lesions in 12% of pterygia. 14,16
Very low recurrence rates (<1%) have been reported for this approach despite varying excision techniques. 2,17,18 Recently a Cochrane meta-analysis demonstrated that conjunctival autograft is associated with a lower risk of recurrence at 6 months after surgery than amniotic membrane transplantation. 19 Furthermore patients with recurrent pterygia had a lower risk of recurrence when they receive conjunctival autograft surgery compared with amniotic membrane transplantation. Thus, such techniques are considered the gold standard surgical procedure for pterygium.
Management of Tenon’s capsule remains the subject of debate. 2 The notion that Tenon’s capsule gives rise to recurrences has resulted in two approaches, both utilizing wide excision of the pterygium and scar tissue if present:
Extensive tenonectomy as advocated by Hirst. 12,20 This large, single surgeon, prospective study was associated with a very low recurrence rate of 0.4%.
Medial fornix reconstruction to “seal the gap,” creating a barrier between Tenon’s capsule and the ocular surface. 21
An important aspect of both of these techniques is the adequate covering of underlying tissue by epithelial or membranous structures. This raises the possibility that the wound healing response is being modulated, in part by epithelial—stromal interactions 22 and perhaps, with adequate graft cover; such extensive excisions are not necessary, particularly if the inflammatory response is controlled. 2,23 Thus in a prospective, randomized controlled study of 36 eyes (34 patients) with primary pterygium treated with the bare sclera technique (a surgical “worst case” scenario), use of topical cyclosporine reduced the recurrence rate by ~ 50% in the control group versus 22 to 44% in the cyclosporine group. 24
Intensive postoperative anti-inflammatory treatment was used in the extensive tenonectomy studies. 12,20
7.3 First Principles
The history of conjunctival grafts is appropriately linked to skin grafting both in historical and pathophysiological senses. Skin harvesting and transplantation, described approximately 3,000 years ago with the Hindu Tilemaker Caste (Koomá), in which skin grafting was used to reconstruct noses that were amputated as a means of judicial punishment, has had a central role in reconstructive procedures. 25,26 The development of split-thickness skin grafts by Ollier in 1872 27 and Thiersch in 1886 28 utilizing knowledge of cellular proliferation to expedite the healing process (and to minimize scarring) had important implications for pterygium surgeons. Scarring is a consequence of failure of the wound to properly transition from the regenerative phase to a resolving phase with aberrant repair 29,30—healing by secondary intention. The resultant uncontrolled inflammatory mediator and growth factor production, deficient generation of anti-inflammatory macrophages, or failed communication between macrophages and epithelial cells, endothelial cells, fibroblasts, and stem or tissue progenitor cells contribute to a state of persistent injury, which may result in fibrosis. 31 Fibrosis and chronic inflammation are deleterious to the health of the ocular surface, and tactics to minimize these phenomena are central to successful pterygium surgery.
7.3.1 Epithelial Mesenchymal Transition
Epithelial mesenchymal transition (EMT), a process where epithelial cells take on characteristics of mesenchymal cells, has been described in both acute and fibrotic cutaneous wound healing of human skin 32 as well as in pterygium. 3,33,34,35 Studies have shown pterygium epithelial cells concurrently expressed epithelial and mesenchymal markers and signaling molecules associated with EMT. 33,34 We have speculated that pterygium fibroblasts may have originated from limbal epithelium via EMT under the influences of transforming growth factor-β (TGF-β) and fibroblast growth factor (FGF)-2 or ultraviolet (UV) exposure. 3 Alternatively, it has been hypothesized that pterygium fibroblasts are recruited from myofibroblasts in the periorbital fibroadipose tissue. 35 This may explain in part why extensive tenonectomy appears to be effective.
Thus a graft of skin or conjunctiva has the potential to reduce the risk of scarring and chronic inflammation via several mechanisms, and this was likely evident to pterygium surgeons early, as surgery for pterygium was being developed.
Traditionally, the standard surgical approach to diseased tissue has been excision, often associated with other destructive/adjunctive therapies. This approach has been reflected in the evolution of pterygium surgery, and at first sight, it would appear that it has only been in recent times that a surgically reconstructive approach has resulted in improved outcomes. Given the well-known restrictive effect of the scarring element of advanced pterygium, with consequent effects on ocular rotation and resultant diplopia, it is perhaps surprising that the widespread adoption of conjunctival autografting had not occurred earlier. In his 1920 book, Tropical Ophthalmology, Elliot noted that even primary pterygia could tether the eye medially causing significant binocular diplopia. 36 Yet for many years, the bare sclera technique remained standard practice, despite the eventual documentation of unacceptably high recurrence rates. In 1948, D’Ombrain 37 in describing his technique stated, “A recurrent pterygium is a major ophthalmic problem, often necessitating the use of skin or mucous membrane grafts” and for second recurrences, referred to “the free graft method” described by North. 38 He reasoned that, “in such cases, the sub-conjunctival adhesions are of such density and toughness that tissue planes are lost, and there is danger of actually perforating the sclera if dissection or even separation of the pterygium tissues from the sclera is attempted.” North 38 described the use of a limbal skin graft as previously described by Hotz 39,40,41 or a mucous membrane graft as described by Gifford. 42,43 In his experience, the mucous membrane graft was inferior to the skin graft as it encroached onto the cornea and became red and vascularized. He also noted that the skin graft could also grow onto the cornea, forming an “ugly pearly white patch,” but this could be safely excised after 6 months.
There is, however, an earlier history of graft repair post pterygium surgery. As early as 1876, Klein 44 described the use of free, full thickness mucous membrane grafts in cases of recurrent pterygium excision where it is impossible to leave sufficient bulbar conjunctiva to cover the defect. Pollet described a similar technique in 1906. 45 According to Forbes et al, 46 this technique was used for recurrent pterygium but subsequently it was applied both in primary and recurrent cases. Such grafts must have been bulky and given what we know appreciate in regard to the cosmetic implications of this disease 2,12,13; there would have been motivation to develop more refined and alternative techniques.
As noted, split-skin Ollier–Thiersch grafts were soon adopted for pterygium surgery by Hotz in 1892. 39 It appears that Hotz built on earlier work in which conjunctiva for grafting was harvested from rabbit eyes. Hotz described the healing process after pterygium surgery and noted that where granulation tissue develops in large bare areas post excision, cicatrization is inevitable. In order to “fill the gap” and to “keep the conjunctiva away from the cornea,” he noted that as compared to mucous membrane grafts, Thiersch grafts “are much easier to handle and to fit, they need no sutures, they grow better and shrink much less. 41 And as to the appearance, they are as smooth as the surrounding conjunctiva, but look paler, more whitish” and that the “color would blend well with the white of the eyeball.” These factors remain of importance to this day. Hotz reported on three cases 39 with grafts as large as 10 × 12 mm being used, with apparent success. Hotz had been of the view that “the substitution of epidermal flaps for epithelial tissue [was] a makeshift to be abandoned as soon as we should find a convenient method of obtaining epithelial grafts, for the ideal aim of our plastic work should be to replace mucous membrane by flaps of the same histologic character.” 41 He too referred to work by Gifford, 42 who cut very thin labial flaps using a razor blade (as per Thiersch) with the aid of a specially designed clamp, 43 “so thin that only the epithelium and very little of the sub-adjacent tissue is removed.” When healed, the flap “is so much like the conjunctiva that its boundaries are hard to trace.” This work can be seen as the prelude to the development of cultivated oral mucosal epithelial transplantation (COMET) for severe ocular surface disease. 47,48 Hotz went on to adopt this technique that was eventually used for both primary and recurrent disease. It is of interest that both of these techniques had some longevity—split-skin grafts were still in use in 1977 49 and split-buccal mucous membrane grafts in 1998. 46
Various, sometimes ingenious, forms of sliding or rotated conjunctival grafts 50 were developed, but it seems that they were attended by unacceptable recurrence rates and are no longer widely used. In 1888, Hobby devised a sliding conjunctival flap of mobilized adjacent superonasal bulbar conjunctiva to cover bare sclera. 51 This technique also enjoyed some longevity. 52 In Hirst’s review, 53 the reported recurrence rates were from <1 to >5%, with minimal complications apart from flap retraction. However, as with many pterygium studies, none were prospective, were without controls, and had poor population descriptions. 53 The advantage of this technique, however, is that it attempts to retain limbal integrity, which may be advantageous in preventing recurrence. Another technique for recurrent pterygium was described by Elschnig in 1926 54 in which, after resection, a conjunctival bridge from the opposite limbus was then brought across the cornea in a “bucket handle” to cover the defect, again repairing the limbus (even though the actual temporal limbus was now displaced nasally). These various techniques suffered from the fact that there was still a relative shortage of conjunctival tissue, which could either result in restriction of ocular rotations or a “gap” where healing by second intention, EMT, could result in scarring, chronic inflammation, and lead to recurrence.
The era of autologous conjunctival grafting began with the use of free fragments of conjunctiva used by de Gama Pinto, Gomez-Marques, and de Paula Xavier as described by Rosenthal. 45 Interestingly, this is reminiscent of the modern technique of simple limbal epithelial transplantation (SLET) that is used in the treatment of limbal failure. 55,56 This technique, “mini-SLET,” has recently been shown to be efficacious in a small series of 10 patients who underwent pterygium excision with limbal-sparing SLET and who had been followed up on for up to 8 months. 56 It is our impression that with our autologous limbal-conjunctival grafting technique (described later in this chapter), supported with postoperative hyperbaric oxygen, the graft seems to expand after the healing phase. This is contrary to the accepted wisdom of graft shrinkage, but is consistent with observations of minimal skin graft shrinkage under ideal conditions. 57
A landmark paper in the development of both excision techniques and autoconjunctival grafting was published in 1931 by Gómez-Márquez. 58 He described extensive excision from the pterygium head to the caruncle and subsequent covering of the conjunctival wound with a piece of healthy conjunctiva (▶ Fig. 7.1), taken from the bulbar conjunctiva of the other eye. He cited Duverger, 59 (who also used skin and buccal grafts) who used a similar autograft technique 60 but who referenced da Gama Pinto who harvested conjunctiva from the same eye. This is perhaps the earliest reference to autoconjunctival grafting as we carry out today.
Fig. 7.1 Gómez-Márquez surgical technique, 1931. 58 (a) Extensive excision of the pterygium to the caruncle. (b, c) Conjunctival graft sites from the superior bulbar conjunctiva. Long-axis alignment was determined by extent of conjunctival laxity. (d) Conjunctival graft sutured into position.
Before a modern understanding of pterygium pathophysiology, which he acknowledges, 58 he intuited that its propensity to persist, grow, and recur once extirpated, suggested a neoplastic, tumor-like condition, but with distinctive growth patterns (centripetal and superficial). He reasoned that an ideal strategy would be to “completely extirpate the whole tumor mass.” While this was possible in its corneal aspect, there is no line of demarcation in its conjunctival aspect, particularly at the caruncular base of the pterygium. Limited excision of the pterygium head he saw “as illusory and unsafe as it would be to destroy an army by simply annihilating its guerrillas, the reserves intended to rebuild the front line.” He was concerned about “tumor cells that may exist in the limbal conjunctiva” and “germ”-like cells in apparently normal-looking conjunctiva. He states, “It is therefore necessary to remove this and a large part of the seemingly healthy conjunctiva that prolongs it, in order to prevent the recurrence of pterygium. Proceeding thus, we will have great probability of excising all cells endowed with the specific faculty of replicating in the direction of the cornea and the pterygium will not recur.” He took this to the extreme by preferentially harvesting conjunctiva from the unaffected eye (where possible). This method is still used in cases where insufficient conjunctiva remains in eye being treated, usually because of multiple previous failed surgeries. Gómez-Márquez successfully used this technique on recurrent pterygia and noted, “the appearance of an eye operated according to this technique is perfect.”
These observations bear some semblance to modern day theory. As discussed, large excisions are recommended by some with extensive tenonectomy 12,20 or “walling off” of Tenon’s capsule by over sewing. 21 Such procedures are associated with very low recurrence rates and improved, if not “perfect” cosmesis. The notion of invasive “altered” limbal basal cells has also been confirmed. 61,62,63,64 In excising pterygia, we routinely take ~ 0.5 to 1.00 mm of extra tissue at the limbus in an attempt to clear the limbus of these cells. We have previously postulated 5,65,66 that limbal cells are damaged by peripheral focusing of light being the cornea to the distal limbus, and it is possible that a good part of the limbal circumference is affected. This might account for the types of recurrences in which tissue invades the cornea above and below the graft (▶ Fig. 7.2). This view is also consistent with the not uncommon finding of dysplastic cells in pterygium specimens. 14,15
Fig. 7.2 Pterygium recurrence. A tongue of recurrent pterygium invading the cornea inferior to the lower border of a conjunctival graft. This type of recurrence could be due to inadequate excision of diseased limbal tissue above or below where the pterygium body crosses the limbus. It has been hypothesized that altered/diseased limbal stem cells could reside in these locations and give rise to this aspect of the recurrence. (Image courtesy of Dr Laurence Sullivan.)
Gómez-Márquez acknowledged some drawbacks of his technique—patient reluctance to have surgery on the unaffected eye and the need to occlude both eyes. He also noted the technical difficulty of the flap rolling up on itself—now soluble by the use of Trypan blue that preferentially stains the stromal aspect of the graft. 2 For these reasons, he confined the technique to recurrent cases.
By 1947, Tagle described a technique in which the excision defect was “covered by a graft obtained from the conjunctival tissue along the superior limbus and anchored with suitable sutures. 67 The donor area is closed with a running stitch.” Almost certainly, more widespread usage of this technique occurred with further studies by Barraquer 68 and subsequently, 69 as an understanding of the importance of the limbus and corneal epithelial stem cell biology grew, 70,71,72 and LCAGs evolved (see following text).
Popularization of the CAG technique in North America came with work by Kenyon et al in the 1980s 73 who demonstrated very low recurrence rates in patients with primary or recurrent pterygium. He also described LCAGs, but these were used for other causes of limbal failure. 74 One reason for this may have been concern about inducing limbal failure at the donor site, and since results with conjunctival grafting can be excellent, limbal-conjunctival grafting has not been widely adopted. However, the long-standing concept that corneal epithelial stem cells reside mainly in the limbus has been challenged 75,76; it has been shown that basal cells of bulbar conjunctival epithelium share a similar expression pattern of stem cell–associated markers to the limbal epithelium. 77 This may help explain the success of CAGs and why a limbal area appears to reform when conjunctival grafts are used and that there are apparently few cases of limbal failure at the donor site when LCAGs have been used.
The LCAG technique has evolved, and in general, its use is associated with very low recurrence rates with few reports of complications. 18,78,79,80,81,82 Many authors use this procedure for recurrent pterygium, and in one randomized, prospective, parallel-group clinical trial, the recurrence rate for the CAG group was significantly higher at 10.0% versus the LCAG group with a recurrence rate of 1%. 83 In another randomized controlled trial with a decade of follow-up, patients treated with excision, intraoperative 0.02% MMC, and local conjunctival closure had a recurrence rate of 25.5% as compared to the LCAG group with a recurrence rate of 6.9%. Similar data was presented with 12 years follow-up. 84 Post MMC, eyes had more fibrovascular tissue in the conjunctival bed. This study also found that in the LCAG group all of the recurrences took place within 1 year, for the MMC group there was an ongoing recurrence rate beyond the first year. 85 This is consistent with concerns previously raised about MMC (see later in this chapter).
7.3.2 Author’s Technique
Anesthesia is usually achieved with a peribulbar block using a long-acting anesthetic. Brimonidine eye drops are applied to act as a vasoconstrictor and have the advantage of not dilating the pupil like adrenaline. Half-strength povidone is applied and careful lash-excluding draping is carried out.
The pterygium head is excised using a 23-gauge (G) needle-tip technique 86 (▶ Fig. 7.3a and Video 7.1) minimizing the amount of corneal tissue excised. Smoothing is not carried out since this removes more tissue unnecessarily, given that perhaps the best smoothing agent is resurfacing with corneal epithelium. This technique also allows complete removal of pterygium tissue from the cornea; this is of importance since residual tissue left behind can be a cause of concern of patient dissatisfaction. If the surgeon is right-handed, when operating on a left eye, the microscope is moved so that the pterygium head dissection is carried out from the side, following which it is then moved so that the superior bulbar conjunctiva and limbus can be approached from above. Care is taken as the limbus is approached, as given the change in radius of curvature, it is important to avoid damage to the sclera. Instead, the corneal dissection merges into the limbal subconjunctival/sub-pterygium space. The body of the pterygium is then excised, taking care to excise tissue for about 1 mm above and below these sites where the pterygium crosses the limbus as described above.
Carrying out an extensive but superficial excision, removing inflamed tissue as far as the plica, if it is still present and if not, just short of the caruncle. Check ligaments and the medial rectus muscle capsule are avoided in primary pterygium. Excess Tenon’s capsule underlying this excision area is also trimmed, but “Tenon’s fishing” by pulling on exposed connective tissue is avoided.
Bleeding is controlled with judicious use of an intraocular diathermy (▶ Fig. 7.3b)—particular care is taken to check for bleeding along and under the cut edges of the conjunctiva, especially in the region of the caruncle. Subgraft hemorrhages, while uncommon, have originated form this area.
The size of excised area is measured with calipers (▶ Fig. 7.3c) and its horizontal extent is estimated with the eye in forced abduction and graft size marked (▶ Fig. 7.3d) on the superior bulbar conjunctiva. Graft dissection to the superior limbus is not infrequent, and an assistant can be helpful.
A thin flap is cut with Westcott-style spring scissors (Martin [35–822–11]) (Fig. M5) (▶ Fig. 7.3e) and oversized by ~ 0.5 mm. The dissected graft is then folded down over the cornea, and if necessary, excess Tenon’s capsule can be trimmed from this exposed graft under surface. A stroll wedge (short end) is used to retract the graft inferiorly, and the limbus is further dissected using the tip of a fresh 23-G needle (▶ Fig. 7.3f). Perforating vessels in this area can result in bleeding, controlled with a second wedge. The aim with this maneuver is to split the limbus, leaving behind sufficient palisadal tissue and stem cells to obviate deficiency. The graft is then detached from the limbus (▶ Fig. 7.3g) with Vannas scissors, taking care to include tissue with a crenellated appearance, an indication that tissue is from the palisades of Vogt. The graft is then glided across a balanced salt-lubricated cornea, using non-toothed tying forceps (▶ Fig. 7.3h), flipped over and placed in the wound created by the pterygium excision (▶ Fig. 7.3i). As indicated, if graft orientation is lost, Trypan blue can be used to check. The graft is sutured in place commencing with limbal sutures, and this, I preferred to glue, given the large size of these grafts.
To reduce postoperative pain and to protect the limbus, an extended-wear contact lens is placed at the end of the procedure (▶ Fig. 7.3j). Lenses made of balafilcon A are used since these lenses do not do well supporting the cell growth that occurs in the week that the lens is left in situ, 87 and this is likely to be less “sticky” to the ocular surface. We have also commenced the use of cyclopentolate 1% eye drops at the end of the case, in an attempt to further reduce postoperative pain. 88 Interestingly, in an earlier era, atropine was used for this purpose. 37
Postoperatively, chloramphenicol, Prednefrin forte, and preservative-free diclofenac eye drops are used four times daily and usually tapered after 1 week, with treatment lasting typically 4 to 6 weeks. Topical treatment is prolonged if excessive inflammation occurs, and topical cyclosporine may also be introduced or restarted if the patient was being treated for dry eye syndrome preoperatively. This treatment regime has been used since a key factor in recurrence is postoperative inflammation (see below), and this treatment regime aims to keep this at a minimum. Patients need to be watched carefully since nonsteroidal anti-inflammatory drugs (NSAIDs) can induce corneal melting, and this seems to occur at the junction of the graft with the limbus. At the first sign of an epithelial defect, topical NSAID treatment is ceased, steroid treatment is further reduced, and lubrication is increased. If the graft edge is raised, an extended-wear contact lens is placed.
Fig. 7.3 (a) Pterygium head excision using a 23-gauge needle tip. After the leading edge of the pterygium head is delineated, vertical retraction reveals the point of attachment of the pterygium to the cornea. Touching this junction with the needle tip causes a “splitting” in a plane that is neither too superficial nor too deep. (b) Bleeding is controlled by using an intraocular diathermy tip, to minimize collateral tissue and particularly, vascular damage. (c) Excision defect size is measured with calipers—the horizontal extent is measured with the eye in forced abduction. (d) A typically trapezoidal conjunctival graft is marked out at the superior limbus and on the bulbar conjunctiva, reflecting the size and shape of the excision defect and oversizing by ~ 0.5 mm in length and breadth. (e) A thin conjunctival graft is dissected with spring scissors.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
