1040-5488/14/9107-0747/0 VOL. 91, NO. 7, PP. 747Y751 OPTOMETRY AND VISION SCIENCE Copyright * 2014 American Academy of Optometry
ORIGINAL ARTICLE
Threshold to Predict Astigmatism Reduction after Pterygium Excision Pere Pujol*, Gemma Julio†, Amelia M. de Carvalho‡, Belchin Kostov§, and Toma`s Martı´-Huguet†
ABSTRACT Purpose. To compare corneal astigmatism after pterygium excision, using limbal-conjunctival autograft (LCA) with and without mitomycin C (MMC) and to establish a threshold for postoperative astigmatism reduction. Methods. Sixty-eight eyes with primary pterygium were consecutively sampled and assigned to LCA with MMC (MMC+) or LCA without MMC (MMCj). Corneal lesion length, corneal lesion area, preoperative corneal astigmatism (PRCA), and postoperative corneal astigmatism (POCA) at 1, 3, and 6 months were assessed. Results. Thirty-six men and 24 women (60 eyes), aged between 24 and 65 years (mean T SD, 41 T 8.2 years), completed the study. Thirty-one eyes were included in the MMC+ group and 29 were in the MMCj group. Both groups showed astigmatism reduction (p G 0.05; paired, two-tailed t test) 1 month after the surgery and remained stable. No differences were found between groups in corneal lesion length, corneal lesion area, or astigmatism results at the four time points (p 9 0.05, independent t test). Preoperative corneal astigmatism showed a significant correlation with POCA at 6 months (r = 0.529; p G 0.01). According to receiver operating characteristic curve analysis, the better threshold for astigmatism reduction with the surgery was 1.05 diopters of PRCA, with 82.5% sensitivity and 80.5% specificity. Conclusions. Both surgical procedures could have similar astigmatism results. Pterygium patients with more than 1.05 diopters of PRCA could reduce it after the surgery. Direct relationship between PRCA and POCA revealed that postoperative astigmatism reduction was partial. Therefore, for minimizing final astigmatism, preoperative values should be slightly above the threshold. (Optom Vis Sci 2014;91:747Y751) Key Words: cornea, astigmatism, pterygium, mitomycin C, limbo-conjunctival autograft
P
terygium, from the Greek pterygos, meaning ‘‘wing,’’ is a common ocular surface lesion characterized by elastotic degeneration of collagen, fibrovascular proliferation, with angiogenesis, and an overlying covering of epithelium. It originates on the bulbar conjunctiva, within the palpebral fissure, and grows across the limbus onto the cornea. Lesion excision is the current treatment and multiple surgical approaches have been described.1Y3 There is evidence that a combination of conjunctival or limbal autograft with mitomycin C (MMC) reduces
*MsOphth † PhD ‡ MD § MSc Department of Ophthalmology, Terrassa Hospital, Terrassa Health Consortium, Terrassa, Spain (PP, GJ, AMdC); Ocular Surface Research Group, Optics and Optometry Department, Universitat Polite`cnica de Catalunya, Terrassa, Spain (PP, GJ, TM-H); Transverse Group for Research in Primary Care, IDIBAPS, Barcelona, Spain (BK); Department of Statistics and Operational Research, Universitat Polite`cnica de Catalunya, Barcelona, Spain (BK); and Department of Ophthalmology, Bellvitge University Hospital, Barcelona, Spain (TM-H).
recurrence rate after pterygium excision compared with bare sclera, conjunctival or limbal autograft, or MMC alone, although further assessments are required to confirm the risk/benefit of this combined therapy.4 Ophthalmologists have traditionally regarded this proliferative disorder as a benign lesion, because it frequently grows slowly. Decisions to treat are often based on a patient’s discomfort and cosmetic concerns, unless the lesion is large enough to obscure the visual axis. However, the local invasiveness and the high rate of recurrence when pterygia are inappropriately managed are grounds for arguments against this point of view.5 Additionally, previous reports have demonstrated that, before entering the optical zone, pterygia can cause visual impairment by locally flattening the cornea and inducing with-the-rule astigmatism.6Y9 Several preoperative and postoperative astigmatism comparisons for different surgical techniques have been published,6Y15 confirming that topographic changes induced by pterygium tend to be partially reversed following removal of the lesion. In fact, Yilmaz et al.8 found that bare sclera excision (with or without MMC) was more effective than autograft (conjunctival or
Optometry and Vision Science, Vol. 91, No. 7, July 2014
Copyright © American Academy of Optometry. Unauthorized reproduction of this article is prohibited.
748 Predicting Astigmatism Reduction after PterygiumVPujol et al.
limbal-conjunctival) in reducing corneal astigmatism. However, to the best of our knowledge, measurement of astigmatism has not been reported for the combination of limbal-conjunctival autograft (LCA) with MMC. It has been found that the larger the pterygium, the greater the preoperative and postoperative corneal astigmatism (PRCA and POCA, respectively),7,12,16,17 and the minimum pterygium size with clinical significant astigmatism was also calculated.18 In addition, some pterygia with negligible astigmatism could produce high discomfort and quick lesion growth that justify the excision. Including these cases in the analysis of surgical benefits could be necessary. Apparently, early surgical excision may result in lower postoperative astigmatism; however, it is reasonable to think that small lesions could maintain or increase corneal astigmatism after surgery, because of surgical traumatism. The current study was conducted to compare corneal astigmatism before and 1, 3, and 6 months after pterygium excision using LCA with or without MMC and to determine the critical point for surgical benefit, in terms of astigmatism reduction. This threshold value could predict which patients have a high probability of postoperative astigmatism reduction.
METHODS Study Design A consecutive series of 68 patients (68 eyes) with symptomatic primary pterygium, undergoing removal surgery from January 2012 through December 2012, were enrolled in this prospective study. Patients with a history of ocular surgery or trauma, contact lens use, corneal scarring, or anterior segment diseases other than pterygium (including preexisting corneal endothelial problems) were excluded. Lesion recurrence during the study was also cause for exclusion because it could influence the topographic analysis and the patients with recurrence may be overrepresented when MMC was not used, thus biasing the results if they were included. Subjects were recruited from Terrassa Hospital, and all gave written, informed consent. The clinical protocol and informed consent form were approved by the Terrassa Health Consortium Ethics Committee. The research adhered to the tenets of the Declaration of Helsinki. Standardized photographs of pterygia were taken and the corneal lesion length (LL) and corneal lesion area (LA) were calculated, in a masked manner, using the Analyze/Measure command of the free software ImageJ (W. Rasband, National Institutes of Health, Bethesda, MD; http://rsb.info.nih.gov/ij/). Corneal lesion length was the imaginary straight line between the vertex of each pterygium head and the sclerocorneal limbus. Corneal lesion area was calculated as the area within the corneal outline demarcation of the lesion. A ruler was used, as scale bar in each image, to convert the squared pixels calculated by the program into millimeters or square millimeters. Eyes with a lesion with an advancing edge reaching the central 4-mm zone of the cornea were also excluded because of the potential for inaccurate corneal topography measurements. The degree of astigmatism at the 3-mm central zone of the cornea was measured by a single masked ophthalmologist using computerized videokeratography (* 2009 Carl Zeiss Meditech
Atlas 9000 Corneal Topographer) before and 1, 3, and 6 months after surgery. The topographic variables were PRCA and POCA. The eyes were randomly assigned to two groups according to the surgical technique used: LCA with MMC (MMC+) or LCA without MMC (MMCj). During topical and subconjunctival anesthesia, the pterygium was removed, starting with a conjunctival incision on the body of the lesion using Westcott scissors (Miltex). The Tenon capsule tissue under the body of the pterygium was also removed. For the MMC+ group, when the excision was completed, regular-tip microsponges (Mitomycin-C Inibsa Hospital SLU) soaked with 0.025% MMC solution were placed on the exposed sclera surface for 5 minutes and then removed. The exposed sclera was irrigated with 40 ml of balanced salt solution, and an LCA was obtained from the upper conjunctiva. The graft dissection was extended in all the lesions approximately 0.5 mm into the clear cornea to include the Vogt palisades and limbal stem cells and moved to cover the defective area. Then, it was secured with nylon sutures placing the limbal end of the autograft directly over the limbal area. All surgical procedures were performed by a single surgeon. Damaged tissue was sent for pathological analysis. Postoperatively, patients were treated with topical tobramycin 3 mg/ml and dexamethasone 1 mg/ml (Tobradex; Alcon Cusi, S.A., El Masnou, Spain) tapered off in 1 month.
Statistical Analysis We calculated that a minimum sample size of 29 eyes in each group was necessary to detect an astigmatism difference of 1.1 diopters (D) between groups, assuming an SD of 1.47 D, with 80% statistical power and 0.05 probability of type 1 error (two tailed). The magnitude of the effect chosen was the minimum significant difference in postoperative astigmatism previously found between pterygium surgical techniques.8 After an exploratory analysis of the data, normal distribution was confirmed using the Kolmogorov-Smirnov test. For each group, PRCA was compared with POCA at the different time points using the paired t test. Results were compared between the two groups (MMC+ and MMCj) using independent t test. The relationship between the variables was analyzed with Pearson correlation, and receiver operating characteristic curve analysis was used to determine the critical point for surgery benefit in terms of astigmatism reduction. This is an objective and widely used statistical method to represent the results of diagnostic tests. It graphically compares, point by point, sensitivity and specificity values, allowing one to choose the point that shows the best combination between them predicting the binary result of some process, in this case, astigmatism postoperative reduction or increase. p G 0.05 was considered statistically significant. Statistical analysis was performed using R (version 2.14.2) statistical software (http://www.r-project.org).
RESULTS Of the 68 eyes included initially, 3 (4.4%) developed pterygium recurrence and were excluded and 60 eyes of 60 patients completed the 6 months follow-up. These included 36 men and 24 women, aged between 24 and 65 years (mean T SD, 41 T 8.2 years). Twenty-nine eyes were assigned to the MMCj group and 31 were
Optometry and Vision Science, Vol. 91, No. 7, July 2014
Copyright © American Academy of Optometry. Unauthorized reproduction of this article is prohibited.
Predicting Astigmatism Reduction after PterygiumVPujol et al.
assigned to the MMC+ group. None of them had clinical complications attributed to MMC in the observation period. Table 1 summarizes the statistics for the two groups. Tendency for astigmatism reduction was shown in both MMC+ and MMCj (p G 0.05; paired, two-tailed t test) 1 month after the surgery and the astigmatism remained stable. No significant differences in lesion size, PRCA, or astigmatism results at the three time points were found between the groups (p 9 0.05, independent t test). At the end of the study, the entire sample showed a mean decrease of 0.49 D (95% confidence interval, 0.26 to 0.72 D). In spite of this global tendency, 20 (33.33%) eyes displayed a slight astigmatism increase 6 months after pterygium excision (Fig. 1) (mean increase, 0.27 D; 95% confidence interval, 0.15 to 0.39 D), whereas 40 (66.67%) eyes presented astigmatism reduction (Fig. 1) (mean decrease, 0.87 D; 95% confidence interval, 0.60 to 1.14 D). It is worth mentioning that eyes with PRCA values greater than 1 D clearly showed a tendency for astigmatism reduction. Receiver operating characteristic curve analysis was used to determine the critical point for surgery, distinguishing eyes with astigmatism reduction from those without. The entire sample (N = 60) was used in this analysis because no significant differences between MMC+ and MMCj were found. Previously, the relationship between the variables was studied and only PRCA and LA showed significant correlation with POCA at 6 months (r = 0.529 and 0.478, respectively; p G 0.01 in both cases). Therefore, a higher PRCA, or LA, could mean a higher POCA. Consequently, we applied two receiver operating characteristic curve analyses: one to establish a PRCA threshold that predicts astigmatism reduction 6 months after lesion excision and another to establish an LA threshold with the same purpose. The area under the curve determined by PRCA was 0.873 (Fig. 2), indicating a good predictive capacity, in terms of astigmatism reduction after surgery. The best threshold was 1.05 D of PRCA, with 82.5% sensitivity and 80.5% specificity. In the second analysis, the area under the curve determined by LA was 0.633, showing a very poor predictive capacity for all
749
FIGURE 1. Scatterplot of the astigmatism change versus the preoperative astigmatism value. The solid line represents the better threshold for astigmatism reduction.
the possible threshold values. In this case, there were two possibilities regarding the relationship between surgery benefit and LA (Fig. 3). Eyes with an LA larger than 9 mm2 displayed a clear tendency for astigmatism reduction. However, a disparity in postoperative results was found for eyes with smaller lesions, with some having a decrease in corneal astigmatism whereas others had an increase. This distribution causes uncertainty and indicates reduced predictive capacity for any LA threshold values. All in all, only PRCA seems to have the ability to predict pterygium surgery benefit in terms of astigmatism reduction.
DISCUSSION It is widely accepted that pterygium excision tends to improve corneal astigmatism induced by the lesion,6Y15,19Y21 through
TABLE 1.
Summary of statistics for each group Mean Median SD Minimum Maximum MMCj (17 men and 12 women) Age 40.4 39.5 LL, mm 1.8 1.8 5.8 5.1 LA, mm2 PRCA, D 1.4 1.2 POCA at 1 mo, D 1.2 0.9 POCA at 3 mo, D 0.9 0.9 POCA astigmatism 0.9 0.8 at 6 mo, D MMC+ (19 men and 12 women) Age 41.9 41.5 LL, mm 2.1 2.0 LA, mm2 6.6 6.0 PRCA, D 1.6 1.3 POCA at 1 mo, D 1.2 1.0 POCA at 3 mo, D 1.1 0.9 POCA astigmatism 1.1 1.0 at 6 mo, D
6.7 0.7 3.4 1.1 1.0 0.6 0.5
29 0.6 1.6 0.2 0.2 0.1 0.0
56 3.7 16.0 5.7 5.5 3.1 2.1
9.5 0.7 4.0 1.0 0.8 0.6 0.6
24 0.7 0.3 0.2 0.2 0.2 0.2
65 3.4 18.2 3.4 4.1 3.2 3.2
FIGURE 2. Receiver operating characteristic curve analysis of preoperative astigmatism for distinguishing cases with surgical benefit in terms of astigmatism reduction.
Optometry and Vision Science, Vol. 91, No. 7, July 2014
Copyright © American Academy of Optometry. Unauthorized reproduction of this article is prohibited.
750 Predicting Astigmatism Reduction after PterygiumVPujol et al.
FIGURE 3. Scatterplot of the astigmatism change versus the LA. The solid line represents the lesion area under which astigmatism reduction was uncertain.
remarkable changes at the front corneal surface.13 Our findings agree with the previously described evolution of astigmatism after LCA,11,13 because the improvements were evident and stable 1 month after the surgery. Astigmatism reductions in different studies are difficult to compare because of the different ranges of lesion size and PRCA. Indeed, according to our results, and previous works,22,23 the larger the pterygium, the greater the PRCA. For this reason, it is probable that the mean astigmatism reduction in our study was relatively modest when compared with other studies using the LCA technique for larger pterygia.8,15 In addition, astigmatism results differed among surgical procedures. However, differences were found only when autograft (conjunctival or limbal-conjunctival) and bare sclera (with or without MMC) were compared.8 The current study supports these findings because the use of MMC as an adjuvant does not seem to affect final astigmatism. However, drug combination with LCA seems to further reduce recurrences4 and may avoid some potential corneal scarring attributed to the inhibitory effect of MMC on fibroblast proliferation, although this last question needs new studies. The route of administration, dose, and duration of the drug treatment could be determining factors for obtaining a good risk/ benefit relationship in pterygium surgery.4 The dosage of the drug used in the current study has been reported to induce postoperative corneal endothelial loss,24 but no patients had clinical complications in the observation period, perhaps because of the absence of preexisting endothelial alterations. A recent study has confirmed that treatment with intraoperative MMC (under similar conditions to this study) was not associated with long-term corneal endothelial cell loss.25 Despite the clear tendency toward reduced POCA, there were some eyes with postoperative increases in astigmatism, although the mean increase was slight. Yilmaz et al.8 suggested that grafting the limbal conjunctiva with nylon sutures after excising the lesion creates a flattened effect of the cornea, mimicking the effect of a small pterygium. This effect could be one of the causes of the
global astigmatism increase found with small lesions. In fact, according to the results of the present study, postoperative astigmatism reduction was highly uncertain for eyes with an LA less than 9 mm2. Further studies are needed to evaluate the determining factors for surgery-related astigmatism changes in patients with small lesions. The main contribution of this work is the determination of a good threshold value for surgery benefit, in terms of astigmatism reduction. The findings suggest that, among the two variables directly correlated with POCA at 6 months (PRCA and LA), only PRCA seems to have a good predictive capacity. The better threshold value was 1.05 D of PRCA, with 82.5% sensitivity and 80.5% specificity. Applying this threshold would select 82.5% of the patients with postoperative astigmatism reduction and 19.5% (false-positive rate = 100 j 80.5) of the patients with postoperative astigmatism increase, although the mean increase in the latter patients seems to be mild, according to our results. However, surgery could provide other benefits. For instance, it could reduce ocular discomfort, which seems to be more frequent with small lesions.26 For this reason, all factors should be considered when deciding whether to operate and this threshold seems to be a suitable tool to assess one of those factors. Mohammad-Salih and Sharif18 found that lesion area is the best size indicator of PRCA in pterygium patients. According to our findings, LA not only correlated better than LL with PRCA but also was associated with POCA at 6 months. Despite this fact, LA showed a very low potential to predict surgical benefit. This seems to be attributable to the great uncertainty in regard to results for small lesions, as previously mentioned. There were some limitations to this study. We did not include recurrent pterygia and thus cannot make conclusions about outcomes for such cases. Another limitation could be the small number of large lesions that were included. However, larger pterygia showed a very clear tendency to improve corneal postoperative astigmatism.12,16,23 It is plausible that including a larger number of lesions with this characteristic would not fundamentally change the final conclusion of this work. In fact, this threshold determined the critical point where surgery clearly tends to produce astigmatism reduction with minimal final astigmatism, taken into account the direct relationship between PRCA and POCA reported in this and other studies.7,12,16,17 Indeed, despite reduction, astigmatism is partially reversed because patients with PRCA values higher than the threshold could tend to show higher final corneal astigmatism than patients with PRCA closer to it. In summary, combined surgical technique, one of the better strategies to reduce recurrences,4 seems to have astigmatism results similar to those of autograft alone. The great majority of pterygium patients with PRCA greater than 1.05 D could have substantial benefit, that is, reducing their corneal astigmatism after excision with LCA, with or without MMC. Patients with values below this threshold could tend to experience a slight increase in astigmatism after surgery, whereas patients with higher values, despite reduction, could tend to show higher final corneal astigmatism. This threshold value is only applicable to LCA or LCA combined with MMC, two surgical procedures with low recurrence rates. Its application in other excision surgical techniques could be suitable, but further assessments are required.
Optometry and Vision Science, Vol. 91, No. 7, July 2014
Copyright © American Academy of Optometry. Unauthorized reproduction of this article is prohibited.
Predicting Astigmatism Reduction after PterygiumVPujol et al.
ACKNOWLEDGMENTS All the authors declare no conflicts of interest, including relevant financial interests, activities, relationships, and affiliations. Received January 20, 2014; accepted April 15, 2014.
REFERENCES 1. Al Fayez MF. Limbal versus conjunctival autograft transplantation for advanced and recurrent pterygium. Ophthalmology 2002;109: 1752Y5. 2. Hirst LW. The treatment of pterygium. Surv Ophthalmol 2003;48: 145Y80. 3. Ozkurt YB, Kocams O, Comez AT, Uslu B, Dogan OK. Treatment of primary pterygium. Optom Vis Sci 2009;86:1178Y81. 4. Kaufman SC, Jacobs DS, Lee WB, Deng SX, Rosenblatt MI, Shtein RM. Options and adjuvants in surgery for pterygium: a report by the American Academy of Ophthalmology. Ophthalmology 2013;120: 201Y8. 5. Chui J, Coroneo MT, Tat LT, Crouch R, Wakefield D, Di Girolamo N. Ophthalmic pterygium: a stem cell disorder with premalignant features. Am J Pathol 2011;178:817Y27. ¨ zcan AA, Sari S, Erso¨z TR. Visual acuity and corneal 6. Yagmur M, O topographic changes related with pterygium surgery. J Refract Surg 2005;21:166Y70. 7. Bahar I, Loya N, Weinberger D, Avisar R. Effect of pterygium surgery on corneal topography: a prospective study. Cornea 2004;23: 113Y7. 8. Yilmaz S, Yuksel T, Maden A. Corneal topographic changes after four types of pterygium surgery. J Refract Surg 2008;24:160Y5. 9. Cinal A, Yasar T, Demirok A, Topuz H. The effect of pterygium surgery on corneal topography. Ophthalmic Surg Lasers 2001;32: 35Y40. 10. Ozdemir M, Cinal A. Early and late effects of pterygium surgery on corneal topography. Ophthalmic Surg Lasers Imaging 2005;36: 451Y6. 11. Tomidokoro A, Miyata K, Sakaguchi Y, Samejima T, Tokunaga T, Oshika T. Effects of pterygium on corneal spherical power and astigmatism. Ophthalmology 2000;107:1568Y71. 12. Fong KS, Balakrishnan V, Chee SP, Tan DT. Refractive change following pterygium surgery. CLAO J 1998;24:115Y7. 13. Kheirkhah A, Safi H, Molaei S, Nazari R, Behrouz MJ, Raju VK. Effects of pterygium surgery on front and back corneal astigmatism. Can J Ophthalmol 2012;47:423Y8.
751
14. Oh JY, Wee WR. The effect of pterygium surgery on contrast sensitivity and corneal topographic changes. Clin Ophthalmol 2010;4: 315Y9. 15. Errais K, Bouden J, Mili-Boussen I, Anane R, Beltaif O, Meddeb Ouertani A. Effect of pterygium surgery on corneal topography. Eur J Ophthalmol 2008;18:177Y81. 16. Lin A, Stern G. Correlation between pterygium size and induced corneal astigmatism. Cornea 1998;17:28Y30. 17. Avisar R, Loya N, Yassur Y, Weinberger D. Pterygium-induced corneal astigmatism. Isr Med Assoc J 2000;2:14Y5. 18. Mohammad-Salih PA, Sharif AF. Analysis of pterygium size and induced corneal astigmatism. Cornea 2008;27:434Y8. 19. Soriano JM, Janknecht P, Witschel H. [Effect of pterygium operation on preoperative astigmatism. Prospective study]. Ophthalmologe 1993;90:688Y90. 20. Stern GA, Lin A. Effect of pterygium excision on induced corneal topographic abnormalities. Cornea 1998;17:23Y7. 21. Budak K, Khater TT, Friedman NJ, Koch DD. Corneal topographic changes induced by excision of perilimbal lesions. Ophthalmic Surg Lasers 1999;30:458Y64. 22. Hansen A, Norn M. Astigmatism and surface phenomena in pterygium. Acta Ophthalmol (Copenh) 1980;58:174Y81. 23. Tomidokoro A, Oshika T, Amano S, Eguchi K, Eguchi S. Quantitative analysis of regular and irregular astigmatism induced by pterygium. Cornea 1999;18:412Y5. 24. Kheirkhah A, Izadi A, Kiarudi MY, Nazari R, Hashemian H, Behrouz MJ. Effects of mitomycin C on corneal endothelial cell counts in pterygium surgery: role of application location. Am J Ophthalmol 2011;151:488Y93. 25. Young AL, Ho M, Jhanji V, Cheng LL. Ten-year results of a randomized controlled trial comparing 0.02% mitomycin C and limbal conjunctival autograft in pterygium surgery. Ophthalmology 2013; 120:2390Y5. 26. Julio G, Lluch S, Pujol P, Merindano D. Ocular discomfort in pterygium patients. Optom Vis Sci 2013;90:269Y74.
Gemma Julio Optics and Optometry Department Violinista Vellsola` 37 08222 Terrassa Spain e-mail: [email protected]
Optometry and Vision Science, Vol. 91, No. 7, July 2014
Copyright © American Academy of Optometry. Unauthorized reproduction of this article is prohibited.
ORIGINAL ARTICLE
Threshold to Predict Astigmatism Reduction after Pterygium Excision Pere Pujol*, Gemma Julio†, Amelia M. de Carvalho‡, Belchin Kostov§, and Toma`s Martı´-Huguet†
ABSTRACT Purpose. To compare corneal astigmatism after pterygium excision, using limbal-conjunctival autograft (LCA) with and without mitomycin C (MMC) and to establish a threshold for postoperative astigmatism reduction. Methods. Sixty-eight eyes with primary pterygium were consecutively sampled and assigned to LCA with MMC (MMC+) or LCA without MMC (MMCj). Corneal lesion length, corneal lesion area, preoperative corneal astigmatism (PRCA), and postoperative corneal astigmatism (POCA) at 1, 3, and 6 months were assessed. Results. Thirty-six men and 24 women (60 eyes), aged between 24 and 65 years (mean T SD, 41 T 8.2 years), completed the study. Thirty-one eyes were included in the MMC+ group and 29 were in the MMCj group. Both groups showed astigmatism reduction (p G 0.05; paired, two-tailed t test) 1 month after the surgery and remained stable. No differences were found between groups in corneal lesion length, corneal lesion area, or astigmatism results at the four time points (p 9 0.05, independent t test). Preoperative corneal astigmatism showed a significant correlation with POCA at 6 months (r = 0.529; p G 0.01). According to receiver operating characteristic curve analysis, the better threshold for astigmatism reduction with the surgery was 1.05 diopters of PRCA, with 82.5% sensitivity and 80.5% specificity. Conclusions. Both surgical procedures could have similar astigmatism results. Pterygium patients with more than 1.05 diopters of PRCA could reduce it after the surgery. Direct relationship between PRCA and POCA revealed that postoperative astigmatism reduction was partial. Therefore, for minimizing final astigmatism, preoperative values should be slightly above the threshold. (Optom Vis Sci 2014;91:747Y751) Key Words: cornea, astigmatism, pterygium, mitomycin C, limbo-conjunctival autograft
P
terygium, from the Greek pterygos, meaning ‘‘wing,’’ is a common ocular surface lesion characterized by elastotic degeneration of collagen, fibrovascular proliferation, with angiogenesis, and an overlying covering of epithelium. It originates on the bulbar conjunctiva, within the palpebral fissure, and grows across the limbus onto the cornea. Lesion excision is the current treatment and multiple surgical approaches have been described.1Y3 There is evidence that a combination of conjunctival or limbal autograft with mitomycin C (MMC) reduces
*MsOphth † PhD ‡ MD § MSc Department of Ophthalmology, Terrassa Hospital, Terrassa Health Consortium, Terrassa, Spain (PP, GJ, AMdC); Ocular Surface Research Group, Optics and Optometry Department, Universitat Polite`cnica de Catalunya, Terrassa, Spain (PP, GJ, TM-H); Transverse Group for Research in Primary Care, IDIBAPS, Barcelona, Spain (BK); Department of Statistics and Operational Research, Universitat Polite`cnica de Catalunya, Barcelona, Spain (BK); and Department of Ophthalmology, Bellvitge University Hospital, Barcelona, Spain (TM-H).
recurrence rate after pterygium excision compared with bare sclera, conjunctival or limbal autograft, or MMC alone, although further assessments are required to confirm the risk/benefit of this combined therapy.4 Ophthalmologists have traditionally regarded this proliferative disorder as a benign lesion, because it frequently grows slowly. Decisions to treat are often based on a patient’s discomfort and cosmetic concerns, unless the lesion is large enough to obscure the visual axis. However, the local invasiveness and the high rate of recurrence when pterygia are inappropriately managed are grounds for arguments against this point of view.5 Additionally, previous reports have demonstrated that, before entering the optical zone, pterygia can cause visual impairment by locally flattening the cornea and inducing with-the-rule astigmatism.6Y9 Several preoperative and postoperative astigmatism comparisons for different surgical techniques have been published,6Y15 confirming that topographic changes induced by pterygium tend to be partially reversed following removal of the lesion. In fact, Yilmaz et al.8 found that bare sclera excision (with or without MMC) was more effective than autograft (conjunctival or
Optometry and Vision Science, Vol. 91, No. 7, July 2014
Copyright © American Academy of Optometry. Unauthorized reproduction of this article is prohibited.
748 Predicting Astigmatism Reduction after PterygiumVPujol et al.
limbal-conjunctival) in reducing corneal astigmatism. However, to the best of our knowledge, measurement of astigmatism has not been reported for the combination of limbal-conjunctival autograft (LCA) with MMC. It has been found that the larger the pterygium, the greater the preoperative and postoperative corneal astigmatism (PRCA and POCA, respectively),7,12,16,17 and the minimum pterygium size with clinical significant astigmatism was also calculated.18 In addition, some pterygia with negligible astigmatism could produce high discomfort and quick lesion growth that justify the excision. Including these cases in the analysis of surgical benefits could be necessary. Apparently, early surgical excision may result in lower postoperative astigmatism; however, it is reasonable to think that small lesions could maintain or increase corneal astigmatism after surgery, because of surgical traumatism. The current study was conducted to compare corneal astigmatism before and 1, 3, and 6 months after pterygium excision using LCA with or without MMC and to determine the critical point for surgical benefit, in terms of astigmatism reduction. This threshold value could predict which patients have a high probability of postoperative astigmatism reduction.
METHODS Study Design A consecutive series of 68 patients (68 eyes) with symptomatic primary pterygium, undergoing removal surgery from January 2012 through December 2012, were enrolled in this prospective study. Patients with a history of ocular surgery or trauma, contact lens use, corneal scarring, or anterior segment diseases other than pterygium (including preexisting corneal endothelial problems) were excluded. Lesion recurrence during the study was also cause for exclusion because it could influence the topographic analysis and the patients with recurrence may be overrepresented when MMC was not used, thus biasing the results if they were included. Subjects were recruited from Terrassa Hospital, and all gave written, informed consent. The clinical protocol and informed consent form were approved by the Terrassa Health Consortium Ethics Committee. The research adhered to the tenets of the Declaration of Helsinki. Standardized photographs of pterygia were taken and the corneal lesion length (LL) and corneal lesion area (LA) were calculated, in a masked manner, using the Analyze/Measure command of the free software ImageJ (W. Rasband, National Institutes of Health, Bethesda, MD; http://rsb.info.nih.gov/ij/). Corneal lesion length was the imaginary straight line between the vertex of each pterygium head and the sclerocorneal limbus. Corneal lesion area was calculated as the area within the corneal outline demarcation of the lesion. A ruler was used, as scale bar in each image, to convert the squared pixels calculated by the program into millimeters or square millimeters. Eyes with a lesion with an advancing edge reaching the central 4-mm zone of the cornea were also excluded because of the potential for inaccurate corneal topography measurements. The degree of astigmatism at the 3-mm central zone of the cornea was measured by a single masked ophthalmologist using computerized videokeratography (* 2009 Carl Zeiss Meditech
Atlas 9000 Corneal Topographer) before and 1, 3, and 6 months after surgery. The topographic variables were PRCA and POCA. The eyes were randomly assigned to two groups according to the surgical technique used: LCA with MMC (MMC+) or LCA without MMC (MMCj). During topical and subconjunctival anesthesia, the pterygium was removed, starting with a conjunctival incision on the body of the lesion using Westcott scissors (Miltex). The Tenon capsule tissue under the body of the pterygium was also removed. For the MMC+ group, when the excision was completed, regular-tip microsponges (Mitomycin-C Inibsa Hospital SLU) soaked with 0.025% MMC solution were placed on the exposed sclera surface for 5 minutes and then removed. The exposed sclera was irrigated with 40 ml of balanced salt solution, and an LCA was obtained from the upper conjunctiva. The graft dissection was extended in all the lesions approximately 0.5 mm into the clear cornea to include the Vogt palisades and limbal stem cells and moved to cover the defective area. Then, it was secured with nylon sutures placing the limbal end of the autograft directly over the limbal area. All surgical procedures were performed by a single surgeon. Damaged tissue was sent for pathological analysis. Postoperatively, patients were treated with topical tobramycin 3 mg/ml and dexamethasone 1 mg/ml (Tobradex; Alcon Cusi, S.A., El Masnou, Spain) tapered off in 1 month.
Statistical Analysis We calculated that a minimum sample size of 29 eyes in each group was necessary to detect an astigmatism difference of 1.1 diopters (D) between groups, assuming an SD of 1.47 D, with 80% statistical power and 0.05 probability of type 1 error (two tailed). The magnitude of the effect chosen was the minimum significant difference in postoperative astigmatism previously found between pterygium surgical techniques.8 After an exploratory analysis of the data, normal distribution was confirmed using the Kolmogorov-Smirnov test. For each group, PRCA was compared with POCA at the different time points using the paired t test. Results were compared between the two groups (MMC+ and MMCj) using independent t test. The relationship between the variables was analyzed with Pearson correlation, and receiver operating characteristic curve analysis was used to determine the critical point for surgery benefit in terms of astigmatism reduction. This is an objective and widely used statistical method to represent the results of diagnostic tests. It graphically compares, point by point, sensitivity and specificity values, allowing one to choose the point that shows the best combination between them predicting the binary result of some process, in this case, astigmatism postoperative reduction or increase. p G 0.05 was considered statistically significant. Statistical analysis was performed using R (version 2.14.2) statistical software (http://www.r-project.org).
RESULTS Of the 68 eyes included initially, 3 (4.4%) developed pterygium recurrence and were excluded and 60 eyes of 60 patients completed the 6 months follow-up. These included 36 men and 24 women, aged between 24 and 65 years (mean T SD, 41 T 8.2 years). Twenty-nine eyes were assigned to the MMCj group and 31 were
Optometry and Vision Science, Vol. 91, No. 7, July 2014
Copyright © American Academy of Optometry. Unauthorized reproduction of this article is prohibited.
Predicting Astigmatism Reduction after PterygiumVPujol et al.
assigned to the MMC+ group. None of them had clinical complications attributed to MMC in the observation period. Table 1 summarizes the statistics for the two groups. Tendency for astigmatism reduction was shown in both MMC+ and MMCj (p G 0.05; paired, two-tailed t test) 1 month after the surgery and the astigmatism remained stable. No significant differences in lesion size, PRCA, or astigmatism results at the three time points were found between the groups (p 9 0.05, independent t test). At the end of the study, the entire sample showed a mean decrease of 0.49 D (95% confidence interval, 0.26 to 0.72 D). In spite of this global tendency, 20 (33.33%) eyes displayed a slight astigmatism increase 6 months after pterygium excision (Fig. 1) (mean increase, 0.27 D; 95% confidence interval, 0.15 to 0.39 D), whereas 40 (66.67%) eyes presented astigmatism reduction (Fig. 1) (mean decrease, 0.87 D; 95% confidence interval, 0.60 to 1.14 D). It is worth mentioning that eyes with PRCA values greater than 1 D clearly showed a tendency for astigmatism reduction. Receiver operating characteristic curve analysis was used to determine the critical point for surgery, distinguishing eyes with astigmatism reduction from those without. The entire sample (N = 60) was used in this analysis because no significant differences between MMC+ and MMCj were found. Previously, the relationship between the variables was studied and only PRCA and LA showed significant correlation with POCA at 6 months (r = 0.529 and 0.478, respectively; p G 0.01 in both cases). Therefore, a higher PRCA, or LA, could mean a higher POCA. Consequently, we applied two receiver operating characteristic curve analyses: one to establish a PRCA threshold that predicts astigmatism reduction 6 months after lesion excision and another to establish an LA threshold with the same purpose. The area under the curve determined by PRCA was 0.873 (Fig. 2), indicating a good predictive capacity, in terms of astigmatism reduction after surgery. The best threshold was 1.05 D of PRCA, with 82.5% sensitivity and 80.5% specificity. In the second analysis, the area under the curve determined by LA was 0.633, showing a very poor predictive capacity for all
749
FIGURE 1. Scatterplot of the astigmatism change versus the preoperative astigmatism value. The solid line represents the better threshold for astigmatism reduction.
the possible threshold values. In this case, there were two possibilities regarding the relationship between surgery benefit and LA (Fig. 3). Eyes with an LA larger than 9 mm2 displayed a clear tendency for astigmatism reduction. However, a disparity in postoperative results was found for eyes with smaller lesions, with some having a decrease in corneal astigmatism whereas others had an increase. This distribution causes uncertainty and indicates reduced predictive capacity for any LA threshold values. All in all, only PRCA seems to have the ability to predict pterygium surgery benefit in terms of astigmatism reduction.
DISCUSSION It is widely accepted that pterygium excision tends to improve corneal astigmatism induced by the lesion,6Y15,19Y21 through
TABLE 1.
Summary of statistics for each group Mean Median SD Minimum Maximum MMCj (17 men and 12 women) Age 40.4 39.5 LL, mm 1.8 1.8 5.8 5.1 LA, mm2 PRCA, D 1.4 1.2 POCA at 1 mo, D 1.2 0.9 POCA at 3 mo, D 0.9 0.9 POCA astigmatism 0.9 0.8 at 6 mo, D MMC+ (19 men and 12 women) Age 41.9 41.5 LL, mm 2.1 2.0 LA, mm2 6.6 6.0 PRCA, D 1.6 1.3 POCA at 1 mo, D 1.2 1.0 POCA at 3 mo, D 1.1 0.9 POCA astigmatism 1.1 1.0 at 6 mo, D
6.7 0.7 3.4 1.1 1.0 0.6 0.5
29 0.6 1.6 0.2 0.2 0.1 0.0
56 3.7 16.0 5.7 5.5 3.1 2.1
9.5 0.7 4.0 1.0 0.8 0.6 0.6
24 0.7 0.3 0.2 0.2 0.2 0.2
65 3.4 18.2 3.4 4.1 3.2 3.2
FIGURE 2. Receiver operating characteristic curve analysis of preoperative astigmatism for distinguishing cases with surgical benefit in terms of astigmatism reduction.
Optometry and Vision Science, Vol. 91, No. 7, July 2014
Copyright © American Academy of Optometry. Unauthorized reproduction of this article is prohibited.
750 Predicting Astigmatism Reduction after PterygiumVPujol et al.
FIGURE 3. Scatterplot of the astigmatism change versus the LA. The solid line represents the lesion area under which astigmatism reduction was uncertain.
remarkable changes at the front corneal surface.13 Our findings agree with the previously described evolution of astigmatism after LCA,11,13 because the improvements were evident and stable 1 month after the surgery. Astigmatism reductions in different studies are difficult to compare because of the different ranges of lesion size and PRCA. Indeed, according to our results, and previous works,22,23 the larger the pterygium, the greater the PRCA. For this reason, it is probable that the mean astigmatism reduction in our study was relatively modest when compared with other studies using the LCA technique for larger pterygia.8,15 In addition, astigmatism results differed among surgical procedures. However, differences were found only when autograft (conjunctival or limbal-conjunctival) and bare sclera (with or without MMC) were compared.8 The current study supports these findings because the use of MMC as an adjuvant does not seem to affect final astigmatism. However, drug combination with LCA seems to further reduce recurrences4 and may avoid some potential corneal scarring attributed to the inhibitory effect of MMC on fibroblast proliferation, although this last question needs new studies. The route of administration, dose, and duration of the drug treatment could be determining factors for obtaining a good risk/ benefit relationship in pterygium surgery.4 The dosage of the drug used in the current study has been reported to induce postoperative corneal endothelial loss,24 but no patients had clinical complications in the observation period, perhaps because of the absence of preexisting endothelial alterations. A recent study has confirmed that treatment with intraoperative MMC (under similar conditions to this study) was not associated with long-term corneal endothelial cell loss.25 Despite the clear tendency toward reduced POCA, there were some eyes with postoperative increases in astigmatism, although the mean increase was slight. Yilmaz et al.8 suggested that grafting the limbal conjunctiva with nylon sutures after excising the lesion creates a flattened effect of the cornea, mimicking the effect of a small pterygium. This effect could be one of the causes of the
global astigmatism increase found with small lesions. In fact, according to the results of the present study, postoperative astigmatism reduction was highly uncertain for eyes with an LA less than 9 mm2. Further studies are needed to evaluate the determining factors for surgery-related astigmatism changes in patients with small lesions. The main contribution of this work is the determination of a good threshold value for surgery benefit, in terms of astigmatism reduction. The findings suggest that, among the two variables directly correlated with POCA at 6 months (PRCA and LA), only PRCA seems to have a good predictive capacity. The better threshold value was 1.05 D of PRCA, with 82.5% sensitivity and 80.5% specificity. Applying this threshold would select 82.5% of the patients with postoperative astigmatism reduction and 19.5% (false-positive rate = 100 j 80.5) of the patients with postoperative astigmatism increase, although the mean increase in the latter patients seems to be mild, according to our results. However, surgery could provide other benefits. For instance, it could reduce ocular discomfort, which seems to be more frequent with small lesions.26 For this reason, all factors should be considered when deciding whether to operate and this threshold seems to be a suitable tool to assess one of those factors. Mohammad-Salih and Sharif18 found that lesion area is the best size indicator of PRCA in pterygium patients. According to our findings, LA not only correlated better than LL with PRCA but also was associated with POCA at 6 months. Despite this fact, LA showed a very low potential to predict surgical benefit. This seems to be attributable to the great uncertainty in regard to results for small lesions, as previously mentioned. There were some limitations to this study. We did not include recurrent pterygia and thus cannot make conclusions about outcomes for such cases. Another limitation could be the small number of large lesions that were included. However, larger pterygia showed a very clear tendency to improve corneal postoperative astigmatism.12,16,23 It is plausible that including a larger number of lesions with this characteristic would not fundamentally change the final conclusion of this work. In fact, this threshold determined the critical point where surgery clearly tends to produce astigmatism reduction with minimal final astigmatism, taken into account the direct relationship between PRCA and POCA reported in this and other studies.7,12,16,17 Indeed, despite reduction, astigmatism is partially reversed because patients with PRCA values higher than the threshold could tend to show higher final corneal astigmatism than patients with PRCA closer to it. In summary, combined surgical technique, one of the better strategies to reduce recurrences,4 seems to have astigmatism results similar to those of autograft alone. The great majority of pterygium patients with PRCA greater than 1.05 D could have substantial benefit, that is, reducing their corneal astigmatism after excision with LCA, with or without MMC. Patients with values below this threshold could tend to experience a slight increase in astigmatism after surgery, whereas patients with higher values, despite reduction, could tend to show higher final corneal astigmatism. This threshold value is only applicable to LCA or LCA combined with MMC, two surgical procedures with low recurrence rates. Its application in other excision surgical techniques could be suitable, but further assessments are required.
Optometry and Vision Science, Vol. 91, No. 7, July 2014
Copyright © American Academy of Optometry. Unauthorized reproduction of this article is prohibited.
Predicting Astigmatism Reduction after PterygiumVPujol et al.
ACKNOWLEDGMENTS All the authors declare no conflicts of interest, including relevant financial interests, activities, relationships, and affiliations. Received January 20, 2014; accepted April 15, 2014.
REFERENCES 1. Al Fayez MF. Limbal versus conjunctival autograft transplantation for advanced and recurrent pterygium. Ophthalmology 2002;109: 1752Y5. 2. Hirst LW. The treatment of pterygium. Surv Ophthalmol 2003;48: 145Y80. 3. Ozkurt YB, Kocams O, Comez AT, Uslu B, Dogan OK. Treatment of primary pterygium. Optom Vis Sci 2009;86:1178Y81. 4. Kaufman SC, Jacobs DS, Lee WB, Deng SX, Rosenblatt MI, Shtein RM. Options and adjuvants in surgery for pterygium: a report by the American Academy of Ophthalmology. Ophthalmology 2013;120: 201Y8. 5. Chui J, Coroneo MT, Tat LT, Crouch R, Wakefield D, Di Girolamo N. Ophthalmic pterygium: a stem cell disorder with premalignant features. Am J Pathol 2011;178:817Y27. ¨ zcan AA, Sari S, Erso¨z TR. Visual acuity and corneal 6. Yagmur M, O topographic changes related with pterygium surgery. J Refract Surg 2005;21:166Y70. 7. Bahar I, Loya N, Weinberger D, Avisar R. Effect of pterygium surgery on corneal topography: a prospective study. Cornea 2004;23: 113Y7. 8. Yilmaz S, Yuksel T, Maden A. Corneal topographic changes after four types of pterygium surgery. J Refract Surg 2008;24:160Y5. 9. Cinal A, Yasar T, Demirok A, Topuz H. The effect of pterygium surgery on corneal topography. Ophthalmic Surg Lasers 2001;32: 35Y40. 10. Ozdemir M, Cinal A. Early and late effects of pterygium surgery on corneal topography. Ophthalmic Surg Lasers Imaging 2005;36: 451Y6. 11. Tomidokoro A, Miyata K, Sakaguchi Y, Samejima T, Tokunaga T, Oshika T. Effects of pterygium on corneal spherical power and astigmatism. Ophthalmology 2000;107:1568Y71. 12. Fong KS, Balakrishnan V, Chee SP, Tan DT. Refractive change following pterygium surgery. CLAO J 1998;24:115Y7. 13. Kheirkhah A, Safi H, Molaei S, Nazari R, Behrouz MJ, Raju VK. Effects of pterygium surgery on front and back corneal astigmatism. Can J Ophthalmol 2012;47:423Y8.
751
14. Oh JY, Wee WR. The effect of pterygium surgery on contrast sensitivity and corneal topographic changes. Clin Ophthalmol 2010;4: 315Y9. 15. Errais K, Bouden J, Mili-Boussen I, Anane R, Beltaif O, Meddeb Ouertani A. Effect of pterygium surgery on corneal topography. Eur J Ophthalmol 2008;18:177Y81. 16. Lin A, Stern G. Correlation between pterygium size and induced corneal astigmatism. Cornea 1998;17:28Y30. 17. Avisar R, Loya N, Yassur Y, Weinberger D. Pterygium-induced corneal astigmatism. Isr Med Assoc J 2000;2:14Y5. 18. Mohammad-Salih PA, Sharif AF. Analysis of pterygium size and induced corneal astigmatism. Cornea 2008;27:434Y8. 19. Soriano JM, Janknecht P, Witschel H. [Effect of pterygium operation on preoperative astigmatism. Prospective study]. Ophthalmologe 1993;90:688Y90. 20. Stern GA, Lin A. Effect of pterygium excision on induced corneal topographic abnormalities. Cornea 1998;17:23Y7. 21. Budak K, Khater TT, Friedman NJ, Koch DD. Corneal topographic changes induced by excision of perilimbal lesions. Ophthalmic Surg Lasers 1999;30:458Y64. 22. Hansen A, Norn M. Astigmatism and surface phenomena in pterygium. Acta Ophthalmol (Copenh) 1980;58:174Y81. 23. Tomidokoro A, Oshika T, Amano S, Eguchi K, Eguchi S. Quantitative analysis of regular and irregular astigmatism induced by pterygium. Cornea 1999;18:412Y5. 24. Kheirkhah A, Izadi A, Kiarudi MY, Nazari R, Hashemian H, Behrouz MJ. Effects of mitomycin C on corneal endothelial cell counts in pterygium surgery: role of application location. Am J Ophthalmol 2011;151:488Y93. 25. Young AL, Ho M, Jhanji V, Cheng LL. Ten-year results of a randomized controlled trial comparing 0.02% mitomycin C and limbal conjunctival autograft in pterygium surgery. Ophthalmology 2013; 120:2390Y5. 26. Julio G, Lluch S, Pujol P, Merindano D. Ocular discomfort in pterygium patients. Optom Vis Sci 2013;90:269Y74.
Gemma Julio Optics and Optometry Department Violinista Vellsola` 37 08222 Terrassa Spain e-mail: [email protected]
Optometry and Vision Science, Vol. 91, No. 7, July 2014
Copyright © American Academy of Optometry. Unauthorized reproduction of this article is prohibited.
