- Email: [email protected]
Total ocular surface amniotic membrane transplantation for paraquat-induced ocular surface injury
ORIGINAL ARTICLE
Total ocular surface amniotic membrane transplantation for paraquat-induced ocular surface injury Ting Wang, MD, PhD, Changsen Liang, MD, Xiaohan Xu, MD, Weiyun Shi, MD, PhD ABSTRACT ● RÉSUMÉ Objective: To evaluate the therapeutic efficacy of modified amniotic membrane transplantation (MAMT) for paraquat-induced ocular surface injury. Design: Retrospective case series. Participants: Thirty patients (30 eyes) with paraquat-induced ocular surface injury. Methods: Among the patients, 8 underwent MAMT, 14 received conventional amniotic membrane transplantation (AMT), and 8 were treated with simple drug therapy (DT). Features related to the damage, corneal epithelial defect closure time, visual acuity, stromal haze, and complications were recorded. Results: In the MAMT and AMT groups, visual acuity in all eyes recovered to the preinjury level; in the DT group, visual acuity in 3 eyes (37.5%) recovered to the preinjury level. The mean corneal epithelial defect closure time was 7.6 ⫾ 2.7 days in the MAMT group, 9.8 ⫾ 3.6 days in the AMT group, and 18.2 ⫾ 5.2 days in the DT group (p o 0.05). There was a significant difference in the symblepharon rate after treatment among the 3 groups (MAMT: 0%, AMT: 35.7%, DT: 87.5%; p o 0.05). Although the tear secretion was reduced in all groups, it was significantly lower in the DT group compared with the MAMT and AMT groups (p o 0.05). Conclusions: Paraquat-induced ocular injuries can lead to whole ocular surface damage. MAMT treatment in a timely manner can effectively promote the repair of the ocular surface and reduce the complications from symblepharon. Objet : Évaluer l’efficacité thérapeutique de la greffe de membrane amniotique modifiée (GMAM) dans des cas de lésion de la surface oculaire causée par le paraquat. Nature : Étude de cas rétrospective. Participants : Trente patients (30 yeux) ayant une lésion de la surface oculaire causée par le paraquat. Méthodes : Parmi les patients, 8 ont subi une greffe de membrane amniotique modifiée (GMAM), 14 ont reçu une greffe de membrane amniotique classique (GMA) et 8 ont été traités par pharmacothérapie. Ont été consignés les détails relatifs aux lésions, le délai de rétablissement de l’épithélium cornéen, l’acuité visuelle, la présence d’une opacification du stroma et les complications. Résultats : Dans les groupes GMAM et GMA, tous les yeux ont retrouvé leur niveau d’acuité visuelle prélésionnel; dans le groupe traité par pharmacothérapie, 3 yeux (37,5 %) ont retrouvé leur acuité visuelle prélésionnelle. Le délai moyen de rétablissement de l’épithélium cornéen était de 7,6 ± 2,7 jours pour le groupe GMAM, de 9,8 ± 3,6 jours pour le groupe GMA et de 18,2 ± 5,2 jour pour le groupe traité par pharmacothérapie (p o 0,05). Les trois groupes affichaient des taux de symblépharon significativement différents après le traitement (GMAM: 0 %, GMA : 35,7 %, pharmacothérapie : 87,5 %) (p o 0,05). La sécrétion lacrymale a diminué pour tous les groupes, mais elle était beaucoup plus faible pour le groupe traité par pharmacothérapie que pour les groupes GMAM et GMA (p o 0,05). Conclusions : Les lésions oculaires causées par le paraquat peuvent entraîner l’atteinte de toute la surface oculaire. Réalisée sans délai, la GMAM peut favoriser la réparation de la surface oculaire et réduire les complications associées au symblépharon.
Paraquat, a fast-acting sterilant herbicide widely used in agriculture, is among the most common causes of chemical eye injury. After a splashing incident, paraquat can cause severe ocular damage, including epithelial defect, pannus, symblepharon, ankyloblepharon, trichiasis, entropion, and punctual stenosis.1–4 Two types of treatment, medication2,5 and amniotic membrane transplantation (AMT), have been previously reported.5–7 AMT treatment can shorten the time required for epithelial defect closure compared with medication treatment, but it cannot prevent symblepharon and ankyloblepharon in severe cases. In this study, we observed a total conjunctival pseudomembrane at the acute phase of damage. When the pseudomembrane detached, symblepharon gradually formed
after medication or AMT treatment. Therefore, we carried out modified AMT (cornea, fornix, and palpebral conjunctiva total ocular surface AMT) to treat paraquat-induced ocular injury. Our data showed that MAMT treatment can effectively promote the repair of the ocular surface and reduce the complications from symblepharon.
From the Shandong Eye Hospital, Shandong Eye Institute, Shandong Academy of Medical Sciences, Qingdao, China
Can J Ophthalmol 2015;50:461–465 0008-4182/15/$-see front matter & 2015 Canadian Ophthalmological Society. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcjo.2015.08.002
Originally received Jun. 14, 2014. Final revision Jul. 22, 2015. Accepted Aug. 11, 2015
METHODS Patients
Thirty patients (30 eyes) with acute, severe chemical eye injury caused by paraquat were retrospectively studied at Shandong Eye Hospital from January 2009 to December 2012. The diagnosis was made on the basis of the clinical
Correspondence to Weiyun Shi, MD, Shandong Eye Institute, 5 Yanerdao Road, Qingdao 266071, China; [email protected] CAN J OPHTHALMOL — VOL. 50, NO. 6, DECEMBER 2015
461
AMT for paraquat-induced injury—Wang et al. features of paraquat-induced ocular chemical injuries. All patients suffered corneal epithelial defect and conjunctival pseudomembrane. All patients received drug therapy (DT). In addition, 8 patients (8 eyes) also underwent modified amniotic membrane transplantation (MAMT; cornea, bulbar conjunctiva, palpebral conjunctiva, and fornix AMT) after January 2012 (the MAMT group). Fourteen patients (14 eyes) also received conventional AMT treatment between March 2010 and December 2011 (the AMT group), and 8 patients (8 eyes) received only DT treatment between January 2009 and February 2010 (the DT group). This study was approved by the Ethics Committee of Shandong Eye Institute. Surgical techniques and medication treatment
The amniotic membrane was tiled on the surface of the cornea so that the conjunctiva with the epithelial surface was facing up. An amniotic membrane patch was sutured onto the surface to cover the entire cornea with continuous 10–0 nylon sutures placed 1 and 5 mm away from the corneal limbus.
Amniotic membrane transplantation surgical technique.
membrane, the dosage and the frequency of drug usage could be gradually reduced, and the sutures could be removed at the surgeon’s discretion. In the DT group, TobraDEX eye drops and artificial tears were also used 4 times daily. Glucocorticoids were given to reduce inflammation and promote the recovery. Observational index and statistical analysis
Slit-lamp microscopy was used preoperatively to observe the features of the damage to the eyelid margin, conjunctiva, and cornea. Visual acuity, ocular surface recovery, and the Schirmer test were recorded. The Statistical Package for Social Science 17.0 was used for statistical analysis. To rule out observation bias, we performed comparisons among the MAMT group, the AMT group, and the DT group with respect to general conditions such as age, sex, and the time of seeking medical advice. The significance of the differences was determined by the χ2 test. Categorical data were compared with the Fisher exact test. Differences were considered statistically significant if p o 0.05.
RESULTS Modified amniotic membrane transplantation surgical technique.
The pseudomembrane was removed with the use of a surgical microscope. The congestive conjunctiva was exposed, and the symblepharon was separated. The amniotic membrane was tilted on the surface of the cornea so that the conjunctiva with the epithelial surface was facing up. A single 10–0 running nylon suture was used to fasten a circle 1 and 5 mm away from the corneal limbus. After the upper and lower eyelids were opened, interrupted 10–0 nylon sutures were used to secure the amniotic membrane near the fornix. Finally, the amniotic membrane was secured to the upper and lower palpebral conjunctiva with 10–0 nylon sutures using a variable combination of interrupted and running suture techniques depending on the surgeon’s preference (Fig. 1). TobraDex eye drops (Alcon, Fort Worth, Tex.) and artificial tears were used 4 times daily after surgery. After the dissolution of the amniotic Postoperative care.
Among the 30 patients (30 eyes) with paraquat-induced ocular surface injury, 21 were males and 9 were females. There was no difference among the 3 groups in demographic features, the length of the injury history, and the size of epithelial defect (Table 1). The features of the damage at the acute phase were as follows: (A) palpebral edema, hyperemia, and obstruction of meibomian gland orifices; (B) conjunctiva congestion and edema, and conjunctiva (including palpebral conjunctiva, bulbar conjunctiva, and fornix conjunctiva) covered with pseudomembrane; and (C) corneal epithelial defect with cornea edema (Fig. 2). Treatment outcome
In all cases, the epithelial defect was fully recovered within the follow-up period. The average healing time was the shortest in the MAMT group compared with the other 2 groups (p o 0.05). The incidence of haze was lower in
Fig. 1 — A, Amniotic membrane was tiled on the surface of the cornea. B and C, The upper and lower eyelids were opened, and 10–0 nylon interrupted sutures were used to secure the amniotic membrane near the fornix to the upper and lower palpebral conjunctiva.
462
CAN J OPHTHALMOL — VOL. 50, NO. 6, DECEMBER 2015
AMT for paraquat-induced injury—Wang et al. Table 1—Demographic data p
Mean Value Group Patients, n Male Female Mean age (SD), y Mean time of seeking medical advice (SD), days Mean size of epithelial defect (SD), mm Initial best corrected visual acuity
MAMT
AMT
DT
5 3 49.00 (7.84) 4.50 (3.46) 7.25 (0.71) 0.24 (0.17)
10 4 49.29 (5.76) 5.29 (4.01) 6.86 (0.77) 0.22 (0.13)
6 2 48.00 (5.26) 5.00 (3.78) 6.63 (1.06) 0.26 (0.18)
MAMT to AMT
MAMT to DT
AMT to DT
0.51
0.5
0.61
0.92 0.205 0.25 0.77
0.77 0.217 0.19 0.83
0.61 0.751 0.56 0.57
MAMT, modified amniotic membrane transplantation; AMT, amniotic membrane transplantation; DT, drug therapy.
the MAMT and AMT groups compared with the DT group. The tear secretion was reduced in all groups, and it was significantly lower in the DT group compared with the other 2 groups (p o 0.05). Symblepharon did not occur in the MAMT group during the 3-month follow-up period, and the conjunctiva was smooth and flat. The rate of symblepharon was the lowest in the MAMT group (p o 0.05) (Fig. 3). The final best corrected visual acuity in the MAMT and AMT groups was higher than that in the DT group (p o 0.05). Visual acuity improved in all eyes, except for 2 eyes with severe haze in the DT group (Table 2). In the MAMT group, partial meibomian gland atresia developed in 2 eyes, and punctal stenosis developed in 1 eye, which underwent lacrimal intubation. Three patients from the AMT group suffered meibomian gland atresia, and 3 patients from the DT group suffered partial meibomian gland atresia.
DISCUSSION Paraquat toxicity is caused by the production of superoxide radicals, which have strong corrosive effects on the conjunctiva and other mucosal tissues. They can cause mucosal congestion, hemorrhage, edema, hyaline membrane formation, degeneration, hyperplasia, and fibrosis.3,8,9 Within the cells, the main toxicologic effect of paraquat is caused by oxygen-free radicals, which affect
cell metabolism through the depletion of NADPH.2 Ocular toxicity may also be associated with surfactants that can reduce the effectiveness of lavage, aid the spread of the agent on the ocular surface, and promote absorption.3 It has been previously reported that poor outcomes are associated with serious paraquat injuries, such as chronic ocular surface inflammation, conjunctivalization of the cornea, haze, and symblepharon. Medications such as corticosteroids, antibiotics, lubricants, and serum eye drops can be used topically to reduce inflammation, prevent infection, and promote re-epithelialization. Moreover, Yoon et al.10 investigated the therapeutic efficacy of AMT and found that it can shorten the epithelial defect closure time compared with medication treatment. However, AMT treatment could not prevent symblepharon. From 2009 to 2010, patients with serious paraquat injuries were treated with medications in our hospital; AMT was used to treat patients from 2010 to 2011. After 2012, patients were treated with MAMT by covering the whole ocular surface with the amniotic membrane. We found that MAMT treatment can effectively promote the repair of the ocular surface. The average epithelial defect healing time was the shortest in the MAMT group. This may be attributed to the antiangiogenic, antiscarring, and anti-inflammatory properties of the amniotic membrane.11–14 MAMT treatment can reduce inflammation and promote wound
Fig. 2 — A–C, Features of paraquat-induced ocular surface injury: (A) palpebral edema, hyperemia, and obstruction of meibomian gland orifices; (B) conjunctival congestion and edema, and whole conjunctiva (including palpebral conjunctiva, bulbar conjunctiva, and fornix conjunctiva) covered with pseudomembrane; and (C) corneal epithelial defect with corneal edema. CAN J OPHTHALMOL — VOL. 50, NO. 6, DECEMBER 2015
463
AMT for paraquat-induced injury—Wang et al.
Fig. 3 — A, Patients treated with drug therapy only developed severe symblepharon. B, Patients treated with amniotic membrane transplantation developed symblepharon. C, Symblepharon did not occur in patients treated with modified amniotic membrane transplantation.
flat and smooth planting bed for the proliferation and differentiation of residual conjunctival epithelium, thus leading to the rebuilding of the conjunctival surface and the effective prevention of symblepharon and scarring.14 Consistent with these features, the rate of symblepharon was the lowest in the MAMT group. We also found that severe paraquat-induced ocular injury is associated with the pseudomembrane. This can be easily confused with acute pseudomembranous conjunctivitis. Three patients referred to our hospital had been misdiagnosed with acute pseudomembranous conjunctivitis and treated with antibiotics only. The pseudomembrane of the patients with paraquat-induced ocular chemical injuries is found throughout the palpebral conjunctiva, bulbar conjunctiva, and fornix conjunctiva, whereas the pseudomembrane of acute conjunctivitis is restricted to the palpebral conjunctiva. Use of antiinfection treatment alone will not suffice for patients with paraquat-induced ocular chemical injuries. However, there are a few limitations in our study. In this retrospective study, different treatments, from medication to surgery, were selected over time. In addition, an observation bias may exist because patients were not randomly assigned for treatment. Further studies are needed with a randomized design, a larger sample size, and long-term follow-up. In conclusion, paraquat-induced ocular chemical injuries may cause damage to the whole ocular surface. MAMT treatment in a timely manner can effectively promote the repair of the ocular surface and reduce complications from symblepharon.
healing of the whole ocular surface, thus leading to rapid epithelial healing. Moreover, the amniotic membrane and quick epithelial healing can suppress the haze after injury. Connon et al.15 reported that the amniotic membrane used for ocular surface reconstruction can affect corneal transparency. In our study, the incidence of haze was lower in the MAMT and AMT groups than in the DT group. Increased haze in the DT group also led to the least final best corrected visual acuity as compared with the other 2 groups. Tear secretion has been reported to be decreased in cases with chemical injuries. In this study, which focused on paraquat-induced chemical injury, tear secretion was reduced in all groups. However, tear function was significantly higher in the MAMT group. It has previously been reported that treating a chemical burn with AMT can stop ulceration, promote corneal epithelialization, and reduce symblepharon. However, AMT treatment cannot release the symblepharon completely when it has already been formed.14–16 In this study, we observed a total conjunctiva pseudomembrane with congestion at the severe acute phase. When the pseudomembrane detached, the symblepharon gradually formed in the DT and AMT groups. To solve this problem, we modified the surgical technique by removing pseudomembrane with the aid of a surgical microscope, exposing the congestive and bleeding conjunctiva, separating the symblepharon, and tilting the amniotic membrane on the whole ocular surface, covering the cornea, bulbar conjunctiva, palpebral conjunctiva, and fornix. The amniotic membrane not only covered and repaired the conjunctival wound induced by paraquat but also provided a Table 2—Results after treatment at 3 months
p Group
MAMT
Mean healing time of corneal epithelial defect (SD), days Patients with haze, n (%) Schirmer test (SD), mm Patients with symblepharon, n (%) Mean final best corrected visual acuity (SD)
7.63 0/8 4.75 0/8 0.85
(2.00) (0%) (1.17) (0%) (0.09)
AMT 10.14 2/14 3.50 7/14 0.78
(2.07) (14.3%) (1.22) (50.0%) (0.19)
DT 19.63 5/8 2.31 7/8 0.59
MAMT, modified amniotic membrane transplantation; AMT, amniotic membrane transplantation; DT, drug therapy.
464
CAN J OPHTHALMOL — VOL. 50, NO. 6, DECEMBER 2015
(5.40) (62.5%) (0.84) (87.5%) (0.20)
MAMT to AMT
MAMT to DT
AMT to DT
0 0.52 0.03 0.02 0.33
0 0.03 0 0.001 0.04
0 0.05 0.03 0.17 0.04
AMT for paraquat-induced injury—Wang et al.
Disclosure: The authors have no proprietary or commercial interest in any materials discussed in this article. Supported by: This work was supported by the National Natural Science Foundation of China, Beijing, China (grants 81170815, 81200726); the National Basic Research Program of China, Beijing, China (grant 2013CB967004); Shandong Province Science and Technology Department of Science and Technology Development Plan, Jinan, China (grant 2013GSF11816); and Qingdao Science and Technology Plan, Qingdao, China (13-1-4-240-jch). The authors thank Ms. Ping Lin for her editorial assistance.
REFERENCES 1. Cant JS, Lewis DR. Ocular damage due to paraquat and diquat. Br Med J. 1968;2:224. 2. McKeag D, Maini R, Taylor HR. The ocular surface toxicity of paraquat. Br J Ophthalmol. 2002;86:350-62. 3. Vlahos K, Goggin M, Coster D. Paraquat causes chronic ocular surface toxicity. Aust N Z J Ophthalmol. 1993;21:187-90. 4. Yoon KC, Im SK, Park YG, et al. Application of umbilical cord serum eyedrops for the treatment of dry eye syndrome. Cornea. 2006;25:268-72. 5. Gomes JA, Romano A, Santos MS, et al. Amniotic membrane use in ophthalmology. Curr Opin Ophthalmol. 2005;16:233-40.
6. Dua HS, Gomes JA, King AJ, et al. The amniotic membrane in ophthalmology. Surv Ophthalmol. 2004;49:51-77. 7. Fernandes M, Sridhar MS, Sangwan VS, et al. Amniotic membrane transplantation for ocular surface reconstruction. Cornea. 2005;24: 643-53. 8. Nordquist RE, Nguyen H, Poyer JL, et al. The role of free radicals in paraquat-induced corneal lesions. Free Radic Res. 1995;23: 61-71. 9. Yamamoto I, Saito T, Harunari N, et al. Correlating the severity of paraquat poisoning with specific hemodynamic and oxygen metabolism variables. Crit Care Med. 2000;28:1877-83. 10. Yoon KC, Im SK, Kim JC, Yoon KW, Choi SK. Prognosis of paraquat-induced ocular surface injury: therapeutic effect of amniotic membrane transplantation. Cornea. 2009;28:520-3. 11. Shimmura S, Shimazaki S, Ohashi Y, et al. Antiinflammatory effects of amniotic membrane transplantation in ocular disorders surface. Cornea. 2001;20:408-13. 12. Tseng SCG, Espana EM, Kawakita T, et al. How does amniotic membrane word. Ocul Surf. 2004;2:177-8. 13. Ma DHK, Wang SF, Su WY, et al. Amniotic membrane graft for the management of sclera melting and corneal perforation in recalcitrant infectious scleral and corneoscleral ulcers. Cornea. 2002;21:275-83. 14. Tandon R, Gupta N, Kalaivani M, et al. Amniotic membrane transplantation as an adjunct to medical therapy in acute ocular burns. Br J Ophthalmol. 2011;95:199-204. 15. Connon CJ, Doutch J, Chen B, et al. The variation in transparency of amniotic membrane used in ocular surface regeneration. Br J Ophthalmol. 2010;94:1057-61. 16. Iakimenko SA, Buznyk OI, Rymgayllo-Jankowska B. Amniotic membrane transplantation in treatment of persistent corneal ulceration after severe chemical and thermal eye injuries. Eur J Ophthalmol. 2013;23:496-503.
CAN J OPHTHALMOL — VOL. 50, NO. 6, DECEMBER 2015
465
Total ocular surface amniotic membrane transplantation for paraquat-induced ocular surface injury Ting Wang, MD, PhD, Changsen Liang, MD, Xiaohan Xu, MD, Weiyun Shi, MD, PhD ABSTRACT ● RÉSUMÉ Objective: To evaluate the therapeutic efficacy of modified amniotic membrane transplantation (MAMT) for paraquat-induced ocular surface injury. Design: Retrospective case series. Participants: Thirty patients (30 eyes) with paraquat-induced ocular surface injury. Methods: Among the patients, 8 underwent MAMT, 14 received conventional amniotic membrane transplantation (AMT), and 8 were treated with simple drug therapy (DT). Features related to the damage, corneal epithelial defect closure time, visual acuity, stromal haze, and complications were recorded. Results: In the MAMT and AMT groups, visual acuity in all eyes recovered to the preinjury level; in the DT group, visual acuity in 3 eyes (37.5%) recovered to the preinjury level. The mean corneal epithelial defect closure time was 7.6 ⫾ 2.7 days in the MAMT group, 9.8 ⫾ 3.6 days in the AMT group, and 18.2 ⫾ 5.2 days in the DT group (p o 0.05). There was a significant difference in the symblepharon rate after treatment among the 3 groups (MAMT: 0%, AMT: 35.7%, DT: 87.5%; p o 0.05). Although the tear secretion was reduced in all groups, it was significantly lower in the DT group compared with the MAMT and AMT groups (p o 0.05). Conclusions: Paraquat-induced ocular injuries can lead to whole ocular surface damage. MAMT treatment in a timely manner can effectively promote the repair of the ocular surface and reduce the complications from symblepharon. Objet : Évaluer l’efficacité thérapeutique de la greffe de membrane amniotique modifiée (GMAM) dans des cas de lésion de la surface oculaire causée par le paraquat. Nature : Étude de cas rétrospective. Participants : Trente patients (30 yeux) ayant une lésion de la surface oculaire causée par le paraquat. Méthodes : Parmi les patients, 8 ont subi une greffe de membrane amniotique modifiée (GMAM), 14 ont reçu une greffe de membrane amniotique classique (GMA) et 8 ont été traités par pharmacothérapie. Ont été consignés les détails relatifs aux lésions, le délai de rétablissement de l’épithélium cornéen, l’acuité visuelle, la présence d’une opacification du stroma et les complications. Résultats : Dans les groupes GMAM et GMA, tous les yeux ont retrouvé leur niveau d’acuité visuelle prélésionnel; dans le groupe traité par pharmacothérapie, 3 yeux (37,5 %) ont retrouvé leur acuité visuelle prélésionnelle. Le délai moyen de rétablissement de l’épithélium cornéen était de 7,6 ± 2,7 jours pour le groupe GMAM, de 9,8 ± 3,6 jours pour le groupe GMA et de 18,2 ± 5,2 jour pour le groupe traité par pharmacothérapie (p o 0,05). Les trois groupes affichaient des taux de symblépharon significativement différents après le traitement (GMAM: 0 %, GMA : 35,7 %, pharmacothérapie : 87,5 %) (p o 0,05). La sécrétion lacrymale a diminué pour tous les groupes, mais elle était beaucoup plus faible pour le groupe traité par pharmacothérapie que pour les groupes GMAM et GMA (p o 0,05). Conclusions : Les lésions oculaires causées par le paraquat peuvent entraîner l’atteinte de toute la surface oculaire. Réalisée sans délai, la GMAM peut favoriser la réparation de la surface oculaire et réduire les complications associées au symblépharon.
Paraquat, a fast-acting sterilant herbicide widely used in agriculture, is among the most common causes of chemical eye injury. After a splashing incident, paraquat can cause severe ocular damage, including epithelial defect, pannus, symblepharon, ankyloblepharon, trichiasis, entropion, and punctual stenosis.1–4 Two types of treatment, medication2,5 and amniotic membrane transplantation (AMT), have been previously reported.5–7 AMT treatment can shorten the time required for epithelial defect closure compared with medication treatment, but it cannot prevent symblepharon and ankyloblepharon in severe cases. In this study, we observed a total conjunctival pseudomembrane at the acute phase of damage. When the pseudomembrane detached, symblepharon gradually formed
after medication or AMT treatment. Therefore, we carried out modified AMT (cornea, fornix, and palpebral conjunctiva total ocular surface AMT) to treat paraquat-induced ocular injury. Our data showed that MAMT treatment can effectively promote the repair of the ocular surface and reduce the complications from symblepharon.
From the Shandong Eye Hospital, Shandong Eye Institute, Shandong Academy of Medical Sciences, Qingdao, China
Can J Ophthalmol 2015;50:461–465 0008-4182/15/$-see front matter & 2015 Canadian Ophthalmological Society. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcjo.2015.08.002
Originally received Jun. 14, 2014. Final revision Jul. 22, 2015. Accepted Aug. 11, 2015
METHODS Patients
Thirty patients (30 eyes) with acute, severe chemical eye injury caused by paraquat were retrospectively studied at Shandong Eye Hospital from January 2009 to December 2012. The diagnosis was made on the basis of the clinical
Correspondence to Weiyun Shi, MD, Shandong Eye Institute, 5 Yanerdao Road, Qingdao 266071, China; [email protected] CAN J OPHTHALMOL — VOL. 50, NO. 6, DECEMBER 2015
461
AMT for paraquat-induced injury—Wang et al. features of paraquat-induced ocular chemical injuries. All patients suffered corneal epithelial defect and conjunctival pseudomembrane. All patients received drug therapy (DT). In addition, 8 patients (8 eyes) also underwent modified amniotic membrane transplantation (MAMT; cornea, bulbar conjunctiva, palpebral conjunctiva, and fornix AMT) after January 2012 (the MAMT group). Fourteen patients (14 eyes) also received conventional AMT treatment between March 2010 and December 2011 (the AMT group), and 8 patients (8 eyes) received only DT treatment between January 2009 and February 2010 (the DT group). This study was approved by the Ethics Committee of Shandong Eye Institute. Surgical techniques and medication treatment
The amniotic membrane was tiled on the surface of the cornea so that the conjunctiva with the epithelial surface was facing up. An amniotic membrane patch was sutured onto the surface to cover the entire cornea with continuous 10–0 nylon sutures placed 1 and 5 mm away from the corneal limbus.
Amniotic membrane transplantation surgical technique.
membrane, the dosage and the frequency of drug usage could be gradually reduced, and the sutures could be removed at the surgeon’s discretion. In the DT group, TobraDEX eye drops and artificial tears were also used 4 times daily. Glucocorticoids were given to reduce inflammation and promote the recovery. Observational index and statistical analysis
Slit-lamp microscopy was used preoperatively to observe the features of the damage to the eyelid margin, conjunctiva, and cornea. Visual acuity, ocular surface recovery, and the Schirmer test were recorded. The Statistical Package for Social Science 17.0 was used for statistical analysis. To rule out observation bias, we performed comparisons among the MAMT group, the AMT group, and the DT group with respect to general conditions such as age, sex, and the time of seeking medical advice. The significance of the differences was determined by the χ2 test. Categorical data were compared with the Fisher exact test. Differences were considered statistically significant if p o 0.05.
RESULTS Modified amniotic membrane transplantation surgical technique.
The pseudomembrane was removed with the use of a surgical microscope. The congestive conjunctiva was exposed, and the symblepharon was separated. The amniotic membrane was tilted on the surface of the cornea so that the conjunctiva with the epithelial surface was facing up. A single 10–0 running nylon suture was used to fasten a circle 1 and 5 mm away from the corneal limbus. After the upper and lower eyelids were opened, interrupted 10–0 nylon sutures were used to secure the amniotic membrane near the fornix. Finally, the amniotic membrane was secured to the upper and lower palpebral conjunctiva with 10–0 nylon sutures using a variable combination of interrupted and running suture techniques depending on the surgeon’s preference (Fig. 1). TobraDex eye drops (Alcon, Fort Worth, Tex.) and artificial tears were used 4 times daily after surgery. After the dissolution of the amniotic Postoperative care.
Among the 30 patients (30 eyes) with paraquat-induced ocular surface injury, 21 were males and 9 were females. There was no difference among the 3 groups in demographic features, the length of the injury history, and the size of epithelial defect (Table 1). The features of the damage at the acute phase were as follows: (A) palpebral edema, hyperemia, and obstruction of meibomian gland orifices; (B) conjunctiva congestion and edema, and conjunctiva (including palpebral conjunctiva, bulbar conjunctiva, and fornix conjunctiva) covered with pseudomembrane; and (C) corneal epithelial defect with cornea edema (Fig. 2). Treatment outcome
In all cases, the epithelial defect was fully recovered within the follow-up period. The average healing time was the shortest in the MAMT group compared with the other 2 groups (p o 0.05). The incidence of haze was lower in
Fig. 1 — A, Amniotic membrane was tiled on the surface of the cornea. B and C, The upper and lower eyelids were opened, and 10–0 nylon interrupted sutures were used to secure the amniotic membrane near the fornix to the upper and lower palpebral conjunctiva.
462
CAN J OPHTHALMOL — VOL. 50, NO. 6, DECEMBER 2015
AMT for paraquat-induced injury—Wang et al. Table 1—Demographic data p
Mean Value Group Patients, n Male Female Mean age (SD), y Mean time of seeking medical advice (SD), days Mean size of epithelial defect (SD), mm Initial best corrected visual acuity
MAMT
AMT
DT
5 3 49.00 (7.84) 4.50 (3.46) 7.25 (0.71) 0.24 (0.17)
10 4 49.29 (5.76) 5.29 (4.01) 6.86 (0.77) 0.22 (0.13)
6 2 48.00 (5.26) 5.00 (3.78) 6.63 (1.06) 0.26 (0.18)
MAMT to AMT
MAMT to DT
AMT to DT
0.51
0.5
0.61
0.92 0.205 0.25 0.77
0.77 0.217 0.19 0.83
0.61 0.751 0.56 0.57
MAMT, modified amniotic membrane transplantation; AMT, amniotic membrane transplantation; DT, drug therapy.
the MAMT and AMT groups compared with the DT group. The tear secretion was reduced in all groups, and it was significantly lower in the DT group compared with the other 2 groups (p o 0.05). Symblepharon did not occur in the MAMT group during the 3-month follow-up period, and the conjunctiva was smooth and flat. The rate of symblepharon was the lowest in the MAMT group (p o 0.05) (Fig. 3). The final best corrected visual acuity in the MAMT and AMT groups was higher than that in the DT group (p o 0.05). Visual acuity improved in all eyes, except for 2 eyes with severe haze in the DT group (Table 2). In the MAMT group, partial meibomian gland atresia developed in 2 eyes, and punctal stenosis developed in 1 eye, which underwent lacrimal intubation. Three patients from the AMT group suffered meibomian gland atresia, and 3 patients from the DT group suffered partial meibomian gland atresia.
DISCUSSION Paraquat toxicity is caused by the production of superoxide radicals, which have strong corrosive effects on the conjunctiva and other mucosal tissues. They can cause mucosal congestion, hemorrhage, edema, hyaline membrane formation, degeneration, hyperplasia, and fibrosis.3,8,9 Within the cells, the main toxicologic effect of paraquat is caused by oxygen-free radicals, which affect
cell metabolism through the depletion of NADPH.2 Ocular toxicity may also be associated with surfactants that can reduce the effectiveness of lavage, aid the spread of the agent on the ocular surface, and promote absorption.3 It has been previously reported that poor outcomes are associated with serious paraquat injuries, such as chronic ocular surface inflammation, conjunctivalization of the cornea, haze, and symblepharon. Medications such as corticosteroids, antibiotics, lubricants, and serum eye drops can be used topically to reduce inflammation, prevent infection, and promote re-epithelialization. Moreover, Yoon et al.10 investigated the therapeutic efficacy of AMT and found that it can shorten the epithelial defect closure time compared with medication treatment. However, AMT treatment could not prevent symblepharon. From 2009 to 2010, patients with serious paraquat injuries were treated with medications in our hospital; AMT was used to treat patients from 2010 to 2011. After 2012, patients were treated with MAMT by covering the whole ocular surface with the amniotic membrane. We found that MAMT treatment can effectively promote the repair of the ocular surface. The average epithelial defect healing time was the shortest in the MAMT group. This may be attributed to the antiangiogenic, antiscarring, and anti-inflammatory properties of the amniotic membrane.11–14 MAMT treatment can reduce inflammation and promote wound
Fig. 2 — A–C, Features of paraquat-induced ocular surface injury: (A) palpebral edema, hyperemia, and obstruction of meibomian gland orifices; (B) conjunctival congestion and edema, and whole conjunctiva (including palpebral conjunctiva, bulbar conjunctiva, and fornix conjunctiva) covered with pseudomembrane; and (C) corneal epithelial defect with corneal edema. CAN J OPHTHALMOL — VOL. 50, NO. 6, DECEMBER 2015
463
AMT for paraquat-induced injury—Wang et al.
Fig. 3 — A, Patients treated with drug therapy only developed severe symblepharon. B, Patients treated with amniotic membrane transplantation developed symblepharon. C, Symblepharon did not occur in patients treated with modified amniotic membrane transplantation.
flat and smooth planting bed for the proliferation and differentiation of residual conjunctival epithelium, thus leading to the rebuilding of the conjunctival surface and the effective prevention of symblepharon and scarring.14 Consistent with these features, the rate of symblepharon was the lowest in the MAMT group. We also found that severe paraquat-induced ocular injury is associated with the pseudomembrane. This can be easily confused with acute pseudomembranous conjunctivitis. Three patients referred to our hospital had been misdiagnosed with acute pseudomembranous conjunctivitis and treated with antibiotics only. The pseudomembrane of the patients with paraquat-induced ocular chemical injuries is found throughout the palpebral conjunctiva, bulbar conjunctiva, and fornix conjunctiva, whereas the pseudomembrane of acute conjunctivitis is restricted to the palpebral conjunctiva. Use of antiinfection treatment alone will not suffice for patients with paraquat-induced ocular chemical injuries. However, there are a few limitations in our study. In this retrospective study, different treatments, from medication to surgery, were selected over time. In addition, an observation bias may exist because patients were not randomly assigned for treatment. Further studies are needed with a randomized design, a larger sample size, and long-term follow-up. In conclusion, paraquat-induced ocular chemical injuries may cause damage to the whole ocular surface. MAMT treatment in a timely manner can effectively promote the repair of the ocular surface and reduce complications from symblepharon.
healing of the whole ocular surface, thus leading to rapid epithelial healing. Moreover, the amniotic membrane and quick epithelial healing can suppress the haze after injury. Connon et al.15 reported that the amniotic membrane used for ocular surface reconstruction can affect corneal transparency. In our study, the incidence of haze was lower in the MAMT and AMT groups than in the DT group. Increased haze in the DT group also led to the least final best corrected visual acuity as compared with the other 2 groups. Tear secretion has been reported to be decreased in cases with chemical injuries. In this study, which focused on paraquat-induced chemical injury, tear secretion was reduced in all groups. However, tear function was significantly higher in the MAMT group. It has previously been reported that treating a chemical burn with AMT can stop ulceration, promote corneal epithelialization, and reduce symblepharon. However, AMT treatment cannot release the symblepharon completely when it has already been formed.14–16 In this study, we observed a total conjunctiva pseudomembrane with congestion at the severe acute phase. When the pseudomembrane detached, the symblepharon gradually formed in the DT and AMT groups. To solve this problem, we modified the surgical technique by removing pseudomembrane with the aid of a surgical microscope, exposing the congestive and bleeding conjunctiva, separating the symblepharon, and tilting the amniotic membrane on the whole ocular surface, covering the cornea, bulbar conjunctiva, palpebral conjunctiva, and fornix. The amniotic membrane not only covered and repaired the conjunctival wound induced by paraquat but also provided a Table 2—Results after treatment at 3 months
p Group
MAMT
Mean healing time of corneal epithelial defect (SD), days Patients with haze, n (%) Schirmer test (SD), mm Patients with symblepharon, n (%) Mean final best corrected visual acuity (SD)
7.63 0/8 4.75 0/8 0.85
(2.00) (0%) (1.17) (0%) (0.09)
AMT 10.14 2/14 3.50 7/14 0.78
(2.07) (14.3%) (1.22) (50.0%) (0.19)
DT 19.63 5/8 2.31 7/8 0.59
MAMT, modified amniotic membrane transplantation; AMT, amniotic membrane transplantation; DT, drug therapy.
464
CAN J OPHTHALMOL — VOL. 50, NO. 6, DECEMBER 2015
(5.40) (62.5%) (0.84) (87.5%) (0.20)
MAMT to AMT
MAMT to DT
AMT to DT
0 0.52 0.03 0.02 0.33
0 0.03 0 0.001 0.04
0 0.05 0.03 0.17 0.04
AMT for paraquat-induced injury—Wang et al.
Disclosure: The authors have no proprietary or commercial interest in any materials discussed in this article. Supported by: This work was supported by the National Natural Science Foundation of China, Beijing, China (grants 81170815, 81200726); the National Basic Research Program of China, Beijing, China (grant 2013CB967004); Shandong Province Science and Technology Department of Science and Technology Development Plan, Jinan, China (grant 2013GSF11816); and Qingdao Science and Technology Plan, Qingdao, China (13-1-4-240-jch). The authors thank Ms. Ping Lin for her editorial assistance.
REFERENCES 1. Cant JS, Lewis DR. Ocular damage due to paraquat and diquat. Br Med J. 1968;2:224. 2. McKeag D, Maini R, Taylor HR. The ocular surface toxicity of paraquat. Br J Ophthalmol. 2002;86:350-62. 3. Vlahos K, Goggin M, Coster D. Paraquat causes chronic ocular surface toxicity. Aust N Z J Ophthalmol. 1993;21:187-90. 4. Yoon KC, Im SK, Park YG, et al. Application of umbilical cord serum eyedrops for the treatment of dry eye syndrome. Cornea. 2006;25:268-72. 5. Gomes JA, Romano A, Santos MS, et al. Amniotic membrane use in ophthalmology. Curr Opin Ophthalmol. 2005;16:233-40.
6. Dua HS, Gomes JA, King AJ, et al. The amniotic membrane in ophthalmology. Surv Ophthalmol. 2004;49:51-77. 7. Fernandes M, Sridhar MS, Sangwan VS, et al. Amniotic membrane transplantation for ocular surface reconstruction. Cornea. 2005;24: 643-53. 8. Nordquist RE, Nguyen H, Poyer JL, et al. The role of free radicals in paraquat-induced corneal lesions. Free Radic Res. 1995;23: 61-71. 9. Yamamoto I, Saito T, Harunari N, et al. Correlating the severity of paraquat poisoning with specific hemodynamic and oxygen metabolism variables. Crit Care Med. 2000;28:1877-83. 10. Yoon KC, Im SK, Kim JC, Yoon KW, Choi SK. Prognosis of paraquat-induced ocular surface injury: therapeutic effect of amniotic membrane transplantation. Cornea. 2009;28:520-3. 11. Shimmura S, Shimazaki S, Ohashi Y, et al. Antiinflammatory effects of amniotic membrane transplantation in ocular disorders surface. Cornea. 2001;20:408-13. 12. Tseng SCG, Espana EM, Kawakita T, et al. How does amniotic membrane word. Ocul Surf. 2004;2:177-8. 13. Ma DHK, Wang SF, Su WY, et al. Amniotic membrane graft for the management of sclera melting and corneal perforation in recalcitrant infectious scleral and corneoscleral ulcers. Cornea. 2002;21:275-83. 14. Tandon R, Gupta N, Kalaivani M, et al. Amniotic membrane transplantation as an adjunct to medical therapy in acute ocular burns. Br J Ophthalmol. 2011;95:199-204. 15. Connon CJ, Doutch J, Chen B, et al. The variation in transparency of amniotic membrane used in ocular surface regeneration. Br J Ophthalmol. 2010;94:1057-61. 16. Iakimenko SA, Buznyk OI, Rymgayllo-Jankowska B. Amniotic membrane transplantation in treatment of persistent corneal ulceration after severe chemical and thermal eye injuries. Eur J Ophthalmol. 2013;23:496-503.
CAN J OPHTHALMOL — VOL. 50, NO. 6, DECEMBER 2015
465