- Email: [email protected]
Vitreous Incarceration Complicating Cataract Surgery
Vitreous Incarceration Complicating Cataract Surgery A Light and Electron Microscopic Study PETER J. McDONNELL, MD, ZENAIDA C. de la CRUZ, MS, W. RICHARD GREEN, MD
Abstract: Vitreous incarceration in the cataract wound may contribute to development of aphakic cystoid macular edema, vitreoretinal traction and retinal detachment, and corneal decompensation. Thirty-one eyes in which vitreous was incarcerated in the cataract wound were examined postmortem and the corneal, anterior segment, and vitreoretinal changes were reported. Light and electron microscopy specimens demonstrated migration of corneal endothelium onto the adherent vitreous with production of basement membrane (descemetization). Fibrous ingrowth was present in 84% of the eyes. Iridovitreal synechiae were seen in 87% of the eyes. Cystoid macular edema, present in six eyes (19%), was usually accompanied by retinal phlebitis and often by distortion of the pars plicata. Vitreoretinal traction was seen in four eyes (13%), retinal tears were seen in two eyes (6%), and one eye had a total retinal detachment. Preretinal membranes were present in five eyes (16%) without other macular pathology. Persistent cystoid macular edema appears to occur in a minority of eyes with vitreous incarcerated in the cataract wound. If a vitrectomy is to be performed, the surgeon should excise vitreous adherent to the iris and in the anterior vitreous cavity in addition to the vitreous in the region of the wound. [Key words: cataract extraction, cystoid macular edema, descemetization, endothelialization, histopathology, myoblastic differentiation, retinal detachment, vitreoretinal traction, vitreous incarceration.] Ophthalmology 93:247-253, 1986
Vitreous incarceration in the cataract wound may contribute to the development of corneal decompensation, vitreoretinal traction and retinal detachment. I Irvine2 described eyes having cystoid macular edema, photophobia, irritability, injection, reduced visual acuity, and formed vitreous adherent to the corneoscleral wound, often associated with a distorted pupil. Our study offers an assessment of the frequency of these complications in a series of eyes examined postmortem, with insight into the From the Eye Pathology Laboratory, Wilmer Ophthalmological Institute, The Johns Hopkins Medical Institutions, Baltimore. Reprint requests to W. Richard Green, MD, Eye Pathology Laboratory, Johns Hopkins Hospital, 600 N. Wolfe Street, Baltimore, MD 21205.
pathophysiologic changes induced by incarcerated vitreous.
MATERIALS AND METHODS We examined the Wilmer Eye Pathology Laboratory records from January 1970 through December 1984 and identified 31 eyes of 27 autopsied patients with aphakic eyes with vitreous incarcerated within the cataract wound. The eyes were sectioned vertically, examined with a dissecting microscope, and processed and sectioned for light microscopy. Sections from the temporal callotte were obtained in each eye through the macula. Retinal trypsin digestions were performed in two eyes. 3 247
OPHTHALMOLOGY
•
FEBRUARY 1986
Table 1. Vitreous Incarceration in the Wound: Demographic Data Total Age (years) Sex Race Surgery-death interval (years) Diabetes
27 patients (31 eyes) Avg 76 (range, 56-93) Male 17, female 10 White 21, Black 5, Oriental 1 Avg 14 (range, 4-30) 1 (4%)
Table 2. Histopathologic Findings No. Eyes Corneal edema Stripped Descemet's membrane Iris incarceration Fibrous tissue ingrowth Endothelializationjdescemetization Iritis Iridovitreal synechiae Peripheral anterior synechiae Cyclitis Traction on pars plicata Vitritis Posterior vitreous detachment Traction on retina Retinal tear Retinal detachment Cystoid macular edema Phlebitis Preretinal membrane Background diabetic retinopathy Total
5
16
21
26 25
9
27
19 11
5 6
31 4
2 1 6 5 5 2
(%)
16 52 68 84 81
29 87 61
35
16
19 100
13
6 3 19
16 16
6
31
Selected specimens for transmission electron microscopy were processed in the usual manner and 1 #m-thick sections were stained with paraphenylenediamine for phase-contrast microscopy.
RESULTS The demographic data for the data for the patients are presented in Table 1. Most patients were elderly and the majority were white males. The interval between surgery and death averaged 14 years. One patient had a known history of diabetes. The histopathologic findings in the 31 eyes with vitreous incarceration are recorded in Table 2. Five eyes showed corneal edema. This was most apparent superiorly. Other corneal changes relating to the cataract wound included stripping of a small portion of Descemet's membrane in 52% of the eyes, iris incarceration in 68%, and fibrous tissue ingrowth in 84%. The amount of fibrocellular tissue ingrowth into the attached vitreous varied from minimal (Figs 1, 2) to marked (Figs 3-5). In most eyes, corneal endothelial cells had migrated onto the surface of the adherent vitreous strands (Figs 10-13). Light and transmission electron mi248
•
VOLUME 93
•
NUMBER 2
croscopy demonstrated formation of a thick periodic acidSchiff-positive basement membrane (Figs 2, 11, 12A, 13A). The endothelial cells displayed occasional surface microvilli (Fig 11), and junctional complexes (Fig 13); some had subplasmalemmal aggregates of microfilaments with associated fusiform bodies (Figs 11, 12B). Light and transmission electron microscopy demonstrated the presence of spindle-shaped fibroblastic cells in the vitreous strands (Figs 6, 12A). Vitreous collagen fibrils, measuring about 100° in diameter, were present in all cases. In addition, larger collagen fibrils, measuring 200 to 250 A in diameter, were associated with the fibroblastic cells (Fig 13), apparently reflecting new collagen synthesis by those cells. In addition to incarceration in the cataract wound, anterior segment changes relating to the iris included: iritis, iridovitreal synechiae, and peripheral anterior synechiae. The peripheral anterior synechiae were marked in two eyes with extensive glaucomatous changes of the optic nerve head. Iridovitreal synechiae, often extensive, were sometimes associated with a lymphocytic infiltrate of the iris (Fig 3). A total of nine eyes had chronic inflammatory infiltrates (usually mild) within the iris. In four eyes, iris pigment epithelium extended onto the incarcerated vitreous for short distances. There were 11 eyes (35%) with similar inflammatory infiltrates within the ciliary body. Frequently, the infiltrates were centered in the region of the larger arterioles and venules in the anterior ciliary body and iris root. Five eyes had distortion with anterior displacement of the ciliary processes by vitreous that was incarcerated in the cataract wound (Fig 7). Posterior vitreous detachment was present in the 31 eyes. The vitreous detachment extended to or was anterior to the equator in 30 eyes and in one eye to a point posterior to the equator. Inflammatory cells were identified within the vitreous in six eyes (19%). Typically, vitritis was minimal, but two eyes had moderate inflammation including the presence oflymphocytes and occasional macrophages on the internal aspect of the internal limiting membrane in the posterior pole area. Retinal findings included vitreoretinal traction in four eyes (Fig 8), with tears in two eyes and a total retinal detachment in one eye (Fig 4). Six eyes (19%) had multiple small cystoid spaces in the outer plexiform layer and occasionally in the inner nuclear layer of the macula (Fig 9). Eosinophilic transudate was often present within these cystic spaces. In no instance could any apparent vitreo~ macular adhesions be discerned. Of the six eyes with CME, five showed retinal phlebitis on histologic sections and/or trypsin digestion preparations. The infiltrates were focal, and consisted primarily of lymphocytes with occasional plasma cells. Capillaries and arterioles did not show involvement. Five eyes showed delicate unicellular, or occasionally two-ceIl-layers thick preretinal membranes, associated with fine folds of the internal limiting membrane in the macular region.
McDONNELL, et
al • VITREOUS INCARCERATION
Fig I. Top left E.P. 56790. Low-power photomicrograph demon trating vitreou strand e tending to limbal wound. lrido~treal adh ion has resulted in traction on pupillary margin of iri (periodic acid- hiff, 10). Fig 2. Top nglll .P. 56790. Higher power of itreou trand extending to limbal wound. Thick periodic acidchiff-po iti e ba ement membrane (descemetization) h been produced by migrating endothelial cell (periodic acid- hiff, X40). Fig 3. econd row left. .P. 56790. lridovitreal ynechia at pupillary margin ofiri asociated with moderate I mphocytic infiltrate of iri (periodic acid- hiff, 40). Fig 4. I!cond rOIl nghl .P. 56220. ro appearance of cye ith \ itreou incarceration in cataract ~ound . Fibrou ingrowth and traction rctinal detachment re ulted. Fig 5. Tlurd roll' left E.P. 56220. Photomicrograph demon tratc fibrous ingrowth into vitreou trand (periodic acid- hiff, X I0). Fig 6. Tlurd row nglll .P. 56220. Higher power of itreou trand demon trate pindle- hapcd fibrobla tic cell within the vitreou (periodic acid-Schiff, X 160). Fig 7. FOurlh rOIl' left. .P. 33079. Vitrcou incarceratortion of ciliary proce which are di placed anteriorl toward the uperior limbal wound (periodic acidSchiff, X5). Fig 8, Fourtll row "glll .P. 393 6. itreoretinal traction on inferior peripheral retina with hole formation in eye with vitreou incarceration (hemato ylin-eo in, X 10). Fig 9, BOl10m left. E.P. 57133. ti pac in outer plexiform and inner nuclear layers of parafo cal retina in eye with itreou incarceration and macular edema (hematoxylineo in. XIOO).
249
OPHTHALMOLOGY
•
FEBRUARY 1986
Fig 10. E.P. 51445. A, gross photograph shows thin strand of vitreous (arrowhead) extending to limbal wound. Traction on iris has resulted in peaked pupil and ectropion uveae. B, phase-contrast photomicrograph of inner aspect of cataract wound shows markedly thinned, irregularly spaced endothelial cells (arrowheads) lining Descemet's membrane (asterisk) and vitreous strand (arrow) (paraphenylenediamine, X320).
DISCUSSION In normal healing oflimbal cataract wounds, bridging of the inner aspect of the wound by endothelium is accomplished within two weeks after the surgical procedure. This is thought to confine fibroblast migration to the stromal defect, even when there is gaping of the inner edges. 4 Histopathologic study has demonstrated regeneration of Descemet's membrane by the endothelial cells as early as four months postoperatively.s The presence of vitreous in the wound prevents this key reparative step from occurring, and may allow fibroblastic cellular proliferation into the vitreous strand. Rather than restoring a continuous endothelial monolayer, the endothelial cells instead migrate onto the surface of the vitreous strand. In time, this may lead to exuberant production of basement membrane (descemetization) onto the surface of vitreous. 250
•
VOLUME 93
•
NUMBER 2
Fig 11. E.P. 51447. Transmission electron micrograph shows endothelial cells lining both sides of 5-micron-thick vitreous strand. Endothelial cells have surface microvilli (arrowheads and inset) and subplasmalemmal aggregates of microfilaments measuring 6 nm in diameter (circle and inset). Center of vitreous strand consists primarily of randomly oriented collagen fibrils measuring 100 Ain diameter. A zone ofloosely compacted. multilaminated basement membrane-like material (arrows) is present beneath the endothelium (X 17,400; inset, X60,000).
Corneal endothelial cell migration and possibly proliferation onto adjacent structures has been described in a number of conditions6 including: essential iris atrophy,7,8 iris-nevus syndrome,9-11 iris neovascularization due to long-standing retinal detachment, 12.13 central retinal vein occlusion,14 diabetic retinopathy;I3,14 post-contusion deformity of the anterior chamber angle,14-16 Chandler's syndrome;17 posterior polymorphous corneal dystrophy;18 anterior chamber intraocular lens implantation;19 after argon laser trabeculoplasty;20 in the late stages of aniridia; and with chronic iridocyclitis. Donaldson and Smith21 observed "tubular-appearing structures" clinically when vitreous strands were incarcerated in corneal surgical wounds, and postulated that a new Descemet's membrane may be forming on the strands. W olter2 demonstrated growth of endothelium onto vitreous in an eye that had been seriously traumatized. Thus, our finding of endothelialization and descemetization of vitreous strands incarcerated in the limbal cataract wound is not surprising,
McDONNELL, et al • VITREOUS INCARCERATION
B Fig 12. .P. 67 O. ,region of catara t wound with cut end of Dc cemet' membrane (D). corneal troma ( ). and vllreou trand ( ) e tending to "ound. lIenuatcd endothelial cellla)er (arrowheads) e tends onto Itreou and i iated with production of basemen I membrane (arrow). ibrobla I-like cell (). me of hich contain pigment granul • are present in the vltreou trand ( 3(00). B, higher pow rofvitreous trand demon trate endothelial cell on urfa e. ell contaIn dense ubpi malemmal aggregate of mlcrofilamenlS (between arrowhead) ( 60.(00).
but rather appears to represent a stereotyped response of the corneal endothelium to injury and irritation. This phenomenon explains the "glassy membrane" appearance of vitreous strands that are observed through biomicroscopic examination of strands of incarcerated vitreous present for several years. Endothelial cell migration may be one of the mechanisms accounting for continued loss of corneal endothelium and eventual corneal edema. This phenomenon should be considered by the surgeon at the time of anterior vitrectomy because the vitreous strand may be quite tough, reinforced by a thick basement membrane produced by the endothelium, and efforts to break the strand may result in undesired traction on the retina and possibly retinal breaks and detachment. Iliff23 noted the great tenacity of these strands and proposed using a bent 30-gauge needle to pull the vitreous strands through a corneal section and cutting them with a scissors. Norton and Machemer! noted that inferior retinal detachments may complicate vitreous incarceration. They stated that the "sheet of vitreous apparently shrinks, gradually pulling the inferior retina into the pupillary area." The presence of subplasmalemmal microfilaments in the endothelial cells which have migrated onto the vitreous strands is similar to those observed in endothelial cells on the surface of the iris in various conditionsY·17 This "myoblastic differentiation" presumably reflects contrac-
tile filaments which are capable of generating traction, much as fibroblasts generate traction during the woundhealing process.2 4 Similarly, the cells within preretinal membrane demonstrate such filaments, apparently accounting for the ability of the membranes to contract and distort the inner retinal surface and contribute to retinal detachment in cases of proliferative vitreoretinopathy.25 We propose that the migration of fibroblasts into, and endothelium onto, the adherent vitreous and the generation of traction by these cells, may contribute to the vitreoretinal traction and retinal detachments that occur as delayed complications of vitreous incarceration. The presence of fibrocytes in the incarcerated vitreous, presumably having migrated from the corneal stroma, would appear to have two primary clinical consequences. First, it would "fortify" the vitreous strand and make it more difficult to tear or sweep from the wound. Second, contraction of these cells would presumably generate tractional forces that might result in peripheral retinal breaks and retinal detachment. Thus, the retinal traction that occurs in such cases may not be due to "vitreous shrinkage,"! but rather, it may be partly a consequence oftractional forces generated by endothelial cells and fibroblasts. The pathogenesis of cystoid macular edema is unknown, but several clinicopathologic studies26- 28 suggest that ocular inflammation plays a role in cases of macular edema after cataract extraction. Vitreous aspirates may 251
OPHTHALMOLOGY
•
FEBRUARY 1986
*
•
VOLUME 93
•
NUMBER 2
Table 3. Eyes with Vitreous Incarceration and Cystoid Macular Edema No, Eyes/Patients
Diabetic retinopathy Iris incarceration Fibrous ingrowth Endothelialization/descemetization Iritis Iridovitreal synechiae Peripheral anterior synechiae CYClitis Ciliary body traction Retinal phlebitis Vitritis Vitreoretinal traction Total
Fig 13. E.P. 62128. A, endothelial cells. joined by junctional complexes (circle), line surface of vitreous strand. M ultilaminated basement membrane is seen beneath endothelial cell (between arrowheads). Fibroblastic cells (asterisks) are present in vitreous strand (X5700). B, high power of fibroblast within vitreous strand shows numerous 70 to 90 nm cytoplasmic filaments (asterisk) and collagen fibrils (arrowheads) that measure 200 A in diameter (X74,000).
show acute and chronic inflammatory cells. 29 It has been suggested that prostaglandins, produced in the anterior uvea, may be the mediators responsible for the retinal vascular leakage. 3D,31 Martin and co-workers,28 in a series of 152 eyes examined pathologically and analyzed statistically, found the presence of intraocular inflammation (iritis, cyclitis, vitritis, retinal phlebitis) to significantly correlate with the presence of macular edema. Among the various anterior segment operative complications, only iridovitreal synechiae correlated significantly with the presence of macular edema. This suggests that mechanical irritation of the iris may be of pathophysiological importance. Our results are consistent with previous observations. Of the six eyes with macular edema in our series, all had one or more foci of inflammation (Table 3). Two eyes of one patient had macular edema in the setting of background diabetic retinopathy. The remaining four eyes all had chronic inflammatory infiltrates ofthe iris and ciliary body. Two of the four eyes had vitritis and all four showed 252
2/2 5 6 5 5 6 5 5 4 5
2 2
6/5
phlebitis or periphlebitis in histologic section or retinal digest preparations. All six eyes had iridovitreal synechiae. Thus, our findings are consistent with the proposal that uveal inflammation results in the breakdown of the retinal blood-ocular barrier. We also found an additional histopathologic feature in the four eyes with macular edema without diabetic retinopathy. In those eyes, the pars plicata was markedly distorted with the ciliary processes displaced anteriorly. This may be a mechanical effect of the anteriorly displaced vitreous body. Flaxel32 examined an eye histopathologically 24 hours after cataract extraction and noted zonule entrapment in the limbal wound. His report does not state whether there was distortion of the pars plicata. To our knowledge, such distortion of the pars plicata has not been previously noted clinically or pathologically in association with aphakic cystoid macular edema. The ciliary epithelium absorbs prostaglandins produced by the anterior uvea. 33 It is unknown whether mechanical distortion of the pars plicata by anteriorly displaced vitreous may alter prostaglandin absorption. In a subset of eyes, strands of vitreous connecting the detached vitreous body with the macula results in chronic cystoid macular edema. Vitreous attachments were observed in a number of patients. 34-38 Fung38 reported that out of 130 eyes in a prospective study, 4 demonstrated the existence of a vitreous strand attached to the macula on biomicroscopic examination. In all eyes in our series, except for one, the vitreous was detached to or anterior to the equator. In the remaining eye, the vitreous detachment did not extend to the equator. In no eye could a persistent vitreomacular adhesion be discerned; only six eyes in our series had macular edema. Thus, vitreomacular traction appears to account for only a minority of the cases of aphakic cystoid macular edema. Anterior vitrectomy is of value in the management of selected cases of chronic aphakic cystoid macular edema. 39 The surgeon should be aware that almost all eyes with vitreous to the wound will have iridovitreal synechiae which may distort the pupil and be associated with chronic inflammatory cell infiltrates. Thus, in addition to excising
MCDONNELL, et al •
vitreous within the anterior chamber, it is advisable to cut all iridovitreal adhesions around the pupillary margin and behind the iris plane and thereby restore "normal aphakic anatomy,,40 to the eyes. The neodymium:YAG laser has been used to cut vitreous strands in eyes with vitreous incarceration in the corneoscleral wound and cystoid macular edema. 41 Bahn and co-workers42 were able to lyse a vascularized vitreous strand to cornea by the argon laser in one patient. This procedure does not address the iridovitreal adhesions which are present in the majority of these eyes. Because the laser study included patients with macular edema of short duration, and because cystoid macular edema resolves spontaneously in most eyes, it is not clear whether Y AG laser treatment improves the natural history of the condition. This is a relatively safe and noninvasive procedure. It may well prove to be valuable means of cutting the vitreous strand allowing the endothelium to "seal" the wound, limit fibroblastic ingrowth, and possibly avoid the later consequences of peripheral vitreoretinal traction.
REFERENCES 1. Norton EWD, Machemer R. New approach to the treatment of selected retinal detachments secondary to vitreous loss at cataract surgery. Am J Ophthalmol1971; 72:705-7. 2. Irvine SR. A newly defined vitreous syndrome following cataract surgery interpreted according to recent concepts of the structure of the vitreous. Am J Ophthalmol1953; 36:599-619. 3. Kuwabara T, Cogan DG. Studies of retinal vascular pattems. Part I. Normal architecture. Arch Ophthalmol 1960; 64:904-11. 4. Swan KC. Fibroblastic ingrowth following cataract extraction. Arch Ophthalmol1973; 89:445-9. 5. Heller MD, Irvine SR, Straatsma BR, Foos RY. Wound healing after cataract eltlraction and position of the vitreous in aphakic eyes as studied postmortem. Trans Am Ophthalmol Soc 1971; 69:245-62. 6. Green WR. The uveal tract. In: Spencer WH, ed. Ophthalmic Pathology; An Atlas and Textbook, 3d ed. Philadelphia: WB Saunders, 1985; 1448. 7. Heath P. Essential atrophy of the iris: a histopathologic study. Trans Am Ophthalmol Soc 1953; 51:167-92. 8. Scheie HG, Yanoff M, Kellogg WT. Essential iris atrophy; report of a case. Arch Ophthalmol1976; 94:1315-20. 9. Wolter JR, Makley TA Jr. Cogan-Reese syndrome; formation of a glass membrane on an iris nevus clinically simulating tumor growth. J Pediatr Ophthalmol Strabismus 1972; 9:102-5. 10. Scheie HG, Yanoff M. Iris nevus (Cogan-Reese) syndrome; a cause of unilateral glaucoma. Arch Ophthalmol1975; 93:963-70. 11. Cogan DG, Reese AB. A syndrome of iris nodules, ectopic Descemet's membrane, and unilateral glaucoma. Doc ~hthalmoI1969; 26:42433. 12. Wolter JR, Fechner PU. Glass membranes on the anterior iris surface. Am J Ophthalmol1962; 53:235-43. 13. Harris M, Tso AY, Kaba FW, et al. Comeal endothelial overgrowth of angle and iris; evidence of myoblastic differentiation in three cases. Ophthalmology 1984; 91: 1154-60. 14. Schulze RR. Rubeosis iridis. Am J Ophthalmol1967; 63:487-95. 15. Gartner S, Taffet S, Friedman AH. The association of rubeosis iridis with endothelialisation of the anterior chamber: report of a clinical case with histopathological review of 16 additional cases. Br J OphthalmoI1977; 61 :267-71. 16. Wolff SM, Zimmerman LE. Chronic secondary glaucoma associated with retrodisplacement of iris root and deepening of the anterior
VITREOUS INCARCERATION
chamber angle secondary to contusion. Am J Ophthalmol1962; 54: 547-63. 17. Patel A. Kenyon KR, Hirst LW, et al. Clinicopathologic features of Chandler's syndrome. Surv Ophthalmol1983; 27:327-44. 18. Rodrigues MM, Phelps CD, Krachmer JH, et al. Glaucoma due to endothelialization of the anterior chamber angle; a comparison of posterior polymorphous dystrophy of the comea and Chandler's syndrome. Arch Ophthalmol1980; 98:688-96. 19. McDonnell PJ, Green WR, Maumenee AE,lIiff WJ. Pathology of intraocular lenses in 33 eyes examined postmortem. Ophthalmology 1983; 90:386-403. 20. Rodrigues MM, Spaeth GL, Donohoo P. Electron microscopy of argon laser therapy in phakic open-angle glaucoma. Ophthalmology 1982; 89:198-210. 21. Donaldson DD, Smith TR. Descemet's membrane tubes. Trans Am Ophthalmol Soc 1966; 64:89-109. 22. Wolter JR. Proliferating comeal endothelium causing a vitreous membrane and incurable retinal detachment. J Pediatr Ophthalmol 1970; 7:162-6. 23. Iliff CEo Treatment of the vitreous-tug syndrome. Am J Ophthalmol 1966; 62:856-9. 24. Gabbiani G, Hirschel BJ, Ryan GB, et al. Granulation tissue as a contractile organ; a study of structure and function. J Exp Med 1972; 135:719-34. 25. Kampik A, Kenyon KR, Michels RG, et al. Epiretinal and vitreous membranes; comparative study of 56 cases. Arch Ophthalmol1981; 99: 1445-54. 26. Michels RG, Green WR, Maumenee AE. Cystoid macular edema following cataract extraction (the Irvine-Gass syndrome); a case studied clinically and histopathologically. Ophthalmic Surg 1971; 2:217-21. 27. Norton AL, Brown WJ, Carlson M, et al. Pathogenesis of aphakic macular edema. Am J Ophthalmol1975; 80:96-101. 28. Martin NF, Green WR, Martin LW. Retinal phlebitis in the Irvine-Gass syndrome. Am J Ophthalmol1977; 83:377-86. 29. Federman JL, Annesley WH Jr, Sarin LK, Remer P. Vitrectomy and cystoid macular edema. Ophthalmology 1980; 87:622-8. 30. Miyake K. Prevention of cystoid macular edema after lens extraction by topical indomethacin. (I) A preliminary report. Albrecht von Graefes Arch Klin Exp OphthalmoI1977; 203:81-8. 31. Miyake K. Prophylaxis of aphakic cystoid macular edema using topical indomethacin. Am Intraocullmplant Soc J 1978; 4:174-9. 32. Flaxel JT. Histology of cataract extractions. Arch Ophthalmol 1970; 83:436-44. 33. Hanna C, Sharp JD. Ocular absorption of indomethacin by the rabbit. Arch Ophthalmol1972; 88:196-8. 34. Ho PC, Tolentino FI. The role of vitreous in aphakic cystoid macular edema: a review. Am Intraocullmplant Soc J 1982; 8:258-64. 35. Maumenee AE. Further advances in the study of the macula. Arch Ophthalmol1967; 78:151-65. 36. Reese AB, Jones IS, Cooper WC. Macular changes secondary to vitreous traction. Trans Am Ophthalmol Soc 1966; 64:123-34. 37. Tolentino FI, Schepens CL. Edema of posterior pole after cataract extraction; a biomicroscopic study. Arch Ophthalmol 1965; 74:781-6. 38. Fung WE. The national, prospective, randomized vitrectomy study for chronic aphakic cystoid macular edema. Progress report and comparison between the control and non randomized groups. Surv Ophthalmol1984; 28:569-75. 39. Fung WE. Results of the vitrectomy/aphakic cystoid macular edema study. Abstract. Ophthalmology 1984; 91 (Suppl):67. 40. Fung WE. Anterior vitrectomy for chronic aphakic cystoid macular edema. Ophthalmology 1980; 87:189-93. 41. Katzen LE, Fleischman JA, Trokel S. YAG laser treatment of cystoid macular edema. Am J Ophthalmol1983; 95:589-92. 42. Bahn CF, Vine AJ, Wolter JR, Martonyi CL. Argon laser photocoagulation of a vitreo-comeal adhesion after trauma. Ophthalmic Surg 1982; 13:53-5.
253
Abstract: Vitreous incarceration in the cataract wound may contribute to development of aphakic cystoid macular edema, vitreoretinal traction and retinal detachment, and corneal decompensation. Thirty-one eyes in which vitreous was incarcerated in the cataract wound were examined postmortem and the corneal, anterior segment, and vitreoretinal changes were reported. Light and electron microscopy specimens demonstrated migration of corneal endothelium onto the adherent vitreous with production of basement membrane (descemetization). Fibrous ingrowth was present in 84% of the eyes. Iridovitreal synechiae were seen in 87% of the eyes. Cystoid macular edema, present in six eyes (19%), was usually accompanied by retinal phlebitis and often by distortion of the pars plicata. Vitreoretinal traction was seen in four eyes (13%), retinal tears were seen in two eyes (6%), and one eye had a total retinal detachment. Preretinal membranes were present in five eyes (16%) without other macular pathology. Persistent cystoid macular edema appears to occur in a minority of eyes with vitreous incarcerated in the cataract wound. If a vitrectomy is to be performed, the surgeon should excise vitreous adherent to the iris and in the anterior vitreous cavity in addition to the vitreous in the region of the wound. [Key words: cataract extraction, cystoid macular edema, descemetization, endothelialization, histopathology, myoblastic differentiation, retinal detachment, vitreoretinal traction, vitreous incarceration.] Ophthalmology 93:247-253, 1986
Vitreous incarceration in the cataract wound may contribute to the development of corneal decompensation, vitreoretinal traction and retinal detachment. I Irvine2 described eyes having cystoid macular edema, photophobia, irritability, injection, reduced visual acuity, and formed vitreous adherent to the corneoscleral wound, often associated with a distorted pupil. Our study offers an assessment of the frequency of these complications in a series of eyes examined postmortem, with insight into the From the Eye Pathology Laboratory, Wilmer Ophthalmological Institute, The Johns Hopkins Medical Institutions, Baltimore. Reprint requests to W. Richard Green, MD, Eye Pathology Laboratory, Johns Hopkins Hospital, 600 N. Wolfe Street, Baltimore, MD 21205.
pathophysiologic changes induced by incarcerated vitreous.
MATERIALS AND METHODS We examined the Wilmer Eye Pathology Laboratory records from January 1970 through December 1984 and identified 31 eyes of 27 autopsied patients with aphakic eyes with vitreous incarcerated within the cataract wound. The eyes were sectioned vertically, examined with a dissecting microscope, and processed and sectioned for light microscopy. Sections from the temporal callotte were obtained in each eye through the macula. Retinal trypsin digestions were performed in two eyes. 3 247
OPHTHALMOLOGY
•
FEBRUARY 1986
Table 1. Vitreous Incarceration in the Wound: Demographic Data Total Age (years) Sex Race Surgery-death interval (years) Diabetes
27 patients (31 eyes) Avg 76 (range, 56-93) Male 17, female 10 White 21, Black 5, Oriental 1 Avg 14 (range, 4-30) 1 (4%)
Table 2. Histopathologic Findings No. Eyes Corneal edema Stripped Descemet's membrane Iris incarceration Fibrous tissue ingrowth Endothelializationjdescemetization Iritis Iridovitreal synechiae Peripheral anterior synechiae Cyclitis Traction on pars plicata Vitritis Posterior vitreous detachment Traction on retina Retinal tear Retinal detachment Cystoid macular edema Phlebitis Preretinal membrane Background diabetic retinopathy Total
5
16
21
26 25
9
27
19 11
5 6
31 4
2 1 6 5 5 2
(%)
16 52 68 84 81
29 87 61
35
16
19 100
13
6 3 19
16 16
6
31
Selected specimens for transmission electron microscopy were processed in the usual manner and 1 #m-thick sections were stained with paraphenylenediamine for phase-contrast microscopy.
RESULTS The demographic data for the data for the patients are presented in Table 1. Most patients were elderly and the majority were white males. The interval between surgery and death averaged 14 years. One patient had a known history of diabetes. The histopathologic findings in the 31 eyes with vitreous incarceration are recorded in Table 2. Five eyes showed corneal edema. This was most apparent superiorly. Other corneal changes relating to the cataract wound included stripping of a small portion of Descemet's membrane in 52% of the eyes, iris incarceration in 68%, and fibrous tissue ingrowth in 84%. The amount of fibrocellular tissue ingrowth into the attached vitreous varied from minimal (Figs 1, 2) to marked (Figs 3-5). In most eyes, corneal endothelial cells had migrated onto the surface of the adherent vitreous strands (Figs 10-13). Light and transmission electron mi248
•
VOLUME 93
•
NUMBER 2
croscopy demonstrated formation of a thick periodic acidSchiff-positive basement membrane (Figs 2, 11, 12A, 13A). The endothelial cells displayed occasional surface microvilli (Fig 11), and junctional complexes (Fig 13); some had subplasmalemmal aggregates of microfilaments with associated fusiform bodies (Figs 11, 12B). Light and transmission electron microscopy demonstrated the presence of spindle-shaped fibroblastic cells in the vitreous strands (Figs 6, 12A). Vitreous collagen fibrils, measuring about 100° in diameter, were present in all cases. In addition, larger collagen fibrils, measuring 200 to 250 A in diameter, were associated with the fibroblastic cells (Fig 13), apparently reflecting new collagen synthesis by those cells. In addition to incarceration in the cataract wound, anterior segment changes relating to the iris included: iritis, iridovitreal synechiae, and peripheral anterior synechiae. The peripheral anterior synechiae were marked in two eyes with extensive glaucomatous changes of the optic nerve head. Iridovitreal synechiae, often extensive, were sometimes associated with a lymphocytic infiltrate of the iris (Fig 3). A total of nine eyes had chronic inflammatory infiltrates (usually mild) within the iris. In four eyes, iris pigment epithelium extended onto the incarcerated vitreous for short distances. There were 11 eyes (35%) with similar inflammatory infiltrates within the ciliary body. Frequently, the infiltrates were centered in the region of the larger arterioles and venules in the anterior ciliary body and iris root. Five eyes had distortion with anterior displacement of the ciliary processes by vitreous that was incarcerated in the cataract wound (Fig 7). Posterior vitreous detachment was present in the 31 eyes. The vitreous detachment extended to or was anterior to the equator in 30 eyes and in one eye to a point posterior to the equator. Inflammatory cells were identified within the vitreous in six eyes (19%). Typically, vitritis was minimal, but two eyes had moderate inflammation including the presence oflymphocytes and occasional macrophages on the internal aspect of the internal limiting membrane in the posterior pole area. Retinal findings included vitreoretinal traction in four eyes (Fig 8), with tears in two eyes and a total retinal detachment in one eye (Fig 4). Six eyes (19%) had multiple small cystoid spaces in the outer plexiform layer and occasionally in the inner nuclear layer of the macula (Fig 9). Eosinophilic transudate was often present within these cystic spaces. In no instance could any apparent vitreo~ macular adhesions be discerned. Of the six eyes with CME, five showed retinal phlebitis on histologic sections and/or trypsin digestion preparations. The infiltrates were focal, and consisted primarily of lymphocytes with occasional plasma cells. Capillaries and arterioles did not show involvement. Five eyes showed delicate unicellular, or occasionally two-ceIl-layers thick preretinal membranes, associated with fine folds of the internal limiting membrane in the macular region.
McDONNELL, et
al • VITREOUS INCARCERATION
Fig I. Top left E.P. 56790. Low-power photomicrograph demon trating vitreou strand e tending to limbal wound. lrido~treal adh ion has resulted in traction on pupillary margin of iri (periodic acid- hiff, 10). Fig 2. Top nglll .P. 56790. Higher power of itreou trand extending to limbal wound. Thick periodic acidchiff-po iti e ba ement membrane (descemetization) h been produced by migrating endothelial cell (periodic acid- hiff, X40). Fig 3. econd row left. .P. 56790. lridovitreal ynechia at pupillary margin ofiri asociated with moderate I mphocytic infiltrate of iri (periodic acid- hiff, 40). Fig 4. I!cond rOIl nghl .P. 56220. ro appearance of cye ith \ itreou incarceration in cataract ~ound . Fibrou ingrowth and traction rctinal detachment re ulted. Fig 5. Tlurd roll' left E.P. 56220. Photomicrograph demon tratc fibrous ingrowth into vitreou trand (periodic acid- hiff, X I0). Fig 6. Tlurd row nglll .P. 56220. Higher power of itreou trand demon trate pindle- hapcd fibrobla tic cell within the vitreou (periodic acid-Schiff, X 160). Fig 7. FOurlh rOIl' left. .P. 33079. Vitrcou incarceratortion of ciliary proce which are di placed anteriorl toward the uperior limbal wound (periodic acidSchiff, X5). Fig 8, Fourtll row "glll .P. 393 6. itreoretinal traction on inferior peripheral retina with hole formation in eye with vitreou incarceration (hemato ylin-eo in, X 10). Fig 9, BOl10m left. E.P. 57133. ti pac in outer plexiform and inner nuclear layers of parafo cal retina in eye with itreou incarceration and macular edema (hematoxylineo in. XIOO).
249
OPHTHALMOLOGY
•
FEBRUARY 1986
Fig 10. E.P. 51445. A, gross photograph shows thin strand of vitreous (arrowhead) extending to limbal wound. Traction on iris has resulted in peaked pupil and ectropion uveae. B, phase-contrast photomicrograph of inner aspect of cataract wound shows markedly thinned, irregularly spaced endothelial cells (arrowheads) lining Descemet's membrane (asterisk) and vitreous strand (arrow) (paraphenylenediamine, X320).
DISCUSSION In normal healing oflimbal cataract wounds, bridging of the inner aspect of the wound by endothelium is accomplished within two weeks after the surgical procedure. This is thought to confine fibroblast migration to the stromal defect, even when there is gaping of the inner edges. 4 Histopathologic study has demonstrated regeneration of Descemet's membrane by the endothelial cells as early as four months postoperatively.s The presence of vitreous in the wound prevents this key reparative step from occurring, and may allow fibroblastic cellular proliferation into the vitreous strand. Rather than restoring a continuous endothelial monolayer, the endothelial cells instead migrate onto the surface of the vitreous strand. In time, this may lead to exuberant production of basement membrane (descemetization) onto the surface of vitreous. 250
•
VOLUME 93
•
NUMBER 2
Fig 11. E.P. 51447. Transmission electron micrograph shows endothelial cells lining both sides of 5-micron-thick vitreous strand. Endothelial cells have surface microvilli (arrowheads and inset) and subplasmalemmal aggregates of microfilaments measuring 6 nm in diameter (circle and inset). Center of vitreous strand consists primarily of randomly oriented collagen fibrils measuring 100 Ain diameter. A zone ofloosely compacted. multilaminated basement membrane-like material (arrows) is present beneath the endothelium (X 17,400; inset, X60,000).
Corneal endothelial cell migration and possibly proliferation onto adjacent structures has been described in a number of conditions6 including: essential iris atrophy,7,8 iris-nevus syndrome,9-11 iris neovascularization due to long-standing retinal detachment, 12.13 central retinal vein occlusion,14 diabetic retinopathy;I3,14 post-contusion deformity of the anterior chamber angle,14-16 Chandler's syndrome;17 posterior polymorphous corneal dystrophy;18 anterior chamber intraocular lens implantation;19 after argon laser trabeculoplasty;20 in the late stages of aniridia; and with chronic iridocyclitis. Donaldson and Smith21 observed "tubular-appearing structures" clinically when vitreous strands were incarcerated in corneal surgical wounds, and postulated that a new Descemet's membrane may be forming on the strands. W olter2 demonstrated growth of endothelium onto vitreous in an eye that had been seriously traumatized. Thus, our finding of endothelialization and descemetization of vitreous strands incarcerated in the limbal cataract wound is not surprising,
McDONNELL, et al • VITREOUS INCARCERATION
B Fig 12. .P. 67 O. ,region of catara t wound with cut end of Dc cemet' membrane (D). corneal troma ( ). and vllreou trand ( ) e tending to "ound. lIenuatcd endothelial cellla)er (arrowheads) e tends onto Itreou and i iated with production of basemen I membrane (arrow). ibrobla I-like cell (). me of hich contain pigment granul • are present in the vltreou trand ( 3(00). B, higher pow rofvitreous trand demon trate endothelial cell on urfa e. ell contaIn dense ubpi malemmal aggregate of mlcrofilamenlS (between arrowhead) ( 60.(00).
but rather appears to represent a stereotyped response of the corneal endothelium to injury and irritation. This phenomenon explains the "glassy membrane" appearance of vitreous strands that are observed through biomicroscopic examination of strands of incarcerated vitreous present for several years. Endothelial cell migration may be one of the mechanisms accounting for continued loss of corneal endothelium and eventual corneal edema. This phenomenon should be considered by the surgeon at the time of anterior vitrectomy because the vitreous strand may be quite tough, reinforced by a thick basement membrane produced by the endothelium, and efforts to break the strand may result in undesired traction on the retina and possibly retinal breaks and detachment. Iliff23 noted the great tenacity of these strands and proposed using a bent 30-gauge needle to pull the vitreous strands through a corneal section and cutting them with a scissors. Norton and Machemer! noted that inferior retinal detachments may complicate vitreous incarceration. They stated that the "sheet of vitreous apparently shrinks, gradually pulling the inferior retina into the pupillary area." The presence of subplasmalemmal microfilaments in the endothelial cells which have migrated onto the vitreous strands is similar to those observed in endothelial cells on the surface of the iris in various conditionsY·17 This "myoblastic differentiation" presumably reflects contrac-
tile filaments which are capable of generating traction, much as fibroblasts generate traction during the woundhealing process.2 4 Similarly, the cells within preretinal membrane demonstrate such filaments, apparently accounting for the ability of the membranes to contract and distort the inner retinal surface and contribute to retinal detachment in cases of proliferative vitreoretinopathy.25 We propose that the migration of fibroblasts into, and endothelium onto, the adherent vitreous and the generation of traction by these cells, may contribute to the vitreoretinal traction and retinal detachments that occur as delayed complications of vitreous incarceration. The presence of fibrocytes in the incarcerated vitreous, presumably having migrated from the corneal stroma, would appear to have two primary clinical consequences. First, it would "fortify" the vitreous strand and make it more difficult to tear or sweep from the wound. Second, contraction of these cells would presumably generate tractional forces that might result in peripheral retinal breaks and retinal detachment. Thus, the retinal traction that occurs in such cases may not be due to "vitreous shrinkage,"! but rather, it may be partly a consequence oftractional forces generated by endothelial cells and fibroblasts. The pathogenesis of cystoid macular edema is unknown, but several clinicopathologic studies26- 28 suggest that ocular inflammation plays a role in cases of macular edema after cataract extraction. Vitreous aspirates may 251
OPHTHALMOLOGY
•
FEBRUARY 1986
*
•
VOLUME 93
•
NUMBER 2
Table 3. Eyes with Vitreous Incarceration and Cystoid Macular Edema No, Eyes/Patients
Diabetic retinopathy Iris incarceration Fibrous ingrowth Endothelialization/descemetization Iritis Iridovitreal synechiae Peripheral anterior synechiae CYClitis Ciliary body traction Retinal phlebitis Vitritis Vitreoretinal traction Total
Fig 13. E.P. 62128. A, endothelial cells. joined by junctional complexes (circle), line surface of vitreous strand. M ultilaminated basement membrane is seen beneath endothelial cell (between arrowheads). Fibroblastic cells (asterisks) are present in vitreous strand (X5700). B, high power of fibroblast within vitreous strand shows numerous 70 to 90 nm cytoplasmic filaments (asterisk) and collagen fibrils (arrowheads) that measure 200 A in diameter (X74,000).
show acute and chronic inflammatory cells. 29 It has been suggested that prostaglandins, produced in the anterior uvea, may be the mediators responsible for the retinal vascular leakage. 3D,31 Martin and co-workers,28 in a series of 152 eyes examined pathologically and analyzed statistically, found the presence of intraocular inflammation (iritis, cyclitis, vitritis, retinal phlebitis) to significantly correlate with the presence of macular edema. Among the various anterior segment operative complications, only iridovitreal synechiae correlated significantly with the presence of macular edema. This suggests that mechanical irritation of the iris may be of pathophysiological importance. Our results are consistent with previous observations. Of the six eyes with macular edema in our series, all had one or more foci of inflammation (Table 3). Two eyes of one patient had macular edema in the setting of background diabetic retinopathy. The remaining four eyes all had chronic inflammatory infiltrates ofthe iris and ciliary body. Two of the four eyes had vitritis and all four showed 252
2/2 5 6 5 5 6 5 5 4 5
2 2
6/5
phlebitis or periphlebitis in histologic section or retinal digest preparations. All six eyes had iridovitreal synechiae. Thus, our findings are consistent with the proposal that uveal inflammation results in the breakdown of the retinal blood-ocular barrier. We also found an additional histopathologic feature in the four eyes with macular edema without diabetic retinopathy. In those eyes, the pars plicata was markedly distorted with the ciliary processes displaced anteriorly. This may be a mechanical effect of the anteriorly displaced vitreous body. Flaxel32 examined an eye histopathologically 24 hours after cataract extraction and noted zonule entrapment in the limbal wound. His report does not state whether there was distortion of the pars plicata. To our knowledge, such distortion of the pars plicata has not been previously noted clinically or pathologically in association with aphakic cystoid macular edema. The ciliary epithelium absorbs prostaglandins produced by the anterior uvea. 33 It is unknown whether mechanical distortion of the pars plicata by anteriorly displaced vitreous may alter prostaglandin absorption. In a subset of eyes, strands of vitreous connecting the detached vitreous body with the macula results in chronic cystoid macular edema. Vitreous attachments were observed in a number of patients. 34-38 Fung38 reported that out of 130 eyes in a prospective study, 4 demonstrated the existence of a vitreous strand attached to the macula on biomicroscopic examination. In all eyes in our series, except for one, the vitreous was detached to or anterior to the equator. In the remaining eye, the vitreous detachment did not extend to the equator. In no eye could a persistent vitreomacular adhesion be discerned; only six eyes in our series had macular edema. Thus, vitreomacular traction appears to account for only a minority of the cases of aphakic cystoid macular edema. Anterior vitrectomy is of value in the management of selected cases of chronic aphakic cystoid macular edema. 39 The surgeon should be aware that almost all eyes with vitreous to the wound will have iridovitreal synechiae which may distort the pupil and be associated with chronic inflammatory cell infiltrates. Thus, in addition to excising
MCDONNELL, et al •
vitreous within the anterior chamber, it is advisable to cut all iridovitreal adhesions around the pupillary margin and behind the iris plane and thereby restore "normal aphakic anatomy,,40 to the eyes. The neodymium:YAG laser has been used to cut vitreous strands in eyes with vitreous incarceration in the corneoscleral wound and cystoid macular edema. 41 Bahn and co-workers42 were able to lyse a vascularized vitreous strand to cornea by the argon laser in one patient. This procedure does not address the iridovitreal adhesions which are present in the majority of these eyes. Because the laser study included patients with macular edema of short duration, and because cystoid macular edema resolves spontaneously in most eyes, it is not clear whether Y AG laser treatment improves the natural history of the condition. This is a relatively safe and noninvasive procedure. It may well prove to be valuable means of cutting the vitreous strand allowing the endothelium to "seal" the wound, limit fibroblastic ingrowth, and possibly avoid the later consequences of peripheral vitreoretinal traction.
REFERENCES 1. Norton EWD, Machemer R. New approach to the treatment of selected retinal detachments secondary to vitreous loss at cataract surgery. Am J Ophthalmol1971; 72:705-7. 2. Irvine SR. A newly defined vitreous syndrome following cataract surgery interpreted according to recent concepts of the structure of the vitreous. Am J Ophthalmol1953; 36:599-619. 3. Kuwabara T, Cogan DG. Studies of retinal vascular pattems. Part I. Normal architecture. Arch Ophthalmol 1960; 64:904-11. 4. Swan KC. Fibroblastic ingrowth following cataract extraction. Arch Ophthalmol1973; 89:445-9. 5. Heller MD, Irvine SR, Straatsma BR, Foos RY. Wound healing after cataract eltlraction and position of the vitreous in aphakic eyes as studied postmortem. Trans Am Ophthalmol Soc 1971; 69:245-62. 6. Green WR. The uveal tract. In: Spencer WH, ed. Ophthalmic Pathology; An Atlas and Textbook, 3d ed. Philadelphia: WB Saunders, 1985; 1448. 7. Heath P. Essential atrophy of the iris: a histopathologic study. Trans Am Ophthalmol Soc 1953; 51:167-92. 8. Scheie HG, Yanoff M, Kellogg WT. Essential iris atrophy; report of a case. Arch Ophthalmol1976; 94:1315-20. 9. Wolter JR, Makley TA Jr. Cogan-Reese syndrome; formation of a glass membrane on an iris nevus clinically simulating tumor growth. J Pediatr Ophthalmol Strabismus 1972; 9:102-5. 10. Scheie HG, Yanoff M. Iris nevus (Cogan-Reese) syndrome; a cause of unilateral glaucoma. Arch Ophthalmol1975; 93:963-70. 11. Cogan DG, Reese AB. A syndrome of iris nodules, ectopic Descemet's membrane, and unilateral glaucoma. Doc ~hthalmoI1969; 26:42433. 12. Wolter JR, Fechner PU. Glass membranes on the anterior iris surface. Am J Ophthalmol1962; 53:235-43. 13. Harris M, Tso AY, Kaba FW, et al. Comeal endothelial overgrowth of angle and iris; evidence of myoblastic differentiation in three cases. Ophthalmology 1984; 91: 1154-60. 14. Schulze RR. Rubeosis iridis. Am J Ophthalmol1967; 63:487-95. 15. Gartner S, Taffet S, Friedman AH. The association of rubeosis iridis with endothelialisation of the anterior chamber: report of a clinical case with histopathological review of 16 additional cases. Br J OphthalmoI1977; 61 :267-71. 16. Wolff SM, Zimmerman LE. Chronic secondary glaucoma associated with retrodisplacement of iris root and deepening of the anterior
VITREOUS INCARCERATION
chamber angle secondary to contusion. Am J Ophthalmol1962; 54: 547-63. 17. Patel A. Kenyon KR, Hirst LW, et al. Clinicopathologic features of Chandler's syndrome. Surv Ophthalmol1983; 27:327-44. 18. Rodrigues MM, Phelps CD, Krachmer JH, et al. Glaucoma due to endothelialization of the anterior chamber angle; a comparison of posterior polymorphous dystrophy of the comea and Chandler's syndrome. Arch Ophthalmol1980; 98:688-96. 19. McDonnell PJ, Green WR, Maumenee AE,lIiff WJ. Pathology of intraocular lenses in 33 eyes examined postmortem. Ophthalmology 1983; 90:386-403. 20. Rodrigues MM, Spaeth GL, Donohoo P. Electron microscopy of argon laser therapy in phakic open-angle glaucoma. Ophthalmology 1982; 89:198-210. 21. Donaldson DD, Smith TR. Descemet's membrane tubes. Trans Am Ophthalmol Soc 1966; 64:89-109. 22. Wolter JR. Proliferating comeal endothelium causing a vitreous membrane and incurable retinal detachment. J Pediatr Ophthalmol 1970; 7:162-6. 23. Iliff CEo Treatment of the vitreous-tug syndrome. Am J Ophthalmol 1966; 62:856-9. 24. Gabbiani G, Hirschel BJ, Ryan GB, et al. Granulation tissue as a contractile organ; a study of structure and function. J Exp Med 1972; 135:719-34. 25. Kampik A, Kenyon KR, Michels RG, et al. Epiretinal and vitreous membranes; comparative study of 56 cases. Arch Ophthalmol1981; 99: 1445-54. 26. Michels RG, Green WR, Maumenee AE. Cystoid macular edema following cataract extraction (the Irvine-Gass syndrome); a case studied clinically and histopathologically. Ophthalmic Surg 1971; 2:217-21. 27. Norton AL, Brown WJ, Carlson M, et al. Pathogenesis of aphakic macular edema. Am J Ophthalmol1975; 80:96-101. 28. Martin NF, Green WR, Martin LW. Retinal phlebitis in the Irvine-Gass syndrome. Am J Ophthalmol1977; 83:377-86. 29. Federman JL, Annesley WH Jr, Sarin LK, Remer P. Vitrectomy and cystoid macular edema. Ophthalmology 1980; 87:622-8. 30. Miyake K. Prevention of cystoid macular edema after lens extraction by topical indomethacin. (I) A preliminary report. Albrecht von Graefes Arch Klin Exp OphthalmoI1977; 203:81-8. 31. Miyake K. Prophylaxis of aphakic cystoid macular edema using topical indomethacin. Am Intraocullmplant Soc J 1978; 4:174-9. 32. Flaxel JT. Histology of cataract extractions. Arch Ophthalmol 1970; 83:436-44. 33. Hanna C, Sharp JD. Ocular absorption of indomethacin by the rabbit. Arch Ophthalmol1972; 88:196-8. 34. Ho PC, Tolentino FI. The role of vitreous in aphakic cystoid macular edema: a review. Am Intraocullmplant Soc J 1982; 8:258-64. 35. Maumenee AE. Further advances in the study of the macula. Arch Ophthalmol1967; 78:151-65. 36. Reese AB, Jones IS, Cooper WC. Macular changes secondary to vitreous traction. Trans Am Ophthalmol Soc 1966; 64:123-34. 37. Tolentino FI, Schepens CL. Edema of posterior pole after cataract extraction; a biomicroscopic study. Arch Ophthalmol 1965; 74:781-6. 38. Fung WE. The national, prospective, randomized vitrectomy study for chronic aphakic cystoid macular edema. Progress report and comparison between the control and non randomized groups. Surv Ophthalmol1984; 28:569-75. 39. Fung WE. Results of the vitrectomy/aphakic cystoid macular edema study. Abstract. Ophthalmology 1984; 91 (Suppl):67. 40. Fung WE. Anterior vitrectomy for chronic aphakic cystoid macular edema. Ophthalmology 1980; 87:189-93. 41. Katzen LE, Fleischman JA, Trokel S. YAG laser treatment of cystoid macular edema. Am J Ophthalmol1983; 95:589-92. 42. Bahn CF, Vine AJ, Wolter JR, Martonyi CL. Argon laser photocoagulation of a vitreo-comeal adhesion after trauma. Ophthalmic Surg 1982; 13:53-5.
253