SURVEY OF OPHTHALMOLOGY
VOLUME 31 . NUMBER 1 . JULY-AUGUST
AFTERIMAGES
JONATHAN
1966
D. WIRTSCHAFTER,
EDITOR
Infectious Crystalline Keratopathy GEORGE
R. REISS, M.D.,’
R. JEAN CAMPBELL,
M.D.,‘32 AND WILLIAM
Departments of ‘Ophthalmology and ‘Surgical Pathology, Mqo
M. BOURNE,
M.D.’
Clinic and Foundation, Rochester. Minnesota
Abstract. A 57-year-old
white female was noted to have unusual intrastromal crystalline-like opacities in her failing right cornea1 graft five months after transplant surgery. Cultures grew Streptococcus viridans; since the graft was failing it was replaced and sent for pathologic esamination. H&E stains revealed intrastromal pockets of basophilic material between the lamellae: there was a notable absence of inflammatory cells. Gram’s stain showed the basophilic material to be gram positive and transmission electron microscopy confirmed the presence of intrastromal pockets of bacteria. The term “infectious crystalline keratopathy” was coined by Meisler for this entity, which occurs following longterm topical steroid usage. The quiet clinical appearance may make its infectious etiology unapparent. Management consists of discontinuation of the steroids and the administration of antibiotics; continued infection, cornea1 scars, or graft failure may occur and transplantation is then required. (Surv Ophthalmol 31:69-72, 1986)
Key words.
herpetic Streptococcus viridans
keratitis
l
penetrating
A 57-year-old white woman with retinitis pigmentosa and chronic open angle glaucoma had an intracapsular cataract extraction of the right eye in 1971 and of the left eye in 1973. Bilateral aphakic bullous keratopathy developed and a left penetrating keratoplasty was performed in March 1984 without sequelae. A penetrating keratoplasty was performed on the right eye in December 1984. In the immediate postoperative period, pachymetry revealed persistent progressive thickening of the graft, which appeared hazy, particularly in the posterior stroma. Stria, 3 + were seen but no keratic precipitates were present. Light projection was poor. The diagnosis of early graft failure was made and the patient was treated with tapering doses of topical steroids over the subsequent months. In addition, she received Timoptic 0.5% bid bilaterally for the chronic open angle glaucoma. A pulmonary embolus developed two weeks following surgery and required anticoagulant therapy; thus, a repeat keratoplasty was delayed. When next seen by the ophthalmologist in April 1985, the right graft was thick and mildly hazy. Folds of Descemet’s membrane were present but no keratic precipitates and no vessels were seen. Prednisolone acetate 1o/o
keratoplasty
l
steroids,
topical
l
drops were continued once daily; repeat keratoplasty was delayed for one month until anticoagulant therapy could be safely discontinued. By May 5, the patient complained of persistent irritation of the right eye accompanied by photophobia. The graft appeared clear centrally, but the epithelium was edematous. The central cornea1 thickness was 0.66 mm. Slit lamp examination showed a large epithelial defect and a white anterior stromal lesion at the 7 o’clock position (Figs. l-3). This lesion appeared crystalline on retroillumination. Fingerlike projections were seen extending from these intrastromal lesions. The edematous epithelium stained slightly with fluorescein. Cornea1 scrapings grew Streptococcusuiridans in significant numbers. The topical steroids were discontinued and topical and subconjunctival AnceP were given. A repeat penetrating keratoplasty was performed. The patient has done well with the second graft.
Pathology and Microscopy cn
A portion of the cornea1 button was submitted to microbiology and Streptococcus viridans was cultured.
70
3 I( 1) July-August
Surv Ophthalmol
Fig. 1. (Slit Lamp Biomicroscopy) reveals whitish infiltrates in cornea.
Direct
1986
REISS ET AL
illumination
Fig. 2. (Slit Lamp Biomicroscopy) cornea1 location of infiltrates.
The remaining tissue was fixed in Trump’s solution, a portion was processed for transmission electron microscopy and a portion for light microscopy. Transmission electron microscopy showed bacterial organisms with a distinct laminated cell wall. Mesosomes and ribosomes were present within the cytoplasm (Figs. 4 and 5). H&E sections showed an area of ulceration with focal destruction of Bowman’s membrane. Pockets of granular basophilic material were localized in the superficial stroma and were particularly prominent at the ulcer base (Fig. 6). These stained positive with the Brown and Best gram stain (Fig. 7). Of note was the paucity of inflammatory cells. There was severe loss of endothelial cells.
Slit shows
superficial
Fig. 3. (Slit Lamp Biomicroscopy) Retro-illumination accentuates the crystalline-like projections at the edges of the lesions.
Discussion Gorovoy and associates5 in 1983 were the first to report crystalline opacities in the cornea1 transplant of a patient receiving topical corticosteroids for a prolonged period. Cultures were not obtained because the infectious etiology of the lesion was not recognized, but transmission electron microscopic studies demonstrated the organisms. The patient was a 69-year-old woman who had undergone penetrating keratoplasty six months earlier for aphakic
bullous keratopathy. Included in the postoperative medication were topical chloramphenicol and 0.1% dexamethasone sodium phosphate. The antibiotic had been terminated three months after surgery, but the steroid treatment was maintained daily until the patient presented six months post-surgery with needle-like stromal opacities. The authors commented on the lack of an inflammatory response and
1
TABLE
Summay of Findings in Eight Reported Patients With Injtktious C’ry~tallineKeratopathy No. of patients
R eport
Surgery
Topical
Herpes
Bacterial
steroids
simplex
Culture -
1983 Gorovoy et al 1984 Meisler et al
5*
4
1
1 4
-
1985/N Samples 1986 Reiss et al
1 1
1 I
1
-
I
-
7
3
1
et al
Total *hleisler’s report separately here.
8 included
7 live of his patients,
3
plus GoroLroy’s and Samplrs’
3
-
TEM -
1 1
1 7 patients,
1 3 who are reported
INFECTIOUS
CRYSTALLINE
KERATOPATHY
71
Fig. 4. Transmission
electron microscopy. Aggregates of streptococci lie between stromal lamellae. ( X 13,500)
Fig. 5. Transmission electron microscopy. Organisms show well defined cell wall and cytoplasm contains mesosomes and ribosomes. ( x 25,000)
suggested that it had been suppressed by the longterm topical corticosteroid therapy. Interestingly, a localized area of epithelial downgrowth was found within the cornea1 stroma adjacent to a suture track and the authors suggested that this presented a route through which the bacteria could subsequently invade the stroma. Meisler et al7 coined the term “infectious crystalline keratopathy” for this entity, which aptly describes the clinical appearance and its etiology. These authors described three patients with the condition. All had received topical steroids for a prolonged period. Two patients had had penetrating keratoplasty and one of these patients had Herpes simplex keratitis. The third nonsurgical patient also had Herpes simplex keratitis. In a later report,” Meisler summarized the findings in seven patients: his own three previously reported cases’ plus two patient;” and one who would new ones; Goroway’s Our patient be reported subsequently by Samples.‘” represents the eighth reported case. [Linreportc(l
cases exist (Personal communications: ,J .R. Samples, E.L. Stock, D. East)).] Findings in these eight patients are summarized in Table 1. All patients but one had received topical steroids for a significant period of time. Cultures of Streptococcus uiriduns were obtained from the cornea in four of the eight patients. Histology in seven patients demonstrated that the bacteria tend to lie in the anterior cornea1 stroma and that there is a notable lack of inflammatory cells. Transmission rlectron microscopy was performed in three patients and findings were similar.‘-H Corncal transplant had and three been performed in seven of‘ the patients. patients had a history of Hrrpcs simplex infection. Growth of the crystals has bt‘en ohscbrvvd in all patients, indicating that the organism is \.ial)lc within the (‘orma. The crystalline appearanc’e is prohably due to the bacterial masses t)cbt\\:rrn the. corncal c*lec,!ron drnsc lamellar. Samples et all1 obsCrvt4 bodies with ncedlr-like projrctions antI suggested tll;lt th(ssc mav by rcspc,ns;I,tc. fi)r ttif. cr\.staltinr
72
Surv Ophthalmol
3 l(1) July-August
1986
REISS ET AL
Fig. 6. Cornea1 epithelium above is partially and artifactitiously detached from stroma. Granular deposits, ofirregular size, lie between stromal lamellae. Note lack of inflammatory cell response. (H&E, X 400 original magnification)
Fig. 7. Gram positive cocci lie in clumps between stromal lamellae. (Gram stain, X 400 original magnification)
it is doubtful if the size of appearance. However, these individual structures is responsible for the appearance, which is more likely attributable to the pockets of bacteria between the lamellar structures. The appearance of crystals within a transplanted cornea or in the cornea of a patient who has had Herpes simplex treated with longterm topical steroids should suggest the infectious nature of this condition’s4 A cornea1 scraping is recommended. of the normal Streptococcus viridans, *v3s4a constituent flora of the mouth, conjunctiva and upper respiratory tract, is a slowly growing, opportunistic pathogen. 4,g,1’The ability to elaborate dextran may contribute to the organism’s ability to adhere to tissues. It is likely that local immunosuppression plays a role in this clinical entity. Discontinuation of the topical steroids together with aggressive therapy with appropriate antibiotics may suffice, but continued infection, vascularization or scar formation may affect the final visual acuity and require keratoplasty.
2. Bourgault A-M, Wilson WR, Washington JA II: Antimicrobial susceptibilities of species of Viridans Streptococci. J Infect Dis 140:316-321, 1979 3. Facklam RR: Physiological differentiation of Viridunr Streptococci. J Clin Microbial 5:184-201, 1977 4. Friedlaender MH, Masi RJ, Osumoto M, et al: Ocular microbial flora in immunodeftcient patients. Arch Ophthalmol 98: 121 I-1213, 1980 5. Gorovoy MS, Stern GA, Hood CI, Allen C: Intrastromal noninflammatory bacterial colonization of a cornea1 graft. Arch Ophthalmol 101: 1749-l 752, 1983 Holt SC: Bacteria, in Johannsson JV (ed:) Electron Microscopy in Human Medicine Vol3. New York, McGraw Hill, pp 91-278, 1980 Meisler DM, Langston RHS, Naab TJ, et al: Infectious crystalline keratopathy. Am J Ophthalmol 97:337-343, 1984 Meisler DM, Langston RHS, Aaby AA, et al: Infectious cornea1 crystalline formation. Inuest Ophthalmol Vis Sci (Suppl) 25:23, 1984 Ostler HB, Okumoto M, Wilkey C: The changing pattern of the etiology ofcentral bacterial cornea1 (hypopyon) ulcer. Tram Pat Coast Otoophthalmol Sot Ann Meeting 57:23!+246, 1976 SD, Binder PS: Infectious crystalline 10. Samples JR, Baumgartner keratopathy: An Electron Microscope Analysis. Cornea 4; 118-126, 1985 11. Thygeson P, Spencer WH: The Verhoeff Symposium: The changing character of infectious cornea1 disease: Emerging opportunistic microbial forms (1928-1973). Truer Am Ophthalmol Sot 71:246-253, 1973
References 1. Boisjoly
HM, Pavan-Langston A: Superinfections in Herpes %t354-361, 1983
D, Kenyon KR, Sullivan Baker simplex keratitis. AmJ Ophthalmol
Reprint requests should be addressed to R. Jean Campbell, M.D., Department of Ophthalmology, Mayo Clinic, Rochester, MN 55905.