Local Antivirals in
a
Herpes Simplex Stromal James I.
Keratitis Model
McNeill, MD, Herbert E. Kaufman, MD
\s=b\ A herpes simplex stromal keratitis rabbit model, which was produced by intrastromal injection of live virus, was used to evaluate the effects of local antivirals on the natural course of the disease. Topical trifluridine (trifluorothymidine) and vidarabine monophosphate (adenine arabinoside monophosphate), when given early and frequently, suppressed the disease, indicating that viral replication was important in initiating the disease. However, seven days after the stromal disease had begun to develop, neither drug had an appreciable effect. Since the early drug effect had suggested adequate drug penetration, the absence of drug effect later in the disease indicates that viral multiplication may not be important in maintaining the disease. Immunologic reactions may control the disease once the cornea is antigenically altered by the initial infection. Subconjunctivally injected trifluridine was not effective.
(Arch Ophthalmol 97:727-729, 1979) rPhe stromal disease that can follow herpes simplex virus (HSV) infec¬ tion sometimes involves the presence of virus particles.1-' At other times, when stromal edema is the major finding, viral particles are rarely Accepted
publication May 21, 1978. From the Ophthalmology Section, Department of Surgery, Loma Linda (Calif) University School of Medicine (Dr McNeill), and the Department of Ophthalmology, Louisiana State University Eye Center, New Orleans (Dr Kaufman). for
Presented in part before the Ocular Microbiology and Immunology Group, Las Vegas, Oct 6, 1976. Reprint requests to Louisiana State University Eye Center, 136 S Roman St, New Orleans, LA 70112 (Dr Kaufman).
demonstrated24; however, immuno¬ logie processes have been strongly implicated.-'1 The relationship be¬ tween the active viral replication that is seen primarily when necrosis is occurring and the purely edematous form is not completely understood. A rabbit model of stromal herpes has been recently described.' The
corneal disease in this animal model produced by the intrastromal injection of the RE strain of HSV. The model was found to be highly repro¬ ducible and did not require presensitized animals. Disciform edema and stromal necrosis were both observed. The kind of severe prolonged reaction with stromal necrosis and vascularization was much more pronounced than that seen after injecting nonliving antigens and seemed to be a different type of phenomenon. This present study was designed to evaluate the role of the multiplying virus by utilizing the effects of locally administered trifluridine (trifluorothymidine)"·7 and vidarabine monophosphate (adenine arabinoside monophosphate)^" on the natural course of the disease in the rabbit model. was
MATERIALS AND METHODS Inoculation of Rabbit Corneas
New Zealand white rabbits weighing 2 to 3 kg were given intramuscular injections of chlorpromazine (Thorazine, 25 mg/kg) one hour prior to inoculation of the corneas. Following the topical application of 0.5%
proparacaine hydrochloride (Ophthaine), the virus suspension (RE strain HSV) was injected under magnification into the central corneal stroma with a 27-gauge needle attached to a tuberculin syringe. The injection produced a corneal bleb 4 mm
in diameter with 0.2 mL of fluid. All eyes of
Downloaded From: http://archopht.jamanetwork.com/ by a New York University User on 05/21/2015
treated and control animals by day 0. Control
were
injected
Eyes
eyes of the 28 control rabbits were treated with topically applied normal saline solution at the same time as treated eyes received medica¬ tion.
Six times
a
day both
Treated
Eyes
The treated rabbits were divided into the following three major groups: (1) Twentyseven rabbits were given 1% trifluridine in normal saline solution topically six times a day. Seven of these rabbits had the corneal epithelium scratched twice a day with a sterile disposable needle to make certain that epithelial denudation did not in¬ fluence the therapeutic results. (2) Twentynine rabbits were given 2% vidarabine monophosphate in sterile water topically six times a day. (3) Fifteen rabbits were injected subconjunctivally in the upper culde-sac with 2.5 mg of trifluridine once each
day.
Both eyes of each treated rabbit received the same medication. Half of the rabbits in each group received treatment from day 1 following virus injection and the remain¬ ing half of each group were treated from day 7. Treatment was continued until days 21 to 24.
Biomicroscopy
Biomicroscopic observations were made on each eye at regular intervals in a single-controlled fashion. A slit-lamp biomicroscope with white illumination was
used to evaluate the corneal stroma and anterior chamber structures. Fluorescein was placed on the cornea and observations were made with the cobalt blue filter to quantitate epithelial ulcération.
Slit-Lamp Observation Grading All
observations were observer-con¬ trolled and graded on a scale of 0 to 4 to
correspond with increasing pathological findings. The stromal disease was scored as
2
follows: 0, clear and thin stroma; 1, detect¬ able corneal edema, iris details clearly visi¬ ble; 2, gross corneal edema with stromal swelling, iris details still distinct; 3, pupilla¬ ry border no longer distinctly visible, some cellular infiltration possibly present; and 4, an opaque cornea, anterior chamber struc¬ tures not visible.
·
Statistical Treatment Started
Analysis
The Kruskal-Wallis one-way variance was used to test for differences on a daily basis.1"
analysis of significant
RESULTS
Day 1.—Mean corneal stromal disease scores for treatment started on day 1 showing disease to be significantly suppressed (dotted [2% vidarabine monophosphate applied topically six times daily] and dashed [1% trifluridine applied topically six times daily] lines). Almost no disease developed in rabbits receiving trifluridine (dashed line). Solid line indicates normal saline solution applied topically six times daily to eyes of control rabbits.
Fig
When either the trifluridine or the vidarabine monophosphate therapy was started the day after intrastromal inoculation of the live virus, the disease was significantly suppressed (Fig 1). Almost no disease developed in the rabbits treated with trifluridine and they scored significantly better throughout the study period than did those rabbits in the group treated with vidarabine monophosphate. If treatment were delayed until day 7 after viral inoculation, no benefit resulted from the treatment. There were no significant differences be¬ tween the control rabbits and either treatment group started on day 7 (Fig
2).
Scratching the corneal epithelium a day to maintain constant epithelial defects provided some con¬ trol of drug penetration. The course of the disease in the group subjected to twice
2.—Mean corneal stromal disease scores for treatment delayed until day 7 showing significant differences between control rabbits receiving normal saline solution topically six times daily (solid line) and either treatment group. Dotted line indicates 2% vidarabine monophosphate applied topically six times daily, and dashed line, 1% trifluridine applied topically six times daily.
Fig no
corneal scratching was not different from that of the group in which the epithelium was left intact (Fig 3). The amount of trifluridine that was
injected subconjunctivally once a day starting on day 1 or day 7 had no significant effect on the course of the disease (Fig 4). Both of these treated groups were always a little better than the controls; however, this differ¬ ence, which was analyzed daily, may well have been a chance (P > .05 for each day).
occurrence
COMMENT
Fig 3.—Maintaining constant epithelial defects by scratching cornea twice daily in rabbits receiving 1% trifluridine topically six times daily (dashed line) did not alter disease, as is seen in comparison of this group with identical group (solid line) in which epithelium was left intact.
Downloaded From: http://archopht.jamanetwork.com/ by a New York University User on 05/21/2015
Trifluridine was very effective in suppressing the development of stro¬ mal disease when it was started the day after injection of the live virus. Since injecting heat-inactivated virus does not cause disease," antigen alone in the absence of live virus seems insufficient to initiate the disease. Because trifluridine inhibits viral rep¬ lication by forming abnormal DNA," it seems that viral replication within the stroma is necessary to initiate the stromal disease process. It is interesting that as early as day
designed to give a drug dose roughly equivalent to the topically adminis¬ tered drugs. From the lack of response when the injections were given from day 1, it seems that the drug was probably rapidly hydrolyzed by hematogenous enzymes with the once-a-day frequency since its intravenous drug
Fig 4.—Subconjunctival administration of 2.5 mg of trifluridine once daily from day 1 (dashed line) and from day 7 (dotted line) to rabbit corneal stromal herpes model did not significantly affect course of disease. Drug was probably hydrolyzed rapidly by hematogenous enzymes. Solid line indicates normal saline solution applied topically six times daily to eyes of control rabbits. 7 the disease
was
refractory
to the
antiviral treatment that was so effective when started earlier. Small amounts of virus could be cultured on day 7,1- but virus multiplication was probably no longer important in the disease progression and immune mechanisms became predominant. The small epithelial defect at the injection site had healed well by the same
following day, which
the earliest any treatment was begun. The re¬ sponse to trifluridine therapy shown in Fig 1 indicates that trifluridine can adequately cross the intact epithe¬ lium. Also maintaining constant epi¬ thelial defects later did not change the effect of the drug (Fig 3). The subconjunctival injections of trifluridine in infected eyes were was
half-life is about 20 minutes. It seems that once a sufficient amount of viral antigen has formed in the cornea or when the cell mem¬ branes have been antigenically al¬ tered, antiviral chemotherapy can suppress further viral proliferation but cannot eliminate the fixed anti¬ ' '' gens.- The intensity of the corneal disease probably depends on the level of host response and the antigenicity of the cornea. This investigation was supported in part by Public Health Service grants EY-07012, EY00446, and EY-00007 from the National Eye Institute, National Institutes of Health. Names and Trademarks of Drugs
Nonproprietary
Chlorpromazine—Chlor-PZ, Cromedazine, Promachel, Thorazine.
Proparacaine hydrochloride—Alcaine, Ophthaine, Ophthetic.
References 1. Dawson C, Togni B, Moore TE: Structural changes in chronic herpetic keratitis: Studies by light and electron microscopy. Arch Ophthalmol
79:740-747, 1968. 2. Dawson CR, Togni B: Herpes simplex eye infections: Clinical manifestations, pathogenesis and management. Surv Ophthalmol 21:121-135,
1976. 3. Kaufman HE: Disease of the corneal stroma after herpes simplex infection. Am J Ophthalmol 63:878-882, 1967. 4. Sugar A, Kaufman HE: Herpetic Keratitis: Current Concepts in Ophthalmololgy. St Louis, CV Mosby Co, 1976, pp 1-16. 5. Metcalf JF, McNeill JI, Kaufman HE: Experimental disciform edema and necrotizing keratitis in the rabbit. Invest Ophthalmol 15:979\x=req-\
985, 1976.
6. Wellings PC, Awdry PH, Bors FH, et al: Clinical evaluation of trifluorothymidine in the treatment of herpes simplex corneal ulcers. Am J Ophthalmol 73:932-942, 1972. 7. Hyndiuk RA, Kaufman HE: Newer compounds in therapy of herpes simplex keratitis. Arch Ophthalmol 78:600-605, 1967. 8. Kaufman HE, Varnell ED: Effect of 9-B-D arabinofuranosyladenine 5\m='\-monophosphate and 9-B-D arabinofuranosylhypoxanthine 5\m='\-monophosphate on experimental herpes simplex keratitis. Antimicrob Agents Chemother 10:885-888, 1976. 9. Trobe JD, Centifanto Y, Zam ZS, et al: Anti-herpes activity of adenine arabinoside monophosphate. Invest Ophthalmol 15:196-198,
Downloaded From: http://archopht.jamanetwork.com/ by a New York University User on 05/21/2015
1976. 10. Hollander M, Walf DA: Nonparametric Statistical Methods. New York, John Wiley & Sons Inc, 1973, pp 128-130. 11. Fujiwara Y, Heidelberger C: Fluorinated pyrimidines: XXXVIII. The incorporation of 5-trifluoromethyl-2\m='\-deoxyuridineinto the deoxyribonucleic acid of vaccinia virus. Mol Pharmacol 6:281, 1970. 12. Metcalf JF, Kaufman HE: Herpetic stromal keratitis\p=m-\evidencefor cell-mediated immunopathogenesis. Am J Ophthalmol 82:827-834, 1976. 13. Gerstein DD, Dawson CR, Oh JO: Phosphonacetic acid in the treatment of experimental herpes simplex keratitis. Antimicrob Agents Chemother 7:285-288, 1975.
a
Herpes Simplex Stromal James I.
Keratitis Model
McNeill, MD, Herbert E. Kaufman, MD
\s=b\ A herpes simplex stromal keratitis rabbit model, which was produced by intrastromal injection of live virus, was used to evaluate the effects of local antivirals on the natural course of the disease. Topical trifluridine (trifluorothymidine) and vidarabine monophosphate (adenine arabinoside monophosphate), when given early and frequently, suppressed the disease, indicating that viral replication was important in initiating the disease. However, seven days after the stromal disease had begun to develop, neither drug had an appreciable effect. Since the early drug effect had suggested adequate drug penetration, the absence of drug effect later in the disease indicates that viral multiplication may not be important in maintaining the disease. Immunologic reactions may control the disease once the cornea is antigenically altered by the initial infection. Subconjunctivally injected trifluridine was not effective.
(Arch Ophthalmol 97:727-729, 1979) rPhe stromal disease that can follow herpes simplex virus (HSV) infec¬ tion sometimes involves the presence of virus particles.1-' At other times, when stromal edema is the major finding, viral particles are rarely Accepted
publication May 21, 1978. From the Ophthalmology Section, Department of Surgery, Loma Linda (Calif) University School of Medicine (Dr McNeill), and the Department of Ophthalmology, Louisiana State University Eye Center, New Orleans (Dr Kaufman). for
Presented in part before the Ocular Microbiology and Immunology Group, Las Vegas, Oct 6, 1976. Reprint requests to Louisiana State University Eye Center, 136 S Roman St, New Orleans, LA 70112 (Dr Kaufman).
demonstrated24; however, immuno¬ logie processes have been strongly implicated.-'1 The relationship be¬ tween the active viral replication that is seen primarily when necrosis is occurring and the purely edematous form is not completely understood. A rabbit model of stromal herpes has been recently described.' The
corneal disease in this animal model produced by the intrastromal injection of the RE strain of HSV. The model was found to be highly repro¬ ducible and did not require presensitized animals. Disciform edema and stromal necrosis were both observed. The kind of severe prolonged reaction with stromal necrosis and vascularization was much more pronounced than that seen after injecting nonliving antigens and seemed to be a different type of phenomenon. This present study was designed to evaluate the role of the multiplying virus by utilizing the effects of locally administered trifluridine (trifluorothymidine)"·7 and vidarabine monophosphate (adenine arabinoside monophosphate)^" on the natural course of the disease in the rabbit model. was
MATERIALS AND METHODS Inoculation of Rabbit Corneas
New Zealand white rabbits weighing 2 to 3 kg were given intramuscular injections of chlorpromazine (Thorazine, 25 mg/kg) one hour prior to inoculation of the corneas. Following the topical application of 0.5%
proparacaine hydrochloride (Ophthaine), the virus suspension (RE strain HSV) was injected under magnification into the central corneal stroma with a 27-gauge needle attached to a tuberculin syringe. The injection produced a corneal bleb 4 mm
in diameter with 0.2 mL of fluid. All eyes of
Downloaded From: http://archopht.jamanetwork.com/ by a New York University User on 05/21/2015
treated and control animals by day 0. Control
were
injected
Eyes
eyes of the 28 control rabbits were treated with topically applied normal saline solution at the same time as treated eyes received medica¬ tion.
Six times
a
day both
Treated
Eyes
The treated rabbits were divided into the following three major groups: (1) Twentyseven rabbits were given 1% trifluridine in normal saline solution topically six times a day. Seven of these rabbits had the corneal epithelium scratched twice a day with a sterile disposable needle to make certain that epithelial denudation did not in¬ fluence the therapeutic results. (2) Twentynine rabbits were given 2% vidarabine monophosphate in sterile water topically six times a day. (3) Fifteen rabbits were injected subconjunctivally in the upper culde-sac with 2.5 mg of trifluridine once each
day.
Both eyes of each treated rabbit received the same medication. Half of the rabbits in each group received treatment from day 1 following virus injection and the remain¬ ing half of each group were treated from day 7. Treatment was continued until days 21 to 24.
Biomicroscopy
Biomicroscopic observations were made on each eye at regular intervals in a single-controlled fashion. A slit-lamp biomicroscope with white illumination was
used to evaluate the corneal stroma and anterior chamber structures. Fluorescein was placed on the cornea and observations were made with the cobalt blue filter to quantitate epithelial ulcération.
Slit-Lamp Observation Grading All
observations were observer-con¬ trolled and graded on a scale of 0 to 4 to
correspond with increasing pathological findings. The stromal disease was scored as
2
follows: 0, clear and thin stroma; 1, detect¬ able corneal edema, iris details clearly visi¬ ble; 2, gross corneal edema with stromal swelling, iris details still distinct; 3, pupilla¬ ry border no longer distinctly visible, some cellular infiltration possibly present; and 4, an opaque cornea, anterior chamber struc¬ tures not visible.
·
Statistical Treatment Started
Analysis
The Kruskal-Wallis one-way variance was used to test for differences on a daily basis.1"
analysis of significant
RESULTS
Day 1.—Mean corneal stromal disease scores for treatment started on day 1 showing disease to be significantly suppressed (dotted [2% vidarabine monophosphate applied topically six times daily] and dashed [1% trifluridine applied topically six times daily] lines). Almost no disease developed in rabbits receiving trifluridine (dashed line). Solid line indicates normal saline solution applied topically six times daily to eyes of control rabbits.
Fig
When either the trifluridine or the vidarabine monophosphate therapy was started the day after intrastromal inoculation of the live virus, the disease was significantly suppressed (Fig 1). Almost no disease developed in the rabbits treated with trifluridine and they scored significantly better throughout the study period than did those rabbits in the group treated with vidarabine monophosphate. If treatment were delayed until day 7 after viral inoculation, no benefit resulted from the treatment. There were no significant differences be¬ tween the control rabbits and either treatment group started on day 7 (Fig
2).
Scratching the corneal epithelium a day to maintain constant epithelial defects provided some con¬ trol of drug penetration. The course of the disease in the group subjected to twice
2.—Mean corneal stromal disease scores for treatment delayed until day 7 showing significant differences between control rabbits receiving normal saline solution topically six times daily (solid line) and either treatment group. Dotted line indicates 2% vidarabine monophosphate applied topically six times daily, and dashed line, 1% trifluridine applied topically six times daily.
Fig no
corneal scratching was not different from that of the group in which the epithelium was left intact (Fig 3). The amount of trifluridine that was
injected subconjunctivally once a day starting on day 1 or day 7 had no significant effect on the course of the disease (Fig 4). Both of these treated groups were always a little better than the controls; however, this differ¬ ence, which was analyzed daily, may well have been a chance (P > .05 for each day).
occurrence
COMMENT
Fig 3.—Maintaining constant epithelial defects by scratching cornea twice daily in rabbits receiving 1% trifluridine topically six times daily (dashed line) did not alter disease, as is seen in comparison of this group with identical group (solid line) in which epithelium was left intact.
Downloaded From: http://archopht.jamanetwork.com/ by a New York University User on 05/21/2015
Trifluridine was very effective in suppressing the development of stro¬ mal disease when it was started the day after injection of the live virus. Since injecting heat-inactivated virus does not cause disease," antigen alone in the absence of live virus seems insufficient to initiate the disease. Because trifluridine inhibits viral rep¬ lication by forming abnormal DNA," it seems that viral replication within the stroma is necessary to initiate the stromal disease process. It is interesting that as early as day
designed to give a drug dose roughly equivalent to the topically adminis¬ tered drugs. From the lack of response when the injections were given from day 1, it seems that the drug was probably rapidly hydrolyzed by hematogenous enzymes with the once-a-day frequency since its intravenous drug
Fig 4.—Subconjunctival administration of 2.5 mg of trifluridine once daily from day 1 (dashed line) and from day 7 (dotted line) to rabbit corneal stromal herpes model did not significantly affect course of disease. Drug was probably hydrolyzed rapidly by hematogenous enzymes. Solid line indicates normal saline solution applied topically six times daily to eyes of control rabbits. 7 the disease
was
refractory
to the
antiviral treatment that was so effective when started earlier. Small amounts of virus could be cultured on day 7,1- but virus multiplication was probably no longer important in the disease progression and immune mechanisms became predominant. The small epithelial defect at the injection site had healed well by the same
following day, which
the earliest any treatment was begun. The re¬ sponse to trifluridine therapy shown in Fig 1 indicates that trifluridine can adequately cross the intact epithe¬ lium. Also maintaining constant epi¬ thelial defects later did not change the effect of the drug (Fig 3). The subconjunctival injections of trifluridine in infected eyes were was
half-life is about 20 minutes. It seems that once a sufficient amount of viral antigen has formed in the cornea or when the cell mem¬ branes have been antigenically al¬ tered, antiviral chemotherapy can suppress further viral proliferation but cannot eliminate the fixed anti¬ ' '' gens.- The intensity of the corneal disease probably depends on the level of host response and the antigenicity of the cornea. This investigation was supported in part by Public Health Service grants EY-07012, EY00446, and EY-00007 from the National Eye Institute, National Institutes of Health. Names and Trademarks of Drugs
Nonproprietary
Chlorpromazine—Chlor-PZ, Cromedazine, Promachel, Thorazine.
Proparacaine hydrochloride—Alcaine, Ophthaine, Ophthetic.
References 1. Dawson C, Togni B, Moore TE: Structural changes in chronic herpetic keratitis: Studies by light and electron microscopy. Arch Ophthalmol
79:740-747, 1968. 2. Dawson CR, Togni B: Herpes simplex eye infections: Clinical manifestations, pathogenesis and management. Surv Ophthalmol 21:121-135,
1976. 3. Kaufman HE: Disease of the corneal stroma after herpes simplex infection. Am J Ophthalmol 63:878-882, 1967. 4. Sugar A, Kaufman HE: Herpetic Keratitis: Current Concepts in Ophthalmololgy. St Louis, CV Mosby Co, 1976, pp 1-16. 5. Metcalf JF, McNeill JI, Kaufman HE: Experimental disciform edema and necrotizing keratitis in the rabbit. Invest Ophthalmol 15:979\x=req-\
985, 1976.
6. Wellings PC, Awdry PH, Bors FH, et al: Clinical evaluation of trifluorothymidine in the treatment of herpes simplex corneal ulcers. Am J Ophthalmol 73:932-942, 1972. 7. Hyndiuk RA, Kaufman HE: Newer compounds in therapy of herpes simplex keratitis. Arch Ophthalmol 78:600-605, 1967. 8. Kaufman HE, Varnell ED: Effect of 9-B-D arabinofuranosyladenine 5\m='\-monophosphate and 9-B-D arabinofuranosylhypoxanthine 5\m='\-monophosphate on experimental herpes simplex keratitis. Antimicrob Agents Chemother 10:885-888, 1976. 9. Trobe JD, Centifanto Y, Zam ZS, et al: Anti-herpes activity of adenine arabinoside monophosphate. Invest Ophthalmol 15:196-198,
Downloaded From: http://archopht.jamanetwork.com/ by a New York University User on 05/21/2015
1976. 10. Hollander M, Walf DA: Nonparametric Statistical Methods. New York, John Wiley & Sons Inc, 1973, pp 128-130. 11. Fujiwara Y, Heidelberger C: Fluorinated pyrimidines: XXXVIII. The incorporation of 5-trifluoromethyl-2\m='\-deoxyuridineinto the deoxyribonucleic acid of vaccinia virus. Mol Pharmacol 6:281, 1970. 12. Metcalf JF, Kaufman HE: Herpetic stromal keratitis\p=m-\evidencefor cell-mediated immunopathogenesis. Am J Ophthalmol 82:827-834, 1976. 13. Gerstein DD, Dawson CR, Oh JO: Phosphonacetic acid in the treatment of experimental herpes simplex keratitis. Antimicrob Agents Chemother 7:285-288, 1975.
