Purpose
To evaluate the clinical characteristics and therapeutic outcomes of cytomegalovirus (CMV)-positive Posner-Schlossman syndrome patients undergoing topical ganciclovir treatment.
Design
Retrospective, comparative, and interventional case series.
Methods
One eye of each of 126 consecutive Posner-Schlossman syndrome patients was investigated using aqueous polymerase chain reaction (PCR) between January 2006 and June 2013. The initial presentations and follow-up data of the CMV-positive patients (68 eyes) and CMV-negative patients (58 eyes) were compared.
Results
Severe endothelial cell loss ( P < .001) and a higher number of eyes requiring glaucoma filtering surgery ( P = .017) were observed in CMV-positive Posner-Schlossman syndrome patients. All CMV-infected eyes treated with continual topical 2% ganciclovir exhibited an undetectable CMV level at the following taps. During follow-up, the average number of antiglaucomatous agents decreased, and a similar frequency of intraocular pressure (IOP) spikes was observed in both groups ( P = .358). Patients with CMV-positive eyes with a disease duration over 5 years were likely to require glaucoma surgery ( P = .024, log-rank test). All patients receiving surgery exhibited CMV-negative PCR during the IOP attack, but experienced severe peripheral anterior synechiae and pigment clogging. Both groups exhibited a similar endothelial cell decrease ( P = .243) and probability of progressive endothelial cell loss ( P = .219, log-rank test).
Conclusion
Ganciclovir treatment was effective for clearing the viral load, assisting the IOP control, and preserving the corneal endothelium of CMV-positive Posner-Schlossman syndrome patients. Early diagnosis and proper treatment could decrease the risk of advanced glaucoma and avoid glaucoma surgery in long-lasting cases.
Cytomegalovirus (CMV) infection in the anterior chamber of the eyes is a newly identified clinical entity that is the result of CMV reactivation in immunocompetent patients. Clinical manifestations of CMV infection include anterior uveitis with anterior chamber (AC) inflammation and keratic precipitates (KPs), elevated intraocular pressure (IOP), and corneal endothelial cell damage. CMV infection in the anterior chamber has been detected in patients diagnosed with Posner-Schlossman syndrome or Fuchs heterochromic iridocyclitis, diseases that were previously believed to be idiopathic.
The corneotrabecular endothelium is a vulnerable tissue in this virus-related anterior segment infection. Although the route of infection remains unclear, CMV particles have been detected in the corneal endothelium and trabeculum in patients with acquired immunodeficiency syndrome, and the condition has been diagnosed as CMV panuveitis. A marked loss of corneal endothelial cells has been detected in correlation to viral loads in CMV-associated iridocyclitis and corneal endotheliitis. A high CMV copy number was also identified as a substantial risk factor for IOP elevations. In 1 recent study, Sobolewska and associates reported that 36.4% of patients required surgical treatment in addition to oral valganciclovir administration to stabilize the IOP in CMV-related Posner-Schlossman syndrome.
Proper and early ganciclovir treatments can halt CMV activity; thus, the damage to the trabecular meshwork and corneal endothelium may be avoided. However, this disease is notorious for its high recurrence rate. Continual administration of ganciclovir was considered to control the reactivation of CMV in an immunocompetent group. In our previous report, we determined that topical administration of 2% ganciclovir eye drops could effectively clear the viral load in CMV endotheliitis. A continual topical ganciclovir treatment avoids the side effects of long-term systemic and intravitreal ganciclovir treatments, including pancytopenia, renal impairment, retinal detachment, and endophthalmitis.
In the literature, induction therapy for CMV infection included the systemic administration of ganciclovir and valganciclovir and the intravitreal injection of ganciclovir. Patients experienced a high recurrence rate after ceasing the treatment. Topical antiviral agents such as acyclovir, ganciclovir, and valganciclovir have been used as a maintenance therapy in previous studies. In our daily practice, we use topical 2% ganciclovir as a long-term maintenance therapy without interruption for Posner-Schlossman syndrome patients with CMV-positive eyes.
Until now, no large-scale study of the long-term clinical outcomes in CMV-positive Posner-Schlossman syndrome patients has been reported. The influence of CMV infection and the therapeutic effect of topical ganciclovir should be discussed. To identify the clinical features and long-term prognosis of Posner-Schlossman syndrome patients with CMV-positive eyes, their demographic data and clinical manifestations were compared with those of Posner-Schlossman syndrome patients with CMV-negative eyes.
In this report, we investigated the efficacy of a topical ganciclovir treatment on CMV-related Posner-Schlossman syndrome. The long-term changes in the corneal endothelial count and IOP of Posner-Schlossman syndrome patients with CMV-positive eyes were compared with those of Posner-Schlossman syndrome patients with CMV-negative eyes.
Materials and Methods
Patients
From January 1, 2005 to June 30, 2013, aqueous tapping for viral polymerase chain reaction (PCR) was routinely performed in patients diagnosed as having Posner-Schlossman syndrome at the Uveitis Clinic of the National Taiwan University Hospital. The National Taiwan University Hospital is the main tertiary referral center in Taiwan. All cases were aqueous PCR-negative for herpes simplex virus (HSV) and varicella zoster virus (VZV) and positive for CMV. The eligible diagnostic criteria of Posner-Schlossman syndrome were as follows: (1) unilateral, recurrent episodes of mild anterior chamber inflammation; (2) the presence of corneal edema and endothelial keratic precipitates; (3) elevated IOP during the attack; (4) the absence of posterior synechiae or posterior inflammation; (5) a duration of attack lasting from a few days to a few weeks; and (6) normal eye examinations between attacks. Aqueous taps were performed in the presence of active inflammation during attack upon presentation. Patients with negative PCR results who presented with persistent anterior chamber inflammation, corneal edema, elevated IOP for more than 2 weeks, or progressive endothelial loss during the follow-up period underwent repeated AC taps for PCR. Among the 126 Posner-Schlossman syndrome patients who were HSV-negative and VZV-negative according to PCR diagnosis, 68 consecutive patients (68 eyes) were CMV-positive and were treated with topical ganciclovir. Another 58 consecutive patients (58 eyes) who were CMV-negative according to PCR diagnosis during the same period were selected as the control group.
The medical records of the CMV-positive and CMV-negative Posner-Schlossman syndrome patients were reviewed. The demographics and clinical findings of each patient were recorded, including their age, sex, duration of disease, best-corrected visual acuity, IOP, and crystalline lens status. The anterior chamber reaction was recorded using slit-lamp microscopy. Changes in the anterior chamber cells were observed after the treatment and accompanied steroid use. The IOP and number of antiglaucomatous medications prescribed at each visit were also compiled. Any glaucoma filtering surgery was highlighted and indicated as poor IOP control. Any self-reported recurrence of uveitis or unexpected IOP was indicated. Corneal endothelial cell density was examined by using confocal microscopy of both the diseased eyes and healthy eyes, and the observations during each visit were recorded. This research protocol was approved by the Ethics Review Board of the National Taiwan University Hospital.
Topical Ganciclovir Treatment of Patients With Cytomegalovirus-Positive Posner-Schlossman Syndrome
All patients were treated with a topical 2% ganciclovir solution after receiving a CMV-positive diagnosis according to the results of PCR of aqueous humor taps. The 2% ganciclovir solution was prepared by dissolving 500 mg of Cymevene lyophilized IV powder (Roche, Basel, Switzerland) in 25 mL of distilled water. The treatment was applied every 2–3 hours daily as an induction therapy, and every 4 hours as a long-term maintenance therapy, as previously reported. After the continual administration of topical ganciclovir for 3 months, all patients underwent another aqueous humor tap to perform a PCR test for detecting the presence of CMV. Topical ganciclovir was prescribed as a long-term antiviral therapy, without discontinuation. Once the conditions of high IOP and corneal edema occurred during the follow-up period, repeated aqueous taps were performed to determine the CMV viral load.
Control of Intraocular Pressure and Need for Glaucoma Surgery
Before referral, the duration of disease exposure, number of antiglaucomatous agents used, and number of eyes that received glaucoma filtering surgery were compared between the 2 groups. During the follow-up period, the occurrence of an IOP spike (IOP > 30 mm Hg) and need for glaucoma filtering surgery were evaluated in the 2 groups. For both groups, the Kaplan-Meier survival analysis was used to estimate the probability of not requiring glaucoma filtering surgery in the diseased eye during the follow-up period.
Response to therapy was defined as remission of anterior chamber inflammation, resolution of corneal edema, and good IOP control with or without antiglaucomatous medications after topical ganciclovir had been received for 1 month. Early treatment failure was defined as persistence of anterior chamber inflammation, corneal edema, or poor IOP control despite the use of antiglaucomatous medications within the first 3 months of treatment. Late treatment failure was defined as relapse of anterior chamber inflammation, the presence of corneal edema, or elevated IOP after 3 months of treatment. Recurrence of CMV-positive Posner-Schlossman syndrome was defined as the presence of active inflammation, corneal edema, endothelial KPs, and elevated IOP with a positive CMV viral load after repeated taps during treatment. Recurrence of CMV-negative Posner-Schlossman syndrome was defined as the presence of active inflammation, corneal edema, endothelial KPs, and elevated IOP with a negative CMV viral load after repeated taps during treatment. These definitions were adapted from a previous study of CMV anterior uveitis.
Evaluation of Corneal Endothelial Cell Density
The corneal endothelial cell density was examined in both the diseased eye and the healthy eye during each visit, and the observations were recorded. The initial corneal endothelial cell loss of the diseased eye was measured upon diagnosis. The corneal endothelial cell loss was calculated according to the formula: corneal endothelial cell loss (%) = (1 − endothelial cell density in the diseased eye/endothelial cell density in the healthy eye) × 100%. The initial corneal endothelial cell loss of the CMV-positive Posner-Schlossman syndrome patients and CMV-negative Posner-Schlossman syndrome patients was compared. Progressive endothelial cell loss was defined as a 20% decrease in the initial endothelial cell count in the diseased eye. For both groups, the Kaplan-Meier survival analysis was used to estimate the probability of progressive endothelial cell loss in the diseased eye during the follow-up period.
Statistics
Data were reported as means ± standard deviation (SD). To compare the differences between the 2 groups, a statistical analysis of age, visual acuity, duration of disease, number of antiglaucomatous agents, endothelial cell density, endothelial cell loss, and follow-up time was performed using the Student t test. All other categorical variables, including sex, lens status, and eyes receiving glaucoma surgery, were analyzed using the χ 2 test (the Fisher exact test was used if n < 5); P < .05 was considered to be statistically significant.
Results
Clinical Manifestations of Patients in the 2 Groups
The demographic data and initial clinical manifestations of the CMV-positive and CMV-negative Posner-Schlossman syndrome patients are listed in Table 1 .
| Characteristics | CMV-Positive Patients (N = 68) | CMV-Negative Patients (N = 58) | P Value |
|---|---|---|---|
| Number of diseased eyes (n) | 68 | 58 | |
| Age at diagnosis (mean ± SD) (y) | 57.69 ± 10.55 | 54.05 ± 12.59 | .08 |
| Male sex (n) | 38 | 36 | .486 |
| Lens status of the diseased eye (n) | .633 | ||
| Phakic | 54 | 44 | |
| Pseudophakic | 14 | 14 | |
| Duration of disease (mean ± SD) (y) | 5.44 ± 4.74 | 3.72 ± 3.79 | .054 |
| Endothelial cell density of the diseased eye (mean ± SD) (cell/mm 2 ) | 1498 ± 743 | 2040 ± 593 | <.001 |
| Endothelial cell density of the healthy eye (mean ± SD) (cell/mm 2 ) | 2664 ± 449 | 2524 ± 509 | .103 |
| Corneal endothelial cell loss of the diseased eye (mean ± SD) (%) | 45.24 ± 25.38 | 18.48 ±18.19 | <.001 |
| Number of antiglaucomatous agents used for the diseased eye at diagnosis (mean ± SD) (n) | 1.78 ± 1.13 | 1.73 ± 1.13 | .17 |
| Eyes receiving filtering surgery (n) | 9 | 1 | .017 |
| Corrected visual acuity in the diseased eye at diagnosis (mean ± SD) (logMAR) | 0.33 ± 0.29 | 0.18 ± 0.12 | .023 |
| Follow-up time (mean ± SD) (mo) | 39.79 ± 14.96 | 36.81 ± 11.32 | .17 |
The mean ages of the CMV-positive Posner-Schlossman syndrome patients and CMV-negative Posner-Schlossman syndrome patients were not significantly different (57.69 and 54.05 years, respectively; P = .08). The proportion of pseudophakia was also similar between the CMV-positive and CMV-negative Posner-Schlossman syndrome patients ( P = .633). The 2 groups did not differ in sex composition ( P = .486). Five eyes (5/68; 7.35%) in patients who were CMV-positive and 2 eyes (2/58; 3.45%) in patients who were CMV-negative exhibited diffuse iris atrophy.
Although the duration of the glaucomatocyclitic crisis was longer in CMV-positive eyes than in CMV-negative eyes, this difference was not significant. (5.44 ± 4.74 years in CMV-positive eyes and 3.72 ± 3.79 years in CMV-negative eyes; P = .054). Upon diagnosis, significantly more CMV-positive Posner-Schlossman syndrome patients received glaucoma filtering surgery than did CMV-negative Posner-Schlossman syndrome patients because of uncontrollable IOP (13.24% and 1.72%, respectively; P = .017). However, the number of antiglaucomatous agents used was similar for both groups ( P = .17).
The corneal endothelial cell density of the diseased eye in CMV-positive and CMV-negative Posner-Schlossman syndrome patients exhibited significant differences (1498 ± 743 cells/mm 2 and 2040 ± 593 cells/mm 2 , respectively; P < .001). Endothelial cell loss in the diseased eye relative to the healthy eye was significantly higher in CMV-positive patients than in CMV-negative patients (45.24% and 18.48%, respectively; P < .001).
The best-corrected visual acuity at diagnosis was slightly lower for CMV-positive eyes than for CMV-negative eyes (0.33 and 0.18 in logMAR, respectively; P = .023). The follow-up time from diagnosis to the final visit was similar in CMV-positive and CMV-negative patients (39.79 ± 14.96 months and 36.81 ± 11.32 months, respectively; P = .17).
Clinical Outcomes After Topical Ganciclovir Treatment
The clinical outcomes of CMV-positive and CMV-negative Posner-Schlossman syndrome patients are listed in Table 2 .
| Characteristics | CMV-Positive Patients (N = 68) | CMV-Negative Patients (N = 58) | P Value |
|---|---|---|---|
| Number of diseased eyes | 68 | 58 | |
| Endothelial cell density of the diseased eye (mean ± SD) (cell/mm 2 ) | 1445 ± 705 | 2000 ± 583 | <.001 |
| Endothelial cell density of the healthy eye (mean ± SD) (cell/mm 2 ) | 2636 ± 450 | 2482 ± 517 | .077 |
| Decrease of endothelial cell density of the diseased eye (mean ± SD) (cell/mm 2 ) | 53 ± 21 | 40 ± 19 | .243 |
| Number of antiglaucomatous agents use at the final visit (mean ± SD) (n) | 0.75 ± 0.5 a | 0.68 ± 0.4 | .258 |
| Eyes on no antiglaucomatous medication (n) | 29 | 34 | .074 |
| IOP crisis higher than 30 mm Hg (n) | 25 | 26 | .358 |
| Eyes receiving filtering surgery during the follow-up period (n) | 8 | 0 | .007 |
| Follow-up time (mean ± SD) (mo) | 39.79 ± 14.96 | 36.81 ± 11.32 | .17 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
