Purpose
To determine the incidence, risk factor, and outcomes of elevated intraocular pressure (IOP) during active microbial keratitis.
Design
Retrospective cohort study.
Methods
One hundred eighty-four patients with culture-proven microbial keratitis examined from January 2003 through December 2007 were included. High IOP was defined as IOP of 22 mm Hg or higher measured during the episode of active keratitis. The control group consisted of eyes with microbial keratitis whose IOP remained less than 22 mm Hg. Twelve factors were evaluated by univariate and multivariate analyses to determine whether any were associated with increased IOP. The incidence, risk factors, microbial profile, the necessity of therapeutic surgery, time to resolution, and final visual acuity were compared between the high IOP group and the control group.
Results
High IOP (mean, 29.1 mm Hg; range, 22 to 51 mm Hg) occurred in 52 (28%) of 184 patients with active corneal infection. Prior ocular surgery, diabetes mellitus, and ulcer size of 4.0 mm or larger were associated with IOP elevation ( P ≤ .013). Surgical interventions were necessary in 19 (39%) of 49 patients in the high IOP group and in 14 (11%) of 129 patients in the control group ( P < .0001). Time to ulcer resolution was longer in the high IOP group (mean, 50.1 ± 53.2 days) than in the control (mean, 31.6 ± 42.0 days; P = .005). Final visual acuity of 20/40 or better was achieved by more patients in the control group (47%) than in the high IOP group (20%; P < .001).
Conclusions
Elevated IOP was detected in a significant proportion of patients with active microbial keratitis and was associated with poorer outcomes. Routine IOP check should be performed to avoid possible optic nerve damage.
In many developing countries, microbial keratitis is the second most common cause of preventable blindness after cataracts. In Uganda, 22% of cases of visual impairment in children are related directly to corneal ulceration. In southern India, Gonzales and associates determined that the annual incidence of corneal ulceration is 11.3 per 10 000 people. Even in developed countries, the incidence and prevalence of microbial keratitis are increasing. In a population-based cohort study in Olmsted county, Minnesota, USA, Erie and associates showed that the incidence of ulcerative keratitis has increased from 2.5 per 100 000 people in the 1950s to 11.0 per 100 000 people in the 1980s. In a recent study in northern California, Jeng and associates found that the annual incidence of ulcerative keratitis was 27.6 per 100 000 people.
Common complications of microbial keratitis include persistent epithelial defect; corneal haze, thinning, scarring, melting, and perforation; cataract; and, rarely, endophthalmitis. Some of these complications have been evaluated and reported previously. Studies of these complications have focused mainly on the causative pathogens, risk factors, response to treatment, final visual acuity (VA), and therapeutic surgical intervention.
Infectious keratitis often causes anterior chamber reaction that can range from the presence of a few cells to significant hypopyon. Intraocular pressure (IOP) elevation secondary to uveitis is a well-known consequence of anterior chamber inflammation, and in most cases of uveitis associated with high IOP, infection is often the cause. These findings raise concern about an increase in IOP in patients with active keratitis resulting in inflammatory cell blockage of the trabecular meshwork, trabeculitis, or peripheral anterior synechiae. To the best of our knowledge, the effect of acute infectious keratitis on IOP has not been reported. The objective of our study was to investigate the incidence, microbial profile, risk factors, and clinical outcomes of acute microbial keratitis associated with increased IOP.
Methods
We conducted a retrospective chart review of culture-proven microbial keratitis cases that occurred from January 2003 through December 2007 at 1 tertiary referral center (Jules Stein Eye Institute, University of California, Los Angeles). All cases of culture-proven microbial keratitis in which IOP was measured during the active infection by either Tono-Pen (Medtronic Solan, Jacksonville, Florida, USA) or pneumotonometer (Medtronic Solan) were included. Corneal specimens were collected by the corneal scraping technique. Blood agar plates, chocolate agar plates, or both were used for bacterial culture. Fungi and acanthamoeba cultures were grown on Sabouraud plates and on nonnutrient agar, overlaid with Escherichia coli , respectively. Data on medical and ocular history reviewed included a history of diabetes mellitus, glaucoma, prior ocular surgery, recent ocular trauma, use of topical corticosteroids, systemic immunosuppressive therapy, and contact lens wear. Information about VA at presentation, resolution of the ulcer, and the time of the last follow-up visit was collected. In case of therapeutic surgical intervention, the preoperative VA was defined as the final VA. VA was measured by using the Snellen chart. In preschool children, the Lea symbols chart was used to measure VA.
Ulcer size was measured in every patient at each visit during the acute phase. The longest diameter of the ulcer at presentation, measured by slit-lamp examination, was defined as the size of the ulcer. A large ulcer was defined as one with a diameter of 4 mm or larger; a medium ulcer was defined as one with a diameter of 2 mm or larger and less than 3.9 mm; and small ulcer was defined as one with a diameter of less than 2 mm. The location of the ulcer was defined as central if its infiltration involved the 3-mm zone of the central cornea, paracentral if it involved between the 3- and 5-mm central zone, and peripheral if it involved the corneal tissue outside the 5-mm zone. Complete resolution was defined as resolution of stromal infiltration and epithelial defect with or without scar formation.
High IOP was defined as an IOP of 22 mm Hg or higher measured by a Tono-Pen or pneumotonometer at any time before the resolution of the ulcer. The highest IOP recorded was used for data analysis. In patients with glaucoma, previously prescribed glaucoma medications were continued. To identify the possible risk factors associated with the increase in IOP, we defined the high IOP group as those patients with high IOP during active microbial keratitis and the control group as those with active microbial keratitis but without high IOP. The pathogens that accounted for the infection also were compared between these 2 groups.
The statistical software SAS version 9.1 (SAS Institute, Cary, North Carolina, USA) was used. Univariate statistical tests (Fisher exact text and Wilcoxon rank-sum test) and multivariate statistical tests (multiple logistic regression model) were used to compare variables between groups, and only statistically significant variables in the univariate analysis were included in the multiple logistic regression model. A P value ≤ .05 indicated statistical significance.
Results
A total of 184 patients with microbial keratitis were included in the study. The high IOP group comprised 52 patients (28%; i.e., the group of patients with an IOP ≥ 22 mm Hg). The mean IOP was 29.1 ± 6.9 mm Hg (range, 22 to 51 mm Hg) in the high IOP group and 16.3 ± 2.7 mm Hg (range, 10 to 21 mm Hg) in the control group (n = 132). The mean age was 55.44 ± 22.86 years (range, 0.37 to 89.73 years) in high IOP group and 51.34 ± 26.52 years (range, 0.07 to 97.6 years) in the control group ( P = .33). Thirty patients (58%) in the high IOP group and 74 patients (56%) in the control group were male ( P = .87). Age and gender were not significantly different between these 2 patient groups.
The most commonly isolated organisms are listed in Table 1 . Twelve patients (23.1%) in the high IOP group and 22 patients (16.6%) in the control group had polymicrobial infections. Coagulase-positive staphylococci (6/52, 11.5%) and Streptococcus species (6/52, 11.5%) were the most common isolates in the high IOP group. Coagulase-positive staphylococci (20/132, 15.2%) and Pseudomonas aeruginosa (20/132, 15.2%) were the most common pathogens detected in the control group. Fungal infection accounted for 9.6% (5/52) of ulcers in the high IOP group and for 15.2% (20/132) of ulcers in the control group. Candida species were the most common cause of fungal keratitis in both groups. Acanthamoeba was isolated in 4 cases in the control group and in 1 case in the high IOP group. Neither gram-positive nor gram-negative organisms were associated with the risk of developing elevated IOP in patients with acute microbial keratitis ( P = .98).
| Culture Result | Keratitis with High IOP, n (%) | Keratitis with Normal IOP, n (%) |
|---|---|---|
| Uniorgansim | 40 (76.92%) | 110 (83.33%) |
| Gram-positive bacteria | 20 (38.46%) | 58 (43.93%) |
| Coagulase-positive staphylococci | 6 (11.53%) | 20 (15.15%) |
| Coagulase-negative staphylococci | 5 (9.61%) | 18 (13.63%) |
| Streptococcus species | 6 (11.53%) | 10 (7.57%) |
| Other | 3 (5.76%) | 10 (7.57%) |
| Gram-negative bacteria | 14 (26.92%) | 28 (21.21%) |
| Pseudomonas aeruginosa | 4 (7.96%) | 20 (15.15%) |
| Other | 8 (15.38%) | 8 (6.06%) |
| Fungi | 5 (9.61%) | 20 (15.15%) |
| Candida species | 3 (5.76%) | 8 (6.06%) |
| Mold | 2 (3.84%) | 12 (9.09%) |
| Acanthamoeba | 1 (1.92%) | 4 (3.03%) |
| Multiorganism | 12 (23.07%) | 22 (16.55%) |
Using univariate analysis, we found that the risk factors associated with high IOP were poor VA at presentation ( P = .002; Figure 1 , Right), diabetes mellitus ( P = .004), contact lens wear ( P = .030), prior ocular surgery ( P < .001), and glaucoma ( P = .020), whereas ocular trauma, exposure keratitis, history of ocular herpes virus infection, topical corticosteroid use, and systemic immunosuppressant therapy were not associated with increased IOP ( Table 2 ). Sixty-five percent of (34/52) ulcers were categorized as large in the high IOP group, whereas only 32% (48/132) were defined as large in the control group. Ulcer size, but not ulcer location, was statistically associated ( P < .001) with IOP elevation ( Table 3 ).
| Risk Factors | Presence | High IOP Group, n (%) | Normal IOP Group, n (%) | P Value a |
|---|---|---|---|---|
| Contact lens wear b | Yes | 14 (27%) | 59 (45%) | .030 |
| No | 38 (73%) | 73 (55%) | ||
| Recent ocular trauma | Yes | 4 (8%) | 15 (11%) | .60 |
| No | 48 (92%) | 117 (89%) | ||
| History of ocular surgery b | Yes | 32 (62%) | 42 (32%) | < .001 |
| No | 20 (38%) | 90 (68%) | ||
| Diabetes mellitus b | Yes | 15 (29%) | 14 (11%) | .004 |
| No | 37 (71%) | 118 (89%) | ||
| History of neurotrophic keratopathy | Yes | 4 (8%) | 9 (7%) | .76 |
| No | 48 (92%) | 123 (93%) | ||
| History of exposure keratopathy | Yes | 9 (17%) | 16 (12%) | .35 |
| No | 43 (83%) | 116 (88%) | ||
| Glaucoma b | Yes | 15 (29%) | 18 (14%) | .020 |
| No | 37 (71%) | 114 (86%) | ||
| Immunosuppression therapy | Yes | 4 (8%) | 16 (12%) | .44 |
| No | 48 (92%) | 116 (88%) | ||
| Use of topical corticosteroids | Yes | 10 (19%) | 20 (15%) | .51 |
| No | 42 (81%) | 112 (85%) |
| Risk Factors | High IOP Group, n (%) | Normal IOP Group, n (%) | P Value a |
|---|---|---|---|
| Ulcer size (mm) b | |||
| Small (< 2) | 9 (17%) | 42 (32%) | < .001 |
| Medium (2 to 3.9) | 9 (17%) | 48 (36%) | |
| Large (≥ 4) | 34 (65%) | 42 (32%) | |
| Ulcer location | |||
| Central | 17 (33%) | 34 (26%) | .35 |
| Paracentral | 24 (46%) | 57 (43%) | |
| Peripheral | 11 (21%) | 41 (31%) |
Multivariate analysis was performed to investigate whether each significant risk factor identified in the univariate analysis increased the risk of IOP elevation independently. We found that size of the ulcer, history of diabetes mellitus, VA at presentation, and a history of ocular surgery were associated with an increase in IOP, whereas contact lens wear and glaucoma were not ( Table 4 ).
| Risk Factors | Odds Ratio | 95% Confidence Interval | P Value |
|---|---|---|---|
| Contact lens wear | 0.73 | 0.33 to 1.63 | .44 |
| Glaucoma | 1.20 | 0.46 to 3.11 | .71 |
| History of ocular surgery a | 2.80 | 1.24 to 6.30 | .013 |
| Diabetes mellitus a | 3.33 | 1.32 to 8.41 | .011 |
| Ulcer size a | |||
| Large (≥ 4 mm) | 1.0 (Referent) | ||
| Medium (2 to 3.9 mm) | 0.29 | 0.12 to 0.71 | .007 |
| Small (< 2 mm) | 0.20 | 0.081 to 0.51 | < .001 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
