Introduction
Infectious keratitis remains a major cause of visual impairment worldwide.1–3 In tropical environments, fungal keratitis makes up a large proportion of keratitis cases.4, 5 Relative to bacterial corneal ulcers, fungal ulcers are typically thought to be more difficult to treat and to result in worse outcomes.6 We previously reported that at 3 months after diagnosis, fungal ulcers were more likely to have a larger scar size and to have perforated than were bacterial ulcers, but did not have significantly worse best spectacle corrected visual acuity (BSCVA).7 In a separate report, we found that bacterial corneal ulcers experienced significant improvements in BSCVA after 3 months.8 Thus, it is possible that the differences between bacterial and fungal ulcers that we observed at 3 months could change with longer follow-up. As an attempt to better describe the long-term visual outcomes of bacterial and fungal corneal ulcers, we called back a subset of study participants from the earlier comparative study for a four-year follow-up visit. Here, we compare presenting visual acuity, BSCVA, and hard contact lens corrected visual acuity (CLVA) in bacterial and fungal ulcers to better understand the natural history of infectious corneal ulcers.
Methods
Study design.
This is a prospective cohort study that drew participants from two randomized clinical trials conducted at Aravind Eye Hospital in Madurai, India—one of which enrolled fungal corneal ulcers and one of which enrolled bacterial ulcers. The two trials overlapped in time and had similar inclusion and exclusion criteria and outcomes, which allowed a direct comparison. The first trial was a multi-center therapeutic exploratory trial that randomized participants with smear-positive fungal keratitis to topical natamycin or voriconazole (clinicaltrials.gov #NCT00557362).9 The second trial was the multi-center Steroids for Corneal Ulcers Trial (SCUT), which randomized participants with culture-positive bacterial keratitis who had been treated with at least 48 hours of moxifloxacin to either topical corticosteroid or placebo (clinicaltrials.gov #NCT00324168).10 Aside from the causative organism, inclusion and exclusion criteria for the trials were similar. In both trials, patients with No Light Perception (NLP) vision in the affected eye or BSCVA worse than 20/200 in the unaffected eye were excluded.9, 11 In both trials, best spectacle corrected visual acuity (BSCVA) was assessed at enrollment and 3 months. For the present study, we invited for a 4-year follow-up visit any study participants who had been enrolled for either trial at Aravind Eye Hospital Madurai from October 8, 2007 to August 18, 2008. Note that comparison of the two trials was facilitated by restricting cases to only those enrolled at a single study site and to only those enrolled during a common time window. The study received ethical approval from the Aravind Eye Hospital Institutional Review Board and the University of California, San Francisco Committee on Human Research.
Assessment.
We used similar methods for the current study as for the original trials, including refraction protocols and slit lamp examination protocols.9, 11 Certified refractionists assessed presenting visual acuity (i.e., visual acuity corrected with the participant’s current eyeglasses, or uncorrected visual acuity if the participant did not use eyeglasses), BSCVA, and CLVA in logMAR units using the ETDRS visual acuity chart.12 Refractionists evaluated subjects with visual acuity worse than logMAR 1.6 (Snellen equivalent 20/800) for counting fingers, hand motions, light perception, or no light perception visual acuity, which were assigned a logMAR of 1.7, 1.8, 1.9, or 2.0, respectively. Participants whose BSCVA was worse than logMAR 0.22 (Snellen equivalent 20/32) received hard contact lens over-refraction. We performed slit lamp biomicroscopy and dilated fundus examinations for all study participants. Refractionists were masked to the treatment allocation at each visit and to the causative organism at the four-year visit.
Statistical analysis.
We performed descriptive statistics and made comparisons between the fungal keratitis group and bacterial keratitis group using the Wilcoxon rank sum test for continuous variables and Fisher’s exact test for categorical variables. We compared the proportion of participants meeting the World Health Organization threshold for visual impairment (visual acuity worse than 20/60) and blindness (visual acuity worse than 20/400) in a logistic regression adjusted for baseline visual acuity. We conducted all visual acuity analyses with logMAR acuities but used Snellen equivalents when describing the results.13 We reported scar size as the geometric mean of the largest scar diameter and its perpendicular width, and imputed scar sizes at the 4-year visit for eyes that had undergone keratoplasty by using the last observation carried forward (i.e., the 3-month visit for all eyes).
Results
Study population.
Between October 8, 2007 and August 18, 2008, Aravind Eye Hospital Madurai enrolled 80 study participants with bacterial keratitis and 72 study participants with fungal keratitis into the respective trials. Of these, we conducted 4-year follow-up examinations for 50 (62.5%) participants with bacterial keratitis and 50 (69.4%) participants with fungal keratitis. The baseline characteristics of study participants with fungal keratitis who were lost to follow-up were not significantly different from those who had a 4-year follow-up visit in terms of age, sex, visual acuity, or scar size. As reported elsewhere, the study participants with bacterial keratitis who were lost to follow-up had worse enrollment visual acuity compared with those who returned for the 4-year visit.14 Baseline characteristics are shown in Table 1; visual acuity was similar between bacterial and fungal ulcers, but fungal ulcers tended to be larger.
Table 1.
Enrollment characteristics of study participants and those lost to follow-up (LTFU) at four years
| Fungal | Bacterial | ||||
|---|---|---|---|---|---|
| Enrollment Characteristic | Included N=50 |
LTFU N=22 |
Included N=50 |
LTFU N=30 |
|
| Female sex, N (%) | 20 (40.0%) | 6 (27.3%) | 25 (50%) | 10 (33.3%) | |
| Age, median (IQR) | 48 (33–58) | 55 (48–60) | 45 (38–60) | 56 (37–65) | |
| logMAR BSCVA,a median (IQR) | 0.61 (0.30–1.7) | 0.80 (0.24–1.8) | 0.65 (0.36–1.6) | 1.36 (0.64–1.7) | |
| Infiltrate size, mm, median (IQR) | 3.5 (2.2–5.1) | 3.6 (2.7–4.8) | 2.5 (1.9–3.7) | 3.2 (1.9–5.1) | |
| Organism, N (%) | |||||
| Fusarium species | 17 (34.0%) | 6 (27.3%) | -- | -- | |
| Aspergillus species | 6 (12.0%) | 3 (13.6%) | -- | -- | |
| Bipolaris species | 6 (12.0%) | 0 (0%) | -- | -- | |
| Curvularia species | 4 (8.0%) | 3 (13.6%) | -- | -- | |
| Other fungus | 17 (34.0%) | 10 (45.5%) | -- | -- | |
| S. pneumoniae | -- | -- | 21 (42.0%) | 15 (51.7%) | |
| Nocardia species | -- | -- | 14 (28.0%) | 4 (13.3%) | |
| P. aeruginosa | -- | -- | 12 (24.0%) | 7 (23.3%) | |
| Other bacteria | -- | -- | 3 (6.0%) | 4 (13.3%) | |
Abbreviations: LTFU=lost to follow-up; IQR=interquartile range; N=number
Best spectacle corrected visual acuity, assessed in logarithm of the minimum angle of resolution (logMAR) units
Presenting vision.
Of the 100 study participants who had 4-year follow-up, 9 were wearing eyeglasses at the time of the study visit (5 in the fungal keratitis group and 4 in the bacterial keratitis group). The median presenting logMAR visual acuity (i.e., acuity with current spectacle correction for those with eyeglasses and uncorrected visual acuity for those without eyeglasses) at 4 years in the affected eye was 0.7 (Snellen equivalent 20/100; interquartile range [IQR] 0.3 to 1.5 logMAR units) for the fungal keratitis group and 0.6 (Snellen equivalent 20/80; IQR 0.3 to 1.0 logMAR units) for the bacterial keratitis group. Blindness of the affected eye, defined as visual acuity worse than 20/400, was more common in the fungal group than the bacterial group (Table 2). The median logMAR presenting vision in the fellow eye at 4 years was 0.3 (Snellen equivalent 20/40) for both the fungal and bacterial group. Bilateral visual impairment, defined as presenting visual acuity worse than 20/60 in the better-seeing eye, was present in 12 (24.0%) individuals in the fungal group and in 10 (20.0%) in the bacterial group.
Table 2.
Visual impairment in eyes with fungal and bacterial keratitis four years following diagnosis and treatment
| Fungal N=50 |
Bacterial N=50 |
OR (95%CI)* | P-valuea | |
|---|---|---|---|---|
| Presenting vision | ||||
| Worse than 20/60 | 32 | 30 | 1.18 (0.48–2.86) | 0.72 |
| Worse than 20/400 | 13 | 4 | 5.00 (1.30–19.2) | 0.02 |
| BSCVA | ||||
| Worse than 20/60 | 16 | 12 | 1.54 (0.54–4.43) | 0.42 |
| Worse than 20/400 | 12 | 4 | 4.19 (1.11–15.8) | 0.04 |
| CLVA | ||||
| Worse than 20/60 | 14 | 7 | 2.71 (0.85–8.59) | 0.09 |
| Worse than 20/400 | 12 | 3 | 5.74 (1.37–24.1) | 0.02 |
Abbreviations: BSCVA, best spectacle-corrected visual acuity; CI, confidence interval; CLVA, hard contact lens-corrected visual acuity; OR, odds ratio
Logistic regression adjusted for best spectacle corrected visual acuity at baseline; odds ratio compares fungal relative to bacterial ulcers
BSCVA.
Best-spectacle correction improved visual acuity in the affected eye for most participants in the fungal group (median 1.9-line improvement over presenting visual acuity, IQR 0 to 4.4) and bacterial group (median 2.8-line improvement over presenting visual acuity, IQR 1.0 to 4.6) at the 4-year study visit (P=0.27 comparing fungal with bacterial ulcers). As shown in Supplemental Figure 1, spectacle correction was generally not effective for those whose presenting vision was worse than 20/400. Supplemental Figure 2 depicts the BSCVA in the affected eye for each study participant at enrollment, 3 months, and 4 years. Although the median BSCVA was similar in the fungal and bacterial groups (median Snellen equivalent 20/32 in each; P=0.90; Table 3), the fungal ulcer group was more likely to have BSCVA of worse than 20/400 at the 4-year follow-up visit (12 of 50 fungal ulcers compared with 4 of 50 bacterial ulcers; OR 4.19, 95%CI 1.11 to 15.8, P=0.03, logistic regression adjusting for BSCVA at enrollment; Table 2 and Figure 1). We refracted both eyes of all patients; appropriate eyeglasses would reduce the number of bilaterally visually impaired (i.e., worse than 20/60) study participants to 4 (8.0%) in the fungal group and zero in the bacterial group (P=0.06, Fisher’s exact test).
Table 3:
Visual acuity four years after treatment for infectious keratitis
| logMAR Visual Acuity Median (Interquartile Range) |
||
|---|---|---|
| Fungal N=50 |
Bacterial N=50 |
|
| Presenting Visual Acuitya | ||
| Affected Eye | 0.7 (0.3 to 1.5) | 0.6 (0.3 to 1.0) |
| Fellow Eye | 0.3 (0 to 0.78) | 0.3 (0 to 0.48) |
| Better Seeing Eye | 0.19 (0 to 0.48) | 0.3 (0 to 0.48) |
| BSCVA | ||
| Affected Eye | 0.16 (0.02 to 0.78) | 0.22 (0.02 to 0.42) |
| Fellow Eye | 0 (0 to 0.18) | 0.1 (0 to 0.2) |
| Better Seeing Eye | 0 (0 to 0.18) | 0.05 (0 to 0.2) |
Abbreviations: BSCVA, best spectacle corrected visual acuity
Presenting visual acuity is defined by the World Health Organization as visual acuity obtained with currently available refractive correction, if any. Here, 5 participants in the fungal group and 4 participants in the bacterial group had visual acuity tested with their eyeglasses and the remainder were tested without eyeglasses.
Figure 1. Presenting visual acuity, best spectacle-corrected visual acuity (BSCVA), and contact lens-corrected visual acuity (CLVA) at 4 years in eyes with fungal versus bacterial keratitis.
The distribution of logMAR visual acuity is shown for each vision outcome as a dotplot superimposed on a violin plot, stratified by causative organism. Each point represents the visual acuity of a single eye. CLVA was assessed only for eyes with a BSCVA worse than logMAR 0.22; for the purposes of this graph the BSCVA was substituted for eyes not undergoing CLVA.
CLVA.
Contact lens over-refraction was performed for participants with a logMAR BSCVA of >0.22 (Snellen equivalent 20/32; N=21 in fungal group and N=21 in bacterial group). Of these, contact lenses improved visual acuity beyond BSCVA in 10 (47.6%) eyes in the fungal keratitis group and 14 (66.7%) in the bacterial keratitis group. Among those who underwent contact lens over-refraction, contact lens visual acuity improved upon BSCVA by a median of 0 lines (IQR 0 to 0.8 lines) in the fungal group and 0.8 lines (IQR 0 to 2.4) in the bacterial group. After contact lens over-refraction, 32 study eyes (64.0%) from the fungal group and 33 (66.0%) from the bacterial group had a corrected visual acuity of logMAR 0.22 (Snellen equivalent 20/32) or better (P=0.96, logistic regression adjusted for baseline BSCVA). As shown in Table 2, more eyes in the fungal group had CLVA worse than 20/60 at 4 years (14 participants vs. 7; P=0.09 after adjusting for enrollment BSCVA) and CLVA worse than 20/400 at 4 years (12 participants vs. 3, P=0.02 after adjusting for enrollment BSCVA).
Perforation/keratoplasty.
By the time of the 4-year study visit, perforation had occurred in 8 study participants from the fungal group had 0 from the bacterial group (P=0.01, Wilcoxon rank sum test). Moreover, 7 study participants with fungal keratitis had received a penetrating keratoplasty and none with bacterial keratitis had (P=0.01, Wilcoxon rank sum test). Each perforation and keratoplasty occurred before the 3-month visit. The post-operative vision of eyes undergoing keratoplasty generally remained poor at 4 years: while 1 participant had a BSCVA of 0.12 logMAR units, the remaining 6 participants had a BSCVA of worse than 20/400, and none of these achieved substantive improvement with CLVA. Of these, 4 had a failed graft, 1 had a dense pupillary membrane, and 1 had a central corneal scar. The other major surgical intervention since enrollment in the respective studies was cataract surgery, which had been performed in the 4-year study interval in 7 (14.0%) of the fungal group and 6 (12.0%) of the bacterial group (P=0.50, Fisher’s exact test).
Corneal scars.
The median scar size at the 4-year follow-up visit was 3.3mm (IQR 2.1 to 5.4) in the fungal group and 2.9mm (IQR 1.9 to 4.4) in the bacterial group (P=0.20, Wilcoxon rank sum test). Scar size and BSCVA were correlated at the 4-year visit, with each 1 mm increase in size associated with a 2.1-line improvement in vision (95%CI 1.6 to 2.6 lines). However, scar size was not significantly different between the two groups when accounting for infiltrate size at enrollment (scars in the fungal group were on average 0.1mm smaller, 95%CI 0.6mm smaller to 0.4mm larger; P=0.67; see Supplemental Figure 3 for representative images). On slit lamp examination, a substantial number of study participants had multiple corneal scars in the study eye: 8 (16.0%) in the fungal group and 14 (28.0%) in the bacterial group. Moreover, many study participants had scars in the fellow eye: 21 (42.0%) in the fungal group and 27 (54.0%) in the bacterial group. Bilateral visual impairment, defined as presenting acuity worse than 20/60 in each eye, was more common in those who also had scars in the fellow eye (13/48; 27%) relative to those without scars in the fellow eye (9/52; 17%), though the difference was not statistically significant (P=0.33).
Discussion
Four years after being enrolled in a bacterial or fungal keratitis trial at a single center in South India, the majority of study participants experienced visual impairment in the affected eye, and almost one-quarter experienced bilateral visual impairment. Spectacle correction would greatly reduce the number of bilateral visual impairment, yet only 9% of study participants were wearing eyeglasses at the time of the 4-year follow-up visit. Study participants treated for fungal keratitis were more likely to have counting fingers vision or worse compared with those treated for bacterial keratitis, even after accounting for the worse visual acuity in fungal ulcers at trial enrollment.
Fungal ulcers had worse outcomes than bacterial ulcers in this study, with a greater proportion having visual impairment or blindness after spectacle correction and contact lens correction and a greater proportion experiencing perforation and subsequent therapeutic penetrating keratoplasty. In addition, a greater proportion in the fungal group had bilateral visual impairment after refraction, although this was not statistically significant. A previous study comparing 3-month outcomes in a much larger group of patients from the same set of randomized trials found that fungal corneal ulcers had larger scar sizes and were more likely to perforate relative to bacterial ulcers.7 The present study found similar results in terms of perforation, but not scar size, perhaps because of the limited sample size.
Fungal keratitis cases were generally worse at study enrollment, which could account for these findings; however, poor visual outcomes were more likely in the fungal group at 4 years even after adjusting for baseline visual acuity. The difference between bacterial and fungal cases could also be due in part to poor outcomes in perforated or nearly perforated eyes: the 7 eyes with fungal keratitis that underwent therapeutic keratoplasty had poor long-term visual outcomes, with only 1 of 7 patients achieving a post-operative visual acuity of better than 20/400 at the 4-year study visit, and none improving with a hard contact lens. In contrast, none of the bacterial keratitis cases in this study population experienced perforation. Perforation did not explain the entire difference, however, as 6 of the 11 fungal cases with BSCVA worse than 20/400 at the 4-year study visit had not experienced a perforation. Supplemental Figure 2 shows that even as early as the 3-month visit, the fungal keratitis cases with poor enrollment visual acuity were less likely to experience improvement in BSCVA than were the bacterial cases. This lack of improvement could have been due to inadequate clearance of organisms and/or difficulty managing the host immune response, both of which are relatively more challenging with fungal keratitis than with bacterial keratitis.
It is notable that many patients with poor visual acuity in this South Indian setting did not pursue optical penetrating keratoplasty once the keratitis had resolved, and over 90% did not wear spectacle correction. This resulted in a considerable amount of correctable visual impairment in this population. The vast majority of study participants achieved excellent visual acuity with spectacle correction alone, and would not require hard contact lenses. This is especially encouraging in a place like India, where it may not be practical for the agricultural workers most at risk for infectious keratitis to wear hard contact lenses. Nonetheless, the low uptake of eyeglasses demonstrated in this study highlights the unmet need for refractive error correction in this population.
Corneal ulcers are thought of as a unilateral condition that does not cause bilateral blindness. However, we show here that approximately 20% of study participants with a corneal ulcer had bilateral visual impairment worse than 20/60 four years after developing the ulcer. This may be due in part to corneal ulcers in both eyes, since almost half of participants had corneal scars in the non-study eye at the 4-year follow-up visit. Participants with scars in the fellow eye were more likely to have bilateral visual impairment at 4 years, though this was not statistically significant. These results are especially noteworthy because an exclusion criterion for both the fungal and bacterial trials was visual acuity worse than 20/200 in the non-affected eye, and many patients excluded from the trials had corneal scars in the non-affected eye.15 These findings (i.e., the high prevalence of bilateral scars and bilateral blindness) suggest that corneal ulceration is frequently a bilateral condition with the potential to cause bilateral visual impairment. We speculate that study participants may have been at continued risk for corneal trauma and infection in this South Indian setting as a result of agricultural work.
In this study, cases of fungal keratitis and bacterial keratitis were enrolled into separate clinical trials at a single center during the same time period, allowing us to compare outcomes between the two groups four years after enrollment. The inclusion and exclusion criteria, study design, and outcomes of the trials were similar and enrollment occurred during the same time window, making the two groups inherently comparable. Moreover, we masked the refractionists to the causative organism to reduce bias. We acknowledge several limitations. The study has a relatively high rate of loss to follow-up, especially in the bacterial keratitis group. It is possible that participants with bacterial ulcers had less of an incentive to return for a 4-year follow-up visit due to better visual outcomes, although such reasoning is speculative. Patients lost to follow-up appeared to have worse ulcers in both the bacterial and fungal groups, but the possibility of differential loss to follow-up and subsequent bias cannot be ruled out. The study participants do not form a population-based sample: the most severe ulcers were excluded from the trials. It is possible that one of the groups was worse off at baseline but not enrolled into the respective trial, in which case we could have missed a larger difference between the bacterial and fungal ulcer groups. We did not perform contact lens visual acuity for all participants due to logistical constraints, and so could not compare this outcome among all study participants. Because the results come from a single center in India, the generalizability of the findings to other settings with a different spectrum of causative organisms is not clear.
In summary, we showed that correctable visual impairment was relatively common in a South Indian population treated 4 years prior for infectious keratitis. Fungal ulcers were more likely to have counting fingers vision or worse and were more likely to have perforated than were bacterial ulcers. These results support the notion that early and accurate diagnosis of microbial keratitis, especially of fungal keratitis, is important to prevent long-term visual impairment. Better treatments for fungal corneal infections and their inflammatory sequelae are needed.
Supplementary Material
Supplemental Figure 1. Spectacle and contact lens correction in eyes with an episode of fungal or bacterial keratitis four years prior. Each bar represents an eye previously affected with (A) bacterial keratitis or (B) fungal keratitis. The presenting visual acuity (current spectacle correction for those with eyeglasses and uncorrected visual acuity for those without eyeglasses) is shown as a black line at the top of each bar. Visual acuity improvements afforded by spectacle correction are shown in light blue and by hard contact lenses in dark blue. Light grey horizontal lines represent lines of logMAR visual acuity, from no light perception (NLP; logMAR 2.0) at the top to 20/15 (logMAR −0.1) at the bottom.
Supplemental Figure 2. Best spectacle corrected visual acuity (BSCVA) at enrollment, 3 months, and four years following treatment for fungal or bacterial keratitis. Each horizontal row represents one of the 72 eyes with fungal keratitis or 80 eyes with bacterial keratitis eligible during the study period. Boxes are shaded proportional to logMAR BSCVA, with yellow boxes representative of 20/20 vision and red boxes representative of No Light Perception vision. Eyes are ordered based on enrollment BSCVA, from best on the bottom to worst at top. An asterisk indicates that a therapeutic penetrating keratoplasty occurred between study visits. Study visits with missing data for BSCVA are grey.
Supplemental Figure 3. Corneal opacity at enrollment, 3 months, and 4 years. Photographs were chosen randomly for one participant with a bacterial ulcer and one participant with a fungal ulcer.
Acknowledgements
a. Funding/support:
The research was supported by the National Institutes of Health – National Eye Institute (Bethesda, MD; grants U10EY015114, K23EY019071, and P30EY002162), That Man May See (San Francisco, CA), the South Asia Research Fund (San Francisco, CA), and Research to Prevent Blindness (New York, NY).
b. Financial Disclosures:
No financial disclosures.
c. Other Acknowledgements:
Subbaiah Rengasamy Sumithra, Ophthalmic Assistant, and Thirukkonda Subramanian Chandravathi, MA, coordinated both the original trial activities and this four-year follow-up study at Aravind Eye Hospital, Madurai.
References
- 1.Pascolini D, Mariotti SP. Global estimates of visual impairment: 2010. Br J Ophthalmol 2012;96(5):614–618. [DOI] [PubMed] [Google Scholar]
- 2.Flaxman SR, Bourne RRA, Resnikoff S, et al. Global causes of blindness and distance vision impairment 1990–2020: a systematic review and meta-analysis. Lancet Glob Health 2017;5(12):e1221–e1234. [DOI] [PubMed] [Google Scholar]
- 3.Ung L, Bispo PJ, Shanbhag SS, Gilmore MS, Chodosh J. The Persistent Dilemma of Microbial Keratitis: Global Burden, Diagnosis, and Antimicrobial Resistance. Surv Ophthalmol 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Shah A, Sachdev A, Coggon D, Hossain P. Geographic variations in microbial keratitis: an analysis of the peer-reviewed literature. Br J Ophthalmol 2011;95(6):762–767. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Lalitha P, Prajna NV, Manoharan G, et al. Trends in bacterial and fungal keratitis in South India, 2002–2012. Br J Ophthalmol 2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Wong TY, Ng TP, Fong KS, Tan DT. Risk factors and clinical outcomes between fungal and bacterial keratitis: a comparative study. CLAO J 1997;23(4):275–281. [PubMed] [Google Scholar]
- 7.Prajna NV, Srinivasan M, Lalitha P, et al. Differences in clinical outcomes in keratitis due to fungus and bacteria. JAMA Ophthalmol 2013;131(8):1088–1089. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Srinivasan M, Mascarenhas J, Rajaraman R, et al. Visual recovery in treated bacterial keratitis. Ophthalmology 2014;121(6):1310–1311. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Prajna NV, Mascarenhas J, Krishnan T, et al. Comparison of natamycin and voriconazole for the treatment of fungal keratitis. Arch Ophthalmol 2010;128(6):672–678. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Srinivasan M, Mascarenhas J, Rajaraman R, et al. Corticosteroids for bacterial keratitis: the Steroids for Corneal Ulcers Trial (SCUT). Arch Ophthalmol 2012;130(2):143–150. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Srinivasan M, Mascarenhas J, Rajaraman R, et al. The steroids for corneal ulcers trial: study design and baseline characteristics. Arch Ophthalmol 2012;130(2):151–157. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Resnikoff S, Pascolini D, Mariotti SP, Pokharel GP. Global magnitude of visual impairment caused by uncorrected refractive errors in 2004. Bull World Health Organ 2008;86(1):63–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Diabetic Retinopathy Clinical Research Network. Visual Acuity Score Conversion Chart.
- 14.McClintic SM, Prajna NV, Srinivasan M, et al. Visual outcomes in treated bacterial keratitis: four years of prospective follow-up. Invest Ophthalmol Vis Sci 2014;55(5):2935–2940. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Oldenburg CE, Sy A, Srinivasan M, et al. Pre-existing blindness in a cohort of patients with bacterial keratitis. Br J Ophthalmol 2011;95(11):1612–1613. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental Figure 1. Spectacle and contact lens correction in eyes with an episode of fungal or bacterial keratitis four years prior. Each bar represents an eye previously affected with (A) bacterial keratitis or (B) fungal keratitis. The presenting visual acuity (current spectacle correction for those with eyeglasses and uncorrected visual acuity for those without eyeglasses) is shown as a black line at the top of each bar. Visual acuity improvements afforded by spectacle correction are shown in light blue and by hard contact lenses in dark blue. Light grey horizontal lines represent lines of logMAR visual acuity, from no light perception (NLP; logMAR 2.0) at the top to 20/15 (logMAR −0.1) at the bottom.
Supplemental Figure 2. Best spectacle corrected visual acuity (BSCVA) at enrollment, 3 months, and four years following treatment for fungal or bacterial keratitis. Each horizontal row represents one of the 72 eyes with fungal keratitis or 80 eyes with bacterial keratitis eligible during the study period. Boxes are shaded proportional to logMAR BSCVA, with yellow boxes representative of 20/20 vision and red boxes representative of No Light Perception vision. Eyes are ordered based on enrollment BSCVA, from best on the bottom to worst at top. An asterisk indicates that a therapeutic penetrating keratoplasty occurred between study visits. Study visits with missing data for BSCVA are grey.
Supplemental Figure 3. Corneal opacity at enrollment, 3 months, and 4 years. Photographs were chosen randomly for one participant with a bacterial ulcer and one participant with a fungal ulcer.