Author Affiliations: Department of Ocular Inflammation and Immunology, Singapore National Eye Centre (Drs Bacsal, Chee, and Cheng); Scientific Committee, Singapore Eye Research Institute (Dr Chee); Department of Ophthalmology, National University Hospital (Dr Chee); and Department of Ophthalmology, Yong Loo Lin School of Medicine (Dr Chee), National University of Singapore; and Quality Management Department, Singapore Health Services (Dr Flores).
To describe the clinical spectrum of fundus manifestations and angiographic and optical coherence tomographic features of dengue-associated maculopathy in a large series.
We reviewed clinical records of patients diagnosed as having dengue maculopathy at the Singapore National Eye Centre between January 1, 2002, and December 31, 2005.
We identified 41 patients with serological evidence of dengue fever who had ocular signs and symptoms not attributable to other diseases within 1 month after onset of symptoms of dengue. Seventy-one eyes had maculopathy. Mean best-corrected visual acuity in the affected eye was 20/40 (range, hand motions to 20/20). Intraretinal hemorrhages were seen in 45% of eyes, usually in association with venous sheathing. Fundus fluorescein angiography demonstrated venular occlusion in 25% or arteriolar and/or venular leakage in 3% and 13%, respectively. Yellow subretinal dots were an unusual finding in 28%. Of these, 50% showed corresponding hypofluorescent spots on indocyanine green angiography. Central or paracentral scotomas were observed in 63%. Twenty-eight patients received steroid treatment. Mean visual acuity showed significant improvement between weeks 2 and 4, with an increasing proportion of eyes achieving a best-corrected visual acuity of 20/40 or better across time.
Fundus fluorescein and indocyanine green angiography, optical coherence tomography, and visual field testing are useful tools in the diagnosis of dengue maculopathy.
Dengue fever, the most common mosquito-borne viral disease in humans, is a multisystemic disease with known complications.1- 7 It has also been found to affect the eye.8- 12 The report of Wen et al8 on the ocular complications of dengue in 24 patients is the largest series thus far. Common findings include retinal hemorrhages, retinal edema, optic disc swelling, and vitritis. Fundus fluorescein angiography (FFA) showed leakage at optic disc and retinal vessels, macular nonperfusion, and retinal pigment epithelium (RPE) defects, suggesting that the retina and RPE were the primary sites of involvement. Poor choroidal flushing suggested choroidal involvement as well, supported by Lim et al,13 who also performed indocyanine green angiography (ICG) in their small series years after that by Wen et al.8
Despite all these reports, the pathophysiologic mechanism of dengue fever and its effect on the eye is still unclear. We present a large, comprehensive series that describes not only FFA and visual field results but also optical coherence tomography (OCT) and ICG findings associated with dengue fever to further our understanding of the ocular sequelae of this condition.
This study involved patients who were seen at the Singapore National Eye Centre between January 1, 2002, and December 31, 2005, and were diagnosed as having dengue maculopathy based on the following criteria: (1) clinical diagnosis of dengue fever according to the criteria set by the World Health Organization14 and (2) presence of fundus lesions that were not attributable to other conditions with known ocular involvement, such as diabetes mellitus, within 1 month after onset of symptoms of dengue fever. A senior ophthalmologist (S.-P.C.) confirmed the presence of maculopathy. Patients who satisfied both criteria were included in this series.
Patients diagnosed as having maculopathy based on fundus examination underwent the following baseline ancillary tests: FFA and ICG (Topcon 50IA retina camera; Topcon, Tokyo, Japan), automated perimetry (Zeiss Humphrey Field Analyzer 750; Carl Zeiss Ophthalmic Systems, Dublin, Calif) or Amsler grid testing, and OCT (Zeiss OCT 3 version 4; Carl Zeiss Ophthalmic Systems). In cases of abnormal baseline results, field testing was performed at every visit to assess progression or resolution of disease. Although Humphrey visual field (HVF) testing was preferred, Amsler grid testing was performed if patients refused to give consent for the automated test. Institutional review board approval was obtained before data were collected.
Medical records were reviewed for demographic data (age, sex, and race), ocular symptoms, ocular findings, Snellen visual acuity, treatment, and outcome. Angiographic and OCT findings were reported based on the consensus of 3 masked investigators (K.E.B., S.-P.C., and C.-L.C.). The size and severity of scotoma on HVF tests were assessed based on our modification of the Advanced Glaucoma Intervention Study scoring system.15 We arbitrarily selected to study only the central field, composed of the inner 2 rows of each hemifield, excluding the blind spot and nasal fields. The number of adjacent depressions in the central field was counted using a reference template. To account for the density of the scotoma, additional points were awarded based on the Advanced Glaucoma Intervention Study system. The final score, reflective of the size and density of the scotoma, was referred to as the dot score.
Systemic screening to exclude autoimmune-mediated disorders (including antinuclear antibodies, double-stranded DNA antibodies, anticytoplasmic antibodies, C3, C4, and urine microalbumin) was performed on the initial patients. Subsequent patients were tested for C3, C4, and urine microalbumin alone, since only these 3 parameters were found to occasionally produce abnormal levels. Serial determinations were performed in the event of abnormal levels on initial testing. Platelet counts, when available, were obtained from clinical records.
Patients were treated with corticosteroids based on the postulated immune-mediated pathogenesis of disease. In general, we treated patients who were symptomatic or had visual acuity of 20/40 or worse in the affected eye and subjectively had persistent or progressive deterioration of vision. Unless contraindicated, patients with bilateral disease received daily oral prednisolone, 1 mg/kg, tapered according to ocular response or development of adverse drug reactions. If visual acuity was severely depressed (20/200 or worse) and/or in the absence of improvement after 3 days of initial treatment, intravenous methylprednisolone, 1 g/d for 3 days, was given followed by oral prednisolone. Additional treatment with intravenous immunoglobulins, 400 mg/kg, was given for 5 days in cases in which no clinical improvement was noted after 3 days of intravenous methylprednisolone. Unilateral disease was treated with periocular steroids unless contraindicated (low platelet counts). No treatment was given if the initial visual acuity was better than 20/40 or if the patient reported subjective improvement from the initial decrease in vision. Descriptive statistics were generated using Epi Info statistical software, version 6.04d (Centers for Disease Control and Prevention, Atlanta, Ga). Both χ2 tests and analysis of variance were used where appropriate. Results are presented as mean ± SD.
A 33-year-old Chinese woman was hospitalized for dengue fever. On the second day of admission (6 days after fever onset), she complained of an acute decrease in vision in both eyes and was referred for ophthalmic evaluation.
Her best-corrected visual acuity (BCVA) at initial examination was 20/400 OU. Pupillary responses and intraocular pressure were normal. Anterior chamber cells were occasionally found in both eyes. She had bilateral swollen and hyperemic optic discs and venular and arteriolar sheathing in the posterior pole. A small vitreous hemorrhage was found in the right eye (Figure 1A and B). Amsler grid testing detected bilateral large, dense central scotomas.
Eyes of patient 1 at initial examination. A and B, Bilateral swollen and hyperemic optic discs, venular and arteriolar sheathing in the posterior pole, and a small vitreous hemorrhage in the right eye. C and D, Fundus fluorescein angiography shows severe venular and arteriolar leakage over the entire posterior pole. E and F, Optical coherence tomography shows severe neurosensory retinal detachments in both eyes.
We found early segmental leakage of the retinal arterioles and venules over the entire macula, which later leaked floridly on FFA (Figure 1C and D). The ICG findings were unremarkable. The OCT showed exudative detachment over the fovea in both eyes (Figure 1E and F). She had a low platelet count (26 × 103/μL) and C4 level (18 mg/mL) and an elevated urine microalbumin level (34 mg/L) at the time of examination.
The patient was treated with intravenous methylprednisolone, 1 g/d for 3 days. On day 2 of treatment, vision decreased to counting fingers OD despite a marked decrease in subretinal fluid on OCT. Visual acuity in the left eye was unchanged. In view of progressive eye disease, immunoglobulin therapy was given for 5 days. The BCVA was 20/200 on day 7. Intravenous steroids were shifted to 1 mg/kg daily of oral prednisolone and tapered. On day 9, visual acuity improved to 20/100 OD and 20/200 OS. Vision improved to 20/25 OD and 20/30 OS at 10 weeks and remained stable throughout follow-up. By week 2, vascular leakage and neurosensory retinal detachment had resolved (Figure 2A-D). The HVF had a dot score of 8 OD and 13 OS. At month 12, the visual acuity was unchanged, and a small central scotoma remained in the right eye despite complete resolution in the left eye.
Eyes of patient 1 after 2 weeks of treatment. A and B, Fundus fluorescein angiography confirms resolution of vascular leakage. C and D, Optical coherence tomography shows flattening of the retina with resolution of subretinal fluid.
A 16-year-old girl was referred for ophthalmic evaluation when she complained of blurring of vision (7 days after fever onset) during her hospitalization for dengue fever. On examination, BCVA was 20/40 OD and 20/50 OS. Mild anterior segment inflammation was found in both eyes, with vitritis of 3+ OD and 1+ OS. She had swollen, hyperemic optic discs. Several small, yellowish subretinal macular dots were found in both eyes (Figure 3A and B), which corresponded to areas of early hyperfluorescence that did not fade in later images on FFA (Figure 3C and D), and hypofluorescent spots in the middle to late phases of the ICG (Figure 3D and E). Additional areas of choroidal involvement that were not clinically evident were detected on ICG. The HVF test findings were unremarkable.
Eyes of patient 2 at initial examination. A and B, Multiple small, yellowish subretinal macular dots in both eyes. C and D, Fundus fluorescein angiography shows areas of early hyperfluorescence that do not fade in later images. E and F, Hypofluorescent spots in the middle to late phases are seen on indocyanine green angiography.
The patient received a course of oral prednisolone for 2 months. Visual acuity had improved to 20/30 OD and 20/25 OS by week 2. At month 2, BCVA was 20/20 OD and 20/25 OS.
Forty-one patients who satisfied the inclusion criteria were identified from January 1, 2002, to December 31, 2005. Of these, 24 (58%) were female and 35 (85%) were Chinese. The mean age was 28.8 ± 11.4 years (range, 11-61 years).
Seventy-one eyes had dengue maculopathy. Ocular involvement was bilateral but asymmetric in 30 patients (73%); the rest were unilateral. Mean follow-up was 5.4 ± 4.8 months (range, 1-21 months).
Symptoms included generalized blurring of vision in 62 eyes (87%), scotoma in 45 eyes (63%), and floaters in 1 eye (1%). Symptoms were noted a mean ± SD of 6.9 ± 4.0 days (range, 0-30 days) after the onset of fever. There was a mean ± SD delay in consultation of 8.6 ± 15.6 days (range, 0-90 days) from the onset of symptoms.
The summary of fundus and angiographic findings is provided in Table 1 and Table 2. Intraretinal hemorrhages were seen in 32 eyes (45%), usually in association with venous sheathing. Fundus fluorescein angiography demonstrated nonischemic venular occlusion in 16 eyes (25%) or arteriolar and/or venular leakage in 2 eyes (3%) and 18 eyes (13%), respectively. Yellow subretinal dots were an unusual finding in 20 eyes (28%), usually without corresponding angiographic findings. In 10 eyes (50%) with yellow dots, the lesions corresponded to hypofluorescent spots in the middle to late phases of ICG and RPE hyperfluorescence on FFA in 3 eyes (15%). The rest had no corresponding abnormality on angiography. Anterior chamber cells were observed in 12 eyes (17%), vitreous cells in 22 eyes (31%), and anterior chamber and vitreous cells in 8 eyes (11%). Among the 42 eyes tested, OCT showed exudative retinal detachment in 4 eyes (10%), outer neurosensory retina–RPE thickening at the fovea (Figure 4A) in 11 eyes (26%), and neurosensory retina–RPE atrophy in 1 eye (2%) (Figure 4B). A combination of lesions frequently occurred in the same eye. Conversely, different lesions were identified in either eye of the same patient in some cases. The details of ocular characteristics according to predominant involvement are presented in Tables 3, 4, 5, 6, 7, and 8.
Optical coherence tomography of the tested eyes. A, Outer neurosensory retina–retinal pigment epithelial thickening at the fovea. B, Neurosensory retina–retinal pigment epithelial atrophy in dengue maculopathy.
Among eyes with maculopathy, the initial BCVA was 20/45 (mean ± SD, 0.47 ± 0.42 logMAR equivalent); 37 eyes (52%) had a BCVA of 20/40 (0.30 logMAR equivalent) or better. Those without maculopathy had a BCVA of 20/25 or better (mean ± SD, 0.05 ± 0.06 logMAR equivalent), which was significantly different from maculopathy eyes (analysis of variance; P = .001). Among the 64 eyes with visual field test results, the mean ± SD HVF parameters were as follows: mean deviation of 5.62 ± 5.43, pattern standard deviation of 4.38 ± 2.94, and dot score of 5.25 ± 5.31.
Improvement in mean BCVA was noted in all eyes (Figure 5). A significant improvement in mean BCVA from baseline was noted between weeks 2 and 4 (Tukey multiple comparisons testing; P = .05). Although visual acuity continued to improve thereafter, the difference between subsequent follow-up intervals was not significant. Table 9 further illustrates that the proportion of eyes with BCVA of 20/40 or better increased at every follow-up, from 37 eyes (52%) at initial examination to 61 eyes (86%) at final follow-up. In 3 eyes (4%) that had only 1 month of follow-up, BCVA was worse than 20/200 although better than at initial examination. The lack of adequate serial HVF observations precluded the same analysis for visual fields.
Improvement in mean best-corrected visual acuity (VA) was noted in all eyes.
Among 19 patients tested, the mean ± SD C4 level was low at 17 ± 6 mg/mL (range, 8-38 mg/mL), whereas in 14 patients a high mean ± SD urine microalbumin level was reported at 68.06 ± 144.56 mg/L (range, <11-565 mg/L). The mean ± SD C3 level of 86 ± 24 mg/mL (range, 47-121 mg/mL) was within the reference range. In 37 patients, the mean ± SD platelet count at the time of ocular examination was 133.47 ± 117.82 × 103/μL (range, 11-670 × 103/μL).
Twenty-eight (68%) of the 41 patients received treatment in the following manner: intravenous methylprednisolone followed by oral prednisolone in 12, oral prednisolone alone in 11, and intravenous immunoglobulins with intravenous hydrocortisone followed by oral steroids in 3 patients with poor visual acuity who did not respond to initial intravenous steroids. Other patients received periocular methylprednisolone injections in 10 eyes, intravitreal triamcinolone acetonide in 2 eyes, and topical prednisolone acetate in 1 eye.
Dengue maculopathy has myriad manifestations that may be inflammatory or occlusive in nature or both. Lesions in one eye of the same patient may not necessarily be the same in the other eye. Conversely, a combination of lesions may be observed in the same eye. Fundus fluorescein angiography, ICG, and OCT are crucial in the assessment of the type and severity of ocular involvement, since identification may not be evident clinically and may therefore be missed.
Retinal vasculopathy, as evidenced by intraretinal hemorrhages and vascular sheathing, was the most common ocular manifestation of dengue. Hemorrhage, a consistent finding in patients with diseases that result in thrombocytopenia,16 was found in 32 eyes (45%). In a third of these cases, venular occlusion was an associated finding. Focal to extensive vascular sheathing involved the venules in 32 eyes (45%) and arterioles in 3 eyes (4%). Fundus fluorescein angiography showed nonischemic venular occlusion in most eyes. Vasculitis (phlebitis more than arteriolitis) and knobby capillary hyperfluorescence were less commonly observed. In rare cases (3% of eyes), FFA was unremarkable despite clinically observed sheathing on fundus examination. Both occlusive and inflammatory lesions may be found in the same eye. In all eyes with retinal vascular sheathing, ICG was usually unremarkable. Regardless of whether an inflammatory or occlusive lesion was present, more severe central field and visual loss were associated with maculopathy involving larger retinal vessels, covering more extensive areas or areas adjacent to the foveal center. Furthermore, eyes with occlusive disease tended to have residual scotomas despite apparent recovery of vision.
The presence of single or multiple yellow subretinal dots is an unusual clinical feature best studied with ICG. Although early RPE hyperfluorescence was seen in 3 eyes with yellow dots (15%), FFA was frequently unremarkable. In half of cases, ICG showed hypofluorescent spots corresponding to these yellow dots and additional spots in areas without clinically evident dots. These findings are in keeping with choriocapillary-RPE involvement. Whether these lesions are inflammatory or occlusive in nature is unclear. Nevertheless, they resolved with time in all eyes. In eyes with yellow dots, central scotomas were not observed on HVF test results and visual loss was less pronounced at initial examination compared with those with predominant retinal vascular involvement. Eyes with isolated choroidal inflammation, typically in the form of large-vessel hyperfluorescence on ICG, also developed less severe visual disability.
Optical coherence tomography is an indispensable diagnostic tool in eyes in which no apparent lesion is observed clinically or if clinical findings cannot account for poor visual acuity. In 11 eyes in which a well-circumscribed, pale yellow-orange lesion at the foveal center was seen, a corresponding area of focal outer neurosensory retina–RPE thickening at the foveal center was identified on OCT. We have coined the term foveolitis to describe this lesion. Foveolitis was noted either as a solitary finding or in the presence of other fundus lesions. Except for 1 eye in which FFA showed early hyperfluorescence that persisted in the late phase, the foveal lesion had no corresponding FFA finding. However, foveolitis occurred in the presence of other lesions that were extrafoveal in location: vascular occlusion in 3 eyes and venular leakage with occlusion in 1 eye. A dense central scotoma was identified in all eyes. The severity of visual acuity and field loss corresponded with the size of the lesion and degree of thickening.
Distinct from foveolitis, OCT was also useful in documenting and assessing the severity of macular swelling, which was seen in 11 eyes (less than 15%). This ranged from mild to severe, the latter being associated with neurosensory retinal detachment identified in eyes with severe and widespread retinal vascular leakage. The presence of cystoid spaces was starkly absent in these eyes. Foveolitis and macular swelling were found to resolve over time on serial OCT. Hence, we suggest that serial OCT and HVF testing be performed to monitor maculopathy.
The pathophysiologic mechanism of dengue fever is still not completely understood. Several theories have been suggested.17- 21 The 5- to 7-day delay in onset of ocular symptoms supports an immune-mediated mechanism rather than direct viral infection as previously reported.13 The transient decrease in levels of C4 in several patients in this series also suggests an immune complex–mediated mechanism, which explains the occlusive vasculopathy seen in a third of eyes in this series.
In view of this immune-mediated hypothesis, we opted to treat patients primarily with corticosteroids. In our opinion, the prevention of structural damage and permanent visual loss due to ocular inflammation justifies treatment unless contraindicated. In 3 cases with severe visual loss that persisted despite initial anatomical improvement with intravenous steroids, immunoglobulin therapy, currently used to treat conditions such as idiopathic thrombocytopenic purpura and systemic vasculitides, was given. Immunoglobulins modulate complement activation, suppress idiotypic antibodies, and inhibit complement-mediated immune damage.22 Common to these 3 cases is a persistently low level of C4. Whether such patients are more susceptible to developing more severe disease is unclear.
Analysis regarding therapy and its effect on visual function is difficult because of the retrospective nature of this study. Treatment, field testing, and follow-up schedule were not standardized. Nevertheless, it is our clinical impression that cases with intraretinal vascular or choroidal leakage, evidence of active ocular inflammation, and foveal swelling are more likely to benefit from steroid treatment compared with those with vaso-occlusive disease, which were associated with residual dense scotomas.
The question of whether the disease is self-limiting is also a valid concern. Those who had a better initial BCVA were not treated, whereas those who had worse involvement, and therefore had poor BCVA at initial examination, were treated. Hence, it is unclear whether visual recovery was the result of treatment or part of the natural course of the disease. It is imperative that the effect of treatment on rate and amount of recovery be addressed in future studies, since dengue maculopathy is a potentially disabling disease, particularly in patients with severe involvement. Clearly, a prospective randomized controlled study on therapy for dengue maculopathy may be necessary to devise a protocol for the early diagnosis and treatment of maculopathy for the prevention of permanent visual impairment.
In conclusion, dengue fever may cause maculopathy with a wide spectrum of manifestations. Fundus fluorescein angiography, ICG, OCT, and visual field testing are useful tools in the diagnosis and assessment of the extent and severity of this ocular disease.
Correspondence: Soon-Phaik Chee, FRCOphth, FRCS(G), Singapore National Eye Centre, 11 Third Hospital Ave, Singapore 168751 (firstname.lastname@example.org).
Submitted for Publication: February 22, 2006; final revision received July 18, 2006; accepted July 24, 2006.
Author Contributions: Kristine Enrile Bacsal, MD, had full access to all the data in the study and takes responsibility for the integrity of the data and accuracy of the data analysis.
Financial Disclosure: None reported.
Previous Presentation: Presented in part at the Asia-Pacific Academy of Ophthalmology Meeting; March 23, 2005; Kuala Lumpur, Malaysia.
Thank you for submitting a comment on this article. It will be reviewed by JAMA Ophthalmology editors. You will be notified when your comment has been published. Comments should not exceed 500 words of text and 10 references.
Do not submit personal medical questions or information that could identify a specific patient, questions about a particular case, or general inquiries to an author. Only content that has not been published, posted, or submitted elsewhere should be submitted. By submitting this Comment, you and any coauthors transfer copyright to the journal if your Comment is posted.
* = Required Field
Disclosure of Any Conflicts of Interest*
Indicate all relevant conflicts of interest of each author below, including all relevant financial interests, activities, and relationships within the past 3 years including, but not limited to, employment, affiliation, grants or funding, consultancies, honoraria or payment, speakers’ bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued. If all authors have none, check "No potential conflicts or relevant financial interests" in the box below. Please also indicate any funding received in support of this work. The information will be posted with your response.
Register and get free email Table of Contents alerts, saved searches, PowerPoint downloads, CME quizzes, and more
Subscribe for full-text access to content from 1998 forward and a host of useful features
Activate your current subscription (AMA members and current subscribers)
Purchase Online Access to this article for 24 hours
Some tools below are only available to our subscribers or users with an online account.
Download citation file:
Web of Science® Times Cited: 22
Customize your page view by dragging & repositioning the boxes below.
and access these and other features:
Enter your username and email address. We'll send you a link to reset your password.
Enter your username and email address. We'll send instructions on how to reset your password to the email address we have on record.
Athens and Shibboleth are access management services that provide single sign-on to protected resources. They replace the multiple user names and passwords necessary to access subscription-based content with a single user name and password that can be entered once per session. It operates independently of a user's location or IP address. If your institution uses Athens or Shibboleth authentication, please contact your site administrator to receive your user name and password.