One thousand HIV-positive patients were included in the study. Their age ranged between 18 months to 75 years (median age, 35 years). Male-to-female ratio was 7:3. The most common presenting complaints were decreased vision and floaters. Overall, ocular lesions attributable to HIV/AIDS were found in 68.5% of the patients (Table ). Interval between diagnosis of HIV and onset of ocular lesions ranged from 10 days to 10.24 years (mean, 2.43 years). CD4 counts range from 2 to 1,110 cells/mm3
213.36, median 224.50) at the time of inclusion into the study. Mean follow-up was 67 months (range between 1 month and 11 years), and median duration of HIV disease was 360 days. More than 65% of the patients had at least 1 year follow-up. About 51.3% of the patients had an underlying systemic disease. Systemic tuberculosis, the most common opportunistic systemic infection in our study, was seen in 43%, of which pulmonary tuberculosis was seen in 38.01% of patients (Table ). Other systemic diseases included Cryptococcus
infection, pneumonia, central nervous system (CNS) toxoplasmosis, and others. Out of 48.70% without any underlying systemic disease, 74.74% were without ocular lesions, while 25.26% had ocular lesions attributable to HIV (Table ).
Ocular lesions due to HIV in the pre- and post-HAART era
Underlying systemic disease associations in patients with HIV
Ocular lesions in correlation with systemic disease
Among patients with ocular lesions, CMV retinitis (Figure ) was the commonest and was seen in 248 patients (36.2%). All patients with active CMV retinitis were treated with induction therapy with intravenous ganciclovir (5 mg/kg of body weight every 12 h) along with an intravitreal injection of ganciclovir, in most cases followed by maintenance therapy. In our study, two patients with CMV retinitis on HAART presented with neovascularization and vitreous and subretinal hemorrhage and were managed accordingly.
Composite color fundus photographs showing a case of active and healed CMV retinitis. Pre- (left) and post-treatment (right) with ganciclovir.
Ocular toxoplasmosis (Figure ) was the next most common ocular opportunistic infection and was seen in 28 patients (4.09%). Most of them were diagnosed based on clinical features and supportive anterior chamber tap findings, positive polymerase chain reaction (PCR) or antibodies for toxoplasmosis. All patients were treated with combined anti-toxoplasma therapy based on clinical and serological evidence.
Composite color fundus photographs showing a case of active (left) and healed (right) toxoplasmic retinochoroiditis.
Ocular tuberculosis (Figure ) was seen in 26 patients (3.8% patients), presenting as choroidal tubercles, subretinal abscess, conjunctival tuberculosis, and panophthalmitis. All cases had evidence of pulmonary tuberculosis. Coexistent central nervous system and abdominal tuberculosis were the associations. CD4+ cell counts in patients with ocular tuberculosis were between 14 to 560 cells/μl (mean, 160.85 cells/μl). HIV retinopathy was seen in 6.28% of the patients. Other lesions noted significantly were acute retinal necrosis in 1.6%, progressive outer retinal necrosis in (PORN) in 1.17%, syphilis (1.2%), endophthalmitis (1.02%), fungal corneal ulcer (0.3%), retinal vessel occlusions (0.44%), and anemic retinopathy (0.73%).
Color fundus photograph showing a case of active choroidal granuloma due to tuberculosis.
One hundred three patients (15.04%) had anterior segment, adnexal, and/or ocular surface lesions. CD4 count range in this subgroup of patients was 19 to 964 cells/mm3
. HZO was the commonest, seen in 21.4% of patients with anterior segment manifestations (Figure ). Others included anterior uveitis in 16 patients (15.57%), ulcerative blepharitis, keratitis, conjunctival microvasculopathy, keratoconjunctivitis, conjunctival squamous cell carcinoma, molluscum contagiosum, episcleritis, Stevens-Johnson syndrome, lid abscess, and herpetic sclerouveitis. Most viral anterior segment lesions had higher CD4 counts of >
. Significantly, 12 patients (13.3%) were not known to be HIV-positive and were diagnosed based on clinical suspicion.
External photograph showing herpes zoster ophthalmicus.
Neuro-ophthalmic lesions were seen in 61 patients. Optic atrophy, seen in 49.45%, and disk edema, seen in 21.97%, were the most prominent manifestations. Others included optic neuritis in 14.28%, cranial nerve palsies in 9.89%, retrobulbar neuritis in 2.19%, and optic nerve head infiltration and cortical blindness in one patient each. Cryptococcus infection was the commonest underlying infection in 66% patients.
Multiple ocular opportunistic infections in the same eye/patient were seen in ten patients. The commonest combination was presence of toxoplasmic and cytomegalovirus retinitis, seen in four patients. Others included ocular TB and CMVR, and herpetic retinitis and CMVR in two patients each, and ocular toxoplasmosis with ocular syphilis and ocular TB in one patient each.
Confirmation of the etiological agent was done in select cases by intraocular fluid evaluation in 8.6% patients with a 72% yield. Specific ocular medical and surgical therapy was administered based on the diagnosis.
The commonest combination regimen of HAART was lamivudine, nevirapine, and stavudine in 84% of patients. Drug combinations were individualized based on response to treatment or development of resistance under care of an AIDS care physician.
When comparing patients on antiretroviral therapy (ART) and those not on ART with respect to the diagnosis of ocular lesions, it was found that 52.2% of patients were on ART at the time of ocular lesion diagnosis, 62% of them had CD4
200 while the rest had CD4 counts more than 200. While among patients on ART and without ocular lesions, 31% had CD4 counts less than 200, and significantly, in spite of this, no ocular opportunistic infection was noted even during the period of follow-up.
There were 47.7% of patients who were not on HAART at the time of ocular lesion diagnosis; 69% of them had CD4 counts less than 200. While among patients not on ART and without ocular lesions, 73% of patients understandably had CD4 counts more than 200, while significantly 27% of them had CD4 counts less than 200 (Table ).
Comparison of presence of ocular lesions in patients with/without HAART in relation to CD4 counts
Immune recovery uveitis (IRU) was seen in 17.4% of the patients on HAART. Interval between the start of HAART and onset of IRU was 4 months to 2.5 years, and 52% of the patients developed IRU when the increase in CD4 counts was between 100 and 150 cells. Seventeen patients were treated with topical steroids and cycloplegics, one patient was treated with topical and periocular steroid injection, and five patients were treated with an additional short course of oral steroids along with topical steroids.
In our series of patients, immune recovery uveitis was seen predominantly post-CMVR in 88% of patients, while IRU post-tuberculous and post-cryptococcal infection was noted in 6% and 2% of patients with IRU, respectively. Whereas in 4% of patients, there was inflammation post-immune recovery; the cause remained undetermined. Recurrence of ocular infection was seen in 2.53% (post-HAART) and >
20% (pre-HAART) of patients.
Based on the severity of visual impairment, the patients were divided into groups with visual acuity better than 6/12, between 6/12 and 6/60, and worse than 6/60. Pre- and post-treatment (i.e., at presentation and at final follow-up) were analyzed (Table ). At presentation, 57% had visual acuity better than 6/12, while at final follow-up 58.3% fell in this group; 13.9% had moderate visual impairment, and at the end of treatment, 13% were in this group. At presentation, 29.1% had vision poorer than 6/60, and 28.8% belonged to this group at the final analysis. The visual outcome among various groups, pre- and post-treatment, is given in Table .
Pre- and post-treatment visual acuity: comparison between various groups based on severity of visual impairment
Visual outcome pre- and post-treatment in the various groups
Considering a two-line improvement in Snellen’s visual acuity, overall visual outcome showed improvement in about 14.3% and was maintained in 71.6% of patients. While in 14.1% of patients, vision deteriorated due to complications such as retinal detachment, optic atrophy, macular scarring, and others.