Background
Immune reconstitution inflammatory syndrome (IRIS) is a group of disorders that has been described in HIV patients who are on highly active antiretroviral therapy (HAART). This phenomenon is associated with a paradoxical or exaggerated reactivation of the immune system in the background of a wide range of pathologies such as Kaposi’s sarcoma, tuberculosis, cryptococcal infection and cytomegalovirus [
1].
Immune recovery uveitis (IRU) is one manifestation where immune-deficient patients with prior cytomegalovirus retinitis (CMVR) experience an increase in intraocular inflammation following HAART. It is presumed to be mediated by the recovery of immune responses specific to residual cytomegalovirus (CMV) antigens in the eye [
2]. IRU is an important cause of visual morbidity in HIV-infected patients with CMVR in the era of HAART [
3]. Although immune recovery associated with HAART has allowed some patients to discontinue specific anti-CMV therapy, the rejuvenated immune response can be associated with chronic sight-threatening inflammation. Visual complications of IRU include cataract, cystoid macular edema (CME), glaucoma and epiretinal membrane formation [
4‐
10]. Reported risk factors include a low CD4 count at the time of initiation of HAART and larger areas of CMVR [
3]. With the rising prevalence of HIV patients and greater accessibility and availability of HAART worldwide, IRU is an increasingly important disease entity. We aim to analyse the characteristics of IRU in HIV patients on HAART in Singapore, to identify risk factors for the development of IRU and to determine their visual outcomes.
Methods
A retrospective analysis of clinical records of 30 consecutive IRU patients presenting to the Eye Clinic at the Communicable Disease Centre (CDC), Singapore, between March 2004 and September 2009 was performed. Ethics approval was obtained from the local ethics review board in accordance with the Declaration of Helsinki.
The CDC is the main site of referral and management for HIV in Singapore. IRU was diagnosed when there is an increase in any form of intraocular inflammation from an inactive state of CMV retinitis secondary to immune recovery after initiation of HAART, as evidenced by an increase in CD4 count by 50 or more cells/μL from the nadir CD4 count to a level of more than 100 cells/μL [
3]. Despite a CD4 count of less than 100 cells/μL at diagnosis, a small subset of patients are also deemed to have IRU. These patients would have an increase in intraocular inflammation, which is correlated with a rise in CD4 count that could not be clinically attributed to a reactivation of CMVR.
All of the patients had prior CMV retinitis treated with an intravitreal ganciclovir regimen, as previously described [
11]. Two patients were also treated with intravenous ganciclovir, and one received oral valganciclovir. Patients with concomitant infective or other autoimmune causes of inflammation were excluded from this study. All patients underwent Snellen best-corrected visual acuity (BCVA) assessment, tonometry and dilated slit-lamp biomicroscopy and indirect ophthalmoscopy examinations. Clinical data obtained included patient demographics (age, gender and ethnicity); duration between HIV, CMVR and IRU diagnoses; BCVA at presentation and at 6 months; corresponding CD4 counts and CD4/CD8 ratios; duration from initiation of HAART to IRU development; presenting signs and symptoms; CMVR zone and area of involvement; IRU severity and location; and treatment regimes. The location of CMVR lesion in each eye was categorised into three zones. Zone 1 was defined as the area within 1500 μm of the optic nerve or within 3000 μm of the fovea. Zone 2 extended from the borders of zone 1 to the vortex veins, and zone 3 involves the retina peripheral to zone 2 [
12]. For patients with CMVR involving multiple zones, the zone nearest to the macula was reported.
Descriptive analysis was performed for demographics, visual acuity, complications and IRU and CMVR characteristics, with categorical variables summarised with frequencies and percentages and continuous variables summarised as either means with standard deviations or medians with ranges. Spearman’s rho was calculated to check for association between CD4 counts and durations. Independent sample t test and Mann-Whitney U test were used for comparison of age, durations and CD4 counts between patient groups with and without complications, respectively. Logistic regression was used to understand the relationship between variables. A p value less than 0.05 was considered to indicate statistical significance. Data analysis was carried out using IBM SPSS Statistics (version 19, IBM Corp., NY, USA).
Discussion
HIV-related mortality has decreased dramatically in the era of HAART. However, it is now evident that the commencement of HAART can also potentially lead to considerable morbidity and mortality in the form of IRIS, especially in the first 6 months [
14]. IRU is an important presentation in this broad disease group which also includes tuberculosis-associated IRIS and Kaposi’s sarcoma IRIS [
1]. This is the first paper discussing the characteristics of IRU and its complications in Singapore and Southeast Asia. As the Communicable Disease Centre in Singapore manages and treats the majority of local HIV patients [
15], our patient cohort is highly reflective of IRU diagnosed locally. The number of HIV-positive patients in Singapore has been gradually increasing, with 454 newly diagnosed cases in 2013 and a total of 6229 patients diagnosed since 1985 [
16]. The number of patients with CMVR has increased correspondingly, with 224 patients diagnosed at the CDC from 2005 to 2010 [
17]. IRU is most commonly associated with previous CMVR in patients who have experienced immune recovery.
Similar to previous studies, vitritis is the most common presentation in our cohort [
5,
10,
18,
19]. There are also a significant proportion of patients who had anterior uveitis. Although anterior uveitis is commonly reported [
5,
9,
10,
20], most studies have focused their diagnosis of IRU on the presence of vitritis and associated CME [
4,
6,
9,
18,
19].
Patients with larger areas of CMVR involvement were reported to have higher risk of IRU development [
9]. In contrast, the majority of eyes in our cohort had <25 % area of previous CMVR involvement. Arevalo et al. did not find an association between CMVR area and risk of IRU development [
21]. This suggests that small CMVR lesions are common and eyes without large areas of CMVR invovement are similarly at risk of developing IRU.
Within our cohort, we did not find any association between CD4 counts at HIV diagnosis, at initiation of HAART or at IRU diagnosis, with the development of IRU. However, the duration from the initiation of HAART to IRU was found to be longer in younger patients (<50 years old). Hartigan-O’Connor et al. previously reported that IRU is most likely to develop in patients with the greatest degree of immune dysfunction prior to HAART [
22]. Additionally, older age groups have been correlated with quicker seroconversion to AIDS in the pre-HAART era, as well as higher rates of immunologic failure and persistent HIV viraemia in the HAART era [
23]. Thus, it is plausible that younger HIV patients have a more intact immune system that may not have such heightened sensitivity to CD4 T cell responses, consequently delaying their development of IRU until the occurrence of further immune dysfunction.
The most common complications of IRU in our cohort were cataract formation, glaucoma and ocular hypertension. Our findings differ from other reports where CME and epiretinal membrane were the most common complications. For instance, Robinson et al. reported that of the IRU eyes in their cohort, 91 % developed CME, 52 % developed cataract and 30 % developed epiretinal membrane [
5]. Cataract, CME and epiretinal membrane typically caused moderate visual loss [
3,
5,
7].
No cases of CME were reported during the course of this study. Although CME is often described and is a serious sight-threatening complication of IRU, we believe that it is not a common feature, especially if the inflammation is treated earlier. This is particularly pertinent as there is currently no consensus on the ideal time to start treatment of IRU. In fact, the majority of studies have suggested that mild and asymptomatic ocular inflammation usually resolves without treatment; treatment is instituted in more severe inflammation and in patients who become symptomatic [
6,
8].
IRU has been described as a transient form of inflammation [
19]. Still, others report that the persistence of inflammation in IRU is the cause of complications like CME and epiretinal membrane [
6,
18]. Based on our findings, we propose the early treatment of ocular inflammation in IRU patients to prevent sight-threatening complications of CME and epiretinal membrane. Nonetheless, the judicious use of topical steroids in treating asymptomatic patients must be balanced against the possible complications of steroid use.
The risk of IRU complications was found to be higher with greater CD4 count increase from HIV to IRU diagnosis. The greater inflammatory response mounted in these immune reconstituted patients, together with a robust increase in the number of viable T lymphocytes, results in severe and chronic inflammation [
24]. The increased inflammatory response can lead to cataract formation, raised intraocular pressure and glaucoma. Conversely, the development of cataract and raised intraocular pressure may also be the result of aggressive steroid therapy, which is often instituted in these patients. However, the mode of steroid therapy did not appear to be associated with the risk of development of complications in this study.
The presence of IRU at baseline was reported to be associated with an increased risk of moderate vision loss to 6/15 [
3]. The visual outcome of our IRU patient cohort was, however, fair with only eight eyes losing two or more lines of Snellen VA. 73.6 % of eyes (
n = 39) had either maintained or improved in vision. The importance of early diagnosis with aggressive and appropriate treatment can improve and preserve vision. Moreover, a number of patients were asymptomatic and were found to have IRU only on routine follow-up examinations. All HIV patients with treated CMVR should continue to have regular reviews at the eye clinic at regular intervals, even if they have high CD4 counts. Our study showed the median time from commencement of HAART to IRU development to be about 7 months. The longest time taken was 8.8 years in a 25-year-old female. Without regular follow-up, early recognition and treatment, the visual outcomes of these patients may be less favourable.
The main limitations of this study were its retrospective design and relatively small sample size. There was also no control group. Unlike other studies, we did not routinely perform angiogram on our patients. Hence, it is possible that early CME not apparent on physical examination was missed. Nonetheless, our cohort of patients was diagnosed consecutively from the CDC, which is reflective of the HIV patient profile in Singapore, as the CDC is the main centre for HIV treatment.