Corresponding Author: Jong Soo Lee, MD, PhD. Department of Ophthalmology, Pusan National University Hospital, 179 Gudeok-ro, Seogu, Busan 49241, Korea. Immune reconstitution inflammatory syndrome (IRIS) is a paradoxical clinical worsening in HIV-infected patients after initiating highly active antiretroviral therapy (HAART). This deterioration is known to arise from the restoration of immunity to specific agents such as cytomegalovirus (CMV), mycobacterium tuberculosis, cryptococcal infection, and various other pathogens [1-4].
Ocular IRIS is mainly composed of immune recovery uveitis (IRU) associated with CMV. IRU involves the activation of ocular inflammation in patients with inactive CMV retinitis after the initiation of HAART [5]. Some studies have also broadened the definition of ocular IRIS to include the development or worsening of active CMV retinitis after the initiation of HAART [6-8]. Ocular IRIS also extends to encompass other opportunistic infections such as mycobacterium related choroiditis [9]. Ocular IRIS can result in visual impairment, constituting a significant cause of visual morbidity in HIV-infected patients [10].
Regarding T cell immunity in ocular IRIS, low CD4+ T cell count prior to HAART is known to be one of the risk factors for IRU in HIV patients [10,11]. Additionally, CD8+ T cells have been implicated in the pathogenesis of CMV retinitis in HIV patients [12,13]. The histological analyses of vitreous humor and central nervous system tissue of IRIS patients demonstrated CD8+ T cell predominance [12,14].
However, only a limited number of studies have investigated the clinical significance of CD8+ T cell in ocular IRIS patients after the initiation of HAART. Thus, we planned to investigate ocular manifestation of IRIS in HIV patients after starting HAART and the clinical relationship between the ocular manifestation and T cell immunity.
Materials and Methods
Ethics statement
The study was approved by the Institutional Review Board of Pusan National University Hospital (No. 2304-004-125). The requirement for informed consent was waived due to the retrospective nature of the study.
Patient recruitment
The study included HIV-infected patients who initiated HAART and developed ocular IRIS between 2004 and 2023 at Pusan National University Hospital (Busan, Korea). Medical records of consecutive patients meeting the inclusion criteria were retrospectively reviewed. The primary objective of our study was to analyze ocular manifestation of IRIS in HIV patients after starting HAART and the clinical relationship between the ocular manifestation and T cell immunity.
The diagnosis of IRIS in HIV-infected patients who had started HAART was based on the consensus criteria of the International Network for the Study of HIV-Associated IRIS, which requires virologic response to HAART (decrease of HIV RNA with >1 log10 copies/mL), clinical deterioration of infectious or inflammatory condition temporally related to HAART initiation, and symptoms that cannot be explained by expected clinical course of a previously recognized and treated infection, medication side effect, toxicity, treatment failure, and complete nonadherence [2,3,6,15,16].
Ocular IRIS was diagnosed if any ocular inflammation was examined in previously inactive opportunistic infection; if it was newly developed; or if it was paradoxically exacerbated in previously improving active opportunistic infection, after the initiation of HAART [6,17]. The ocular inflammation included anterior uveitis, vitritis, retinitis, vasculitis, macular edema, and epiretinal membrane. Patients without evidence of previous opportunistic infection were excluded.
Ophthalmic assessment
All patients underwent a complete ophthalmic examination. Anterior segment and fundus examination was performed by slit-lamp biomicroscopy and indirect ophthalmoscopy. Best-corrected visual acuity (BCVA) at ocular IRIS diagnosis and at 6-month follow-up were examined. BCVA was measured by Snellen chart which was then converted to the logarithm of the minimum angle of resolution (logMAR). For each ocular IRIS developed, previous opportunistic infection, duration from initiation of HAART to diagnosis of ocular IRIS, and ocular complication after ocular IRIS were reviewed. Blood CD4+, CD8+ T cell count, and HIV RNA copies before HAART and at diagnosis of ocular IRIS were collected.
Statistical analysis
Descriptive analysis was performed to show clinical manifestation of ocular IRIS. Correlation between T cell count and other clinical parameters was investigated by Spearman correlation analysis. To determine the cutoff value of CD8+ T cell count before HAART for anticipating the onset of ocular IRIS, receiver operating characteristic (ROC) curve was derived. To compare clinical manifestation between two groups with CD8+ T cell count above and below the cutoff value, Mann-Whitney test and Fischer exact test were performed. Statistical analysis was performed using IBM SPSS ver. 19.0 (IBM Corp), and the statistical significance was set at p < 0.05.
Results
A total of 19 patients (15 men, 4 women; 27 eyes) were included in the study. The median age was 53 years (interquartile range [IQR], 47-60 years). The most common previous opportunistic infection was CMV retinitis (17 patients, 89.5%), followed by tuberculosis choroiditis (2 patients, 10.5%). The median duration from initiation of HAART to ocular IRIS was 70 days (IQR, 46-160 days). Ocular IRIS was unilateral in 11 patients and bilateral in 8 patients. Twenty eyes (74.0%) presented with vitritis; 14 eyes presented with retinitis (51.9%); and 5 eyes presented with anterior uveitis. The median visual acuity at the diagnosis of ocular IRIS and at 6-month follow-up did not differ statistically. At 6 months after ocular IRIS diagnosis, 6 eyes (22.2%) showed improvement in visual acuity by more than 1 line in Snellen chart, while 10 eyes (37.0%) showed deterioration by more than 1 line. Epiretinal membrane was the most common complication after ocular IRIS in six eyes (22.2%), followed by four cataracts, three retinal detachments, two posterior synechiaes, and one macular hole cases (Tables 1, 2).
The median CD4+ T cell count increased from 35/mm3 before HAART to 142/mm3 at the diagnosis of ocular IRIS (p < 0.001). The median CD8+ T cell count increased from 263/mm3 before HAART to 630/mm3 at the diagnosis of ocular IRIS (p = 0.007). The median HIV RNA copies decreased from 517,800/mL before HAART to 49/mL at the diagnosis of ocular IRIS (p < 0.001) (Table 1).
The CD8+ T cell count before HAART was negatively correlated with duration from HAART to ocular IRIS (Spearman coefficient, −0.847; p < 0.001) (Table 3, Fig. 1). The CD4+ T cell count and HIV RNA copies did not show any correlation with duration from HAART to IRIS. There was no other variable reviewed that had multicollinearity with CD8+ T cell count before HAART. The distribution of duration from HAART to ocular IRIS showed two peaks, which could be divided into two groups by the median value of 70 days (Fig. 2). To analyze the predictive performance of CD8+ T cell count before HAART for anticipating either the early or late onset of ocular IRIS, cutoff value of CD8+ T cell count before HAART was estimated by the ROC curve. The area under the curve was 0.944 (p = 0.001), and its cutoff value of CD8+ T cell count was 258/mm3 with sensitivity of 90% and specificity of 89%. Fig. 3 shows the boxplot of duration from HAART to ocular IRIS according to CD8+ T cell count of 258/mm3 before HAART. When divided into two groups by the CD8+ T cell count of 258/mm3, 90% patients with CD8+ T cell count higher than 258/mm3 before HAART developed ocular IRIS within 70 days. Characteristics of patients were compared between the two groups with CD8+ T cell count above and below 258/mm3 in Table 4. Patients with higher CD8+ T cell counts tended to present with earlier but not statistically different clinical manifestations of IRIS compared to patients with lower CD8+ T cell count.
Discussion
In summary, ocular IRIS was mostly related to CMV infection, and the median duration from HAART initiation to ocular IRIS diagnosis is 2.3 months. The CD4+ and CD8+ T cell counts at the time of ocular IRIS diagnosis were higher than before HAART initiation. This study showed a negative relationship between CD8+ T cell count before HAART initiation and duration from HAART to ocular IRIS. Patients with a higher CD8+ T cell count before HAART initiation developed ocular IRIS symptoms earlier, as CD8+ T cells play a role in controlling viral replication and killing infected cells [18]. Thus, monitoring CD8+ T cells may be useful in cases with low CD4+ T cell counts who may be at higher risk for IRIS.
Our study consecutively reviewed all HIV-infected patients who visited the ophthalmology clinic after the initiation of HAART and analyzed all types of ocular IRIS, including IRU in inactive CMV retinitis, active CMV retinitis-associated IRIS, and mycobacterial choroiditis. Since the introduction of HAART, many researchers have investigated the incidence, risk factor, visual outcome, and complication of IRU in inactive CMV retinitis [11,19-23]. Recently, the concept of active CMV retinitis-associated IRIS has been suggested, describing active CMV retinitis occurring in the context of successful HAART initiation [6-8]. Moreover, choroiditis related to mycobacterial infection was also reported as another form of IRU [9], albeit less frequent than CMV-related ocular IRIS. Here, we reviewed IRU in inactive CMV retinitis, active CMV retinitis-associated IRIS, and mycobacterial choroiditis altogether and analyzed their overall clinical characteristics. Yeo et al. [23] reported median durations from HAART initiation to IRU diagnosis of 214 days while Ruiz-Cruz et al. [6] reported the durations from HAART initiation to immune recovery retinitis (IRR) diagnosis of 28 days. Our study’s median duration from HAART initiation to ocular IRIS diagnosis which included both IRU and IRR was 70 days, which lies in between the average value of the previous studies. Also, median increase of CD4+ T cell count from HAART initiation to diagnosis was 155.5/mm3 in Yeo et al. [23]’s IRU study and 54/mm3 in Ruiz-Cruz et al. [6]’s IRR study. In our study, the median increase was 107/mm3, which again is approximately the average of the two previous studies. This shows our study’s inclusion of both IRU and IRR and integration of these two entities into ocular IRIS by the definition.
There was a negative relationship between CD8+ T cell count before HAART and the duration from HAART initiation to ocular IRIS diagnosis. This correlation suggests that patients with higher CD8+ T cell count showed earlier manifestation of ocular IRIS. Research into the relationship between CD8+ T cell count and ocular IRIS after HAART has been limited to histologic examination, with Mutimer et al. [14] demonstrating a predominance of CD8+ CMV-specific T cells in IRU patients. Other research has shown that CD8+ T cells are related to the IRIS process in extraocular manifestations. Gray et al. [24] examined brain tissue of central nervous system IRIS and found that inflammation was mostly composed of CD8+ lymphocytes, especially in lethal cases. In IRIS related to tuberculosis, activated CD8+ T cells transiently peaked during IRIS [25]. As immune reconstitution process itself is mediated by CD8+ T cell, it can be inferred that if the baseline CD8+ T cell count is higher before HAART, the immune recovery process is more readily initiated, thus resulting in earlier manifestation of ocular IRIS.
The recommended interval for routine ophthalmic examinations in patients with low CD4+ T cell counts may vary across different studies. Jabs et al. [26] recommended that patients with CMV retinitis undergoing HAART receive ophthalmic examination every 3 months. Later, Ruiz-Cruz et al. [6] defined the term CMV IRR and suggested weekly ophthalmic follow-up for patients with CD4+ T cell count lower than 50/mm3 after starting HAART. Our study divided ocular IRIS patients into two groups based on CD8+ T cell count (>258/mm3 and <258/mm3) and found that 90% of patients with CD8+ T cell count higher than 258/mm3 before HAART developed ocular IRIS within 70 days (median, 47 days; IQR, 38-56 days), while 88.9% of patients with CD8+ T cell count lower than 258/mm3 before HAART developed ocular IRIS after 70 days (median, 160 days; IQR, 130-162 days). This result is comparable to Ruiz-Cruz et al. [6]’s study where median CD8+ T cell count was 428/mm3 (IQR, 223-697/mm3) and median time to IRIS after HAART initiation was 4 weeks (IQR, 3-8 weeks). Therefore, authors speculate that patients with high risk of ocular IRIS (e.g., CD4+ T cell count lower than 50/mm3) may require ophthalmologic examination at 4 to 8 weeks after HAART initiation if they have CD8+ T cell counts higher than 258/mm3 prior to initiation of HAART. In this study, 10 of 19 patients (52.6%) had CD8+ T cell count higher than 258/mm3, which suggests that this follow-up interval can both achieve close ophthalmic screening and prevent excessive follow-up schedule. However, to suggest a recommendation for ophthalmologic examinations follow-up interval, further validation in larger and prospective cohorts, including patients with HAART who did not develop ocular IRIS, is necessary in future study.
This study has some limitations, such as recruitment of patients from a single tertiary medical center, resulting in a small sample size and selection bias. Furthermore, since it was retrospective study, not all patients initiating HAART was referred to the ophthalmology clinic, and not all patients diagnosed with IRIS had undergone CMV activity test such as polymerase chain reaction test. Additionally, the 6-month follow-up period may not fully capture long-term outcomes and recurrence rates. Future studies should consider longer observation periods to provide a comprehensive understanding of IRIS outcomes.
Overall, this study aimed to clarify the relationship between ocular manifestation and T cell immunity in patients with ocular IRIS in a broad sense. The duration from HAART initiation to ocular IRIS diagnosis was negatively correlated to CD8+ T cell count prior to HAART. Ocular IRIS with higher CD8+ T cell count before HAART developed earlier after HAART initiation compared to those with lower CD8+ T cell count.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
