Key findings:

Key findings of the abstract include a hospital-based cross-sectional study investigating the correlation between CD4 count and dry eye severity in HIV-positive patients. Significant mild positive correlations were found between CD4 count and Schirmer's test without anesthesia, Schirmer's test with anesthesia, and Tear Film Breakup Time, highlighting the importance of dry eye treatment in HIV patients and the potential use of CD4 count as an indicator of dry eye severity.

What is known and what is new?
Ophthalmic manifestations of HIV infection, including dry eye, are known, but the correlation between CD4 count and dry eye severity is less explored. This study establishes a significant mild positive correlation, suggesting CD4 count as a potential indicator of dry eye severity in HIV patients

What is the implication, and what should change now?
The study underscores significant correlations between CD4 count and dry eye severity in HIV-positive patients, suggesting CD4 count as a potential indicator. It emphasizes the importance of recognizing and treating dry eye in HIV patients, urging integration of dry eye management into routine care protocols to enhance patient outcomes.
 

HIV (Human Immunodeficiency Virus) continues to be a major global public health issue, having claimed more than  39 million lives so far [1]. Acquired Immunodeficiency syndrome (AIDS) is one of the most feared infectious diseases of  the late 20th century. Since its discovery in 1981, AIDS has emerged as a global health problem. Thirty years after its  discovery, the virus has reached virtually every corner of the globe [2]. India has a large number of patients with AIDS, which  is the third largest population of this group in the world. [3] 

According to a report which was made by the National AIDS Control Organization, 2017-2018, adult HIV prevalence in India was estimated at 0.26% (0.22% – 0.32%). The total number of people living with HIV (PLHIV) in India is estimated to be around 21.40 lakhs (15.90 lakhs –28.39 lakhs) in 2017. India was estimated to have around 87.58 (36.45  – 172.90) thousand new HIV infections in 2017, showing a decline in HIV infection by 85% since the peak of 1995 and  by 27% between 2010-2017. [4] 

HIV causes a wide spectrum of diseases and it is undoubtedly a multisystem disorder, but the ophthalmic disease does affect 70-80% of the patients with HIV infection. Various studies have shown that about 40-45% of the HIV infected patients have some or the other ophthalmic manifestations. Lifetime cumulative rate of at least one abnormal ocular lesion in HIV positive patients is 52%–100% in different studies. [5] 

Though most HIV infected patients live in developing countries; there are only a few reports on ophthalmic  manifestations of HIV from the developing countries of the world [6].  Ophthalmic manifestations of HIV infection are  diverse. Both anterior and posterior segments of the eye can be involved and it may even lead to blindness. [7] 

Keratoconjunctivitis sicca or dry eye occurs in later stages of the disease in 10 -20% of the patients. An abnormal Rose Bengal staining and Schirmer’s test is invariably detected in these patients. The cause of dry eye in HIV patients is  \complex. Combined effects of HIV-mediated inflammation and destruction of the lacrimal and salivary glands and direct  HIV infection of the conjunctiva is the proposed mechanism. [7] 

As per literature there are very few studies in which they tried to correlate the severity of dry eye with CD4  count especially in North-western part of India  [7].  Therefore, a present study was planned to see the  correlation between the level of CD4 count with the  severity of dry eye in HIV seropositive patients. 
 

Study Design, Settings and Participants: 

It was a hospital based cross-sectional study  conducted over a period of one year from June 2018 to  May 2019 in the ophthalmology department of a tertiary  care teaching hospital in Himachal Pradesh, India. All HIV-positive patients between 18 to 50 years of age group, registered at the ART centre and referred to  Ophthalmology OPD for ocular complaints and  asymptomatic patients from ART centre were selected  randomly for the study. Patient with confounding  factors and comorbidities like diabetes, corneal scars,  lid abnormalities, contact wearers, SJS, arthritis, any  other connective tissue disorder; patients who were  already on treatment for DES, oral contraceptive pills,  pregnant females and those taking any medications like antidepressants, antihistamines, anxiolytics which can  independently cause dry eye and seriously ill patients  who were not able to cooperate for ophthalmological  examination were excluded from the study. Total 346 HIV patients were examined but after considering inclusion and exclusion criterias 120 participants were  finally included in our study. Every patient was on Highly Active Antiretroviral Therapy (HAART). 
 

After taking written informed consent from the study subjects, data was collected in a predesigned, pre tested, semi-structured interview schedule in which socio-demographic profiles like age, gender, occupation, duration of ART, duration of disease etc were collected.  After taking history as per proforma, each patient has undergone detailed ophthalmological examination in the following sequence: 

⮚ Dry Eye Questionnaire  

Ocular Surface Disease Index Score (OSDI) 
The OSDI questionnaire was administered to all patients and by a single examiner. To those who were non conversant in English, the questions were explained to the patients in their local language. The OSDI questionnaire has 12 items, with each question given a score ranging from 0 (none of the time) to 4 (all of the  time). The final score was calculated by multiplying the sum of all the scores by 25 and then dividing the total  by the number of questions answered. Scores range  from 0 to 100 with 0–12 representing normal, 13–22  representing mild DED, 23–32 representing moderate  DED, and ≥33 representing severe DED. [8,9] .

⮚ Slit Lamp Biomicroscopy 
Detailed evaluation from anterior segment to posterior segment was done. Relevant to our study detailed evaluation of lid. Eye lashes, conjunctiva, cornea, sclera, tear film height, lens was done. 

⮚ TBUT (Tear Film Break Up Time)  
Tear break-up time was performed by moistening a fluorescein strip with sterile non-preserved saline and applying it to the inferior tarsal conjunctiva. After several blinks, the tear film was examined using a broad beam of the slit-lamp biomicroscopy with a cobalt blue filter. The time lapse between the last blink and the appearance of the first randomly distributed dark discontinuity in the fluorescein -stained tear film is the tear break-up time. Break-up times less than 20 seconds was considered abnormal. This procedure was repeated three times on both eyes.  

⮚ Schirmer-1 Test Without Anaesthesia 
The Schirmer test was performed by placing a  narrow filter-paper strip in the inferior cul-de-sac.  Aqueous tear production was measured by the length in  millimeters that the strip wets during the test period,  generally 5 minutes. Test was performed with eyes  closed. Serin et al. suggested that administering the  Schirmer test with the patient’s eyes closed produces  less variable results and greater repeatability.[13] While  an isolated abnormal result can be nonspecific, serially  consistent low results are highly suggestive of aqueous  deficiency. [9,11,12] Severity grading was done as follows; >15 mm/5 min (Normal); 10-15 mm/5min (Mild Dry  Eye); 5-9 mm/5min (Moderate Dry Eye) and <5  mm/5 min (Severe Dry Eye). 

⮚ Schirmer's Test 1 With Anaesthesia  
The Schirmer test with anesthesia, also referred to as  a basic secretion test, has been reported to give more  variable results than the Schirmer test done without  anesthesia. If topical anesthesia is applied, excess fluid  should be gently removed from the cul-de-sac prior to  insertion of the filter paper. This test is performed after 15 minutes of Schirmer’s test 1. After putting topical  proparacaine drops, Schirmer's strip was applied as for  Schirmer’s test 1. Results noted down after 5 minutes.  <6mm/5 min was considered abnormal and >6 mm of  wetting after 5 minutes was considered normal. 

⮚ Rose Bengal Staining  
Rose Bengal staining was done and graded as van  bijsterveld scoring. Rose Bengal score > 4 was  considered abnormal and < 4 was considered normal.  Maximum score was 9. Intensity scored in 2 exposed  nasal conjunctival zones and cornea. It was started after  putting anaesthetic drop to avoid irritation as this stain  is slightly toxic to epithelial cells. [8,9] 

Posterior segment examination was also performed  in every patient. Those patients who were found to have  abnormalities other than dry eye were treated  accordingly. 

Data were analyzed and statistically evaluated using  SPSS software, version 17 (Chicago II, USA).  Quantitative data was expressed in mean, standard  deviation while qualitative data were expressed in  percentage. Statistical differences between the  proportions were tested by chi square test or Fisher’s  exact test. Spearman correlation coefficient was used to see the correlation between two quantitative variables.  ‘P’ value less than 0.05 was considered statistically  significant. 

Ethical Issues 
All participants were explained about the purpose of  the study. Confidentiality was assured to them along  with informed written consent. The study was approved by the Institutional Ethical Committee 
 

Out of 120 patients, 67 were males (55.8%) and 33 were females (44.2%). In 40 (33.3%) patients, CD4  count was > 500 cells/mm3 where, in 50 (41.7%) patients  it was between 200 to 500 cells/mm3and in 30 (25%)  patients, CD4 cell count was <200 cells/mm3.  

OSDI score was found to be positive for dry eye in  67 patients (56%). Out of 40 patients with CD4 Count >  500 cells/mm3 OSDI was abnormal in 21 (52.5%)  patients, in patients with CD4 count between 200 to 500  cells/mm3 OSDI score was abnormal in 27 (54.0%)  patients while in patients with CD4 count <200  cells/mm3 OSDI score was abnormal in 11 (36.7%)  patients. However, trends for mild and severe dry eye  cases have shown that patients' severe dry eye were  increasing as CD4 cell count diseases but this  association was found statistically not significant  (p=0.24) (Table 1). 
 

Table 1: Association of CD4 Count with OSDI Score 

TBUT (Tear Film Breakup Time) was normal in  124 (51.7%) eyes while in 116 (48%) eyes TBUT was  abnormal. In patients with CD4 Count > 500 cells/mm³,  between 200 to 500 cells/mm³and <200 cells/mm³ 

TBUT <5 seconds was seen in 2.5%, 5.0% and 10.0%  patients respectively. This association was found  statistically significant (p=0.04) (Table 2). 

Table 2: Association of CD4 count with Tear Film Breakup Time (TBUT) 

Schirmer-1 test with anaesthesia was found abnormal in 81 (33.7%) eyes while on Schirmer-1 test without anaesthesia 31% (75) were found to be abnormal. No significant association was seen between Schirmer-1 test without anaesthesia  with CD4 count but CD4 count was significantly associated with Schirmer-1 test with anaesthesia (p <0.001) (Table 3). 

Table 3: Association of CD4 Count with Schirmer -1 Test With and Without Anaesthesia

On rose Bengal staining 25 (10.4%) eyes were  found positive. Highest number of positive eyes  (16.7%) were seen in patients with CD4 count <200  cells/mm³ compared to patients with CD4 Count > 500  cells/mm³ (12.5%) and CD4 count between 200 to 500  cells/mm³(5.0%). This association was also found  statistically significant (p=0.04) (Figure 1). 

On using spearman correlation coefficient,  significant mild positive correlation was seen between  CD4 count and Schirmer-1 test without anaesthesia  score (r value =0.23; p value <0.001), Schirmer -1 with  anaesthesia score (r value =0.22; p value <0.001), Tear  Film Breakup Time (TBUT) (r value =0.28; p value  <0.01) but no correlation was seen with OSDI score (r  value = -0.01; p value =0.83). 

Figure 1: Association of CD4 Count with Rose Bengal Staining  

Kerato-conjunctivitis sicca has been reported as one  of the most common anterior segment manifestations.  Various studies conducted on the normal population have  shown that the prevalence of dry eye ranges between  10% and 20% . [7,14-16]

The etiology of dry eye disease in HIV patients is  usually thought to be due to HIV-mediated lymphocytic  infiltration of the lacrimal gland. This leads to the  destruction of lacrimal acini and the ductal system, as  well as direct conjunctival damage [17]. Dry eye can also be due to blepahritis in these patients, pathogenesis of  which is explained as reduced ability in these patients to  control normal flora and more complex changes in  cutaneous glands of eyelids. Meibomitis has been the  causative factor in lipid layer dysfunction in these  patients.[18] The resulting keratoconjunctivitis  contributes to a chronic inflammatory state, further  promoting cytokine secretion, destruction and  dysfunction of the lacrimal gland, and loss of tear  production.[19] Since HIV has been isolated from retina,  it is believed that ocular complication results either  from a general decrease of host immunity as seen with  decreasing CD4 cell count more likely, direct ocular infection by HIV through the blood-aqueous barrier  and/or blood-retinal barrier. [17,20] 

In our study, association between OSDI and CD4  cell count for detecting the dry eye severity as disease  progresses was not found to be statistically significant  (p>0.005), but the fact that 56% of our study subjects  had dry eye symptoms should not be ignored and it  clearly indicates the magnitude of problem in these  HIV/AIDS patients. To the best of my knowledge no  study has been done so far which has used an OSDI score  for diagnosing dry eye in PLWHA. 

In our study it was observed that in patients with  CD4 Count > 500 cells/mm3, between 200 to 500  cells/mm3and <200 cells/mm3 TBUT <5 seconds was  seen in 2.5%, 5.0% and 10.0% patients respectively.  This association was found statistically significant  (p=0.04). Findings of our study were in concordance  with Gowda HT et al [7] in which patients with CD4  count of 101-500 cells/mm3 the tear film breakup time  was <10 s only in 12 eyes (27.27%). In patients with 0- 100 cells/mm3and >500 cells/mm3only 20.59% and  4.5% of the eyes had dry eyes. Our results were also  comparable with Mathebula et al. they have also found  increasing dry eye cases and disease severity is  increasing tested with TBUT. [21] 

In the present study, Schirmer-1 test with anaesthesia  was found abnormal in 81 (33.7%) eyes while on  Schirmer-1 test without anaesthesia 31% (75) were  found to be abnormal. No significant association was  seen between Schirmer-1 test without anaesthesia with  CD4 count but CD4 count was significantly associated with Schirmer-1 test with anaesthesia (p <0.001). Similar  significant association was found in study by Gowda  HT et al. [7] .

On rose Bengal staining 25 (10.4%) eyes were found positive. Highest number of positive eyes  (16.7%) were seen in patients with CD4 count <200  cells/mm3 compared to patients with CD4 Count > 500  cells/mm3 (12.5%) and CD4 count between 200 to 500  cells/mm3(5.0%). This association was also found  statistically significant (p=0.04). In a study by Gowda HT  et al[7], Rose Bengal tested in the count of 0-100  cells/mm3, only 7 eyes were positive for Rose Bengal  staining which constitutes 20.59% and in 101-500  cells/mm3only 29.55% eyes were positive. But, with a count of >500 cells/mm3 100% were negative (P =  0.018). This shows concordance with our study.  
 

Present study concluded that positive and  statistically significant correlation was found between  CD4 cell count and Schirmer-1 test without and with  anaesthesia and TBUT only. Hence dry eyes in PLWHA  should not be looked down upon and should be treated  for the same. More studies with larger sample size are  needed to see the ocular involvement and correlation of  CD-4 cell count with dry eye severity for treatment and  better visual prognosis. 

Acknowledgement: We are grateful to all the  participants of the study.

Funding: No funding sources.

Conflict of interest: None declared.

Ethical approval: The study was approved by the Institutional Ethics Committee of Indira Gandhi Medical College Shimla.
 

1. World Health Organisation. Factsheet No 360:  HIV/AIDS, 2014 [Internet]. Geneva; WHO; 2015  [cited 2015 Nov 7]. Available from:  http://www.who.int/mediacentre/factsheets/fs360/e 

n/. 

2. Acharya, Pavana Krishnaraj, et al. "Ocular manifestations in patients with HIV infection/AIDS who were referred from the ART Centre, Hassan, Karnataka, India." Journal of clinical and diagnostic research: JCDR 6.10 (2012): 1756. https://doi.org/10.7860%2FJCDR%2F2012%2F4738.2637 

3. World Health Organization. AIDS Epidemic  Update. December 2017-2018; Geneva: WHO. 4. National AIDS Control Organization Annual  Report 2017-18. Department of AIDS Control.  Ministry of Health and Family Welfare. 

5. Sharma, Ram Lal, et al. "Ocular manifestations in human immunodeficiency virus/acquired immunodeficiency syndrome patients and their correlation with CD4+ T-lymphocyte count." Annals of Tropical Medicine and Public Health| Sep-Oct 5.5 (2012). https://cyberleninka.org/article/n/1182614.pdf 

6. Hodge, William G., Stuart R. Seiff, and Todd P. Margolis. "Ocular opportunistic infection incidences among patients who are HIV positive compared to patients who are HIV negative." Ophthalmology 105.5 (1998): 895-900. https://doi.org/10.1016/S0161-6420(98)95033-3 

7. Gowda HK, Tanushree V, Nayak S. Correlation of  CD4 count and severity of dry eye disease in  human immunodeficiency virus positive patients.  Int. J. Sci. Study. 2015;3:68-71. 

8. TFOS DEWS II definition and classification report.  Acta Ophthalmologica. 2018; 96(S261):136-40. 

9. Introduction to the Report of the International Dry Eye Workshop 2007. The Ocular Surface.  2007;5(2):69–70. Available at: https://www.sciencedirect.com/science/article/pii/S 1542012412700782 

10. Jakobiec, Frederick A., et al. Albert & Jakobiec's Principles & Practice of Ophthalmology. W.B. Saunders Company, (2008).

11. Krachmer J, Mannis M, Holland E. Cornea.  Philadelphia [u.a.]: Elsevier. 2017. (Krachmer,  Mannis and Holland, 2017). 

12. Listed NA. The definition and classification of dry  eye disease: report of the Definition and  Classification Subcommittee of the International  Dry Eye Workshop. Ocul Surf. 2007;5(2):75-92. 

13. Serin, Didem, et al. "A simple approach to the repeatability of the Schirmer test without anesthesia: eyes open or closed?." Cornea 26.8 (2007): 903-906. https://journals.lww.com/corneajrnl/FullText/2007/09000/A_Simple_Approach_to_the_Repeatability_of_the.1.aspx 

 14. Sharma, Mukta, et al. "Ocular manifestations in patients attending antiretroviral therapy centre at a tertiary care hospital in Himachal Pradesh, India." Indian Journal of Medical Research 147.5 (2018): 496-500. https://journals.lww.com/ijmr/fulltext/2018/47050/ocular_manifestations_in_patients_attending.10.aspx 

15. Hashemi, Hassan, et al. "Prevalence of dry eye syndrome in an adult population." Clinical & experimental ophthalmology 42.3 (2014): 242-248. https://onlinelibrary.wiley.com/doi/abs/10.1111/ceo.12183 

16. Sahai, Anshu, and Pankaj Malik. "Dry eye: prevalence and attributable risk factors in a hospital-based population." Indian journal of ophthalmology 53.2 (2005): 87-91. https://journals.lww.com/ijo/fulltext/2005/53020/Dry_Eye__Prevalence_and_Attributable_Risk_Factors.2.aspx 

17. Stewart, Michael W. "Human immunodeficiency virus and its effects on the visual system." Infectious Disease Reports 4.1 (2012): e25. https://www.mdpi.com/2036-7449/4/1/e25 

18. Jeng, Bennie H., et al. "Anterior segment and external ocular disorders associated with human immunodeficiency virus disease." Survey of ophthalmology 52.4 (2007): 329-368. https://www.sciencedirect.com/science/article/pii/S0039625707000586 

19. Biswas, Jyotirmay, H. N. Madhavan, and S. S. Badrinath. "Ocular lesions in AIDS: a report of first two cases in India." Indian journal of ophthalmology 43.2 (1995): 69-72. https://journals.lww.com/ijo/fulltext/1995/43020/ocular_lesions_in_aids__a_report_of_first_two.5.aspx 

20. Amsalu, Anteneh, et al. "Ocular manifestation and their associated factors among HIV/AIDS patients receiving highly active antiretroviral therapy in Southern Ethiopia." International journal of ophthalmology 10.5 (2017): 776. https://doi.org/10.18240%2Fijo.2017.05.20 

21. Mathebula, Solani D., and Prisilla S. Makunyane. "Ocular surface disorder among HIV and AIDS patients using antiretroviral drugs." African Vision and Eye Health 78.1 (2019): 1-7. https://journals.co.za/doi/abs/10.4102/aveh.v78i1.457