Overall, 79 children all HIV1 infected were included. Ocular manifestations were reported in 37 (46.7%) children.

The median age was 8 years (interquartile range; 6 - 12 years). The 6 to10 years age group accounted for 48.1% of children. The sex-ratio was 1.13. None of the children had histories of diabetes, hemoglobinopathies or impaired renal function.

The most common pathologies involved at the time of the examination were malnutrition (12.6%), pneumonia (7.6%), acute media otitis (5.1%), and prurigo (5.1%).

Immune status was normal (CD4 > 500 cells/ml) in 55 (69.6%) children aged above 5 years and in 2 children aged less than 2 years (CD4 > 25%). Viral load was undetectable (<50 copies/ml) in 34 (43%) patients.

Antiretroviral treatment (ART) was prescribed to 73 children (92.4%). Mean duration of ART was 46.69 months ± 26.76 months (range: 7 months to 9 years). Fifty (63.29%) children received concomitant ART and co-trimoxazole prophylaxis. One-third of patients (31.5%) experienced ART failure.

Ocular complaints were reported in 32 (40.5%) children and were as follows: itchy eyes (7), eye discharge (7), eye watering (6), poor eye sight (4), eye pain (3), blurred vision (2), blindness, red eye, and diplopia (1) patient respectively.

Visual acuity testing was normal (>7/10) for the right eye in 65/70 (92.8%) children and for the left eye in 67/70 (95.7%) children. No case of neuro-ophthalmologic disorder was reported. Table 1 summarizes other ocular disorders.

The risk of having ophthalmic manifestations was statically the same for the two sexes (OR = 1.57; IC₉₅ [0.54 - 4.53]).

Patients having immune deficiency and/or very high viral load (≥10 000) were more at risk for ocular manifestations involvement (Table 2). No Child with undectable viral load presented ocular disorders.

Relation between eye segments manifestations and immune status is illustrated on Table 3.

Ocular fundus lesions were found at all stages of immuno-depression. Five (5) children out of 7 who had fundus lesions had CD4 count <350 cells/mm³. Ocular fundus was abnormal in 6 children out of 73 (8.2%) who were on ART, and in 1 out of 6 who were not on ART (16.6%) (OR = 0.45; IC95 [0.05 - 12.35]).

Ocular manifestations were involved in 46.7% of HIV-infected children in this West African cohort. The prevalence is similar to those published elsewhere in Africa particularly in Uganda, Rwanda and Democratic Republic of Congo [3] [5] [7] . Prevalence rates depend on study period and children’s mean age at the time of study. In

*Tropical endemic limbo-conjunctivitis (TELC).

Africa, the highest prevalence rates of up to 75% ware recorded before triple therapy era [5] . Ocular complaints were more common (40%) among our patients who were relatively older, compared to other studies (29.7%) where children were younger (mean age 2.2 years) [8] . Indeed young children are not always capable of expressing visual symptoms nor collaborate in an effective way during ophthalmic examination, hence possible under estimation of cases. In our context, a systematic eye inspection upon HIV detection among children should be highlighted, especially when the child is young and his immune status severely affected.

Ocular manifestations were dominated by adnexia abnormalities and involved one third of our patients. The same observation was made in Tanzania [9] [10] , in the USA [11] and in France [12] . These are primarily mucco-purulent conjunctivitis which are cosmopolitan and more related to poor hygiene and bad weather conditions rather than immuno-depression status during HIV infection. The possibility of eyelashes trichomegaly in children living with HIV/AIDS has also been reported [13] .

Anterior and posterior segments manifestations were rare. Eye segments lesions are usually associated with severe immuno-depression [7] [14] - [16] , hence this was not the case for the majority of our patients on triple therapy at the time of the study. However, we cannot exclude possible recovery of initial lesions with anti-retr- oviral treatment, before our passage. Moreover, ocular fundus pathologies were more common among patients with TCD4 cells count below 350 cells/mm³. Indeed, it is at this stage of severe immuno-depression that one finds opportunistic germs (Herpes virus, toxoplasma, Varicella Zona Virus, Pneumocystis jiroveci, cryptococcus) which may have an impact on eye segments [13] . On the other hand, patients who received effective anti-retr- oviral treatment were less exposed to eye segment lesions [3] .

Contrary to the previous studies on adults [17] which reported neuro-ophthalmologic manifestations in 6% of HIV-infected patients, no case of neuro-ophthalmologic pathology was noted in our study. This is probably because the majority of our patients were on anti-retroviral treatment (92.4%) and opportunistic infections prophylaxis (70.88%). Neuro-ophthalmologic manifestations usually occur during HIV encephalopathy, opportunistic infections and tumor pathology of the central nervous system [17] .

First of all, our study confirms what other authors already described; the risk of ophthalmic lesions involvement is much higher in children with severe immune deficiency and/or high viral load [13] [16] . Indeed all children who had ophthalmic abnormalities had no undetectable viral load. Two assumptions can be made; either the initial eye lesions persist because of ART failure, or eye lesions appeared or reappeared following ART failure. Indeed the purpose of anti-retroviral treatment is to restore immunity and reduce viral replication; this would in turn reduce opportunistic infections hence eye infections. Regular eye inspection during ART is therefore very important, especially in the absence of ART success, since most eye disorders may potentially evolve to blindness [3] [14] [18] .

Secondly, boys had approximately twice more ocular manifestations than girls even if this result was not statistically significant. Indeed boys play more often outdoors (football) exposing themselves to dust.

We were not able to assess the impact of ART on eye lesions as the time of their appearance could not be specified in this type of study. Moreover, infection pathways and duration could not be clarified. Also, the number of children examined was relatively small and all of them were recruited at Yalgado Ouedraogo Teaching Hospital which is a reference hospital; therefore the results may not reflect the reality for the rest of the country.