This is the first population-based study in Kenya to estimate the prevalence of KC and confirms the anecdotal observations of a high prevalence of the disease among adolescents. Our study found the prevalence of KC and KCS to be 1.7% and 3.8%, respectively.
As expected, we found a lower prevalence of KC compared to facility-based studies previously reported in Africa [16] which included high-risk individuals with either allergic conjunctivitis [30, 31] or astigmatism [32] or seeking refractive surgery [33, 34] or CL services [35]. One exception was a study in Ghana, where Kobia-Acquah et al. [36] reported a prevalence of < 1%, possibly due to low uptake of referrals, under-diagnosis and/or misdiagnosis of the condition.
When compared with other studies conducted among adolescents, the prevalence of KC in Kenyans was lower than that found in Saudi Arabians (4.8%) [15], similar to that in Mexicans (1.8%) [37], Maori islanders (2.3%) [20] and Emiratis (2.7%) [21] but higher than that in Caucasians (0.52%) in New Zealand and Norway (0.2%) [20, 22] and Asians in South Korea (0.06%) [7]. The variation in the prevalence of KC in these studies could be explained by differences in a combination of factors such as ethnicity, consanguinity, sunlight exposure, ocular allergy or corneal thickness.
Ethnicity, rather than geographical location, is likely to affect the prevalence of KC. Maori and Pacific ethnicities in New Zealand [20, 38] and Latinos and Blacks in the US [39] were found to have a higher prevalence of KC than Caucasians living in the same country. This implies the contribution of genetic and heredity factors in the aetiology of the disease [2, 40, 41].
Consanguinity is a traditional practice in countries that form a belt spanning from Morocco to Pakistan across North Africa, the Middle East and Western Asia, including South India ranging from 20–50% [42] of all marriages which could account for the high prevalence of KC in these populations [41, 43, 44]. The prevalence of consanguinity across most of Africa is generally unknown [45], with one study conducted in Cameroon reporting a prevalence of 1.9% for first-cousin marriages [46].
The prevalence of KC appears to be higher in countries with hot and dry climates compared to cooler climates [47] which could explain the higher prevalence of KC in Kenya. It is thought that excessive exposure to UV light might cause oxidative stress to the cornea [48]. However, studies in animals [49, 50] and humans [51] have not shown a definitive association. Additionally, confounding variables such as ethnicity and genetics, make it difficult to assess its contribution towards the aetiology of KC.
Higher prevalence rates of vernal keratoconjunctivitis (VKC) have been reported in more African, Mediterranean and Middle Eastern countries than in Western countries [52]. In African children, the prevalence of VKC ranges from 4.0–39.9% [53–55]. In Kenya specifically, allergic conjunctivitis accounts for 27% of outpatient visits in eye clinics [56] and the prevalence of KC in children with allergic conjunctivitis was 30.9% [30]. In this study, more than half of the students with KC had allergic conjunctivitis and reported rubbing their eyes, similar to other studies in children [57]. Kerosene and firewood cooking, dust exposure and animal contact were reported to be significantly associated with VKC in Ethiopia [54, 55]. Ocular allergy is likely an indirect cause of KC and the eye rubbing caused by it is a risk factor for the development of KC and not the allergy itself [13, 18, 58]. Eye rubbing has both bio-mechanical and bio-cellular effects on the cornea. It is thought that the friction from eye rubbing increases epithelial temperature and intraocular pressure resulting in the collagen fibres and the proteoglycans slipping, deforming and thinning the cornea, causing it to lose rigidity and steepen [59, 60]. In addition, eye rubbing causes the release of proteolytic enzymes and inflammatory mediators from epithelial and stromal cells, resulting in the apoptosis of keratocytes, which releases enzymes that degrade the collagen fibrils and extracellular matrix leading to the onset/progression of KC [58, 61]. With such a high prevalence rate in the current study, it is advised that practitioners screen all children with VKC and that all those with a scissors reflex on retinoscopy be referred for corneal tomography. This will aid in early diagnosis and referral for CXL treatment to retard further development in those diagnosed with incipient KC. By managing the ocular allergy with medication and patient education, eye rubbing could be reduced and KC onset/progression prevented.
Similar to many studies in Africa [16], and around the world, we did not find a significant difference between the prevalence of KC and gender [4, 7, 12, 13]. However, there are conflicting results with some studies showing a higher prevalence in men [10, 11, 57, 62–64] and a few in women [5, 14]. Sex hormones and hormonal imbalances in both males and females may play a role in the development of KC [65, 66].
The majority (n = 45; 88%) of students with KC were not aware of their diagnosis, despite half of them having moderate to severe KC on presentation, the reasons for which could be due to the complex nature of the disease, personal factors and the lack of affordable, accessible and quality paediatric eye care services [67]. Rono et al. (2019) reported that in Kenya there is a lack of awareness of eye conditions, low awareness of services available, low parental education and fear of costs associated with accessing services and interventions [68]. KC is asymptomatic in the early stages and asymmetric where one may not appreciate a reduction in vision until the better eye gets affected and there could be a short interval between noticeable symptoms and progression due to the ‘explosive’ nature of the disease in children [69]. Children find strategies to cope with reduced vision by sitting closer to the blackboard, squeezing their eyes or may not report mild visual impairment. Health-systems barriers include a lack of school screening programs, lack of insurance to cover spectacles and the lack of knowledge, skills and equipment among mid-level ophthalmic workers (MLOWs) to screen, diagnose and manage KC [26]. Including KC tests in school screening programmes and during routine eye exams, conducting awareness campaigns among students, teachers and parents, upskilling MLOWs and providing funding for spectacles and CXL in children would result in early identification and prompt management.
A strength of this study is that it is the first to report the prevalence of KC in Kenya and that the sample size was relatively large, randomly selected and representative of the national census ethnicity data. In addition, a final diagnosis of patients with KC was based on both corneal tomography and slit lamp examinations. We had a few limitations; Firstly, the battery of tests used at school could not determine a posterior corneal profile and hence some children with posterior corneal steepening only may have been missed, affecting the final prevalence estimate. Secondly, more than half of the students did not return the consent forms. Strategies to increase participation need to be explored for future studies and more studies on KC prevalence in different parts of the country are recommended.