Dry eye disease (DED) is a:

The disease disrupts the tear film, causing symptoms such as ocular discomfort, itching, tearing and irritation.² The tear film, consisting of the lipid and muco-aqueous layers, protects the eye from infections and foreign particles, creating a moist and healthy surface. Dysfunction of each layer can result in a specific form of DED. Untreated DED may lead to progressive ocular surface diseases, which may lead to corneal scarring; making accurate, timely diagnosis and appropriate management important.³

Dry eye disease is considered a significantly growing public health problem worldwide, with global surveys estimating prevalence, in various age groups, ranging between 5% and > 80% across different countries.⁴ Although there is limited scholarly evidence, the prevalence across the province of KwaZulu-Natal in South Africa has been anecdotally reported as being high, by both private and public sector practitioners.

According to the Tear Film and Ocular Surface Society Dry Eye Workshop (TFOS-DEWS), DED is classified into three categories, which are aqueous deficient dry eye (ADDE), evaporative dry eye (EDE) and mixed, when there is an overlap between the two types.¹ Clinical assessment enables DED to be classified as mild, moderate or severe.⁵ Practitioners generally do the Schirmer and tear break-up time (TBUT) tests or rely on DED symptom questionnaires to diagnose DED.⁶ Siong et al.⁶ advise that while most clinicians still practise the traditional clinical dry eye tests, such as Schirmer and TBUT, there needs to be a shift from traditional clinical testing methods to non-invasive diagnostics, such as utilisation of the Oculus Keratograph^® 5M (Oculus Optikgeräte GmbH, Wetzovir, Germany), a corneal topographer and dry eye analyser.

Despite DED being identified as a common disorder, it remains largely underdiagnosed in many healthcare facilities, even among patients presenting with other ocular disorders. A review of the scholarly literature revealed very few studies having been undertaken in South Africa on dry eye prevalence, of which, three were facility based and revealing a high DED prevalence (40% – 92%).^7,8,9,10 No information is reported on the DED clinical profile of patients within the public health sector in the province of KwaZulu-Natal (KZN).

McCord Provincial Eye Hospital (MPEH) is the only public specialist eye facility in KZN and serves as the referral centre for eleven surrounding districts. Hospitals and clinics refer patients to MPEH as the majority do not have ophthalmic or other eye care services. Approximately 1250 patients with a diverse range of ocular conditions present at the outpatient department each week. The researcher’s clinical observation reveals that most of the glaucoma, keratoconus and vernal keratoconjunctivitis (VKC) patients who present for treatment at MPEH also have DED. To date, there is no policy, defined protocol or clinical guideline on DED management, which may be because of a lack of empirical evidence of DED prevalence and profile to inform planning. This study aimed to determine the clinical profile of DED patients at a tertiary hospital to inform a public sector intervention strategy.

A quantitative, single-centre, cross-sectional, descriptive study design was applied. Six hundred and two MPEH (tertiary eye hospital) patients, seated in the waiting area of the hospital were recruited to participate in the study between July 2023 and August 2023. Males and females aged 7 years and older were included, and patients with any ocular surface pathology (e.g. lid abnormality, corneal scarring, active ocular disease), recent intraocular surgery (< 3 months) or intellectual impairment were excluded. The study complied with the principles of the Helsinki Declaration for human participants, and informed consent and assent (for children) were obtained from study participants.

A total of 602 patients, aged 7 years – 88 years, were included in the study. The mean age was 48.56 years (s.d. = 18.74), and females were the majority (78.6%). ‘Most participants were of African ethnicity (83.2%; n = 501), with the remaining ethnic groups (white people [1%], indian people [13.5%] and mixed-race people [2.3%]) combined, only comprising 16.8% (n = 101)’.¹⁰ The majority of participants (n = 501; 83.2%) were diagnosed with DED. The demographic profile of patients with and without DED is shown in Table 2. Binary logistic regression was performed to identify factors associated with DED. Age, occupation, diseases and medication were significantly associated with DED with the older population being more likely to experience DED.

Data analysis identified several significant risk factors for DED, including systemic diseases, medication, blink rate, contact lens wear, sleep apnea and previous cataract surgery. Although more participants presented with hypertension than other systemic diseases, the association with DED was found to be highest (p = 0.01) in patients with diabetes and glaucoma (Table 3).

Of the risk factors reported, systemic diseases (p = 0.006), medication (p = 0.01)¹⁰ and blink rate (p = 0.002) were statistically significant. There were 327 (65.3%) participants on medication, of which 268 (82%) participants were on systemic (Beta-blockers, Antiretrovirals [ARVS] and non-steroidal anti-inflammatory drugs [NSAIDS]) and 59 (18%) participants on topical medication. The topical medication was mostly antiglaucoma topical drops (78%), with 22% participants using anti-allergic eye drops. Participants with DED had significantly lower blink rates than those without DED (p = 0.002).

The correlation between clinical tests (TBUT and Schirmer) for both eyes and SPEED scores is summarised in Figure 4. Spearman rank-order correlation was used to examine the relationships between subjective symptom scores (SPEED) and objective tear metrics (TBUT and Schirmer 2) for both eyes. There was a very weak negative correlation between SPEED and TBUT for the right (r = −0.13) and left eyes (r = −0.15; p < 0.05), suggesting that higher symptom scores were associated with slightly reduced tear stability. No significant correlation was found between SPEED and Schirmer 2 (r = −0.001; p > 0.05). Although statistically significant, the correlations were weak and unlikely to be clinically relevant.

The mean SPEED score of the DED group was 6.04 which reveals that, on average, there was mild (n = 209; 41.7%) to moderate (n = 137; 27.3%) DED (p = 0.03) (Table 4). Spearman correlation test analysis revealed significant differences in DED severity, with EDE (n = 165; 52.5%) and mixed (n = 58; 34.1%) being mostly moderate, while ADDE had significantly more participants with mild severity (p < 0.001).

The majority (n = 314; 62.7%) of participants with DED had EDE and were between the ages of 41 years – 55 years. There were very few (n = 16; 3.2%) with ADDE, with most of those affected (n = 6) being within the demographic of 19 years – 40 years old.

Meibomian gland dysfunction and meibomian gland dropout revealed a significant association with DED (p < 0.001) and both variables also increased with age. Fisher’s exact test showed no statistically significant association between dry eye type (EDE, ADDE or mixed) and gender (p = 0.54). Corneal and conjunctival staining was only seen in the DED group (p = 0.007). There was a significant association between meibography and corneal staining (p = 0.002). Further, patients with moderate or severe meibography were almost twice as likely to have moderate or severe corneal staining (OR = 1.95; 95% CI: 129–196). There was a significant association (p < 0.001) between TBUT and meibography findings; as the TBUT values decreased, the meibomian gland dropout increased in severity.

Females were at higher risk of having DED, as 79 out of every 100 female participants, aged between 19 years and 65 years, presenting at the hospital had DED (Table 5). The majority (76.24%) of all participants with DED, irrespective of the associated risk factors, had moderate or severe dry eye, with those with glaucoma, hypertension and post-cataract surgery being the most severe (p < 0.001).

Dry eye disease is a common ocular condition affecting a significant portion of the population, leading to discomfort, visual disturbances and diminished standard of living. To ensure optimal patient care and inform clinical eye health services planning, it is important to understand the prevalence and profile of DED among patients attending public health facilities. The study findings revealed that most patients presenting to MPEH, from various referral centres, were females in the 40 years to 70 years age group. Dry eye disease diagnosis was established in the majority (83.2%) of the participants, with a higher prevalence in females (84.1%), a finding concurring with other studies.^14,15 However, Idiculla and Zachariah¹⁶ in a prospective study reported a slightly higher DED prevalence in males (n = 122, 51.9%). Possible reasons for the difference could be the variation in the study selection criteria, as their study included 122 males and 118 females. Further, they only included healthy participants, excluding systemic diseases, contact lens users and post-ocular surgery participants.

As shown in Table 1, DED increased with age, a well-documented DED risk factor, resulting from a decreased lacrimal response which occurs with ageing.^17,18 Further, with increasing age, the chances of having systemic diseases and taking medication increase, conditions which may result in increasing aqueous tear deficiency.¹⁹

A significant association was found between DED and occupation (p = 0.03), even though the scholars and employed group comprised only 25.4% of the total sample. These groups had a much lower blink rate (< 10 blinks per minute) than pensioners and the unemployed. This could be because of increased usage of computers and mobile devices by the employed and scholars, as studies show an increase in DED with a reported decrease in blink rate by 40% – 60% and incomplete blinking.²⁰ Work and study tasks increasingly command the prolonged use of computers, cell phones or other visual display units (VDUs), leading to DED. Furthermore, air-conditioned rooms and low-relative-humidity office environments negatively impact the tear film, exacerbating DED. These negative influences can be alleviated by modifying the ergonomics of the work environment with changes such as reducing screen brightness and glare and using the appropriate screen location. Patients should be advised on blinking exercises, taking frequent breaks and modifiable factors like air-conditioners and low humidity.

Although only 1% of patients of the total sample were contact lens wearers, it is a concern that five had DED. Similarly, Uddin³ found all six contact lens wearers in his study sample to have DED. Possible reasons could be a low blink rate and incomplete blinking in contact lens wearers. Furthermore, poor lens wettability may cause high evaporative loss during contact lens wear and contribute to tear film lipid composition changes.²¹ Contact lenses with high Dk and low water content are recommended to maintain normal homeostasis of the tear film.²² Jimenez et al.²³ cite blinking exercises and the use of preservative-free artificial tears supplements as important to contact lens wearers to reduce DED and improve patient comfort.

Similar to previous studies, such as the ‘Beaver Dam Eye Study’,²⁴ that found DED associated with systemic diseases, this study revealed a significant association with arthritis, hypertension, diabetes, thyroid disease, HIV and glaucoma.^{9,10,25,26,27} The association with diabetes and glaucoma has been found to be the highest when compared to other conditions. Zhang et al.²⁸ assert that poor glycaemic control affects both anterior and posterior segments, increasing the prevalence of diabetes-associated DED. Furthermore, diabetics also have reduced corneal sensitivity, which contributes to DED, as corneal nerves are responsible for the blinking reflex, tear production and secretion.²⁸

The profile of significant DED severity varied by systemic condition, with glaucoma and hypertension being more significantly associated with severe DED as compared to other systemic diseases. Noting that 31 out of every 100 hypertensive patients presenting at the hospital may likely have DED, an intervention by the facility is necessary. Yu et al.²⁶ stated that the association of DED with systemic diseases warrants practitioners obtaining a detailed medical history and understanding which systemic diseases predispose to severe DED to improve overall patient management. Screening for the presence of DED in patients with risk factors helps with an early diagnosis and appropriate management. The study findings support a protocol whereby primary care clinicians, managing patients with related chronic systemic diseases, include referrals for a comprehensive tear function evaluation as part of their diagnostic protocols. Tear function assessments in patients with a history of any of the systemic diseases will help with early detection and treatment, ultimately improving patient comfort and preventing visual disturbances because of a disruptive tear film or ocular surface compromise.²⁹

Ronaldo et al.²⁹ stated that the number of drugs administered systemically increases with age. The majority of study participants were on medication, many of which are known to have DED side effects, resulting in inhibition of the production of efficient tears. Only 18% were on topical medication (antiglaucoma and antihistamines), while 82% were on systemic medication, which included Beta-blockers, ARVS and NSAIDS. Some of these medications alter the tear volume and quality via the protein kinase pathway and may cause dry mouth.³⁰ Clinicians should be aware that if they cause a dry mouth, the chances of them causing dry eye are high.³¹ Fraunfelder et al.³⁰ stated that long-term usage of topical ocular therapy may cause dry eye as well.

Some components of eye drop formulations can be toxic to the ocular surface with preservatives cited as the most common offenders, potentially causing epithelial cell damage and punctate epithelial keratitis, which interferes with ocular surface wetting.^2,32 Glaucoma treatment, which decreases or controls the intraocular pressure, may also disrupt the homeostasis of the tear film.³³ At MPEH, appropriately treating glaucoma patients presenting at the facility will be challenging as they are generally prescribed with antiglaucoma drops and tear supplements with preservatives, which may worsen the DED. This problem is inherent within the health system, as clinics and hospitals only prescribe medication contained in the essential drug list produced by the National Department of Health. Noting the availability of empirical data, it may be necessary for relevant stakeholders to engage to improve care of DED patients presenting at provincial facilities. Optimal clinical protocols should include consideration of disease and medication history, taking cognisance of contraindications when managing all patients presenting at the hospital. A further suggestion is to introduce appropriate DED management clinical guidelines at all facilities. Application of guidelines may warrant prescribing preservative-free tear supplements and alternative ocular or systemic medication to alleviate DED.

Poor correlation between signs and symptoms in DED has been a challenging issue highlighted in DED research. The study revealed that the prevalence of DED was 83.2% with clinical tests while with SPEED was only 56%. The study findings correlate with others which conclude that symptoms do not always correlate with clinical signs.^34,35 Although the screening questionnaires serve to identify potential dry eye patients, comprehensive diagnosis must include clinical assessments. Gierow and Kacz,³⁶ however, found a positive correlation between clinical tests and a Tear Evaluation and Research Tool for Contact Lens dry eye questionnaire (TERTC-DEQ), which may be considered for further studies.

Although the Schirmer test is the most used test to assess aqueous tear production, results revealed no correlation between the Schirmer 2 test and the SPEED score, which is also consistent with studies that used the OSDI questionnaire for comparison.^35,37 However, TBUT, a test commonly used to assess tear film stability, was found to have a significant correlation with SPEED scores (p = 0.002). Similarly, a correlation was revealed in Lee et al.³⁸ study between the ocular surface disease index (OSDI) and non-invasive keratograph break-up time (NIKBUT), supporting the notion that subjective symptoms are more likely to indicate a lipid deficiency as opposed to an aqueous deficiency. Variations in correlations between DED tests and symptoms revealed in this study highlight the multifactorial aetiology of DED and the fact that each diagnostic assessment for DED assesses a particular layer of the tear film, contributing to inconsistent symptoms experienced. Therefore, solely using a symptom questionnaire as a dry eye diagnostic tool may misdiagnose a significant portion of the DED population. Study findings highlight the importance of doing clinical tests together with a dry eye questionnaire to diagnose DED. This approach is also included in the latest TFOS DEWS III recommendation, which states that an abnormal score in the symptomatic dry eye questionnaire (OSDI) plus one of the tear film markers (NIKBUT, TBUT, osmolarity) and ocular surface staining (lid, corneal and conjunctival) is considered as a positive diagnosis of DED.³⁹ This recommendation, endorsed in this study, is further supported by Kyei et al.³⁵ who advise that dry eye diagnosis should focus on symptom assessment and clinical test results.

The study showed an association between TBUT and meibography (p < 0.001); as meibography increased in severity, so did TBUT. However, there was no correlation between Schirmer findings and TMH. This requires further investigation to strengthen the call for a complete shift from traditional clinical testing methods to non-invasive diagnostics, as recommended by Siong et al.⁶ Further, studies are needed to support the inclusion of non-invasive tests in resource constrained public sector facilities to enhance DED diagnosis, classification and management as advised in other studies.⁴⁰

Dry eye disease severity was graded according to the types of clinical tests done. Decreased TBUT indicates a tear film instability and is a vital indicator in diagnosing and treating DED. In this study, moderate dry eye was most prevalent and DED severity was (45.3%). Dry eye severity was significantly associated with age (p = 0.06). This could be because of the beginning of menopause. However, these results are in discordance with Chalmers et al.⁴¹ who underestimated dry eye severity by at least one grade compared to subject’s self-assessment, especially among the elderly and females. Differences may be attributed to clinicians in their study mainly relying on the subject’s global dry eye self-assessment as opposed to the use of clinical tests or both.

Successful management of DED depends on early diagnosis, accurate DED subtype classification and severity determination. Garcia et al.¹³ advise that a patient’s dry eye classification and severity status should be comprehensively diagnosed, as individualised treatment will ensure improvement of the patient’s therapeutic outcome.

Corneal and conjunctival staining were only seen in the DED group and in those with severe DED (p < 0.001). Studies report that DED symptoms may occur before the evidence of ocular surface damage; thus, staining alone is not appropriate to diagnose DED.⁴² However, corneal and conjunctival staining is important in determining dry eye severity and facilitates the measurement of ocular surface inflammation in DED.⁴³

Increased tear evaporation, tear hyposecretion and/or mucin dysfunction are known aetiologies of DED.⁴⁴ In the current study, more than half of the DED group had reduced TBUT, with most having MGD and meibomian gland dropout, contributing to the diagnosis of EDE. As shown in Table 3, EDE was most prevalent, with only 3.2% of participants solely having ADDE. This is not unexpected as EDE, or reduced TBUT, was noted by the Asia Dry Eye Society as a predominant form of DED.⁴² Other studies also reported that more than 80% of DED cases fall into EDE or mixed, with 10% being solely ADDE.^45,46 Findlay et al.⁴⁷ suggest that EDE management should focus more on underlying ophthalmic and systemic conditions and aggressive therapies to protect the ocular surface rather than mainly focusing on artificial tears and warm compressors. Public health sector facilities in KZN could note this treatment recommendation in the DED management resource planning as currently, many only provide artificial tear supplements to treat EDE.

The study revealed high DED prevalence (83.2%), which concludes that a significant number of DED patients are presenting at public sector hospitals in the province of KZN. Evaporative dry eye is the most prevalent type, and significant associations were revealed among DED, age, systemic diseases and medication. Notably, hypertension and glaucoma may lead to patients having a more severe form of DED. As many patients routinely visit their primary care practitioners to manage their systemic diseases and obtain chronic medication, dry eye assessments should be included as part of the comprehensive care of patients with diabetes, hypertension, glaucoma, HIV, arthritis and other related diseases. Given the high DED prevalence and clinical profile data contributed by this study, the authors recommend implementing strategies to achieve early diagnosis, prevention and appropriate treatment for DED patients. A multifaceted approach to manage this complex condition should be considered by appropriate public sector decision-makers and additionally, the optometry minimum competency standards be revised and updated accordingly by the regulator.