Research Review

Does Diabetes Cause Dry Eye?

BY MARIA MARKOULLI, PHD, MOPTOM, GRADCERTOCTHER, FBCLA, FAAO

Diabetes is one of the most common systemic diseases that we encounter in clinical practice, with the American Diabetes Association estimating that 25.8 million children and adults, approximately 8.3% of the U.S. population, have the condition (Charnogursky et al, 2014). Worldwide, diabetes is thought to affect 415 million people and continues to rise, with estimates that this will affect more than 640 million people within a generation (International Diabetes Federation, 2015; Seidell, 2000). Given this significant prevalence, it is important that, as eyecare practitioners (ECPs), we are vigilant for the ocular complications associated with the condition.

Diabetes is associated with both macrovascular complications, such as cardiovascular disease, and microvascular complications, including nephropathy leading to kidney failure, potentially blinding diabetic retinopathy, and diabetic neuropathy. The prevalence of these complications rises with increasing duration of the disease. While the retinal complications of diabetes are well recognized, the effects on the ocular surface are not commonly discussed.

The Ocular Surface in Diabetes

Recent studies have reported on the association between peripheral and corneal neuropathy, showing that a reduction of corneal nerve density is indicative of peripheral neuropathy (Edwards et al, 2012; Misra et al, 2015). Even more importantly, Pritchard et al (2014) reported in a longitudinal study that reduced corneal nerve fiber length is predictive of the development of diabetic peripheral neuropathy. These findings are supported by a meta-analysis conducted by Jiang and colleagues (2016), which included 13 studies and 1,680 participants and found that corneal nerve fiber density, nerve branch density, and nerve fiber length were significantly reduced in people who have diabetic peripheral neuropathy.

Earlier detection of peripheral neuropathy would mean earlier management and prevention of debilitating complications. The fact that we can detect this noninvasively by observing the corneal nerves could translate to ECPs playing a greater role in detecting peripheral disease in the future.

Diabetic peripheral neuropathy is one of the most common complications of diabetes, estimated to affect 60% to 70% of people who have diabetes (Charnogursky et al, 2014). Given its association with corneal neuropathy, this implies that the majority of diabetic patients we see will have reduced corneal nerve density, with implications for the integrity of the ocular surface.

Corneal nerves provide trophic support to the corneal epithelial cells by releasing soluble mediators that stimulate epithelial cell growth, mitosis, differentiation, and migration (Müller et al, 2003). Corneal neuropathy can therefore lead to a disruption in epithelial integrity, predisposing the ocular surface to recurrent erosions (Arentsen and Tasman, 1981) and eventually diabetic neurotrophic keratopathy (Dogru et al, 2001).

The mechanisms that underpin these ocular surface changes to some extent overlap with the underlying mechanisms of dry eye disease (DED). The question then is: Are people who have diabetes prone to dry eye disease?

Dry Eye Disease in Diabetes

Let us consider the assessment of DED as three main categories: symptoms, tear film function, and ocular surface morphology.

Symptoms Diabetic patients reportedly complain of dry eye symptoms, including burning and foreign body sensation (Kaiserman et al, 2005; Lv et al, 2014). With longer disease duration, reduced corneal sensation results in greater tolerance to DED and corneal staining, reflecting the progression of the neuropathy (Alves Mde et al, 2008). Interestingly, dry eye signs and symptoms have also been found to correlate with the degree of peripheral neuropathy and severity of diabetic retinopathy (O’Donnell and Efron, 2012; Creuzot-Garcher et al, 2005; Lv et al, 2014).

Tear Film Function There is evidence that tear film parameters are altered in patients who have diabetes, with a reduction in tear breakup time and tear secretion reported (Goebbels, 2000). Factors thought to contribute to reduced tear production in diabetes include microvascular damage to lacrimal glands from hyperglycemia, reduced lacrimal innervation from autonomic neuropathy, a reduction in trophic support to lacrimal tissue, and reduced reflex tearing due to impairment of corneal sensitivity (Lv et al, 2014).

Reduction in tear film stability and tear breakup time are likely due to reduction in mucin production by goblet cells, which are important in stabilizing the tear film. Goblet cell density is thought to be dependent on corneal innervation, with reductions in corneal innervation reported to result in reduced goblet cell function (Dogru et al, 2001).

Ocular Surface Morphology In a healthy eye, superficial damage to the ocular surface is rapidly healed within a continuous trophic environment contributed by the lacrimal secretions. In DED, the ocular surface does not receive as much assistance in this respect, increasing the risk of corneal erosions and infection (Müller et al, 2003). When this is compounded with diabetic corneal neuropathy reducing the trophic factors normally secreted by the lacrimal gland, the ocular surface is placed at greater risk.

In studies comparing diabetic corneas to non-diabetic controls, diabetic corneas had a multilaminar, irregular, and thickened epithelial basement (Schultz et al, 1981; Azar et al, 1989). This may explain the defective epithelial adhesion to the underlying stroma. The combination of these conditions places these patients at increased risk of persistent epithelial defects (Azar et al, 1989), recurrent erosions (Dursun et al, 2001; Akpek et al, 1997; Schultz et al, 1981), poor wound healing, neurotrophic keratopathy, and a higher risk of microbial keratitis (Alves Mde et al, 2008).

Clinical Management

These findings highlight the importance of several clinical considerations. First, question diabetic patients about control of their glycated hemoglobin concentrations (HbA1c) and their time of diagnosis, because poor control and longer duration have been associated with more complications (Kilpatrick et al, 2008). This indicates for clinicians that we need to be more observant for complications, as improved glycemic control has been shown to reduce the incidence of complications such as retinopathy (Stratton et al, 2001). Advise patients to seek advice on managing their control.

Second, question diabetic patients about dry eye-related symptoms and the presence of any other diabetes-related complications. They should also be questioned about other clinicians whom they are seeing—for example, endocrinologists and podiatrists—as a way of gauging the extent of their complications and the level of management.

Third, as ECPs, we need to monitor patients for signs of DED and implement management sooner rather than later, with intensive management of corneal defects to prevent persistent epithelial defects and more severe sequelae. In addition, these patients need to be monitored for neurotrophic ulcers and retinopathy. In the future, we may also play a role in monitoring corneal neuropathy as a way of detecting peripheral neuropathy.

Finally, patients who have diabetes should be educated about the risk of complications and the possible signs and symptoms related to these. Regular eye examinations can play a critical role in detection and prevention of complications, and we are perfectly placed to manage these. CLS

For references, please visit www.clspectrum.com/references and click on document #248.

Dr. Markoulli is a lecturer at the School of Optometry & Vision Science at the University of New South Wales in Sydney.