Two studies presented this week at ARVO tackled the complex relationship between diabetes and dry eye, adding to the growing body of evidence supporting the need for dry eye screening in patients with diabetes.

In the first study, researchers looked at 81 patients with diabetes, dry eye disease (DED) or both (DMDED), using the Ocular Surface Disease Index (OSDI), Pain Sensitivity Questionnaire, lissamine green staining, meibography, meibomian gland (MG) expression, noninvasive break-up time (NIBUT) and hemoglobin A1c (HbA1c). When comparing the data, they discovered 19 of 37 diabetes patients had dry eye, an undiagnosed rate of 51.3%.1

In addition, those with DMDED reported significantly less ocular redness compared with the DED cohort, which “may be related to vascular changes seen in diabetes and may merit further study,” the researchers said. DMDED patients also had significantly worse conjunctival staining and reported more itching than DED patients. While the researchers found no difference between groups for pain sensitivity, MG structure or NIBUT, they did find higher HbA1c was moderately associated with OSDI in DMDED patients.

“Dry eye is often underdiagnosed in this population and may have different signs and symptoms due to vascular changes when compared to patients without diabetes,” the researchers concluded.

According to Joseph P. Shovlin of Scranton, PA, these results highlight the need for use of a dry eye screening questionnaire. “These are important for all patients, but this shows it would be especially prudent to use them for diabetes patients,” he says.

Another group of researchers discovered meibomian gland dysfunction (MGD) is “a major cause of dry eye syndrome in type 2 diabetics without obvious aqueous tear deficiency.” This investigation included 198 eyes of 99 diabetes patients and 160 eyes of 80 DED patients without diabetes, and collected OSDI, tear film break-up time (TBUT), Schirmer I test, corneal fluorescein staining (FL), lipid layer thickness (LLT), MG parameters, HbA1c levels and duration of diabetes.2

They found tear volume in diabetes patients was close to the normal cut-off value, but higher than that of the DED group; these patients also had lower TBUT and LLT, but higher MG parameters. The researcher were particularly interested to discover differences between diabetes patients with and without DED: in the diabetes-without-DED group, tear volume, BUT and LLT were significantly higher, and all MG parameters and FL scores were significantly lower than the diabetes-with-DED group.2

The four MG parameters were significantly higher in the diabetes-without-DED group than the controls, even though OSDIs were below 13 in both. “This suggested the MGD emerged before ocular discomfort developed,” the researchers realized.

“The tear function and MG parameters were correlated with HbA1c level and diabetic duration. MGD, which appears before the development of ocular discomfort, may be an early sign and the critical factor of dry eye in type 2 diabetics,” the researchers concluded.2

Overall, these studies represent additional identifiable morbidities of diabetes, Dr. Shovlin says. “Many exist, and these seem to have a direct correlation to patients with symptoms of dry eye.”

1. Schwartz S, Halleran C, Doll T, et al. Does diabetes make a difference in dry eye? ARVO 2018. Abstract 956-B0134.
2. Wu H, Xie F, Luo S, et al. Meibomian gland dysfunction is an early sign and major cause of dry eye in Type 2 diabetes. ARVO 2018. Abstract 4900-C0345.