The aim of the current study was to use a nationwide population-based database to evaluate the prevalence of DES in patients with PD and to investigate the associated risk factors for the period 1997 to 2011. We found the prevalence of DES in adult patients with PD was 4.84% (n= 3665/75650).
Among the patients with PD in our study, female gender, older age, and a number of medical comorbidities were associated with higher risk of DES after adjustment, and the result is consistent with prior studies (*6,7). Furthermore, patients with DES had significantly greater risks of schizophrenia (OR = 1.34), bipolar disease (OR = 1.90), depression (OR = 1.54), sleeping disturbance/ insomnia (OR =1.19), and neurotic disorders (OR = 1.24), including anxiety (OR = 1.34), than those without DES. However, prevalence of DES was significantly lower in patients with dementia (OR = 0.73) and mental retardation (OR = 0.25).
Several previous studies demonstrated greater prevalence of DES in people with depression (*3,4,5,6,7). Furthermore, the severity of DES had a greater impact on the depressive symptoms compared with other psychosomatic symptoms (*8). Several mechanisms may explain the association between depression and DES (*9,10,11). First, we believe that inflammation may play a key role in the pathogenesis of both depression and DES. In patients with DES, increased production of inflammatory cytokines was found in the tears and conjunctiva, including TNF-a, IFN-c, IL-1b, IL-2, IL-6, and IL-8.47–51 (*12,13,14). Depressive patients were also reported to have higher levels of inflammatory cytokines and neuropeptides in the blood (*13,14,22). These cytokines and neuropeptides may simultaneously lead to ocular surface inflammation and exacerbation of negative moods. Second, many previous studies reported depressive patients have a lower threshold of pain perception and often complaint worse dry eye symptom compare with patients without depression. “Neuropathic pain” caused by neural dysfunction plays a role in the unreasonable chronic pain in patients with DES and depression (*15). Third, selective serotonin reuptake inhibitors (SSRIs) are a class of antidepressant that have been reported to be significantly associated with DES. It is possible that SSRIs cause an anticholinergic side effect whereby altered serotonin levels affect the sensitivity thresholds of corneal nerves (*16). Further study on the correlation between SSRIs and DES is necessary.
Our study also found that anxiety disorder was correlated with DES, which supports the findings of two previously reported case-control studies (*4,18). Furthermore, Wen et al. found three significant independent predictors of DES in patients with anxiety, including older age, duration of PD, and the use of an SSRI (*4).
There was a significantly increased risk of DES in patients with bipolar disorder (OR = 1.55). Dibajnia et al. showed significant decreased tear break up time in patients with bipolar disorder treated with either lithium carbonate or sodium valproate (*5). However, the mechanism of this pharmacologic effect is not clear.
Older age has been shown to be associated with increased risk of both DES and dementia (*1). Previous studies demonstrated that patients with dementia have a greater risk of developing dry eye and Sjögren’s syndrome (*18,19). However, we found a decreased risk of DES in dementia patients in our multivariate models. This is the first large-scale study to find this association. We believe the risk might have been underestimated because dementia patients may be less likely to report symptoms due to the decline in cognitive function and communication skills.
A major strength of this study was the use of a large cohort study to investigate the association between DES and PD using data from the National Health Insurance Research Database (NHIRD). This database covers approximately 99% of the country’s residents. However, there were several limitations in this study. First, it was a retrospective study and detailed information are not available in the NHIRD, such as socioeconomic status, severity of disease, and laboratory results. Second, we used strict exclusion criteria and definitions for the diagnosis of DES. Only patients with both ICD-9-CM diagnosis code 375.15 and an anatomical therapeutic chemical code for ophthalmological lubricants (ATC code: S01XA) were included in the analysis. This selection bias might have resulted in an underestimation of the prevalence of DES in psychiatric patients, because patients with only mild dry eye symptoms may not need to use any medication. In addition, some patients may seek other medications that are not available on the NHI system.