Background
Continuous efforts by researchers have produced substantial progress in the field of dry eye disease (DED). In 1995, the National Eye Institute/Industry Dry Eye Workshop defined dry eye (DE) as a disorder of the tear film due to tear deficiency or excessive evaporation that damages the interpalpebral ocular surface and is associated with symptoms and discomfort [
1]. The International Dry Eye Workshop refined this definition in 2007, adding tear hyperosmolarity as a newly discovered cause of DE and addressing the instability of the tear film as well as the effect of DE on visual function [
2]. National consensus definitions of DE for Japan were reported in 1995 [
3] and 2007 [
4].
Recently, a new definition of DE by the Asia Dry Eye Society has highlighted the important role of tear instability, based on evidence obtained from epidemiologic studies [
5]. It has been reported that the majority of patients with DE have abnormal tear film break-up time (TBUT); in contrast, abnormal Schirmer tests and keratoconjunctival staining are less prevalent in this group [
6,
7]. In addition to these objective DE tests, validated questionnaires including the Ocular Surface Disease Index [
8], Impact of Dry Eye on Everyday Life [
9], and Dry Eye-Related Quality-of-Life Score (DEQS) [
10] questionnaires, have been developed for subjective evaluation of DE symptoms. The most common ocular symptoms of DED are discomfort and visual disturbance [
2], including dryness, grittiness, ocular fatigue, redness, foreign body sensation, and soreness [
6,
11]. These symptoms are known to reduce patients’ perception of their quality of life (QOL) [
12,
13] and utility values [
14,
15], which have identified a number of DE subtypes, including aqueous-deficient dry eye (ADDE) and short tear film break-up time dry eye (short TBUT-DE). Short TBUT-DE is characterized by severe symptoms with minimal ocular surface damage except for tear film instability [
5,
7,
16]. The severity of the symptoms of short TBUT-DE is reportedly comparable with that of ADDE [
17,
18]. In 2017, the epidemiological results of Dry Eye Cross-Sectional Study (DECS-J) revealed that the two most common DED subtypes were ADDE and short TBUT-DE [
7]. These findings led us to speculate about whether the disease burden of DE subtypes (ADDE and short TBUT-DE) had a similar effect on QOL.
The importance of assessing the impact of disease on QOL from a patient-centric perspective has been established. Quantitative evaluation of patient-reported outcomes requires the use of measurement modalities with confirmed reliability and validity. The DEQS evaluates the severity of DE-associated symptoms and the multifaceted effect of DED on a patient’s daily life and has been evaluated for internal consistency, reproducibility, validity, and responsiveness when used in the Japanese population [
10]. The Health Utilities Index Mark 3 (HUI-3) quantifies patient preferences for various health states and calculates associated health utility values [
19,
20]. The HUI-3 is a standard method for assessing health utility. Schiffman et al. investigated health utility values using the time trade-off (TTO) method [
14] and demonstrated that DE has a substantial impact on patients’ lives. However, there are no reports on the use of the HUI-3 to evaluate health utility values in patients with DED.
This study was performed to determine whether QOL and health utility are affected to the same extent in patients with short TBUT-DE with minimal clinical signs and those with ADDE. We compared the association of QOL and health utility among patients with ADDE and those with short TBUT-DE, using the DEQS and HUI-3. We also explored the associations of QOL and health utility with various ophthalmic parameters.
Discussion
In this study, we evaluated the association of QOL with health utility values between the two major subtypes of DE, ADDE and short TBUT-DE. Short TBUT-DE, which is characterized by severe symptoms with minimal ocular surface damage except for tear film instability, is a new subtype of DE [
11] that has gradually become more accepted. In the present study, we have shown that QOL and health utility were similarly severe in short TBUT-DE and ADDE, which features ocular surface damage accompanied by reduced tear volume. To the best of our knowledge, this is the first report to quantitatively evaluate the impact of DED on QOL and health utility using established questionaires, and to compare the results between patients with ADDE and those with short TBUT-DE.
In the present study, the median DEQS in patients with short TBUT-DE 23.3 (13.3–38.3) was comparable with that in patients with ADDE 23.3 (10.0–40.0). A significant correlation of DEQS scores with the results of the Short Form-8 Health Survey and the 25-item National Eye Institute Visual Functional Questionnaire has been reported [
10]. Furthermore, DEQS scores have been reported to be significantly higher in patients with DED than in controls and to have improved significantly after treatment [
10]. The authors of the presented research have previously found that of all the subtypes of DE, Sjögren syndrome was associated with the most severe score and MGD with the mildest score (38.8 and 18.1, respectively) [
7]. The severity of the subjective symptoms of short TBUT-DE has been assessed using different QOL questionnaires. In one analysis of symptom scores in DE patients who visited the subspecialty outpatient clinic of a university hospital, Shimazaki-Den et al. found that the sum of the mean values of five symptom scores did not differ significantly between the ADDE and short TBUT-DE subgroups [
17]. Another report by Yokoi et al. [
18] used two questionnaires to assess subjective symptoms in VDT workers with either definite DE or short TBUT-DE [
8,
25,
26] and concluded that the DE symptoms in short TBUT-DE were comparable with those of definite DE. Using a validated questionnaire specifically designed for DE, our results have confirmed the previous findings.
It has been reported that the method used to collect utility data affects the utility value [
27]. However, in this study, the median HUI-3 value across all patients was 0.82 (0.69–0.91), which is comparable with the results obtained by Schiffman et al. (0.78 for moderate DE) [
14] and Buchholz et al. (0.68 for moderate DE) [
15] using the TTO method. We used the scores on the HUI-3 for evaluation of utility values because this instrument is validated in Japanese and is administered in questionnaire format, rather than requiring an interview like the TTO and standard gamble (SG) methods. The TTO method directly assesses how long a period of perfect health is equivalent to the given period with a current disease status [
28] and the SG method is based on a paired comparison in which the subject chooses the strategy of perfect health with probability of death or an intermediate health state [
29]. Interestingly, the median health utility values in patients with short TBUT-DE 0.82 (0.74–0.92) and those with ADDE 0.79 (0.69–0.88) were comparable.
The degree of correlation between symptoms and ophthalmic examination findings in these conditions remains uncertain. Our results demonstrate a statistically significant but weak correlation of TBUT, keratoconjunctival staining, and Schirmer test results with DEQS and HUI-3 scores, which is in agreement with several previous reports [
6,
13,
30,
31]. However, Nichols et al. found no association between clinical tests for DE and the five major symptoms of DE [
6], while Sullivan et al. concluded that there was no consistent relationship between clinical tests and the Ocular Surface Disease Index score [
31]. Similarly, Mizuno et al. reported a discrepancy between ocular surface findings and QOL scores assessed using the Visual Functioning Questionnaire and the 8-item Short-Form Health Survey [
13]. Bron et al. [
30] reported that tear osmolarity was the best index for assessing subjective symptoms in patients with DE; in our study, although measurement of tear osmolarity would have been interesting, it would have been prohibitively difficult to provide the required device to all participating clinics. There could be several reasons for the discrepancy between signs and symptoms, including a need for new clinical tests for DED, failure to recognize the signs of the disease, and the relationship with hypersensitivity. Studies are underway to explore these possibilities further.
TBUT was the only index that showed a statistically significant correlation with both DEQS and HUI-3 in the present study. This finding may reflect the fact that a short TBUT is apparent in all types of DE and it is an essential index for diagnosing DED in clinical practice [
5]. However, the extremely low correlation coefficients for these relationships suggest that further interpretation of the results would be difficult; given that the symptoms of DE are influenced by various factors, including psychological state, the impact of DE cannot be explained by “unstable tear film” alone [
32].
Advances in research in the field of DE have revealed that the prevalence of short TBUT-DE is substantial, especially among VDT workers [
16], and that the symptoms of this condition are similar to those of ADDE [
17,
18]. Present study also seem to support the prevalence of short TBUT-DE were substantial, among all patients included in the study. Although the cause of short TBUT-DE is still unclear, tear film instability due to decreased wettability of the corneal and conjunctival epithelia has been proposed; indeed, decreased expression of MUC5AC and MUC16 mRNA has been reported in both ADDE and short TBUT-DE [
17]. The authors of the presented research have previously reported a statistically significant correlation between TBUT and wheat germ agglutinin fluorescence intensity (a marker of ocular surface mucins) [
33]. A decrease in mucins may be induced by underlying or subclinical inflammation [
34‐
36] or by hyperosmolarity [
37], which has been shown to activate nociceptors such as transient receptor potential melastatin subfamily member 8 (TRPM8), a cold thermoreceptor [
38]. TRPM8 is known to regulate ocular surface wettability and also triggers DE symptoms [
39]. In the present study, the severity of DE symptoms in patients with short TBUT-DE seems to support the revised definition of DE proposed by the Asia Dry Eye Society, which the new definition assigns the essential role for TBUT assessment for diagnosing DED [
5].
The present study had several limitations. First, although the 10 clinics included in the study were widely distributed across Japan, the sample may not have been large enough to be representative sample of all Japanese eye clinics and patients. Second, although all the study investigators were DE specialists and followed the same guidelines, there may have been some variations in their evaluations. For example, since the scoring system of vital staining is qualitatively evaluated, it may have influenced the ophthalmic evaluations. Third, patients who had already been treated for DE were included in the study. Three hundred and seventy-seven patients (84%) had already received treatment for DE and 72 (16%) were newly diagnosed with DE and were yet to be treated. DEQS and HUI-3 scores were not significantly different between treated and untreated patients. Finally, the study questionnaires were answered by patients at home, so we cannot exclude the possibility of environmental effects on our findings, such as the order of administration settings. The extent to which mode of administration may have influenced the questionnaire results is unknown; however, it has been reported that mode of administration does not have a major effect on the response to questionnaires regarding health-related quality-of-life measures [
40].
Acknowledgments
The study was conducted jointly by the Dry Eye Society and Santen Pharmaceutical Co., Ltd. Both organizations contributed to the creation of documents (including the study protocol), management of study progress, provision of information and support to study sites, data analysis and study outcome disclosure, and entry/update of the study in the public registration system.
The DECS-J Study Group: The following individuals participated in the study.
Writing Committee:
Chika Shigeyasu, Masakazu Yamada, Motoko Kawashima, Kazuhisa Suwaki, Miki Uchino, Yoshimune Hiratsuka, Norihiko Yokoi, Kazuo Tsubota.
Study group investigators and clinical sites:
Yoshitsugu Tagawa (Kitaichijo Tagawa Eye Clinic, Sapporo, Hokkaido,Japan), Seika Den (Akasaka Tokyu Shimazaki Eye Clinic, Tokyo, Japan), Miki Iwasaki (Ryogoku Eye Clinic, Tokyo, Japan), Hiroshi Saito (Saito Eye Clinic, Saitama, Japan), Reiko Ishida (Ishida Clinic, Shizuoka, Japan), Aoi Komuro (Shijo-karasuma Komuro Eye Clinic, Kyoto, Japan), Naoki Iwasaki (Iwasaki Eye Clinic, Osaka, Japan), Harue Matsumoto (Matsumoto Eye Clinic, Tokushima, Japan), Tomoko Goto (Hanamizuki Eye Clinic, Ehime, Japan), Atsuko Kiyosawa (Kiyosawa Eye Clinic, Fukuoka, Japan).