Dry eye disease is a condition that imposes significant physical and mental burdens on many individuals^1,2,3,4. It leads to direct medical costs and reduced productivity, and diminishes quality of life due to eye pain, irritation, and impaired visual function^3,4. Patients with mild to severe dry eye experience a decrease in quality of life comparable to those with mild to moderate psoriasis or severe angina². Therefore, research to elucidate the causes of dry eye is considered important.

The surface of the eye is constantly exposed to the external environment and is affected by atmospheric conditions and air pollution⁵. Humidity, in particular, has been reported to significantly impact dry eye. The biological mechanisms underlying seasonal variation in dry eye disease prevalence and severity are multifaceted, involving environmental factors, immune responses, and physiological changes. Seasonal fluctuations in humidity and temperature significantly influence ocular surface conditions^6,7,8. Experimental studies have shown that lower relative humidity accelerates tear evaporation, shortens tear film breakup time, and increases ocular surface staining^9,10,11. According to the World Health Organization, major components of air pollution include particulate matter (PM), ozone, nitrogen dioxide(NOâ‚‚), and sulfur dioxide(SOâ‚‚)¹². While air pollution is known to be associated with respiratory diseases such as allergic conjunctivitis, asthma, and allergic rhinitis, various reports exist regarding its association with dry eye^1,3,5,11,13.

The first-line treatment for dry eye is eye drops, but if these are insufficient, surgical treatments such as punctal plugs are considered^1,14,15,16. Punctal plugs are a non-pharmacological treatment used when artificial tears alone do not improve symptoms. Silicone or collagen plugs are inserted into one or both puncta^14,15,16. The insertion of punctal plugs is a relatively simple outpatient procedure performed promptly when dry eye worsens. There is thought to be little time lag between the timing of operation and the worsening of dry eye. Therefore, we examined the relationship between changes in dry eye conditions and seasons using the number of dry eye operations.

The study in Norway have reported that dry eye symptoms and signs are most prevalent from winter to spring and least prevalent in summer¹⁷. In the United States, the prevalence of dry eye among veterans is highest in winter/spring and lowest in summer¹⁸. In Bologna, Italy, eye discomfort peaks from November to February and June to August¹⁹. An international survey conducted in 2013 across five European countries reported seasonal characteristics of dry eye²⁰. Conversely, a multicenter study in Japan found no seasonal variation in the prevalence of dry eye²¹. There have been many reports of seasonal changes in dry eye, but no such changes have been observed in Japan, where atmospheric changes with the seasons are large. The aim of this study is to determine whether there are seasonal differences in the number of dry eye operations in Japan and to investigate whether meteorological conditions and air pollutants are related to the number of dry eye operations. The number of dry eye operations was examined monthly using the National Database of Health Insurance Claims and Specific Health Checkups of Japan (NDB) database. Additionally, data from the Japan Meteorological Agency was used to investigate correlations between meteorological conditions (temperature, atmospheric pressure, relative humidity, wind speed) and air pollutants SOâ‚‚, nitrogen oxides X (NOX), photochemical oxidants (OX), CO and PM2.5. Based on past reports^22,23,24,25, not only relative humidity but also volume humidity was examined. This is expected to deepen the understanding of seasonal variations in dry eye and their causes.

Line graph showing the number of dry eye operations by month from fiscal years 2019 to 2021. The maximum number of dry eye operations were observed in December in all years. April and May 2020 were considered to be low due to the declaration of a state of emergency for COVID-19 pandemic. (Fig.  1).

Line graph of the number of dry eye operations by month from fiscal years 2019 to 2021. The maximum number of dry eye operations were observed in December in all years. April and May 2020 were considered to be low due to the declaration of a state of emergency for COVID-19 pandemic.

Full size image

Figure 2 shows number of dry eye operations box-and-whisker diagram by season from fiscal years 2019 to 2021. The Steel–Dwass test showed a predominant difference in the number of dry eye operations in summer and winter. (p = 0.0023). No significant differences were found among the other seasons.

Dry eye operations’ box-and-whisker diagram by season from fiscal years 2019 to 2021. The number of dry eye operations was significantly higher in winter than in summer. (p = 0.0023). No significant differences were found among the other seasons.

Full size image

Figure 3 shows line graph of monthly air quality index data from 2019 to 2021.

Line graph of monthly air quality index data from 2019 to 2021. Monthly meteorological data was collected from the Japan Meteorological Agency database, monthly data on air pollution was collected from the database of the Ministry of the Environment. Data were averages for Tokyo, Osaka, and Nagoya. sulfur dioxide (SOâ‚‚), nitrogen oxides X (NOX), photochemical oxidants (OX), carbon monoxide (CO) and particulate matter 2.5 (PM2.5).

Full size image

Correlations between climatic conditions and multiple air pollutants with the number of dry eye operations are presented in Fig. 4. While volume humidity and temperature were strong negatively correlated (volume humidity, R = −0.7590; 95%CI,−0.8731 to −0.5662, P < 0.0001; temperature, R = −0.7552; 95% CI,−0.6229 to −0.1442, P < 0.0001), NOX and atmospheric pressure were strong positive correlated. (NOX, R = 0.8105; 95%CI, 0.6506 to 0.9016, P < 0.0001; atmospheric pressure, R = 0.7615; 95% CI, 0.5702 to 0.8745, P < 0.0001).

Correlations between climatic conditions and multiple air pollutants with the number of dry eye operations. Volume humidity and temperature were strong negatively correlated (volume humidity, R = – 0.7590; 95%CI, – 0.8731 to – 0.5662, P < 0.0001; temperature, R = −0.7552; 95% CI, −0.6229 to −0.1442, P < 0.0001), NOX and atmospheric pressure were strong positive correlated. (NOX, R = 0.8105; 95%CI, 0.6506 to 0.9016, P < 0.0001; atmospheric pressure, R = 0.7615; 95% CI, 0.5702 to 0.8745, P < 0.0001). sulfur dioxide (SO₂), nitrogen oxides X (NOX), photochemical oxidants (OX), carbon monoxide (CO) and particulate matter 2.5 (PM2.5).

Full size image

This study conducted from fiscal years 2019 to 2021 in Japan showed a trend of fewer dry eye operations in August and more in December. When comparing seasons, the number of dry eye operations was significantly higher in winter than in summer. No significant differences were found among the other seasons. Previous multicenter studies in Japan did not find seasonal variations in the prevalence of dry eye²¹, but our study observed seasonal fluctuations in the number of operations. There have been conflicting reports in the past, such as Kumar’s report on U.S. veterans¹⁸ which noted seasonal changes in dry eye symptoms, while the report in Norway¹⁷ found no such variations. These inconsistencies may be due to regional climate differences and the lack of standardized methods for evaluating dry eye^{3,4,17,18,20,21}.

There was a strong negative correlation between relative humidity, volume humidity, and temperature, while atmospheric pressure showed a positive correlation. Many reports have discussed the relationship between humidity, temperature, and dry eye. Low humidity is a well-known risk factor for dry eye disease^1,3,4,9,26. Both human and animal studies have reported that exposure to dry environments can lead to a significant decrease in tear production, evaporation rates, tear film lipid layer thickness, corneal epithelial integrity and reduced fluorescein staining, goblet cell density^5,10,27,28. In addition to relative humidity, volume humidity was also examined, and a strong correlation with the number of operations was found in our study. Since relative humidity is influenced by temperature, research using volume humidity, which represents the amount of moisture in the air, has been reported for other diseases such as influenza^24,25, but not for dry eye. Additionally, there are no reports on the relationship between atmospheric pressure and dry eye, which needs further investigation.

In our study, strong positive correlation with NOX was observed. The association between NOâ‚‚ and dry eye has been reported in the past^5,29,30,31. Traffic-related air pollution is becoming increasingly common in urban areas. Such pollution is generally assessed by particulate matter and NOâ‚‚ levels³². NOâ‚‚ is produced when oxygen or ozone in the air oxidizes nitric oxide, with the main sources of NOâ‚‚ in outdoor air being primarily automobiles, followed by fuel combustion from power plants and factories. Traffic-derived particulate matter includes components such as engine exhaust, brake and tire wear, and dust from road surfaces^3,33. In Japan, nitrogen oxides in the air are indicated by the total of nitric oxide and nitrogen dioxide, referred to as NOX, which was used in our study.

A study of 79,866 participants living in northern Netherlands found a strong positive correlation between dry eye disease and air pollutants, this study suggests that the association between air pollutants and dry eye disease may be largely due to the high prevalence of other diseases directly related to air pollution, such as allergies, atopic diseases, arteriosclerosis, and diabetes³⁴. A report from South Korea found that among 16,824 participants, increased ozone concentration and low humidity were significantly associated with dry eye symptoms and diagnosis, while sulfur dioxide and PM10 were not associated⁵. These conflicting reports highlight the significant limitation in human studies of the difficulty in measuring and standardizing exposure to environmental pollutants. While animal experiments may establish more controlled environments, exposure levels are often extreme and do not match actual human exposure. The additive or synergistic deterioration of the ocular surface due to combinations of pollutants also needs to be evaluated. Advanced techniques are needed for measuring environmental factors and identifying sensitive individuals who require early treatment for dry eye disease³.

Our study has several limitations. This study compared only the number of operations, which may not accurately represent the timing of worsening dry eye symptoms. Additionally, meteorological data were averaged from three regions with high operation numbers, which may be affected by localized abnormal weather. Although the data covers three years, it includes the COVID-19 pandemic period. Furthermore, psychological factors related to seasonal affective disorder may also influence patients’ decisions regarding surgery, but this was not examined in the current study. Finally, this study shows correlation, but does not prove causality.

In this study, we investigated the seasonal variation in dry eye operations and whether meteorological conditions and air pollutants are related to the number of dry eye operations. The number of dry eye operations was significantly higher in winter than in summer. While volume humidity and temperature were strong negatively correlated, NOX and atmospheric pressure were strong positive correlated. Understanding the seasonal variation in operations is useful from a medical planning perspective, as it allows for the preparation of a system to smoothly conduct operations, secure personnel, strengthen the supply system for medical resources, and prepare for inventory management.

This retrospective and descriptive study used NDB open data published and managed by the Japanese Ministry of Health, Labor and Welfare³⁵. All the investigations adhered to the principles of the Declaration of Helsinki. Ethical review was exempt because we used a public database for analysis, and the data contained no identifiable personal information.

The number of operations was collected from the NDB database from fiscal years 2019 to 2021. The fiscal month is the same as the actual month. Dry eye operation in Japan is number K200-2, applies to lacrimal plug insertion and lacrimal closure. We used accumulated NDB data from fiscal years 2019 to 2021, because monthly data started in 2019.

Monthly meteorological data was collected from the Japan Meteorological Agency database³⁶. Temperature, atmospheric pressure, relative humidity and wind speed were considered. Due to the large proportion of dry eye operations in the Kanto, Chubu and Kansai regions³⁵, data from Tokyo, Nagoya and Osaka were used as an average.

Volume humidity is calculated from the collected temperature and relative humidity data, and the calculation formula is as follows²³:

Monthly data on air pollution was collected from the database of the Ministry of the Environment^36,37. Air pollutants were studied in SOâ‚‚, NOX, OX, CO and PM2.5. The seasons were divided into spring (March, April, May), summer (June, July, August), fall (September, October, November), and winter (December, January, February), while acknowledging that weather conditions in Japan can be different even during the same season depending on the location of the participating institutions.

The Steel–Dwass test, a nonparametric statistical test, was used to compare each season. Peason’s correlation coefficient was used for meteorological data and air pollution data with the number of dry eye operations. Statistical significance was set at P < 0.05. Statistical analysis was performed using JMP®16 (SAS Institute Inc., Cary, NC).

In Japan, the first COVID-19 case was identified in January 2020. With the spread of the infection, the Japanese government declared a state of emergency for COVID-19 in April 2020. During the COVID-19 pandemic, non-emergency surgery was postponed; thus, the COVID-19 pandemic may have affected the trends in ophthalmic surgery as well. Therefore, April and May 2020 were excluded from the statistical study.