The role of air temperature in Legionella water contamination and legionellosis incidence rates in southern Italy (2018-2023).

Abstract

Background: Legionnaires' disease is caused by inhalation or aspiration of small water droplets contaminated with Legionella, commonly found in natural and man-made water systems and in moist soil. Over the past 5 years, notification rates of this disease have almost doubled in the European Union (EU) / European Environmental Agency (EEA), from 1.4 in 2015 to 2.2 cases per 100,000 population in 2019. Some studies show that the greater presence of the microorganism in the water network and the increase in cases of legionellosis could be related to the variations in some environmental factors, such as air temperature, which may influence the water temperature.

Study design: Climate change is currently a prominent topic worldwide because of its significant impact on the natural environment. It is responsible for the increase in numerous waterborne pathologies. The purpose of this study was to correlate the air temperature recorded in Apulia region from January 2018 to April 2023 with the presence of Legionella in the water networks of public and private facilities and the incidence rates of legionellosis during the same period.

Methods: During the period from January 2018 to April 2023, water samples were collected from facilities involved in legionellosis cases and analyzed for Legionella. During the same period, all the cases notified to the regional epidemiological observatory (OER-Apulia) were included in this study. Statistical analyses were conducted using the Shapiro-Wilk test to determine whether the Legionella load was distributed normally, the Wilcoxon rank sum test to compare the air temperatures (average and range) of the negative and positive samples for Legionella detection, and the multivariate analysis (Poisson regression) to compare the Legionella load with the water sample temperature, average air temperature, and temperature range on the day of sampling. The Wilcoxon test for paired samples was used to compare legionellosis cases between the warmer and colder months.

Results: Overall, 13,044 water samples were analyzed for Legionella and 460 cases of legionellosis were notified. Legionella was isolated in 20.1% of the samples examined. The difference in the air temperature between negative samples and positive samples was statistically significant (p-value < 0.0001): on days when water samples tested positive for Legionella a higher temperature range was observed than on days when water samples tested negative (p-value = 0.004). Poisson regression showed a direct correlation between Legionella load, water temperature, and average air temperature. The incidence of legionellosis cases in warmer months was higher than in colder months (p-value = 0.03).

Conclusions: Our study highlights a significant increase in the load of Legionella in the Apulian water network, and an association between warmer temperatures and legionellosis incidence. In our opinion, further investigations are needed in different contexts and territories to characterize the epidemiology of legionellosis, and to explain its extreme variability in different geographical areas and how these data may be influenced by different risk factors.