Abstract Introduction Association between both short [1] and long-term exposure to air pollution and cardiovascular (CV) morbidity and mortality is known. Lombardy region, Italy is a hot spot for air pollution spikes in particulate matter (PM) and NO2, with over 60% and 45%, respectively, of days/year exceeding the WHO daily permissible limits. This phenomenon is largely due to the regional land use and its geographical conformation, which hinders pollution dispersion. Our aim was to study the short-term relation between both trends and spatial distribution of air pollution and geo-localized CV events reported to the regional emergency services (AREU) in Lombardy region. Method First, time-series of PM2.5, PM10, NO2 and O3 concentration were obtained using daily average of 10 km x 10 km resolution satellite data from Copernicus Atmosphere Monitoring Service (CAMS) for 2015-2020 period, and superimposed to time-series of CV related emergency calls (CVCALLS) in the same period (total calls 1,094,848). For every year, daily peaks both for pollution concentration and CVCALLS were identified as the top 1st percentile or as exceeding three times the standard deviation. Peaks related to SARS-CoV-2 pandemic in 2020 were excluded. Spatial autocorrelation analyses, using both global (G) and local (L) Moran’s I, were conducted to evaluate the monthly clustering patterns of both pollutants and CVCALLS in uniform districts of 100,000 residents. Results Seasonal peaks were observed for PM2.5, PM10 and NO2 in winter, and for O3 in summer. Identified peak days for NO2 and O3 were coinciding with those from CVCALLS, considering temporal lags up to seven days (see Fig 1-2). Global autocorrelation resulted in a high Moran’s I for concentration levels of all the pollutants (G > 0.6), while comparatively lower clustering emerged for CVCALLS (G <0.65). The monthly L Moran’s I showed high value clustering for PM2.5, PM10 and NO2 in the southern, largely urban, part of the region, and low value clusters in the Alpine northern and north-eastern area. The exact opposite pattern was observed for O3. The spatial clustering of the CVCALLS, coherently with the low global autocorrelation, showed less significant patterns, with a small high value cluster in and around the city of Milan and its southern districts throughout the whole year. Conclusion As Spearman’s regression analysis between pollutants’ concentration and CVCALLS resulted in low correlation coefficient (R2 ≤ 0.16), the use of advanced statistical models, accounting for temporal lags and confounders, is suggested to better understand the underlying phenomena. In conclusion, our study found a strong spatial clustering pattern in all the pollutants throughout the year, but this was not observed for the CV events. However, strong correspondences between peaks of NO2 during winter and O3 during summer and CVCALLS were clearly observed, thus confirming short term effect of such pollutants on emergency CV events.