Background: The aim of this work was to study the associations between weather conditions and the occurrence of type B acute aortic dissections (ABAD).
Patients and methods: This study was a retrospective review of all ABAD cases between January 1st, 2006 and December 31st, 2015. Using a time-series design and distributed lag non-linear models (DLNM), we estimated the relative risk (RR) of ABAD presentation associated with mean daily temperatures, including cumulative RR for a 28-day period, and RR for individual daily lags through 28 days.
Results: A total of 213 patients were admitted with ABAD. A significant association was found between the daily maximal temperature and the number of hospital admissions for ABAD. The lower the maximal temperature, the higher the incidence of ABAD (P = 0.044). Furthermore, the onset of ABAD was higher in winter than in summer and autumn (P = 0.009 and P = 0.001). Based on a time-series analysis, this study showed that the associations between mean daily temperature and ABAD presentation were not monotonic. Compared to the centered temperature at 8°C, the cumulative 28-day (lag 0 to lag 27) RR was significantly elevated at - 20 °C and - 19 °C for ABAD (RR = 1.39; 95%CI: 1.02, 1.98 and RR = 1.36; 95%CI: 1.02, 1.98). At the extreme low temperature (- 17.7 °C) in Shenyang, the cumulative 14-day (lag 0 to 13 day) and 21-day (lag 0 to 20 day) RR were remarkably increased for ABAD (RRlag14-day = 1.34; 95%CI: 1.08, 1.40 and RRlag21-day = 1.06; 95%CI: 1.06, 1.23). For the extreme high temperature, however, no particular finding was detected regarding acute and prolonged effects on ABAD.
Conclusions: In general, low ambient temperature was significantly associated with ABAD presentations in comparison with high temperature. The effects of cold were delayed by two weeks and persisted for a few days.
Keywords: Aortic dissection; ambient temperature; biometeorology; distributed lag non-linear models.