OBJECTIVE: This time series study aimed to examine the association between daily air temperature and cause-specific cardiovascular mortality in Bavaria, Southern Germany. METHODS: We obtained data from the cities Munich, Nuremberg and Augsburg and two adjacent administrative districts (Augsburg and Aichach-Friedberg), for the period 1990-2006. Data included daily cause-specific cardiovascular death counts, mean daily meteorological variables and air pollution concentrations. In the first stage, data were analysed for Munich, Nuremberg and the Augsburg region separately using Poisson regression models combined with distributed lag non-linear models adjusting for long-term trend, calendar effects and meteorological factors. In a second stage, we combined city-specific exposure-response relationships through a multivariate meta-analysis framework. RESULTS: An increase in the 2-day average temperature from the 90th (20.0°C) to the 99th centiles (24.8°C) resulted in an increase of cardiovascular mortality by 10% (95% CI 5% to 15%) in the pooled analysis, while for a decrease from the 10th (-1.0°C) to the 1st centiles (-7.5°C) in the 15-day average temperature cardiovascular mortality increased by 8% (95% CI 2% to 14%). Strongest consistent risk estimates were seen for high 2-day average temperatures and mortality due to other heart diseases (including arrhythmias and heart failure) and cerebrovascular diseases, especially in the elderly. CONCLUSIONS: Results indicate that, in addition to low temperatures, high temperatures increase cause-specific cardiovascular mortality in temperature climates. These findings may guide planning public health interventions to control and prevent the health effects of exposure to air temperature, especially for individuals at risk for mortality due to heart failure, arrhythmias or cerebrovascular diseases.