The aim of this study was to determine the temperature threshold that triggers an increase in heat-induced mortality in Zaragoza, Spain to determine the impact of extreme heat on mortality and in-hospital cost. A longitudinal ecological study was conducted according to an autoregressive integrated moving average model of a time series for daily deaths and to determine the relative risk of mortality for each degree that the temperature threshold was exceeded. Mortality showed a statistically significant increase when the daily maximum temperature exceeded 38 degrees C. A Relative Risk was 1.28 with a 95 % confidence interval (95 %CI:1.08-1.57) This threshold temperature didn't change over time. A total of 107 (95 %CI:42-173) heat-attributable deaths were estimated for the period 2002-2006, and the in-hospital estimated cost of these deaths reach euro 426,087(95 %CI.euro167,249-euro688,907). The articulation of preventive measures to minimize the impact of extreme heat on human health is necessary because of the mortality-temperature relationship.