Several studies have estimated the associations between extreme temperatures and mortality and morbidity; however, few have investigated the attributable fraction for a wide range of temperatures on the risk of out-of-hospital cardiac arrest (OHCA). We obtained daily records of OHCA cases in the 47 Japanese prefectures between 2005 and 2014. We examined the relationship between OHCA and temperature for each prefecture using a Poisson regression model combined with a distributed lag non-linear model. The estimated prefecture-specific associations were pooled at the nationwide level using a multivariate random-effect meta-analysis. A total of 659,752 cases of OHCA of presumed-cardiac origin met the inclusion criteria. Overall, 23.93% (95% empirical confidence interval [eCI]: 20.15-26.19) of OHCA was attributable to temperature. The attributable fraction to low temperatures was 23.64% (95% eCI: 19.76-25.87), whereas that of high temperatures was 0.29% (95% eCI: 0.21-0.35). The attributable fraction for OHCA was related to moderate low temperature with an overall estimate of 21.86% (95% eCI: 18.10-24.21). Extreme temperatures were responsible for a small fraction. The majority of temperature-related OHCAs were attributable to lower temperatures. The attributable risk of extremely low and high temperatures was markedly lower than that of moderate temperatures.