To assess the relationship between patients’ age and prognosis outcome after cardiopulmonary resuscitation (CPR) in adults.

Relevant observational studies were identified by a search of PubMed and ISI databases from their beginning to 30 January 2014. Survival to discharge was used as the prognosis outcome. The weighted mean difference (WMD) for patients’ age were calculated for survival to discharge and death in hospital. We also carried out a dose-response meta-analysis for assessing summary odds ratio (OR) of survival by patients’ age.

A total of 27 studies included in the meta-analysis. The WMD of age between those who survived or not was 5.294 (95% CI 3.150 to 7.438; Z=4.84, P=0.000). No publication bias was found (Begg test P=0.228 and Egger test P=0.101). Sensitivity analyses by only included prospective cohorts showed the WMD of age between those who survived or not was 4.236 (95% CI 1.652 to 6.821; Z=3.21, P=0.001). Association between patients’ age and survival to discharge risk after CPR was observed statistically significant (χ²=23.54, P=0.000). In linear model, the summary OR was 0.976 (95% CI 0.966–0.986) for every 1-year increase in patients’ age. In spline model, the OR decreased along with the patients’ age, especially when patients’ age over 70years.

Old age is an important pre-arrest predictor for poor prognosis after CPR in adults. When patients’ age over 70years, the survival to discharge risk decreased rapidly along with the patients’ age increase.