There are still debates on the association of increased cardiovascular risk with the use of nonsteroidal anti-inflammatory drugs (NSAIDs) in patients with rheumatoid arthritis (RA) because of inconsistent results. Therefore, our study aims to evaluate the transient effects of selective and nonselective NSAIDs on the risk of stroke and acute myocardial infarction (AMI) in patients with RA. We conducted a case-crossover study of 5,921 stroke or AMI patients with co-morbidity of RA. All cases were identified from the Taiwan National Health Insurance Database between January 1, 2006, and December 31, 2011, according to the International Classification of Diseases, 9th Revision and Clinical Modification diagnosis codes from inpatient claims. The index date was defined as the date of hospitalization for stroke or AMI. Exposure to NSAIDs was compared during a case period (1 to 30 days before the index date) with a control period (91 to 120 days before the index date). The adjusted odds ratios (ORs) of stroke and AMI were estimated using conditional logistic regression models. Our results showed that overall NSAIDs use increased the risk of stroke by 1.40-fold (95% confidence interval [CI] 1.25 to 1.56) and risk of AMI by 1.73-fold (95% CI 1.29 to 2.32). After classifying NSAIDs into selective and nonselective groups, only nonselective NSAIDs use significantly increased the risks of stroke (adjusted OR 1.39; 95% CI 1.25 to 1.55) and AMI (adjusted OR 1.82; 95% CI 1.37 to 2.41), respectively. In conclusion, nonselective NSAIDs were associated with an increased risk of both stroke and AMI in patients with RA.