The traditional approach to biomedical device design for resistance to fatigue failure is based on a total-life philosophy for predicting safe in vivo operating conditions.  Although this approach is extremely useful for determining safe applied loads and displacements, it cannot predict the critical flaw size that may eventually lead to a cumulative damage and eventual failure in an implanted device subjected to millions of pulsatile cyclic loads.  There exists a dearth of relevant data in the literature on such fracture-mechanics based approaches to fatigue. Furthermore, the available literature shows fatigue response in product forms that are not appropriate for stent manufacture, namely bulk Nitinol bar and sheet.  The results presented here document the fatigue response in Nitinol tubing, similar to that used for medical device manufacture, which has undergone a series of shape setting procedures and has a resultant A_f temperature of approximately 29°C. Fatigue-crack growth and threshold behavior at various frequencies (5 and 50 Hz) and load ratios (R = 0.1, 0.5, and 0.7) are presented; these results indicate that fatigue thresholds in the tubing are higher than previous values quoted in the literature for bulk Nitinol material.   Such results are used to provide quantitative estimates on the potential effect of flaws in limiting the life of endovascular stents and other Nitinol biomedical devices.