
In medical device development, a lifecycle approach is indispensable because devices are modified frequently. In clinical trials, double-blind sham-controlled procedures are difficult because they may cause adverse events. The verification of superiority in medical treatment is essential because of the invasive nature of medical devices, which may raise ethical issues with regard to control groups. Accordingly, typical prospective randomized double-blind studies are less common; the majority of studies are single-arm trials. To assess the clinical trial design of medical devices, 53 items approved in 2016 by the Pharmaceuticals and Medical Devices Agency of Japan (PMDA) were reviewed. The results indicate that more than half involved single-arm trials, whereas only 23% were prospective randomized trials. There was only one sham-controlled prospective randomized trial. With respect to single-arm trials, performance goals were established from raw data on a previous generation of the device in 33% of studies and from multiple publications in 33%. These results suggest that continuous data accumulation, such as a registry, could provide a control group based on patient-level data. The PMDA recognizes that the appropriate method to collect reliable and robust data based on a registry has the potential to facilitate a pre-market and post-market balance because medical device development is a lifecycle. The PMDA intends to move toward a fast-track approval scheme and controlled release of certain new medical devices to the market after conditional approval if benefits and risks can be reasonably proven from the available clinical data and an appropriate risk management plan is prepared.