Safety assessment of active implants is defined in the technical specification ISO/TS 10974. However, the safety of small passive implants must be assessed individually for each device, e.g., via the ASTM F2182-11a standard test method. Computational studies have demonstrated that the mechanism postulated for elongated implants at 64 MHz is not applicable. We propose a new mechanistic theory of general validity applicable across a large frequency range to describe exposure enhancement at critical lead locations. Simulation-based assessment was successfully compared to predictions from the mechanistic model on the impact of several parameters (tissue dielectric properties, insulation thickness, and tip shape).