Cardiac L-type Ca current ( I Ca,L) is controlled not only by voltage but also by Ca 2+-dependent mechanisms. Precise implementation of I Ca,L in cardiac action potential models therefore requires thorough understanding of intracellular Ca 2+ dynamics, which is not yet available. Here, we present a novel formulation of I Ca,L for action potential models that does not explicitly require the knowledge of local intracellular Ca 2+ concentration ([Ca 2+] i). In this model, whereas I Ca,L is obtained as the product of voltage-dependent gating parameters ( d and f), Ca 2+-dependent inactivation parameters ( f Ca: f Ca-entry and f Ca-SR), and Goldman-Hodgkin-Katz current equation as in previous studies, f Ca is not a instantaneous function of [Ca 2+] i but is determined by two terms: onset of inactivation proportional to the influx of Ca 2+ and time-dependent recovery (dissociation). We evaluated the new I Ca,L subsystem in the framework of the standard cardiac action potential model. The new formulation produced a similar temporal profile of I Ca,L as the standard, but with different gating mechanisms. Ca 2+-dependent inactivation gradually proceeded throughout the plateau phase, replacing the voltage-dependent inactivation parameter in the LRd model. In typical computations, f declined to ∼0.7 and f Ca-entry to ∼0.1, whereas deactivation caused fading of I Ca,L during final repolarization. These results support experimental findings that Ca 2+ entering through I Ca,L is essential for inactivation. After responses to standard voltage-clamp protocols were examined, the new model was applied to analyze the behavior of I Ca,L when action potential was prolonged by several maneuvers. Our study provides a basis for theoretical analysis of I Ca,L during action potentials, including the cases encountered in long QT syndromes.