Pressure-based fractional flow reserve (FFR) is used clinically to evaluate the functional severity of a coronary stenosis, by predicting relative maximal coronary flow (Q(s)/Q(n)). It is considered to be independent of hemodynamic conditions, which seems unlikely because stenosis resistance is flow dependent. Using a resistive model of an epicardial stenosis (0-80% diameter reduction) in series with the coronary microcirculation at maximal vasodilation, we evaluated FFR for changes in coronary microvascular resistance (R(cor) = 0.2-0.6 mmHg. ml(-1). min), aortic pressure (P(a) = 70-130 mmHg), and coronary outflow pressure (P(b) = 0-15 mmHg). For a given stenosis, FFR increased with decreasing P(a) or increasing R(cor). The sensitivity of FFR to these hemodynamic changes was highest for stenoses of intermediate severity. For P(b) > 0, FFR progressively exceeded Q(s)/Q(n) with increasing stenosis severity unless P(b) was included in the calculation of FFR. Although the P(b)-corrected FFR equaled Q(s)/Q(n) for a given stenosis, both parameters remained equally dependent on hemodynamic conditions, through their direct relationship to both stenosis and coronary resistance.