Current in vivo models of arterial lesions often lead to unpredictable results in terms of lesion anatomy and hemodynamical significance. This study aimed to evaluate the impact of coronary stenosis using a novel in vivo adjustable stenosis model capable of mimicking advanced human coronary lesions. We developed a series of balloon expandable covered coronary stents with a central restriction, mimicking different intermediate to severe stenosis, and implanted them percutaneously in coronary arteries of eight healthy hybrid Landrace pigs. Optical coherence tomography (OCT) pullbacks and fractional flow reserve (FFR) were acquired along the artery after implantation of the stenotic stents for precise evaluation of anatomy and functional impact. Diameter and area stenosis after deployment of the stenosis implant were, on average, respectively, 54.1 ± 5.9 and 78.4 ± 5.8 % and average FFR value was 0.83 (SD 0.13). There was a low correlation between FFR and MLA evaluated by OCT (r = 0.02, p = 0.94), improved with percentage area stenosis (r = -0.55, p = 0.12), or OCT volumetric evaluation of the stenosis taking into account not only the MLA but also the length of the lesion (r = -0.78, p = 0.01). This study presents a method and proof of concept for percutaneously introducing, and removing, anatomical stenosis of predetermined severity in vivo. Such in vivo model may be used to create and evaluate the impact of focal stenoses on physiological parameters such as FFR.