About three quarters of patients who undergo mitral valve surgery develop aortic valve disease during follow-up, with 5% of them requiring a new intervention [12]. Prior cardiac surgery in SAVR candidates increases operative risk by up to 70% [13]. Furthermore, the presence of a prosthetic mitral valve (PMV) doubles the mortality in patients undergoing SAVR compared with those without a pre-existing PMV [14]. For these reasons, TAVR seems to be a reasonable alternative for this high-risk population in view of its extensive use and proven efficacy in high- and intermediate-risk patients with severe aortic stenosis [1,2,3]. Our experience with the two reported index cases, similar to other reports and registry data included in a recent systematic review [15], suggests the feasibility of this course of action. Furthermore, it seems the presence of a PMV may not significantly increase short term mortality risk post-TAVR [9, 11]. However, randomized clinical trial data and guideline recommendations are still lacking.

In view of the technical concerns and risk of interference between the implanted THV and the pre-existing PMV, careful pre-procedural planning with CT and TEE imaging is highly recommended. Interference between the THV and PMV may lead to acute/chronic malfunction of both prostheses and an increased risk of device embolization [15]. The latter was noted among 6.7% of post-MVR patients in a multicenter Spanish registry [9]; however, this was not significantly higher compared with patients without prior mitral valve surgery. Furthermore, all cases of embolization occurred in subjects with the PMV to aortic annulus distance of < 7 mm by CT imaging. A minimum mitral-aortic distance of 3 mm for the balloon expandable Sapien valve and 4 mm for the self-expandable CoreValve has been recommended [8]. The two index cases had a distance of 6.3 mm and 3.6 mm, respectively, which are within the acceptable minimum range.

The preferential use of self-expandable valves in post-MVR TAVR procedures may seem logical in view of the possibility of partially recapturing the prostheses during deployment. However, recent reports and retrospective analyses have demonstrated similar peri-procedural outcomes with the use of either self or balloon expandable valves [8, 9, 15].

Bleeding complications have been reported to be more frequent in subjects with PMV who undergo TAVR [9]. This is likely due to the need for more aggressive antithrombotic therapy because of a higher rate of atrial fibrillation and the presence of a mechanical PMV. We did not observe any bleeding complication among the 2 index cases.

Our second patient developed complete heart block a few hours post-TAVR. The peri-procedural occurrence of conduction abnormalities is reportedly the most common complication of TAVR with about 13% of subjects requiring permanent pacemaker implantation [16]. This is due to the close proximity of the aortic valve apparatus to the AV node, HIS bundle and the proximal left bundle branch. This explains the high peri-procedural incidence AV block and new onset left bundle branch block [16]. Factors associated with the occurrence of these abnormalities include prior right bundle branch block, transcatheter valve type and implantation depth [16]. Permanent pacemaker implantation is about 5 times more frequent among subjects receiving self-expandable CoreValve (25–28%) compared with subjects with a balloon expandable Edwards SAPIEN/SAPIEN XT valve (5–7%). This observation is likely related to the deeper implantation required for self-expandable valves [16]. It is not clear whether post-MVR subjects are at a higher risk of post-TAVR conduction abnormalities. However, our second case likely had an AV conduction disturbance in view of her prior history atrial fibrillation with slow ventricular response. This might have increased her risk of developing complete heart block post-procedure.