Troubleshooting transcaval access: honoring our commitments

We are grateful to Mano and colleagues (1) for sharing their complication of transcaval TAVI access, which highlights several important lessons.

During their case, a transcaval aortic 0.014” guidewire was introduced but the operators were unable to upsize to a TAVI guidewire, so they withdrew the guidewire and successfully completed transcaval TAVI at a supra-renal aortic target. A postprocedure CT revealed a pseudoaneurysm and aortocaval fistula at the first (abandoned) traversal site. The pseudoaneurysm was excluded with a transfemoral covered stent, and the patient was discharged home. In retrospect the complications appear related to a co-registration error and to an abandoned guidewire traversal.

Elective transcaval TAVI should be planned (2) on pre-procedure CT to identify a sufficiently large calcium-free traversal target for the transfemoral venous introducer sheath, without interposed structures such as bowel, and sufficiently far from renal and iliac branches to allow bailout covered stent implantation. The CT is used to identify specific traversal targets, fluoroscopic projection angles, select appropriately sized bailout devices, and to identify structures for co-registering the CT and fluoroscopy.

The last step, co-registration with known (“fiducial”) structures, can be confusing. We plan transcaval targets in relation to specific lumbar vertebrae, which should be uniquely identified in relation to other structures such as iliac crest or renal parenchyma. One rule-of-thumb is that the top of the iliac crests typically corresponds to the L4-L5 lumbar interspace and to the aorto-iliac bifurcation. We recommend that the aortic target be depicted in digital-subtraction aortography in context of those fiducial structures, and that operators also view un-subtracted or partial-subtraction “landmark” modes before beginning traversal. In this case, the renal parenchymal blush evident in Figure1A should alert the operators of a possible peri-renal transcaval traversal, which is undesirable.

The second challenge described here is inability to advance a transcaval microcatheter into the aorta after successful electrosurgical caval-aortic guidewire entry. Typical traversal equipment assemblies — consisting of a coaxial stiff 0.014” guidewire inside a 0.014” microcatheter inside a 0.035” microcatheter, inside a short curved guiding catheter —are intended to upsize to a rigid Lunderquist-style 0.035” guidewire and thereby deliver a transcaval TAVI introducer sheath. When the 0.014” microcatheter fails to deliver, small coronary or peripheral angioplasty balloon catheters usually successfully cross and dilatate, and allow intended escalation of microcatheter caliber until the 0.035” Lunderquist wire and transcaval introducer sheath are delivered. Countertraction via the aortic snare can be helpful. When angioplasty catheters fail, some have used laser (3) or even rotational atherectomy [J Fredi, Personal Communication] devices across the aortic wall.

We consider a successful caval-aortic guidewire traversal to be a “commitment,” because abandoned traversals can cause pinhole aortic leaks. Absent a corresponding decompressing venous hole, such leaks can generate pseudoaneurysm or (unlike this case) even hemorrhage. Such leaks are usually evident on procedural aortography. The best solution is to re-cross the aorta and complete the transcaval TAVI within millimeters of the abandoned hole, so that the large aorto-caval fistula encompasses the earlier abandoned spot, and the nitinol closure device facilitates thrombosis and closure.

Following these simple procedure and troubleshooting steps, and “honoring our commitments” to transcaval aortic guidewire traversal, can help to prevent and manage complications.