Association of LVAD Cannula Position with Stroke Risk

Abstract

Stroke remains a major cause of morbidity and mortality in LVAD patients, with incidence upto to 22%. We hypothesized that certain LVAD cannula positions are associated with unfavourable flow patterns resulting in altered thrombogenicity, and therefore higher stroke risk. We conducted a retrospective review of patients that underwent LVAD implantation at our institution from 2011 to 2016. All patients who underwent cardiac CT scan, were included. Patients with non-contrast CT scans, and poor quality imaging data were excluded. We studied imaging data from three modalities: X-ray, Echocardiography and cardiac CT. The primary outcome was defined as stroke within one year of LVAD implantation. A total of 78 patients were included in the study. The predominant device was HeartMate II (94.8%), there were three patients with HVAD and one patient with HeartMate III. Of these, 12 patients had stroke within one year of LVAD implantation: 10 were ischemic and 2 were hemorrhagic. Mean time to stroke was 108 days. There were no significant differences between the two groups with regard to X-ray or Echocardiographic variables. The mean outflow cannula angle in the stroke group was 61.19 degrees, whereas that in the non-stroke cohort was 45.92 degrees. The mean diameter of outflow cannula anastomosis in the stroke group was 1.52 cm whereas that in the non-stroke group was 1.34 cm. Our study was the first to show that LVAD outflow cannula angle and diameter of anastomosis are significantly associated with stroke risk. Therefore, optimization of surgical technique/ engineering methods may help mitigate stroke risk in LVAD patients.