FIGURE LEGENDS
Figure 1: Example of preoperative CT-derived anatomic
measurements. A. Axial plane midline-to-LV apex distance. B. Coronal
plane angulation of LV apex-LV outflow tract axis.
Figure 2: Example of inflow cannula (IFC) angular deviation
measurements. Ideal line in each plane is from LV apex through
approximate center of mitral valve. Actual position is line through IFC
center. A. Anterior plane. Deviation is toward superior (+) or inferior
(-) free wall. Illustrated cannula deviates very slightly inferiorly. B.
Lateral plane. Deviation is toward septum (-) or lateral wall (+).
Illustrated cannula deviates very slightly septally.
Figure 3: CF-LVAD inflow cannula position by surgical
technique. Open circles: conventional sternotomy. Black stars: lateral
thoracotomy/hemisternotomy
Figure 4: Significant linear relationships between anatomic
variables and IFC angles. Best-fit linear regression lines (solid) and
95% confidence intervals (dashed) are shown. A. Anterior plane angle vs
LVOT angle (p = 0.01); B. Lateral plane angle as function of LVOT angle
(p = 0.04); C. Lateral plane angle as function of apex-midline distance
(p = 0.04).
Figure 5: Example patient with higher LVOT angle (35.2 deg) on
preoperative CT (A). Postoperative CT demonstrates (B) inferior-wall
(-30.6 deg) and (C) lateral-wall directed (+16.9 deg) inflow malposition
consistent with regression models.
Figure 6. Box and whisker plots of three-month (A) and
six-month (B) VAD flow index (FI) stratified by whether inflow cannula
is angulated toward lateral wall. p = 0.002 for both timepoints.
Figure 7. Box and whisker plots of relationship between
readmission prevalence and inflow malposition. A. Anterior plane angle
and overall heart failure/VAD-related readmissions (HFRAs; p = 0.01). B.
Total angular deviation and 30-day HFRAs (p = 0.04).