The positioning of the left ventricular assist device (LVAD) inflow cannula within the left ventricle (LV) has received considerable attention, due to clinical findings that show a significantly lower heart failure admissions-free survival rate for malpositioned cannula.1 The angulation of the inflow cannula within the LV affects flow dynamics, which can predispose thrombus formation. Adamson et al. and others have identified best practices associated with low thrombus incidence, which have emphasized the importance of aligning the cannula parallel to the intraventricular septum in order to avoid flow obstruction.2,3 Previous studies have demonstrated that patients with thrombus had more angulated inflow cannula and a shallower pump pocket than those without thrombus.4
Orientation of the Heart in the Human Chest
The position and orientation of the human heart in the chest relates to the evaluation of inflow cannula orientation of LVAD patients. When the LVAD is implanted, the positioning of the inflow cannula is relative to the cardiac anatomy. A right-handed cardiac coordinate system can be established in which the origin is located at the apex, and aligned with the short (e1′, e2′) and long (e3′) axes of the LV (Figure 1). Once the coordinate system is designated, measurements of angles and distance are referred to that system.
Alternately, a coordinate system can be established that is aligned with the anatomical planes, with the origin at the heart apex, and axes aligned with the A-P (e1), R-L (e2), and S-I (e3) directions as shown in Figure 1. Measurements of cardiac geometry in this coordinate system must be interpreted in light of the orientation of the heart within the chest. Without patient-specific data, an assumption must be made that standardized measurements adequately describe the rotation of the heart relative to the anatomical planes. A recent study of 185 healthy adults found that the heart is rotated 36° counter-clockwise (CCW) in the coronal plane, 50° CCW in the transverse plane, and 50° CCW in the sagittal plane.5 These data are used to calculate a rotation matrix that can be used to mathematically transform the geometry measurements from one coordinate system to the other (Figure 2).
Coordinate Transformation Basics
The rotation of coordinate systems around an axis is defined with an angle in the plane perpendicular to that axis. The transformation between the coordinate systems is defined with a rotation matrix, with components that are sine and cosine functions of the angle. For example, a rotation of θc in the coronal plane produces a matrix:
Analogous rotation matrices can be identified for rotations in the transverse and sagittal planes. Combining the rotation matrices together produces a complex rotation that defines the transformation from anatomical to cardiac coordinates:
The reverse transformation is calculated from the transpose of this matrix. The transformation matrix is necessary to relate the orientation of a vector, for example, the direction of the inflow cannula, to geometry measurements made in another coordinate plane.
Orientation of the Left Ventricular Assist Device Inflow Cannula
When the LVAD inflow cannula is positioned during the implant surgery, the orientation is based on aligning with cardiac-specific features, such as the mitral valve or intraventricular septum. In this context, the orientation of the inflow cannula can be described as a vector, uIC′, in the cardiac coordinate system, with angles α1′, α2′, and α3′ defining the rotation of the inflow cannula direction with respect to the cardiac coordinate axes. For example, if the inflow cannula is implanted parallel to the intraventricular septum in the LV midplane, the orientation with respect to the long axis direction (e3′) can be described by the angle α1′. The vector defined by this angle is:
This same vector has different components when transformed to the anatomical coordinate system, and is calculated as:
This 3-dimensional vector will produce different values for angle and distance depending on the 2-D measurement plane. To calculate the projection of a 3-D vector onto a 2-D plane, another mathematical operation is needed:
in which uICc is a unit vector in the coronal plane. This sequence of steps (1) to (6) provide the calculation of inflow cannula angle measured from the chest X-ray to be related to the orientation in the cardiac coordinate system.
Bringing It All Together With Published Data
In order to relate findings of the importance of LVAD inflow cannula orientation from different sources, calculation of the same angles must be made for direct comparison. We begin with tabulating the data from 3 studies evaluating the effect of cannula malposition (Table 1).
For the May-Newman study,6uIC′, must be transformed to the anatomical coordinate system using equation (5), with RTOTAL composed of the values from Odille et al. The projection of the new vector, uIC, onto the coronal plane, uICC, and αc are computed from equation (6). This sequence of steps results in values for αc of 86° for the standard cannula position and 122° for the malpositioned cannula.
For the Lima study,7u′ICL must be transformed to the anatomical coordinate system using equation (5), with RTOTAL composed with values from the same study,
from which the angle αc can be computed using equation (6). The result gives the same angles measured in the study, 56° for the standard position, and 70° for the malpositioned orientation. It should be noted that the cardiac coordinate systems differ, as the Lima study orients the z-axis toward the mitral valve, rather than parallel to the intraventricular septum as in the Odille study. This difference results in a 15° rotation between the axis in the coronal plane.
The angle in the coronal plane, αc, differs across studies but shows a consistent trend. The relationship between cannula position and event-free survival are likely related to the optimization of flow architecture, which is also affected by LVAD speed, pulsatility, and native cardiac function. The link between benchtop experiments and clinical studies often includes geometrical descriptions made in different settings. An understanding of these geometrical relationships is needed in order to relate studies using different reference measurements.
1. Imamura T, Nguyen A, Chung B, et al. Association of inflow cannula position with left ventricular unloading and clinical outcomes in patients with HeartMate II left ventricular assist device. ASAIO J 2019.65: 331–335.
2. Adamson RM, Mangi AA, Kormos RL, Farrar DJ, Dembitsky WP. Principles of HeartMate II implantation to avoid pump malposition and migration. J Card Surg 2015.30: 296–299.
3. Klodell CT, Massey HT, Adamson RM, et al. Factors related to pump thrombosis with the Heartmate II left ventricular assist device. J Card Surg 2015.30: 775–780.
4. Taghavi S, Ward C, Jayarajan SN, Gaughan J, Wilson LM, Mangi AA. Surgical technique influences HeartMate II left ventricular assist device thrombosis. Ann Thorac Surg 2013.96: 1259–1265.
5. Odille F, Liu S, van Dam P, et al. Statistical variations of heart orientation in healthy adults. Computing in Cardiology 2017.44: 1–4.
6. May-Newman K, Marquez-Maya N, Montes R, Salim S. The effect of inflow cannula angle
on the intraventricular flow field of the left ventricular assist device-assisted heart: An In Vitro
Flow Visualization study. ASAIO J 2019.65: 139–147.
7. Lima B, Dur O, Chuang J, et al. Novel cardiac
coordinate modeling system for three-dimensional quantification of inflow cannula malposition of HeartMate II LVADs. ASAIO J 2018.64: 154–158.