Effect of IFC Position on LV Unloading and Outcomes
We found that a laterally-directed IFC predicted significantly reduced VAD FI at three and six months (Figure 6). Computational studies8,9 have shown that increasingly lateral IFC position causes poor LV washout and longer particle residence times, consistent with poor LV unloading and compromised flow. Imamura10 found that increasing HVAD pump body area in the coronal x-ray plane was associated with a reduced decrement in LVEDD and pulmonary capillary wedge pressure during ramp studies. Larger pump area would correspond to a more laterally-oriented IFC.
In contrast, we did not find an association between malposition and decrement in MR or LVEDD from pre- to post-LVAD. However, we had only fixed-RPM echo studies. This MR finding is contrast to our previous work, wherein greater total malposition magnitude predict a higher degree of post-LVAD MR11. However, the studies differ on a few key aspects: a) Pasrija included only patients with at least mild-moderate preoperative MR; b) they used a subset of the current population (only patients implanted 2013-2017; N = 41 vs 83 patients in the current study); and c) MR was graded on a more granular scale, which captured intermediate degrees such as mild-moderate. These differences likely explain the discrepancy.
Imamura10 also found that a steeper coronal plane IFC angle (≥ 76 deg) was associated with increased LVEDD and wedge pressure. However, such an angle represents relatively extreme superior-wall directed malposition, which was rare in our study (Figure 3). Imamura’s report has only nine patients in this group and 27 total patients, so their findings may not be translatable to a larger population.
This and other clinical studies of IFC malposition in CF-LVADs10,12,13 rely on two-dimensional measurements of IFC angle relative to the spine or horizontal in the coronal plane, and are unable to identify septal or lateral-wall malposition.
We did find that IFC malposition was associated with readmissions (HFRAs). Patients with total malposition magnitude > 40 deg had an increased incidence of 30-day HFRAs, while patients with a lateral plane angle ≤ -20 deg experienced an increased incidence of overall HFRAs. Imamura associated a higher IFC coronal angle (> 65 deg) with increased HFRAs in HVAD patients10, while an angle > 28 deg was the cut-off in a larger population with the Heartmate 312. It is unclear why the two devices have such different thresholds, since both are intrapericardial with integral IFCs of roughly the same diameter. In contrast, Shih et al.13 found increased “malposition-related” adverse event rates, including HFRAs, in patients with a Heartmate 3 whose median spine-IFC angle on chest x-ray was 66 deg at one month, compared to 59 deg in patients without adverse events. In this study, a higher pump-spine angle would correspond to a shallower angle using the method of Imamura.
Finally, we did not find an association between malposition and either PT/eCVA or survival free of malposition-related surgery. PT and eCVA have multifactorial causes, including hypertension and inadequate antithrombotic therapy14,15. While we did observe an association between malposition and pump thrombosis in the Heartmate II, we observed deviations that were much more extreme than those observed for the HVAD, and which tended to be primarily superior-wall and septum-directed1. The degree of malposition observed here may not be sufficient to independently cause thrombosis. Mortality is similarly multifactorial.