Cannula and Pump Positions Are Associated With Left Ventricular Unloading and Clinical Outcome in Patients With HeartWare Left Ventricular Assist Device

Abstract

Background Cannula and pump positions are associated with clinical outcomes such as device thrombosis in patients with HeartMate II; however, clinical implications of HVAD (HeartWare International, Framingham, Massachusetts) cannula position are unknown. This study aims to assess the relationship among cannula position, left ventricular (LV) unloading, and patient prognosis. Methods and results Twenty-seven HVAD patients (60.0 ± 12.6 years of age and 19 males [70%]) underwent ramp test. Device position was quantified from chest X-ray parameters obtained at the time of the hemodyamic ramp test: (1) cannula coronal angle, (2) pump depth, (3) cannula sagittal angle, and (4) pump area. Lower cannula coronal angle was associated with LV unloading (as measured by smaller LV diastolic dimension and lower pulmonary capillary wedge pressure). Smaller pump area was associated with LV dynamic unloading, as assessed by steeper negative slopes of LV diastolic dimension and pulmonary capillary wedge pressure during incremental rotational speed change. Cannula coronal angle ≤65° was associated with reduced heart failure readmission rate (hazard ratio, 10.33; P = .007 by log-rank test). Conclusion HVAD cannula and pump positions are associated with LV unloading and improved clinical outcomes. Prospective studies evaluating surgical techniques to ensure optimal device positioning and its effects on clinical outcomes are warranted.

Mobile inflow cannula leading to positional ventricular tachycardia

“…Malpositioned LVAD inflow cannula is associated with pump failure, thrombosis, suction events, and VT (3,4). VT is associated with decreased pump flow, right ventricular failure, and increased overall morbidity and mortality in LVAD-supported patients (4).…”

Pulmonary artery wedge pressure respiratory variation is correlated with haemodynamic improvement with increased left ventricular assist system speed

“…4 Similarly, the inability to reduce the PAWP with LVAS speed increase is also associated with worse outcomes. 3 The correlation between PAWP Δramp and TMP LV suggests that biventricular interaction may play a role: patients with elevated PAWP and low CVP (high TMP LV ) have a higher PAWP var and PAWP Δramp , and conversely those with elevated PAWP and high CVP (low TMP LV ) have lower PAWP var and PAWP Δramp . One may hypothesize that the presence of a low TMP LV indicates the presence of a 'pseudoconstrictive' state due to increased cardiac volumes and/or right ventricular failure.…”

“…In one cohort of 35 LVAS patients, only 43% had normal central venous pressure (CVP) and/or pulmonary artery wedge pressure (PAWP) at the time of haemodynamic assessment with right heart catheterization (RHC), with 57% achieving normal filling pressures at the end of ramp speed adjustment . Further studies have suggested that inability to normalize filling pressures during speed adjustment confers a worse prognosis . Even in the hands of LVAS‐trained clinicians with the aid of real‐time invasive haemodynamic data, over 40% of patients remain non‐optimized after LVAS speed adjustment.…”

“…Kussmaul’s sign, or the absence of a decrement in CVP with inspiration, is associated with worse outcomes in heart failure patients . Similarly, the inability to reduce the PAWP with LVAS speed increase is also associated with worse outcomes . The correlation between PAWP Δramp and TMP LV suggests that biventricular interaction may play a role: patients with elevated PAWP and low CVP (high TMP LV ) have a higher PAWP var and PAWP Δramp , and conversely those with elevated PAWP and high CVP (low TMP LV ) have lower PAWP var and PAWP Δramp .…”

“…2 Further studies have suggested that inability to normalize filling pressures during speed adjustment confers a worse prognosis. 3 Even in the hands of LVAS-trained clinicians with the aid of real-time invasive haemodynamic data, over 40% of patients remain non-optimized after LVAS speed adjustment. Which patients will not benefit haemodynamically from speed increase remains unclear.…”