Answers are grounded in this paper and cite their sources.
Limited donor availability and increasing reliance on left ventricular assist devices (LVAD) as bridge to heart transplants have ushered in a new set of clinical problems. The most frequent LVAD complications are infections, embolic events, and hemorrhage. 1, 2 Inflow and outlet cannula obstruction is the next most frequently observed complication, of which inflow obstruction has been described in more detail. Echocardiography and especially transesophageal echocardiography (TEE) have been the most commonly used noninvasive techniques in the evaluation of obstructed conduits. 3–5 Fluoroscopy and selective catheterization of the conduits are also being introduced as additional modalities. 6 However, the optimal method or combination of methods to identify device problems is yet to be defined. In this report, we describe a case of LVAD malfunction that presented a diagnostic dilemma, despite evaluation using different modalities. The device was subsequently explanted and a new device placed. Case Report A 17 year old African-American woman with postpartum cardiomyopathy had placement of a Thermo Cardiosystems, Inc. (TCI) HeartMate LVAD as bridge to cardiac transplant. She had an uncomplicated early postoperative recovery. Two months after discharge, the patient reported frequent alarms with low LVAD beat rates, but with no symptoms. Device interrogation revealed intermittent high power use, suggesting outflow cannula obstruction. Initial evaluation with TEE demonstrated normal LV decompression and the absence of aortic valve opening. There was excellent inflow opening, which in turn indicated that the left ventricle was adequately supported by the device. The peak velocity obtained by TEE Doppler across the outflow cannula was 2.3 m/sec, which was considered normal. Retrograde cannulation of the outflow conduit by means of the femoral artery was then performed. This demonstrated the absence of outflow obstruction (Figure 1). Subselective cannulation of the inflow conduit was done with the catheter positioned near the inflow (Figure 2). Figure 1: Left anterior oblique view (LAO) showing opacification of the outflow cannula from the device to the aorta. The proximal portion (indicated by arrows) of the outflow cannula as visualized in this projection shows no evidence of obstruction.Figure 2: Ventriculogram demonstrating a subselective opacification of the inflow conduit by means of catheter positioned in the left ventricle close to the apical inflow tract.Left ventricular hemodynamics were recorded using a 6F multipurpose catheter (Figure 3). The LV pressure tracings demonstrated peak LV pressures up to 40 mm when the LV and LVAD systole were asynchronous. LV pressures were 22–24 mm with synchronous LV and LVAD systole. The initial decision was made to conservatively follow the patient with periodic VAD system analysis. The cause of these “alarming!” episodes could not be explained, but was presumed to be transient device sensing problems with inappropriate alarms. However, during the subsequent month, the patient returned with increasing frequency of episodes, more so in the upright position. These were now associated with transient reduction in the VAD outputs and intermittent device nonfunction for 10 to 45 seconds at a time. The increasingly frequent episodes of positional drop in flow rates and intermittent device nonfunction prompted surgical exploration, explantation, and insertion of a new TCI HeartMate device. The interval between the initial and subsequent device implantation was 3 months. Intraoperative examination of the device did not reveal any mechanical kinking, thrombus, or abnormal orientation of the conduits to the aortic LV apical implantation sites. With the significant positional reduction in VAD outputs, intermittent kinking of the outflow cannula was the most likely cause of her symptoms. This finding was supported by the intermittent power surges observed during system analysis, most likely occurring when the device had to expend more energy during these episodes to overcome resistance to flow. Her post reimplantation course was unremarkable, with no recurrence of these episodes. Figure 3: Left ventricular hemodynamics measured by means of 6F multipurpose catheter during synchronous phase of ventricular contraction (a) with left ventricular assist device filling showing peak systolic pressure of 24 mm of Hg and (b) during asynchronous phase with a peak pressure of 42 mm of Hg.Discussion There is minimal information in the literature as to the optimal method or combination of methods for evaluation of patients presenting with possible LVAD malfunction. Echocardiography has been reported to be useful in the noninvasive evaluation of inflow obstruction. 3–5, 7 However, there has been no well established echocardiographic criteria for evaluation of such patients. There is very little information regarding Doppler-derived velocities and pressure gradients that have been standardized to differentiate normal from abnormal flow profiles. Velocity > 2.5 m/sec at the inlet is being increasingly recognized as an indicator of inflow obstruction. Given the ever increasing population of LVAD recipients, it would be appropriate to develop a set of standardized data incorporating a series of Doppler velocities and pressure gradients for different LVAD models, as well as to differentiate between inflow versus outflow characteristics. We are currently developing a database from intraoperative and postoperative echocardiographic examinations, so as to have a baseline for future comparison. TEE should be considered the principle modality in the evaluation of LVAD dysfunction (Table 1). Aortic valve opening, aortic forward flow, non laminar flow in the left ventricle, mismatch of Doppler-derived cardiac output at the pulmonic valve, and device output are indicators of device malfunction. 3 Persistent aortic valve opening with each cardiac cycle indicates inadequate hemodynamic support by the LVAD, which could be due to inflow obstruction in addition to intrinsic device malfunction. 3 Incomplete LV decompression and turbulence at the inflow cannula may be suggestive of inflow cannula obstruction. The possible causes of inflow obstruction include malposition at the time of initial implantation or subsequent migration of the cannula, so as to abut the cannula against the lateral wall or septum 6; kinking of the tubing; and thrombus or fungal growth. Two-dimensional echo is ideally suited to determine the orientation of the inflow cannula. Partial obstruction against the LV wall has been postulated to be secondary to ventricular remodeling after the effects of chronic unloading of the LV. Table 1: Evaluation of LVAD DysfunctionFluoroscopy and selective catheterization of the conduits have been described as useful in assessing the orientation of the inflow cannula. 6 Normally, the inflow cannula should be oriented toward the mitral valve. The left anterior oblique view (LAO) is best suited to evaluate abnormal orientation of the conduit against the septum or the lateral wall of the LV. Left ventriculography has not been shown to provide useful information in the assessment of inflow obstruction and is not expected to provide better visualization of the outflow cannula. Therefore, we believe TEE is adequate for demonstration of inflow cannula function. 6 Selective cannulation of the outflow conduits is easily performed and can rapidly eliminate outflow cannula problems, and LV hemodynamics can complement the information obtained from other methods. Peak LV systolic pressures > 40–50 mm would suggest incomplete ventricular decompression. Summary With the increasing number of LVAD recipients due to lack of donor availability and use of an LVAD as destination therapy, there will be a significant proportion of this population presenting with device related complications. Initial evaluation should be with TEE because conduit obstruction would be a significant complication, among others. As observed in this case report, it is sometimes difficult to noninvasively detect device malfunction, and even surgical exploration left questions unanswered. Echocardiography has been the principal modality used so far to assess the LVAD cannulas. More reliance on Doppler velocities and establishment of a normal range of values and flow profiles of the inflow and outflow cannulas are necessary. Baseline echocardiographic studies could be used for future comparisons. Fluoroscopy and selective angiography have also been reported to be helpful. We would also recommend measuring LV hemodynamics, which have not been previously described, as an additional mode of evaluation of this emerging problem. This case report demonstrates the need for additional noninvasive modalities for diagnosis of ventricular assist device malfunction.
No takes yet. Share an insight, caveat, or question.
Ferns et al. (2001) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: