Secondary Logo

Journal Logo

Cardiovascular Anesthesiology: Echo Rounds

Echocardiographic Artifact Induced by HeartWare Left Ventricular Assist Device

Lesicka, Agnieszka MD*; Feinman, Jared W. MD; Thiele, Karl MS; Andrawes, Michael N. MD*

Author Information
doi: 10.1213/ANE.0000000000000664

A 66-year-old man with ischemic cardiomyopathy was admitted for management of worsening heart failure and implantation of a HeartWare (HeartWare International, Framingham, MA) left ventricular assist device (LVAD) as a bridge to heart transplantation. Written consent was obtained from the patient for publication of this report.

A complete transesophageal echocardiography (TEE) examination was performed with the Philips iE33 xMATRIX Ultrasound (Philips Healthcare, Andover, MA) and, most notably, revealed a dilated left ventricle with severe hypokinesis and severe tricuspid regurgitation. A HeartWare LVAD was implanted, and a tricuspid annuloplasty was performed. After bypass, 2-dimensional TEE imaging revealed proper inflow cannula orientation at the left ventricular apex with no evidence of obstruction. Color Doppler, however, revealed a continuous artifact that extended the length of the color sector, from the LVAD inflow cannula toward the origin of the imaging sector (Fig. 1A; Supplemental Digital Content 1, Video 1, The use of spectral Doppler with both pulsed wave and continuous wave (CW) demonstrated an artifact that precluded interrogation of the inflow cannula velocities (Fig. 1B). The alignment of the impeller and the associated artifact with the coaptation point of the mitral valve made evaluation of the mitral valve challenging. Unobstructed laminar flow was seen at the site of the outflow graft anastomosis in the ascending aorta. Despite HeartWare LVAD implantation and tricuspid valve annuloplasty, the right ventricle became dilated, and the tricuspid valve regurgitation remained moderate to severe. The tricuspid valve was then replaced, and a right ventricular assist device was placed for support.

Figure 1
Figure 1:
A, Side-by-side 2D and color Doppler interrogation of the HeartWare left ventricular assist device (LVAD) in the midesophageal 4-chamber view. Note the continuous artifact that extends vertically from the LVAD inflow cannula for the length of the color sector. This artifact is not present when 2-dimensional imaging is used (Supplemental Digital Content 1, Video 1, B, Continuous-wave Doppler interrogation of the HeartWare LVAD. The artifact precludes interrogation of the true inflow velocity. LV = left ventricle; LA = left atrium; RV = right ventricle.


The HeartWare LVAD is a third-generation, centrifugal, continuous-flow device1 (Fig. 2) that was approved by the Food and Drug Administration in November 2012 as bridge to transplant therapy. It has 3 basic components: a short metallic inflow cannula inserted into the left ventricular apex, a magnetically and hydrostatically levitated impeller attached to the inflow cannula, and an outflow graft that courses along the right anterolateral cardiac border and is anastomosed to the ascending aorta.2 The device is implanted within the pericardial space, obviating the need for an abdominal pocket.2 It also has been used off-label for right and biventricular support.3,4

Figure 2
Figure 2:
HeartWare left ventricular assist device. The inflow cannula is secured in the left ventricle apex and connected directly to an impeller. An outflow graft courses along the right anterolateral cardiac border and is anastomosed to the ascending aorta. Image reproduced with permission from HeartWare International.

TEE has been instrumental in the intraoperative evaluation of ventricular assist devices, their function, and position of the cannula and has been described previously.5 In this case, the postimplantation TEE revealed a prominent artifact inline with the HeartWare LVAD impeller that degraded both color and spectral Doppler, making it impossible to assess the velocity of flow into the inflow cannula, evaluate for the presence of turbulence or cannula obstruction, or perform Doppler evaluation of the mitral valve when the valve was inline with the impeller.

This artifact, previously described as the waterfall artifact,6 is the color flow version of an old b-mode artifact, referred to as a Herbie, described by DeMaria et al. in 1980.7 As in b-mode, energy from a previous ultrasonic transmission contaminates the echo from the current transmission, which results in the artifact appearing to originate above its true source. In color flow imaging, the Doppler shift is determined by use of a packet of transmissions in the same direction. Because of the nature of the materials used in the HeartWare LVAD device, the amplitude of the echo can be quite large and can reverberate for a considerable duration after the first contact, much like a bell ringing after being struck. If the second transmission of the color flow packet was to occur before this reverberation died down, the ultrasound system would associate this echo as originating from a shallow depth corresponding to the second transmission (Fig. 3).

Figure 3
Figure 3:
Schematic depiction of the HeartWare artifact. After first transmission of the color flow packet, the echo from the HeartWare left ventricular assist device (LVAD) reverberates for a considerable duration. If the second transmission occurs during the reverberation, the ultrasound system would associate this echo as originating from a shallower depth corresponding to the second transmission.

The mosaic color pattern associated with the HeartWare LVAD impeller, which appears similar to that of regurgitant flow, is caused by the chaotic nature of reverberation. This reverberation will vary significantly from one transmission to the next because of the different angular position of the impeller at the time of insonification. Standard wall filtering will bias these chaotic echoes toward the Nyquist velocities observed in the attached clips.

To verify this interpretation of the mechanism, we were able to vary the location of the artifact and to eliminate the artifact by varying both the velocity scale and the color box depth. Note that the color box depth corresponds to the deepest location of the color flow region of interest. At typical system settings, the LVAD device was observed at a depth of approximately 12 cm. The scale was set to ±57.2 cm/s, the color frequency was set to 4.4 MHz, and the color box depth was set to 12 cm, which corresponds to a pulse repetition frequency of 4.9 KHz and a pulse repetition interval (PRI) of 204 microseconds. Because this PRI corresponds to a transmit depth of 15.7 cm, and because the chaotic impeller reverberation would exceed this depth, the artifact was observed by the second transmission at the shallow depth of 1.5 cm. In a second case, the color box depth was extended to 17 cm, and the scale was reduced to ±23.1 cm/s at 4.4 MHz. Here, the PRI was 379 microseconds, with a corresponding transmit depth of 29 cm (depth at which the next color packet transmission occurs). Because the impeller reverberation had died out by this depth, no artifact was seen (Supplemental Digital Content 2, Video 2,

This artifact only appears when the impeller is inline with the Doppler beam. Off-axis views in which the impeller device is not inline with the Doppler beam eliminate this artifact, allowing interrogation of the mitral valve. This is most easily and consistently obtained using the transgastric 2-chamber or long-axis approach (Supplemental Digital Content 3, Video 3, or a foreshortened midesophageal view. Zooming into the mitral valve to exclude the inflow cannula and impeller device from the imaging sector does not eliminate the artifact because the impeller remains inline with the Doppler beam and thus, despite being off the screen, the interfering signal is still picked up by the TEE probe.

Even when the impeller is inline with the Doppler beam, the artifact can be reduced or eliminated by resizing the color region of interest to extend the color box depth. Often, this step may require increasing the sector depth first beyond what is appropriate for visualizing the heart and LVAD device but is the most robust way of extending the duration of the PRI and of delaying the next transmission beyond the reverberation. Simply decreasing the velocity scale, which also extends the PRI, does not always work because most commercial ultrasonography instruments use a technique referred to as interleaving, such that when the scale gets slow enough, they will fire adjacent color flow lines in the available dead time. These adjacent lines will then pick up the artifact.

The CW Doppler continuously transmits and receives; it will have neither the same range discrimination nor the same issues of range aliasing, as seen by pulsed-wave Doppler or color flow imaging. The CW Doppler artifact (Fig. 1B) likely results as a direct observation of the HeartWare impellor. Because the impellor provides an echo significantly stronger than normal red blood cells, and because the impellor is moving perpendicular to the insonifying ultrasound beam, one would expect a very strong and broad spectral response centered about the 0 velocity.

Intraoperative TEE is a valuable tool in the assessment of LVADs such as the HeartWare. However, it is important to know the unique characteristics of the device being evaluated, because ventricular assist device–induced artifacts may be mistaken for turbulent or partially obstructed flow.

Clinician’s Key Teaching Points

By Nikolaos J. Skubas, MD, Kent H. Rehfeldt, MD, and Martin J. London, MD

  • In cases of end-stage cardiac failure, a left ventricular assist device (LVAD) can provide hemodynamic support until the time of cardiac transplantation. The HeartWare LVAD has a metallic inflow cannula (surgically inserted in the left ventricular apex) with a centrifugal impeller pump attached to it. This pump provides continuous blood flow, via an outflow cannula, to the ascending aorta.
  • Transesophageal echocardiography is used to evaluate position of the cannula and their respective blood flow. With 2-dimensional transesophageal echocardiography imaging in the midesophageal 4-chamber or long-axis views, the inflow cannula should be oriented away from the interventricular septum toward the lateral left ventricular walls. If the blood flow is unobstructed, spectral Doppler imaging should show a blood velocity in the inflow cannula <2 m/s and in the outflow cannula about 1.8 m/s.
  • In this case of a HeartWare LVAD insertion, there was velocity aliasing at the properly oriented LVAD inflow cannula. The flow was impossible to evaluate, even at maximal Doppler scale (±8 m/s). With color Doppler, a mosaic of blood velocities occupied the entire sector, proximal as well as distal to the inflow cannula and the impeller, making examination of the mitral valve impossible in the midesophageal views.
  • Because of its proximity to the inflow cannula, the high-speed rotating impeller of the HeartWare LVAD degrades the Doppler signal when the impeller and the Doppler beam are aligned. Contrary to the conventional reverberation artifact, which occurs distal to a reflector, this artifact occurs proximal and distal to the reflector. Alternate transgastric views that remove the impeller from the imaging sector permit the evaluation of the mitral valve apparatus.


Name: Agnieszka Lesicka, MD.

Contribution: This author helped write the manuscript.

Attestation: Agnieszka Lesicka approved the final manuscript.

Conflicts of Interest: The author has no conflicts of interest to declare.

Name: Jared W. Feinman, MD.

Contribution: This author helped write the manuscript.

Attestation: Jared W. Feinman approved the final manuscript.

Conflicts of Interest: The author has no conflicts of interest to declare.

Name: Karl Thiele, MS.

Contribution: This author helped write the manuscript.

Attestation: Karl Thiele approved the final manuscript.

Conflicts of Interest: Karl Thiele works for Philips Medical Systems as a principal scientist in echocardiography.

Name: Michael N. Andrawes, MD.

Contribution: This author helped write the manuscript.

Attestation: Michael N. Andrawes approved the final manuscript.

Conflicts of Interest: The author has no conflicts of interest to declare.

This manuscript was handled by: Martin J. London, MD.


1. Strueber M, O’Driscoll G, Jansz P, Khaghani A, Levy WC, Wieselthaler GMHeartWare Investigators. . Multicenter evaluation of an intrapericardial left ventricular assist system. J Am Coll Cardiol. 2011;57:1375–82
2. Larose JA, Tamez D, Ashenuga M, Reyes C. Design concepts and principle of operation of the HeartWare ventricular assist system. ASAIO J. 2010;56:285–9
3. Hetzer R, Krabatsch T, Stepanenko A, Hennig E, Potapov EV. Long-term biventricular support with the HeartWare implantable continuous flow pump. J Heart Lung Transplant. 2010;29:822–4
4. Stulak JM, Griffith KE, Nicklas JM, Pagani FD. The use of the HeartWare HVAD for long-term right ventricular support after implantation of the HeartMate II device. J Thorac Cardiovasc Surg. 2011;142:e140–2
5. Chumnanvej S, Wood MJ, MacGillivray TE, Melo MF. Perioperative echocardiographic examination for ventricular assist device implantation. Anesth Analg. 2007;105:583–601
6. Shah NR, Cevik C, Hernandez A, Gregoric ID, Frazier OH, Stainback RF. Transthoracic echocardiography of the HeartWare left ventricular assist device. J Card Fail. 2012;18:745–8
7. DeMaria AN, Bommer W, Joye JA, Mason DT. Cross-sectional echocardiography: physical principles, anatomic planes, limitations and pitfalls. Am J Cardiol. 1980;46:1097–108

Supplemental Digital Content

© 2015 International Anesthesia Research Society