Secondary Logo

Journal Logo

Cardiovascular Anesthesiology: Echo Rounds

Coronary Artery to Left Ventricle Fistulas After Septal Myectomy

Haddy, Steven MD

Author Information
doi: 10.1213/ANE.0000000000000530

A patient presented with dynamic left ventricular (LV) outflow tract (LVOT) obstruction (LVOTO) following mitral valve repair. The resting peak outflow tract gradient was 100 mm Hg by transthoracic echocardiogram. Because the LVOTO and symptoms were refractory to medical management, the patient was scheduled for septal myectomy.

An extensive myectomy was performed though an aortotomy on cardiopulmonary bypass. After release of the aortic cross-clamp, before restoration of spontaneous rhythm, a constant jet was noted draining into the LV at the site of myectomy in the midesophageal 5-chamber view (Video, Supplemental Digital Content 1, Upon separation from cardiopulmonary bypass, the jet was noted to appear only in diastole (Fig. 1; Video, Supplemental Digital Content 2, In this clinical context, a coronary artery–LV fistula was suspected, probably involving the first septal perforating artery. LV contractility was evaluated with particular attention to the area of myocardium perfused by the left anterior descending (LAD) coronary artery, and no new wall motion abnormalities were seen. Careful interrogation of the septum disclosed no ventricular septal defect (VSD). The case was concluded without incident.

Figure 1
Figure 1:
Midesophageal long-axis views in (A) systole and (B) diastole demonstrating flow only in diastole consistent with a coronary to left ventricle fistula (red arrow). Although the electrocardiogram is not visible, timing in the cardiac cycle is indicated by the position of the mitral valve (blue arrow): (A) closed in systole and (B) open in diastole. LVOT = left ventricular outflow tract.

The pathophysiology of dynamic LVOTO is now recognized to be more complex than previously thought.1 Myectomy for relief of LVOTO may be performed alone, or together with procedures on the mitral valve and/or papillary muscles. Removal of myocardium leaves an area of muscle not covered by endocardium with the cut ends of the vessels exposed to the ventricular cavity. Depending on their size and perfusion, these may be seen as small jets emptying into the LV cavity from the myocardium during diastole. The jets are most commonly visualized in the midesophageal 5-chamber and long-axis views but may be seen in any view that demonstrates the LVOT or septum.

Although fistulas are usually easily distinguished from aortic insufficiency or paravalvular leak by clinical context and location, confusion can arise. Aortic paravalvular leaks occur in diastole and can be traced to an area outside of the sewing ring where they may display proximal flow convergence. They tend to be eccentric and often track along the ventricular wall. Fistulas typically arise more proximal in the LVOT (Fig. 2; Video [both taken from another patient], Supplemental Digital Content 3,, are traceable to the muscular septum, and are usually directed into the ventricular cavity.

Figure 2
Figure 2:
Modified midesophageal commissural view from another patient after myectomy demonstrating small jets on either side of the myectomy site (red arrows) in the proximal outflow tract, and the anterior leaflet of the mitral valve (yellow arrow). The blue arrow points to the annuloplasty ring.

Iatrogenic VSDs must always be excluded after myectomy. Septal defects occur after myectomy in up to 2% of cases,2 so a thorough examination of the postmyectomy septum is imperative. The midesophageal 4- and 5-chamber, long-axis, and right ventricular inflow–outflow views, as well as the transgastric basal and midpapillary short-axis views, are interrogated with and without color flow Doppler, searching for systolic transseptal flow.3 This flow is most commonly left to right due to the greater left-sided pressures. However, the shunt may be right to left early in systole if right ventricular pacing causes right ventricular pressure to transiently exceed LV pressure.4 Pulse-wave Doppler may be used to determine the direction of flow if the color flow signal is unclear. Determining whether the jet occurs in systole or diastole (although potentially difficult with tachycardia) is critical. Fistulas and aortic insufficiency are diastolic, whereas a VSD’s primary signal is systolic and flow should be seen in both the LV and right ventricular cavities, often with upstream flow convergence. In this regard, the electrocardiogram timing, as well as slow-motion and step-frame viewing, can be helpful. Driven by systolic LV pressure, VSDs display higher velocities than fistulas or perivalvular leaks, which are driven by diastolic pressure.

One study reported coronary to LV fistulas occurring in 23% of patients after myectomy,5 with most closing spontaneously within several months on transthoracic echocardiogram follow-up. Both myocardial thickness and elevated LV cavity pressure probably contribute to the tendency of these fistulas not to progress over time. Potential complications from a large, persistent fistula include myocardial ischemia in the area perfused by the coronary artery distal to the fistula via a “steal” mechanism, but this is rare.5 Sgalambro et al.5 describe a patient manifesting symptoms of congestive heart failure and decreased contractility of the LV apex after a myectomy that resulted in a persistent fistula from a large first septal perforating coronary artery. After percutaneous placement of a covered stent in the LAD coronary artery, the symptoms and wall motion abnormalities improved.

The area of hypertrophic obstruction is usually perfused by the first septal perforating branch of the LAD, and this corresponds to the area wherein we demonstrated a fistula. Obviously, this cannot be known with certainty without angiographic confirmation. Perfusion of the basal septum can be quite variable, and the first septal perforating coronary artery may supply areas remote from the septum.6 This has been demonstrated during contrast echo-guided alcohol septal ablations.7

In conclusion, these images demonstrate a common finding after septal myectomy. Generally, the only clinical relevance is that it not be confused with pathology requiring acute intervention; however, large fistulas may need to be addressed if symptoms arise. E

Clinician’s Key Teaching Points

By Kent H. Rehfeldt, MD, Roman M. Sniecinski, MD, and Martin J. London, MD

  • Transesophageal echocardiography (TEE) is often used during septal myectomy, a procedure performed to relieve left ventricular (LV) outflow tract (LVOT) obstruction refractory to medical management. In about one-fourth of cases, small coronary artery branches, typically from the first septal perforator, are transected. These small coronary artery-to-LV fistulas are seen on color flow Doppler as small jets entering the LVOT, which disappear during systole when the heart is contracting. Unlike aortic regurgitation, the direction of these jets is usually perpendicular, not parallel, to the LVOT.
  • A less common, but more serious, complication of septal myectomy is creation of a perimembranous ventricular septal defect (VSD). While the direction of the color flow Doppler jet is perpendicular to the LVOT, unlike a transected coronary artery branch, a VSD is characterized by left-to-right flow seen during systole.
  • In this case, TEE pulsed wave Doppler and color M-mode imaging in the midesophageal aortic valve long-axis view were used to distinguish between transected coronary vessel and VSD based on the direction and timing of the color flow jet. The observed color flow signal was toward the transducer in diastole, effectively excluding iatrogenic VSD.
  • After septal myectomy, the interventricular septum should be carefully interrogated. Echocardiographers should be careful to distinguish the benign finding of a transected coronary branch from the more serious complication of an iatrogenic VSD.


Name: Steven Haddy, MD.

Contribution: This author obtained the images and wrote the manuscript.

Attestation: Steven Haddy attests to the content of this manuscript.

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


1. Hymel BJ, Townsley MM. Echocardiographic assessment of systolic anterior motion of the mitral valve. Anesth Analg. 2014;118:1197–201
2. Sherrid MV, Chaudhry FA, Swistel DG. Obstructive hypertrophic cardiomyopathy: echocardiography, pathophysiology, and the continuing evolution of surgery for obstruction. Ann Thorac Surg. 2003;75:620–32
3. Hudson CC, Hudson JK. The interventricular septum: measurement and motion. Anesth Analg. 2013;116:788–92
4. Kurokawa S, Taneoka M, Imai H, Baba H, Nomura M. Transesophageal echocardiography detection of undiagnosed multiple muscular ventricular septal defects with alteration of shunt flow by right ventricular pacing after an arterial switch operation in a neonate. Anesth Analg. 2011;113:233–5
5. Sgalambro A, Olivotto I, Rossi A, Nistri S, Baldini K, Baldi M, Stefano P, Antoniucci D, Garbini F, Cecchi F, Yacoub MH. Prevalence and clinical significance of acquired left coronary artery fistulas after surgical myectomy in patients with hypertrophic cardiomyopathy. J Thorac Cardiovasc Surg. 2010;140:1046–52
6. Singh M, Edwards WD, Holmes DR Jr, Tajil AJ, Nishimura RA. Anatomy of the first septal perforating artery: a study with implications for ablation therapy for hypertrophic cardiomyopathy. Mayo Clin Proc. 2001;76:799–802
7. Hsieh TC, Patel K. Echocardiography-guided alcohol septal ablation for hypertrophic obstructive cardiomyopathy. Anesth Analg. 2011;113:44–6
© 2015 International Anesthesia Research Society