History
An 82-year-old woman who had a myocardial infarction and a coronary artery bypass graft (CABG) in 1997 developed progressive dyspnea and limitation of exercise tolerance to 200 yards (New York Heart Association [NYHA] class III) in 2010. After the diagnosis of systolic heart failure, she was started on medical therapy, but developed bronchoconstriction with beta blockers and a cough on angiotensin-converting enzyme (ACE) inhibitors, and she could tolerate only low doses of an angiotensin receptor blocker. Her medical history included permanent atrial fibrillation and peripheral vascular disease.
Comments
The patient was symptomatic from systolic heart failure and intolerant to medical therapy.
Current Medications
The patient was taking clopidogrel 75 mg daily, furosemide 40 mg daily, losartan 50 mg daily, rosuvastatin 10 mg daily, and isosorbide mononitrate XL 60 mg daily.
Comments
The patient experienced bronchoconstriction while on beta blockers, developed a cough on ACE inhibitors, and was intolerant of doses of losartan higher than 50 mg daily.
Current Symptoms
The patient was in NYHA class III. Her exercise capacity was limited at 200 yards by dyspnea, and she had no angina.
Comments
Dyspnea is the main symptom. The patient rarely experienced intermittent claudication.
Physical Examination
Electrocardiogram
Findings
The electrocardiogram showed atrial fibrillation with a ventricular response of 97 bpm and a QRS duration of 134 ms (Figure 37-1). There was fragmentation of the QRS complex in leads II, III, aVF and V3, with no evidence of a bundle branch block.
Comments
The patient had a high ventricular rate at rest, in the background of intolerance to beta blockers. The QRS duration is in keeping with electrical dyssynchrony. The high ventricular rate raises the possibility of tachyarrhythmia-related left ventricular dysfunction.
Chest Radiograph
Findings
The chest radiograph revealed an increased cardiothoracic ratio, no evidence of pulmonary edema, and sternotomy wires (Figure 37-2).
Comments
The findings on chest radiography were in keeping with heart failure without pulmonary edema.
Echocardiogram
Findings
The echocardiogram showed global left ventricular hypokinesia, inferior akinesis, myocardial thinning at the apex, a pseudospherical left ventricle with severely impaired left ventricular function (LVEF of 29% using Simpson’s method), and biatrial dilation (Figure 37-3, A).
FIGURE 37-2 Pre-implant chest radiograph.
Comments
The findings on echocardiography were in keeping with ischemic cardiomyopathy.
Findings
The transaortic peak gradient on the echocardiogram was 17 mm Hg, and the mean gradient was 14 mm Hg (see Figure 37-3, B).
FIGURE 37-3 Pre-implant transthoracic echocardiogram showing an apical. A, Apical four-chamber view. B, Continuous wave Doppler image through aortic valve. C, Color Doppler image through mitral valve.
Comments
The gradient was consistent with mild aortic stenosis (see Figure 37-3, B). In the background of severely impaired left ventricular function, the severity of aortic stenosis may be underestimated.
Findings
The mitral regurgitant jet occupied 36% of the left atrial area on the four-chamber view. The proximal isovelocity hemispheric surface area radius was 0.4 cm and the effective regurgitant orifice was 0.2 cm2. The valvular leaflets were intact, and dilation of the mitral valve ring and biatrial dilation were present.
Comments
The findings are consistent with moderate mitral regurgitation. The presence of inferior hypokinesis and a previous myocardial infarction raises the possibility of ischemic mitral regurgitation resulting from involvement of the posterior papillary muscle (mitral valve apparatus). Functional dilatation of the mitral valve ring also contributes to the degree of regurgitation.
Coronary Sinus Venography
Right anterior oblique (RAO) and left anterior oblique (LAO) fluoroscopic coronary sinus venography was performed at the time of cardiac resynchronization therapy (CRT) device implantation, showing veins that might be considered options for left ventricular lead deployment (Figure 37-4, A).
Findings
Option 1 in Figure 37-4 is a small-caliber, posterolateral vein draining the mid-lateral segment. This vein has a stenosis in its proximal, tortuous portion (see Figure 37-4, insert, white arrow). Option 2 is a posterolateral vein draining the mid-lateral segment. Option 3 is a small-caliber, anterolateral vein draining the basal anterior segment.
Comments
Option 1 in Figure 37-4 appears to be a reasonable candidate vein for left ventricular lead deployment. A venoplasty could be performed in the proximal portion and the vein then straightened using a buddy wire technique. A small-caliber left ventricular lead could then be deployed distally, in the mid-lateral segment.
FIGURE 37-4 Coronary sinus venography at implantation. A, Right anterior oblique projection. B, Left anterior oblique projection.
Feature Tracking Cardiovascular Magnetic Resonance Imaging
Findings
Feature tracking of the basal slice shows that the latest contracting segments in the lateral free left ventricular wall are the basal posterior and the basal inferior segments (Figure 37-5, A, red arrows ). Peak circumferential strain in these segments, however, is less than 10%, suggesting myocardial scar. The basal lateral segment appears to not be scarred (peak circumferential strain, −15.81%) and also contracts late (time to peak strain, 414 ms). Note that the earliest contracting segments with a high peak circumferential strain are not scarred (mid-septal and mid-anteroseptal).
Comments
The basal lateral segment appears to be an appropriate target for left ventricular lead deployment.
Findings
Within the mid-cavity the latest contracting segment in the lateral free left ventricular wall is the mid-inferior (see Figure 37-5, B, red arrow). The peak circumferential strain in this segment, however, is less than 10%, suggesting myocardial scar. The remaining left ventricular free wall segments do not appear scarred, and the latest contracting segments are the mid-anterior and mid-lateral (time to peak circumferential strain of 338 ms for both). The mid-anterior and the mid-lateral segments therefore do not appear to be scarred and are late contracting (see Figure 37-5, B, green arrows). Note that the earliest contracting segments with a high peak circumferential strain are not scarred (mid-septal and mid-anteroseptal).
Comments
The mid-anterior and the mid-lateral segments (Figure 37-5, B, green arrows) appear to be appropriate targets for left ventricular lead deployment
Findings
At the apex the latest contracting segments in the lateral free left ventricular wall are the apical lateral and the apical inferior segments (see Figure 37-5, B, red arrow and green arrow respectively). The apical lateral segment, however, is likely to be scarred (peak circumferential strain <10%). The apical inferior segment appears not to be scarred (peak circumferential strain, −23.05%) and contracts late. Note that the earliest contracting segment with a high peak circumferential strain is the apical septal segment.
Comments
The apical inferior segment appears to be an appropriate target for left ventricular lead deployment.
FIGURE 37-5 Feature-tracking cardiovascular magnetic resonance imaging of the left ventricular short axis stack at the (A) basal level, (B) mid-cavity level, and C, apical level.
Late Gadolinium Enhancement Cardiovascular Magnetic Resonance Imaging
Findings
The left hand panels of Figure 37-6 show basal, mid-cavity, and apical short-axis late gadolinium-enhancement images of the left ventricle. The relevant segments for left ventricular lead deployment, that is, those on the left ventricular free wall, delineated by endocardial, epicardial, and segmental boundaries, are shown in the center panels. In these images, the myocardial scar appears white and the viable myocardium appears black. The tables on the right describe whether scar is present and its pattern. Subendocardial scar is defined as a scar with less than 50% transmurality, whereas a transural scar is defined as that with a transmurality of 51% or greater.
Comments
The only segments that do not appear scarred on late gadolinium-enhanced cardiovascular magnetic resonance (LGE-CMR) are the basal anterior and basal lateral segments. The apical inferior segment, which had a peak circumferential strain of greater than 10% (−23.05%), also appeared scarred on LGE-CMR. This illustrates that the presence of circumferential strain does not necessarily equate to the absence of scar.
Focused Clinical Questions and Discussion Points
Question
Is this patient likely to benefit from CRT?
FIGURE 37-7 Anteroposterior fluoroscopic view of the final left ventricular lead position.
FIGURE 37-6 Short-axis, late gadolinium enhancement cardiovascular magnetic resonance.
Discussion
This patient was in permanent atrial fibrillation. The major outcome trials of CRT, however, have included only patients in sinus rhythm. Notwithstanding, some studies have shown that CRT in the context of atrial fibrillation improves symptoms.7 Other studies have suggested that CRT in patients in atrial fibrillation is effective only after atrioventricular junction ablation.2 These findings are encouraging for patients with heart failure and atrial fibrillation, who account for 10% to 25% of patients with heart failure in NYHA class II to III and up to 50% in patients in NYHA class IV. This patient also had a prolonged QRS duration (134 ms) and severely impaired left ventricular function as a result of ischemic cardiomyopathy. It is on this basis that the decision was made to proceed to proceed to implant a CRT defibrillator (CRT-D).
Question
What is the best left ventricular lead position in this patient?
Discussion
Fluoroscopy remains the standard imaging modality for guiding left ventricular lead deployment. As shown at coronary sinus venography (see Figure 37-4), several candidate coronary veins were apparent. As expected after CABG, these veins were generally of small caliber. A posterolateral vein (see Figure 37-4, option 1) had a tortuous, proximal segment with a subocclusion that was apparent on the RAO view. This stenosis could be repaired with venoplasty, even in the tortuous segment. After angioplasty, the vein could be straightened using a buddy wire technique. This could allow deployment of a small-caliber (perhaps 4 French) left ventricular lead. A more cranial posterolateral vein (see Figure 37-4, option 2) also is an option. This vein drains the mid-lateral segment.
An anterolateral vein (see Figure 37-4, option 3) also may be appropriate, on the basis of fluoroscopy. This vein, however, overlies the basal anterior segment, which may be regarded as an inappropriate site for left ventricular lead deployment. In this respect, early CRT studies suggested that the lateral free wall is a better left ventricular pacing site than a more anterior position.1,3 These findings make mechanical sense, because it is the lateral wall that is typically activated late in the context of a left bundle branch block. Importantly, clinical studies have failed to show superiority of pacing in lateral or posterolateral sites. In a retrospective study of 567 consecutive patients, a posterolateral position (2 to 5 o’clock on the LAO fluoroscopic view) was not associated with a better clinical outcome or echocardiographic response than other positions.6
On the basis of the factors discussed previously, most current CRT implanters would be satisfied with options 1 and 2 (lateral and posterior veins) (see Figure 37-4). Fewer implanters would choose option 3 (anterolateral vein), particularly because it is a vein of very small caliber.
Question
Should we choose a left ventricular lead position over a late-contracting segment?
Discussion
Single-center echocardiographic studies using tissue Doppler imaging, tissue synchronization imaging, three-dimensional echocardiography, and speckle-tracking echocardiography have shown that a better response to CRT can be achieved if the left ventricular lead is deployed in the area of latest contraction (presumed latest activation).5
In this case, we have used feature-tracking cardiovascular magnetic resonance (FT-CMR) for the quantification of myocardial strain. This is a new CMR technique that has been validated against myocardial tagging4 and uses the same principle as speckle-tracking echocardiography for the quantification of myocardial motion. On the basis of latest contraction segments, we could choose the targets described in the following section.
Basal Segments
As shown in Figure 37-5, A, the basal anterior segment contracts earliest (time to peak systolic circumferential strain, 338 ms), whereas the basal inferior (time to peak systolic circumferential strain, 489 ms) and basal posterior (time to peak systolic circumferential strain, 452 ms) contract the latest. The amplitude of circumferential strain of less than 10% in these segments raises the possibility of myocardial scarring. The only segments that appear not to be scarred are the basal anterior and basal lateral segments, and, of these, the latter contracts latest (time to peak systolic circumferential strain, 414 ms). One of the preferred targets for left ventricular lead deployment using FT-CMR is therefore the basal lateral segment. This site can be reached via the posterolateral vein (see Figure 37-4, option 2).
Mid-segments
As shown in Figure 37-5, B, the mid-inferior segments contract the latest (time to peak systolic circumferential strain, 367 ms). The low amplitude of strain (–8.85%), however, suggests that this segment is scarred. Of the remaining segments, the mid-anterior segment contracts relatively late (time to peak systolic circumferential strain, 338 ms), as does the mid-lateral segment (time to peak systolic circumferential strain, 310 ms). The latest contracting, nonscarred mid-segments are therefore the mid-anterior and the mid-lateral segments.
Apical Segments
As shown in Figure 37-5, C, both the apical lateral and apical inferior segments contract late, in contrast to the apical anterior segment (time to peak systolic circumferential strain of 349 ms and 285, respectively). The apical lateral segment appears scarred (amplitude of peak circumferential strain, −6.67%). Therefore the apical inferior segment is the latest contracting, nonscarred segment on the basis of FT-CMR.
Conclusions from Feature-Tracking Cardiovascular Magnetic Resonance
Candidate targets for left ventricular lead deployment on the basis of FT-CMR are the basal lateral segment, the mid-anterior and the mid-lateral segments, and the apical inferior segment.
Question
Should a left ventricular lead position be chosen over a nonscarred segment, assessed using LGE-CMR?
Discussion
Although myocardial strain can be used as a surrogate for myocardial scarring, the gold standard for detection and quantification of myocardial scarring in vivo is LGE-CMR. Several studies have shown that viability of the paced left ventricular segment also influences the outcome of CRT.
As shown in Figure 37-6, this patient had sustained an extensive myocardial infarction in the territory of the circumflex artery, which extended from the basal to the apical segments, involving the left ventricular free wall but sparing the basal anterior and basal lateral segments. Therefore, on the basis of LGE-CMR alone, the basal anterior and the basal lateral segments are appropriate targets for left ventricular lead deployment. Of these, the basal lateral segment contracts later than the basal anterior segment (see Figure 37-5, A). Using the combination of LGE-CMR and FT-CMR, the basal lateral segment is the latest contracting viable segment.
A shown in Figure 37-6, the mid-anterior and the mid-posterior segments were shown to have a peak circumferential strain of −24.78 and −10.71%, respectively, suggesting active contraction. Importantly, LGE-CMR shows that these segments have subendocardial and transmural scars, respectively. All of the apical segments were scarred.
Final Diagnosis
The final diagnosis is symptomatic systolic heart failure, with severely impaired left ventricular function despite maximum tolerated medical therapy. In addition, the patient had permanent atrial fibrillation.
Plan of Action
For this patient the plan was CRT-D implantation, targeting the basal lateral segment for left ventricular lead deployment. A future option is atrioventricular junction ablation.
Intervention
On the basis of the fluoroscopic images alone, an implanter might have been tempted to deploy the left ventricular lead in a posterolateral vein (see Figure 37-4, option 1). Although the vein was tortuous and stenosed in its proximal portion, this could have been surmounted by venoplasty and buddy wire technique. However, the segments subtended by this vein were the site of a transmural myocardial infarction. This was clear from the LGE-CMR images and FT-CMR strain analyses, which showed that the latest contracting, viable segment was the basal lateral segment.
To reach the basal lateral segment, the left ventricular lead was deployed in the posterolateral vein (see Figure 37-4, option 2). Anticipating some overlap of the pacing electrode with adjacent scar, a quadripolar lead (Quartet, St Jude Medical, St. Paul, Minn.) was selected. Figure 37-7 shows that the distal electrode of the quadripolar left ventricular lead overlies the mid-lateral segment which harbors a transmural scar on LGE-CMR. However, the mid-electrodes overlie the targeted basal lateral segment. At implantation, bipolar pacing thresholds in the distal poles were high (2.75-4.0 V, at a pulse duration of 0.5 ms). Bipolar pacing vectors incorporating the more proximal electrodes were associated with lower thresholds (1.5 to 3.0 V, at a pulse duration of 0.5 ms) but were associated with phrenic nerve stimulation. Pacing from the mid-electrodes (to the right ventricular coil) was associated with the lowest threshold (0.75 V at 0.5 ms, with phrenic nerve stimulation occurring at 4.0 V at 0.5 ms).
Outcome
At a 2-month follow-up examination, the patient was in NYHA class I.
Findings
The left ventricular pacing threshold was 0.75 (at 0.5 ms), and phrenic nerve stimulation occurred at 4 V (at 0.5 ms).
Comments
In this patient the combination of FT-CMR and LGE-CMR was used to identify the latest contracting viable segment over the left ventricular free wall in a patient undergoing CRT. Using fluoroscopy alone, some implanters might have selected the posterolateral vein, but this subtended a transmural myocardial scar.
Despite compelling evidence from observational studies, the utility of LGE-CMR has not been assessed by randomized controlled studies. Furthermore, FT-CMR is in its infancy as a technology for the assessment of cardiac dyssynchrony. Further studies are needed to determine whether the combination of these techniques, which can be applied to routine CMR scanning without additional acquisitions, can be used to guide CRT left ventricular lead deployment.
Selected References
1. Butter C., Auricchio A., Stellbrink C. et al. Effect of resynchronization therapy stimulation site on the systolic function of heart failure patients. Circulation. 2001;104:3026–3029.
2. Gasparini M., Auricchio A., Metra M. et al. Long-term survival in patients undergoing cardiac resynchronization therapy: the importance of performing atrio-ventricular junction ablation in patients with permanent atrial fibrillation. Eur Heart J. 2008;29:1644–1652.
3. Gold M.R., Auricchio A., Hummel J.D. et al. Comparison of stimulation sites within left ventricular veins on the acute hemodynamic effects of cardiac resynchronization therapy. Heart Rhythm. 2005;2:376–381.
4. Hor K.N., Gottliebson W.M., Carson C. et al. Comparison of magnetic resonance feature tracking for strain calculation with harmonic phase imaging analysis. Cardiovasc Imaging. 2010;3:144–151.
5. Khan F.Z., Virdee M.S., Palmer C.R. et al. Targeted left ventricular lead placement to guide cardiac resynchronization therapy: the TARGET study: a randomized, controlled trial. J Am Coll Cardiol. 2012;59:1509–1518.
6. Kronborg M.B., Albertsen A.E., Nielsen J.C. et al. Long-term clinical outcome and left ventricular lead position in cardiac resynchronization therapy. Europace. 2009;11:1177–1182.
7. Linde C., Leclercq C., Rex S. et al. Long-term benefits of biventricular pacing in congestive heart failure: results from the MUltisite STimulation In Cardiomyopathy (MUSTIC) study. J Am Coll Cardiol. 2002;40:111–118.
