OUP user menu

Impact of viability and scar tissue on response to cardiac resynchronization therapy in ischaemic heart failure patients

Claudia Ypenburg, Martin J. Schalij, Gabe B. Bleeker, Paul Steendijk, Eric Boersma, Petra Dibbets-Schneider, Marcel P.M. Stokkel, Ernst E. van der Wall, Jeroen J. Bax
DOI: http://dx.doi.org/10.1093/eurheartj/ehl379 33-41 First published online: 22 November 2006

Abstract

Aims At present, 20–30% of patients do not respond to cardiac resynchronization therapy (CRT). In this study, the relation between the extent of viable myocardium and scar tissue vs. response to CRT was evaluated. In addition, the presence of scar tissue in the left ventricular (LV) lead position was specifically related to response to CRT.

Methods and results A total of 51 consecutive patients with ischaemic heart failure and substantial LV dyssynchrony undergoing CRT were included. All patients underwent gated SPECT before CRT implantation to determine the extent of scar tissue and viable myocardium. Clinical and echocardiographic parameters were assessed at baseline and after 6 months of CRT. The results demonstrated direct relations between the response to CRT and the extent of viable myocardium and scar tissue. In addition, the 15 patients (29%) with transmural scar tissue (< 50% tracer activity) in the region of the LV pacing lead showed no improvement after 6 months of CRT.

Conclusion The extent of scar tissue and viable myocardium were directly related to the response to CRT. Furthermore, scar tissue in the LV pacing lead region may prohibit response to CRT. Evaluation for viability and scar tissue may be considered in the selection process for CRT.

  • Cardiac resynchronization therapy
  • Heart failure
  • Nuclear imaging
  • Viability

Introduction

Cardiac resynchronization therapy (CRT) is an accepted therapy for patients with advanced heart failure. The technique improves heart failure symptoms, exercise capacity, and left ventricular (LV) function, with a reduction in morbidity and mortality.14 The response to CRT, however, varies significantly among individuals, and different predictors of response to CRT have been proposed. One of the most important predictors of response is the presence of LV dyssynchrony.57 In addition, some studies have suggested that aetiology is related to response to CRT.810 Molhoek et al.11 have demonstrated that the percentage of responders was comparable between patients with ischaemic and non-ischaemic cardiomyopathy, but Woo et al.12 showed that the magnitude of benefit was larger in patients with non-ischaemic cardiomyopathy. In particular, the improvement in LV ejection fraction (LVEF) and the reduction in LV volumes after CRT was more outspoken in patients with non-ischaemic cardiomyopathy. Moreover, ischaemic aetiology of heart failure has been identified as a potential predictor of non-response.13 The response to CRT may thus be related to the extent of viable myocardium and inversely related to the extent of scar tissue.14 In addition, not only the extent of scar tissue may be important for the response to CRT, but also the location of scar tissue. Initial data suggested that scar tissue in the postero-lateral wall (as assessed by contrast-enhanced MRI) resulted in non-response to CRT.15 To further evaluate these issues, 51 consecutive patients with substantial LV dyssynchrony underwent nuclear imaging with technetium-99 m tetrofosmin to assess viability and scar tissue, prior to CRT implantation. The extent of viable myocardium and scar tissue were subsequently related to the response to CRT. Also, the influence of scar tissue in the region of the LV lead was evaluated.

Methods

Patients

The study population consisted of 51 consecutive patients with ischaemic cardiomyopathy who were scheduled for CRT implantation. Selection criteria for CRT were severe heart failure [New York Heart Association (NYHA) class III or IV], depressed LVEF (< 35%), and prolonged QRS duration (> 120 ms); patients had substantial LV dyssynchrony, averaging 86 ± 42 ms, as assessed by tissue Doppler imaging.7 Ischaemic aetiology was based on the presence of significant coronary artery disease (> 50% stenosis in one or more of the major epicardial coronary arteries) on coronary angiography and/or a history of myocardial infarction with ECG evidence, prior PCI or prior CABG.

The study protocol was as follows: before pacemaker implantation, a resting single-photon emission computed tomography (SPECT) with technetium-99 m tetrofosmin was performed to assess scar tissue and viable myocardium.16,17 Next, the clinical status was assessed and resting 2D transthoracic echocardiography was performed to measure LV volumes and LVEF. Clinical status and echocardiographic characteristics were re-assessed at 6 months follow-up.

SPECT with Technetium-99 m tetrosfosmin

SPECT imaging with technetium-99 m tetrofosmin (500 MBq, injected at rest) was performed using a triple head SPECT camera system (GCA 9300/HG, Toshiba Corp.) equipped with low-energy general-purpose collimators. Around the 140-KeV energy peak of technetium-99 m tetrofosmin, a 20% window was used. A total of 90 projections (step and shoot mode, 35 s per projection, imaging time 23 min) were obtained over a 360° circular orbit. Data were stored in a 64 × 64, 16-bit matrix. Data were displayed in polar map format (normalized to the maximum tracer activity) and analysed using a 17-segment model.18 Segmental tracer uptake was quantified and a segmental score was appointed with 0 = ≥75% of maximum tracer activity, 1 = 50–75% of maximum tracer activity, 2 = 25–50% of maximum tracer activity, and 3 = ≤25% of maximum tracer activity. Segments with tracer uptake ≥ 75% were considered normal (viable), segments with tracer uptake 50–75% were considered to contain some scar tissue (non-transmural infarction), and segments with tracer uptake < 50% were considered to have extensive scar tissue (transmural infarction). Summation of the segmental scores yielded the total scar score, with the higher scores indicating more extensive scar tissue. In addition to a general score, the presence of regional scar tissue was also evaluated in the area where the LV pacing lead was positioned (see in what follows). Regions with a tracer activity < 50% (score 2 or 3) were considered as having transmural scar formation in the LV pacing lead area.

Echocardiography

Patients were imaged in the left lateral decubitus position, using a commercially available system (Vingmed Vivid Seven, General Electric-Vingmed, Milwaukee, WI, USA). Images were obtained using a 3.5 MHz transducer, at a depth of 16 cm in the parasternal and apical views. Standard 2D and colour Doppler data, triggered to the QRS complex were saved in cine loop format. LV end-diastolic volume (LVEDV), LV end-systolic volume (LVESV), and LVEF were calculated from the conventional apical two- and four-chamber images, using the biplane Simpson's technique.19

The severity of mitral regurgitation was graded semi-quantitatively from colour-flow Doppler images in the apical four-chamber view. Mitral regurgitation was classified as: mild = 1 + (jet area/left atrial area < 10%), moderate = 2 + (jet area/left atrial area 10–20%), moderately severe = 3 + (jet area/left atrial area 20–45%), and severe = 4 + (jet area/left atrial area > 45%).20

Clinical evaluation

Clinical evaluation was performed at baseline and after 6 months of follow-up by an independent physician blinded to all other data. The clinical characteristics included NYHA class, quality-of-life score (using the Minnesota Living with Heart Failure questionnaire) and 6 min walking distance.21,22 In all patients, QRS duration was measured from the surface ECG using the widest QRS complex from the leads II, V1, and V6.

CRT implantation and LV lead position

A coronary sinus venogram was obtained using a balloon catheter, followed by the insertion of the LV pacing lead. An 8F guiding catheter was used to position the LV lead (Easytrak 4512-80, Guidant Corporation, St. Paul, MN, USA; or Attain-SD 4189, Medtronic Inc., Minneapolis, MN, USA) in the coronary sinus, preferably in the lateral or postero-lateral vein.23 The right atrial and ventricular leads were positioned conventionally. All leads were connected to a dual chamber biventricular ICD (Contak CD or Renewal, Guidant Corporation; or Insync III-CD or Marquis, Medtronic Inc.). After implant, the LV lead position was assessed from a chest X-ray. Using the frontal views (scored base, mid, or apex) and lateral views (scored anterior, lateral, or posterior) the LV lead locations were determined.24 To determine whether the LV lead was positioned in a region with scar tissue (score 2 or 3), the LV lead position was related to the 17-segment SPECT model.The preferred locations mid-lateral and mid-posterior corresponded with segments 5 and 11 and 4 and 10, respectively (Figure 1).

Figure 1

The LV pacing lead positions were positioned mid-lateral (n = 22), mid-posterior (n = 27), and mid-anterior (n = 2). These lead positions corresponded with segments 5 and 11, 4 and 10, and 1 and 7, respectively, in this 17-segment model. Adapted from Cerqueira et al.18

Statistical analysis

Most continuous variables had non-normal distribution (as evaluated by Kolmogorov–Smirnov tests). For reasons of uniformity, summary statistics for all continuous variables are therefore presented as medians together with the 25 and 75th percentiles. Categorical data are summarized as frequencies and percentages. Patients were classified as responders of CRT (‘responders’) if they improved at least one level in NYHA class and had an improvement of 25% in exercise distance after 6 months of CRT. The remaining patients, including those who died during the 6-month follow-up period, were classified as ‘non-responders’. Differences in baseline characteristics between responders and non-responders, and between patients with and without scar tissue in the target pacing region, were analysed using Wilcoxon–Mann–Whitney tests, χ2 tests, or Fisher's exact tests, as appropriate.

Changes that occurred over time in clinical (QRS duration), functional (NYHA class, 6 min exercise distance, quality-of-life), and echo (LVEF, LVEDV, LVESV) characteristics were studied by subtracting the baseline values from the values at 6 months follow-up for each individual patient. These changes were then summarized as median values (25 and 75th percentiles). Differences in changes between subgroups of patients were studied by applying the statistical tests that are mentioned earlier.

Linear regression analyses were performed to evaluate the relations between the overall scar score and the number of viable segments (results of SPECT imaging), and changes in LVEF, LVEDV, and LVESV (indicating the magnitude of LV reverse remodelling) after 6 months of CRT. We also aimed to study to what extent SPECT imaging results are associated with response to CRT. For this purpose, two multivariable logistic regression models were constructed, with overall scar score (first model) and the number of viable segments (second model) as main exposure, and age, QRS duration, the presence of LV dyssynchrony, LVEF, LVEDV, and LVESV as confounding factors. Adjusted odds ratios (ORs) with their corresponding 95% confidence intervals (CI) are reported. Note that it was not our intention to formally build an outcome prediction model. Still, all the continuous variables were assessed for linearity by entering a transformed variable in addition to the variable of interest. The natural logarithm and square transformations were used. A significant change in the − 2 log-likelihood was considered as a sign of non-linearity, otherwise the linearity assumption was accepted. All variables met the linearity assumption.

All statistical tests were two-sided. For all tests, a P-value < 0.05 was considered statistically significant.

Results

Patient characteristics

Baseline characteristics of the 51 consecutive patients (40 men, median age 68) included in this study are summarized in Table 1. A total of 43 (84%) patients were in NYHA class III before CRT implantation. The median QRS duration was 166 and the median LVEF was 22%. All patients received optimized medical therapy, if tolerated.

View this table:
Table 1

Baseline characteristics of the study population

Age (years)68 (62, 76)
Gender (M/F)40/11
NYHA class (I/II/III/IV)1/1/43/6
QRS duration (ms)166 (140, 188)
LBBB (%)39 (76)
LV dyssynchrony (ms)80 (70, 110)
Rhythm (SR/AF/paced)44/6/1
LVEF (%)22 (17, 28)
LVEDV (mL)223 (184, 280)
LVESV (mL)176 (139, 238)
Grade 3–4 + MR (%)10 (20)
Medication
 Diuretics (%)48 (94)
 ACE-inhibitors (%)43 (84)
 Beta-blockers (%)29 (57)
 Spironolactone (%)15 (29)
 Amiodarone (%)12 (24)
  • ACE, angiotensin-converting enzyme; AF, atrial fibrillation; LBBB, left bundle branch block; MR, mitral regurgitation; SR, sinus rhythm.

Device implantation was successful in all patients and no procedure-related complications were observed. The LV pacing lead was positioned in the mid-lateral region in 22 (43%) patients, in the mid-posterior region in 27 (53%), and in the mid-anterior region in two (4%) patients. Three patients died of worsening heart failure before the 6-month follow-up evaluation.

Clinical and echocardiographic improvement after CRT

After 6 months of CRT, 27 patients improved one NYHA functional class and three patients improved two NYHA functional classes (McNemar test P < 0.001). The quality-of-life score decreased from 37 (25, 48) to 18 (10, 32) (P < 0.001). In addition, a significant increase in 6 min walking distance was seen [from 300 (220, 400) m to 368 (328, 455) m, P < 0.001]. The LVEF showed a modest improvement from 22 (17, 28)% to 28 (22, 34)% (P < 0.001). Significant reverse remodelling was observed at 6-month follow-up, as evidenced by a decrease in LVEDV from 223 (184, 280) mL at baseline to 199 (164, 245) mL (P < 0.001) after 6 months of CRT. Similarly, LVESV decreased from 176 (139, 238) mL to 138 (107, 185) mL (P < 0.001).

Viability and scar score

A total of 867 segments were evaluated, with 476 (55%) classified as normal or viable (score 0), 128 (15%) having non-transmural scar (score 1), and 263 (30%) having transmural scar (95 with score 2 and 168 with score 3). The median number of normal viable segments was 10 (7, 12). A median scar score of 15 (7, 25) per patient was determined.

Changes in LVEF, LVEDV, and LVESV after 6 months of CRT were significantly correlated with the number of viable segments at baseline (Figure 2). Changes in LV function and LV dimensions were also associated with the total scar score (Figure 3).

Figure 2

Relationship between (A) the number of viable segments and the absolute change in LVEF, (B) the relative change in LVEDV, and (C) the relative change in LVESV after 6 months of CRT.

Figure 3

Relationship between (A) the total scar score and the absolute change in LVEF, (B) the relative change in LVEDV, and (C) the relative change in LVESV after 6 months of CRT.

Transmural scar tissue in the LV pacing target region

Fifteen patients (29%) had transmural scar tissue (score 2 or 3) in the region where the LV pacing lead was positioned (Figure 1). We observed significant differences between patients with and without scar tissue in the target region in QRS duration, the frequency of left bundle branch block, LV dyssynchrony, the number of viable segments, and total scar score (Table 2). No differences were observed in LV function and LV volumes.

View this table:
Table 2

Comparison of patients with and without transmural scar in the LV pacing lead region

Transmural scar (n = 15)No transmural scar (n = 36)P-value
Baseline clinical characteristics
 Age (years)68 (61, 71)68 (62, 77)0.5
 Gender (M/F)13/227/90.5
 QRS duration (ms)132 (116, 166)176 (148, 190)0.002
 LBBB (%)7 (47)32 (89)0.003
 LV dyssynchrony (ms)70 (20, 100)90 (73, 120)0.030
 Rhythm (SR/AF/paced)14/1/030/5/10.8
 Grade 3–4 + MR (%)3 (20)7 (19)1.0
 Number of viable segments7 (3, 10)11 (8, 13)0.001
 Total scar score26 (15, 32)11 (5, 19)< 0.001
Functional characteristics
 NYHA class
  Baseline (I/II/III/IV)0/0/13/21/1/30/41.0
  Follow-up (I/II/III/IV)0/3/9/24/25/5/0
  Δ (−2/ − 1/0/1)a0/3/10/14/24/6/0< 0.001
 6-MWT (m)
  Baseline340 (190, 410)287 (230, 380)0.5
  Follow-up360 (240, 420)393 (340, 500)
  Δa5 (−40, 90)120 (50, 159)0.026
 QOL score
  Baseline37 (25, 48)38 (25, 49)0.8
  Follow-up31 (21, 40)15 (8, 24)
  Δa− 7 (−13, 3)− 17 (−6, − 25)0.015
Echo characteristics
 LVEF (%)
  Baseline22 (17, 30)22 (17, 28)0.7
  Follow-up22 (19, 28)30 (26, 36)
  Δa−2 (−4, 2)7 (4, 14)0.002
 LVEDV (ml)
  Baseline212 (173, 270)227 (187, 299)0.4
  Follow-up225 (200, 263)186 (156, 230)
  Δa13 (−9, 39)−25 (−11, −65)< 0.001
 LVESV (ml)
  Baseline176 (135, 215)174 (139, 251)0.5
  Follow-up170 (134, 219)133 (102, 159)
  Δa8 (1, 36)− 50 (−24, −71)< 0.001
  • 6-MWT, 6 min walk test; QOL, quality-of-life score.

  • aFollow-up minus baseline value.

Immediately after implantation of the CRT device, the QRS duration was reduced from 176 (148, 190) ms to 155 (141, 165) ms in patients without scar tissue in the target region. In the patients with scar tissue, QRS duration increased from 132 (116, 166) ms to 168 (133, 176) ms. The difference in median change between the two groups was statistically significant [−21 (−41, 1) vs. 7 (−4, 48) ms, P = 0.001].

In patients without scar tissue in the target region NYHA class, 6 min walking distance, quality-of-life score, LV function, and LV dimensions had improved at 6-month follow-up compared with the baseline value. No improvement was seen in patients without scar tissue (Table 2). The differences in median change between the two groups were statistically significant for all these variables.

Of note, the overall scar score was significantly higher in patients with than in those without scar tissue in the LV pacing lead region (median 26 vs. 11, P < 0.001, Table 2). A difference was also observed in the number of viable segments (median 7 vs. 11, P = 0.001). The relations between viability and improvement in LVEF and LV volumes after CRT were maintained when patients with scar tissue in the target region of the LV pacing lead were excluded (Figure 4A–C).

Figure 4

Relationship in patients without scar tissue in the LV pacing region between (A) the number of viable segments and the absolute change in LVEF, (B) the relative change in LVEDV, and (C) the relative change in LVESV after 6 months of CRT.

Responders and non-responders

At 6-month follow-up, 27 (53%) patients were classified as responders to CRT (the remaining 21 patients were classified as non-responders). We observed statistically significant differences between responders and non-responders in QRS duration, LV dyssynchrony, the number of viable segments, the total scar score, the presence of scar tissue in the LV pacing target region, and LVEDV at baseline (Table 3).

View this table:
Table 3

Comparison of characteristics of responders and non-responders at baseline and after 6 months of CRT

Responders (n = 27)Non-responders (n = 24)P-value
Baseline clinical characteristics
 Age (years)68 (62, 77)69 (62, 76)0.7
 Gender (M/F)21/619/51.0
 QRS duration (ms)176 (150, 190)145 (119, 180)0.011
 LBBB (%)23 (85)16 (67)0.2
 LV dyssynchrony (ms)90 (80, 120)70 (20, 100)0.003
 Rhythm (SR/AF/paced)24/2/120/4/00.4
 Grade 3–4 + MR (%)4 (15)6 (25)0.5
 Number of viable segments12 (11, 13)8 (5, 10)< 0.001
 Total scar score10 (5, 15)25 (15, 28)< 0.001
 Scar in LV pacing region (%)3 (11)12 (50)0.005
Functional characteristics
 NYHA class
  Baseline (I/II/III/IV)0/1/25/11/0/18/50.07
  Follow-up (I/II/III/IV)3/24/0/01/4/14/2
  Δ (−2/ − 1/0/1)a3/24/0/01/3/16/1< 0.001
 6-MWT (m)
  Baseline280 (220, 340)343 (230, 405)0.2
  Follow-up420 (350, 500)340 (240, 380)
  Δa140 (120, 170)− 30 (−50, 30)< 0.001
 QOL score
  Baseline37 (24, 50)38 (27, 47)1.0
  Follow-up12 (5, 18)32 (24, 44)
  Δa− 21 (−15, −40)2 (−7, 5)< 0.001
Echo characteristics
 LVEF (%)
  Baseline22 (17, 28)22 (17, 31)1.0
  Follow-up31 (26, 36)23 (19, 29)
  Δa9 (5, 16)− 2 (−4, 4)< 0.001
 LVEDV (ml)
  Baseline249 (188, 326)204 (178, 248)0.048
  Follow-up198 (164, 233)204 (164, 251)
  Δa− 45 (−74, −12)0 (−13, 23)< 0.001
 LVESV (ml)
  Baseline190 (153, 259)157 (133, 219)0.1
  Follow-up137 (103, 164)138 (113, 195)
  Δa56 (−74, − 27)3 (−12, 17)< 0.001
  • aFollow-up minus baseline value.

After implantation of the CRT device, QRS duration in responders decreased from 176 (150, 190) to 154 (140, 166) ms. QRS duration in non-responders, however, increased from 145 (131, 176) to 159 (131, 176) ms. The difference in median QRS duration change between responders and non-responders was statistically significant [−23 (−40, − 2) vs. 5 (−15, 40), P = 0.007]. Six months after implantation, quality-of-life score, LV function, and LV dimensions had improved in responders of CRT compared with their baseline values, whereas no improvement was found in non-responders. The differences in median change between responders and non-responders were statistically significant for all these variables.

The overall scar score on SPECT was significantly lower in responders when compared with non-responders (Table 3, Figure 5). Responders also had higher numbers of viable segments, compared with non-responders. Vice-versa, the overall scar score and the number of viable segments were strongly related to the probability of being a responder of CRT. Among the patients with an overall scar score below the median,15 only 33% were classified as responder compared with 88% in those with a value above the median. Similarly, in patients with less than 10 viable segments (median value), only 29% were responders vs. 80% in patient with 10 or more viable segments. After adjustment for multiple confounders, a higher scar score remained associated with a lower probability of response (adjusted OR 0.89 per point and 95% CI 0.81–0.98, P = 0.017); a higher number of viable segments remained associated with a higher probability of response (adjusted OR 1.36 per segment and 95% CI 1.04–1.77, P = 0.013).

Figure 5

(A) SPECT imaging in a responder and (B) non-responder patient. As demonstrated in the short axis, vertical long axis, and horizontal long axis of the LV, the extent of viable myocardium is larger in the responder (total scar score 5, number of viable segments 13) compared with the non-responder patient (total scar score 29, number of viable segments 4). See supplementary material online for a colour version of this figure.

Discussion

The current findings illustrate that extensive scar tissue is frequently present in patients with ischaemic cardiomyopathy and substantial LV dyssynchrony. In these patients, response to CRT is directly related to the extent of viable myocardium and inversely related to the extent of scar tissue. In addition, the location of the scar tissue is important; patients with extensive scar tissue in the region where the LV lead is positioned do not respond to CRT. Consequently, the group of non-responders thus consisted mainly of patients with scar tissue in the region of the LV lead or with viable tissue in the region of the LV lead but with extensive scar tissue.

Viability and scar tissue vs. response to CRT

The observation that 20–30% of patients do not respond to CRT has resulted in a search for factors that may predict response. Various studies have recently demonstrated the value of LV dyssynchrony for prediction of response to CRT.5,7,25 It has also been suggested that the extent of scar tissue on the one hand and the extent of viable myocardium on the other hand are important for response to CRT. At present, only one study has systematically evaluated the relation between viability and the response to CRT.14 In 21 patients with ischaemic cardiomyopathy (mean LVEF 21 ± 5%), contrast echocardiography was performed before pacemaker implantation. With the use of contrast echocardiography, segmental perfusion was evaluated and a perfusion score index was derived, reflecting the extent of viable myocardium. The perfusion score index was directly related to the change in LVEF as assessed immediately after CRT. Similarly, the perfusion score index was also related to LV reverse remodelling (indicated by LV end-diastolic dimension) at 6 months after CRT. The current observations are in line with these findings; the extent of viability (expressed as the number of viable segments) was linearly related to the increase in LVEF, and the decrease in LVESV and LVEDV assessed at 6-months follow-up (Figure 2A–C). These findings are not surprising since one could anticipate that a substantial amount of viable myocardium is needed for improvement in systolic LV function after CRT.

In the present study, nuclear imaging with technetium-99 m tetrofosmin SPECT was used to assess viability. SPECT imaging with technetium-99 m labelled tracers has extensively been used for assessment of viability, and tracer uptake is dependent on a combination of intact perfusion, cell membrane, and mitochondrial integrity.26 However, viability assessment in patients with left bundle branch block may be affected by partial volume effects, particularly in the septum with asynchronous regional wall thickening. Still, SPECT imaging may be an ideal technique for assessment in these patients, since a resting SPECT is sufficient and the technique is widely available.

In addition to viability, the extent of scar tissue is also important for response to CRT, as reflected in the inverse relation between the extent of scar tissue and the change in LVEF and LV volumes (Figure 3A–C). Substantial improvement in LVEF was rare in patients with extensive scar tissue (Figure 3A), reflecting that too much scar tissue does not permit recovery of systolic LV function after CRT. In particular, when the scar score exceeded 15 (median value), the response rate to CRT was only 12%. Future studies are needed to specifically elucidate how much scar tissue and/or viable myocardium is needed to result in improvement of LVEF after CRT.

Scar tissue in the region of the LV pacing lead vs. response to CRT

Besides the extent of scar tissue, the location of scar tissue is also important. In a recent case report,27 it was demonstrated that acute infarction in the region where the LV lead was positioned resulted in acute loss of response to CRT. Moreover, it was demonstrated with contrast-enhanced MRI that transmural scar tissue in the region of the LV pacing lead prohibited response to CRT.28 The findings in the current study are in line with these observations. The patients with a transmural scar on technetium-99 m tetrofosmin SPECT did not improve in LV function, did not show reverse remodelling, and did not improve in clinical characteristics. These observations suggest that extensive scar tissue in the region of the LV pacing lead results in inadequate pacing with no response to CRT.

In the current study (in patients with severe ischaemic LV dysfunction), 15 (29%) patients had transmural scar tissue in the region of the LV pacing lead. Of interest, another recent observational study (with 91 patients with ischaemic cardiomyopathy and QRS > 120 ms) reported a similar percentage of patients with scar tissue in the infero-lateral wall, which is the target region for LV lead positioning.29 Accordingly, evaluation for the extent and location of scar tissue may be considered in the selection process for CRT to avoid non-response. Larger studies are needed to confirm our findings and to fully elucidate the clinical relevance of viability and scar tissue for response to CRT.

Of note, adjustments in V–V intervals were not evaluated in the present study. V–V optimization may be of particular importance in patients with ischaemic cardiomyopathy and scar tissue. This issue needs further study in future trials.

Responders vs. non-responders

Multivariable analysis revealed that a higher scar score was associated with a lower probability of response, and vice-versa, a higher number of viable segments was associated with a higher probability of response. Importantly, both LV dyssynchrony (as assessed with TDI) and the extent of scar tissue (as assessed with SPECT imaging) are of value in the prediction of response to CRT.

Of note, the non-responder rate in our study was quite high when compared with several clinical trials. We feel that this may be due to the fact that we only included patients with previous myocardial infarction, who appeared to have a worse outcome after CRT than patients with a non-ischaemic cardiomyopathy.12,13

Conclusion

Transmural scar formation is frequently observed in patients with ischaemic cardiomyopathy. A higher number of viable segments at baseline was associated with a higher probability of response; vice-versa, a higher total scar score was associated with a lower probability of response. In addition, transmural scar tissue in the region of the LV pacing lead may prohibit response. Evaluation for viability and scar tissue may be considered in the selection process for CRT.

Supplementary material

Supplementary material is available at European Heart Journal online.

Acknowledgements

This work was supported by the Dutch Heart Foundation, grant no: 2002B109.

Conflict of interest: none declared.

References

View Abstract