Skip main navigation

Segment Length in Cine Strain Analysis Predicts Cardiac Resynchronization Therapy Outcome Beyond Current Guidelines

Originally publishedhttps://doi.org/10.1161/CIRCIMAGING.120.012350Circulation: Cardiovascular Imaging. 2021;14

Abstract

Background:

Patients with a class I recommendation for cardiac resynchronization therapy (CRT) are likely to benefit, but the effect of CRT in class II patients is more heterogeneous and additional selection parameters are needed in this group. The recently validated segment length in cine strain analysis of the septum (SLICE-ESSsep) measurement on cardiac magnetic resonance cine imaging predicts left ventricular functional recovery after CRT but its prognostic value is unknown. This study sought to evaluate the prognostic value of SLICE-ESSsep for clinical outcome after CRT.

Methods:

Two hundred eighteen patients with a left bundle branch block or intraventricular conduction delay and a class I or class II indication for CRT who underwent preimplantation cardiovascular magnetic resonance examination were enrolled. SLICE-ESSsep was manually measured on standard cardiovascular magnetic resonance cine imaging. The primary combined end point was all-cause mortality, left ventricular assist device, or heart transplantation. Secondary end points were (1) appropriate implantable cardioverter defibrillator therapy and (2) heart failure hospitalization.

Results:

Two-thirds (65%) of patients had a positive SLICE-ESSsep ≥0.9% (ie, systolic septal stretching). During a median follow-up of 3.8 years, 66 (30%) patients reached the primary end point. Patients with positive SLICE-ESSsep were at lower risk to reach the primary end point (hazard ratio 0.36; P<0.001) and heart failure hospitalization (hazard ratio 0.41; P=0.019), but not for implantable cardioverter defibrillator therapy (hazard ratio, 0.66; P=0.272). Clinical outcome of class II patients with a positive ESSsep was similar to those of class I patients (hazard ratio, 1.38 [95% CI, 0.66–2.88]; P=0.396).

Conclusions:

Strain assessment of the septum (SLICE-ESSsep) provides a prognostic measure for clinical outcome after CRT. Detection of a positive SLICE-ESSsep in patients with a class II indication predicts improved CRT outcome similar to those with a class I indication whereas SLICE-ESSsep negative patients have poor prognosis after CRT implantation.

CLINICAL PERSPECTIVE

Patients with a class I recommendation for cardiac resynchronization therapy are likely to benefit from the therapy, but the effect in class II patients is more heterogeneous and additional selection parameters are needed for identification of suitable cardiac resynchronization therapy candidates from within this group. Our work adds to the accumulating evidence that the septum plays a crucial role in the translation of the detrimental effects of electrical conduction delay on cardiac mechanics and consequently, predicts benefit from resynchronization. The segment length in cine strain technique is a simple tool that can be performed on standard cardiovascular magnetic resonance cine images without the need for specialized commercial software tools. This technique provides the implanting physician rapid quantification of the mechanical consequences of electrical conduction disease. The segment length in cine strain analysis of the septum (ESSsep) strongly predicts clinical outcome after cardiac resynchronization therapy beyond current guideline criteria and is particularly useful in patients with a class II indication for cardiac resynchronization therapy in whom benefit of the therapy is less consistent.

See Editorial by Soman et al

Myocardial strain imaging provides information on cardiac mechanics and can reveal regional deformation abnormalities secondary to left bundle branch block (LBBB). In true LBBB, a typical pattern is observed where the late-activated lateral wall pushes the septum back during systole (ie, discoordination), resulting in reduced left ventricular (LV) pumping efficiency. Cardiac resynchronization therapy (CRT) aims to restore LV pump function by recoordinating contraction of asynchronous LV regions. Approximately one-third of the patients meeting ACCF/AHA/HRS criteria for CRT implantation have no benefit or demonstrate worsening heart failure (HF; nonresponder).1,2 Whereas patients with a class I indication for CRT (LBBB with QRS duration [QRSd] ≥150 ms) are likely to benefit from CRT, the effect in class II patients (LBBB with QRSd <150 ms; intraventricular conduction delay [IVCD]) is more heterogeneous.3 Myocardial strain imaging could potentially improve patient selection in class II patients through identification of dyssynchronous mechanical substrate. Multiple imaging techniques are presently available for measuring strain including cardiac magnetic resonance (CMR) tagging (CMR-TAG), CMR feature tracking (CMR-FT) and speckle tracking echocardiography; however, these require specific pulse-sequences to acquire additional images and/or specialized post-processing software tools. The segment length in cine (SLICE) measure of dyssynchrony can be performed on standard CMR cine images without the need for specialized commercial software tools but we used ImageJ (free available for download) instead. Previously, SLICE was validated against the gold standard CMR-TAG with good accuracy (ICC, 0.76) and reproducibility (intraobserver ICC, 0.94; interobserver ICC, 0.86).4 The subsequent study comparing different SLICE parameters showed that the end-systolic septal strain (ESSsep) parameter was most predictive of functional LV recovery after CRT.5 The present study aims to explore the value of SLICE-ESSsep for the prediction of clinical end points in a large population of patients undergoing CRT.

Methods

Study Population

Patients who underwent CMR examination within 12 months before CRT implantation between 2005 and 2017 were retrospectively included in this 2-center study (Amsterdam University Medical Center, location VU medical center, The Netherlands; Duke University Medical Center, Durham, NC). Inclusion criteria were NYHA class II-IV HF symptoms, LV ejection fraction (LVEF) ≤35% measured by CMR, QRSd ≥120 ms and complete set of CMR late gadolinium enhancement (LGE) images. The presence of LBBB morphology was assessed by the Strauss criteria.6 Patients who did not meet the Strauss criteria were classified as IVCD. Patients with LBBB with QRSd ≥150 ms fulfilled criteria for a class I indication for CRT whereas patients with LBBB with QRSd <150 ms or IVCD fulfilled criteria for a class II indication.2 Exclusion criteria were right bundle branch block and upgrade procedures from other implanted non-CRT devices. Clinical data were collected from medical records. The primary end point was a combined end point of all-cause mortality, LV assist device (LVAD) or heart transplantation (HTx). All-cause mortality data were retrieved from medical records, the municipal civil registry or the Social Security Death Index. All patients were included for the primary end point analysis. Secondary end points included (1) appropriate implantable cardioverter defibrillator (ICD) therapy and (2) HF hospitalization. Appropriate ICD therapy was defined as antitachycardia pacing or defibrillation shock to terminate ventricular arrhythmias. All patients implanted with a CRT device equipped with ICD function (CRT-D) who routinely underwent device interrogation at Amsterdam University Medical Center or Duke University Medical Center with regular intervals of 6 months were included for this secondary end point. HF hospitalization was defined as unplanned admission to the hospital for worsening HF. All patients who received outpatient follow-up at Amsterdam University Medical Center or Duke University Medical Center with regular intervals of 6 months were included for this end point. The institutional review committee of the Amsterdam University Medical Center, location Vrije Universiteit Amsterdam and Duke University Medical Center approved the study protocol. Informed consent was waived due to the retrospective nature of this study. The investigation conforms to the principles outlined in the Declaration of Helsinki. The data that support the findings of this study are available from the corresponding author upon reasonable request.

Image Acquisition: CMR Imaging

Patients underwent CMR examination at Amsterdam University Medical Center, location VU medical center (Amsterdam, the Netherlands) or at Duke Cardiovascular Magnetic Resonance Center (Durham, NC), on a clinical 1.5 Tesla MR scanner (Magnetom Sonata/Avanto, Siemens, Erlangen, Germany) with dedicated phased array cardiac receiver coil. Standard CMR cine images were acquired using a retrospectively ECG-gated balanced steady-state free-precession sequence during end-expiratory breath holding. A stack of 8 to 12 consecutive short-axis cine images was acquired covering the full LV. Typical imaging parameters were: slice thickness 5 mm, slice gap 5 mm, voxel size 1.3×1.6 mm and temporal resolution <50 ms. Subsequently, LV volumes and function were quantified as LV end-diastolic volume (LVEDV), LV end-systolic volume, and LVEF derived from the short-axis images using commercially available software on locally available workstations. Myocardial scar territory was assessed by LGE imaging, and infarct size was measured using the full width at half maximum method in the full stack of LGE images using dedicated offline software (QMass version 7.6, Medis, Leiden, The Netherlands).7 Septal scar localization and transmurality was measured as the maximal percentage hyperenhancement of the LV wall thickness in the basal- and mid-LV level of the anteroseptal and inferoseptal segments. Posterolateral scar was similarly measured at the basal- and mid-LV level of the posterolateral segments. Patients were classified as having ischemic cardiomyopathy in case of a history of myocardial infarction, percutaneous coronary intervention, or coronary artery bypass grafting accompanied by associated findings on LGE-CMR.

Post-Processing: SLICE Analysis

A detailed description of the SLICE post-processing steps has been published previously.4,5 In brief, the mid-LV slice position with short-axis cine images was selected (QMass version 7.6, Medis, Leiden, The Netherlands). Two endocardial anatomic landmarks (trabeculae) near the right ventricular insertion points, delimiting the septal segment, were chosen in the end-diastolic frame and checked for traceability to the end-systolic frame. End-diastolic was defined as the first R-wave triggered image and end-systolic was assessed in the 3-chamber view by aortic valve closure. Marks were placed perpendicular to the myocardium at the anatomic landmarks in each frame. Subsequently, marked cine images were exported to ImageJ to measure segment length between the marks over the myocardium midline in each phase. ESSsep was then calculated in % units by: ([end-systolic septum length−end-diastolic septum length]/end-diastolic septum length)×100 where negative numbers indicate shortening and positive numbers indicate stretching. Normally, the septum demonstrates systolic shortening (ESSsep negative) but in patients with true LBBB activation, a distinct pattern can be observed where the late-activated lateral wall pushes back the early-activated septum (ESSsep positive). SLICE-ESSsep analysis is illustrated in Figure 1.

Figure 1.

Figure 1. Segment length in cine (SLICE) strain measurement of the septum. Illustration of SLICE strain analysis in a cardiac resynchronization therapy candidate. First, endocardial anatomic landmarks near the right ventricular insertion points delimiting the septum are identified (red dotted lines). Second, segment length of the septum is measured between the marks over the myocardium midline (black dotted lines). End-diastolic segment length was 412 pixels and end-systolic segment length was 474 pixels. Subsequently, end-systolic septum strain (ESSsep) was calculated by: ([474–412]/412)×100=15% indicating systolic stretching of the septum in this case (ESSsep positive). ED indicates end diastole; and ES, end systole.

Statistics

The commercially available Statistical Package for Social Sciences software (IBM SPSS Statistics for Windows, Version 20.0. Armonk, NY) was used for statistical analyses. Continuous variables are expressed as mean±SD if normally distributed or median and interquartile range (IQR) otherwise. Categorical variables are presented as frequencies and percentages. The independent-samples t test and a Mann-Whitney U test were used to compare means and distributions of continuous variables between groups. Pearson and Spearman correlations were calculated as a measure of association depending on whether distribution was normal or not. Receiver operating characteristics curve analysis was used to find optimal ESSsep cut-off value by maximizing the Youden’s index. Time to occurrence of the combined end point was compared between groups using Kaplan-Meier analysis and a log-rank test. The association between baseline parameters and time to the combined end point was tested using univariable Cox regression analysis with hazard ratios (HRs) as effect-size. All variables with P<0.1 were then included in a multivariable analysis using a backward elimination procedure (P<0.05 as entry criterion and P>0.10 as removal criterion). To test the additional value of SLICE on top of standard CMR parameters, multivariable analysis was performed where SLICE was added to a model with the other CMR parameters. Last, to test the additional value of CMR imaging on top of clinical/guideline parameters, multivariable analysis was performed where independent CMR predictors were added to a model with clinical and ECG parameters. A P of <0.05 was considered statistically significant.

Results

After screening a total of 372 patients who underwent a preimplantation CMR examination, 32 patients were excluded from the analysis based on absence of LGE imaging, 26 patients based on a LVEF >35%, 41 patients because of right bundle branch block, 42 patients with QRSd <120 ms, 9 patients who were asymptomatic (NYHA class 1), and 4 patients because of missing data. As a result, 218 patients (35% women, median age 67 years) were included. A detailed description of the patient characteristics is given in the Table. Median time between CMR examination and CRT implantation was 40 (IQR, 5–112) days.

Table. Patient Characteristics

VariablesTotal group (n=218)ESSsep negative (n=77)ESSsep positive (n=141)P Value
Age, y67 (59–74)70 (60–77)65 (56–72)P=0.001
Sex (n, % female)77 (35%)20 (26%)57 (40%)P=0.033
NYHA class (n, %)
 Class II65 (30%)29 (38%)36 (26%)
 Class III149 (68%)47 (61%)102 (72%)
 Class IV4 (2%)1 (1%)3 (2%)P=0.167
Device
 ICD function (CRT-D) (n, %)210 (96%)72 (94%)138 (98%)P=0.101
 Quadripolar LV lead (n, %)45 (21%)17 (22%)28 (20%)P=0.699
Medical history (n, %)
 Atrial fibrillation42 (19%)22 (29%)20 (14%)P=0.010
 Diabetes49 (23%)22 (29%)27 (19%)P=0.111
 Ischemic cause69 (32%)36 (47%)33 (23%)P=0.001
Medication (n, %)
 Beta-blockers181 (83%)60 (78%)121 (86%)P=0.138
 Diuretics165 (76%)59 (77%)106 (75%)P=0.812
 ACE/ATII inhibitors191 (88%)66 (86%)125 (89%)P=0.529
 Aldosterone antagonist87 (40%)22 (29%)65 (46%)P=0.041
ECG
 QRS duration ≥150 ms (n, %)132 (61%)29 (38%)103 (73%)P<0.001
 LBBB (Strauss) (n, %)145 (66%)35 (46%)110 (78%)P<0.001
 Class I indication (n, %)108 (50%)20 (26%)88 (62%)P<0.001
Laboratory results
 Creatinine, μmol/L97 (77–118)97 (80–132)92 (77–115)P=0.163
CMR
 LVEDV, mL299±94292±111303±83P=0.389
 LVESV, mL234±88227±104238±77P=0.360
 LVEF, %23±724±722±6P=0.008
 Scar size (% LV mass)1.88 (0.00–7.93)5.82 (1.00–11.69)0.00 (0.00–5.17)P<0.001
 Septal scar (n, %)
  No scar127 (58%)33 (43%)94 (67%)
  <25% TM64 (29%)33 (43%)31 (22%)
  25%–50% TM18 (8%)9 (12%)9 (6%)
  >50% TM9 (4%)2 (3%)7 (5%)P=0.002
 Posterolateral scar (n, %)
  No scar171 (78%)49 (64%)122 (87%)
  <25% TM34 (16%)18 (23%)16 (11%)
  25%–50% TM8 (4%)6 (8%)2 (1%)
  >50% TM5 (2%)4 (5%)1 (1%)P<0.001
 SLICE ESSsep (%)2.75±5.45−2.65±3.195.70±3.97P<0.001

ACE indicates angiotensin-converting enzyme; ATII, Angiotensin II receptor blockers; CMR, cardiovascular magnetic resonance; CRT-D, cardiac resynchronization therapy with defibrillator function; ESSsep, end-systolic septal strain; ICD, implantable cardioverter defibrillator; LBBB, left bundle branch block; LV, left ventricular; LVEDV, left ventricular end-diastolic volume; LVEF, left ventricular ejection fraction; LVESV, left ventricular end-systolic volume; NYHA, New York Heart Association dyspnea class; SLICE, segment length in cine; and TM, transmurality.

Primary End Point Analysis

SLICE-derived ESSsep showed large variability between patients ranging between –14.9% and 19.7% (negative numbers indicate septal shortening and positive numbers indicate stretching). Over a median follow-up time of 3.8 (IQR, 1.7–5.9) years, 66 (30%) patients reached the combined end point of all-cause mortality (n=61), LVAD (n=3), or HTx (n=2). Univariable Cox regression analysis showed that ESSsep was significantly associated with the primary end point (HR, 0.89 per % [95% CI, 0.84–0.93]; P<0.001). Receiver operating characteristics curve analysis revealed an optimal cut-off value of 0.9% predictive for survival free of LVAD or HTx. Two-third of patients (65%) demonstrated ESSsep ≥0.9% (ESSsep positive) and the remaining patients ESSsep <0.9% (ESSsep negative). Comparing patient characteristics between groups, ESSsep positive patients were younger, more often female, less often had atrial fibrillation or ischemic etiology, more frequently used aldosterone antagonists, had wider QRS duration and more often LBBB ECG morphology, and demonstrated lower LVEF and less scar burden by CMR imaging (Table). Patients with a positive ESSsep demonstrated close to 3-fold lower likelihood of death, LVAD, or HTx after CRT implantation compared with ESSsep negative patients (HR, 0.36 [95% CI, 0.22–0.59]; P<0.001), illustrated in Figure 2.

Figure 2.

Figure 2. Kaplan-Meier curves primary end point. Occurrence of the primary end point (freedom from death, left ventricular assist device [LVAD] or heart transplantation [HTx]) differed significantly between ESSsep positive and negative patients (left diagram) also after adjustment for age, renal function, left ventricular end-diastolic volume, and septal scar (right diagram). Median event-free survival after cardiac resynchronization therapy implantation was 9.3 y for segment length in cine strain analysis of the septum (SLICE-ESSsep) positive patients and 4.6 y for SLICE-ESSsep negative patients. HR indicates hazard ratio. ESSsep indicates end-systolic septal strain; and SLICE, segment length in cine.

Secondary End Point Analysis

For the secondary end point of HF hospitalization, a total of 138 patients who received outpatient follow-up at Amsterdam University Medical Center or Duke University Medical Center with regular intervals of 6 months were included. Over a median follow-up time of 3.2 (IQR, 1.1–4.9) years, 28 (20%) patients were admitted to the hospital. Patients with a positive ESSsep showed a longer time to HF hospitalization after CRT compared with ESSsep negative patients (HR, 0.41 [95% CI, 0.20–0.87]; P=0.019), illustrated in Figure 3A. For the secondary end point of appropriate ICD therapy, a total of 177 patients implanted with a CRT-D device who routinely underwent device interrogation at Amsterdam University Medical Center or Duke University Medical Center with regular intervals of 6 months were included. Over a median follow-up time of 3.1 (IQR, 1.0–5.4) years, 30 (17%) patients received appropriate device therapy. Freedom from ICD therapy, however, did not differ between groups (HR, 0.66 [95% CI, 0.31–1.39]; P=0.272) as illustrated in Figure 3B.

Figure 3.

Figure 3. Kaplan-Meier curves secondary end points. Patients with a positive ESSsep experienced lower rates of heart failure (HF) hospitalization (left diagram), but no differences in appropriate implantable cardioverter defibrillator (ICD) therapy (right diagram) compared with ESSsep-negative patients. ESSsep indicates end-systolic septal strain.

Myocardial Scar Tissue

Myocardial scarring (ischemic cardiomyopathy or nonischemic cardiomyopathy pattern) was present in 129 (59%) of the patients with a median scar size of 6.9% of LV mass. ESSsep was the highest in patients without scar (ESSsep: 4.7±5.3%) and gradually decreased from patients with scar burden below the median (ESSsep: 2.5±5.2%) to patients with scar burden above the median (ESSsep: 0.2±4.9%), Figure 4A. Concurrently, global scar (none; below median; above median) was associated with higher likelihood of death, LVAD, or HTx after CRT implantation (HR, 1.78 [95% CI, 1.33–2.39]; P<0.001).

Figure 4.

Figure 4. Effects of myocardial scarring. Plot means with SE bars illustrating that increases in global scar size gradually reduced ESSsep (left diagram). With regionally scarring of the septum, ESSsep also reduced with 0% to 25% scar transmurality (TM) but paradoxically increased at higher scar burden (right diagram). ESSsep indicates end-systolic septal strain.

Myocardial scarring of the septum was present in 91 (42%) patients. ESSsep was highest in patients without septal scar (ESSsep: 4.0±5.6%) and showed lower values in patients with <25% transmurality (ESSsep: 0.6±5.3%). Interestingly, ESSsep paradoxically increased with higher scar transmurality from 25% to 50% transmurality (ESSsep: 1.2±2.9%) to >50% transmurality (ESSsep: 3.2±3.0%; Figure 4B). Septal scar (none; <25% transmurality; 25%–50% transmurality; >50% transmurality) was associated with higher probability of death, LVAD, or HTx after CRT implantation (HR, 1.51 [95% CI, 1.18–1.93]; P=0.001).

Posterolateral scar was present in 47 (22%) patients. ESSsep was the highest in patients without posterolateral scar (3.6%) and progressively decreased from patients with <25% transmurality (0.2%) to 25% to 50% transmurality (−1.5%) and >50% transmurality (−2.6%). Posterolateral scar (none; <25% transmurality; 25%–50% transmurality; >50% transmurality) was also associated with higher likelihood of the primary end point (HR, 1.40 [95% CI, 1.05–1.85]; P=0.022).

LV Lead Location

LV lead location was assessed in 147 (67%) cases based on available fluoroscopy and/or chest X-ray images. LV lead location was basal in 15 (10%), mid-LV in 101 (69%), and apical in 31 (21%) of cases. Apical LV lead location showed a trend toward higher risk for occurrence of the primary end point compared with nonapical positions (HR, 1.82 [95% CI, 0.97–3.42]; P=0.060). Furthermore, LV lead location was anterolateral in 16 (11%), lateral in 53 (36%), and posterolateral in 78 (53%) of cases. Posterolateral LV lead location may be preferred over other positions but also did not reach statistical significance (HR, 0.62 [95% CI, 0.35–1.11]; P=0.107). As LV lead location was incompletely recorded, this parameter was not included in multivariable modeling to prevent drop-out of patients.

Univariable and Multivariable Modeling

Univariable and multivariable Cox regression analysis for the primary end point is illustrated in Figure 5. Significant CMR predictors in univariable Cox regression included ESSsep, LVEDV, LV end-systolic volume, scar size, septal scar and posterolateral scar. Addition of ESSsep to a multivariable model with significant CMR predictors showed that ESSsep was independently related to the primary end point, together with septal scar and LVEDV. Significant clinical/guideline parameters in univariable Cox regression were age, sex, atrial fibrillation, HF cause, renal function, QRSd ≥150 ms, and QRS morphology. Addition of the independent CMR predictors (ESSsep; septal scar; and LVEDV) to a multivariable model with clinical/guideline parameters showed that all CMR predictors were independently related to the primary end point together with age and renal function.

Figure 5.

Figure 5. Multivariable risk modeling. Cox proportional hazard modeling for risk assessment of death, left ventricular assist device (LVAD) or heart transplantation (HTx) after cardiac resynchronization therapy. The relative predictive value of end-systolic septal strain (ESSsep) and other cardiovascular magnetic resonance (CMR) parameters are shown in the top diagram. Independent CMR predictors (left ventricular end-diastolic volume [LVEDV]; septal scar; ESSsep) are added to a multivariable model with clinical/ECG parameters as shown in the bottom diagram. An asterisk indicates that the parameter is not included in multivariable analysis (ie, nonsignificant). HR indicates hazard ratio; LBBB, left bundle branch block; LV, left ventricular; LVEF, left ventricular ejection fraction; LVESV, left ventricular end-systolic volume; NYHA, New York Heart Association dyspnea class; QRSd, QRS duration; and TM, transmurality.

Subgroup Analysis by Guideline Criteria

Half of the patients fulfilled the guideline criteria for a class I recommendation for CRT (LBBB with QRSd ≥150 ms). Within this subgroup, ESSsep positive (81%) compared with negative classification did not differentiate in likelihood of the primary outcome after CRT (HR, 0.83 [95% CI, 0.31–2.27]; P=0.728). Among patients with a LBBB or IVCD and a class II indication, however, those with a positive ESSsep (48%) demonstrated 4-fold better outcome compared with ESSsep-negative patients (HR, 0.23 [95% CI, 0.11–0.48]; P<0.001). Clinical outcome of class II patients with a positive ESSsep was similar to those of class I patients as a reference (HR, 1.38 [95% CI, 0.66–2.88]; P=0.396), see Figure 6.

Figure 6.

Figure 6. Additional value of segment length in cine (SLICE) strain analysis of the septum beyond current guideline criteria. Half of the patients fulfilled criteria for a class I indication for cardiac resynchronization therapy (CRT; left bundle branch block [LBBB] with QRS duration [QRSd] ≥150 ms) and showed good clinical outcome after CRT (irrespective of SLICE-ESSsep). Patients with a class II recommendation (LBBB with QRSd <150 ms; intraventricular conduction delay [IVCD]) underwent additional SLICE analysis and demonstrated systolic stretching of the septum (ESSsep positive) in half of the cases whereas this was lacking in the others (ESSsep negative). Class II ESSsep positive patients demonstrated improved CRT outcome similar to those with a class I indication whereas class II ESSsep-negative patients had poor survival after CRT. ESSsep indicates end-systolic septal strain; and SLICE, segment length in cine.

Discussion

The present study explores the role of SLICE strain analysis of the septum (SLICE-ESSsep) to predict clinical outcome after CRT using standard CMR cine imaging. The main findings of this study were that detection of systolic stretching of the septum (ie, positive ESSsep) was associated with improved clinical outcome after CRT implantation. Furthermore, the ESSsep measure was sensitive to the presence of global myocardial scarring which lowered strain values, corresponding with relatively poor outcome. Regional scarring of the septum, however, paradoxically increased ESSsep despite having a poor prognosis. Both ESSsep and septal scar were independent predictors of the primary end point, indicating that SLICE should be combined with LGE to exclude >50% septal scarring as the cause of septal stretching. Moreover, multivariable modeling showed that CMR predictors (ESSsep; septal scar; and LVEDV) provided incremental prognostic value beyond current guideline criteria (QRS morphology and duration). Detection of a positive ESSsep in patients with a class II indication predicts improved CRT outcome similar to those with a class I indication whereas SLICE-ESSsep negative patients have poor prognosis after CRT implantation.

LBBB and Septal Mechanics

In LBBB, early-activated regions in the septum start contracting first while the lateral wall prestretches. After the delayed activation of the lateral wall, reinforced contraction leads to passive stretching of the septum during ejection. This paradoxical septal movement (ie, discoordination) renders the LV pump function inefficient and subsequently further deterioration of LV function ultimately resulting in HF and premature death.8,9 Electrical resynchronization recruits myocardial work from the septum by converting systolic stretching into shortening, leading to a homogenized work distribution and improved cardiac pump efficiency.10–12 The amount of systolic stretching of the septum during LBBB therefore reflects the potential room for improvement after resynchronization. In line with this concept, strain parameters of the septum (such as systolic rebound stretch) have been linked to CRT outcome in prior studies by various groups.10,13–18 Previously, we compared different SLICE parameters and found ESSsep was most predictive of LV functional improvement after CRT.5 In addition, the ESSsep parameter showed consistent results when using other strain imaging techniques in relation with CRT response (CMR-TAG; CMR-FT; speckle tracking echocardiography).19 The present 2-center study extends the role of SLICE-ESSsep to the prediction of clinical end points in a large population of CRT candidates.

Septal Mechanics and Clinical Outcome After CRT

All CRT candidates in the present study had prolonged electrical activation with QRSd ≥120 ms. Septal discoordination, however, varied widely between patients with SLICE-ESSsep ranging between −14.9% (shortening) and 19.7% (stretching), indicating a fair amount of variation in mechanical discoordination between patients. Receiver operating characteristics-curve analysis revealed an optimal cutoff value of 0.9% for prediction of the primary end point, which is similar to the cutoff of 0.3% that we found previously to predict echocardiographic CRT response.5 Strain values around 0% indicate the tipping point where the septum makes no contribution to LV ejection or stroke volume and starts wasting myocardial work. Patients with inefficient septal mechanics (positive ESSsep) demonstrated better outcome after CRT (Figure 2), presumably by a mechanism of reversal of HF progression whereas no difference was observed in the incidence of ventricular arrhythmias (Figure 3). The latter may be explained by a recent study from Linhart et al20 showing that arrhythmic events depend on the presence of myocardial scar (59% of the patients in this study) irrespective of CRT response.

Myocardial Scar Tissue

Myocardial scarring is associated with worse outcomes after CRT, irrespective of HF etiology.21,22 Computer simulations by Leenders et al14 demonstrated that septal deformation in LBBB depends on both LV activation delay and LV contractility. Reducing global LV contractility (by global scarring) results in less septal discoordination with little systolic stretching, and consequently, smaller benefit from CRT. This was confirmed in our study as we found that global scar size reduced ESSsep and related to poor CRT outcome. Septal discoordination therefore provides integrated information on both intraventricular dyssnchrony and global LV scar size, 2 important determinants of CRT response. Gradual increases in regional scarring of the septum, however, paradoxically influenced ESSsep with increased strain values. A transmural myocardial infarction located at the septum may therefore influence septal strain assessment with positive strain values due to passive stretching (as seen with dyskinetic aneurysm), thus resembling strain patterns seen in patients with explicit discoordination. Septal scar, however, was associated with poor CRT outcome which is in line with previous reports.23–25 In multivariable analysis, both ESSsep and septal scar were independent predictors of the primary end point, indicating that SLICE should be combined with LGE to exclude >50% septal scarring as the cause of septal stretching.

Guideline Criteria

Present guideline criteria for CRT being QRS duration (≥150 versus 120–150 ms) and QRS morphology (LBBB versus IVCD) were found useful as they were both significant predictors of the primary end point (HR, 0.58 and 0.51, respectively).2 Moreover, patients with a class I indication for CRT (LBBB with QRSd ≥150 ms) showed excellent survival, without the need for ESSsep assessment (as the vast majority of class I patients were ESSsep positive). However, half of the population had a class II recommendation (LBBB with QRSd <150 ms or IVCD). Although CRT implantation is more controversial in these patients, a distinct subpopulation of patients seems to benefit from this therapy.3 This may be explained by the fact that only half of the class II patients were ESSsep positive. Additional strain analysis can be of value here to identify patients who may or may not benefit from CRT. Detection of a positive ESSsep predicted improved CRT outcome similar to those with a class I indication (Figure 6). In contrast, class II patients with a negative ESSsep had poor survival after CRT (although benefit from the therapy cannot be excluded since there was no control group without CRT). At present, CMR imaging is increasingly used to screen candidates by measuring LVEF combined with LGE imaging to guide LV lead placement.26,27 Additional SLICE analysis of the septum in patients with a class II recommendation for CRT could potentially improve diagnostic yield of CMR and improve patient selection for CRT. Assessment of SLICE-ESSsep requires 2 simple measurements (Figure 1) and can be performed in <10 minutes without the need for commercial software tools. Instead, images are exported to ImageJ (free available for download) making it a widely available tool.

Limitations

The major limitation of the study is the observational study design, which does not allow any conclusions on causal relationships, only associations. Despite the large sample size and the use of clinical end points, the results should therefore be interpreted with care. As patients were included based on available CMR examination, this may have caused a selection bias. Moreover, CMR imaging is not always available in clinical practice. Also there was no control group without CRT. Although ESSsep negative patients showed relatively poor survival after CRT implantation, benefit from the therapy cannot be excluded since patients may have had even worse outcome without CRT. To prevent over-optimism of the prediction model, results should be validated in an external cohort. Furthermore, thresholds derived within this specific CRT population should not be generalized to other populations. Last, SLICE relies on strain measures from in-plane motion of anatomic landmarks. For this reason, we could only analyze the mid-LV slice, since this plane is relatively motion independent (ie, to avoid through-plane motion). Nevertheless, variation in strain values between basal, mid- and apical LV segments are relatively small.

Conclusions

SLICE strain analysis of the septum (ESSsep) on standard CMR cine imaging is a simple tool that enables rapid quantification of the mechanical consequences of electrical conduction disease and strongly predicts clinical outcome after CRT beyond current guideline criteria. SLICE-ESSsep is particularly useful in patients with a class II indication for CRT in whom benefit of the therapy is less consistent.

Nonstandard Abbreviations and Acronyms

CMR

cardiovascular magnetic resonance

CRT

cardiac resynchronization therapy

HF

heart failure

ICD

implantable cardioverter defibrillator

IQR

interquartile range

IVCD

intraventricular conduction delay

LBBB

left bundle branch block

LGE

late gadolinium enhancement

LV

left ventricular

LVAD

LV assist device

LVEDV

LV end-diastolic volume

LVEF

LV ejection fraction

QRSd

QRS duration

SLICE

segment length in cine

Disclosures None

Footnotes

*C.P. Allaart and R. Nijveldt contributed equally.

For Sources of Funding and Disclosures, see page 617.

Correspondence to: Robin Nijveldt, MD, PhD, Department of Cardiology, Amsterdam University Medical Center, De Boelelaan 1117, 1081 HV Amsterdam, The Netherlands. Email

References

  • 1. Daubert C, Behar N, Martins RP, Mabo P, Leclercq C. Avoiding non-responders to cardiac resynchronization therapy: a practical guide.Eur Heart J. 2017; 38:1463–1472. doi: 10.1093/eurheartj/ehw270CrossrefMedlineGoogle Scholar
  • 2. Yancy CW, Jessup M, Bozkurt B, Butler J, Casey DE, Drazner MH, Fonarow GC, Geraci SA, Horwich T, Januzzi JL, et al; American College of Cardiology Foundation; American Heart Association Task Force on Practice Guidelines. 2013 ACCF/AHA guideline for the management of heart failure: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines.J Am Coll Cardiol. 2013; 62:e147–e239. doi: 10.1016/j.jacc.2013.05.019CrossrefMedlineGoogle Scholar
  • 3. Salden OAE, Vernooy K, van Stipdonk AMW, Cramer MJ, Prinzen FW, Meine M. Strategies to improve selection of patients without typical left bundle branch block for cardiac resynchronization therapy.JACC: Clinical Electrophysiology. 2020; 6:129–142. doi: 10.1016/j.jacep.2019.11.018CrossrefMedlineGoogle Scholar
  • 4. Zweerink A, Allaart CP, Kuijer JPA, Wu L, Beek AM, van de Ven PM, Meine M, Croisille P, Clarysse P, van Rossum AC, et al. Strain analysis in CRT candidates using the novel segment length in cine (SLICE) post-processing technique on standard CMR cine images.Eur Radiol. 2017; 27:5158–5168. doi: 10.1007/s00330-017-4890-0CrossrefMedlineGoogle Scholar
  • 5. Zweerink A, Nijveldt R, Braams NJ, Maass AH, Vernooy K, de Lange FJ, Meine M, Geelhoed B, Rienstra M, van Gelder IC, et al. Segment length in cine (SLICE) strain analysis: a practical approach to estimate potential benefit from cardiac resynchronization therapy.J Cardiovasc Magn Reson. 2021; 23:4. doi: 10.1186/s12968-020-00701-4CrossrefMedlineGoogle Scholar
  • 6. Strauss DG, Selvester RH, Wagner GS. Defining left bundle branch block in the era of cardiac resynchronization therapy.Am J Cardiol. 2011; 107:927–934. doi: 10.1016/j.amjcard.2010.11.010CrossrefMedlineGoogle Scholar
  • 7. Flett AS, Hasleton J, Cook C, Hausenloy D, Quarta G, Ariti C, Muthurangu V, Moon JC. Evaluation of techniques for the quantification of myocardial scar of differing etiology using cardiac magnetic resonance.JACC Cardiovasc Imaging. 2011; 4:150–156. doi: 10.1016/j.jcmg.2010.11.015CrossrefMedlineGoogle Scholar
  • 8. Hawkins NM, Wang D, McMurray JJ, Pfeffer MA, Swedberg K, Granger CB, Yusuf S, Pocock SJ, Ostergren J, Michelson EL, et al; CHARM Investigators and Committees. Prevalence and prognostic impact of bundle branch block in patients with heart failure: evidence from the CHARM programme.Eur J Heart Fail. 2007; 9:510–517. doi: 10.1016/j.ejheart.2006.11.006CrossrefMedlineGoogle Scholar
  • 9. Zweerink A, de Roest GJ, Wu L, Nijveldt R, de Cock CC, van Rossum AC, Allaart CP. Prediction of acute response to cardiac resynchronization therapy by means of the misbalance in regional left ventricular myocardial work.J Card Fail. 2016; 22:133–142. doi: 10.1016/j.cardfail.2015.10.020CrossrefMedlineGoogle Scholar
  • 10. De Boeck BW, Teske AJ, Meine M, Leenders GE, Cramer MJ, Prinzen FW, Doevendans PA. Septal rebound stretch reflects the functional substrate to cardiac resynchronization therapy and predicts volumetric and neurohormonal response.Eur J Heart Fail. 2009; 11:863–871. doi: 10.1093/eurjhf/hfp107CrossrefMedlineGoogle Scholar
  • 11. Prinzen FW, Vernooy K, De Boeck BW, DeBoeck BW, Delhaas T. Mechano-energetics of the asynchronous and resynchronized heart.Heart Fail Rev. 2011; 16:215–224. doi: 10.1007/s10741-010-9205-3CrossrefMedlineGoogle Scholar
  • 12. Russell K, Eriksen M, Aaberge L, Wilhelmsen N, Skulstad H, Gjesdal O, Edvardsen T, Smiseth OA. Assessment of wasted myocardial work: a novel method to quantify energy loss due to uncoordinated left ventricular contractions.Am J Physiol Heart Circ Physiol. 2013; 305:H996–1003. doi: 10.1152/ajpheart.00191.2013CrossrefMedlineGoogle Scholar
  • 13. Kirn B, Jansen A, Bracke F, van Gelder B, Arts T, Prinzen FW. Mechanical discoordination rather than dyssynchrony predicts reverse remodeling upon cardiac resynchronization.Am J Physiol Heart Circ Physiol. 2008; 295:H640–H646. doi: 10.1152/ajpheart.00106.2008CrossrefMedlineGoogle Scholar
  • 14. Leenders GE, Lumens J, Cramer MJ, De Boeck BW, Doevendans PA, Delhaas T, Prinzen FW. Septal deformation patterns delineate mechanical dyssynchrony and regional differences in contractility: analysis of patient data using a computer model.Circ Heart Fail. 2012; 5:87–96. doi: 10.1161/CIRCHEARTFAILURE.111.962704LinkGoogle Scholar
  • 15. Risum N, Tayal B, Hansen TF, Bruun NE, Jensen MT, Lauridsen TK, Saba S, Kisslo J, Gorcsan J, Sogaard P. Identification of typical left bundle branch block contraction by strain echocardiography is additive to electrocardiography in prediction of long-term outcome after cardiac resynchronization therapy.J Am Coll Cardiol. 2015; 66:631–641. doi: 10.1016/j.jacc.2015.06.020CrossrefMedlineGoogle Scholar
  • 16. Gorcsan J, Anderson CP, Tayal B, Sugahara M, Walmsley J, Starling RC, Lumens J. Systolic stretch characterizes the electromechanical substrate responsive to cardiac resynchronization therapy.JACC Cardiovasc Imaging. 2019; 12:1741–1752. doi: 10.1016/j.jcmg.2018.07.013CrossrefMedlineGoogle Scholar
  • 17. Salden OAE, Zweerink A, Wouters P, Allaart CP, Geelhoed B, de Lange FJ, Maass AH, Rienstra M, Vernooy K, Vos MA, et al. The value of septal rebound stretch analysis for the prediction of volumetric response to cardiac resynchronization therapy.Eur Heart J Cardiovasc Imaging. 2021; 22:37–45. doi: 10.1093/ehjci/jeaa190CrossrefMedlineGoogle Scholar
  • 18. Lumens J, Tayal B, Walmsley J, Delgado-Montero A, Huntjens PR, Schwartzman D, Althouse AD, Delhaas T, Prinzen FW, Gorcsan J. Differentiating electromechanical from non-electrical substrates of mechanical discoordination to identify responders to cardiac resynchronization therapy.Circ Cardiovasc Imaging. 2015; 8:e003744. doi: 10.1161/CIRCIMAGING.115.003744LinkGoogle Scholar
  • 19. Zweerink A, van Everdingen WM, Nijveldt R, Salden OAE, Meine M, Maass AH, Vernooy K, de Lange FJ, Vos MA, Croisille P, et al. Strain imaging to predict response to cardiac resynchronization therapy: a systematic comparison of strain parameters using multiple imaging techniques.ESC Heart Fail. 2018; 5:1130–1140. doi: 10.1002/ehf2.12335CrossrefMedlineGoogle Scholar
  • 20. Linhart M, Doltra A, Acosta J, Borràs R, Jáuregui B, Fernández-Armenta J, Anguera I, Bisbal F, Martí-Almor J, Tolosana JM, et al. Ventricular arrhythmia risk is associated with myocardial scar but not with response to cardiac resynchronization therapy.Europace. 2020; 22:1391–1400. doi: 10.1093/europace/euaa142CrossrefMedlineGoogle Scholar
  • 21. Leyva F, Foley PW, Chalil S, Ratib K, Smith RE, Prinzen F, Auricchio A. Cardiac resynchronization therapy guided by late gadolinium-enhancement cardiovascular magnetic resonance.J Cardiovasc Magn Reson. 2011; 13:29. doi: 10.1186/1532-429X-13-29CrossrefMedlineGoogle Scholar
  • 22. Adelstein EC, Althouse AD, Schwartzman D, Jain SK, Soman P, Saba S. Scar burden, not intraventricular conduction delay pattern, is associated with outcomes in ischemic cardiomyopathy patients receiving cardiac resynchronization therapy.Heart Rhythm. 2018; 15:1664–1672. doi: 10.1016/j.hrthm.2018.05.027CrossrefMedlineGoogle Scholar
  • 23. White JA, Yee R, Yuan X, Krahn A, Skanes A, Parker M, Klein G, Drangova M. Delayed enhancement magnetic resonance imaging predicts response to cardiac resynchronization therapy in patients with intraventricular dyssynchrony.J Am Coll Cardiol. 2006; 48:1953–1960. doi: 10.1016/j.jacc.2006.07.046CrossrefMedlineGoogle Scholar
  • 24. Duckett SG, Ginks M, Shetty A, Kirubakaran S, Bostock J, Kapetanakis S, Gill J, Carr-White G, Razavi R, Rinaldi CA. Adverse response to cardiac resynchronisation therapy in patients with septal scar on cardiac MRI preventing a septal right ventricular lead position.J Interv Card Electrophysiol. 2012; 33:151–160. doi: 10.1007/s10840-011-9630-9CrossrefMedlineGoogle Scholar
  • 25. Aalen JM, Donal E, Larsen CK, Duchenne J, Lederlin M, Cvijic M, Hubert A, Voros G, Leclercq C, Bogaert J, et al. Imaging predictors of response to cardiac resynchronization therapy: left ventricular work asymmetry by echocardiography and septal viability by cardiac magnetic resonance.Eur Heart J. 2020; 41:3813–3823. doi: 10.1093/eurheartj/ehaa603CrossrefMedlineGoogle Scholar
  • 26. Romano S, Judd RM, Kim RJ, Kim HW, Klem I, Heitner JF, Shah DJ, Jue J, White BE, Indorkar R, et al. Feature-tracking global longitudinal strain predicts death in a multicenter population of patients with ischemic and nonischemic dilated cardiomyopathy incremental to ejection fraction and late gadolinium enhancement.JACC Cardiovasc Imaging. 2018; 11:1419–1429. doi: 10.1016/j.jcmg.2017.10.024CrossrefMedlineGoogle Scholar
  • 27. Bilchick KC, Dimaano V, Wu KC, Helm RH, Weiss RG, Lima JA, Berger RD, Tomaselli GF, Bluemke DA, Halperin HR, et al. Cardiac magnetic resonance assessment of dyssynchrony and myocardial scar predicts function class improvement following cardiac resynchronization therapy.JACC Cardiovasc Imaging. 2008; 1:561–568. doi: 10.1016/j.jcmg.2008.04.013CrossrefMedlineGoogle Scholar