Prediction of Cardiac Resynchronization Therapy Response Using a Lead Placement Score Derived From 4-Dimensional Computed Tomography

Abstract

Background: Cardiac resynchronization therapy (CRT) is an effective treatment for patients with heart failure; however, 30% of patients do not respond to the treatment. We sought to derive patient-specific left ventricle maps of lead placement scores (LPS) that highlight target pacing lead sites for achieving a higher probability of CRT response.

Methods: Eighty-two subjects recruited for the ImagingCRT trial (Empiric Versus Imaging Guided Left Ventricular Lead Placement in Cardiac Resynchronization Therapy) were retrospectively analyzed. All 82 subjects had 2 contrast-enhanced full cardiac cycle 4-dimensional computed tomography scans: a baseline and a 6-month follow-up scan. CRT response was defined as a reduction in computed tomography-derived end-systolic volume ≥15%. Eight left ventricle features derived from the baseline scans were used to train a support vector machine via a bagging approach. An LPS map over the left ventricle was created for each subject as a linear combination of the support vector machine feature weights and the subject's own feature vector. Performance for distinguishing responders was performed on the original 82 subjects.

Results: Fifty-two (63%) subjects were responders. Subjects with an LPS≤Q₁ (lower-quartile) had a posttest probability of responding of 14% (3/21), while subjects with an LPS≥ Q₃ (upper-quartile) had a posttest probability of responding of 90% (19/21). Subjects with Q₁<LPS<Q₃ had a posttest probability of responding that was essentially unchanged from the pretest probability (75% versus 63%, P=0.2). An LPS threshold that maximized the geometric mean of true-negative and true-positive rates identified 26/30 of the nonresponders. The area under the curve of the receiver operating characteristic curve for identifying responders with an LPS threshold was 87%.

Conclusions: An LPS map was defined using 4-dimensional computed tomography-derived features of left ventricular mechanics. The LPS correlated with CRT response, reclassifying 25% of the subjects into low probability of response, 25% into high probability of response, and 50% unchanged. These encouraging results highlight the potential utility of 4-dimensional computed tomography in guiding patient selection for CRT. The present findings need verification in larger independent data sets and prospective trials.

Keywords: cardiac imaging techniques; cardiac resynchronization therapy; four-dimensional computed tomography; heart failure; heart function tests; support vector machine; ventricular function.

Fig. 1.. LV blood-volume segmentation.

(A) Axial contrast-enhanced cardiac CT image. (B) LV blood-volume segmentation overlaid in red. (C) 3D rendering of the LV blood-volume segmentation with the locations of the right (cyan) and the left (magenta) lead tips. The anterior wall is in view with the septum on the left and the lateral wall on the right. LV: left ventricle.

Fig. 2.. Flow diagram of the subject selection process.

Fig. 3.. Relationship between change in ESV at 6-months follow-up and lead placement score (LPS).

Distribution of LPS values for the lead locations in each subject as a function of the relative change in ESV between baseline and follow-up scans. The vertical dashed line corresponds to the LPS threshold (−0.2) that maximized the g-mean value for response prediction. The horizontal dashed line corresponds to the response definition of a relative reduction in ESV of 15%. ‘+’ represents a true responder and ‘o’ represents a true non-responder. LPS: lead placement score; ESV: end systolic volume.

Fig. 4.. Performance of the trained LPS model.

(A) Histogram of LPS for all 82 subjects. The non-responders are shown in red, and the responders are shown in blue. (B) ROC curve of the trained SVM model to predict responders as a function of changing LPS threshold. LPS: lead placement scores; ROC: receiver operating characteristic; AUC: area under curve; SVM: support vector machine.

Fig. 5.. Lead placement score (LPS) maps of CRT responders and non-responders.

(a-d) Four example subjects with the following information shown for each subject Lead tip locations: blue = right lead, red = left lead TOS: time to onset of shortening, in ms; PRS_CT: peak regional shortening; TPRS_CT: time to peak regional shortening, in ms; MSRS_CT: maximum pre-stretch of regional shortening; LPS map: lead placement score map as a polar map; 3D LPS map: 3D rendering of the LV lateral wall with LPS values mapped onto the endocardial surface. (a) a responder (%ΔESV = −43) with uniformly high LPS values across the entire LV free wall and the left lead placed in this region of high LPS (b) a responder (%ΔESV = −68) with a localized region of high LPS values on the basal inferolateral wall and the left lead placed in this region of high LPS (c) a non-responder (%ΔESV = −13) with globally low LPS values and thus the left lead placed in a region of low LPS (d) a non-responder (%ΔESV = +14) with a localized region of high LPS values on the basal inferolateral wall, but the left lead not placed in this region of high LPS

Fig. 5.. Lead placement score (LPS) maps of CRT responders and non-responders.

(a-d) Four example subjects with the following information shown for each subject Lead tip locations: blue = right lead, red = left lead TOS: time to onset of shortening, in ms; PRS_CT: peak regional shortening; TPRS_CT: time to peak regional shortening, in ms; MSRS_CT: maximum pre-stretch of regional shortening; LPS map: lead placement score map as a polar map; 3D LPS map: 3D rendering of the LV lateral wall with LPS values mapped onto the endocardial surface. (a) a responder (%ΔESV = −43) with uniformly high LPS values across the entire LV free wall and the left lead placed in this region of high LPS (b) a responder (%ΔESV = −68) with a localized region of high LPS values on the basal inferolateral wall and the left lead placed in this region of high LPS (c) a non-responder (%ΔESV = −13) with globally low LPS values and thus the left lead placed in a region of low LPS (d) a non-responder (%ΔESV = +14) with a localized region of high LPS values on the basal inferolateral wall, but the left lead not placed in this region of high LPS

Fig. 5.. Lead placement score (LPS) maps of CRT responders and non-responders.

(a-d) Four example subjects with the following information shown for each subject Lead tip locations: blue = right lead, red = left lead TOS: time to onset of shortening, in ms; PRS_CT: peak regional shortening; TPRS_CT: time to peak regional shortening, in ms; MSRS_CT: maximum pre-stretch of regional shortening; LPS map: lead placement score map as a polar map; 3D LPS map: 3D rendering of the LV lateral wall with LPS values mapped onto the endocardial surface. (a) a responder (%ΔESV = −43) with uniformly high LPS values across the entire LV free wall and the left lead placed in this region of high LPS (b) a responder (%ΔESV = −68) with a localized region of high LPS values on the basal inferolateral wall and the left lead placed in this region of high LPS (c) a non-responder (%ΔESV = −13) with globally low LPS values and thus the left lead placed in a region of low LPS (d) a non-responder (%ΔESV = +14) with a localized region of high LPS values on the basal inferolateral wall, but the left lead not placed in this region of high LPS

Fig. 5.. Lead placement score (LPS) maps of CRT responders and non-responders.

(a-d) Four example subjects with the following information shown for each subject Lead tip locations: blue = right lead, red = left lead TOS: time to onset of shortening, in ms; PRS_CT: peak regional shortening; TPRS_CT: time to peak regional shortening, in ms; MSRS_CT: maximum pre-stretch of regional shortening; LPS map: lead placement score map as a polar map; 3D LPS map: 3D rendering of the LV lateral wall with LPS values mapped onto the endocardial surface. (a) a responder (%ΔESV = −43) with uniformly high LPS values across the entire LV free wall and the left lead placed in this region of high LPS (b) a responder (%ΔESV = −68) with a localized region of high LPS values on the basal inferolateral wall and the left lead placed in this region of high LPS (c) a non-responder (%ΔESV = −13) with globally low LPS values and thus the left lead placed in a region of low LPS (d) a non-responder (%ΔESV = +14) with a localized region of high LPS values on the basal inferolateral wall, but the left lead not placed in this region of high LPS