Results of the Predictors of Response to CRT (PROSPECT) Trial
Background— Data from single-center studies suggest that echocardiographic parameters of mechanical dyssynchrony may improve patient selection for cardiac resynchronization therapy (CRT). In a prospective, multicenter setting, the Predictors of Response to CRT (PROSPECT) study tested the performance of these parameters to predict CRT response.
Methods and Results— Fifty-three centers in Europe, Hong Kong, and the United States enrolled 498 patients with standard CRT indications (New York Heart Association class III or IV heart failure, left ventricular ejection fraction ≤35%, QRS ≥130 ms, stable medical regimen). Twelve echocardiographic parameters of dyssynchrony, based on both conventional and tissue Doppler–based methods, were evaluated after site training in acquisition methods and blinded core laboratory analysis. Indicators of positive CRT response were improved clinical composite score and ≥15% reduction in left ventricular end-systolic volume at 6 months. Clinical composite score was improved in 69% of 426 patients, whereas left ventricular end-systolic volume decreased ≥15% in 56% of 286 patients with paired data. The ability of the 12 echocardiographic parameters to predict clinical composite score response varied widely, with sensitivity ranging from 6% to 74% and specificity ranging from 35% to 91%; for predicting left ventricular end-systolic volume response, sensitivity ranged from 9% to 77% and specificity from 31% to 93%. For all the parameters, the area under the receiver-operating characteristics curve for positive clinical or volume response to CRT was ≤0.62. There was large variability in the analysis of the dyssynchrony parameters.
Conclusion— Given the modest sensitivity and specificity in this multicenter setting despite training and central analysis, no single echocardiographic measure of dyssynchrony may be recommended to improve patient selection for CRT beyond current guidelines. Efforts aimed at reducing variability arising from technical and interpretative factors may improve the predictive power of these echocardiographic parameters in a broad clinical setting.
Received October 24, 2007; accepted February 27, 2008.
In patients with New York Heart Association (NYHA) class III and ambulatory class IV systolic heart failure and electrocardiographic evidence of ventricular dyssynchrony, cardiac resynchronization therapy (CRT) improves quality of life and functional status (eg, NYHA class, exercise capacity),1 reduces heart failure–related hospitalizations,2 and prolongs survival.3 Thus, there is a strong clinical mandate for the use of CRT in eligible patients that is supported by national and international practice guidelines.4–6 Although most treated patients show a benefit from CRT, some patients have been considered nonresponders in clinical trials using a variety of measures of clinical responsiveness.1,7,8 In the Multicenter InSync Randomized Clinical Evaluation (MIRACLE), 34% of patients did not demonstrate an improvement in a heart failure clinical composite score (CCS) that combined all-cause mortality, heart failure hospitalization, NYHA class, and patient global assessment into an outcome measure.1,9
Editorial p 2573 Clinical Perspective p 2616
Recently, several echocardiographic measures of mechanical dyssynchrony have identified responders to CRT before device implantation. Ventricular dyssynchrony has been measured with traditional echocardiographic techniques,10 tissue Doppler imaging (TDI), and other methods.11–18 A few of these parameters have demonstrated the ability to distinguish CRT responders from nonresponders with a high degree of accuracy in multiple small single-center studies. To date, however, no multicenter study has examined the potential of echocardiography to assist in determining patients most likely to respond to CRT. Here, we present the results of Predictors of Response to CRT (PROSPECT), a prospective, multicenter, nonrandomized study designed to evaluate selected, predefined baseline echocardiographic parameters for their ability to predict clinical and echocardiographic response to CRT.
A detailed description of the rationale and methods for PROSPECT has been published previously.19 The patient population and study protocol are described below.
Patients with heart failure symptoms referred for CRT according to the current guidelines for the treatment of chronic heart failure were evaluated for enrollment. Patients were included according to the following criteria: left ventricular ejection fraction (LVEF) ≤35% as assessed by the investigator, NYHA functional status III or IV, and QRS duration ≥130 ms. Medical therapy, unless contraindicated, was to include an angiotensin-converting enzyme inhibitor or angiotensin receptor blocker for at least 1 month before enrollment and a β-blocker started at least 3 months before and unchanged for at least 1 month before enrollment. Patients were enrolled across 53 centers in the United States, Europe, and Hong Kong between March 2004 and December 2005. All centers collected data at preimplant baseline assessment; at the time of implantation; immediately after implantation; at 1, 3, and 6 months after implantation; and every 6 months until study closure. Any Medtronic market–released CRT device with or without implantable cardioverter-defibrillator (ICD) functionality could be used in the study.
Echocardiographic Measures of Ventricular Dyssynchrony
The PROSPECT Steering Committee selected 12 echocardiographic parameters identified from published and unpublished literature as possible predictors of a positive response to CRT.10,11,13–18 Details of each parameter are presented in Table 1.
Echocardiography Core Laboratories
The recording of a baseline echocardiogram was performed according to study protocol and sent to either a US or European core laboratory. The core laboratories used echocardiographic equipment from 1 of 3 manufacturers: General Electric (GE, Milwaukee, Wis), Philips (Andover, Mass), and Siemens (Malvern, Pa). Echocardiogram data obtained on GE machines (37%) were analyzed with GE Echopac software on a standalone workstation V4.0.4 or a GE Vivid 7 ultrasound system running version 3.2.6. Data obtained on Philips (50%) used Sonos 5500, 7500, or IE33, and TDI data were analyzed beginning with Philips QLAB version 2.0 and progressively updated throughout the study to version 4.1. TDI data on Siemens (12%) were analyzed with Tomtec Research Arena software version 1.0. In addition, 6 baseline echocardiograms (1%) were obtained with Aloka ProSound 5500.
The US core laboratory in Atlanta, Ga (Emory University–Crawford Long Hospital), processed all studies from the United States and Hong Kong. For studies from Europe, the core laboratory in Pavia, Italy (Policlinico San Matteo), analyzed all standard 2-dimensional measurements and TDI analysis of studies recorded on machines from GE. The core laboratory in London, UK (Hammersmith Hospital), performed TDI analysis of echocardiograms recorded on machines manufactured by Philips and Siemens.
Training and Quality Control
At all centers, training on the echocardiogram protocol, data acquisition, and storage was completed by means of a written manual and video presentation. In addition, each center was required to obtain accreditation from the echocardiography core laboratory in its region by providing high-quality images before enrolling any study subjects. Furthermore, any subsequent studies judged to be of insufficient quality by the core laboratory were censored and not included in the analysis.
A team of echocardiologists (see the Appendix in the online Data Supplement) assisted in the creation and review of an echocardiographic measurement manual that was followed by the core laboratories to ensure consistent measurement of images. This document was reviewed and approved by the PROSPECT Steering Committee, which was made up of heart failure specialists, echocardiologists, and electrophysiologists (see the Appendix). In addition, an independent echocardiographic review committee (see the Appendix) reviewed the core laboratory data and predictor calculations before statistical analysis. The echocardiographic review committee provided support and expert review during visits to each of the core laboratories.
Repeatability and Reproducibility of Measurements
Variability was evaluated with measurements selected to represent the major echocardiographic methodology categories: 2-dimensional study, M-mode, pulsed Doppler, and TDI based. Intraoperator variability was assessed within each core laboratory with 10 baseline recordings. Measurement of each echocardiographic recording was conducted 3 times (on days 1, 2, and 7). For each patient, left ventricular end-systolic volume (LVESV; study end point), septal-posterior wall motion delay (SPWMD), the SD of time to peak systolic velocity of 12 segments of the left ventricular wall at the basal and medial levels (Ts-SD; used for power calculations), left ventricular preejection interval (measured in MIRACLE and MIRACLE ICD), and maximal time to peak systolic velocity difference of 6 segments at the basal level (Ts-peak) were measured.
Interobserver variability was measured between core laboratories with the same echocardiographic parameters (LVESV, SPWMD, Ts-SD, left ventricular preejection interval, Ts-peak). For this assessment, a total of 12 recordings were exchanged by the US and Italian core laboratories; the US and UK core laboratories exchanged 20 echocardiograms.
Definition of Response to CRT
Response to CRT was evaluated through the use of 2 separately analyzed primary outcomes at 6 months: heart failure CCS and relative change in LVESV. The CCS describes patients regardless of vital status at 6 months and includes both objective and subjective measures of clinical status. A patient’s CCS was classified as one of the following: worsened (the patient died or was hospitalized for or associated with worsening heart failure, demonstrated worsening in NYHA class at last observation carried forward, had moderate or marked worsening of patient global assessment score at last observation carried forward, or permanently discontinued CRT because of or associated with worsening heart failure), improved (the patient had not worsened as defined above and demonstrated improvement in NYHA class at last observation carried forward or had moderate or marked improvement in patient global assessment score at last observation carried forward), or unchanged (the patient was neither improved nor worsened).9 An independent end point committee (see the Appendix) adjudicated all hospitalizations and CRT discontinuations for heart failure relatedness. A positive response to CRT was defined in the case of CCS as a designation of “improved” and with LVESV as a reduction of ≥15% at 6 months compared with baseline (paired LVESV measurements). A reduction in LVESV of ≥15% has been used in previous trials10,11 and was used in this study as an objective measure of cardiac function.
Prespecified cutoff values for each echocardiographic measure were determined from both published and unpublished data, and each parameter was tested against the response outcomes (CCS and LVESV). When predictive cutoff values did not appear in the literature, the cutoff was defined as the median value (Table 1). The proportions of patients with a positive response to CRT above and below the specified cutoff values were compared by Fisher’s exact test. No adjustment for conducting multiple statistical tests was made. Receiver-operating characteristics curves were generated, and areas under the curve are reported as a measure of the ability to predict positive response at any cutoff value.
For the analysis of intraobserver and interoperator echocardiographic variability, we calculated an adjusted coefficient of variation (CV), defined as the ratio of the SD and the mean of absolute readings for each echocardiographic parameter. Agreement of binary echocardiographic predictors was assessed with Cohen’s κ coefficients.
Finally, to further characterize the relationship between echocardiographic measures and the 2 primary outcomes, retrospective analyses of the following subgroups were performed: core laboratory–measured LVEF ≤35%, left ventricular end-diastolic diameter ≥65 mm, and ischemic and nonischemic origin of heart failure.
The authors had full access to and take full responsibility for the integrity of the data. All authors have read and agree to the manuscript as written.
Overall, 498 patients were enrolled in PROSPECT, and after 31 early exits, 467 patients were successfully implanted with CRT devices. After exclusion of 41 patients in a narrow QRS substudy (<130 ms), the 426 remaining patients formed the final study group of the present report. The disposition of subjects is presented in Figure 1. A total of 227 patients were enrolled in the United States, 194 in Europe, and 5 in Hong Kong. Baseline characteristics are shown in Table 2. In general, most patients were male (71%); 54% had ischemic origin of heart failure; and the use of evidence-based heart failure medications at baseline was high, as required. Compared with patients enrolled in the United States, European patients had more nonischemic origin of underlying cardiomyopathy (63% versus 44%; P<0.0001) and, from core laboratory measurements, lower LVEF (25.5±6.2% versus 32.9±11.5%; P<0.0001) and larger LVESV (206.6±91.9 versus 132.5±69.5 mL; P<0.0001).
Variability analyses of echocardiographic measures are shown in Table 3. Intraobserver reproducibility was similar in the 3 core laboratories, with low variability for LVESV and left ventricular preejection interval (CV, 3.8% and 3.7%, respectively), moderate variability for Ts-SD and Ts-peak (CV, 11.4% and 15.8%), and high variability for SPWMD (CV, 24.3%). Interobserver variability was higher for each parameter than intraobserver variability, with high variability for Ts-peak, Ts-SD, and SPWMD (CV, 31.9%, 33.7%, and 72.1%, respectively).
TDI data obtained with the Siemens machines were excluded from analysis because of suboptimal data quality as determined by the core laboratories.
CCS End Point
Overall, based on the CCS, 69% of patients improved, 15% remained unchanged, and 16% worsened (Figure 2). Fourteen patients (3%) died and 45 patients (11%) were hospitalized for worsening heart failure during the 6-month follow-up. Table 4 summarizes the ability of each evaluated echocardiographic parameter to predict clinical response as measured by the CCS. Three of the 5 non-TDI parameters and 1 of the 7 TDI-based methods (Ts onset basal) demonstrated modest, statistically significant value in predicting a higher rate of CCS response for those reaching the cutoff value. In receiver-operating characteristics analyses (Table 5), among the published TDI methods, the SD of time to peak velocity of 12 segments achieves an area under the curve of 0.60 (P=0.03) for the ability to predict improved CCS.
LVESV as an End Point
The 286 patients with available paired LVESV measurements showed a relative LVESV reduction of 19.7±27.3% (mean±SD) at 6 months. One hundred sixty-one patients (56.3%) had a reduction of ≥15%, meeting the prespecified definition of improvement, and 26 patients (9.1%) had an increase of at least 15% in LVESV (Figure 2). The predictive ability of each echocardiographic parameter for LVESV reduction is shown in Table 4. All 4 of the non-TDI methods and 1 of the TDI-based tests (time difference between lateral and septal peak systolic wall velocity) demonstrated a significantly higher level of response among those meeting the cutoff criterion.
Based on the receiver-operating characteristics analysis, the time difference between lateral and septal peak systolic wall velocity achieved an area under the curve of 0.61 (P=0.01) for the ability to predict LVESV reduction of ≥15%.
There were no statistically significant differences in CCS or LVESV responses between patients in the United States and those outside the United States (65.2% and 73.4%, respectively, P=0.07 for CCS; and 54.2% and 58.3%, respectively, P=0.55 for LVESV). The CCS response rate was higher among nonischemic patients compared with those with ischemic origin (75% versus 64%, respectively; P=0.01); also, the LVESV response rate tended to be higher in the nonischemic group compared with the ischemic group (63% versus 50%, respectively; P=0.03).
Although the center reported that mean LVEF was 23.6±7% in PROSPECT, the core laboratory–measured mean LVEF was 29.3±10%; 20.2% of subjects had a core laboratory–measured LVEF >35%. Nonetheless, in the subgroups defined by core laboratory–measured LVEF ≤35% (n=340, 79.8%) or left ventricular end-diastolic diameter ≥65 mm (n=265, 62.2%), there were no substantial differences in the predictive power of the echocardiographic parameters compared with all subjects (data not shown).
The PROSPECT trial represents the first large-scale, multicenter, clinical trial evaluating the performance of echocardiographic measures of mechanical ventricular dyssynchrony to predict responsiveness to CRT. Although several parameters predicted statistically significant improvement in clinical and reverse remodeling outcomes, sensitivity and specificity were modest. In a population of patients whose baseline characteristics fulfill the current indications for CRT and whose clinical response rates are comparable to other trials of CRT,1 the findings of PROSPECT suggest that various echocardiographic measures of ventricular dyssynchrony as applied in this study were unable to distinguish responders from nonresponders to a degree that should affect clinical decision making. Thus, current clinical criteria, including the ECG, remain the standard for CRT patient selection.
Given the background of numerous smaller, single-center studies demonstrating a strong correlation between echocardiographic measures of mechanical dyssynchrony and clinical response to CRT,11–18 the results of PROSPECT are somewhat surprising. There may be several explanations for this discrepancy.
We observed relatively low yield and high variability for the TDI measures. Specifically, the percent of individual parameters deemed interpretable by the core laboratories ranged between 61% and 95% for the routine non-TDI methods and between 37% and 82% for TDI-based tests. Although TDI measurements, performed in a laboratory with special expertise, may be effective in predicting outcomes to CRT, current technology, degree of training standards, and analytic methods do not allow its incorporation in a generalized setting and may currently represent a significant limitation for the widespread use of TDI methods. The use of different echocardiographic platforms and equipment to collect and analyze images may have exacerbated variability in measurements. However, in this study evaluating methods across a wide spectrum of locations and medical practices, we sought to include the major vendors of these technologies. There also appear to be differing practice styles between the United States and Europe with regard to patient selection for CRT, with a tendency toward use of CRT earlier in the disease process in the United States, which may have added to the variability in patient selection characteristics. Note, however, that distance walked in 6 minutes was 246 m on average in US patients versus 308 m in European patients. It also has been suggested that longitudinal dyssynchrony may be subject to higher variability and may correlate less well with ventricular functional recovery after CRT than 2-dimensional strain techniques that evaluate radial dyssynchrony.20 Other potentially interesting methods for further study are novel echocardiographic techniques, including 2-dimensional strain imaging (speckle tracking approach),21 3-dimensional echocardiography,22 anatomic M-mode imaging,23 and magnetic resonance imaging.24
Were the cutoff points for the echocardiographic measurements appropriately selected? The values selected for PROSPECT, although based on published reports when available, were developed in observational single-center studies and may need to be refined when the methodology is translated to the larger community. Do the end points reflect clinically relevant outcomes, and are they robust enough to be accurately predicted by echocardiographic methods? The 2 primary end points were selected to measure clinically relevant outcomes in heart failure patients with a validated tool (CCS) and to assess cardiac structural response (reverse remodeling). The CCS has been used in heart failure clinical trials with success1,7 and consists of easily measured components (survival, hospitalization, NYHA class, patient global assessment) with stringent criteria for improved status. Nonetheless, given that patients enrolled in clinical trials generally improve even in the absence of therapy, it is possible that CCS changes may not be due to events that bear any relationship to echocardiographic measures. Therefore, a measure of cardiac structural change also was incorporated. The reverse remodeling end point (LVESV reduction ≥15%) has been used more specifically in studies evaluating echocardiographic predictors,10,11 and there was general consensus among the steering committee that 15% reduction in LVESV was a robust and clinically relevant outcome measure. The ideal end point to assess response to CRT is currently unclear, and it has been noted that a discrepancy exists between clinical and echocardiographic responses (a reduction in LVESV), with a greater rate of clinical response compared with echocardiographic response.25
The PROSPECT patient population may be somewhat different from those studied for the published echocardiographic parameters, partly accounting for the discrepancy with previously reported results. Indeed, as measured by the core laboratory, 20.2% had LVEF >35% and 37.8% had left ventricular end-diastolic diameter <65 mm, suggesting that perhaps the severity of illness in PROSPECT, a study conducted in the course of clinical practice, was not congruent with previously studied patients in the setting of clinical trials. Nonetheless, the average 6-minute walk distance in PROSPECT was 274±122 m compared with 298±92 m in MIRACLE1 and 243±122 m in MIRACLE ICD.7 More specifically, in 2 TDI methodology studies,11,15 baseline 6-minute walk distances were ≈248 and 320 m, respectively. Furthermore, extensive subgroup analyses looking at those with the lowest LVEF, greatest left ventricular diameter, and ischemic or nonischemic origin do not demonstrate substantial difference in the predictive ability of the echocardiographic parameters. Still, additional studies focusing on patients with LVEF ≤35% and left ventricular end-diastolic diameter ≥65 mm should be performed.
It is possible that mechanisms other than attenuation of systolic dyssynchrony per se underlie the benefits seen with CRT. Chronically, the existence of an abnormally high regional wall stress such as with an infarcted zone may lead to progressive left ventricular dilation.26 Prinzen and colleagues27 have shown that ventricular pacing significantly reduces wall stress and workload by causing the region near the pacing site to contract early in the cycle against a lower ventricular pressure and at a reduced preload. Therefore, the preexcited region does less external work and develops less stress; in a chronic setting, this phenomenon may have favorable effects on the failing heart.
Finally, determining the patients most likely to respond before CRT requires an assessment of factors other than echocardiography-measured dyssynchrony. For example, limited venous anatomy may introduce a mismatch between left ventricular lead position and latest activation site28 that is likely to lead to a variation of CRT response that is unrelated to echocardiographic parameters. In a multicenter setting, this factor is even more difficult to control. In addition, extensive left ventricular scar tissue attenuates clinical and structural response rates to CRT despite the presence of cardiac dyssynchrony and could be a potential confounder of results in ischemic patients.29 Therefore, integration of cardiac dyssynchrony and magnetic resonance imaging evaluation for the presence, location, and size of the scar in patients with ischemic cardiomyopathy may be a useful modality in patient selection for CRT.30,31 In this regard, the interplay between heart failure origin, end-point selection, and echocardiographic parameters should be noted in future study design.
The major limitation of this study is the observational, nonrandomized design. The main purpose of this study was to assess predictive abilities of the echocardiographic parameters, not to test whether the use of these parameters has a clinical impact. If echocardiographic parameters can be identified in a multicenter setting that improve significantly on the current implantation criteria, further randomized studies testing such parameters would be warranted. The relatively high levels of both interobserver and intraobserver variability noted in this multicenter study impair our ability to conclusively assess the potential predictive capabilities of the echocardiographic parameters in an ideal setting. Therefore, the introduction of variability in image acquisition and analysis currently represents a major limitation to translating expert, single-center experience to widespread use.
Despite promising preliminary data from prior single-center studies, echocardiographic measures of dyssynchrony aimed at improving patient selection criteria for CRT do not appear to have a clinically relevant impact on improving response rates when studied in a multicenter setting such as PROSPECT. Thus, at present, the echocardiographic parameters assessing dyssynchrony do not have enough predictive value to be recommended as selection criteria for CRT beyond current indications.
Source of Funding
Medtronic Inc provided funding for this study and manufactured the CRT system used in this research.
Drs Abraham, Bax, Chung, Gorcsan, Nihoyannopoulos, St. John Sutton, Tavazzi, and Yu have served as consultants to and received research grants from Medtronic. Drs De Sutter, Ghio, Leclercq, Leon, Murillo, Nihoyannopoulos and Sun have received honoraria from or consulted for Medtronic.
Abraham WT, Fisher WG, Smith AL, Delurgio DB, Leon AR, Loh E, Kocovic DZ, Packer M, Clavell AL, Hayes DL, Ellestad M, Trupp RJ, Underwood J, Pickering F, Truex C, McAtee P, Messenger J, for MIRACLE Study Group, Multicenter InSync Randomized Clinical Evaluation. Cardiac resynchronization in chronic heart failure. N Engl J Med. 2002; 346: 1845–1853.
Bristow MR, Saxon LA, Boehmer J, Krueger S, Kass DA, De Marco T, Carson P, DiCarlo L, DeMets D, White BG, DeVries DW, Feldman AM, for the COMPANION Investigators. Cardiac-resynchronization therapy with or without an implantable defibrillator in advanced chronic heart failure. N Engl J Med. 2004; 350: 2140–2150.
Cleland JG, Daubert JC, Erdmann E, Freemantle N, Gras D, Kappenberger L, Tavazzi L, for the Cardiac Resynchronization-Heart Failure (CARE-HF) Study Investigators. The effect of cardiac resynchronization on morbidity and mortality in heart failure. N Engl J Med. 2005; 352: 1539–1549.
Hunt SA, Abraham WT, Chin MH, Feldman AM, Francis GS, Ganiats TG, Jessup M, Konstam MA, Mancini DM, Michl K, Oates JA, Rahko PS, Silver MA, Stevenson LW, Yancy CW, Antman EM, Smith SC Jr, Adams CD, Anderson JL, Faxon DP, Fuster V, Halperin JL, Hiratzka LF, Jacobs AK, Nishimura R, Ornato JP, Page RL, Riegel B, for the American College of Cardiology, American Heart Association Task Force on Practice Guidelines, American College of Chest Physicians, International Society for Heart and Lung Transplantation, Heart Rhythm Society. ACC/AHA 2005 guideline update for the diagnosis and management of chronic heart failure in the adult: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Update the 2001 Guidelines for the Evaluation and Management of Heart Failure). Circulation. 2005; 112: e154–e235.
Swedberg K, Cleland J, Swedberg K, Cleland J, Dargie H, Drexler H, Follath F, Komajda M, Tavazzi L, Smiseth OA, Gavazzi A, Haverich A, Hoes A, Jaarsma T, Korewicki J, Lévy S, Linde C, Lopez-Sendon JL, Nieminen MS, Piérard L, Remme WJ, for the Task Force for the Diagnosis and Treatment of Chronic Heart Failure of the European Society of Cardiology. Guidelines for the diagnosis and treatment of chronic heart failure: executive summary (update 2005). Eur Heart J. 2005; 26: 1115–1140.
Young JB, Abraham WT, Smith AL, Leon AR, Lieberman R, Wilkoff B, Canby RC, Schroeder JS, Liem LB, Hall S, Wheelan K, for the Multicenter InSync ICD Randomized Clinical Evaluation (MIRACLE ICD) Trial Investigators. Combined cardiac resynchronization and implantable cardioversion defibrillation in advanced chronic heart failure: the MIRACLE ICD Trial. JAMA. 2003; 289: 2685–2694.
Yu CM, Fung JW, Zhang Q, Chan CK, Chan YS, Lin H, Kum LC, Kong SL, Zhang Y, Sanderson JE. Tissue Doppler imaging is superior to strain rate imaging and postsystolic shortening on the prediction of reverse remodeling in both ischemic and nonischemic heart failure after cardiac resynchronization therapy. Circulation. 2004; 110: 66–73.
Bax JJ, Marwick TH, Molhoek SG, Bleeker GB, van Erven L, Boersma E, Steendijk P, van der Wall EE, Schalij MJ. Left ventricular dyssynchrony predicts benefit of cardiac resynchronization therapy in patients with end-stage heart failure before pacemaker implantation. Am J Cardiol. 2003; 92: 1238–1240.
Søgaard P, Egeblad H, Pedersen AK, Kim WY, Kristensen BO, Hansen PS, Mortensen PT. Sequential versus simultaneous biventricular resynchronization for severe heart failure: evaluation by tissue Doppler imaging. Circulation. 2002; 106: 2078–2084.
Yu CM, Abraham WT, Bax J, Chung E, Fedewa M, Ghio S, Leclercq C, Leon AR, Merlino J, Nihoyannopoulos P, Notabartolo D, Sun JP, Tavazzi L, for the PROSPECT Investigators. Predictors of Response to Cardiac Resynchronization Therapy (PROSPECT): study design. Am Heart J. 2005; 149: 600–605.
Helm RH, Leclercq C, Faris OP, Ozturk C, McVeigh E, Lardo AC, Kass DA. Cardiac dyssynchrony analysis using circumferential versus longitudinal strain: implications for assessing cardiac resynchronization. Circulation. 2005; 31: 111:2760–2767.
Suffoletto MS, Dohi K, Cannesson M, Saba S, Gorcsan J 3rd. Novel speckle-tracking radial strain from routine black-and-white echocardiographic images to quantify dyssynchrony and predict response to cardiac resynchronization therapy. Circulation. 2006; 113: 960–968.
van Dijk J, Knaapen P, Russel IK, Hendriks T, Allaart CP, de Cock CC, Kamp O. Mechanical dyssynchrony by 3D echo correlates with acute haemodynamic response to biventricular pacing in heart failure patients. Europace. 2008; 10: 63–68.
Westenberg JJ, Lamb HJ, van der Geest RJ, Bleeker GB, Holman ER, Schalij MJ, de Roos A, van der Wall EE, Reiber JH, Bax JJ. Assessment of left ventricular dyssynchrony in patients with conduction delay and idiopathic dilated cardiomyopathy: head-to-head comparison between tissue Doppler imaging and velocity-encoded magnetic resonance imaging. J Am Coll Cardiol. 2006; 47: 2042–2048.
Aikawa Y, Rohde L, Plehn J, Greaves SC, Menapace F, Arnold MO, Rouleau JL, Pfeffer MA, Lee RT, Solomon SD. Regional wall stress predicts ventricular remodeling after anteroseptal myocardial infarction in the Healing and Early Afterload Reducing Trial (HEART): an echocardiography-based structural analysis. Am Heart J. 2001; 141: 234–242.
Bleeker GB, Kaandorp TA, Lamb HJ, Boersma E, Steendijk P, de Roos A, van der Wall EE, Schalij MJ, Bax JJ. Effect of posterolateral scar tissue on clinical and echocardiographic improvement after cardiac resynchronization therapy. Circulation. 2006; 113: 969–976.
Ypendburg C, Roes SD, Bleeker GB, Kaandorp TA, de Roos A, Schalij MJ, van der Wall EE, Bax JJ. Effect of total scar burden on contrast-enhanced magnetic resonance imaging on response to cardiac resynchronization therapy. Am J Cardiol. 2007; 99: 567–560.
Chalil S, Stegemann B, Muhyaldeen SA, Khadjooi K, Foley PW, Smith RE, Leyva F. Effect of posterolateral left ventricular scar on mortality and morbidity following cardiac resynchronization therapy. Pacing Clin Electrophysiol. 2007; 10: 1201–1209.
Cardiac resynchronization has revolutionized the care of a substantial portion of patients with advanced heart failure; for those with appropriate indications, cardiac resynchronization therapy appears to improve clinical status in ≈70% of those treated. Although this degree of response is very encouraging, numerous echocardiographic techniques, based largely on tissue Doppler measurements, have been developed to better distinguish responders from nonresponders. Ideally, the ability to accurately predict likelihood of response will enhance the quality and substance of the physician-patient conversation before treatment. These methods attempt to detect mechanical dyssynchrony more accurately than with the QRS under the assumption that correction of dyssynchrony is the main mechanism of action with cardiac resynchronization therapy. However, there had not been a large multicenter study to evaluate these methods prospectively. In 426 patients, Predictors of Response to CRT (PROSPECT) studied 12 echocardiographic methods for their ability to predict response and showed that in the multicenter setting, they lacked sensitivity and specificity to affect clinical decisions. We therefore recommend that these methods not be used routinely in evaluating a patient for cardiac resynchronization therapy. Other methods are under investigation (eg, strain measurements, 3-dimensional imaging, scar location) that may be able to distinguish responders from nonresponders with more accuracy. However, before they can be recommended for general use, their applicability should be tested in a multicenter setting.
Continuing medical education (CME) credit is available for this article. Go to http://cme.ahajournals.org to take the quiz.
The online Data Supplement, which consists of an Appendix, can be found with this article at http://circ.ahajournals.org/cgi/content/full/ CIRCULATIONAHA.107.743120/DC1.
Clinical trial registration information—URL: http://www.clinicaltrials.gov. Unique identifier: NCT00253357.