Events include all-cause mortality, heart failure(HF)-related urgent care (a healthcare utilization visit involving intravenous(IV) therapy for heart failure) or significant increase(at least 15%) in LVESVI (a measure of the volume of a patient's left ventricle) from randomization to a later time point. Time from randomization until the subject experienced one of these events served as the outcome measure. LVESVI endpoints occurred primarily at those visits in which LVESVI measurements were required (6, 12, 18, 24 months). Because endpoints such as death or HF urgent care could occur at any time during follow-up, the subject's outcome measure could range from less than 1 month to 105 months (maximum follow-up duration). Primary endpoints and follow-up data occurring after a subject missed a required LVESVI measurement were excluded from the analysis and the table below. The counts reflect the number of subjects meeting each endpoint, and are not necessarily mutually exclusive.

Only randomized subjects in both device groups were included in the analysis of this endpoint, as the analysis was restricted to post-randomization data. Subjects in the "No Implant Attempt", "Unsuccessful Implants", "CRT-P: Not Randomized", and "CRT-D: Not Randomized" subgroups were excluded.

The endpoint is the time to death from any cause. The rate of mortality, as measure by the hazard rate, in each randomization arm will be compared.

This outcome includes all post-randomization deaths, whereas the reporting of the primary outcome excluded primary endpoints (including deaths) that occurred after the subject had missed a study-required echocardiogram (used to determine if the LVESVI primary endpoint was met).

Only randomized subjects in both device groups were included in the analysis of this endpoint, as the analysis was restricted to post-randomization data. Subjects in the "No Implant Attempt", "Unsuccessful Implants", "CRT-P: Not Randomized", and "CRT-D: Not Randomized" subgroups were excluded.

The endpoint will be a subject's time from randomization to either their first heart failure-related hospitalization, or death.

Only randomized subjects in both device groups were included in the analysis of this endpoint, as the analysis was restricted to post-randomization data. Subjects in the "No Implant Attempt", "Unsuccessful Implants", "CRT-P: Not Randomized", and "CRT-D: Not Randomized" subgroups were excluded.

The endpoint will be the time from randomization to either death or a visit (6, 12, 18, 24 month or interim visit) in which the subject undergoes an echocardiogram and the measured left ventricular end systolic volume index (a measure of the size of the subject's left ventricle normalized over their body surface area) is at least 15% greater than the corresponding measured value at randomization.

Only LVESVI endpoints/deaths and follow-up data occurring before a subject missed an LVESVI measurement (due to missed visit, echo not performed, etc.) were used in the analysis and included in the table below. The counts reflect the number of subjects meeting each endpoint, and are not mutually exclusive.

Only randomized subjects in both device groups were included in the analysis of this endpoint, as the analysis was restricted to post-randomization data. Subjects in the "No Implant Attempt", "Unsuccessful Implants", "CRT-P: Not Randomized", and "CRT-D: Not Randomized" subgroups were excluded.

The endpoint is the time from randomization to a subject's first heart failure (HF)-related hospitalization. For each randomization arm, the number of subjects who met the endpoint, experiencing at least one heart failure-related hospitalization post-randomization, are reported, as well as the number of randomized subjects who did not experience any HF hospitalizations post-randomization.

Only randomized subjects in both device groups were included in the analysis of this endpoint, as the analysis was restricted to post-randomization data. Subjects in the "No Implant Attempt", "Unsuccessful Implants", "CRT-P: Not Randomized", and "CRT-D: Not Randomized" subgroups were excluded.

For each subject the endpoint was the days hospitalized for heart failure per patient year, calculated as the total number of days the subject was hospitalized for heart failure divided by the subject's total follow-up time. Only post-randomization data were used.

Only randomized subjects in both device groups were included in the analysis of this endpoint, as the analysis was restricted to post-randomization data. Subjects in the "No Implant Attempt", "Unsuccessful Implants", "CRT-P: Not Randomized", and "CRT-D: Not Randomized" subgroups were excluded.

The endpoint is a subject's change in New York Heart Association Classification (a measure of the degree of heart failure a subject has on a 4 class scale, with NYHA I being the healthiest score and NYHA IV being the sickest score) from randomization to each of four time points: 6 months, 12 months, 18 months, and 24 months post-randomization. The change categories listed will be relative to randomization.

For each time point(e.g. 6 months), only subjects with NYHA assessed at both randomization and that time point were included in the analysis. Those who could not be analyzed at a time point(for reasons such as death, exit,missed visit,or NYHA not assessed at visit) are listed under the "Comparative data not available" category for that time point.

The endpoint is a subject's change in Heart Failure Stage (a measure of the degree of heart failure a subject has on a 4 stage scale (A, B, C, D), with Class A being the healthiest score and Class D being the sickest score) from randomization to each of four time points: 6 months, 12 months, 18 months, and 24 months.

For each time point(e.g. 6 months), only subjects with HF Stage assessed at randomization and that time point were included in the analysis. Those who could not be analyzed at a time point(for reasons such as death, exit,missed visit,or HF Stage not assessed) are listed under the "Comparative data not available" category for that time point.

The endpoints are what classes of drugs (e.g. Beta blockers, Diuretics, Nitrates, etc.) each subject was on at the time of scheduled visits (e.g Randomization, 6 months, 12 months, etc.)

Because indications for defibrillation devices like CRT-Ds are defined by characteristics that relate to medication guidelines, medication results are presented separately for each device group. Medications were assessed only for subjects who completed visits (denoted as "Subjects with Medications Assessed").

Adverse events that subjects experienced after they were randomized were compared between arms with regard to several categories such as heart failure (HF)-relatedness, relatedness to the implant procedure, and relatedness to the implanted system, including individual components such as the left ventricular (LV) lead and the CRT-P or CRT-D generator.

Cardiovascular-related healthcare utilizations (HCUs), such as hospitalizations, Emergency Department visits, urgent care visits, and clinic visits that subjects experienced after being randomized were summarized for each randomization arm

The endpoint will be a subject's change in Quality of Life score from randomization to 6 months. The Quality of Life score at a time point is calculated using the Minnesota Living with Heart Failure Questionnaire, which consists of 21 questions each on a 6 point scale from 0 to 5. The 21 scores are added up and the final score, ranging from 0 (best) to 105 (worst) is the subject's quality of life score. For each subject the measure will be the randomization visit - 6 month difference in QOL score, with positive values denoting a reduction in score and improvement in Quality of Life.

Subjects with missing QOL scores at one or both time points were excluded from analysis, and so the number of subjects analyzed for this outcome was a subset of the number of randomized subjects.

The endpoint will be a subject's change in Quality of Life score from randomization to 12 months. The Quality of Life score at a time point is calculated using the Minnesota Living with Heart Failure Questionnaire, which consists of 21 questions each on a 6 point scale from 0 to 5. The 21 scores are added up and the final score, ranging from 0 (best) to 105 (worst) is the subject's quality of life score. For each subject the measure will be the randomization visit - 12 month difference in QOL score, with positive values denoting a reduction in score and improvement in Quality of Life.

Subjects with missing QOL scores at one or both time points were excluded from analysis, and so the number of subjects analyzed for this outcome was a subset of the number of randomized subjects.

The endpoint will be a subject's change in Quality of Life score from randomization to 18 months. The Quality of Life score at a time point is calculated using the Minnesota Living with Heart Failure Questionnaire, which consists of 21 questions each on a 6 point scale from 0 to 5. The 21 scores are added up and the final score, ranging from 0 (best) to 105 (worst) is the subject's quality of life score. For each subject the measure will be the randomization visit - 18 month difference in QOL score, with positive values denoting a reduction in score and improvement in Quality of Life.

Subjects with missing QOL scores at one or both time points were excluded from analysis, and so the number of subjects analyzed for this outcome was a subset of the number of randomized subjects.

The endpoint will be a subject's change in Quality of Life score from randomization to 24 months. The Quality of Life score at a time point is calculated using the Minnesota Living with Heart Failure Questionnaire, which consists of 21 questions each on a 6 point scale from 0 to 5. The 21 scores are added up and the final score, ranging from 0 (best) to 105 (worst) is the subject's quality of life score. For each subject the measure will be the randomization visit - 24 month difference in QOL score, with positive values denoting a reduction in score and improvement in Quality of Life.

Subjects with missing QOL scores at one or both time points were excluded from analysis, and so the number of subjects analyzed for this outcome was a subset of the number of randomized subjects.

The endpoint will be a subject's change in LV Ejection Fraction (a measure of the percentage of blood ejected from the left ventricle of the heart with each contraction). A normal range is 55% to 70%. For each subject the measure will be the 6 month - randomization visit difference in LVEF value.

The endpoint will be a subject's change in LV Ejection Fraction (a measure of the percentage of blood ejected from the left ventricle of the heart with each contraction). A normal range is 55% to 70%. For each subject the measure will be the 12 month - randomization visit difference in LVEF value.

The endpoint will be a subject's change in LV Ejection Fraction (a measure of the percentage of blood ejected from the left ventricle of the heart with each contraction). A normal range is 55% to 70%. For each subject the measure will be the 18 month - randomization visit difference in LVEF value.

The endpoint will be a subject's change in LV Ejection Fraction (a measure of the percentage of blood ejected from the left ventricle of the heart with each contraction). A normal range is 55% to 70%. For each subject the measure will be the 24 month - randomization visit difference in LVEF value.

The endpoint will be a subject's change in Left Ventricular End Systolic Volume Index (a measure of the volume of blood in the left ventricle at the end of systole, normalized over body surface area). In other words, a measure of the size of the left ventricle. For each subject the measure is the 6 month - randomization visit difference in LVESVI. Negative values correspond to reductions in LVESVI over time.

The endpoint will be a subject's change in Left Ventricular End Systolic Volume Index (a measure of the volume of blood in the left ventricle at the end of systole, normalized over body surface area). In other words, a measure of the size of the left ventricle. For each subject the measure is the 12 month - randomization visit difference in LVESVI. Negative values correspond to reductions in LVESVI over time.

The endpoint will be a subject's change in Left Ventricular End Systolic Volume Index (a measure of the volume of blood in the left ventricle at the end of systole, normalized over body surface area). In other words, a measure of the size of the left ventricle. For each subject the measure is the 18 month - randomization visit difference in LVESVI. Negative values correspond to reductions in LVESVI over time.

The endpoint will be a subject's change in Left Ventricular End Systolic Volume Index (a measure of the volume of blood in the left ventricle at the end of systole, normalized over body surface area). In other words, a measure of the size of the left ventricle. For each subject the measure is the 24 month - randomization visit difference in LVESVI. Negative values correspond to reductions in LVESVI over time.

The endpoint will be a subject's change in Left Ventricular End Diastolic Volume Index (a measure of the volume of blood in the left ventricle at the end of diastole, normalized over body surface area). In other words, a measure of the size of the left ventricle. For each subject the measure is the 6 month - randomization visit difference in LVEDVI. Negative values correspond to reductions in LVEDVI over time.

The endpoint will be a subject's change in Left Ventricular End Diastolic Volume Index (a measure of the volume of blood in the left ventricle at the end of diastole, normalized over body surface area). In other words, a measure of the size of the left ventricle. For each subject the measure is the 12 month - randomization visit difference in LVEDVI. Negative values correspond to reductions in LVEDVI over time.

The endpoint will be a subject's change in Left Ventricular End Diastolic Volume Index (a measure of the volume of blood in the left ventricle at the end of diastole, normalized over body surface area). In other words, a measure of the size of the left ventricle. For each subject the measure is the 18 month - randomization visit difference in LVEDVI. Negative values correspond to reductions in LVEDVI over time.

The endpoint will be a subject's change in Left Ventricular End Diastolic Volume Index (a measure of the volume of blood in the left ventricle at the end of diastole, normalized over body surface area). In other words, a measure of the size of the left ventricle. For each subject the measure is the 24 month - randomization visit difference in LVEDVI. Negative values correspond to reductions in LVEDVI over time.

The endpoint will be a subject's change in Left Ventricular Mass ( a measure of the size of the left ventricle) from randomization to 6 months. For each subject the measurement was calculated as 6 month - randomization visit difference in LV mass measurement. Negative values correspond to a reduction in LV mass over time.

The endpoint will be a subject's change in Left Ventricular Mass ( a measure of the size of the left ventricle) from randomization to 12 months. For each subject the measurement was calculated as 12 month - randomization visit difference in LV mass measurement. Negative values correspond to a reduction in LV mass over time.

The endpoint will be a subject's change in Left Ventricular Mass ( a measure of the size of the left ventricle) from randomization to 18 months. For each subject the measurement was calculated as 18 month - randomization visit difference in LV mass measurement. Negative values correspond to a reduction in LV mass over time.

The endpoint will be a subject's change in Left Ventricular Mass ( a measure of the size of the left ventricle) from randomization to 24 months. For each subject the measurement was calculated as 24 month - randomization visit difference in LV mass measurement. Negative values correspond to a reduction in LV mass over time.

The endpoint will be a subject's change in LVEDD (a measure of the dimension of the left ventricle at the end of diastole). For each subject the measure was the 6 month - randomization visit difference in LVEDD value. Negative values correspond to reductions in LVEDD.

The endpoint will be a subject's change in LVEDD (a measure of the dimension of the left ventricle at the end of diastole). For each subject the measure was the 12 month - randomization visit difference in LVEDD value. Negative values correspond to reductions in LVEDD.

The endpoint will be a subject's change in LVEDD (a measure of the dimension of the left ventricle at the end of diastole). For each subject the measure was the 18 month - randomization visit difference in LVEDD value. Negative values correspond to reductions in LVEDD.

The endpoint will be a subject's change in LVEDD (a measure of the dimension of the left ventricle at the end of diastole). For each subject the measure was the 24 month - randomization visit difference in LVEDD value. Negative values correspond to reductions in LVEDD.

The endpoint will be a subject's change in LVESD (a measure of the dimension of the left ventricle at the end of systole). For each subject the measure was the 6 month - randomization visit difference in LVESD value. Negative values correspond to reductions in LVESD.

The endpoint will be a subject's change in LVESD (a measure of the dimension of the left ventricle at the end of systole). For each subject the measure was the 12 month - randomization visit difference in LVESD value. Negative values correspond to reductions in LVESD.

The endpoint will be a subject's change in LVESD (a measure of the dimension of the left ventricle at the end of systole). For each subject the measure was the 18 month - randomization visit difference in LVESD value. Negative values correspond to reductions in LVESD.

The endpoint will be a subject's change in LVESD (a measure of the dimension of the left ventricle at the end of systole). For each subject the measure was the 24 month - randomization visit difference in LVESD value. Negative values correspond to reductions in LVESD.

The endpoint will be a subject's change in mitral regurgitation (a measure of how much blood flows backwards into the heart due to the mitral valve not closing properly). The measure for each subject will be the 6 month - randomization visit difference in mitral regurgitation. Negative values reflect reductions in mitral regurgitation over time.

The endpoint will be a subject's change in mitral regurgitation (a measure of how much blood flows backwards into the heart due to the mitral valve not closing properly). The measure for each subject will be the 12 month - randomization visit difference in mitral regurgitation. Negative values reflect reductions in mitral regurgitation over time.

The endpoint will be a subject's change in mitral regurgitation (a measure of how much blood flows backwards into the heart due to the mitral valve not closing properly). The measure for each subject will be the 18 month - randomization visit difference in mitral regurgitation. Negative values reflect reductions in mitral regurgitation over time.

The endpoint will be a subject's change in mitral regurgitation (a measure of how much blood flows backwards into the heart due to the mitral valve not closing properly). The measure for each subject will be the 24 month - randomization visit difference in mitral regurgitation. Negative values reflect reductions in mitral regurgitation over time.

The endpoint will be a subject's change in cardiac index (a measure of how much blood the left ventricle ejects in one minute, normalized over body surface area) from randomization to 6 months. The measure for each subject will be the 6 month - randomization visit value.

The endpoint will be a subject's change in cardiac index (a measure of how much blood the left ventricle ejects in one minute, normalized over body surface area) from randomization to 12 months. The measure for each subject will be the 12 month - randomization visit value.

The endpoint will be a subject's change in cardiac index (a measure of how much blood the left ventricle ejects in one minute, normalized over body surface area) from randomization to 18 months. The measure for each subject will be the 18 month - randomization visit value.

The endpoint will be a subject's change in cardiac index (a measure of how much blood the left ventricle ejects in one minute, normalized over body surface area) from randomization to 24 months. The measure for each subject will be the 24 month - randomization visit value.

The endpoint will be a subject's change in interventricular mechanical delay (a measure of dyssynchrony between ventricles, measured in ms) from randomization to the 6 month visit. The measure will be the 6 month - randomization visit difference in IVMD.

The endpoint will be a subject's change in interventricular mechanical delay (a measure of dyssynchrony between ventricles, measured in ms) from randomization to the 12 month visit. The measure will be the 12 month - randomization visit difference in IVMD.

The endpoint will be a subject's change in interventricular mechanical delay (a measure of dyssynchrony between ventricles, measured in ms) from randomization to the 18 month visit. The measure will be the 18 month - randomization visit difference in IVMD.

The endpoint will be a subject's change in interventricular mechanical delay (a measure of dyssynchrony between ventricles, measured in ms) from randomization to the 24 month visit. The measure will be the 24 month - randomization visit difference in IVMD.

The endpoint will be a subject's change in E:A ratio (a measure of diastolic function) from randomization to 6 months. The measure for each subject will be the 6 month - randomization visit difference in E:A ratio. Typical values for the E:A ratio at a single time point are 1.04 in men and 1.03 in women.

The endpoint will be a subject's change in E:A ratio (a measure of diastolic function) from randomization to 12 months. The measure for each subject will be the 12 month - randomization visit difference in E:A ratio. Typical values for the E:A ratio at a single time point are 1.04 in men and 1.03 in women.

The endpoint will be a subject's change in E:A ratio (a measure of diastolic function) from randomization to 18 months. The measure for each subject will be the 18 month - randomization visit difference in E:A ratio. Typical values for the E:A ratio at a single time point are 1.04 in men and 1.03 in women.

The endpoint will be a subject's change in E:A ratio (a measure of diastolic function) from randomization to 24 months. The measure for each subject will be the 24 month - randomization visit difference in E:A ratio. Typical values for the E:A ratio at a single time point are 1.04 in men and 1.03 in women.

The endpoint will be a subject's Clinical Composite Score at 6 months. The Clinical Composite Score is a 3 point score (Worsened, Unchanged, and Improved) based on a number of factors including: whether the subject has died, whether the subject has experienced a heart failure-related hospitalization, whether the subject has discontinued their therapy due to worsening heart failure, whether their New York Heart Association classification has improved or worsened since randomization, and whether they feel moderately or markedly better since randomization.

Only randomized subjects in both device groups were included in the analysis of this endpoint, as the analysis was restricted to post-randomization data. Subjects in the "No Implant Attempt", "Unsuccessful Implants", "CRT-P: Not Randomized", and "CRT-D: Not Randomized" subgroups were excluded.

The endpoint will be a subject's Clinical Composite Score at 12 months. The Clinical Composite Score is a 3 point score (Worsened, Unchanged, and Improved) based on a number of factors including: whether the subject has died, whether the subject has experienced a heart failure-related hospitalization, whether the subject has discontinued their therapy due to worsening heart failure, whether their New York Heart Association classification has improved or worsened since randomization, and whether they feel moderately or markedly better since randomization.

Only randomized subjects in both device groups were included in the analysis of this endpoint, as the analysis was restricted to post-randomization data. Subjects in the "No Implant Attempt", "Unsuccessful Implants", "CRT-P: Not Randomized", and "CRT-D: Not Randomized" subgroups were excluded.

The endpoint will be a subject's Clinical Composite Score at 18 months. The Clinical Composite Score is a 3 point score (Worsened, Unchanged, and Improved) based on a number of factors including: whether the subject has died, whether the subject has experienced a heart failure-related hospitalization, whether the subject has discontinued their therapy due to worsening heart failure, whether their New York Heart Association classification has improved or worsened since randomization, and whether they feel moderately or markedly better since randomization.

Only randomized subjects in both device groups were included in the analysis of this endpoint, as the analysis was restricted to post-randomization data. Subjects in the "No Implant Attempt", "Unsuccessful Implants", "CRT-P: Not Randomized", and "CRT-D: Not Randomized" subgroups were excluded.

The endpoint will be a subject's Clinical Composite Score at 24 months. The Clinical Composite Score is a 3 point score (Worsened, Unchanged, and Improved) based on a number of factors including: whether the subject has died, whether the subject has experienced a heart failure-related hospitalization, whether the subject has discontinued their therapy due to worsening heart failure, whether their New York Heart Association classification has improved or worsened since randomization, and whether they feel moderately or markedly better since randomization.

Only randomized subjects in both device groups were included in the analysis of this endpoint, as the analysis was restricted to post-randomization data. Subjects in the "No Implant Attempt", "Unsuccessful Implants", "CRT-P: Not Randomized", and "CRT-D: Not Randomized" subgroups were excluded.

The endpoint will be whether each subject who underwent an implant attempt of a Cardiac Resynchronization Therapy device, be it a pacing only device (CRT-P) or a pacing device with defibrillation capability (CRT-D), had a successful procedure (i.e. the generator, left ventricular lead, and right ventricular lead were successfully implanted). Only one implant attempt was allowed.

For this outcome measure, only subjects in the "No Implant Attempt" subgroup were excluded.

Among subjects implanted with a Cardiac Resynchronization Therapy with Defibrillation device (CRT-D) and randomized, the endpoint was the time from randomization until the subject experienced a ventricular tachyarrhythmia. For each randomization arm, the number of CRT-D subjects who experienced at least one ventricular tachyarrhythmia post-randomization is reported, as well as the number of CRT-D subjects who did not experience one or more ventricular tachyarrhythmias post-randomization.

Only randomized subjects in the CRT-D device group were included in the analysis of this endpoint, as the analysis was restricted to post-randomization data recorded by CRT-D devices. Subjects in the "No Implant Attempt", "Unsuccessful Implants", "CRT-D: Not Randomized", and all CRT-P subgroups were excluded.