REsynchronization reVErses Remodeling in Systolic Left vEntricular Dysfunction (REVERSE): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for REsynchronization reVErses Remodeling in Systolic Left vEntricular Dysfunction (REVERSE) (NCT00271154), a randomized na study. The primary outcome is Percentage of Patients Worsened for Clinical Composite Response.

2. Clinical Question

The trial addresses whether CRT activation is associated with a different pattern of clinical worsening in this population. The registry reports 21.5% worsened in CRT OFF and 16.0% in CRT ON. Its reported chi-squared comparison has p = 0.10. These figures describe the posted endpoint and comparison; they do not by themselves establish the size or clinical importance of a treatment effect. The record does not report a confidence interval for this primary comparison, so the precision of the difference cannot be assessed from a registry interval. The result should also be considered alongside the secondary measure of left ventricular end systolic volume indexed to body surface area, which captures a different, structural aspect of cardiac remodeling.

3. Trial Design

REVERSE was registered as a randomized, double-blind, parallel-group trial sponsored by Medtronic Cardiac Rhythm and Heart Failure. The registry lists the phase as NA. The condition was heart failure, and the study status is completed. Enrollment is recorded as 684. The start date is 2004-09 and primary completion is recorded as 2007-10.

The two described arms differed in whether CRT was active: CRT ON combined activation with optimal medical therapy, whereas CRT OFF combined deactivation with optimal medical therapy. The intervention description also identifies a CRT device or an implantable cardioverter defibrillator with CRT. Thus, the treatment contrast in the posted results concerns CRT activation within the device and medical-therapy setting, rather than a contrast between device implantation and no device procedure.

Random assignment and double masking are registry-recorded design features. The parallel model means participants were assigned to one arm rather than being described as receiving both conditions in sequence. The registry does not report the details of sequence generation, concealment, who was masked, or how masking was maintained. It also does not report the planned allocation ratio or explain the imbalance between the primary endpoint counts of 191 and 419. Those details should not be inferred from the observed counts.

The primary outcome was the percentage of participants worsened for clinical composite response at 12 months. The primary population is described as all randomized patients analyzed by intent to treat. A secondary posted outcome was change in left ventricular end systolic volume indexed, from baseline to 12 months. The registry lists two outcome measures and two analyses. Other details of the protocol, including the sample-size justification and any planned interim analyses, are not reported in the registry record. The sample-size figures associated with the continuous secondary result are planning illustrations based on the observed effect, not the trial's design calculation.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: A total of 684 patients were enrolled in the study. Of these, 642 patients were attempted for cardiac resynchronization therapy (CRT) device implant. 21 were unsuccessful, leaving 621 patients successfully implanted with a CRT device. Of these 621 patients, 610 were randomized.

Overall Study

Analysis population for the primary outcome, as defined in the registry: All randomized patients were included using Intent to Treat (ITT).

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Patients Worsened for Clinical Composite Response, measured over 12 Months.

The primary endpoint is the percentage of participants classified as worsened for clinical composite response at 12 months. The registry defines worsening as meeting any of several criteria: death; hospitalization for worsening heart failure; crossover to the other arm; deterioration in New York Heart Association functional class; or reporting moderately or markedly worse heart failure symptoms compared with before CRT implantation.

A composite endpoint combines multiple kinds of outcomes into one classification. This can capture a broader range of clinically relevant changes than a single-event measure. It also means that the result does not identify which component chiefly accounts for the overall percentage. Component-specific counts or results are not reported in the registry record, so the contribution of death, hospitalization, crossover, functional-class change, and symptom worsening cannot be disentangled from the posted primary result.

The registry reports the endpoint as a percentage of participants worsened and states that all randomized patients were included using intent to treat. This approach aims to preserve the randomized comparison by analyzing participants according to their assigned group, even if events such as crossover occur. The record does not give additional detail about how missing endpoint observations were handled. It reports 21.5% worsened in CRT OFF and 16.0% in CRT ON, with a chi-squared p-value of 0.10. No confidence interval for the primary comparison is reported, limiting assessment of the range of differences compatible with the data.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the primary comparison test?

The registry reports a chi-squared test for the primary endpoint, which is a count-based outcome expressed as the percentage worsened in each arm. In general, a chi-squared test evaluates whether the observed distribution of categories differs between groups, under assumptions about how observations are collected and classified. The registry reports p = 0.10 for this comparison. The test result summarizes compatibility between the observed group pattern and a no-difference model; it is not a measure of the magnitude or clinical importance of a difference. A primary confidence interval is not reported in the registry record.

The endpoint's definition matters for interpreting that test. It combines death, heart-failure hospitalization, crossover, worsening NYHA functional class, and worsening symptoms. A participant can meet the composite definition through different components, but the record does not give component-specific results. Consequently, the overall percentage cannot reveal whether the observed pattern was driven by one component or distributed across several.

What does the secondary mean difference represent?

The secondary outcome is the change in indexed left ventricular end systolic volume from baseline to 12 months. It is reported as a mean with a standard deviation in each arm. The recalculation compares the two posted arm means using an unadjusted mean difference. The estimate is -17.1 milliliters per meters squared, with a 95% confidence interval from -21.46 to -12.74. The negative direction follows the order used in the recalculation, CRT ON compared with CRT OFF. The posted arm means are -18.4 in CRT ON and -1.3 in CRT OFF.

In general, a confidence interval describes a range of values compatible with the data and analysis assumptions. A narrower interval indicates greater precision than a wider interval, but an interval does not include every source of uncertainty, such as bias from missing data or limitations in outcome measurement. The interval for the recalculated secondary contrast does not include zero, and the recalculated unadjusted p-value is p < 0.001. These are statistical descriptions of the posted summary data, not a measure of clinical importance.

The recalculation is unadjusted: it uses the posted summary statistics rather than participant-level data or a model adjusting for other variables. The record reports a two-sided t-test for the sponsor's analysis, with p < 0.001. The sponsor's test and the simple recalculation are not identical analytical procedures. Their p-values need not match exactly even when they point in the same direction. Details such as covariate adjustment or handling of missing values are not reported in the registry record.

Why do standard deviations and sample sizes matter?

A standard deviation describes spread among individual measurements around an arm mean; it is not the same as the uncertainty around the mean. The secondary arm standard deviations are 23.4 for CRT OFF and 29.5 for CRT ON. The outcome counts are 191 and 419, respectively. In general, a mean comparison's precision depends on both the variability of measurements and the number of observations contributing to the estimate. The registry summary does not describe whether all randomized participants contributed to this continuous outcome or how missing measurements were handled.

Unequal group sizes can affect the precision of a comparison, particularly when outcome variability differs across groups. The observed counts alone do not explain why the groups differ in size; the planned allocation and reasons for the reported counts are not reported in the registry record. No adjusted analysis or sensitivity analysis for this secondary endpoint is included in the posted results.

How should the planning illustrations be read?

The planning illustrations use the observed secondary effect as a reference and show participants per arm for 80% and 90% power at different fractions of that observed effect. For planning for 100% of the observed effect, the illustrations give 42 participants per arm for 80% power and 56 for 90% power. For planning for 75% of the observed effect, they give 74 and 99 participants per arm, respectively. For planning for 50% of the observed effect, they give 166 and 222 participants per arm, respectively.

These are hypothetical planning calculations, not observed power and not the original trial's sample-size justification. They assume a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation, and they are based on fractions of the observed effect. Planning from an observed effect can be optimistic if that estimate is imprecise or unusually large. The registry's own sample-size justification is not reported in the registry record. The illustrations should therefore be understood as conditional scenarios rather than as evidence that the trial was designed or powered in a particular way.

Why do endpoint type and test choice matter?

The primary endpoint is categorical: each participant is classified as worsened or not under the composite definition, and the reported analysis uses a chi-squared test. The secondary endpoint is continuous: it is summarized with arm means and standard deviations and compared using a two-sided t-test in the sponsor's analysis. Different endpoint types call for different summaries and tests. They also answer distinct questions: a categorical composite concerns the proportion classified as worsened, whereas a continuous volume measure concerns average change in a structural measurement.

In general, over-dispersion occurs when count data vary more than a model expects. It matters because a model that understates variability can produce uncertainty estimates that are too small. The primary outcome is reported as a two-group percentage comparison, and no over-dispersion analysis is reported in the registry record. This general issue should not be mistaken for a documented problem with the trial's analysis.

10. Analyses Reported by the Sponsor

For the primary endpoint, the registry reports a chi-squared comparison with p = 0.10. The record does not provide a sponsor-reported effect estimate or confidence interval for that comparison, so there is no posted primary recalculation to set alongside the sponsor's test result.

For the secondary indexed-volume outcome, the sponsor's reported analysis is a two-sided t-test with p < 0.001. The simple unadjusted recalculation from the posted arm means and standard deviations gives a mean difference of -17.1 milliliters per meters squared, a 95% confidence interval from -21.46 to -12.74, and p < 0.001. Both analyses indicate a difference in the same direction. The p-values differ in their reported precision, and the recalculation is not a replacement for the sponsor's analysis: it is based on summary statistics and is unadjusted.

The registry record does not provide enough detail to explain any analytical differences, including whether the sponsor used additional modeling or other data-handling procedures. No discrepancy is flagged in the posted information. Where a sponsor analysis and a recalculation conflict, the sponsor's own analysis takes precedence; the results do not identify such a discrepancy.

11. Multiplicity, Missing Data and Analysis Populations

The registry record does not report whether the trial used a multiplicity adjustment, a hierarchy of endpoints, or a prespecified sequence for testing the primary and secondary outcomes. It lists two outcome measures and two analyses, including the primary clinical composite and the secondary indexed-volume change. Without the full planned analysis strategy, it is not possible to determine how the secondary p-value should be interpreted in relation to other potential tests or whether additional outcomes were planned but not posted.

In general, multiplicity arises when several hypotheses are tested. If each is assessed without adjustment, the chance of at least one apparently unusual result can increase across the set of tests. Adjustment methods or a prespecified hierarchy can address that issue, but neither a method nor an endpoint hierarchy is reported in the registry record. This limitation is especially relevant when considering the secondary p-value as supportive evidence rather than as a stand-alone confirmatory result.

The primary population is described as all randomized patients analyzed using intent to treat. This is a useful design principle because it keeps participants in their randomized groups for analysis, including when crossover is part of the endpoint definition. However, the record does not report the number of missing primary outcome observations, how any missing outcomes were classified, or what methods were used to handle missing data. For the continuous secondary outcome, missing-data handling and the number of measurements contributing to each mean are also not reported in the registry record.

The primary endpoint counts are 191 for CRT OFF and 419 for CRT ON, while registry enrollment is 684. The registry does not explain the difference between these quantities, and the reason for the unequal endpoint counts is not reported. It would be inappropriate to infer whether this reflects the randomized allocation, endpoint availability, or another feature of reporting. The adverse-event section separately lists a Not Randomized group, but that does not resolve the primary outcome count difference.

Details of interim analyses, sample-size justification, and the full statistical analysis plan are not reported in the registry record. Their absence limits assessment of whether the analyses were prespecified and how the final results relate to the original design. These omissions do not invalidate the posted results, but they constrain how confidently the reported p-values and endpoint pattern can be interpreted.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: cardiac failure acute (15.7%), medical device change (10.7%), atrioventricular block complete (7.9%), ventricular tachycardia (7.2%), lv lead dislodgement (6.8%).

The safety record lists serious adverse events affecting 122 participants among 191 at risk in CRT OFF and 268 among 419 at risk in CRT ON. For other adverse events, it lists 61 affected among 119 at risk in CRT OFF and 140 among 491 at risk in CRT ON. These are counts with different denominators across the reported categories. The registry does not provide a comparative statistical analysis for these figures, and they should not be treated as a randomized estimate of relative safety without further information about definitions, follow-up, and ascertainment.

The most frequently listed serious adverse event terms include Cardiac failure acute (15.7%), Medical device change (10.7%), Atrioventricular block complete (7.9%), Ventricular tachycardia (7.2%), LV lead dislodgement (6.8%), and Cardiac failure chronic (5.3%). The record does not specify in this summary the denominator used for each listed percentage or provide event-level timing and recurrence information. The terms describe reported events; they do not by themselves establish causation by CRT activation or device implantation.

A Not Randomized group is also listed for safety, with 13 of 38 at risk for serious adverse events and 2 of 38 at risk for other adverse events. Death counts are not reported in the registry record. The safety summary should therefore be read as descriptive registry reporting, not as a complete causal or comparative assessment.

13. What the Result Does, and Does Not, Mean

The primary result describes the percentage of participants classified as worsened under the registry's composite definition at 12 months. The reported values are 21.5% in CRT OFF and 16.0% in CRT ON, with a chi-squared p-value of 0.10. This does not provide strong statistical evidence of a difference under the reported test, but it does not establish equivalence or rule out a meaningful difference. The lack of a primary confidence interval makes the uncertainty around the comparison harder to judge.

The secondary outcome reports a different kind of information: average change in indexed left ventricular end systolic volume. Its posted analysis indicates a clear statistical contrast, but that does not convert the primary result into a positive primary finding. Nor does a structural measurement alone establish how participants felt or functioned. The clinical composite and the volume measure should be interpreted as distinct outcomes, each with its own definition and limitations.

Statistical evidence and clinical importance are related but not interchangeable. A p-value summarizes the compatibility of data with a specified statistical model; it does not describe the size, importance, or probability of benefit. A confidence interval can help convey precision, but even an interval cannot address every source of bias or uncertainty. The primary endpoint lacks a reported interval, while the secondary interval applies to an unadjusted recalculation from summary statistics.

The results also do not establish that CRT caused every observed event or change, and the safety counts are descriptive rather than a reported comparative causal analysis. Details of missing-data handling, multiplicity, allocation procedures, and the full analysis plan are not reported in the registry record. Those omissions matter when assessing the strength and scope of conclusions. The findings are best stated narrowly: the registry reports no statistically persuasive primary comparison by its chi-squared test, alongside a statistically clear secondary volume comparison. Neither result alone supports a broad claim about all clinical outcomes.

Nothing in these registry results constitutes medical advice. The meaning of any intervention for an individual depends on clinical circumstances and evidence beyond this statistical summary.

14. Why This Trial Matters Statistically

REVERSE illustrates why a trial can produce different statistical stories across endpoint types. Its primary outcome is a clinical composite intended to capture worsening through several events and changes in status. The secondary outcome is a continuous structural measure. The primary comparison has p = 0.10, while the sponsor's secondary analysis reports p < 0.001. Treating these as interchangeable would obscure the distinction between a clinical classification and an average change in ventricular volume.

The example also shows why an effect estimate and its precision matter alongside a p-value. The primary result has no reported confidence interval, leaving its precision difficult to assess from the registry. By contrast, the secondary recalculation has a mean difference and confidence interval, though that calculation is unadjusted and based on posted summary statistics. The level of detail available for each outcome is not the same, and conclusions should reflect that difference.

Composite endpoints can broaden the range of outcomes captured, but they can also hide which component drives the overall result. Here, the definition includes death, hospitalization, crossover, functional-class worsening, and symptom worsening, while component-specific results are not reported in the registry record. This matters because a single composite percentage cannot explain the clinical pattern behind it.

Finally, the record demonstrates the limits of registry summaries. The design identifies randomization, double masking, and parallel groups, but does not report important details such as the allocation procedure, missing-data methods, multiplicity strategy, or sample-size justification. A statistical reading can describe the posted evidence carefully, but it should not fill those gaps by assumption. Clear separation of reported findings, simple recalculation, and unreported design details is essential for interpreting trial results responsibly.

15. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

Several trial-specific limitations constrain interpretation of the registry results. The primary outcome is a composite that combines death, heart-failure hospitalization, crossover, NYHA functional-class worsening, and symptom worsening. Component-level counts and comparisons are not reported in the registry record, so the overall percentage cannot show which types of events or changes contributed most to classification as worsened.

The primary arm counts are 191 for CRT OFF and 419 for CRT ON, while enrollment is 684. The registry does not explain this difference or report the planned allocation ratio. The reason for the unequal counts cannot be determined from the record. Although the primary population is described as intent to treat, the handling of missing primary outcomes is not reported. The record also does not give the number of observations contributing to each secondary mean or describe missing-data handling for that measure.

The primary result lacks a reported confidence interval, so its precision cannot be assessed directly. For the secondary outcome, the recalculation is unadjusted and based on summary statistics; it is not a participant-level analysis. The sponsor's analysis is reported as a two-sided t-test, but additional analysis details are not available in the registry record. No multiplicity strategy, complete outcome hierarchy, or interim-analysis plan is reported.

Other design details, including the randomization procedure, allocation concealment, the people masked, and how masking was maintained, are not reported in the registry record. The sample-size justification is also not reported. The planning illustrations based on the observed secondary effect cannot fill that gap because they are not the trial's own design calculation. These limitations do not erase the observed results, but they make the registry record an incomplete basis for assessing bias, precision, and the full prespecified analysis.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: REsynchronization reVErses Remodeling in Systolic Left vEntricular Dysfunction (REVERSE) — NCT00271154.
• Registry record last updated 2012-01-30; status last verified 2012-01.
• Results first posted 2009-05-13.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

The REVERSE registry record describes a completed randomized, double-blind, parallel-group trial sponsored by Medtronic Cardiac Rhythm and Heart Failure. It lists heart failure as the condition and enrollment as 684. The primary endpoint was the percentage worsened for clinical composite response at 12 months, analyzed in all randomized patients using intent to treat. The reported values were 21.5% for CRT OFF and 16.0% for CRT ON, with a chi-squared p-value of 0.10.

The posted secondary outcome was change in indexed left ventricular end systolic volume from baseline to 12 months. Its sponsor-reported two-sided t-test had p < 0.001; an unadjusted recalculation from the posted summary statistics gave a mean difference of -17.1 and a 95% confidence interval from -21.46 to -12.74. The record does not report key details including the primary confidence interval, missing-data procedures, multiplicity strategy, or sample-size justification. The primary and secondary outcomes therefore warrant separate interpretation.