Clinical Study of Lipoic Acid on Ischemic Heart Failure: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Clinical Study of Lipoic Acid on Ischemic Heart Failure (NCT03491969), a randomized phase 4 study. The primary outcome is Number of Participants That Had First Occurrence of the Composite Endpoint.

2. Clinical Question

The primary question is about a composite outcome: an event is counted when either of its specified components occurs first. It therefore combines death from any cause with hospitalization for heart failure. The registry lists 32 participants with the endpoint in the alpha-lipoic-acid group and 40 in the placebo group. The corresponding outcome denominators are 138 and 139 participants. These counts describe the posted endpoint results; they do not by themselves establish the size or certainty of a treatment effect.

The registry lists the study as randomized and double-blind. Those design features are relevant when interpreting a comparison between groups, but the record does not describe how the randomization sequence was generated, how allocation was concealed, or how masking was maintained. Details about the primary analysis population are also not reported in the registry record. The clinical question can thus be stated from the intervention, comparator, and endpoint descriptions, while several design and analysis details remain unspecified.

3. Trial Design

The registry record identifies NCT03491969 as a phase 4, randomized, double-blind study with a sequential model. Shanghai Zhongshan Hospital is listed as the sponsor. The condition is ischemic heart failure, and the record lists an enrollment of 300. The study status is completed. The recorded start date is 2019-01-01, and the primary completion date is 2025-02-01.

The intervention group received alpha-lipoic acid (α-LA) 200 mg three times daily alongside standard heart-failure treatments. The comparator group received placebo 200 mg three times daily alongside those treatments. Both arm descriptions state a total treatment duration of 24 months during the double-blind treatment period. The registry describes the target population in its brief summary as patients with ischemic heart failure, NYHA Class II–IV, and an ejection fraction of 50% or less.

The posted primary outcome is the number of participants with a first occurrence of the composite endpoint, defined as all-cause death or heart-failure hospitalization, up to 24 months. The outcome results list 138 participants in the alpha-lipoic-acid group and 139 in the placebo group. The record does not explain how these endpoint denominators relate to the listed enrollment of 300, or why they differ from that enrollment figure. Those reasons are not reported in the registry record.

The record lists one outcome measure and no secondary outcome results. It lists no analyses in its analysis records, although a between-group estimate and an unadjusted p-value are posted for the primary outcome. Further information about the randomization procedure, allocation concealment, masking procedures, follow-up completeness, analysis population, and prespecified statistical plan is not reported in the registry record. The label “sequential” is present, but the record does not explain what it means for the conduct or analysis of this study.

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.

Overall Study

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.

Population: The primary efficacy endpoints were analyzed using the full analysis set (FAS), which followed the Intent-to-Treat (ITT) principle. Inclusion in the FAS required that participants receive the investigational drug for a minimum of 1 month. 300 patients were randomized in a 1:1 ratio to receive either ALA or a matching placebo. In the ALA group, 12 patients were excluded from the primary analysis. In the placebo group, 11 participants were excluded.

6. Primary Endpoint

The registry lists the primary outcome as Number of Participants That Had First Occurrence of the Composite Endpoint, measured over up to 24 months.

The primary endpoint counts participants who had a first occurrence of either all-cause death or hospitalization for heart failure. The timeframe is up to 24 months. This is a composite endpoint: a participant meeting either component qualifies for the combined outcome. The posted results give participant counts for the composite, rather than separate counts for each component.

In general, a composite endpoint can capture more than one clinically important event within a single outcome definition. It can also make the overall result harder to interpret when components differ in importance, frequency, or response to an intervention. A combined count does not show which component contributed most to the result. The registry record does not provide component-specific results for this primary endpoint, so the relative contribution of death and heart-failure hospitalization cannot be determined from the posted outcome.

The endpoint is expressed as a count of participants with an event, not as a time-to-event estimate. Although the outcome is defined over a period of up to 24 months, the record does not post a hazard ratio, event-time analysis, or details about censoring. The listed risk difference and relative measures compare the participant-level event results between the intervention and placebo groups. They should be interpreted as comparisons of the posted endpoint data, not as component-specific or time-to-event findings.

The registry record does not describe the endpoint’s adjudication process or provide a more detailed definition of heart-failure hospitalization. Those details may matter when considering how consistently the endpoint was identified, but they are not reported in the registry record.

7. Results Posted in the Registry

• Alpha-Lipoic Acid(α-LA): 23.2% (32/138) versus 28.8% (40/139) on control; risk difference -5.6 percentage points (95% CI -15.9 to 4.7), risk ratio 0.81 (0.54–1.20); unadjusted two-proportion z-test p = 0.289.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is binary at the participant level: a participant either had a first occurrence of the composite endpoint during the stated timeframe or did not. The registry reports the number of participants with an event and the total number in each group. A comparison of these proportions can be expressed on an absolute scale as a risk difference, or on relative scales as a risk ratio and an odds ratio. The record posts all three measures for alpha-lipoic acid compared with placebo.

What does the risk difference express?

In general, a risk difference subtracts one group’s event proportion from the other group’s event proportion, following a stated direction for the comparison. It expresses the contrast on the absolute probability scale and is anchored at zero when the proportions are equal. The posted risk difference is -0.056, with a 95% confidence interval from -0.159 to 0.047. The interval includes zero, so the data are compatible with a range of absolute differences that includes no difference. The sign and interpretation depend on which group is subtracted from which; the posted comparison is for alpha-lipoic acid against placebo.

How do relative measures differ?

In general, a risk ratio compares the event proportions as a ratio, with 1 representing equal proportions. An odds ratio compares the odds of an event between groups, also with 1 representing equal odds. These measures use different scales and need not have the same numerical value. Here, the posted risk ratio is 0.806, with a 95% confidence interval from 0.540 to 1.203, and the posted odds ratio is 0.747, with a 95% confidence interval from 0.436 to 1.282. Both intervals include 1. The registry record does not give details of any adjustment or model used for these estimates.

What does the p-value contribute?

The posted unadjusted p-value is 0.289. In general, a p-value quantifies how unusual data at least as discrepant as those observed would be under a specified null model and the assumptions of the analysis. It is not the probability that the null hypothesis is true, nor does it measure the size or clinical importance of a difference. The term “unadjusted” identifies the posted value as unadjusted; the record does not report a multiplicity adjustment or a detailed test specification. The confidence interval provides complementary information by displaying a range of effect estimates compatible with the data under the interval method’s assumptions.

Why does the composite definition matter?

The event count combines first occurrences of all-cause death and heart-failure hospitalization. In general, a composite may increase the number of events available for an overall comparison, but the combined estimate can conceal differences between components. One component could contribute more events than another, or component-specific patterns could differ. The registry record gives no separate primary-outcome counts for these components, so no component-level interpretation can be drawn from the posted composite result.

What analysis details are documented?

The record posts participant counts, an absolute risk difference, relative estimates, and an unadjusted p-value. It lists zero analyses in its analysis records. It does not report the statistical test or model specification, the analysis population, covariate adjustment, missing-data methods, or an approach to censoring. It also does not describe how outcome events were confirmed. These details are needed to evaluate fully whether the posted estimate follows the prespecified analysis and to judge the assumptions behind its uncertainty measures.

In general, randomization supports a fair comparison by assigning participants to groups according to a chance-based process, while masking can reduce some forms of influence on care or outcome assessment. The registry labels the study randomized and double-blind, but does not describe the sequence-generation, concealment, or masking procedures. The record therefore identifies these design features without documenting how they were implemented.

The sample-size figures are planning illustrations based on the observed effect, not the trial’s original design calculations. Their stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation; the effect to detect is the stated fraction of the observed effect. The original design assumptions are not reported in the registry record. Such illustrations address hypothetical future planning and should not be interpreted as a measure of the trial’s evidential strength.

10. Confidence Intervals and Precision

The risk-difference estimate is -0.056, and its 95% confidence interval runs from -0.159 to 0.047. In general, a confidence interval describes the range of effect values reasonably compatible with the observed data under the statistical method and its assumptions. It conveys uncertainty around an estimate; it is not a guarantee that the true value lies in the interval, nor does it assign a probability to a particular treatment effect after the data have been observed.

For the risk difference, zero represents no difference between the compared proportions. Because the posted interval includes zero, it is compatible with no difference as well as with differences in either direction. The interval is broad enough that the point estimate alone should not be read as a precise estimate of the size or direction of the underlying effect. The posted unadjusted p-value of 0.289 is consistent with the uncertainty conveyed by this interval, but it does not replace the interval’s information about the range of possible effects.

The relative estimates also have intervals spanning their no-difference reference value. The risk ratio is 0.806, with a 95% confidence interval from 0.540 to 1.203; the odds ratio is 0.747, with a 95% confidence interval from 0.436 to 1.282. Their point estimates are below 1, but their intervals include 1 and values above it. Thus, neither relative interval gives a narrowly determined direction of the comparison.

In general, interval width is influenced by the amount of information in the data and by the variability of the outcome. The registry record does not report an interval method or an adjusted analysis, so the assumptions underlying the posted intervals cannot be assessed in detail. Precision should be judged from the full intervals, not from the point estimates alone or from whether a p-value crosses a chosen threshold.

11. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are expressed as participants per arm for the binary participant-event outcome. They are not total event counts. The stated planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set at the stated fraction of the observed effect. The figures are illustrations based on the observed effect; the trial’s own original design assumptions are not reported in the registry record.

The 80% power column gives the participants per arm that the planning calculation estimates would be needed for an 80% chance of detecting the specified effect under those assumptions, if that effect were the true one and the analysis assumptions held. The 90% power column gives the corresponding per-arm requirement for a 90% chance. These are planning targets, not observed outcomes and not guarantees that a future study will detect an effect.

The rows scale the effect used for planning: 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. For each of those planning scenarios, the figures give the per-arm requirement at both power levels. A smaller effect is harder to distinguish from random variation, so detecting a smaller effect requires more participants. The illustrations are therefore sensitive to the observed effect used as their starting point; they should not be treated as the original rationale for the enrollment listed in the registry.

This is not observed power. It is a forward-looking planning calculation under stated assumptions, not a measure of how likely the completed trial was to produce a statistically significant result. The registry record does not report the original sample-size justification or whether its own planned target was based on different assumptions.

12. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.288; treatment proportion p₁ = 0.232; average p̄ = 0.260.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6202 = 1.2155.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6189 = 0.5209.
4. n = (1.2155 + 0.5209)² / (-0.0559)² = 965.4, rounded up to 966 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
p0 <- 0.287769784172662                          # control proportion
p1 <- 0.231884057971014                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(3491969)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 966 (80% power), 1,292 (90% power)
 75% of observed effect: n per arm = 1,747 (80% power), 2,338 (90% power)
 50% of observed effect: n per arm = 3,997 (80% power), 5,350 (90% power)
power.prop.test: 965.4 
Simulated power at n = 966 per arm: 0.800 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

13. Multiplicity, Missing Data and Analysis Populations

The registry record lists one outcome measure and no secondary outcome results. It also lists zero analyses in its analysis records, while posting a primary-outcome comparison and an unadjusted p-value. The record does not describe a prespecified hierarchy of outcomes, a testing sequence, or any correction for multiple comparisons. Consequently, it is not possible to assess from the registry record how multiplicity was handled in the study’s full analysis plan. The presence of a single posted outcome does not establish that no other outcomes were planned or measured.

Information about missing outcome data is not reported in the registry record. The record does not state how many participants had incomplete endpoint follow-up, whether any participants were excluded from the endpoint analysis, how missing status was handled, or whether sensitivity analyses were performed. The listed enrollment is 300, while the endpoint groups have denominators of 138 and 139. The reason for the difference is not reported, and the posted record does not explain whether it reflects eligibility, follow-up, analysis inclusion, or another factor.

The analysis population is also not described. The registry record does not say whether the primary comparison used all randomized participants, participants who received treatment, participants with available outcome data, or another defined group. It would therefore be inappropriate to label the posted comparison as an intention-to-treat analysis or to infer a per-protocol analysis. No details are reported about covariate adjustment, subgroup analyses, imputation, or sensitivity analyses.

These gaps matter because outcome selection, missingness, and analysis-population definitions can affect how a treatment comparison is interpreted. Without the relevant information, readers cannot determine whether the posted estimate reflects the intended analysis or evaluate the impact of incomplete data. The registry record supports interpretation of the numbers it posts, but not reconstruction of an undocumented analysis strategy.

14. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: hospitalization for worsened heart failure (4.9%), death (3.5%), hospitalization for stroke (2.1%), hospitalization for fracture (0.7%), hospitalization for arterial occlusive disease (0.7%).

The registry lists serious adverse events for 12 participants at risk of 145 in the alpha-lipoic-acid group and 14 participants at risk of 144 in the placebo group. It lists other adverse events for 31 participants at risk of 145 and 25 participants at risk of 144, respectively. These are descriptive counts from the safety record. They are separate from the primary composite endpoint, and the record does not provide enough detail to equate the safety counts with the primary-outcome counts.

The listed serious adverse-event terms and percentages are hospitalization for worsened heart failure, 4.9%; death, 3.5%; hospitalization for stroke, 2.1%; hospitalization for fracture, 0.7%; and hospitalization for arterial occlusive disease, 0.7%. These are presented as registry terms, not as a complete account of all event types or a comparison of event rates between arms. The record does not give arm-specific counts for each listed term in the information summarized by these terms.

Adverse-event counts do not by themselves establish whether an event was caused by treatment. The registry record does not report event definitions, severity beyond the serious classification, timing, relatedness assessments, or comparative statistical tests. The safety figures should therefore be read descriptively, with the posted at-risk denominators and the limits of the event detail in mind.

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

The posted data show 32 participants with the primary composite endpoint among 138 in the alpha-lipoic-acid group and 40 among 139 in the placebo group. The risk-difference point estimate is -0.056, but its 95% confidence interval, from -0.159 to 0.047, includes zero and extends in both directions. The unadjusted p-value is 0.289. The relative risk and odds-ratio intervals also include their no-difference reference value. Together, these statistics describe an uncertain comparison rather than a precisely determined difference.

The result does not establish which group had a definitively lower underlying risk, and it does not establish that the groups were equivalent. A confidence interval that includes no difference can be compatible with a range of effects; it should not be translated into proof that there is no effect. Conversely, a point estimate below the no-difference value is not sufficient to establish a reliable reduction. The degree of uncertainty is important to the interpretation.

The endpoint combines all-cause death and heart-failure hospitalization, so the composite result does not show how either component behaved separately. The registry record does not post secondary outcome findings, subgroup results, or an analysis population. It also does not report missing-data handling, an adjudication process, or a detailed statistical plan. Those omissions limit judgments about the robustness and scope of the comparison.

The record identifies a completed, randomized, double-blind study, but does not describe how randomization or masking was implemented. It lists enrollment of 300 and endpoint denominators of 138 and 139 without explaining the relationship between them. These details should be considered alongside the statistical uncertainty when interpreting the result. The posted figures describe the recorded outcome comparison; they do not by themselves support medical advice or a definitive conclusion about treatment benefit or harm.

In general, a study result is most informative when its estimate, uncertainty, outcome definition, design, and analysis details are considered together. In this case, the point estimate is accompanied by intervals that span the no-difference values, and important analysis details are not reported in the registry record. The interpretation should remain confined to the posted composite endpoint and its documented comparison.

16. Why This Trial Matters Statistically

This trial is statistically instructive because its primary outcome is a composite of events with different meanings: all-cause death and heart-failure hospitalization. A single summary estimate can be useful for describing the combined endpoint, but it cannot show whether the components contributed similarly. Readers need component-level results to understand the clinical makeup of a composite. Those results are not reported in the registry record.

The posted absolute and relative measures provide different views of the same comparison. The risk difference gives an absolute-scale contrast, while the risk ratio and odds ratio use relative scales. Their point estimates are below their respective no-difference values, but all the posted confidence intervals include those reference values. This illustrates why conclusions should not be based solely on the direction of a point estimate. The uncertainty intervals are central to understanding how much the data determine the comparison.

The endpoint counts also highlight the importance of transparent denominators. The record lists an enrollment of 300 but reports endpoint denominators of 138 and 139. Without an explanation of the relationship, readers cannot determine why the numbers differ or how the analysis population was defined. Clear reporting of participant flow, follow-up, and missing outcome data would help assess the completeness and representativeness of the endpoint analysis.

Finally, the planning illustration is distinct from the original trial design. The original sample-size assumptions are not reported, while a separate illustration uses the observed effect and stated planning assumptions. This distinction matters: planning calculations can help frame future study sizes, but they do not explain how the completed trial was designed or quantify its evidential strength. The record’s limited analysis details make careful, bounded interpretation especially important.

17. 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 details needed to evaluate the posted comparison are not reported in the registry record. Although the record identifies the study as randomized and double-blind, it does not describe sequence generation, allocation concealment, or how masking was maintained. It also does not explain the sequential model label. These missing details limit assessment of the design’s implementation.

The registry lists enrollment of 300, but the primary endpoint results use denominators of 138 and 139. The reason for this difference is not reported. Participant-flow information, withdrawals, follow-up completeness, exclusions, and missing endpoint data are also not described. Without these details, readers cannot evaluate whether the endpoint analysis included the intended population or determine how incomplete follow-up may have affected the result.

The primary outcome is a composite of all-cause death and heart-failure hospitalization, but no component-specific primary results are posted. The record gives no event-adjudication details or further definition of heart-failure hospitalization. It also does not report secondary outcome results, subgroup analyses, or a statistical analysis plan. Thus, the posted comparison cannot be checked against a documented analysis strategy, and the component-level pattern cannot be assessed.

The record does not provide the original sample-size justification, analysis population, missing-data method, model specification, or multiplicity procedure. The observed-effect planning illustration is not a substitute for these original design details. Safety reporting is also limited: counts and selected serious-event terms are posted, but the registry record does not provide a full description of event timing, relatedness, or arm-specific counts for each listed term. These limitations do not change the posted numbers, but they constrain how fully those numbers can be interpreted.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Clinical Study of Lipoic Acid on Ischemic Heart Failure — NCT03491969.
• Registry record last updated 2025-07-14; status last verified 2025-07.
• Results first posted 2025-07-11.
• 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.

21. Record Summary

NCT03491969 is listed as a completed, phase 4, randomized, double-blind study of alpha-lipoic acid for ischemic heart failure, sponsored by Shanghai Zhongshan Hospital. The registry lists enrollment of 300, a start date of 2019-01-01, and a primary completion date of 2025-02-01. The brief summary describes patients with NYHA Class II–IV and an ejection fraction of 50% or less.

The comparison was alpha-lipoic acid 200 mg three times daily plus standard heart-failure treatment versus placebo 200 mg three times daily plus standard treatment, for a stated treatment duration of 24 months. The primary endpoint was first occurrence of all-cause death or heart-failure hospitalization up to 24 months. The endpoint counts were 32 of 138 and 40 of 139, respectively. The posted risk difference was -0.056, with a 95% confidence interval from -0.159 to 0.047 and an unadjusted p-value of 0.289.

The interval includes zero, and the record does not report component-specific primary results or key analysis details. The result should be read as an uncertain comparison of the posted composite endpoint, not as a definitive conclusion about either treatment.