Metformin, Muscle Energetics, and Vascular Function in Older Adults With Peripheral Artery Disease: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Metformin, Muscle Energetics, and Vascular Function in Older Adults With Peripheral Artery Disease (NCT01901224), a randomized phase 4 study. The primary outcome is Change in PCr Recovery Time.

2. Clinical Question

The study’s broader purpose was to investigate whether metformin could improve walking ability in people with PAD. The registered primary endpoint, however, was change in PCr recovery time, a muscle energetics measure; walking ability is not identified as the primary endpoint. Flow-mediated dilation (FMD) was listed as a secondary outcome.

The registry record says the study was terminated prematurely because it was not funded, that randomization remained blinded, and that outcome-measure data were not collected. Consequently, the clinical question cannot be answered from the posted results: there is no reported comparison of change between the assigned groups. The registry enrollment field lists 2, while the study summary described an aim to enroll up to 100 subjects to find 60 subjects with PAD at Brigham and Women’s Hospital. The record does not establish how many participants were randomized or what information, if any, was collected outside the stated outcome measures.

3. Trial Design

This was a Phase 4 randomized, triple-masked, parallel-group study sponsored by Brigham and Women’s Hospital. The listed interventions were metformin 1000 mg and placebo. A parallel design assigns participants to a group and compares groups over the study period, rather than having each participant receive both interventions. Randomization is intended to make assigned groups comparable on average; it cannot ensure comparability in a small or incomplete study.

The registry describes a titration schedule: the metformin group was to start at 500 mg twice daily and increase to 1000 mg twice daily after one week. The placebo group was to follow a matching tablet schedule to maintain masking during titration. The listed primary endpoint was change in PCr recovery time measured at baseline and 12 weeks. FMD was listed as a secondary outcome. Further details about assessment procedures, the full schedule of visits, adherence monitoring, or how the masking was implemented are not reported in the registry record.

The registry lists a study start of 2013-07 and primary completion of 2015-06, and gives the status as terminated. Its enrollment field lists 2. Separately, the study summary described a plan to enroll up to 100 subjects in order to find 60 subjects with PAD at Brigham and Women’s Hospital. These are different registry statements; the record does not explain their relationship or establish the number randomized, treated, or included in any outcome assessment.

The record explicitly states that the study was not funded, was terminated prematurely, and did not collect data on the outcome measures. It also states that the randomization blind was never broken. A statistical analysis plan, stopping rules, allocation ratio, and details of any interim monitoring are not reported in the registry record. The design description therefore identifies the intended comparison, but it does not establish that a treatment comparison was carried out.

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: Randomization blind was never broken, so only overall study participant flow is available.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Study was not funded and was terminated prematurely. Randomization blind was never broken, and data was not collected on the outcome measures.

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: Randomization blind was never broken, so only overall study participant baseline measures are available.

6. Primary Endpoint

The registry lists the primary outcome as Change in PCr Recovery Time, measured over baseline, 12 weeks.

The primary endpoint was change in PCr recovery time, measured in seconds, from baseline to 12 weeks. PCr is described in the registry as a transport molecule and reservoir of high-energy phosphate bonds involved in cellular energetics. Recovery after exercise depends in part on skeletal-muscle mitochondrial oxidative phosphorylation. The endpoint was intended as an in vivo measure of mitochondrial function in skeletal muscle.

The registry’s description gives the direction of interpretation: shorter PCr recovery times are better, and higher PCr relative to P(i) during recovery is better. This endpoint offers a physiological measure related to muscle energy metabolism. It is not itself a direct measure of how far a participant can walk or of symptoms during walking. The study summary frames walking ability as a broader research interest, but the registered primary endpoint is muscle recovery time.

Change is ordinarily assessed by relating a follow-up measurement to a baseline measurement, then comparing that change between groups. The exact statistical definition of change, the instrument and measurement protocol, and the analysis method are not reported in the registry record. Most importantly, the record states that outcome-measure data were not collected. It therefore contains no observed endpoint values, group changes, or estimate of the difference in change. The biological rationale explains why the measure was chosen; it does not establish what the study would have found.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What can randomization establish?

In general, random allocation is used to reduce systematic differences between groups at the start of a trial. It does not guarantee that groups will be identical, and its value depends on carrying out the allocation and analyzing the resulting data. Here, the registry describes a randomized design, but it also states that outcome-measure data were not collected and that the blind was never broken. Randomization by itself cannot produce an estimate of treatment effect when there are no reported endpoint measurements to compare. The allocation sequence, concealment procedures, and analysis population are not reported in the registry record.

What does masking contribute?

In general, masking can reduce the influence of knowledge about assigned treatment on participant expectations, care, outcome assessment, or analysis decisions. The registry labels the design triple-masked and describes a placebo schedule intended to maintain masking during titration. It does not identify the masked roles or give operational details. Because the blind was never broken, the registry does not report unmasking for outcome analysis. Masking can help protect a comparison from some biases, but it cannot replace outcome measurement or resolve the absence of results.

How would change in the endpoint be compared?

In general, a trial comparing change from baseline can estimate change within each group and then compare those changes between groups. Another approach can model follow-up values while accounting for baseline values. These approaches target related questions but may differ in their assumptions and precision. The registry identifies baseline and 12 weeks as the measurement times, but does not define the exact change calculation or specify a statistical model. It reports no endpoint values or estimated group contrast, so no method can be reconstructed from the record.

How should uncertainty be read?

In general, a confidence interval describes a range of effects that are reasonably compatible with the data and the method used to calculate it. Its width reflects statistical uncertainty as well as the variability and amount of information in the data. A p-value, when used, describes how unusual data at least as inconsistent with a specified null hypothesis would be under that hypothesis and the model assumptions. Neither statistic indicates the probability that a treatment works. No confidence interval, p-value, or primary-endpoint estimate is reported in the registry record. With no outcome data, there is no trial-specific uncertainty to quantify.

What do planning calculations mean?

In general, sample-size planning asks how much information a future study may need to detect a specified effect under stated assumptions. A planning illustration is not an observed treatment effect, and it is not observed power. The registry’s planning description refers to two-sided significance level 0.05, equal allocation, a normal approximation, and detection of a stated fraction of an observed effect; it also distinguishes participants per arm from total events for time-to-event outcomes. These are assumptions for illustrations, not evidence that the terminated study completed an analysis. Since the record says outcome measures were not collected, it reports no trial-derived observed effect that can answer the clinical question. The actual design justification, target effect, and sample-size rationale are not reported in the registry record.

Could other statistical issues affect a future analysis?

In general, outcome distributions, repeated measurements, departures from model assumptions, and missing observations can influence estimates and their uncertainty. Over-dispersion, for example, occurs when variability exceeds what a statistical model assumes; accounting for it can improve uncertainty estimates when applicable. Such considerations are general, not descriptions of analyses used in this study. The registry does not report an analysis plan, model diagnostics, covariate adjustments, subgroup analyses, or interim analyses. It gives no outcome observations with which to evaluate distributional assumptions or fit a model.

For this record, the most consequential methodological point is not a choice between statistical models. It is that the planned outcome measurements were not collected, according to the registry. Methods such as baseline adjustment, repeated-measures analysis, or robust variance estimation require data and a defined analysis strategy. None can recover a treatment contrast from absent outcome measurements. The registered design and endpoint describe what investigators intended to study; they do not establish an observed effect.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists change in PCr recovery time as the primary outcome and change in FMD as a secondary outcome. It also reports six outcome measures in total. The other outcome descriptions and any planned hierarchy among them are not reported in the registry record. A multiplicity strategy—such as prioritizing one primary comparison, adjusting for multiple tests, or treating secondary analyses as exploratory—is not reported in the registry record. With no outcome data collected, the record reports no endpoint tests whose multiplicity could be assessed.

Missing data usually refers to measurements that were intended but are absent for some participants or time points. This record describes a broader absence: data were not collected on the outcome measures. It does not report participant-level reasons for missingness, withdrawals, adherence, or any methods for handling missing observations. Standard approaches such as imputation or complete-case analysis cannot be inferred, and no such approach is reported in the registry record.

The analysis population is also not reported in the registry record. It does not specify whether a planned analysis would include all randomized participants, participants who began treatment, or another defined population. The enrollment field lists 2, but it does not establish randomization or endpoint availability. A participant flow, exclusions, and analysis denominators are not reported.

The overall-study safety entry lists zero participants at risk and zero events. Because the registry explicitly says outcomes were not collected, those entries do not establish that no events occurred. They should not be interpreted as a complete safety follow-up or as an analysis population with no events. The record does not describe how adverse events were sought or summarized.

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

The study record establishes a research question and an intended randomized comparison, not a clinical finding. Investigators planned to compare metformin with placebo in older adults with PAD, using change in PCr recovery time as the primary endpoint. The registry says the study was terminated prematurely because it was not funded and that outcome-measure data were not collected. Therefore, the posted results do not indicate whether PCr recovery time changed, whether the groups differed, or whether walking ability or FMD differed.

Absence of a result is not evidence of no effect. It is also not evidence of benefit, harm, equivalence, or safety. Those interpretations require observed data and a defined analysis. The study’s physiological rationale for PCr recovery time and the proposed metformin regimen do not substitute for a measured treatment contrast.

The record’s enrollment field and its recruitment target describe different aspects of the study and do not establish who was randomized or analyzed. The statement that the blind was never broken does not resolve that uncertainty. Likewise, the zero entries in the overall-study adverse-event listing should not be treated as proof of no harms because the record says outcome data were not collected.

For readers, the appropriate conclusion is about the limits of the evidence in this registry record: it documents a planned trial that did not yield reported outcome results. It cannot support a conclusion about the effect of metformin on muscle energetics, vascular function, walking ability, or safety in this study population. This limitation concerns what the record can establish; it is not a recommendation about treatment.

12. Why This Trial Matters Statistically

This trial matters statistically because it illustrates the difference between a registered design and an evaluable result. A randomized, masked design can be a framework for reducing bias, but its inferential value depends on outcome collection and analysis. When the record says outcome measures were not collected, there is no estimate to interpret, regardless of the intended allocation or masking.

It also illustrates why enrollment figures and treatment targets should not be treated as interchangeable. The registry enrollment field lists 2, while the study summary describes an aim to enroll up to 100 subjects to find 60 subjects with PAD at Brigham and Women’s Hospital. The record does not explain the discrepancy or establish the number randomized or assessed. Readers should not infer an analysis sample from either statement.

The primary endpoint is a physiological measure, while the study summary emphasizes walking ability. That distinction matters: an outcome may be biologically relevant without directly measuring the broader experience or function of interest. The registry identifies PCr recovery time as primary and FMD as secondary, but no collected outcomes are reported for either.

Finally, an empty or zero-filled safety entry is not the same as evidence of no adverse events when ascertainment is not documented and the record states that outcomes were not collected. Clear separation of planning information, trial conduct, endpoint measurement, and analysis is essential for sound interpretation. In this case, the most informative statistical conclusion is that no treatment-effect inference can be drawn from the posted record.

13. 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.

The principal limitation is explicit: the registry states that the study was terminated prematurely because it was not funded and that outcome-measure data were not collected. Thus, the primary endpoint and listed secondary outcome have no reported measurements or comparative estimates. No confidence intervals, p-values, or adjusted results for the clinical outcomes are reported in the registry record.

The enrollment information is difficult to interpret. The registry enrollment field lists 2, while the study summary describes plans to enroll up to 100 subjects to find 60 subjects with PAD at Brigham and Women’s Hospital. The record does not reconcile these statements, and neither establishes how many participants were randomized, treated, followed, or eligible for an endpoint analysis. A participant-flow account is not reported in the registry record.

Additional limitations include the absence of a reported statistical analysis plan, a defined analysis population, allocation details, and a multiplicity strategy. The full eligibility criteria and operational details of endpoint measurement are not reported in the registry record. The registry also lists zero participants at risk and zero adverse events in an overall-study entry, but the stated lack of collected outcome data prevents interpreting those entries as evidence about safety.

These limitations prevent estimation of efficacy, precision, or safety from this study record. They do not demonstrate that the planned design would necessarily have produced a particular result, and they do not establish how the intervention would perform in other settings or populations.

14. Related Tutorials

15. Related Statistical Calculators

16. Sources

• ClinicalTrials.gov: Metformin, Muscle Energetics, and Vascular Function in Older Adults With Peripheral Artery Disease — NCT01901224.
• Registry record last updated 2018-09-24; status last verified 2018-09.
• Results first posted 2018-09-24.
• 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.

17. Record Summary

NCT01901224 is a Phase 4, randomized, triple-masked, parallel study sponsored by Brigham and Women’s Hospital. It planned to compare metformin 1000 mg with placebo in older adults with peripheral artery disease. The primary outcome was change in PCr recovery time from baseline to 12 weeks; change in FMD was listed as a secondary outcome.

The registry lists a start of 2013-07, primary completion of 2015-06, and status as terminated. It states that the study was not funded, the randomization blind was never broken, and outcome-measure data were not collected. The registry enrollment field lists 2; a separate study-summary statement describes a recruitment aim of up to 100 subjects to find 60 subjects with PAD. The record does not explain the relationship between these figures or report a treatment-effect analysis. No conclusion about the clinical effect of metformin can be drawn from the posted results.