Does Pioglitazone Increase the Production of 15-EPI-Lipoxin A4?: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Does Pioglitazone Increase the Production of 15-EPI-Lipoxin A4? (NCT01040819), a randomized phase 4 study. The primary outcome is Plasma 15-epi-lipoxin A4.

2. Clinical Question

The trial title asks whether pioglitazone increases production of 15-epi-lipoxin A4, while the official title also names prostacyclin (PGI2). The posted primary outcome focuses on plasma 15-epi-lipoxin A4 at two months. The study description explains that the investigators were interested in possible anti-inflammatory effects and potential biological mechanisms. Those motivations are context, not outcomes established by the primary comparison.

The specific clinical question addressed by the posted results is narrower than whether pioglitazone increases the measure relative to no treatment: both groups received pioglitazone, but followed different dose schedules. The comparison therefore describes the measured levels in these two groups under the reported regimens. It cannot, by itself, distinguish changes attributable to either regimen from changes that might have occurred without pioglitazone. A separate placebo or untreated group is not reported in the registry record.

The registry describes a randomized, single-masked, parallel-group phase 4 trial with enrollment of 25. It does not report detailed eligibility criteria, participant characteristics, or the identity of the masked party. These omissions limit what can be said about who the results represent or how masking was implemented.

3. Trial Design

The registry identifies this as a randomized, single-masked, parallel-group phase 4 trial sponsored by Baylor College of Medicine. It records the study as completed, with enrollment of 25. The record lists a start of 2010-02 and primary completion of 2011-10. It does not describe the method used to generate or conceal the random allocation sequence, or identify who was masked.

The two groups differed in pioglitazone dosing. Participants assigned to the 15 mg group were scheduled to receive 15 mg/d for two months. Participants in the 30 mg/d group were scheduled to receive 15 mg/d for one month, followed by 30 mg/d for an additional month. Both regimens included planned sampling at baseline, one month, and two months. The descriptions name serum and urine samples for 6-keto-PGF1a and 15-epi-lipoxin A4. The posted primary endpoint, however, is plasma 15-epi-lipoxin A4 at two months.

The registry lists 13 participants in the 15 mg group and 12 in the 30 mg/d group for the posted primary outcome. The arm means and standard errors are reported at the two-month endpoint. The record does not provide participant flow, explain whether all enrolled participants contributed outcome data, or state how many participants were included in the analysis beyond the arm counts attached to the outcome.

This is an active-regimen comparison, not a comparison with placebo or no treatment. Randomization can help balance factors between groups, but the registry record does not report details needed to assess the allocation process or baseline balance. It also does not report a protocol-level sample-size justification, original analysis plan, interim analyses, adherence, or treatment discontinuations. Those details should not be inferred from the posted group results.

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.

6. Primary Endpoint

The registry lists the primary outcome as Plasma 15-epi-lipoxin A4, measured over 2 months.

The primary endpoint is plasma 15-epi-lipoxin A4 at two months, measured in ng/ml and summarized as a mean with a standard error. The registry describes the outcome as plasma 15-epi-LXA4 levels. The two-month time point is the endpoint specified for the primary outcome; the arm descriptions also schedule samples at baseline and one month, but those time points are not posted as primary results.

15-epi-lipoxin A4 is identified in the study description as a lipid mediator with anti-inflammatory properties. The description gives a biological rationale for measuring it: the investigators were interested in possible mechanisms associated with pioglitazone. This explains why the marker was selected, but the reported endpoint measures a blood concentration, not cardiovascular events, atherosclerosis progression, or a clinical benefit.

A continuous measure retains information about the measured level rather than reducing the endpoint to a yes-or-no event. A group mean gives a central summary, while the standard error describes the precision of that estimated mean. It does not describe the full spread of individual participant values. The registry does not report individual-level values, a standard deviation, or a distribution plot in the posted outcome record.

The two-month endpoint compares the reported group means following the specified dosing schedules. Because both groups received pioglitazone, this endpoint does not directly estimate the difference between pioglitazone and no pioglitazone. The registry does not report a primary outcome for prostacyclin or for clinical outcomes.

7. Results Posted in the Registry

• Pioglitazone 30 mg/d: mean 1.08 versus 1.05 on control; difference 0.03 (95% CI -0.25 to 0.31), standardized effect d = 0.08; z-test p = 0.835.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is a continuous measurement, and the registry summarizes it with a mean and standard error in each group. A mean is the arithmetic average of the observed measurements. It gives a useful description of the group’s central level, but it does not show whether individual observations cluster near that value or vary widely. A standard deviation would describe participant-to-participant spread; the posted dispersion measure is instead the standard error, which describes uncertainty in the estimated mean.

What does the mean difference represent?

The reported mean difference compares the two group means at the two-month endpoint. Its direction is defined by the comparison of the 30 mg/d group with the 15 mg group. A positive estimate means the reported mean was higher in the 30 mg/d group; the estimate is 0.03 ng/ml. This is an unadjusted comparison based on the posted group summaries. The registry does not describe covariate adjustment or a detailed statistical model.

Random allocation is intended to make treatment groups comparable on average, including for factors that are not measured. In a small trial, however, randomization does not guarantee that the groups start with similar characteristics. The registry does not report baseline characteristics or the allocation procedure, so it is not possible to assess those matters from the posted record. The registry also does not state whether the primary comparison used a model beyond the reported summary comparison.

How should the confidence interval be read?

A confidence interval expresses the uncertainty around an estimated effect under the statistical method used to calculate it. The posted 95% confidence interval for the mean difference extends from -0.253 to 0.313 ng/ml. Values on both sides of zero are included. Zero corresponds to no difference between the group means. The interval therefore does not identify a clear direction of difference, and its width signals limited precision. A confidence interval is not a range containing a specified proportion of individual participant measurements.

The registry does not state the confidence-interval method or assumptions used for the posted comparison. The interval should therefore be read as the reported uncertainty estimate, without assuming a particular model beyond what the record documents. The group standard errors and sample sizes help describe the reported summaries, but they do not replace information about the analysis method or the underlying distribution of measurements.

What does the p-value tell the reader?

The unadjusted p-value is 0.835. In general, a p-value measures how unusual results at least as incompatible with a specified null model would be if that model and the test assumptions were correct. It is not the probability that the null hypothesis is true, nor the probability that the finding is due to chance. It also does not measure the size or clinical importance of an effect. Here, the p-value accompanies a small estimated mean difference and an interval that includes zero; it does not prove the regimens have identical effects.

Because the record labels the p-value unadjusted, it should not be interpreted as accounting for multiple comparisons. The registry reports zero posted analyses and zero posted secondary outcomes, but does not report a multiplicity strategy or explain the relationship between the reported comparison and any prespecified analysis plan. It is not possible to determine from the record whether other endpoints or analyses were considered.

Why can a small study give an uncertain estimate?

Precision depends on the amount of information in the data, including the number of observations and the variability of measurements. With only 13 and 12 participants reported in the groups, estimates can be sensitive to individual observations. A wider interval means a broader range of effects remains compatible with the data under the stated interval method. It does not mean that every value in the interval is equally plausible, nor does it show which value is true.

Standard errors are not interchangeable with standard deviations. The standard error pertains to how precisely a mean is estimated; the standard deviation describes variation among participants. The registry posts standard errors for the group means and a pooled standard deviation of 0.355 for the comparison. It does not describe the method used to obtain that pooled value. The posted comparison lists an approximation flag as false, but the registry does not provide additional details about how the comparison was calculated.

What information is missing from the analysis record?

The registry does not report the original statistical analysis plan, the method for handling missing observations, the analysis population, or whether the analysis followed intention-to-treat principles. It does not report baseline outcome values or an adjusted analysis. It also does not report interim analyses or a protocol-based sample-size calculation. These absences do not show that particular methods were or were not used; they mean those details are not reported in the registry record.

For the same reason, the planning illustrations based on the observed effect should not be mistaken for the trial’s original design. They translate the observed estimate into hypothetical group sizes under stated assumptions. They do not describe the actual enrollment rationale, and they are not observed power. The statistical result itself remains the posted comparison, its confidence interval, and its unadjusted p-value, interpreted in light of the limited group sizes and incomplete methods reporting.

10. Confidence Intervals and Precision

The estimated mean difference in two-month plasma 15-epi-lipoxin A4 was 0.03 ng/ml, comparing the 30 mg/d group with the 15 mg group. The 95% confidence interval runs from -0.253 to 0.313 ng/ml. Because the interval crosses zero, the data do not pin down whether the mean was higher or lower under the dose-increase schedule. The interval also permits differences appreciably larger in either direction than the point estimate.

Precision concerns how narrowly the data locate an estimated effect. A narrow interval would indicate a more tightly estimated comparison under the method used; a wide interval leaves more uncertainty about the magnitude and direction. The reported interval is broad relative to the estimate, so the point estimate should not be treated as a precise measure of the difference between regimens. The small group sizes—13 and 12—are important context for interpreting that uncertainty, although the registry does not report the interval calculation method.

The group means were 1.05 ng/ml and 1.08 ng/ml, with standard errors of 0.08 and 0.12, respectively. These standard errors describe uncertainty in the group means, not the range of participant values. The registry does not report standard deviations for each arm or individual measurements in the primary outcome record. The pooled standard deviation listed with the comparison is 0.355.

The confidence interval is not a statement that there is a 95% probability that the true difference lies within these particular limits. In general, a confidence interval is generated by a procedure that, under its assumptions, would cover the target quantity at the stated rate across repeated studies. The specific method and assumptions for this interval are not reported in the registry record. The interval is therefore most useful as the posted estimate of uncertainty, not as proof of equivalence or as a measure of individual variation.

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 participants per arm, not total events. This is a continuous-outcome comparison, so the displayed quantities refer to the number of participants needed in each group under the stated planning assumptions. The curve represents the same type of planning illustration across the effect sizes considered; it is not a record of additional trial results.

The 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 as a stated fraction of the observed effect. The scenarios are labeled as 100%, 75%, and 50% of the observed effect. For 100% of the observed effect, the illustration gives 2,199 participants per arm for 80% power and 2,944 participants per arm for 90% power. For 75% of the observed effect, it gives 3,909 participants per arm for 80% power and 5,233 participants per arm for 90% power. For 50% of the observed effect, it gives 8,795 participants per arm for 80% power and 11,774 participants per arm for 90% power.

The 80% power column describes the planned chance of detecting the specified effect under the assumptions if that effect were present; the 90% column uses a higher target chance of detection. These figures are planning illustrations calculated from the observed effect, not observed power. Smaller effects require more participants because they are harder to distinguish from random variation under the same assumptions.

The record does not report the original trial’s sample-size justification or design assumptions. These figures should therefore not be read as the enrollment target used by the investigators. They also do not imply that the trial had an adequate or inadequate original design; they address hypothetical future planning based on the observed comparison.

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, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = 0.030; pooled standard deviation σ = 0.355.
2. n = 2 × (1.96 + 0.842)² × 0.355² / 0.030² = 2,198.6, rounded up to 2,199 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
delta <- 0.03                      # observed difference in means
sigma <- 0.355038271054042                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^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 (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(1040819)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 2,199 (80% power), 2,944 (90% power)
 75% of observed effect: n per arm = 3,909 (80% power), 5,233 (90% power)
 50% of observed effect: n per arm = 8,795 (80% power), 11,774 (90% power)
power.t.test: 2199.6 
Simulated power at n = 2,199 per arm: 0.800 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-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

Multiplicity arises when several outcomes, time points, groups, or statistical tests are examined. Conducting multiple tests can increase the chance that at least one result appears notable by chance alone. A prespecified strategy may define a primary comparison and explain how additional tests are handled. The registry identifies plasma 15-epi-lipoxin A4 at two months as the primary outcome, but it does not report a multiplicity plan.

The record lists zero posted secondary outcomes and zero posted analyses. Those entries describe what is posted in the registry record; they do not establish whether other outcomes or analyses were planned, collected, or examined. The reported p-value is unadjusted. Without a complete analysis plan, it is not possible to assess whether the displayed test was the only planned test or whether any adjustment was intended.

Missing-data handling is also not reported. The arm counts attached to the outcome are 13 and 12, but the record does not provide a participant flow account or state whether every enrolled participant contributed a two-month measurement. It does not describe any imputation, exclusion, or other approach to missing observations. The planned baseline and one-month sampling schedule does not itself show which measurements were available for analysis.

The analysis population is not reported in the registry record. It is therefore unclear whether the posted comparison included all randomized participants, only participants who completed the assigned schedule, or another defined group. The record does not state whether the analysis was intention-to-treat, per-protocol, or otherwise specified. These distinctions can affect interpretation, especially when adherence or follow-up differs between groups.

The original statistical analysis plan, interim-analysis procedures, and prespecified sample-size rationale are also not reported. Their absence should not be taken as evidence that no plan existed; the registry record simply does not document them. Together, the missing information limits assessment of whether the analysis was prespecified, how incomplete measurements were handled, and how the reported test fits within the full set of analyses.

14. 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 0% reporting threshold.

The registry reports serious adverse events for 13 participants at risk in the pioglitazone 15 mg group and 12 at risk in the 30 mg/d group. The serious adverse event count is 0 in each group. No serious adverse event terms are listed. The record also reports 0 other, non-serious adverse events in each group, with the same numbers at risk.

These are the adverse-event figures posted in the registry record. They do not establish that no adverse events occurred outside the reporting period or in participants whose data were not captured. The record does not describe the collection or reporting procedures, follow-up duration for safety, or whether all enrolled participants were included in these counts. Death counts are not reported in the registry record.

With the reported group sizes of 13 and 12, zero recorded events should be read as a descriptive count, not proof that an event is impossible or that the regimens have equivalent safety. The record does not provide enough detail to assess uncommon events or compare safety reliably. These figures should be considered separately from the primary biomarker result.

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

The posted result describes plasma 15-epi-lipoxin A4 at two months in two groups receiving different pioglitazone schedules. The mean was 1.05 ng/ml in the 15 mg group and 1.08 ng/ml in the group that increased to 30 mg/d. The estimated mean difference was 0.03 ng/ml, with a 95% confidence interval from -0.253 to 0.313 and an unadjusted p-value of 0.835. The interval includes zero and indicates substantial uncertainty about the direction and size of the difference.

This does not establish that pioglitazone increases 15-epi-lipoxin A4 compared with no treatment. There is no placebo or untreated group in the posted comparison. It also does not establish that increasing the dose changes the biomarker: the reported estimate is imprecise and compatible with differences in either direction. A non-significant p-value does not show that the schedules are equivalent.

The endpoint is a biomarker concentration, not a cardiovascular event or a measure of atherosclerosis progression. The study description sets out a biological rationale involving inflammation and possible mechanisms, but those mechanisms are not demonstrated by the posted primary outcome. The registry does not report clinical outcomes that would establish effects on cardiovascular morbidity or mortality.

The result applies to the participants and regimens represented in the record, but detailed eligibility and baseline characteristics are not reported. The small group sizes and missing information about missing data and analysis populations further limit interpretation. The posted safety counts are also limited descriptive information and cannot establish comparative safety.

The most defensible reading is that the registry reports similar point estimates of the two-month group means, while the comparison remains too uncertain to support a precise conclusion about the difference between schedules. The result is not evidence for or against a clinical benefit, and it does not answer the broader question of pioglitazone versus no pioglitazone.

16. Why This Trial Matters Statistically

This trial illustrates the difference between a biological question and the comparison a study design can answer. The title raises whether pioglitazone increases 15-epi-lipoxin A4, but the posted comparison is between two pioglitazone dosing schedules. Without a placebo or untreated group, the primary result does not estimate the effect of pioglitazone relative to no treatment. Identifying the actual comparator is essential to interpreting any treatment contrast.

It also illustrates why an effect estimate should be read alongside its uncertainty. The mean difference is 0.03 ng/ml, but the confidence interval extends from -0.253 to 0.313. The point estimate alone can make the contrast look simple; the interval shows that the data do not locate its direction or magnitude precisely. A p-value of 0.835 does not resolve this uncertainty or establish equivalence.

The outcome is a continuous biomarker. Such measures can help describe biological responses, but a difference in a marker is not itself proof of a change in a clinical outcome. The study description offers mechanistic context, while the posted primary outcome remains a plasma level at two months. Separating rationale, measured outcome, and clinical implication prevents conclusions from extending beyond the evidence.

The registry also highlights how reporting detail affects statistical interpretation. The record gives group means, standard errors, sample sizes, a confidence interval, and an unadjusted p-value, but does not report the analysis population, missing-data approach, multiplicity plan, or original sample-size rationale. These omissions make it harder to evaluate completeness and prespecification, particularly in a small study.

Finally, planning calculations based on an observed effect can be useful for thinking about future study size, but they are not a substitute for the original design rationale and are not observed power. Keeping planning illustrations separate from trial findings helps avoid treating hypothetical sample sizes as evidence about the strength of the completed study.

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.

The registry record has important limitations for interpreting this trial. The primary endpoint is reported at two months, but baseline and one-month measurements scheduled in the arm descriptions are not posted as primary results. Baseline endpoint values are not reported, so the posted comparison cannot be interpreted as a comparison of changes from baseline.

The groups are small, with 13 participants in the 15 mg arm and 12 in the group scheduled to increase to 30 mg/d. The confidence interval is broad and includes differences in either direction. Individual observations and arm-specific standard deviations are not reported alongside the primary outcome, limiting assessment of the measurement distribution.

The record does not report detailed eligibility criteria, participant characteristics, allocation procedures, or who was masked. It also does not report participant flow, adherence, treatment discontinuations, missing-data handling, or the analysis population. Consequently, the record does not allow a full assessment of whether the reported outcome represents all randomized participants or a subset.

The comparison is between two active dosing schedules; a placebo or untreated group is not reported. This limits conclusions about whether pioglitazone itself changes the biomarker. The posted endpoint is a biomarker, and clinical cardiovascular outcomes are not reported as primary results. The record also does not describe a multiplicity plan, original statistical analysis plan, interim analyses, or trial-specific sample-size justification.

Safety reporting is limited to the posted counts, with no serious or other adverse events recorded in either group. The record does not provide details about safety follow-up or event ascertainment, and death counts are not reported in the registry record. These gaps call for cautious interpretation rather than assumptions about unreported procedures or outcomes.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Does Pioglitazone Increase the Production of 15-EPI-Lipoxin A4? — NCT01040819.
• Registry record last updated 2016-06-22; status last verified 2016-05.
• Results first posted 2016-06-22.
• 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

The completed phase 4 trial was randomized, single-masked, and parallel-group, with enrollment of 25. It compared pioglitazone 15 mg/d for two months with a schedule of 15 mg/d for one month followed by 30 mg/d for an additional month. The registry lists Baylor College of Medicine as sponsor.

The primary outcome was plasma 15-epi-lipoxin A4 at two months, measured in ng/ml. The group means were 1.05 in the 15 mg arm and 1.08 in the 30 mg/d arm. The mean difference was 0.03, with a 95% confidence interval from -0.253 to 0.313 and an unadjusted p-value of 0.835. The interval includes zero and leaves substantial uncertainty about the difference between schedules.

The record reports zero serious and zero other adverse events in both groups, but provides limited safety detail. It does not report the trial’s original sample-size rationale, analysis population, missing-data methods, or multiplicity plan. The active-regimen comparison does not establish the effect of pioglitazone relative to no treatment or a clinical effect on cardiovascular outcomes.