Follicular Activation in Poor Responders: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Follicular Activation in Poor Responders (NCT02354963), a randomized na study. The primary outcome is Number of Metaphase II (MII).

2. Clinical Question

3. Trial Design

The registry describes a randomized, single-masked, parallel-group trial titled “Follicular Activation in Poor Responders.” It is listed as phase NA and as completed. The record gives a start date of 2014-11 and a primary completion date of 2019-02. The sponsor is César Díaz García.

The control arm had no intervention and proceeded to antral follicle count assessment and IVF treatment. The experimental arm underwent laparoscopy and in vitro fragmentation of ovarian tissue, followed by antral follicle count assessment and IVF treatment. Thus, the planned contrast was between the fragmentation strategy and no intervention within the trial's IVF context.

The registry lists enrollment as 34. For the primary endpoint, it reports data for 16 control participants and 15 experimental participants. The endpoint description says that 2 patients from an initial control group of 18 dropped out, one because she did not want follow-up and the other because she divorced. It also says that one patient from an experimental group of 16 left before treatment because she moved to another city. These descriptions establish that attrition occurred, but do not fully explain how every enrolled participant contributed to each endpoint analysis.

The record lists 10 outcome measures and 10 analyses, while 9 secondary outcomes are posted. It does not report the randomization procedure, allocation concealment, masking details, or a sample-size justification. The registry record also does not report a prespecified statistical analysis plan or whether any interim analyses were planned. Those details should not be inferred from the randomized label or from the reported 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

Analysis population for the primary outcome, as defined in the registry: Of the 18 patients in the initial control group, 2 dropped out of the study (one because she did not want to follow up and the other because she divorced). Of the 16 patients in the experimental group, one patient left the study before treatment because she moved to another city.

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: Two participants in the control group and one in the surgery group were lost after assignment.

6. Primary Endpoint

The registry lists the primary outcome as Number of Metaphase II (MII), measured over 36 hours after triggering.

The primary endpoint is the number of mature metaphase II (MII) oocytes retrieved in IVF cycles after triggering. The stated measurement time is 36 hours after triggering. An MII oocyte count records the number of retrieved oocytes at the mature stage described by the registry; it is a count outcome rather than a pregnancy or live-birth outcome.

The endpoint is relevant to the study's stated aim because the intervention was intended to increase the pool of antral follicles that could potentially be stimulated. Counting mature oocytes after triggering captures a later step in the IVF process than counting antral follicles. It can therefore describe ovarian response during the IVF cycle, while remaining distinct from outcomes such as fertilization, embryo transfer, pregnancy, and live birth.

The registry summarizes the primary outcome using the median and inter-quartile range. The control arm has a median of 4, with an inter-quartile range from 1 to 8; the experimental arm has a median of 2, with an inter-quartile range from 1 to 3. The posted comparison uses the Wilcoxon (Mann-Whitney) method and reports p = 0.302. No confidence interval for the primary comparison is reported in the registry record.

As a general statistical point, a count outcome can be skewed, with many observations concentrated at low values and some larger counts. A median and inter-quartile range describe the center and spread without requiring a symmetric distribution. They do not, by themselves, show the full range of individual responses or quantify a precise between-arm effect.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary comparison uses the Wilcoxon (Mann-Whitney) test, and the endpoint is summarized with medians and inter-quartile ranges. The registry reports a p-value of 0.302. The record does not report a confidence interval or an effect estimate for the primary comparison.

What does the rank-based test compare?

In general, the Wilcoxon (Mann-Whitney) method compares the ordering of observations between two groups. It uses ranks rather than requiring the outcome to follow a normal distribution. This makes it a common choice for outcomes that may be skewed, such as counts of retrieved oocytes. Its result is not automatically a test of the difference between medians alone: interpretation as a location shift requires additional assumptions about the distributions. If the shapes or spreads differ, the test can reflect more than a simple median contrast.

The median is the center of the observed values, while the inter-quartile range describes the middle half of the distribution. These summaries are useful when extreme values could make a mean less representative. They still leave information out: readers cannot determine from the summaries how many participants had each count, the full distribution, or how individual outcomes were related to other measures.

How should p-values and confidence intervals be read?

In general, a p-value measures how unusual data at least as inconsistent with a specified null model would be, if that model and the test assumptions held. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of an effect. The primary p-value of 0.302 therefore should not be interpreted as proof of no difference. Nor does the observed difference between the sample medians establish a causal effect.

In general, a confidence interval expresses uncertainty around an estimated effect under the method used to construct it. A wide interval indicates that a broad range of effects remains compatible with the data. For live birth, the unadjusted recalculation gives a risk difference of -0.121 and a 95% confidence interval from -0.35 to 0.108. Since zero lies within that interval, it does not distinguish clearly between a lower, similar, or higher event probability in the experimental arm. The interval is not a probability statement about the true effect.

What do the live-birth measures represent?

Live birth is a binary outcome at the participant level in the posted results: the registry reports counts of participants with the outcome. A risk difference expresses the absolute contrast in the observed event probabilities, while a risk ratio compares those probabilities proportionally. These are different scales and can convey different aspects of the same data. The recalculated risk difference is unadjusted; the sponsor's analysis reports a risk ratio. Small event counts make both types of estimates vulnerable to substantial uncertainty.

The registry's sponsor analysis for live birth uses a Chi-squared method and reports a risk ratio of 0.36, a 95% confidence interval from 0.04 to 3.05, and p = 0.512. The posted recalculation reports an unadjusted risk difference and p = 0.316. These are not identical estimands or necessarily identical testing procedures, so their point estimates and p-values should not be treated as interchangeable. Neither analysis yields a precise estimate in this record.

Why does over-dispersion matter for count outcomes?

In general, count data are over-dispersed when their variability exceeds what a simple count model would expect. Over-dispersion can make standard errors too small if a model assumes less variability than the data contain. The registry does not report a count regression model for the primary endpoint, so there is no basis to say that a particular over-dispersion adjustment was or was not used. The rank-based comparison avoids specifying a particular count distribution, but does not remove uncertainty from limited observations or solve every issue of interpretation.

The registry also lists planning illustrations for the live-birth effect. These are not observed power calculations and are not the trial's own sample-size justification. The stated assumptions are a two-sided significance level of 0.05, equal allocation, and a normal approximation; the planning effect is a stated fraction of the observed effect. For planning at 100% of the observed effect, the illustration gives 119 participants per arm for 80% power and 158 participants per arm for 90% power. For planning at 75% of the observed effect, it gives 232 participants per arm for 80% power and 311 participants per arm for 90% power. For planning at 50% of the observed effect, it gives 569 participants per arm for 80% power and 762 participants per arm for 90% power. Such illustrations depend strongly on the observed effect, which is uncertain in a small trial, and should not be read as evidence about the power of the completed study.

Finally, the registry does not report enough information to assess all modeling or analysis choices, including how missing outcomes were handled for each endpoint. These limits matter when interpreting both the reported test and the recalculated summaries.

10. Analyses Reported by the Sponsor

For the primary MII oocyte endpoint, the registry reports a Wilcoxon (Mann-Whitney) comparison with p = 0.302. It does not report a sponsor effect estimate or confidence interval for that comparison. The arm-level medians and inter-quartile ranges describe the observed counts, but they do not supply a confidence interval for the between-arm contrast. No separate recalculation of the primary comparison is reported in the posted results.

For live birth, the sponsor reports a Chi-squared analysis of the risk ratio: 0.36, with a 95% confidence interval from 0.04 to 3.05 and p = 0.512. The unadjusted recalculation uses the risk-difference scale and gives -0.121, with a 95% confidence interval from -0.35 to 0.108 and p = 0.316. The risk ratio and risk difference answer related but distinct questions, and the calculations need not yield the same p-value.

The registry's discrepancy indicator is false. The posted recalculation should therefore be understood as a supplementary unadjusted calculation, not as evidence that the sponsor's analysis is incorrect. The sponsor's own analysis remains the reported trial analysis, while the recalculation makes the absolute contrast explicit. Both estimates are imprecise, and neither should be taken as a definitive treatment effect.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 10 outcome measures and 10 analyses, with 9 secondary outcomes posted. Multiple outcomes can be useful for describing different stages of a clinical process, but examining many outcomes also increases the chance of finding an apparently unusual result even when no underlying effect exists. This is the general multiplicity problem. A prespecified outcome hierarchy or adjustment procedure can help manage it, but the registry record does not report whether either was used.

The primary endpoint is identified as the number of MII oocytes. That designation helps distinguish the main outcome from secondary measures, but it does not by itself establish how the study handled multiple comparisons or whether the statistical plan was fixed before outcomes were reviewed. The analysis plan and any multiplicity strategy are not reported in the registry record. Secondary p-values should consequently be interpreted cautiously, and an isolated result should not be treated as confirmatory without further design information.

Attrition is documented for the primary endpoint description: 2 patients from the initial control group of 18 left the study, and one patient from the experimental group of 16 left before treatment. The reasons are also described. The primary outcome summaries list 16 control participants and 15 experimental participants. The record does not fully describe the analysis population for each outcome or how missing measurements were handled. It does not report whether an intention-to-treat analysis was used, whether missing outcomes were imputed, or whether sensitivity analyses assessed the effect of missingness.

In general, missing data can affect an estimate when participants with unavailable outcomes differ from those with observed outcomes. The direction and size of any such effect depend on the missingness pattern and the unobserved outcomes, which cannot be recovered from the posted summaries. The reasons recorded for departures do not establish that missingness is unrelated to outcomes.

The registry does not report a sample-size justification, the assumptions used to plan the trial, or interim analyses. Planning illustrations based on the observed live-birth effect are not a substitute for the trial's own design rationale. Without the analysis plan, missing-data approach, and multiplicity strategy, the results can be described as posted but cannot be fully evaluated against all prespecified decisions.

12. 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 non-serious adverse event terms

The registry reports serious adverse events for 16 control participants at risk and 15 experimental participants at risk, with zero serious adverse events in each arm. It reports deaths as zero in both arms. For other adverse events, the record lists zero affected control participants and one affected experimental participant, using the same at-risk counts stated for the serious-event summaries.

These figures are descriptive counts in the registry record. They do not establish that the intervention caused or prevented an event. The record does not provide a serious adverse event term because none is listed, and it does not give further detail about the single other adverse event in the experimental arm. It also does not report a comparison test or confidence interval for safety outcomes.

Zero reported serious events in these groups should not be interpreted as proof that serious events cannot occur. Safety estimates based on limited numbers of participants may be imprecise, and the record's counts alone cannot characterize the full range of possible risks. The appropriate reading is limited to the events and denominators posted, without assuming that all potential harms were ruled out.

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

The primary result describes the observed MII oocyte counts in the randomized arms. The control median was 4 and the experimental median was 2; the reported Wilcoxon (Mann-Whitney) p-value was 0.302. These results do not establish a statistically clear difference under the reported test, but they also do not demonstrate equivalence. A failure to show a clear difference can reflect uncertainty as well as a genuinely small effect.

The live-birth estimates are also uncertain. The recalculated risk-difference interval ranges from -0.35 to 0.108, and the sponsor's risk-ratio interval ranges from 0.04 to 3.05. Both intervals include effects that point in different directions. The event counts are small, so the direction and magnitude of the underlying contrast cannot be determined precisely from this record.

These endpoints address different questions. MII oocyte count describes an intermediate IVF outcome, while live birth is a later outcome. A result for one does not automatically predict the other. The trial's stated hypothesis concerned increasing the pool of potentially stimulable antral follicles; the primary endpoint measures mature oocytes retrieved after triggering, not the hypothesized follicle pool itself. Secondary measures include antral follicle counts and other IVF outcomes, but the available registry record does not provide a complete basis for combining them into a single conclusion.

The findings apply to the participants and procedures represented in this completed trial. The registry does not provide enough information about participant characteristics, analysis populations, missing-data handling, or the statistical plan to judge how well the results generalize to other settings. The record also does not provide a precise primary effect estimate with a confidence interval.

Accordingly, the posted data are best read as limited evidence with substantial uncertainty. They neither demonstrate that the intervention improves clinical outcomes nor establish that it has no effect. They describe what was observed and the level of precision available in the registry record, without supporting a broader clinical recommendation.

14. Why This Trial Matters Statistically

This trial raises a useful statistical question: how should a study be interpreted when its primary outcome is a count, its posted summary is based on medians and inter-quartile ranges, and the key comparison is reported through a rank-based test? The answer requires keeping the descriptive summaries, test result, and uncertainty distinct. A difference between sample medians does not by itself quantify the treatment effect, and a p-value does not indicate whether any difference would be important.

The trial also illustrates why an intermediate outcome and a later clinical outcome should not be conflated. MII oocytes retrieved, follicle counts, pregnancy, and live birth represent different stages or measures. An intervention can be evaluated across these outcomes, but they do not become interchangeable simply because they are part of the same study.

Its small group sizes and documented attrition make precision especially important. The live-birth estimates have broad confidence intervals, and the primary comparison has no reported confidence interval in the registry. That contrast highlights the value of reporting effect estimates with uncertainty, alongside p-values and descriptive summaries.

Finally, this record illustrates how incomplete reporting limits statistical appraisal. The registry identifies randomization and the primary outcome, but does not report a sample-size justification, a full statistical analysis plan, a multiplicity approach, or complete details about missing-data handling. Those omissions do not invalidate the posted results; they constrain what can be learned from them. Reading the evidence cautiously is therefore not merely a matter of the p-values, but also of the precision and transparency of the available analysis.

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.

The primary result is reported as arm-level medians and inter-quartile ranges, together with a Wilcoxon (Mann-Whitney) p-value. The registry does not report a confidence interval or effect estimate for the primary comparison, which limits direct assessment of its precision. The test result also does not, by itself, quantify the size of a difference in MII oocyte counts.

The primary endpoint description documents attrition, and the posted endpoint denominators are 16 in the control arm and 15 in the experimental arm. The registry does not explain the analysis population for every outcome or report how missing values were handled. This leaves uncertainty about how departures and unavailable outcomes may have influenced the reported estimates.

The live-birth outcome has few events in the posted results, and both the recalculated risk difference and sponsor-reported risk ratio have broad confidence intervals. The planning illustrations based on the observed effect do not address that uncertainty in a definitive way: they are planning calculations, not the trial's own design justification or observed power. The registry does not report a sample-size rationale.

Additional design and analysis details are not reported in the registry record, including how randomization was generated, allocation concealment, what single masking entailed, the prespecified multiplicity strategy, and whether interim analyses were planned. The study is described as randomized and single-masked, but those labels do not supply these missing details.

The registry lists 9 posted secondary outcomes, and the available outcomes cover different points in the IVF process. The record does not report a complete basis for interpreting all measures together or establishing a hierarchy among secondary findings. These limitations, combined with uncertain precision, mean the results should be treated as descriptive and not as a definitive estimate of the intervention's effects.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Follicular Activation in Poor Responders — NCT02354963.
• Registry record last updated 2024-01-08; status last verified 2023-03.
• Results first posted 2024-01-08.
• 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 completed randomized, single-masked, parallel-group trial evaluated in vitro fragmentation of ovarian tissue alongside IVF treatment in patients described as poor responders. The primary endpoint was the number of MII oocytes retrieved 36 hours after triggering. The posted medians were 4 in the control arm and 2 in the experimental arm, with a Wilcoxon (Mann-Whitney) p-value of 0.302.

For live birth, the registry reports 3 participants in the control arm and 1 in the experimental arm. The unadjusted recalculation is imprecise, as is the sponsor's reported risk-ratio analysis. The record documents attrition and does not report a primary confidence interval, sample-size justification, full statistical plan, or complete missing-data approach. Overall, the registry describes the observed outcomes but does not establish a precise or definitive treatment effect.