Retrospective Study of Patients Who Were Treated With Fondaparinux Pre-, Peri- and/or Postpartum for Prophylaxis or Treatment of Venous Thromboembolism: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Retrospective Study of Patients Who Were Treated With Fondaparinux Pre-, Peri- and/or Postpartum for Prophylaxis or Treatment of Venous Thromboembolism (NCT01004939), a registered phase not stated study. The primary outcome is Number of Participants Receiving Fondaparinux in the Indicated Therapy Intervals.

2. Clinical Question

The study’s stated purpose was descriptive: to gather information about fondaparinux use in this setting and address an information gap concerning its off-label use during pregnancy. It included women treated for either prevention in the context of elevated risk or treatment of acute thromboembolism. Those are different clinical reasons for treatment, and the registry description does not say that their outcomes were analyzed as equivalent groups.

The primary measure is a count of participants receiving fondaparinux in the specified intervals, rather than a comparative estimate of treatment effect. The registry record does not report how many participants were in each interval, how many received more than one interval of treatment, or how the counts were distributed by reason for treatment. Consequently, the stated question can be identified, but the extent or pattern of use cannot be quantified from the primary results recorded.

3. Trial Design

This is a retrospective study, with status recorded as completed. The registry lists GlaxoSmithKline as sponsor, enrollment of 120, a start date of 2010-03, and a primary completion date of 2010-07. The cases covered by the primary outcome description occurred between 2004 and 2010. These dates describe the study record and case period; they do not supply information about how records were selected or when individual participants received treatment.

The intervention is fondaparinux. The population is described as pregnant women who received prophylaxis because of elevated thromboembolic risk or therapeutic treatment of acute thromboembolism. The record does not describe a randomized assignment. Allocation is not stated, and the registry warns that the arms are not a like-for-like control comparison. Therefore, the record should be read as descriptive rather than as a controlled estimate comparing fondaparinux with another treatment.

The primary endpoint concerns the number of participants receiving fondaparinux within prenatal, perinatal, and postnatal therapy intervals. Prenatal is defined as the interval until 3 days before birth; perinatal runs from 2 days before birth to one day after birth; and postnatal begins 2 days after birth. The overall timeframe is recorded as 4 months, with all cases occurring between 2004 and 2010. The registry does not provide the primary counts for these intervals.

There are 37 outcome measures listed, including secondary measures concerning hospitalization, complications, thromboembolisms, bleeding, skin changes, and the timing of skin changes under UFH or LMWH therapy. The registry record lists no analyses. It does not report a statistical analysis plan, a sample-size justification for the study, or details of the data collection and record-review procedures. Those omissions limit what can be concluded about the design beyond its retrospective, descriptive purpose.

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: Pregnant women who received prophylaxis because of an elevated thromboembolic risk or therapeutic treatment of acute thromboembolism between the years of 2004 and 2010

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 Number of Participants Receiving Fondaparinux in the Indicated Therapy Intervals, measured over 4 months (all cases occurred between 2004 and 2010).

The primary endpoint is the number of participants receiving fondaparinux during the indicated therapy intervals. It is a count-based measure intended to describe when treatment occurred in relation to birth. The intervals are defined in the registry as prenatal, perinatal, and postnatal, with boundaries specified relative to the date of birth.

A count can address a utilization question directly: it can describe how many participants were recorded as receiving treatment in a given category. It does not, by itself, estimate a risk of an outcome, compare treatment strategies, or show whether fondaparinux caused a subsequent event. This distinction matters because the study’s stated objective was to gather information about use, not to report a randomized treatment comparison.

The endpoint description identifies pregnant women treated for prophylaxis because of elevated thromboembolic risk or for acute thromboembolism. These indications represent different reasons for treatment; the record does not report separate interval counts for each indication. It also does not explain whether a participant treated in more than one interval was counted in each applicable interval or how overlapping interval membership was handled.

Most importantly, the registry record does not report the numerical primary endpoint results. It therefore does not establish how many participants received fondaparinux in any of the intervals or how treatment was distributed across them. The endpoint definition shows what was intended to be counted, but cannot substitute for the missing counts.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Statistical interpretation depends first on knowing what was counted, who was eligible for each count, and what comparison—if any—was intended. In this study, the primary endpoint is a count of participants receiving fondaparinux in therapy intervals defined relative to birth. The record supplies the interval definitions but not the interval-specific results. Without those counts, there is no posted primary estimate to interpret or re-express.

What does a participant count tell the reader?

As a general statistical concept, a count reports the number of observations meeting a stated definition. A count is meaningful alongside its denominator and the rules used to decide who belongs in the group. Counts can describe use or burden, but a count alone does not measure a rate unless an appropriate denominator and time basis are also specified. It does not establish causation or compare outcomes between treatments.

Here, the primary measure is explicitly a number of participants. The registry names prenatal, perinatal, and postnatal intervals, but does not give their participant counts or clarify how repeated or overlapping treatment across intervals was handled. It also does not report results stratified by prophylaxis versus treatment of acute thromboembolism. Those missing details matter because a count can change meaning depending on whether categories are mutually exclusive, whether a participant can contribute to several categories, and which participants were eligible for each one.

Why is the retrospective design important?

In general, a retrospective study uses information about past events rather than assigning treatment prospectively for the purpose of a randomized comparison. Such a design may be useful for describing treatment use in a specified setting. It can also leave important questions about record completeness, participant selection, and consistency of outcome ascertainment. These are general design considerations, not findings that the registry confirms occurred in this study. The record does not report the record-review procedures or how missing or ambiguous entries were handled.

A treatment comparison requires more than listing treatment groups. If treatment choices are related to participants’ clinical circumstances, observed group differences can reflect those circumstances as well as treatment. Random allocation can help balance such factors in expectation, while a nonrandomized design may require explicit adjustment and careful assumptions. Allocation is not stated, and the registry says the arms are not a like-for-like control comparison. It does not report adjusted estimates or an analysis plan that would support a causal comparison.

How should event counts be read?

For a binary outcome, each participant is classified as having or not having the event under the chosen definition and follow-up period. The number with an event is then interpreted in relation to the group at risk. In general, the denominator should match the population actually assessed for that outcome; it need not be identical across endpoints if eligibility, follow-up, or data availability differ. The registry record does not explain the analysis population or denominator rules for each of the 37 listed outcome measures.

The safety entries list affected-participant counts and denominators for mother and child groups. These are descriptive records, not adjusted comparisons. A count of events or participants does not by itself account for timing, duration of exposure, clinical context, or differences in who was observed. The registry does not report time-to-event analyses, adjustment methods, or a comparative safety model.

What do confidence intervals and significance tests mean in general?

In general, a confidence interval describes the uncertainty around an estimated quantity under a specified statistical model and sampling framework. A narrower interval indicates greater precision under that framework, while a wider interval indicates less precision. A confidence interval does not remove bias from participant selection, measurement, or confounding, and it does not convert an association into a causal effect. A significance test evaluates compatibility of data with a specified null model; its interpretation depends on the model and analysis plan, not only on a threshold.

The registry record does not report confidence intervals, p-values, effect estimates, or an inferential analysis for the primary outcome. It lists no analyses. Therefore, there is no reported precision interval or test result on which to base a claim of statistical significance. It would be inappropriate to infer such results from the enrollment figure or the outcome descriptions.

What do the planning assumptions mean?

The record describes a planning illustration based on an observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The stated effect to detect is a specified fraction of the observed effect. These are assumptions for hypothetical planning calculations, not evidence about the design actually used for this retrospective study. The record does not provide numerical planning results in the study details or a sample-size justification for the original study.

In general, a sample-size calculation estimates how many observations would be needed under chosen assumptions about the effect, variability, allocation, and error rates. Such a calculation is not observed power and does not describe the probability that the completed study has demonstrated an effect. Planning based on a fraction of an observed effect is especially dependent on the reliability and relevance of that underlying estimate. Since the primary outcome results are not reported, the observed primary effect cannot be checked from this record.

Over-dispersion is a general issue for count data: it occurs when observed variability exceeds what a simple count model assumes. If present, a model that ignores it can understate uncertainty. The registry does not report a count model, a dispersion assessment, or any other statistical modeling choices. General methodological considerations should not be mistaken for methods confirmed in this study.

Finally, the registry’s list of outcomes raises a general multiplicity consideration: when many outcomes are examined, some apparent findings may arise by chance. Whether and how to address that concern depends on which outcomes were prespecified and which analyses were performed. Those details are not reported in the registry record, so no multiplicity adjustment or hierarchy of endpoints can be assumed.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 37 outcome measures, including the primary endpoint and secondary measures concerning hospitalization, complications, thromboembolism, bleeding, and skin changes. It lists no analyses. The record does not describe a prespecified hierarchy of outcomes, a multiplicity adjustment, or which outcomes were considered confirmatory versus descriptive. With many outcomes, multiplicity is a general concern because examining numerous results can increase the chance of observing apparently notable findings by chance. The record does not provide enough information to determine how that concern was addressed.

Missing-data handling is not reported in the registry record. There is no description of how incomplete medical records, unavailable outcome information, or loss of follow-up were treated. A missing value should not be interpreted as absence of an event, and the posted safety counts do not establish whether every participant had complete information for every measure.

The record also does not report analysis populations for the primary and secondary outcomes. It does not state whether each analysis included all enrolled participants, only participants with a particular exposure history, or a subset with complete records. Enrollment of 120 does not by itself identify the denominator for each outcome. The mother and child safety entries have separately stated denominators, but the record does not give a broader account of population definitions or denominator selection across endpoints.

Because allocation is not stated and the registry cautions against treating the arms as a like-for-like control comparison, no randomized intention-to-treat population can be assumed. The analysis plan, participant flow, and any methods for adjusting for differences between groups are not reported in the registry record. These gaps limit assessment of both precision and potential bias.

11. Safety Results

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

Other (non-serious) adverse events

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

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: vaginal haemorrhage (3.3%), abdominal wall haematoma (0.8%), abortion (0.8%), abortion induced (0.8%), haemorrhage (0.8%).

The registry’s adverse-event entries list serious and other adverse events for mother and child groups. For mothers, serious adverse events are recorded for 9 of 120, and other adverse events for 6 of 120. For children, serious adverse events are recorded for 1 of 124, and other adverse events for 0 of 124. These are descriptive participant counts with their listed denominators; they are not treatment-versus-control estimates.

The listed serious adverse-event terms include Vaginal haemorrhage at 3.3%, Abdominal wall haematoma at 0.8%, Abortion at 0.8%, Abortion induced at 0.8%, Haemorrhage at 0.8%, and Heparin-induced thrombocytopenia at 0.8%. The registry record does not report the term-specific denominators or explain whether a participant could contribute to more than one term. These percentages should therefore be read as the registry lists them, without inferring a common denominator or adding them together.

Deaths are not reported in the registry record. The adverse-event counts do not establish that fondaparinux caused or prevented any event. Without a suitable comparison and stated methods for assessing event occurrence, these figures describe recorded events only; they do not establish comparative safety or risk.

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

The record’s clearest contribution is to state a descriptive question about fondaparinux use in pregnancy and around birth. It defines prenatal, perinatal, and postnatal intervals and identifies a population that received prophylaxis for elevated thromboembolic risk or treatment for acute thromboembolism. It also lists related outcome topics and adverse-event counts. These elements show what the registry intended to describe.

The central limitation for interpretation is that the primary interval-specific counts are not reported in the registry record. Without them, readers cannot determine how many participants received fondaparinux in each interval or characterize the pattern of use from the posted primary results. The enrollment figure cannot fill that gap, because it does not specify the count in any interval or the denominator used for each endpoint.

The record also does not support an estimate of whether fondaparinux changes the likelihood of thromboembolism, bleeding, hospitalization, or another outcome. It is retrospective, allocation is not stated, and the registry says the arms are not a like-for-like control comparison. The outcome descriptions and safety counts are not substitutes for a controlled treatment contrast. No causal conclusion follows from the fact that an event was recorded during treatment.

In general, observational descriptions can help identify how a treatment was used in a particular setting and what outcomes were recorded. Their interpretation depends on the completeness and selection of records, clear outcome definitions, and appropriate handling of differences among participants. Those methodological details are not reported, so the extent to which the entries represent all eligible cases or comparable groups cannot be assessed.

The safety figures should be read as descriptive entries with the stated denominators. They do not provide a comparative risk estimate, and the listed serious-event terms cannot be summed to create a total because the record does not explain whether terms overlap. The absence of a reported death count is not evidence of zero deaths; deaths are not reported in the registry record.

Accordingly, the posted record establishes the study’s setting, purpose, endpoint definition, and selected adverse-event counts, but not the primary result or an estimate of treatment effect. It supports cautious description of what was recorded and does not justify claims about efficacy, harm, or comparative benefit.

13. Why This Trial Matters Statistically

This study matters statistically because it illustrates the difference between defining an endpoint and reporting an estimable result. The registry specifies a primary count of participants receiving fondaparinux during intervals around birth, with explicit timing definitions. Yet the numerical counts for those intervals are not reported in the registry record. Readers can understand the intended measure without being able to evaluate its observed distribution.

That distinction is particularly important for descriptive studies. Counts can be useful for documenting treatment patterns, especially when the stated purpose is to address an information gap about use. But the counts need clearly defined populations and denominators, and interpretation depends on complete reporting of the measured results. A study’s enrollment figure alone cannot answer a question about how use was distributed among intervals or indications.

The record also highlights why design labels matter. Retrospective data can describe past treatment and outcomes, while causal interpretation generally requires stronger design features or explicit analytical assumptions. Here, allocation is not stated, and the registry cautions that the arms are not a like-for-like control comparison. This prevents readers from treating the record as a randomized comparison or interpreting listed outcomes as effects attributable to fondaparinux.

Finally, the record lists many outcome measures but no analyses and no statistical analysis plan. This makes it difficult to assess how the outcomes were prioritized, how missing information was handled, or whether multiplicity was considered. The example reinforces a general principle: endpoint definitions, denominators, analysis populations, and complete numerical reporting are all necessary to understand what a clinical study can and cannot establish.

14. 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 needed to answer the stated question—the number of participants receiving fondaparinux in the prenatal, perinatal, and postnatal intervals—is not reported in the registry record. The record therefore does not permit assessment of the treatment-use pattern, including whether counts differed by interval or by indication. It also does not state how participants treated across more than one interval were counted.

The study is retrospective, and the registry does not describe how records were identified, how eligibility was determined, or how treatment and outcomes were abstracted. It does not report whether outcome definitions were applied consistently or whether data were complete. These are specific uncertainties for a retrospective record-based study; the record does not establish whether any particular problem occurred.

No randomized allocation is described. Allocation is not stated, and the registry warns that the arms are not a like-for-like control comparison. There is no reported statistical analysis plan, adjusted treatment estimate, or analysis population definition. As a result, the study cannot be interpreted from this record as a causal or comparative assessment.

The registry lists 37 outcome measures but no analyses. It does not report how missing data or multiple outcomes were handled, nor does it explain denominator choices across outcomes. The adverse-event term percentages have no term-specific denominators described, and the record does not clarify possible overlap among terms. Deaths are not reported in the registry record.

The stated timeframe for the primary measure is 4 months, while cases occurred between 2004 and 2010. The record does not explain how that timeframe relates to each participant’s observation period. Consequently, interpretation of follow-up and exposure duration is limited. These omissions do not show that the study’s observations are invalid; they constrain what can be concluded from the registry entry alone.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Retrospective Study of Patients Who Were Treated With Fondaparinux Pre-, Peri- and/or Postpartum for Prophylaxis or Treatment of Venous Thromboembolism — NCT01004939.
• Registry record last updated 2011-08-26; status last verified 2011-07.
• Results first posted 2011-08-26.
• 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.

18. Record Summary

NCT01004939 is a completed retrospective study sponsored by GlaxoSmithKline, with enrollment of 120. The registry records a start date of 2010-03 and primary completion date of 2010-07. Its objective was to describe fondaparinux use for prophylaxis or treatment of venous thromboembolism before, around, and after birth, addressing an information gap about off-label use during pregnancy.

The primary endpoint is the number of participants receiving fondaparinux in prenatal, perinatal, and postnatal intervals. The registry defines those intervals but does not report the interval-specific counts. It lists 37 outcome measures and no analyses. The registry also records adverse-event counts for mother and child groups and lists serious adverse-event terms.

Allocation is not stated, and the registry cautions that the arms are not a like-for-like control comparison. Statistical methods, analysis populations, missing-data handling, and multiplicity procedures are not reported in the registry record. The available information supports a description of the study’s purpose and recorded safety entries, but not a primary outcome estimate or a comparative treatment conclusion.