Post-Operative Drainage Following Lymph Node Dissection: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Post-Operative Drainage Following Lymph Node Dissection (NCT00324272), a randomized phase 4 study. The primary outcome is Post-operative Wound Drainage..

2. Clinical Question

The clinical question is about drainage after surgery, not a single pooled comparison across all participants. Groin and axillary procedures are presented separately, and the reported results do not combine those sites into one overall estimate. This matters because each comparison concerns a different surgical setting and has its own drainage summaries and p-value. The record also lists hospital stay, drain duration, complications, postoperative pain, disease recurrence, and death as secondary outcomes. Their presence as listed outcomes does not mean that numerical results are available for each one. The record does not report a detailed eligibility definition, participant characteristics, or a description of the analysis population, so those details cannot be used to broaden the interpretation of the clinical question.

3. Trial Design

The registry describes a prospective randomized, double-blind, parallel trial, classified as phase 4. Its official title identifies fibrin sealant and postoperative drainage following lymph node dissection. The registry lists Oxford University Hospitals NHS Trust as sponsor, enrollment of 74, and study status as completed. The recorded start is 2003-01 and primary completion is 2006-12.

Arms are organized by surgical site and by whether sealant was used. The reported drainage summaries include groin dissection with sealant, groin dissection without sealant, axillary dissection with sealant, and axillary dissection without sealant. The intervention is identified as fibrin sealant (Tisseel). Although the registry labels the study randomized and double blind, further details about the randomization procedure, concealment, who was masked, and whether masking remained effective are not reported in the registry record. The allocation ratio is also not reported in the registry record.

The primary outcome is postoperative wound drainage in millilitres, measured from the day of surgery until removal of the last wound drain. The stated timeframe is from surgery to drain removal, typically approximately one week. Secondary outcomes listed in the registry include length of inpatient stay, time drains remain in place, postoperative complications excluding lymphoedema, pain on the first postoperative day, disease recurrence, and death. The record gives numerical results for complications, but numerical results for the other listed secondary outcomes are not reported in the registry record.

Information about the clinical protocol beyond these recorded design features—including the sample-size justification, planned interim analyses, detailed statistical analysis plan, and analysis population—is not reported in the registry record. The enrollment figure describes the record’s enrollment; it should not automatically be interpreted as the number analyzed for every outcome. Arm-level outcome counts and summary statistics are presented separately, and these vary by surgical site and outcome.

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: All 74 patients who were enrolled in the trial were followed up until 1.6.10 (unless they died prior to this date).

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 Post-operative Wound Drainage., measured over From date of surgery to date of wound drain removal (typically a period of approximately one week)..

The primary endpoint is postoperative wound drainage volume, measured in millilitres from the day of surgery until the last wound drain was removed. It is intended to quantify the amount of fluid collected during the postoperative drainage period. The record describes drain removal as typically occurring approximately one week after surgery, while the endpoint’s actual measurement interval is tied to the date of removal for each participant.

Drainage is relevant to the stated study question because the intervention was intended to reduce postoperative drainage. The record reports a median for each of the four surgical-site and treatment groups, accompanied by a 95% confidence interval. For groin dissection, the median was 892.5 ml with sealant and 762.5 ml without sealant; the corresponding intervals were 265.0 to 2895.0 ml and 25.0 to 3255.0 ml. For axillary dissection, the median was 565.0 ml with sealant and 590.0 ml without sealant; the intervals were 30.0 to 1835.0 ml and 230.0 to 9605.0 ml.

General: a median describes the midpoint of an ordered set of observations and can be useful when values are unevenly distributed or include unusually large observations. A confidence interval expresses uncertainty around an estimated quantity; it does not show the full range of individual participants’ outcomes. The record does not explain how the confidence intervals around these medians were constructed. It also does not report individual drainage measurements or a treatment-effect estimate with a confidence interval.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies the trial as randomized, double blind, and parallel. Those labels describe the reported design, but they do not supply operational detail. For example, the random sequence method, allocation concealment, masking procedures, and any assessment of masking are not reported in the registry record. The outcome comparisons are presented by surgical site rather than as a single pooled treatment contrast.

What do the medians and confidence intervals describe?

The median is a measure of the center of a distribution: it marks the middle position when observations are ordered. It is distinct from the mean, which uses the magnitude of every value in its calculation. When outcome values are highly uneven or include extreme observations, the median can offer a useful description of a typical observation without being pulled as strongly by the largest values. The record reports medians and 95% confidence intervals for drainage volume in each arm. These intervals concern uncertainty in the reported estimates, not the range containing all participant drainage volumes. The method used to construct these intervals is not reported in the registry record.

The arm-level intervals are broad, and their endpoints differ considerably. This signals that the reported summaries are not very precise. A confidence interval should be read alongside the point estimate and the study context; it does not assign a probability to a particular treatment effect after the data have been observed. The record does not give a confidence interval for a direct between-group difference in drainage volume, so the intervals around individual arm medians should not be mistaken for that treatment-effect interval.

What does the t-test report mean?

The primary comparisons are listed as two-sided t-tests, with p = 0.704 for groin dissection and p = 0.217 for axillary dissection. A p-value describes how unusual data at least as incompatible with a specified null hypothesis would be under the assumptions of the test. It is not the probability that the null hypothesis is true, and it does not measure the probability that the result occurred by chance. A larger p-value is not proof that groups are identical; it can reflect a small effect, noisy measurements, limited information, or some combination of these.

General: a t-test is commonly used to compare means. Here, the registry’s descriptive summaries are medians. Because the record does not explain how the t-tests were applied or how they relate to the median summaries, the reported p-values should be treated as the registry’s stated tests rather than reconstructed from the medians. The record does not provide the test statistic, degrees of freedom, a treatment-effect estimate, or a confidence interval for the difference.

Why can over-dispersion matter?

General: over-dispersion means that outcome variability is greater than a statistical model assumes. If a model underestimates variability, its uncertainty estimates may be too narrow and its tests may appear more conclusive than warranted. Highly dispersed observations can also make estimates less stable. The drainage intervals in this record are broad, but the registry does not report a distributional assessment or characterize the outcome as over-dispersed. The concept is relevant as a general consideration, not as a documented diagnosis of the trial data.

How should the binary secondary comparisons be read?

For postoperative complications excluding lymphoedema, the record gives event counts by arm and sponsor-reported Fisher exact tests. It also includes simple recalculations from the posted counts: risk differences with confidence intervals, risk ratios, odds ratios, and unadjusted p-values. These recalculations are unadjusted comparisons; they do not account for other characteristics or the trial design beyond the displayed groups. Their uncertainty intervals are broad and include no difference for both recalculated risk differences. Such intervals indicate that the observed arm counts do not yield a precise estimate of the contrast.

General: a risk difference expresses the contrast between event proportions, while a risk ratio compares those proportions as a ratio; an odds ratio compares odds rather than proportions. These measures answer related but distinct questions, so their numerical values are not interchangeable. Fisher’s exact test evaluates evidence under a null hypothesis for a two-group categorical comparison. With limited counts, exact methods can be useful, but a p-value still does not express effect size or precision. The posted comparisons should be interpreted as exploratory evidence about the observed counts, not as a definitive account of all complications or all secondary outcomes.

What do the planning illustrations represent?

The sample-size illustrations are based on the observed effect and state their assumptions: a two-sided significance level of 0.05, equal allocation between the compared arms, and a normal approximation. The rows describe planning for 100%, 75%, and 50% of the observed effect. For the groin complication comparison, the illustration lists 172 participants per arm for 80% power and 230 participants per arm for 90% power when planning for 100% of the observed effect; the corresponding figures for 75% are 318 and 425, and for 50% are 741 and 991. For the axillary comparison, the figures are 1,079 and 1,443 for 100%, 1,942 and 2,599 for 75%, and 4,422 and 5,919 for 50%.

These are planning illustrations, not observed power and not evidence that the trial was designed around these assumptions. The record explicitly distinguishes them from the trial’s own design. The sample-size justification for the actual trial is not reported in the registry record. In general, planning calculations depend on assumptions about the effect worth detecting, variability or event rates, significance level, and allocation. Changing those assumptions changes the required sample size.

10. Analyses Reported by the Sponsor

The registry’s primary analyses are two-sided t-tests, with p = 0.704 for groin dissection and p = 0.217 for axillary dissection. No recalculated primary treatment-effect estimate is reported because the posted primary summaries do not include the information needed to reconstruct one. In particular, the record gives arm medians and confidence intervals but no direct estimate and interval for the between-group contrast.

For complications excluding lymphoedema, the registry reports Fisher exact p-values of 0.351 for groin and 0.480 for axillary dissection. The simple recalculations from the posted counts give unadjusted p-values of 0.351 for groin and 0.717 for axillary dissection. The groin p-values agree as displayed; the axillary recalculated value differs from the registry-reported Fisher exact result. The recalculated risk differences and their intervals are also descriptive unadjusted contrasts, not replacements for the registry analysis.

The record does not explain why the axillary p-values differ or document the precise calculation settings used for the sponsor-reported tests. Thus, the posted registry analysis remains the reported analysis, while the recalculation is best read as a separate check based on the displayed counts. Neither the p-values nor the recalculations establish equivalence, and the available secondary data do not justify treating either comparison as definitive.

11. Multiplicity, Missing Data and Analysis Populations

The record lists a primary drainage endpoint and multiple secondary outcomes, with analyses reported for drainage and postoperative complications. It does not report whether the statistical analysis plan specified a hierarchy among outcomes, adjusted for multiple comparisons, or identified a limited set of confirmatory tests. Multiplicity matters because, as the number of statistical tests increases, the chance of at least one apparently unusual result can increase even when there is no underlying effect. General: adjustment methods or a prespecified testing order can address this issue, but the record does not document their use. The reported p-values should therefore be understood as the individual analyses presented, not as evidence that the overall family of tests was controlled.

Information about missing observations is also not reported in the registry record. The drainage endpoint depends on measurement through removal of the last drain, but the record does not state whether every participant had an observed drainage value, how missing measurements were handled, or whether any participants were excluded from the analysis. The enrollment figure is 74, while the listed arm counts for the primary summaries are 18, 20, 18, and 18. These arm-level counts should not be assumed to describe the population analyzed for every other outcome. The record does not explain the differences between enrollment and the displayed arm counts or give an outcome-specific accounting of missing data.

The analysis population is not reported in the registry record. In particular, the record does not say whether analyses followed intention-to-treat, per-protocol, or another defined population, or how departures from assigned treatment were handled. Randomization can support a fair comparison when participants are analyzed appropriately, but that benefit depends in part on the analysis approach and complete outcome accounting. Without these details, the reported comparisons cannot be assessed for the effects of exclusions, crossover, or missing outcome data.

General: selective non-reporting can also complicate interpretation when planned outcomes or analyses are unavailable. The registry lists secondary outcomes for which numerical results are not included in the record. It does not explain whether results were unavailable, not estimable, or omitted for another reason. No interim analyses, stopping rules, or protocol-level missing-data procedures are reported in the registry record. These omissions do not establish that such procedures were absent; they mean that their role cannot be evaluated from the record.

12. 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

Serious events with the highest arm-level rate: death (60.0%), surgical seroma (30.0%), wound seroma (27.8%), wound cellulitis (15.0%), wound necrosis (11.1%).

The adverse-event information lists serious events for each surgical-site and treatment group. The number affected is 10 of 18 in groin dissection with sealant, 12 of 20 in groin dissection without sealant, 9 of 18 in axillary dissection with sealant, and 9 of 18 in axillary dissection without sealant. The record lists zero affected for other, non-serious events in each group. These are descriptive counts; they do not by themselves establish that the intervention caused or prevented an event.

The serious-event term list includes Death at 60.0%, Surgical seroma at 30.0%, Wound Seroma at 27.8%, Wound cellulitis at 15.0%, Wound necrosis at 11.1%, and Infected seroma at 11.1%. The registry does not provide the denominator or a clear explanation of how those percentages relate to the arm-level counts. Death counts by group are not reported in the registry record, despite Death appearing in the serious-event term list. The percentages should therefore not be converted into arm-specific death counts or treated as directly comparable group rates.

The record does not provide event timing, severity grading beyond the serious/non-serious categorization, or an attribution assessment. The safety figures are best read as a limited description of recorded events, not as a complete comparative safety analysis.

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

The result addresses whether the registry’s reported comparisons show evidence of a difference in postoperative drainage between sealant and no-sealant groups within groin and axillary dissection. The reported medians differ in direction across the surgical sites, and the registry reports p = 0.704 for groin and p = 0.217 for axillary dissection. The 95% confidence intervals around the arm medians are broad. Taken together, the record does not provide a precise estimate of a treatment effect on drainage.

These findings do not prove that fibrin sealant has no effect. A p-value that is not small does not establish equivalence, and the record reports no equivalence margin or non-inferiority analysis. Conversely, the results do not establish that sealant reduces drainage. The group medians and their intervals are not a direct estimate of the difference between groups, and the record does not provide a confidence interval for that contrast.

The primary analysis also requires care because the outcome is summarized by medians, while the registry identifies t-tests. The record does not explain the relationship between the descriptive summaries and the tests. As a result, it is not possible to assess from the posted information whether the test assumptions fit the observed outcome or to reproduce the primary analysis.

Secondary evidence adds limited context. Complication results are reported, but the record’s other listed secondary outcomes lack numerical results. The complication comparisons are imprecise and should not be treated as confirming or contradicting the primary analysis. The safety listings likewise do not provide a complete account of arm-specific death counts or the denominators underlying the event-term percentages.

The appropriate statistical conclusion is therefore bounded: the registry reports no statistically persuasive difference in the primary comparisons, but the estimates are imprecise and the record lacks important details about analysis populations, missing data, and methods. That combination leaves room for effects not resolved by the posted evidence. It neither supports a claim of benefit nor a claim of no effect.

14. Why This Trial Matters Statistically

This trial illustrates why a result cannot be reduced to a p-value. The primary outcome is a volume measured over a postoperative interval, and the registry presents medians with broad confidence intervals. The reported tests do not give a direct estimate of the between-group difference or its uncertainty. Readers therefore need to distinguish the descriptive group summaries from the inferential test and recognize that each answers a different statistical question.

The surgical-site comparisons also highlight the importance of defining the estimand—the specific treatment contrast a study seeks to estimate. Groin and axillary dissection are reported separately, not as a single combined contrast. A pooled result could obscure differences between settings, but the record does not report a plan for pooling or comparing sites. The posted data support interpretation within the stated strata, while broader claims are not justified by the available results.

The record also shows how incomplete outcome reporting affects interpretation. Several secondary outcomes are named but lack numerical results, and analysis populations and missing-data handling are not reported. These gaps limit the ability to assess whether results reflect all randomized participants and whether the observed evidence covers the full set of outcomes considered.

Finally, the sample-size illustrations demonstrate the distinction between retrospective planning calculations and a trial’s actual design. They use fractions of the observed effect and stated assumptions, but are explicitly not the trial’s own design justification and are not observed power. This distinction matters because a study should not be judged as having adequate or inadequate power by treating a post hoc calculation as a record of its original planning. Here, the actual sample-size justification is not reported in the registry record.

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.

Several limitations specific to this record constrain interpretation. The primary endpoint is reported as a median with a 95% confidence interval, but the registry identifies a t-test for each comparison. The analytical relationship between these summaries and tests is not explained, and no direct treatment-effect estimate or confidence interval is reported. That makes the magnitude and precision of the primary contrast difficult to evaluate.

The arm-level primary summaries are based on the counts listed for each surgical-site group, but the registry does not describe the analysis population or account for missing observations. Although enrollment is 74, the record does not reconcile enrollment with the arm counts displayed for the primary outcome. This does not establish that observations were missing; it means the available record does not explain the accounting.

Reporting for secondary outcomes is incomplete. The record names outcomes for inpatient stay, drain duration, pain, recurrence, and death, but numerical results for these are not reported in the registry record. The complication comparison is available, but it is based on limited counts and has wide recalculated uncertainty intervals. The reported axillary Fisher exact p-value also differs from the simple recalculated p-value, and the reason for that difference is not reported.

The safety information has additional limits: the denominator for percentages in the serious-event term list is not explained, death counts by arm are not reported, and event timing and attribution are not described. The registry also does not report detailed eligibility criteria, participant characteristics, randomization procedures, masking procedures, multiplicity handling, the actual sample-size justification, interim analyses, or a full statistical analysis plan. These are limitations of what can be evaluated from the registry record, rather than proof that the underlying study did not address them.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Post-Operative Drainage Following Lymph Node Dissection — NCT00324272.
• Registry record last updated 2011-08-11; status last verified 2006-05.
• Results first posted 2011-08-11.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

NCT00324272 is a completed, phase 4, randomized, double-blind, parallel trial sponsored by Oxford University Hospitals NHS Trust. The record gives enrollment of 74, a start of 2003-01, and primary completion of 2006-12. It concerns fibrin sealant (Tisseel) following groin or axillary lymph node dissection.

The primary endpoint is postoperative wound drainage in millilitres from surgery until removal of the last drain. The registry reports medians and 95% confidence intervals by surgical site and treatment, with two-sided t-test p-values of 0.704 for groin dissection and 0.217 for axillary dissection. The summaries are imprecise, and the record does not provide a direct treatment-effect estimate with a confidence interval.

Numerical results are also reported for postoperative complications excluding lymphoedema. Other listed secondary outcomes lack numerical results in the registry record. Important details about missing data, analysis populations, multiplicity, and the statistical plan are not reported, limiting the conclusions that can be drawn from the posted evidence.