To Evaluate the Safety and Efficacy of the Saline-coupled Bipolar Sealer in Primary Unilateral Total Knee Arthroplasty: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for To Evaluate the Safety and Efficacy of the Saline-coupled Bipolar Sealer in Primary Unilateral Total Knee Arthroplasty (NCT03952546), a randomized na study. The primary outcome is Hemostasis.

2. Clinical Question

The study is described as prospective and randomized. Its brief description states that the purpose was to investigate whether the saline-coupled bipolar sealer provides superior hemostasis, while the official title identifies a non-inferiority comparison. These are different statistical questions: superiority asks whether results favor one approach, whereas non-inferiority asks whether the intervention is not worse than the comparator by more than an accepted margin. The registry record does not explain the relationship between those descriptions or report the margin used for the non-inferiority objective.

The registry lists functional outcomes and wound infection as secondary outcomes. The posted primary outcome is a continuous measure of estimated blood loss, and its arm-level summaries describe the observed results. Those summaries alone do not establish which statistical objective was prespecified or whether it was met. Interpretation of the sponsor-reported test depends on design details that are not reported in the registry record.

3. Trial Design

The registry describes a randomized, parallel-group trial of two devices used during primary unilateral total knee arthroplasty. The record lists enrollment of 211 and a completed status. It identifies Surgical Blood Loss as the condition and gives the intervention and comparator as saline-coupled bipolar sealer and unipolar electrocautery, respectively. The posted primary outcome results report 82 participants in each arm; those arm-level counts should not be treated as a complete account of enrollment or as evidence about why some participants are not represented in the posted endpoint summaries.

The study description calls the trial prospective, randomized, and single-blinded. The registry metadata lists double masking. Those descriptions do not match, and the registry record does not clarify which participants or personnel, if any, were masked. The practical meaning of the masking designation therefore cannot be determined from this record. The brief description says that restricting the study to two surgeons was intended to limit outcome variation due to differences in surgical technique. It does not report how that restriction was implemented or whether surgical technique was assessed as part of the analysis.

The record lists the primary completion date as 2021-06-09 and the start date as 2019-05-07. The primary endpoint is postoperative day 1 estimated blood loss, calculated using Gross' Formula. Functional outcomes and wound infection are listed as secondary outcomes. The registry states that there are three outcome measures and two analyses, but those counts do not by themselves specify the full testing strategy or which outcomes were included in each analysis.

The trial's detailed protocol, allocation-concealment procedures, sequence-generation method, intended analysis population, and handling of missing outcome values are not reported in the registry record. Nor is the planned non-inferiority margin reported. These omissions limit assessment of how closely the posted results correspond to the original trial plan. The reported randomization and parallel-group structure provide the broad design, but not all details needed to evaluate risk of bias or reproduce the analysis.

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: 47 patients were screen failures due to either the blood work not meeting the study enrollment criteria, an exclusion for spinal anesthesia, or the patient decided not to proceed with enrollment in the study.

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 Hemostasis, measured over Postoperative day 1.

The primary endpoint is estimated blood loss on postoperative day 1, expressed in milliliters and calculated by Gross' Formula. The registry identifies the reported summary as a mean with a 95% confidence interval. Estimated blood loss is a continuous outcome: it can take a range of values, rather than being limited to a yes-or-no event. A mean describes the central level of the measured quantity across the participants included in an arm's summary.

In general, a blood-loss measure can provide a quantitative way to compare hemostasis between groups. For a comparison of means to be informative, the measurement method, timing, analysis population, and treatment of unavailable measurements all matter. The record names the formula and timing but does not provide its calculation details or report how missing postoperative day 1 measurements were handled. The registry also does not describe whether the endpoint was assessed by a person unaware of treatment assignment.

The posted mean is 924 milliliters for the saline-coupled bipolar sealer arm, with a 95% confidence interval from 178 to 1861. For the unipolar electrocautery arm, the mean is 1062 milliliters, with a 95% confidence interval from 218 to 2053. These arm-specific intervals represent uncertainty in the estimates under the analysis used to construct them; they are not ranges containing every individual participant's blood-loss value. The registry does not post a confidence interval for the between-arm contrast, so the arm-level intervals should not be mistaken for an interval estimate of that contrast.

The non-inferiority objective requires interpretation of the contrast against a prespecified margin, not merely a comparison of the two arm means. That margin and the full decision rule are not reported in the registry record. The endpoint summaries therefore describe the observed group-level estimates, while the sponsor's reported test supplies a separate inferential result whose interpretation is constrained by missing design details.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a continuous measure summarized as a mean. The registry reports an arm-level mean and a 95% confidence interval for each treatment group, along with a one-sided t-test labeled as non-inferiority. These methods answer related but distinct questions: descriptive summaries characterize each arm, while a hypothesis test evaluates a specified comparison under assumptions and a decision rule.

What does a mean describe?

In general, the mean is the sum of measured values divided by the number of observations included in the calculation. It provides a measure of the average outcome, but it does not show the distribution of individual observations. A mean can be influenced by unusually high or low measurements. For a clinical measure such as estimated blood loss, understanding the distribution can be helpful alongside the mean, especially if values are skewed or include outliers. The registry does not report the arm-level standard deviations, medians, ranges, or distribution plots as part of the registry entries available for this analysis, so the shape and spread of individual values cannot be assessed from the reported means alone.

The registry gives 95% confidence intervals for the means. General explanation: a confidence interval is an interval produced by a procedure that, under its assumptions and repeated use, captures the target quantity at the stated coverage rate. It conveys uncertainty in an estimate; it is not a range expected to contain a stated proportion of individual participants' measurements. Nor is it the probability that the fixed population value lies in the particular calculated interval. The posted intervals are for the arm means, not for the difference between the two arms.

What does a non-inferiority test ask?

General explanation: a non-inferiority design asks whether the intervention's outcome is not worse than the comparator by more than a prespecified amount considered acceptable for the trial's purpose. That amount is the non-inferiority margin. The test is framed around a boundary: evidence must be sufficient to rule out a difference that crosses the unacceptable side of that margin, according to the design's chosen method and significance level. Merely observing close arm means does not establish non-inferiority, and a non-significant result in a standard superiority test does not establish it either.

The registry labels the reported analysis a one-sided t-test and a non-inferiority analysis. It gives p = 0.125. However, the margin, exact null hypothesis, direction of the contrast, and criterion for declaring non-inferiority are not reported in the registry record. Without those details, the p-value cannot be translated into a complete conclusion about whether the prespecified non-inferiority criterion was met. It is appropriate to report the test as the registry does, while recognizing that its design context is incomplete.

Why does the test direction matter?

General explanation: a one-sided test evaluates a directional hypothesis, in contrast to a two-sided test that allows departures in either direction. The choice affects the test's interpretation and should be specified before examining results, because it depends on the scientific question and the planned decision rule. The registry identifies the sponsor's test as one-sided but does not report the direction coding or the full hypotheses. The study title describes non-inferiority, while the brief description also refers to investigating superiority. Those statements are not enough to reconstruct the exact hypothesis tested.

A t-test typically compares means under assumptions about the observations and their variability. General explanation: the standard form of a t-test relies on an appropriate model for the sampling distribution of the mean difference. The relevant assumptions and any adaptations depend on the design and data. The registry does not report whether the test used a pooled or unequal-variance approach, whether it was adjusted for covariates, or whether any alternative model was used. No test statistic or degrees of freedom are posted in the registry record.

Why are arm means not the whole comparison?

A treatment comparison focuses on a contrast between arms, together with uncertainty around that contrast. The registry gives each mean and its confidence interval, but not the estimated contrast or its confidence interval. General explanation: uncertainty for a contrast depends on variability and sample information in both groups and on the analysis method. Thus, two arm-specific confidence intervals cannot be used as a replacement for a direct interval for the contrast. The sponsor's p-value is a result for a stated test, but it does not provide the missing effect estimate or show the range of contrast values consistent with the data.

There is no reported recalculated comparison for the continuous endpoint in the registry record. Accordingly, no independent unadjusted contrast or p-value is available to compare with the sponsor's result. The arm-level means and intervals remain useful descriptions, but they do not substitute for the unavailable contrast calculation.

What is over-dispersion, and is it relevant?

General explanation: over-dispersion occurs when observed variability is greater than a statistical model assumes. It can make standard errors too small and confidence intervals too narrow if the model does not account for the extra variability. It is often discussed for count outcomes and certain generalized models. The primary endpoint is reported as a continuous mean, and the registry does not identify an over-dispersion model or provide the distributional information needed to assess this issue. It should not be assumed that over-dispersion was present or that a particular correction was required.

The registry lists planning assumptions for a separate sample-size illustration: two-sided significance level 0.05, equal allocation, normal approximation, and an effect to detect defined as a stated fraction of the observed effect. Such a calculation is a planning illustration based on the observed effect, not the trial's own design justification and not observed power. The registry does not report a sample-size justification from the trial protocol. A planning calculation cannot establish that the trial was adequately powered for its original objective, particularly when the non-inferiority margin and original assumptions are not reported in the registry record.

10. Analyses Reported by the Sponsor

The registry's sponsor-reported analysis is a one-sided t-test labeled as non-inferiority, with p = 0.125. The record also posts the arm means and their 95% confidence intervals. No recalculated comparison is available for this continuous endpoint: the registry record indicates that values for a two-arm comparison were missing or not estimable. There is therefore no independent unadjusted contrast estimate or p-value to place beside the sponsor's analysis.

The sponsor's result should be reported as an analysis conducted by the sponsor, not reconstructed from the arm-level summaries. A comparison of the two means, without the appropriate variance information and analysis specification, cannot reproduce the test. The record does not report a contrast estimate, test statistic, degrees of freedom, or interval for the between-arm difference. It also does not report the non-inferiority margin needed to judge the test against the intended boundary.

The posted p-value does not, by itself, establish whether the non-inferiority criterion was met. Its meaning depends on the null hypothesis, margin, analysis population, and prespecified decision rule, details that are not reported in the registry record. The trial record lists the test as one-sided and non-inferiority, so that description can be stated; a stronger interpretation cannot be independently verified from the posted statistics.

No discrepancy between a sponsor result and a recalculated comparison is identified in the registry record. In this case, the limitation is the lack of a recalculated two-arm comparison, rather than a conflict between two reported estimates.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists functional outcomes and wound infection as secondary outcomes, and states that three outcome measures and two analyses are posted. Those details do not establish how many hypotheses were planned, whether all outcomes were tested, or which results were designated confirmatory. The multiplicity strategy is not reported in the registry record. It is therefore not possible to determine whether statistical significance was adjusted across outcomes, analyses, or interim looks, or whether the primary endpoint was protected as the sole confirmatory test.

General explanation: when several hypotheses are tested, the chance of obtaining at least one apparently notable result can rise unless the analysis plan accounts for the number and relationship of the tests. Common approaches include prespecifying a hierarchy, adjusting significance thresholds, or treating secondary analyses as exploratory. These approaches have different interpretations, and none can be attributed to this trial without documentation. The registry's count of outcome measures and analyses is not enough to infer a correction or testing order.

The number and timing of interim analyses are not reported in the registry record. The record also does not report the protocol's statistical analysis plan or whether any outcomes were changed after study commencement. These omissions limit evaluation of selective reporting and the overall false-positive risk associated with multiple comparisons.

Missing-data handling is also not reported. In general, missing outcome measurements can affect both the estimated group means and the uncertainty around them, particularly if the probability of missingness is related to outcome values or treatment assignment. The record does not state whether measurements were unavailable, whether participants were excluded from the endpoint analysis, or whether imputation or sensitivity analyses were used. The posted arm counts of 82 each describe the outcome summaries, but the registry does not explain how those counts relate to the enrollment figure of 211.

The analysis population is not reported in the registry record. It is therefore unclear whether the primary analysis included all randomized participants, only participants with an available postoperative day 1 estimate, or another defined population. For a non-inferiority question, population definition and missing-data assumptions can be especially consequential because different analysis populations may affect the comparison. The registry record does not provide enough information to assess those choices or their influence on the sponsor-reported test.

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 1.2% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports serious adverse events for 82 participants at risk in each arm. It records zero affected participants with serious adverse events in the saline-coupled bipolar sealer arm and zero in the unipolar electrocautery arm. It reports zero deaths in both arms. For other adverse events, the record lists one affected participant in the saline-coupled bipolar sealer arm and three in the unipolar electrocautery arm, with 82 participants at risk in each arm.

These are the counts posted in the registry, not evidence that no adverse events of any kind occurred. The record distinguishes serious adverse events from other adverse events, and the reported figures should be read within those categories. The registry does not provide serious-event terms, and no top serious-event terms are listed. It also does not provide additional details about event definitions, severity grading, duration of follow-up, or how adverse events were collected.

General explanation: small event counts provide limited information about uncommon harms. A zero count in a group does not prove that the underlying risk is zero, and differences in sparse counts can be imprecise. The record's counts do not support a complete comparative safety assessment without further context, including the event definitions, observation period, and analysis plan. They should not be converted into a claim that either device is safer.

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

The registry record shows two arm-level estimates for postoperative day 1 estimated blood loss and reports a sponsor analysis labeled as a one-sided non-inferiority t-test. The saline-coupled bipolar sealer arm mean is 924 milliliters, with a 95% confidence interval from 178 to 1861. The unipolar electrocautery arm mean is 1062 milliliters, with a 95% confidence interval from 218 to 2053. The sponsor-reported test has p = 0.125.

These data do not, on their own, establish that one approach is superior, equivalent, or non-inferior to the other. Non-inferiority requires a defined margin and a test or interval interpreted against that margin. The margin and the full decision rule are not reported in the registry record. The record also does not post the estimated between-arm contrast or its confidence interval. Therefore, the available summary statistics do not show how precisely the difference between groups was estimated or whether it met the intended criterion.

A p-value is not a measure of the size or clinical importance of an effect. It is a measure calculated under a specified null hypothesis and model. The sponsor's value cannot be given a complete non-inferiority interpretation without knowing the margin, contrast direction, and analysis details. It is also not the probability that the null hypothesis is true or that the intervention is non-inferior.

The registry enrollment of 211 and the endpoint counts of 82 in each arm are reported separately. The record does not explain why the numbers differ or identify the analysis population, so no conclusions about participant flow or exclusions can be drawn from those counts alone. Missing-data procedures and the handling of incomplete outcomes are also not reported.

The record lists functional outcomes and wound infection as secondary outcomes, but their detailed results and the approach to multiple testing are not included in the registry record for this analysis. The safety information reports zero serious adverse events and zero deaths in each arm, and counts other adverse events in each arm. Such figures describe the recorded events but cannot rule out uncommon outcomes or settle a comparative safety question.

The appropriate interpretation is limited and descriptive: the registry posts the arm-specific endpoint summaries and a sponsor-reported test, but does not provide enough of the statistical plan or comparison detail to independently verify a non-inferiority conclusion. This is not medical advice and does not determine which device should be used.

14. Why This Trial Matters Statistically

This trial raises a useful statistical distinction between a study's broad clinical motivation and its formal inferential objective. The brief description refers to investigating superior hemostasis, whereas the official title describes a non-inferiority study. Superiority and non-inferiority are not interchangeable labels. Each needs a clearly specified hypothesis, analysis direction, and interpretation. When the registry record contains both descriptions but omits the non-inferiority margin, readers cannot fully reconstruct the intended statistical decision.

The outcome is a continuous estimate of blood loss, with arm means and confidence intervals posted. This makes the difference between separate group summaries and a direct treatment contrast especially important. Readers may be tempted to judge the comparison by looking only at the two means. Statistical interpretation instead requires an estimate of the contrast and its uncertainty, linked to the prespecified objective. That direct contrast is not reported in the registry record.

The record also illustrates why trial metadata and outcome results need to be read together. It describes the study as single-blinded in the brief study description, but lists double masking in its metadata. The discrepancy is unresolved. Because the outcome is an estimated quantity and the record does not describe who calculated or assessed it, the masking information cannot be used to assess the possibility of measurement influence.

The posted counts invite another important distinction: enrollment in the registry is not automatically the same as the number included in a particular endpoint analysis. The registry lists enrollment of 211 and primary outcome summaries with 82 participants in each arm, but does not describe participant flow or the analysis population. Without those details, readers should avoid attributing the difference to any specific reason.

Finally, the stated planning illustration is based on the observed effect rather than the trial's own design justification. General explanation: post hoc planning calculations cannot replace prospective sample-size assumptions or demonstrate observed power. The registry does not report the original justification or non-inferiority margin. These limitations make transparent reporting of the protocol, analysis plan, participant flow, and direct contrast estimates important for interpreting the findings.

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 trial-specific reporting gaps restrict interpretation. Most importantly, the registry does not report the non-inferiority margin, even though the sponsor's analysis is labeled non-inferiority. Without that margin and the precise test direction, the p-value cannot be assessed against the intended decision boundary. The brief description's reference to superior hemostasis also differs from the non-inferiority framing in the official title; the registry record does not explain how the objectives relate.

The endpoint summaries include arm means and 95% confidence intervals, but no direct estimate or confidence interval for the between-arm contrast. The registry also omits the test statistic, degrees of freedom, and details of the t-test implementation. The record indicates that a recalculated two-arm comparison was not available, so the sponsor result cannot be independently checked using a recalculation from the posted summaries.

The registry lists enrollment of 211, while the primary outcome results report 82 participants in each arm. Participant flow, reasons for non-inclusion in endpoint summaries, and the primary analysis population are not reported. Missing-data procedures and sensitivity analyses are also not reported. These omissions leave uncertainty about how the endpoint analysis relates to the randomized population.

The masking description is inconsistent: the study description calls the trial single-blinded, while registry metadata lists double masking. The record does not clarify who was masked or how masking was maintained. Allocation concealment and random-sequence generation are not reported. The stated restriction to two surgeons is intended to limit variation in surgical technique, but the record does not describe how surgeon-related factors were handled in the analysis.

The multiplicity strategy, interim analyses, and detailed statistical analysis plan are not reported. Secondary outcomes are named, but their results and role in the inferential strategy are not established by the available information. Finally, the safety summaries do not include event definitions, detailed serious-event terms, or follow-up context. These limitations call for a restrained reading of the registry results rather than conclusions extending beyond the reported summaries and test.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: To Evaluate the Safety and Efficacy of the Saline-coupled Bipolar Sealer in Primary Unilateral Total Knee Arthroplasty — NCT03952546.
• Registry record last updated 2023-02-15; status last verified 2023-02.
• Results first posted 2023-02-15.
• 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 registry describes a completed, randomized, parallel-group comparison of saline-coupled bipolar sealer with unipolar electrocautery in primary unilateral total knee arthroplasty. Enrollment is listed as 211. The primary endpoint is postoperative day 1 estimated blood loss in milliliters, calculated by Gross' Formula and summarized as a mean with a 95% confidence interval.

The posted mean is 924 milliliters in the saline-coupled bipolar sealer arm, with a 95% confidence interval from 178 to 1861; the unipolar electrocautery arm mean is 1062 milliliters, with a 95% confidence interval from 218 to 2053. The sponsor reports a one-sided t-test labeled non-inferiority, with p = 0.125. The record does not report the non-inferiority margin or a direct interval estimate for the arm contrast.

Functional outcomes and wound infection are listed as secondary outcomes. The record lists zero serious adverse events and zero deaths in each arm, and reports other adverse events in both arms. The masking description, analysis population, missing-data procedures, multiplicity strategy, and detailed statistical plan are not fully reported in the registry record.