Disposable Powered Articulating Linear Cutter Stapler in Total-Thoracoscopic Anatomic Lobectomy (Segmentectomy): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Disposable Powered Articulating Linear Cutter Stapler in Total-Thoracoscopic Anatomic Lobectomy (Segmentectomy) (NCT05329402), a randomized na study. The primary outcome is Number of Participants With Anastomosis Success.

2. Clinical Question

The question is specific to the devices and surgical context studied. The endpoint reflects the registry’s definition of successful cutting and anastomosis during surgery; it does not by itself measure longer-term patient outcomes or compare all aspects of the devices. The record reports 164 enrolled participants and describes the trial as randomized. It does not report enough detail to determine how participants were selected, how the randomization sequence was generated, or whether the observed results apply to other populations or surgical settings.

3. Trial Design

The registry describes a randomized, quadruple-masked, parallel-group device trial. It lists an enrollment of 164 and a completed status. The study start date is 2018-10-22, and the primary completion date is 2019-09-11. The record identifies lung cancer as the condition under study and names Fengh Medical Co., Ltd. as the sponsor. The phase is listed as NA.

Participants were assigned to one of two device groups: the Disposable Powered Articulating Endoscopic Linear Cutter Stapler or the ECHELON Flex Powered Articulating Endoscopic Linear Cutters. The posted primary outcome includes 79 participants in the intervention arm and 83 in the comparator arm. The registry record does not describe the randomization sequence, allocation concealment, the individuals masked, or how masking was maintained. Those details matter when assessing the risk of bias, especially for outcomes that involve intraoperative judgment.

The primary endpoint was assessed during surgery. It classified the device’s cutting and anastomosis as successful when all staple lines were complete and there was no air leakage or bleeding. Persistent air leakage and bleeding at the anastomosis site, with conversion to thoracotomy required, met the stated definition of failure. The registry reports one outcome measure and one analysis. It does not report a detailed protocol, planned enrollment rationale, analysis populations, or other design features such as interim analyses. No study phase beyond NA is reported in the registry record.

The design targets a comparison between two devices in a defined operative setting, rather than a comparison between a device and no device. Its randomized structure is intended to support a fair comparison, but the available record does not give the procedural details needed to evaluate how randomization and masking were implemented. The posted results therefore describe the recorded endpoint and analysis, while questions about the trial’s operational conduct remain unanswered.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Number of Participants With Anastomosis Success, measured over During surgery.

The primary endpoint is the number of participants with anastomosis success during surgery. The registry defines success by the condition of the staple lines after the device is withdrawn: all lines must be complete, with no air leakage or bleeding. Persistent leakage and bleeding at the anastomosis site that requires conversion to thoracotomy is defined as failure. The outcome is therefore a binary participant-level measure, recorded as success or failure.

This endpoint is used to assess the immediate technical performance of the devices in the procedure for which they were studied. It focuses on the cutting and anastomosis step, rather than on outcomes measured after surgery or on a broader assessment of patient health. The stated trial objective was to evaluate whether the success rate with the intervention device was not inferior to the success rate with the comparator.

The posted counts are 79 successes among 79 participants in the intervention arm and 83 among 83 in the comparator arm. These counts describe the recorded primary outcome; they do not establish how outcomes were handled for participants who might have had missing endpoint data, since missing-data handling is not reported in the registry record. The record also does not describe endpoint adjudication, assessor training, or whether the success judgment was independently reviewed. Those details would help readers understand how consistently this intraoperative endpoint was applied.

7. Results Posted in the Registry

• ECHELON Flex Powered Articulating Endoscopic Linear Cutters: 100.0% (83/83) versus 100.0% (79/79) on control; risk difference 0.0 percentage points (95% CI 0.0 to 0.0), risk ratio 1.00 (1.00–1.00); unadjusted two-proportion z-test p = n/a.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is binary at the participant level: each participant is classified as having anastomosis success or not having success according to the registry definition. For a binary outcome, a group’s event proportion is often compared with that of another group. A risk difference expresses the absolute difference between those proportions. In this trial’s posted counts, both groups have a success proportion of 1.0, and the unadjusted risk difference is 0.0.

What does the sponsor’s analysis report?

The registry lists a Wald method and labels the parameter as Mean Difference (Final Values). It identifies the analysis type as noninferiority and reports an estimate of 0, a 95% confidence interval from -0.029 to 0.027, and p = 0.973. These are the sponsor’s posted inferential results. The registry record does not describe the calculation in enough detail to determine how the mean-difference analysis was implemented for this endpoint, which analysis population it used, or how it relates to the simple comparison of the posted counts.

A Wald analysis generally uses an estimate and an estimated standard error to form an interval and test statistic. This is a general explanation of the method, not a claim about undocumented implementation details in this trial. The method’s behavior depends on the data and assumptions used for the estimate and standard error. For a binary outcome with no observed failures in either group, an ordinary unadjusted calculation encounters a boundary case: there is no observed within-group variation in the outcome. That helps explain why a simple recalculation can produce a degenerate interval, but it does not explain the sponsor’s different interval. The registry does not report enough detail to resolve that difference.

What does noninferiority require?

In general, a noninferiority trial asks whether the intervention is not worse than the comparator by more than a prespecified, clinically acceptable margin. The margin is part of the decision rule, not something that can be inferred from the observed estimate or p-value. The registry record does not report the margin or its rationale. Without it, readers cannot determine whether the sponsor’s confidence interval meets the trial’s stated noninferiority criterion. The posted label identifies the type of analysis, but the label alone is not enough to verify the conclusion.

A confidence interval gives a range of values compatible with the data and the statistical method under its assumptions. For noninferiority, the interval’s position relative to the margin is central. The registry reports the interval endpoints, but not the margin. A p-value summarizes evidence against a specified null hypothesis under a statistical model; it is not a direct probability that the null is true and does not independently establish noninferiority. The registry does not state the precise null hypothesis or testing conventions beyond identifying the analysis as noninferiority.

Why distinguish the recalculation?

The unadjusted recalculation uses the posted group counts and is a basic check of the observed absolute difference. It is not the sponsor’s analysis and should not be mistaken for an alternative formal trial conclusion. The observed outcome is at the success boundary in both groups, which limits what an ordinary variance estimate can say about uncertainty. The recalculation’s interval from 0 to 0 is a consequence of that simple calculation, not a guarantee that the underlying success probabilities are known without uncertainty.

Statistical methods for binary outcomes can behave differently when event counts are at a boundary, especially when a group has no observed failures. General approaches may use different assumptions or corrections, but the registry record does not say whether any such choices were made. It also does not report adjusted analyses, covariates, sensitivity analyses, or methods for handling missing outcomes. Those details should not be inferred from the Wald label alone.

What is not documented?

The record reports one primary outcome measure and one analysis. It does not report a sample-size justification, planned noninferiority margin, statistical analysis plan, interim analyses, stopping rules, or a description of the analysis population. The posted sample-size illustrations are planning scenarios based on the observed effect, not evidence about the trial’s original design assumptions and not observed power. Because the observed effect is zero in the simple recalculation, such illustrations should not be interpreted as demonstrating that the trial had adequate power to establish noninferiority.

In general, power is a design property calculated for a specified effect, error rate, and analysis plan. It is not a retrospective probability that the trial has demonstrated its hypothesis. A planning calculation based on an observed estimate is particularly limited when the estimate is at a boundary or when the margin and original assumptions are unknown. No numerical power conclusion about this trial should be drawn from the registry record.

10. Confidence Intervals and Precision

The sponsor-reported 95% confidence interval for the mean difference extends from -0.029 to 0.027, with a reported estimate of 0. A confidence interval is a range produced by a statistical method to represent uncertainty around an estimate under that method’s assumptions. It is not a statement that the true effect has a specified probability of lying inside the particular interval, and its interpretation depends on the analysis and the data used.

For this trial’s noninferiority question, the interval must be interpreted in relation to the prespecified noninferiority margin. That margin is not reported in the registry record. The interval therefore indicates the range reported by the sponsor, but the record does not allow readers to judge whether its lower limit satisfies the trial’s decision threshold. The point estimate of 0 is not enough to make that judgment.

The recalculated interval is 0 to 0 for the simple unadjusted risk difference based on the posted counts. This apparent precision is a feature of the calculation when every participant in each arm has a recorded success, not proof that the underlying device performance is known exactly. With no observed failures, the data provide little direct information about how often failures might occur outside the observed participants. The sponsor-reported interval is wider and comes from the sponsor’s stated analysis; the registry record does not explain why the two intervals differ.

Precision should also be considered alongside design and reporting details. The registry does not report the analysis population, handling of missing data, or the noninferiority margin. These omissions limit interpretation even though the posted counts are complete within the listed arm totals. The confidence intervals describe statistical uncertainty under their respective methods; they do not address possible selection, measurement, or generalizability limitations.

11. Analyses Reported by the Sponsor

The sponsor’s own analysis is reported as a Wald analysis of Mean Difference (Final Values), labeled as a noninferiority analysis. Its estimate is 0, its 95% confidence interval runs from -0.029 to 0.027, and its p-value is 0.973. These figures are distinct from the simple unadjusted recalculation based directly on the posted endpoint counts, which gives a risk difference of 0 and an interval from 0 to 0.

The recalculation is a basic comparison of the recorded binary outcomes. It does not reproduce the sponsor’s stated inferential method, and it should not replace that analysis. The sponsor’s analysis is the formal result posted in the registry, but the record does not explain its detailed implementation or how it relates to the group counts. The difference in interval width should not be resolved by assuming an undocumented method or data set.

The registry record does not report the noninferiority margin. As a result, the sponsor’s estimate, confidence interval, and p-value cannot by themselves establish whether the prespecified noninferiority criterion was satisfied. The reported p-value is not a substitute for the margin-based comparison required for a noninferiority interpretation. The posted result can be described accurately, while the conclusion remains difficult to verify from the registry record.

12. Multiplicity, Missing Data and Analysis Populations

The registry lists one outcome measure and one analysis, and it reports no secondary outcomes. This limits the visible scope for assessing multiple outcome comparisons in the posted results, but it does not establish that no other analyses were conducted. A multiplicity strategy, including any adjustment or hierarchy for testing, is not reported in the registry record. The record also does not provide a statistical analysis plan that would show which analyses were prespecified.

Missing-data handling is not reported in the registry record. The arm results list 79 participants for the intervention device and 83 for the comparator, with success counts matching those group totals. That describes the posted endpoint counts; it does not establish whether there were no missing data, whether all randomized participants were included, or how participants with unavailable outcomes would have been treated. The analysis population, such as intention-to-treat or per-protocol, is also not reported.

These omissions are relevant to noninferiority trials in general because exclusions and missing outcomes can affect comparisons between groups. This is a general methodological consideration, not evidence that such problems occurred in this trial. The record does not describe withdrawals, exclusions after randomization, protocol deviations, or sensitivity analyses for missing outcomes. Without those details, the available results cannot show how robust the sponsor’s conclusion would be to alternative analysis populations or assumptions.

The record reports a randomized parallel design and a quadruple-masked designation, but does not identify who was masked or describe the procedures. It also does not report interim analyses or stopping rules. These items should be treated as unreported details, not as evidence that a particular procedure was or was not used.

13. 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 serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: pneumothorax (2.5%), right pneumothorax (2.4%), lung leak gas (2.4%), pulmonary artery bleeding (1.3%), aerodermectasia (1.3%).

The adverse-event results list serious adverse events affecting 10 of 79 participants in the intervention-device group and 8 of 83 in the comparator-device group. Other adverse events are listed for 44 of 79 and 54 of 83 participants, respectively. Deaths are listed as 0 in each group. These are the counts posted in the registry record; they should be read as descriptive safety information rather than as proof of a difference between devices.

The registry lists pneumothorax at 2.5%, Right pneumothorax at 2.4%, Lung leak gas at 2.4%, Pulmonary artery bleeding at 1.3%, Aerodermectasia at 1.3%, and Thoracic wall lung hernia at 1.3% among the top serious terms. The record does not give enough detail to determine whether terms overlap, how events were attributed, or whether the listed terms represent all serious events. Their percentages are reproduced as reported.

Counts of affected participants do not by themselves establish that an event was caused by a device, and the registry record does not report a causal assessment. The adverse-event figures also do not replace the primary endpoint: success at the staple line and safety events are different outcome measures. No comparative statistical analysis of the adverse-event data is reported. Readers should therefore treat these counts as a description of the posted safety results, with attribution and analysis details not reported in the registry record.

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

The posted results show that every participant counted in the primary endpoint was classified as having anastomosis success in both device groups. The simple risk difference is 0, while the sponsor reports a noninferiority analysis with an estimate of 0 and a 95% confidence interval from -0.029 to 0.027. These results describe the observed endpoint and the sponsor’s reported statistical analysis. They do not establish that the devices are exactly equivalent.

Noninferiority is a specific claim: it depends on whether the result is within a prespecified margin. Because that margin is not reported in the registry record, the formal noninferiority decision cannot be checked from the available information. A p-value of 0.973 does not settle that question. The unadjusted recalculation’s interval from 0 to 0 is also not evidence of perfect certainty; it reflects the absence of observed failures in the listed groups under that simple calculation.

The endpoint concerns technical success during surgery. It does not answer questions about longer-term outcomes, recovery, quality of life, or other aspects of care, which are not reported as outcomes in the registry record. Nor does a single endpoint establish that the findings apply to other devices, populations, surgeons, or operative settings. The record describes a lung cancer population and thoracoscopic anatomic lobectomy or segmentectomy, but provides limited detail about participant characteristics and study conduct.

The adverse-event counts add safety information but do not by themselves support a causal comparison. The registry lists serious and other adverse events, along with zero deaths in both arms, but does not report comparative safety analyses or event-attribution details. The strongest warranted reading is therefore descriptive: the posted groups had the stated primary success counts and adverse-event counts, and the sponsor posted a noninferiority analysis whose criterion cannot be verified without the margin and additional methodological details.

15. Why This Trial Matters Statistically

This trial illustrates why a result that appears identical at the level of observed counts can still require careful statistical interpretation. Every participant listed in each arm had a recorded primary-endpoint success, so the unadjusted risk difference is 0. That observation is clear, but it does not provide direct evidence about the frequency of failures beyond the participants observed. A boundary result can make a simple uncertainty calculation look more certain than the underlying evidence warrants.

The distinction between a difference test and a noninferiority test is central. Noninferiority depends on a prespecified margin and a confidence interval judged against that margin. The registry reports a noninferiority analysis and its interval, but not the margin. Without that essential design element, readers cannot independently assess whether the formal claim follows from the posted result. This is a reporting limitation, not a basis for assuming that the analysis did or did not meet its criterion.

The comparison also shows why it is useful to keep a transparent recalculation separate from the sponsor’s analysis. A direct comparison of the posted counts is easy to describe, while the sponsor reports a Wald mean-difference analysis with a different interval. The record does not explain the difference. Retaining both results, with their methods distinguished, makes clear what can be checked from the counts and what depends on undocumented analysis choices.

Finally, the trial’s statistical value is shaped by more than its endpoint estimate. Information about the analysis population, missing-data procedures, randomization implementation, masking, sample-size rationale, and prespecified plan would help readers evaluate reliability. The registry record’s limitations make cautious interpretation necessary, even though its primary counts are straightforward.

16. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The registry record does not report the noninferiority margin, its rationale, or the rule used to declare noninferiority. This is a central limitation because the sponsor’s interval cannot be compared with the decision threshold without it. The record also does not explain the relationship between the sponsor’s Wald mean-difference analysis and the simple recalculation from the posted success counts.

Details needed to assess trial conduct are limited. The record identifies randomization and quadruple masking, but does not report the sequence-generation method, allocation concealment, who was masked, or how masking was maintained. It does not report the analysis population, handling of missing outcomes, participant withdrawals, exclusions, protocol deviations, or sensitivity analyses. The absence of these details in the registry record does not establish that the procedures were absent; it means their implementation cannot be assessed from this record.

The endpoint is an intraoperative technical measure. The record does not describe independent adjudication or how judgments about staple-line integrity, air leakage, bleeding, and conversion were standardized. It does not report longer-term outcomes as part of the posted results. The primary endpoint counts also show no observed failures in either arm, a boundary situation that limits the information available for a straightforward unadjusted estimate of uncertainty.

The record reports no sample-size justification or original planning assumptions. The planning illustrations based on the observed effect are not a substitute for those details and do not demonstrate observed power. The record does not report interim analyses or stopping rules. Generalizability is also difficult to assess because detailed participant characteristics, eligibility criteria, and study-site information are not reported in the registry record. These limitations call for restraint in interpreting the findings beyond the recorded endpoint and trial groups.

17. Related Tutorials

18. Related Statistical Calculators

19. Sources

• ClinicalTrials.gov: Disposable Powered Articulating Linear Cutter Stapler in Total-Thoracoscopic Anatomic Lobectomy (Segmentectomy) — NCT05329402.
• Registry record last updated 2024-05-02; status last verified 2023-11.
• Results first posted 2024-05-02.
• 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.

20. Record Summary

NCT05329402 is a completed randomized, quadruple-masked, parallel-group device trial sponsored by Fengh Medical Co., Ltd. It lists an enrollment of 164 and studies lung cancer surgery involving thoracoscopic anatomic lobectomy or segmentectomy. The intervention was the Disposable Powered Articulating Endoscopic Linear Cutter Stapler; the comparator was the ECHELON Flex Powered Articulating Endoscopic Linear Cutters.

The primary endpoint was the number of participants with anastomosis success during surgery. The posted counts were 79 of 79 in the intervention arm and 83 of 83 in the comparator arm. The sponsor reported a noninferiority analysis with an estimate of 0, a 95% confidence interval from -0.029 to 0.027, and p = 0.973. The noninferiority margin is not reported in the registry record, so the formal decision cannot be independently assessed from the posted information.

The registry also lists serious and other adverse-event counts and zero deaths in both groups. Analysis populations, missing-data procedures, detailed masking and randomization methods, and the sample-size rationale are not reported in the registry record.