Safety of Endoscopic Resection of Large Colorectal Polyps: A Randomized Trial.: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Safety of Endoscopic Resection of Large Colorectal Polyps: A Randomized Trial. (NCT01936948), a randomized na study. The primary outcome is Number of Participants With Delayed Bleeding Complications.

2. Clinical Question

The clinical context described in the registry is that resection of large polyps can be followed by complications, including delayed bleeding. The primary endpoint focuses on a defined subset of bleeding events, rather than every episode of bleeding: the event had to occur within 30 days and meet one or more specified clinical criteria. Those criteria include a hemoglobin decrease of at least 2gm, hemodynamic instability, an emergency department presentation, hospitalization, repeat colonoscopy, or another intervention.

The primary comparison groups participants by whether the defect was clipped, combining the two electrocautery modes within each closure category. That is the comparison addressed by the posted primary results. The registry also describes the four procedure combinations, but the primary result does not separately estimate each combination or establish whether the closure comparison differs by electrocautery mode. Such interaction results are not reported in the registry record.

3. Trial Design

The registry describes a randomized, parallel trial with no masking listed. It records an enrollment of 928 and a completed status. The listed start is 2013-04, and the primary completion date is 2018-06-30. The sponsor is White River Junction Veterans Affairs Medical Center. A phase is not applicable in the record.

Four procedure combinations are described: clip closure with EndoCut, clip closure with Coagulation, no clip closure with EndoCut, and no clip closure with Coagulation. Each description concerns resection of a non-pedunculated study polyp ≥20mm. The primary analysis, however, compares clip closure with no clip closure, pooling across the electrocautery modes. The record does not describe the randomization procedure, allocation concealment, or whether assignments were balanced within the electrocautery modes.

The primary outcome is a count of participants experiencing delayed bleeding complications during the 30 days following a study polyp resection. The analysis groups recorded 455 participants for clip closure and 464 for no clip closure; the endpoint description notes that data for 9 patients was not available. How those unavailable data were handled, and the analysis population definition beyond this description, are not reported in the registry record.

The registry lists secondary outcomes concerning the overall number of complications, complete study polyp resection, polyp recurrence, and complications associated with clip use. A detailed posted comparison is available for overall complications, while formal reported-analysis entries are absent. The original design’s statistical plan, including its planned sample-size justification, is not reported in the registry record. The planning illustrations associated with the observed primary effect are separate from the trial’s original design assumptions.

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: Some of the inclusion/exclusion criteria, e.g. polyp size, poor bowel preparation, could not be assessed until after the start of the colonoscopy. Therefore, informed consent was obtained and the patient enrolled in the study prior to the procedure. Only when it was determined that the patient was eligible for the study were they then randomized.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: As per study protocol, the primary analysis was done using clip/no clip closure as the study groups. Data for 9 patients was not available.

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 Delayed Bleeding Complications, measured over 30 days following a study polyp resection.

The primary endpoint counts participants with a qualifying delayed bleeding complication. The time window is 30 days following a study polyp resection. The registry defines a qualifying event as bleeding within 30 days after completion of the colonoscopy with a study polyp resection that is associated with at least one specified clinical consequence: a hemoglobin decrease of at least 2gm, hemodynamic instability, presentation to the emergency department, hospitalization, repeat colonoscopy, or another intervention.

This definition makes the endpoint narrower than bleeding of any severity. It emphasizes bleeding linked to a clinically consequential change or need for evaluation or treatment. Counting participants with an event also gives an interpretable binary outcome: each participant is classified according to whether the defined event occurred during the stated follow-up period. The registry reports counts, not a continuous measurement of bleeding severity or time until bleeding.

The endpoint is relevant to the stated study question because delayed bleeding is identified in the registry as a potential complication after large-polyp resection. Applying a specified event definition and follow-up window helps make the comparison more consistent. It does not capture outcomes outside that definition or after the stated 30-day period. Whether all qualifying events were independently adjudicated, and any additional endpoint validation procedures, are not reported in the registry record.

7. Results Posted in the Registry

• No Clip Closure: 7.1% (33/464) versus 3.5% (16/455) on control; risk difference 3.6 percentage points (95% CI 0.7 to 6.5), risk ratio 2.02 (1.13–3.62); unadjusted two-proportion z-test p = 0.015.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

The Overall Number of Complications

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • No Clip Closure: 9.5% (44/464) versus 4.8% (22/455) on control; risk difference 4.6 percentage points (95% CI 1.3 to 8.0), risk ratio 1.96 (1.20–3.22); unadjusted two-proportion z-test p = 0.006.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for No Clip Closure than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The registry lists secondary outcomes spanning overall complications, complete study-polyp resection, recurrence, and complications associated with clip use. The detailed posted secondary comparison concerns overall complications. Its direction is consistent with the primary comparison: the clip-closure group has fewer recorded participants with the outcome than the no-clip group. This broad pattern can be considered alongside the primary result, but it does not replace or independently confirm it.

The secondary outcomes cover different aspects of the procedure and follow-up. Overall complications is broader than the primary delayed-bleeding definition, while resection completeness, recurrence, and complications associated with clip use concern other outcome domains. Results for those additional outcomes are not reported in the registry record. Their omission means the overall secondary pattern cannot be generalized to every listed outcome, and the primary endpoint should not be used as a proxy for them.

Secondary findings generally carry less weight when they are unadjusted and when the number of outcomes examined is not accounted for. Testing multiple outcomes creates more opportunities for a result to appear notable by chance. The registry does not report a multiplicity adjustment or a detailed prespecified hierarchy for secondary analyses. The overall-complications result is therefore best treated as supportive descriptive evidence, not as a standalone confirmatory finding. The registry’s formal reported-analysis entries do not provide further context about how the secondary comparisons were planned or interpreted.

The different outcome definitions also matter. A broad complication outcome can include events that are not part of the primary endpoint, and the same event might contribute to more than one outcome. Whether and how outcomes overlap is not reported in the registry record. Consequently, agreement in direction across posted results does not necessarily represent independent evidence from unrelated events or analyses.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• The Overall Number of Complications
• Complete Study Polyp Resection Rate
• Polyp Recurrence Rate
• The Number of Complications Associated With Clip Use

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is binary at the participant level: a participant either had a qualifying delayed bleeding complication during the stated follow-up window or did not. The posted counts and group sizes therefore support comparisons of event occurrence. They are not measurements of the total number of bleeding episodes, the severity of each event, or the time to an event.

What does the risk difference describe?

A risk difference subtracts one group’s event risk from the other group’s event risk. In this comparison, the estimated difference is expressed in the direction of no clip closure relative to clip closure. A positive estimate therefore indicates a higher observed event risk in the no-clip group. The reported estimate is 0.036. The confidence interval is an uncertainty range around that estimate; its lower and upper limits are both positive, so the reported interval does not include a zero difference.

The group estimates associated with the primary comparison are 0.035 for clip closure and 0.071 for no clip closure. These are displayed as risk-scale estimates. They are distinct from the participant counts, and the risk difference is the reported contrast between the groups. No new estimate is needed to interpret the direction: the no-clip group has the higher posted event count and the higher group risk estimate.

How do relative measures differ?

The risk ratio expresses one group’s risk relative to the other group’s risk. The posted risk ratio of 2.022 is above 1, which is consistent with a higher observed risk in the no-clip group. Its confidence interval, 1.129 to 3.623, also lies above 1. The odds ratio is another relative measure, based on odds rather than risks. Its estimate is 2.101, with a confidence interval from 1.14 to 3.873. Since event occurrence is represented by counts and group sizes, the risk difference, risk ratio, and odds ratio each summarize the same comparison on a different scale. They are not different outcomes.

What does the p-value contribute?

The primary unadjusted p-value is 0.015. A p-value describes how compatible the observed comparison is with a statistical null model of no group difference, under the assumptions of the test used. A smaller value indicates less compatibility with that null model, but it is not the probability that the null hypothesis is true and it does not measure the size or importance of the difference. The registry record does not identify the test used, so details such as its exact assumptions cannot be verified from the posted record.

The word unadjusted matters. It indicates that the reported p-value is not presented as corrected for the number of outcomes or comparisons. The registry record does not report a multiplicity procedure. When many outcomes or comparisons are considered, chance alone can produce small p-values more often than when a single prespecified comparison is considered. This is one reason to give the designated primary outcome more weight than exploratory or unadjusted secondary findings.

Why can over-dispersion matter?

Over-dispersion is a general statistical issue in count models: variation in counts may exceed what a model assumes. If that occurs and the model does not account for it, uncertainty can be understated and intervals or tests may be too optimistic. This primary endpoint is posted as participants with an event, rather than a count of repeated events per participant, and the registry does not report a count model or an over-dispersion assessment. Therefore, over-dispersion is a general concept, not a documented analysis feature of this trial.

What cannot be confirmed from the record?

The registry does not report the original statistical plan, the sample-size justification, the method for interval estimation, adjustment variables, or an analysis population definition beyond the statement that data for 9 patients was not available. It also does not report interim analyses. These absences mean that the posted summary estimates can be described, but the full analysis workflow and the extent to which it followed a prespecified plan cannot be assessed from the registry record.

11. Confidence Intervals and Precision

A confidence interval describes a range of values compatible with the observed data under the statistical method used. It conveys uncertainty around an estimate; it does not guarantee that the true effect lies within the interval or account for every possible source of bias. The method used for the posted intervals is not reported in the registry record.

For the primary risk difference, the reported estimate is 0.036, and both confidence limits are above zero. The interval therefore supports a direction in which the no-clip group has the higher estimated risk, while also showing uncertainty about the size of that difference. The reported limits are 0.007 and 0.065. Because those limits are more finely stated than the three-decimal display limit used for other values, the interpretation emphasizes their relation to zero rather than restating them.

The relative-effect intervals are also informative about precision. The risk ratio is 2.022, with limits 1.129 and 3.623. The odds ratio is 2.101, with limits 1.14 and 3.873. Both intervals are above 1, the null value for a ratio. Their breadth indicates that the data are compatible with a range of relative effect sizes, even though the estimates point in the same direction.

Precision is not the same as clinical importance. The risk difference communicates an absolute contrast, while the ratios communicate relative contrasts. The registry record does not define a threshold for a clinically important difference, so statistical uncertainty cannot be translated into a judgment about importance. Confidence intervals also do not address missing-data bias, outcome misclassification, or limitations in how participants were selected or analyzed. Those issues require information beyond the intervals themselves.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are participants per arm, not total participants across the trial and not total events. They are illustrations calculated from the observed primary effect, not the original trial’s sample-size justification. The planning assumptions listed in the registry record are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. Each scenario uses the stated percentage of the observed effect as the effect to detect.

For planning at 100% of the observed effect, the illustration gives 610 participants per arm for 80% power and 816 participants per arm for 90% power. Planning for 75% of the observed effect gives 998 participants per arm for 80% power and 1,336 participants per arm for 90% power. Planning for 50% of the observed effect gives 2,049 participants per arm for 80% power and 2,743 participants per arm for 90% power.

In plain terms, the 80% power column describes the estimated participants per arm needed, under the stated assumptions, for an analysis to have an 80% chance of detecting the specified effect if that effect were the true one. The 90% column uses a higher target chance of detection and therefore requires more participants in these illustrations. Smaller effects are harder to distinguish from random variation, so planning for a smaller fraction of the observed effect requires more participants per arm.

Any curve depicting these planning scenarios should be read as a relationship between the assumed effect size and the required per-arm sample size, not as a description of what happened during the trial. This is not observed power. The registry record does not report the trial’s original design assumptions, so these figures should not be taken as evidence about the rationale for its enrollment target.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.035; treatment proportion p₁ = 0.071; average p̄ = 0.053.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.3172 = 0.6218.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.3162 = 0.2661.
4. n = (0.6218 + 0.2661)² / (0.0360)² = 609.8, rounded up to 610 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
p0 <- 0.0351648351648352                          # control proportion
p1 <- 0.0711206896551724                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(1936948)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 610 (80% power), 816 (90% power)
 75% of observed effect: n per arm = 998 (80% power), 1,336 (90% power)
 50% of observed effect: n per arm = 2,049 (80% power), 2,743 (90% power)
power.prop.test: 609.8 
Simulated power at n = 610 per arm: 0.806 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Multiplicity, Missing Data and Analysis Populations

The registry record does not report a multiplicity strategy for the primary and secondary outcomes. In general, multiplicity arises when several outcomes, comparisons, or analyses are examined: without an adjustment or prespecified hierarchy, the chance of at least one apparently notable result can increase. The primary endpoint is identified, but details of a testing hierarchy, family-wise error control, or other multiplicity procedure are not reported in the registry record. The unadjusted secondary results should therefore be interpreted cautiously.

The record lists four secondary outcome areas, but detailed results are available for overall complications only. The registry’s reported-analysis entries do not describe the analyses. Results for complete study-polyp resection, polyp recurrence, and complications associated with clip use are not reported in the registry record. The presence of an outcome in the registry is not evidence that a result was statistically significant or that an analysis was completed and posted.

For the primary endpoint, the record states that data for 9 patients was not available. It does not describe the reasons for missingness, whether the unavailable data differed across groups, or the method used to handle missing outcomes. No imputation approach, sensitivity analysis, or missing-data assumption is reported. As a result, the impact of missing endpoint data on the posted comparison cannot be assessed from the registry record.

The primary outcome description says the analysis used clip/no-clip closure as the study groups. It does not provide a fuller definition of the analysis population or state whether all randomized participants were analyzed in their assigned groups. The record also does not report deviations from assigned procedures, exclusions after randomization, or protocol-adherence analyses. These limitations do not change the posted counts, but they restrict assessment of how closely the analysis preserves the randomized comparison.

The original statistical plan, interim analyses, and sample-size justification are not reported in the registry record. Without those details, it is not possible to determine whether the displayed analyses were all planned in advance or how the study addressed repeated looks at the data. Such information is important for assessing the confirmatory status of results, particularly when multiple outcomes are listed.

15. 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: postprocedure hemorrhage (7.4%), perforation (1.3%), intraprocedural bleeding (0.4%), postpolypectomy syndrome (0.4%), abdominal pain (0.4%).

The registry presents serious adverse-event figures separately for the four procedure combinations. The serious-event counts are 11 for Clip Closure + EndoCut, 11 for Clip Closure + Coagulation, 21 for No Clip Closure + EndoCut, and 23 for No Clip Closure + Coagulation. The corresponding numbers at risk are 227, 228, 234, and 230. Deaths are recorded as 0, 0, 2, and 0 in those same groups. These are descriptive counts; the registry does not provide a comparative statistical analysis of these serious-event figures.

The listed serious-event terms include postprocedure hemorrhage at 7.4%, perforation at 1.3%, intraprocedural bleeding at 0.4%, postpolypectomy syndrome at 0.4%, abdominal pain at 0.4%, and other at 0.4%. The denominator used for these percentages is not reported in the registry record. The term-level percentages should not be assumed to be group-specific or directly comparable with the primary endpoint.

The primary endpoint has its own event definition and 30-day window. The serious adverse-event listing is not interchangeable with the primary delayed-bleeding comparison, and the record does not establish how those categories overlap. These figures describe reported safety events; they do not by themselves identify causation or quantify a treatment effect.

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

The posted primary data show fewer recorded participants with the defined delayed bleeding outcome in the clip-closure group than in the no-clip-closure group. The estimated risk difference is 0.036, and the reported interval lies above zero. The relative measures also point toward higher observed event occurrence in the no-clip group. These are statistical descriptions of the recorded comparison, not a complete account of the procedures’ effects in every setting or population.

The result applies to the registered comparison of closure groups for resection of non-pedunculated study polyps ≥20mm, with the primary analysis combining the electrocautery modes. It does not separately answer whether the same contrast holds within each mode. Nor does the primary endpoint describe all complications, resection completeness, recurrence, or events outside the stated 30-day period.

A randomized design can support causal comparisons when allocation and analysis preserve the randomized groups, outcome ascertainment is appropriate, and missing data and other departures are handled adequately. The registry record identifies randomization, but it does not describe enough implementation or analysis detail to assess all of those conditions. In particular, the primary endpoint description notes unavailable data for 9 patients, with no missing-data method reported.

The estimates are subject to statistical uncertainty, as reflected in their confidence intervals. Those intervals do not incorporate every concern about study conduct, generalizability, or outcome measurement. The registry record does not provide participant characteristics needed to judge how closely the enrolled population resembles other patient groups, nor does it report a threshold for clinical importance.

Accordingly, the posted results support a description of the observed difference and its statistical precision, while leaving important questions about analysis choices and broader applicability unresolved. They should not be read as medical advice or as a universal conclusion about all polyp resections or electrocautery settings.

17. Why This Trial Matters Statistically

This trial illustrates why a clinical endpoint needs a precise definition. The outcome is not simply any bleeding; it is a bleeding event within a specified time window and associated with defined clinical consequences. That distinction affects which events count and what the resulting estimate means. Clear endpoint definitions help readers understand the target of a statistical comparison and avoid extending a result to outcomes that were not measured in the same way.

It also demonstrates the importance of distinguishing an absolute measure from relative measures. The risk difference gives the contrast on the risk scale, while the risk ratio and odds ratio express relative contrasts. Each can be informative, but they answer different questions. Confidence intervals add information about uncertainty that a point estimate alone cannot convey.

The trial’s four procedure combinations and pooled primary comparison highlight another statistical issue: an overall contrast does not necessarily describe each combination separately. When treatment features vary together, separate estimates or an interaction analysis may be needed to assess whether the association differs by feature. Such results are not reported in the registry record, so the primary pooled comparison should be kept within its stated scope.

Finally, the record illustrates how conclusions depend not only on observed counts but also on missing-data handling, multiplicity, and the analysis population. The registry notes unavailable primary endpoint data but does not describe the handling method, and it does not report a multiplicity plan. These details shape how confidently a result can be treated as confirmatory. Statistical interpretation therefore requires attention to both the estimate and the information available about the process that produced it.

18. 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 affect interpretation. The primary comparison pools participants across EndoCut and Coagulation electrocautery modes. Although the four combinations are described, separate results for each combination and an interaction analysis are not reported in the registry record. The pooled result therefore cannot establish whether the closure comparison is similar across modes.

The primary endpoint is restricted to defined delayed bleeding events during 30 days following resection. It does not describe later bleeding or all possible complications. The registry lists other outcomes, but detailed results are not reported for complete resection, recurrence, or complications associated with clip use. The record also does not explain whether outcome events overlap across the primary, secondary, and adverse-event listings.

Data for 9 patients was not available for the primary endpoint, and the handling of those data is not described. Reasons for missingness, group-specific missingness information, and sensitivity analyses are not reported. The record does not fully define the analysis population or describe how deviations from assigned procedures were handled.

No masking is listed, and the registry does not report how outcome assessment was conducted or whether event classification was blinded. The randomization method, allocation concealment, participant characteristics, and detailed statistical plan are also not reported. These omissions limit assessment of potential bias, applicability, and whether analysis choices were prespecified.

Finally, secondary findings are unadjusted in the posted comparison, with no multiplicity strategy reported. The registry’s overall-complications result should therefore be treated as less conclusive than a prespecified, multiplicity-aware analysis would be. These limitations describe gaps in the registry record; they do not establish that a particular bias occurred.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Safety of Endoscopic Resection of Large Colorectal Polyps: A Randomized Trial. — NCT01936948.
• Registry record last updated 2024-07-12; status last verified 2024-06.
• Results first posted 2021-06-22.
• 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.

22. Record Summary

The registry record describes a completed randomized, parallel trial with no masking listed and an enrollment of 928. It compares clip closure with no clip closure after resection of a non-pedunculated study polyp ≥20mm, using two electrocautery modes in the four procedure combinations. The primary endpoint counts participants with a qualifying delayed bleeding complication within 30 days.

The posted primary comparison records 16 events with clip closure and 33 with no clip closure. The estimated risk difference is 0.036, with an unadjusted p-value of 0.015; the reported relative measures point in the same direction. Overall complications is the only secondary outcome with a detailed posted comparison. Other listed secondary results, the original statistical plan, the sample-size justification, and methods for handling unavailable primary endpoint data are not reported in the registry record.