SpHincterotomy for Acute Recurrent Pancreatitis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for SpHincterotomy for Acute Recurrent Pancreatitis (NCT03609944), a randomized na study. The primary outcome is Reduce the Risk of Subsequent Acute Pancreatitis Episodes by 33%.

2. Clinical Question

The clinical question concerns whether adding ERCP with minor papilla endoscopic sphincterotomy to an endoscopic ultrasound procedure changes the occurrence of a subsequent acute pancreatitis episode in people with pancreas divisum and recurrent pancreatitis. The primary outcome is described as the next acute pancreatitis attack occurring more than 30 days after treatment allocation, assessed as a time-to-event outcome through a maximum follow-up of 48 months. The title of the primary outcome states an aim to reduce the risk of subsequent episodes by 33%; that wording describes the stated aim, not the observed result.

The comparison is between two procedure pathways rather than between a procedure and no endoscopic care: both groups are described as receiving EUS. The sham comparator is intended to provide a procedural comparison, but the registry record does not specify which participants, clinicians, or outcome assessors were masked. The posted outcome counts describe participants recorded with an event, while the stated primary outcome also has a time-to-event definition. Those are related but not identical summaries: counts alone do not convey when events occurred or how follow-up time contributed to the comparison.

3. Trial Design

The registry describes a randomized, double-masked, parallel-group trial. The intervention comparison is EUS + ERCP with miES versus EUS + Sham. The trial is recorded as completed, with an enrollment of 181. The registry lists a study start date of 2018-09-27 and a primary completion date of 2025-02-28. Phase is recorded as NA.

The brief description says participants were assigned at random to ERCP with sphincterotomy or a sham procedure. The detailed arm descriptions place EUS in both groups: one arm receives EUS plus ERCP with miES, while the other receives EUS plus sham. The registry record does not describe the random sequence generation, allocation concealment, or the masking procedures in enough detail to establish who remained unaware of group assignment. The label “double” is recorded, but the two masked roles are not identified.

The described follow-up schedule includes a visit 30 days after the procedure, then a visit at 6 months, with visits continuing every 6 months until a maximum follow-up period of 48 months. The primary outcome timeframe begins 30 days after treatment and extends through that maximum follow-up. The brief summary also describes participants as having follow-up visits over time; it does not establish that every participant completed the maximum period.

The trial concerns pancreatitis and pancreas divisum, including acute and idiopathic pancreatitis in the registry’s condition listing. Oregon Health and Science University is identified as the sponsor. The posted results list outcome counts for EUS + Sham and EUS + ERCP with miES, with outcome denominators of 73 and 75, respectively. These arm-level outcome denominators differ from the enrollment figure, and the registry record does not explain how enrollment relates to the denominators used for this outcome. Randomization supports a comparative design, but interpretation still depends on the analysis population, follow-up, and handling of missing outcome information, details that are not reported in the registry record.

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

Reasons for not completing this period:

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 Reduce the Risk of Subsequent Acute Pancreatitis Episodes by 33%, measured over This is a time-to-event outcome that is assessed starting 30 days after treatment through a maximum follow-up of 48 months..

The primary endpoint is the occurrence of the next acute pancreatitis attack more than 30 days after treatment allocation. The registry defines it as a time-to-event outcome and states that assessment begins 30 days after treatment and continues through a maximum follow-up of 48 months. The posted group outcome values are participant counts: 32 in EUS + Sham and 26 in EUS + ERCP with miES, with arm denominators of 73 and 75, respectively.

A time-to-event endpoint captures both whether an event occurs and when it occurs. This can be useful when participants have different lengths of follow-up, because a simple event count does not show the timing or duration of observation. Time-to-event methods can account for incomplete follow-up through censoring, but the registry record does not report the specific analysis method, censoring rules, or whether the posted counts represent the full analysis population for the primary endpoint. It therefore is important not to treat the raw counts as a complete time-to-event analysis.

The title states a goal of reducing the risk of subsequent episodes by 33%. The observed outcome should be interpreted from the posted estimates and their uncertainty, rather than by treating that target as a demonstrated effect. The registry also lists an outcome comparison of incidence rates between groups as a secondary outcome. Details of how that secondary comparison was analyzed are not reported in the registry record.

7. Results Posted in the Registry

• EUS + ERCP with miES: 34.7% (26/75) versus 43.8% (32/73) on control; risk difference -9.2 percentage points (95% CI -24.8 to 6.5), risk ratio 0.79 (0.53–1.19); unadjusted two-proportion z-test p = 0.253.

8. Statistical Interpretation

9. Statistical Methods Explained

Statistical interpretation begins by separating the design, endpoint, and reported summary. The trial is recorded as randomized and parallel group. The primary outcome is specified as time to the next acute pancreatitis episode after the first 30 days. The posted numeric comparison, however, is a count-based comparison of participants with an event. These views answer overlapping but different questions: the count summary describes how many participants were recorded with an event, whereas time-to-event analysis also uses event timing and follow-up information.

What does the risk difference describe?

The risk difference compares the event proportions in the two arms by subtracting the comparator proportion from the intervention proportion. The posted estimate is -0.092, with the intervention compared with EUS + Sham. A negative value indicates a lower observed proportion in the intervention arm in this unadjusted summary. The interval from -0.248 to 0.065 describes uncertainty around that estimate. Because it spans zero, the data are compatible with no difference as well as with differences in either direction. The interval does not identify which value is most likely with certainty, and it does not establish a clinical threshold for importance.

Why are risk ratios and odds ratios also reported?

A risk ratio compares the event proportions as a ratio; a value of 1 represents equal proportions. An odds ratio compares odds rather than probabilities, also with 1 as its null value. The posted risk ratio is 0.791, with an interval from 0.527 to 1.186. The posted odds ratio is 0.680, with an interval from 0.350 to 1.32. Both intervals include 1. Different effect scales express the comparison differently, and the odds ratio should not be described as a risk ratio. Because these estimates are unadjusted recalculations from summary counts, they do not incorporate an adjusted model.

How should the p-value be read?

The unadjusted p-value is 0.253. In general, a p-value measures how compatible the observed data are with a specified null hypothesis under the assumptions of the statistical test. It is not the probability that the null hypothesis is true, nor does it measure the probability that results will reproduce. A value above a conventional threshold does not prove there is no effect. It can reflect uncertainty, including limited information or variability. The registry record does not report the test specification or a sponsor analysis, so the posted p-value should be treated as an unadjusted summary rather than as a fully documented primary time-to-event analysis.

Why does time to event matter?

For a time-to-event endpoint, analysis considers when the event occurs and how long each participant is observed. Participants who have not experienced the event by their last available assessment may be censored, meaning their event time is not known beyond that point. Appropriate methods can use the available follow-up without treating such participants as if they had the event or as if they were observed indefinitely. The particular method, censoring definition, event adjudication procedures, and assumptions are not reported in the registry record. The listed outcome counts alone do not reveal whether or how censoring was handled.

In general, a Kaplan–Meier curve can describe event-free probability over time, while a hazard ratio from a proportional hazards model compares instantaneous event rates under that model. These are general concepts, not descriptions of analyses confirmed in this registry record. No hazard ratio or time-to-event analysis result is posted in the registry record for the primary outcome. The secondary outcome mentions an incidence rate ratio, but its estimate and analysis are not reported in the registry record.

What do randomization and masking contribute?

Random assignment is designed to reduce systematic differences between groups at the start of a trial. It does not ensure that every measured or unmeasured characteristic is exactly balanced in a particular sample. Masking can reduce some influences on behavior, assessment, and reporting, depending on who is masked and how procedures are conducted. The record labels the trial randomized and double masked, but sequence generation, concealment, masked roles, and masking success are not reported in the registry record. These labels are useful design information, not a substitute for procedural detail.

The posted comparison is unadjusted and based on summary counts. It cannot establish whether adjustment was planned, whether an intention-to-treat population was used, or how protocol deviations and missing outcomes were handled. The registry record lists zero posted analyses. That absence means detailed inferential methods cannot be verified from the posted analysis information; it does not establish that the trial had no statistical plan or that no analysis was performed. The statistical plan and the trial’s original sample-size justification are not reported in the registry record.

10. Confidence Intervals and Precision

Precision concerns how narrowly or broadly the data constrain an effect estimate. The risk-difference estimate is -0.092, and its 95% confidence interval ranges from -0.248 to 0.065. The interval is broad enough to include zero and values on either side of it. This means the posted summary does not locate the underlying difference precisely. A confidence interval is not a range that contains a fixed effect with a 95% probability after the data have been observed. In general, it describes the behavior of an interval-producing method across repeated studies under its assumptions.

The ratio estimates are also imprecise. The risk ratio of 0.791 has a 95% confidence interval from 0.527 to 1.186, and the odds ratio of 0.680 has a 95% confidence interval from 0.350 to 1.32. Both intervals include 1, the value corresponding to no difference on those scales. These intervals allow effects that differ in magnitude and direction from the point estimates. A point estimate alone would give an incomplete account of the uncertainty.

Confidence intervals reflect sampling uncertainty under the analysis used; they do not automatically incorporate uncertainty from missing data, measurement, departures from assumptions, or choices made during analysis. In this case, the posted comparison is an unadjusted recalculation from summary counts, while the primary outcome is defined as time to event. The registry record does not report a confidence interval for a time-to-event analysis, nor does it describe censoring or the primary analysis method. Therefore, the posted intervals quantify uncertainty for the listed unadjusted comparisons, not necessarily every aspect of the intended primary endpoint.

Interval width is also not a direct statement about clinical importance. Determining whether a range of possible differences would matter requires a prespecified threshold and clinical context; neither is reported for interpreting the observed effect. The stated 33% reduction in the endpoint title is an aim, not a demonstrated threshold for interpreting the confidence interval. The intervals support a cautious reading: the estimate is not sufficiently precise to rule out a range of possible effects, including no difference.

11. 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 for 80% power and participants per arm for 90% power. They are planning illustrations based on fractions of the observed effect, not the original trial’s sample-size justification. The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set at 100%, 75%, or 50% of the observed effect. The listed outputs therefore describe hypothetical planning scenarios, not enrollment targets established by the trial team.

In general, power is the probability that a study will detect an effect of a specified size under the assumptions used, if that effect is present. The 80% power column means the illustrated sample size per arm for an 80% chance of detecting the chosen effect under those assumptions. The 90% column gives the corresponding planning size for a 90% chance. Higher planned power typically requires more participants when other assumptions are held fixed. The figures use the observed effect as their reference, so their interpretation depends on an estimate that itself has uncertainty.

The planning rows for 100%, 75%, and 50% of the observed effect illustrate that smaller effects require more participants to detect under the stated setup. For example, the listed participants per arm for 80% power are 445 for planning for 100% of the observed effect, 799 for 75%, and 1,813 for 50%. The corresponding participants per arm for 90% power are 594, 1,068, and 2,427. These are scenario outputs, not trial results or recommendations.

The sample-size assumptions for time-to-event tables would instead specify total events, but the planning information attached to this comparison is for participants per arm. The curve should be read as a visual expression of the same planning relationship: as the assumed effect becomes smaller, the required sample size rises. This is not observed power. The trial’s original design assumptions, including its own target effect and sample-size rationale, are not reported in the registry record.

12. 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 = 0.438; treatment proportion p₁ = 0.347; average p̄ = 0.393.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6906 = 1.3535.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6875 = 0.5786.
4. n = (1.3535 + 0.5786)² / (-0.0917)² = 444.1, rounded up to 445 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 <- 0.438356164383562                          # control proportion
p1 <- 0.346666666666667                          # 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(3609944)
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 = 445 (80% power), 594 (90% power)
 75% of observed effect: n per arm = 799 (80% power), 1,068 (90% power)
 50% of observed effect: n per arm = 1,813 (80% power), 2,427 (90% power)
power.prop.test: 444.1 
Simulated power at n = 445 per arm: 0.798 (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.

13. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when a study evaluates multiple outcomes, comparisons, time points, or analyses. Conducting many tests can increase the chance of at least one apparently notable result even when the underlying effects are absent. A prespecified strategy may identify a primary comparison, control error rates, or establish how secondary outcomes will be interpreted. The registry record identifies a primary endpoint and lists one secondary outcome: the incidence rate ratio of acute pancreatitis between treatment groups. It does not report a multiplicity adjustment or a hierarchy for interpreting secondary analyses. No secondary analysis estimate is posted in the registry record.

The record lists zero posted analyses, so details such as the exact statistical model, testing procedure, covariate adjustment, and analysis population cannot be confirmed. The trial’s statistical analysis plan and whether it included interim analyses are not reported in the registry record. Their absence from the posted record should not be taken as evidence that no such procedures existed; it means the reader cannot assess them from the available registry information.

Missing data can affect a trial when participants do not complete follow-up or when outcomes are unavailable. The registry describes follow-up visits through a maximum of 48 months, but it does not report the amount or pattern of missing outcome information, withdrawals, loss to follow-up, or imputation methods. For a time-to-event endpoint, censoring rules and the assumptions used to handle incomplete follow-up are especially relevant. Those details are not reported in the registry record.

The outcome denominators are 73 for EUS + Sham and 75 for EUS + ERCP with miES, while enrollment is recorded as 181. The registry record does not explain the relationship between enrollment and these outcome denominators. It also does not state whether the posted counts use an intention-to-treat population, a per-protocol population, or another analysis set. Without those details, it is not possible to determine how exclusions or missing measurements may affect the comparison.

These reporting gaps limit the ability to evaluate robustness and reproducibility. They do not invalidate the posted estimates, but they make clear that the estimates should be read as unadjusted count-based summaries rather than as a complete account of the planned primary analysis.

14. 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: pancreatitis acute (43.8%), abdominal pain (33.3%), pneumonia (4.1%), acute kidney injury (4.0%), clostridium difficile infection (2.7%).

The registry lists serious adverse event figures for each arm. EUS + Sham has 48 participants with a serious adverse event among 73 at risk, while EUS + ERCP with miES has 48 among 75 at risk. The record also lists deaths: 3 in EUS + Sham and 2 in EUS + ERCP with miES. These are descriptive counts. They do not, by themselves, establish that an event was caused by a study procedure or that a difference between groups is statistically meaningful.

The registry’s listed serious-event terms include Pancreatitis acute at 43.8%, Abdominal pain at 33.3%, Pneumonia at 4.1%, Acute kidney injury at 4.0%, Clostridium difficile infection at 2.7%, and Missing at 2.7%. The denominator and method used to calculate these percentages are not reported in the registry record. The terms should therefore be read as registry listings, not as arm-specific risks or a basis for comparing treatment groups.

General safety interpretation depends on definitions, event severity, timing, duration, relatedness assessments, and the completeness of reporting. Those details, including whether the deaths were considered related to a procedure or treatment, are not reported in the registry record. The serious-event counts are not the same as counts of all adverse events, and the record’s “other” adverse-event figures are separate from the serious-event figures. The posted information supports a description of reported counts only; it does not support causal conclusions about safety.

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

The posted count-based summary records 32 participants with an acute pancreatitis event in EUS + Sham and 26 in EUS + ERCP with miES. The unadjusted risk difference is -0.092, with a 95% confidence interval from -0.248 to 0.065 and an unadjusted p-value of 0.253. The interval includes zero, and the ratio confidence intervals include 1. The results therefore do not establish a clear difference in the posted summary, while also leaving a range of possible effects compatible with the data.

This does not mean the groups have been shown to be equivalent. A failure to meet a conventional significance threshold is not proof of no effect, and an interval that includes the null may still include differences that could matter. Equivalence or noninferiority requires prespecified margins and appropriate analyses; those are not reported in the registry record. Nor does the point estimate alone establish a treatment benefit. Its direction is only a feature of this unadjusted summary and should be interpreted alongside its uncertainty and the specified time-to-event endpoint.

The endpoint is the next acute pancreatitis episode more than 30 days after treatment allocation, followed through a maximum of 48 months. The posted results do not include a reported time-to-event estimate, such as a hazard ratio, or the details needed to understand censoring and follow-up. Accordingly, the count comparison cannot fully describe differences in when attacks occurred or how long participants were observed.

The trial’s stated aim of reducing subsequent episodes by 33% is not itself a finding. The registry record does not report the original sample-size justification or a formal statistical analysis plan, and it lists no posted analyses. The planning illustrations based on the observed effect are not evidence about the trial’s power or the truth of the treatment effect. They should be understood as hypothetical study-planning calculations only.

Safety counts are also descriptive. Similar serious adverse event counts are listed in both arms, along with deaths, but the record does not report event relatedness or enough detail to infer causation. Overall, the posted information permits a careful description of observed outcomes and uncertainty; it does not support medical advice or a definitive claim that the intervention does or does not work.

16. Why This Trial Matters Statistically

This trial raises several statistical issues that matter for interpreting procedural studies. First, the comparator is an active procedural context: both arms are described as receiving EUS, with sham in one arm and ERCP with miES in the other. This structure focuses the comparison on the added intervention pathway rather than on a procedure versus no procedure. The registry labels the trial randomized and double masked, but does not identify the masked roles or give details of allocation concealment. Those details are important when assessing how well the design limits bias.

Second, the primary outcome is time to a subsequent pancreatitis episode, while the posted result is a count-based summary. A time-to-event endpoint uses timing and follow-up, not just whether an event was recorded. If follow-up differs or is incomplete, analysis choices about censoring can influence interpretation. The lack of a posted time-to-event estimate and censoring details makes it necessary to distinguish the endpoint as specified from the count comparison available for interpretation.

Third, uncertainty is central. The risk-difference confidence interval spans zero, and both ratio intervals include their null value. The p-value is 0.253. These features do not prove no effect; they show that the posted unadjusted summary does not precisely establish the direction and size of a difference. Confidence intervals convey information that a binary significant/not-significant label would omit.

Finally, the sample-size illustrations show how planning depends on the target effect. Smaller effects call for larger planned samples under the stated assumptions. Because those illustrations use the observed effect as a reference and do not reproduce the trial’s own design rationale, they cannot be used as a retrospective judgment of trial adequacy or observed power. The original design assumptions are not reported in the registry record. These considerations make transparent reporting of the analysis plan, outcome follow-up, and missing-data handling particularly valuable for evaluating the trial’s results.

17. 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 leaves important details unavailable for a full assessment. Although the trial is labeled randomized and double masked, it does not describe sequence generation, allocation concealment, which roles were masked, or whether masking was maintained. The arm descriptions are abbreviated, and the procedure descriptions do not provide enough information to assess consistency of delivery across participants or sites.

The primary endpoint is defined as time to the next attack after the initial 30-day period, but the posted comparison is based on participant counts. No primary time-to-event estimate, analysis method, censoring rules, or event-adjudication process is reported in the registry record. The follow-up schedule allows visits through a maximum of 48 months, but actual follow-up duration and completeness are not reported. This limits interpretation of the count results for an endpoint that depends on event timing.

The record lists enrollment of 181 and outcome denominators of 73 and 75. It does not explain the relationship between these figures, identify the analysis population, or describe missing outcomes, exclusions, withdrawals, or protocol deviations. Baseline characteristics and adjusted analyses are not reported in the registry record. The posted comparisons are unadjusted recalculations from summary counts and may not reflect the method used in any full trial analysis.

The registry record lists no posted analyses and does not report the statistical analysis plan, multiplicity procedures, interim analyses, or the original sample-size justification. The safety information gives counts and selected event terms, but event relatedness, definitions, and percentage denominators are not reported. These limitations do not replace the posted findings; they define what can and cannot be assessed from the registry record.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: SpHincterotomy for Acute Recurrent Pancreatitis — NCT03609944.
• Registry record last updated 2025-12-30; status last verified 2025-12.
• Results first posted 2025-12-30.
• 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.

21. Record Summary

NCT03609944, titled “SpHincterotomy for Acute Recurrent Pancreatitis,” is recorded as a completed randomized, double-masked, parallel-group trial sponsored by Oregon Health and Science University. Enrollment is listed as 181. The study began on 2018-09-27, and primary completion is recorded as 2025-02-28.

The comparison is EUS + ERCP with miES versus EUS + Sham in people with pancreas divisum and recurrent pancreatitis. The primary outcome concerns the next acute pancreatitis episode more than 30 days after treatment, through a maximum follow-up of 48 months. Posted participant event counts are 26 in the intervention arm and 32 in the sham arm. The unadjusted risk difference is -0.092, with a 95% confidence interval from -0.248 to 0.065 and p = 0.253.

The registry also lists serious adverse event counts and planning illustrations based on the observed effect. Detailed primary time-to-event analyses, missing-data handling, the original sample-size rationale, and the statistical analysis plan are not reported in the registry record. The posted count comparison is therefore best read as an uncertain unadjusted summary, not as a complete account of the specified time-to-event analysis.