This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for EUS-CNB Versus EUS-SINK for Diagnosis of Upper Gastrointestinal (UGI) Subepithelial Tumors (NCT02282111), a randomized na study. The primary outcome is Diagnostic Accuracy.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT02282111 |
| Phase | NA |
| Condition | Esophageal Subepithelial Lesions |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Single |
| Enrollment | 56 |
| Arms in the primary outcome | 2 |
| Lead sponsor | Johns Hopkins University |
| Results posted | Yes |
2. Clinical Question
Population
Patients with upper gastrointestinal subepithelial lesions were the intended population. The registry lists esophageal subepithelial lesions as the condition. Further eligibility details and the characteristics of enrolled participants are not reported in the registry record.
Intervention
EUS-guided core needle biopsy (EUS-CNB). The registry describes sampling with a 22-gauge beveled needle using slow capillary suction and fanning, with 4 passes.
Comparator
EUS-guided single-incision with needle knife (SINK). The description refers to making a 6-12mm incision under direct endoscopic vision and introducing biopsy forceps through the opening.
Primary question
How accurately did each method classify biopsy specimens as true positive or true negative by 30 days? The primary outcome was diagnostic accuracy, expressed as the percentage of all samples that were true positive or true negative.
The clinical question concerns the diagnostic yield of two tissue-sampling approaches, rather than a therapeutic outcome. Histological diagnosis is relevant to the trial’s stated rationale because subepithelial lesions may represent different lesion types and diagnosis depends on tissue. The registry describes the study as a multicenter randomized controlled trial, but does not report further detail about the diagnostic reference standard, adjudication process, or which personnel were blinded.
3. Trial Design
NCT02282111 is listed as a randomized, single-blinded, parallel-group trial. The registry title identifies it as multicenter. The listed enrollment is 56. The record lists a start of 2014-07, primary completion of 2017-11, and trial status as terminated. The reason for termination is not reported in the registry record.
Participants were assigned to EUS-CNB or SINK. The EUS-CNB description specifies a 22-gauge beveled needle and 4 sampling passes, using slow capillary suction and fanning. The SINK description specifies a needle-knife incision from the lesion periphery toward its highest convexity zone, followed by deep introduction of biopsy forceps. The registry’s intervention descriptions end mid-sentence, so further procedural details are not reported in the registry record.
The listed primary outcome is diagnostic accuracy, assessed at 30 days. Secondary outcomes are histological yield, technical failure rate, procedure time, and percent sample contribution to immunohistochemistry diagnosis. The registry lists five outcome measures, but only four secondary outcomes are named. The fifth measure is not identified in the registry record.
Primary outcome results are reported for 19 EUS-CNB participants and 22 SINK participants. The safety counts use different denominators: 26 for EUS-CNB and 30 for SINK. Why these analysis counts differ from one another and from the listed enrollment is not reported in the registry record. Details of the randomization sequence, allocation concealment, trial protocol, and planned analysis are also 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.
Before assignment: Patients were excluded from the study because a predominantly cystic lesion was found; the lack of an appreciable mass at Endoscopic Ultrasound scan (EUS).
Overall Study
| Stage | EUS-CNB | EUS-SINK |
|---|---|---|
| STARTED | 26 | 30 |
| COMPLETED | 19 | 22 |
| NOT COMPLETED | 7 | 8 |
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.
| Characteristic | Unit | EUS-CNB | EUS-SINK | Total |
|---|---|---|---|---|
| Participants analysed | 26 | 30 | 56 | |
| Age, Continuous | years | 68.7 (SD 11.9) | 66.2 (SD 13.4) | 67.4 (SD 12.7) |
| Sex: Female, Male: Female | Participants | 11 | 14 | 25 |
| Sex: Female, Male: Male | Participants | 15 | 16 | 31 |
| Race and Ethnicity Not Collected | Participants | — | — | 0 |
| Layer of origin: Submucosa | participants | 7 | 9 | 16 |
| Layer of origin: Muscularis propria | participants | 17 | 17 | 34 |
| Layer of origin: Indeterminate | participants | 2 | 4 | 6 |
| Echogenicity: Hypoechoic | participants | 19 | 22 | 41 |
| Echogenicity: Isoechoic | participants | 5 | 7 | 12 |
| Echogenicity: Hyperechoic | participants | 1 | 1 | 2 |
| Echogenicity: Mixed echogenicity | participants | 1 | 0 | 1 |
| Location of the lesion: Esophagus | participants | 1 | 1 | 2 |
| Location of the lesion: Gastric fundus | participants | 7 | 9 | 16 |
| Location of the lesion: Gastric body | participants | 12 | 12 | 24 |
| Location of the lesion: Antrum | participants | 4 | 7 | 11 |
| Location of the lesion: Doudenum | participants | 2 | 1 | 3 |
| Calcification: Yes | participants | 0 | 4 | 4 |
| Calcification: No | participants | 26 | 26 | 52 |
| Greatest diameter of the lesion | millimeters | 23.5 (20 to 39) | 25 (20 to 30) | 24.5 (20 to 33) |
6. Primary Endpoint
The registry lists the primary outcome as Diagnostic Accuracy, measured over 30 days.
| Element | Registry entry |
|---|---|
| Result type | Number |
| Measure of spread | 95% Confidence Interval |
| Unit | percentage of all samples |
| Endpoint classification | not reduced to a single comparison |
Diagnostic accuracy is defined in the registry as the percentage of true positive and true negative biopsy specimens combined, divided by the total number of specimens. It is reported at 30 days, with a 95% confidence interval. This endpoint summarizes whether samples correctly classify lesions as positive or negative according to the study’s diagnostic reference.
For a tissue-sampling comparison, accuracy addresses whether the specimen yields a correct classification, rather than simply whether tissue was obtained. It is therefore distinct from histological yield, which is listed separately as a secondary outcome. A sample may contribute tissue without necessarily establishing the correct diagnosis; conversely, accuracy depends on how correctness is determined. The registry does not report the reference standard or the details used to verify true positive and true negative classifications.
The posted primary results are 94.7% for EUS-CNB, with a 95% confidence interval from 74.2% to 99.0%, among 19 participants; and 95.5% for SINK, with a 95% confidence interval from 77.3% to 99.2%, among 22 participants. The estimates are high, but their confidence intervals are broad. Those intervals express uncertainty around each arm’s estimate; they do not by themselves establish whether the methods differ. A formal comparative estimate or test for the primary endpoint is not reported in the registry record.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The posted primary outcome estimates are 94.7% for EUS-CNB and 95.5% for SINK. The corresponding 95% confidence intervals are 74.2% to 99.0% and 77.3% to 99.2%. These are arm-specific estimates based on 19 and 22 participants, respectively. The point estimates are close, but the intervals span a substantially wider range of plausible accuracy values. The small analyzed groups therefore leave considerable uncertainty about the underlying accuracy in the populations represented by the trial.
The registry does not report a between-group estimate, confidence interval, or statistical test for the primary outcome. The two arm-specific confidence intervals should not be treated as a substitute for a direct comparison. In particular, close point estimates do not establish equivalence, and the record does not report an equivalence or noninferiority design or margin. Nor does the absence of a reported comparative test show that the methods are identical.
The primary outcome denominators differ from the listed enrollment of 56, and the arms’ safety denominators differ from their primary outcome denominators. The reasons for these differences and the analysis populations used are not reported in the registry record. Without information about participant flow, missing outcome data, or exclusions, the effect of incomplete outcome ascertainment cannot be assessed.
The secondary outcome of histological yield has a simple unadjusted recalculated risk difference of 0.097 for SINK relative to EUS-CNB, with a 95% confidence interval from -0.105 to 0.3 and an unadjusted p-value of 0.342. This interval includes both no difference and differences in either direction. It is a secondary result, and the registry does not list reported analyses. It should not be interpreted as a definitive treatment comparison or as evidence of equivalence.
Overall, the registry posts high arm-level primary accuracy estimates but does not provide the direct primary comparison needed to quantify evidence for a difference between sampling methods. The results are also limited by the small outcome denominators and incomplete reporting of analysis methods, participant flow, and missingness. The registry status is terminated; the reason is not reported in the registry record.
9. Statistical Methods Explained
Diagnostic accuracy is a proportion: the number of specimens classified as true positive or true negative, divided by the total number of specimens. The registry reports each arm’s accuracy with a 95% confidence interval. In general, a confidence interval describes uncertainty in an estimate under the assumptions of the method used to construct it. It is not a range that guarantees where the true value lies, and it does not directly answer whether two groups differ. The registry does not state the interval-construction method.
What does a between-group estimate add?
A comparative analysis estimates the contrast between randomized groups directly. For binary outcomes, a risk difference expresses the difference between the proportions with the outcome in the two groups. A confidence interval around that contrast communicates both its direction and precision. For the primary diagnostic accuracy outcome, the registry posts arm-level estimates but no comparative estimate or test. Consequently, the reader cannot use the registry record to assess the precision of a direct primary-outcome contrast.
A simple unadjusted recalculation is posted for histological yield. The risk difference for SINK relative to EUS-CNB is 0.097, with a 95% confidence interval from -0.105 to 0.3 and an unadjusted p-value of 0.342. This calculation compares the observed proportions without adjustment for other factors. Its interval includes the null value of no difference and also allows effects in either direction. It is not the primary diagnostic accuracy result, and it is not evidence that the groups are equivalent.
How should the p-value be read?
In general, a p-value describes how compatible data are with a specified null hypothesis under a statistical model. It is not the probability that the null hypothesis is true, nor does it measure the clinical importance of an effect. The histological-yield p-value is unadjusted, so it does not account for testing multiple outcomes. The registry does not report a prespecified statistical analysis plan, the test used for the primary outcome, or reported analyses.
What do the planning illustrations mean?
The registry includes sample-size planning illustrations based on the observed histological-yield effect. The stated assumptions are a two-sided significance level of 0.05, equal allocation, and a normal approximation. The illustrations give participants per arm for 80% and 90% power under scenarios planning for 100%, 75%, and 50% of the observed effect. For 100% of the observed effect, the figures are 246 and 328 participants per arm; for 75%, they are 459 and 614; for 50%, they are 1,083 and 1,449.
These are planning illustrations, not the trial’s original sample-size justification and not observed power. The registry does not report the trial’s sample-size assumptions or justification. In general, basing a future planning scenario on an observed effect is uncertain because the observed estimate can be imprecise, particularly when the analyzed groups are small. Planning for a smaller effect requires a larger sample under otherwise comparable assumptions. The registry does not report whether these illustrations reflect the original protocol.
What can be said about multiplicity and over-dispersion?
Multiplicity arises when several outcomes or comparisons are examined: the chance of at least one apparently notable result can increase unless the analysis accounts for the testing plan. The registry lists several secondary outcomes, but does not report a multiplicity strategy. The histological-yield p-value is unadjusted. No count model or over-dispersion estimate is reported. In general, over-dispersion matters when count data vary more than the assumed statistical model allows, because standard errors and resulting intervals may then be too small. That general concern cannot be evaluated for this trial from the posted results.
The record gives no details about the primary test, covariate adjustment, analysis population, missing-data methods, or sensitivity analyses. These omissions limit assessment of how the reported estimates were generated and whether the analysis followed a prespecified plan.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists several secondary outcomes and reports five outcome measures overall. Four secondary outcomes are named: histological yield, technical failure rate, procedure time, and percent sample contribution to immunohistochemistry diagnosis. The additional outcome is not identified in the registry record. A prespecified hierarchy among outcomes, adjustment for multiple comparisons, or other multiplicity strategy is not reported in the registry record. The posted histological-yield p-value is unadjusted and should be interpreted with that limitation in mind.
Missing-data methods and the analysis population are not reported in the registry record. The listed enrollment is 56, while the primary outcome results use 19 EUS-CNB participants and 22 SINK participants. Safety counts use 26 and 30 participants, respectively. The record does not explain how these denominators were formed, how many participants lacked each outcome, whether exclusions occurred, or whether the analyses followed intention-to-treat or another approach.
In general, missing outcomes can affect the interpretation of a randomized comparison when the probability of having an observed result is related to treatment assignment or outcome status. Without participant-flow information and a description of missingness, it is not possible to judge whether the analyzed groups remain comparable or whether the estimates may be affected by selective availability of results. No imputation approach, sensitivity analysis, or missing-data assumption is reported in the registry record.
The trial is listed as terminated, but the reason and any implications for planned enrollment or analyses are not reported in the registry record. The record also does not report interim analyses, stopping criteria, or whether the posted analyses were planned before results were examined. These gaps prevent a full assessment of multiplicity, selective reporting, and the analysis populations.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| EUS-CNB | 0 | 26 | 0.0% |
| SINK | 0 | 30 | 0.0% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| EUS-CNB | 0 of 26 |
| SINK | 0 of 30 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| EUS-CNB | 2 of 26 |
| SINK | 3 of 30 |
Most frequent non-serious adverse event terms
| Event term | Organ system | EUS-CNB | SINK |
|---|---|---|---|
| Delayed bleeding | Surgical and medical procedures | 2 of 26 | 2 of 30 |
| Abdominal pain | Gastrointestinal disorders | 0 of 26 | 1 of 30 |
The posted serious adverse event counts are 0 among 26 EUS-CNB participants at risk and 0 among 30 SINK participants at risk. Deaths are also reported as 0 in each group. The registry lists no serious adverse event terms. These counts describe what was recorded in the listed groups; they do not establish that no serious events could occur.
Other adverse events are reported as 2 among 26 EUS-CNB participants at risk and 3 among 30 SINK participants at risk. The registry separates these counts from serious adverse events, but further definitions, event descriptions, severity details, and attribution to the procedure are not reported in the registry record. No percentages should be inferred from the counts without a stated analysis method.
The adverse-event denominators differ from the primary-outcome denominators, and the record does not explain the difference. With these group sizes and limited event detail, the figures offer only a descriptive account of recorded events. They do not establish comparative safety or rule out uncommon harms. The trial’s terminated status does not, by itself, explain the safety findings; the reason for termination is not reported in the registry record.
12. What the Result Does, and Does Not, Mean
The registry reports high arm-specific diagnostic accuracy estimates for both sampling methods: 94.7% for EUS-CNB and 95.5% for SINK. The 95% confidence intervals are broad, and no direct comparative analysis for this primary endpoint is reported. The posted results therefore describe the observed estimates but do not quantify a precise difference between methods.
These results do not establish that the methods are equivalent. Equivalence requires a defined margin and an analysis designed to assess whether the difference lies within that margin; neither is reported in the registry record. Nor do the results establish superiority of either approach. The arm estimates alone are not a substitute for a comparative confidence interval or test.
The results also do not show how accuracy would generalize beyond the population represented in the trial. The registry identifies esophageal subepithelial lesions as a condition, while the title refers to upper gastrointestinal subepithelial lesions. Further eligibility criteria and participant characteristics are not reported in the registry record. The reference standard and the procedures for determining correct diagnoses are also not described.
Histological yield and adverse-event counts provide additional descriptive information, but the available comparisons are limited and secondary analyses are unadjusted. The record does not report analyses for several named secondary outcomes, missing-data methods, or the analysis populations. Accordingly, the results should be interpreted as limited registry evidence about diagnostic sampling, not as a complete assessment of diagnostic performance or safety.
The trial status is terminated, but no reason is stated. That status does not identify why the trial ended or establish that the posted estimates were affected. The registry record leaves those questions unresolved.
13. Why This Trial Matters Statistically
This trial illustrates why a high point estimate is not enough to judge a diagnostic comparison. The primary accuracy estimates are close, but the confidence intervals are broad and the analyzed groups are small. Precision matters: an estimate can appear favorable while still allowing substantial uncertainty about the value that would be observed in a broader population.
It also demonstrates the distinction between arm-level estimates and a direct treatment contrast. Randomization is intended to support comparisons between assigned groups, but the registry posts no comparative result for the primary endpoint. Without a contrast and its uncertainty, the primary result cannot establish how much diagnostic accuracy differs between the methods.
The record further shows why outcome definitions and analysis populations are essential. Diagnostic accuracy depends on how true positive and true negative classifications are established, yet the reference standard is not described. The primary outcome and safety denominators differ, and the reasons are not stated. Such details are necessary for readers to understand which participants contributed to each estimate and whether outcomes were missing.
Finally, secondary outcomes and planning illustrations need careful interpretation. An unadjusted secondary comparison can inform a question, but multiple outcomes raise multiplicity concerns, and an observed-effect planning scenario is not a measure of the trial’s observed power. The registry does not report the original sample-size justification or multiplicity plan. These omissions make the trial useful as an example of how results and reporting limitations jointly shape statistical interpretation.
14. Limitations
- The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
- Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.
Several limitations specific to this record constrain interpretation. Although the registry lists a randomized, single-blinded, parallel-group multicenter trial, details of sequence generation, allocation concealment, and who was blinded are not reported. The incomplete procedural descriptions also leave aspects of how each sampling method was carried out unclear.
The primary outcome lacks a posted direct comparison, and the registry does not state the statistical method used to construct its confidence intervals. The diagnostic reference standard and the process for deciding whether a specimen was true positive or true negative are not reported. Those details are particularly important when interpreting a diagnostic-accuracy endpoint.
Outcome denominators vary across the primary results and safety counts, with no participant-flow explanation. Missing-data handling and analysis populations are not reported. The status is terminated, but the reason is not reported in the registry record. The record also does not report a protocol, statistical analysis plan, sample-size justification, multiplicity approach, or interim analysis information.
The secondary results are incomplete relative to the named outcomes, and the record lists five outcome measures while identifying four secondary outcomes. The posted histological-yield comparison is unadjusted and does not replace the missing primary comparison. Together, these limits mean that the estimates are best understood as a partial account of trial results rather than a full statistical report.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: EUS-CNB Versus EUS-SINK for Diagnosis of Upper Gastrointestinal (UGI) Subepithelial Tumors — NCT02282111.
- Registry record last updated 2018-10-09; status last verified 2018-10.
- Results first posted 2018-10-09.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
NCT02282111 compared EUS-guided core needle biopsy with EUS-guided single-incision needle-knife sampling for upper gastrointestinal subepithelial lesions. The registry describes a randomized, single-blinded, parallel-group multicenter trial, with enrollment of 56 and status listed as terminated.
The primary outcome was diagnostic accuracy at 30 days. Posted arm-level estimates were 94.7% for EUS-CNB and 95.5% for SINK, with broad 95% confidence intervals. A comparative primary analysis is not reported in the registry record. Histological yield is reported as a secondary outcome, alongside limited adverse-event counts.
The record does not report why the trial was terminated, how differing outcome denominators arose, or key analysis details including missing-data methods and the primary statistical plan. These gaps limit the conclusions that can be drawn from the posted results.