Damage Control for Duodenal and Combined Duodenal-Pancreatic Injuries: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Damage Control for Duodenal and Combined Duodenal-Pancreatic Injuries (NCT00937118), a registered phase not stated study. The primary outcome is Duodenal-related Complications.

2. Clinical Question

3. Trial Design

This is registered as a retrospective review, not a randomized trial. The official title describes a retrospective review of severe duodenal and combined duodenal-pancreatic injuries. According to the summary, investigators planned to use the OHSU Trauma Laparotomy Outcomes Database to identify patients who received trauma laparotomy for duodenal or duodenal-pancreatic injury during 1989-2009. Medical-record information was to include injury grade, mechanism, Injury Severity Score, and other data. The registry record does not report how those data were abstracted, checked, or handled when unavailable.

The study description says the review was intended to quantify a change in practice and report outcomes compared with previous practice. The listed group, however, is described as patients with full-thickness duodenal laceration who underwent laparotomy and survived more than 72 hours. Patients requiring pancreaticoduodenectomy were excluded. These criteria matter when interpreting the results: the listed group does not represent every patient described in the broader database search, and survival beyond 72 hours is part of its description. The record does not explain whether the same criteria were applied to any historical comparator.

Enrollment is listed as 43. The outcome and adverse-event fields separately show 41 at risk. The registry does not reconcile these figures or explain the relationship between enrollment and the analyzed population. It also does not state the allocation method, and no randomized assignment is described. The record marks the comparison as not a like-for-like control comparison.

The study is marked completed. Its listed start is 2009-07 and its primary completion is 2010-11. The registry record does not report the protocol's sample-size justification, prespecified statistical analysis plan, interim analyses, or detailed definitions of study groups. Consequently, the recorded design and posted outcome support a descriptive reading; they do not establish a causal estimate of the effect of changing management.

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

Analysis population for the primary outcome, as defined in the registry: Patients with duodenal related complications

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 Duodenal-related Complications, measured over 20 years.

The primary outcome is duodenal-related complications, expressed as a percentage of subjects. The description includes leak, obstruction, and abscess. These events are clinically relevant measures of complications associated with duodenal injury and its management. Counting whether participants experience a specified complication can give a direct description of the burden of these events in the population under study.

The registry assigns the outcome a 20-year timeframe. The record does not clarify whether that timeframe means the period over which patients were identified, the follow-up period for each patient, or another study interval. Nor does it state how each complication was defined, whether events were adjudicated, or whether the outcome included only the first event per participant. Those details affect how an event percentage should be interpreted, so they cannot be assumed from the outcome label alone.

The posted result lists zero events among 41 participants at risk. This is a descriptive result for the group represented in that outcome field. The registry does not provide a comparison estimate, confidence interval, or statistical test for this endpoint. Without details on follow-up completeness and event ascertainment, the zero-event entry should not be read as proof that complications were absent in all eligible patients or across every possible follow-up period.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The available methods information is limited. The registry describes a retrospective review using an existing trauma laparotomy database and medical records. A retrospective design examines care and outcomes already recorded rather than assigning participants prospectively to treatments. In general, this can be useful for describing practice over time, but it makes the quality and completeness of historical records important. The registry record does not report the abstraction procedures or the approach to missing medical-record information.

What does the outcome count represent?

The primary endpoint is a binary event measure: whether a participant had a duodenal-related complication, with leak, obstruction, and abscess named as examples. The registry describes the unit as a percentage of subjects, and the posted entry lists zero events among 41 participants at risk. The measure is intended to describe how often participants in the specified group experienced a qualifying event. It does not indicate the number of events per participant, the timing of complications, or their severity. Those details are not reported in the registry record.

In general, a binary outcome is informative only when the event definition, observation period, and analyzed population are clear. If ascertainment differs between records or across historical periods, the recorded event frequency may not be directly comparable. If follow-up is incomplete, events occurring outside documented observation may be missed. The registry does not report how follow-up was established or whether event assessment was consistent throughout the study period.

Why does the comparison structure matter?

A comparison of outcomes across management strategies can be difficult to interpret when groups differ in their underlying characteristics. Injury grade, mechanism, and Injury Severity Score are among the data points the investigators intended to retrieve, but the record does not provide their distributions or any adjusted analysis. In general, adjustment can address measured differences when a suitable model and adequate data are available. It cannot account for unmeasured differences, and the registry record does not report a model, covariate set, or adjusted effect estimate.

The record does not state allocation, and the group descriptions do not establish a randomized or contemporaneous controlled comparison. The registry's warning that the arms are not a like-for-like control comparison is therefore important. A change in practice across time may coincide with changes in patient selection, documentation, surgical care, or other aspects of trauma management. These are general possibilities in retrospective research; the record does not say whether they occurred in this study.

How should uncertainty be considered?

A confidence interval, in general, describes statistical uncertainty around an estimate under the assumptions of the analysis. It can help readers distinguish a precise estimate from one compatible with a broad range of values. The registry record does not report a confidence interval for the primary outcome, nor does it list an analysis for that outcome. The event count and number at risk alone do not provide the missing interval or a comparative test.

The accompanying planning illustrations use the observed effect as a basis for hypothetical sample-size calculations. The stated assumptions are a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation. The target effect is a stated fraction of the observed effect. For time-to-event calculations, the convention is to report total events; for other outcomes, the convention is participants per arm. These are planning assumptions, not evidence that the original study used those methods. The registry record does not report an original sample-size justification, and planning illustrations should not be described as observed power.

What can zero observed events establish?

In general, an observed zero count means no qualifying event is entered for the specified group and outcome in the posted data. It does not establish that the underlying event probability is exactly zero. The interpretation depends on the size and definition of the analyzed group, the completeness of outcome capture, and the observation period. Here, the record lists zero events and 41 at risk, but does not report an uncertainty interval or enough follow-up detail to judge how completely events could have been detected.

It is also important to distinguish the primary outcome from adverse-event reporting. These fields may use different definitions and reporting processes. The adverse-event entries list no affected participants in their serious and other event fields, but the record does not explain ascertainment or whether these fields overlap with the primary endpoint. They should not be treated as a substitute for a complete account of complications.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("No Diversion Decompression or Exclusion", "Diversion Decompression or Exclusion", "Damage Control", "Fascial Closure")
n      <- c(23, 18, 25, 16)
events <- c(3, 2, 2, 3)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  No Diversion Decompression or Exclusion  23  3  13.0  2.8  33.6  4.5  32.1  4.7  32.4
ARM  Diversion Decompression or Exclusion  18  2  11.1  1.4  34.7  3.1  32.8  3.4  33.1
ARM  Damage Control  25  2  8.0  1.0  26.0  2.2  25.0  2.4  25.1
ARM  Fascial Closure  16  3  18.8  4.0  45.6  6.6  43.0  6.8  43.4

11. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary outcome measure, duodenal-related complications, and zero statistical analyses. It does not report a multiplicity plan, a hierarchy of outcomes, or adjustments for testing multiple hypotheses. Because no inferential analyses are listed, the record does not support an assessment of how multiplicity was handled in statistical testing. In general, multiplicity matters when several outcomes or comparisons are tested, because the chance of a seemingly notable result can increase across repeated tests. That general concern does not establish that such testing occurred in this review.

The record does not report a missing-data plan, the amount or pattern of missing outcome information, or the method used to handle incomplete records. It also does not explain why enrollment is 43 while the primary outcome field lists 41 at risk. That difference should not be attributed to missingness or exclusions without documentation. The registry record does not state whether all participants with relevant records contributed to the outcome analysis.

The analysis population is not fully characterized. The broader description covers patients identified through trauma laparotomy records during 1989-2009; the listed management group requires full-thickness duodenal laceration, laparotomy, and survival beyond 72 hours, with pancreaticoduodenectomy excluded. The relationship between those criteria and the endpoint population is not reported in the registry record.

In general, a clearly stated analysis population helps readers understand who contributes to an estimate and how exclusions affect interpretation. Here, there is no reported participant flow, no account of exclusions by reason, and no description of how patients with incomplete follow-up were treated. Consequently, the posted count should be read according to its displayed at-risk group, without assuming that the group represents every patient screened or identified.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

The adverse-event fields list 41 participants at risk for serious events and 41 at risk for other events. The corresponding affected counts are zero in both fields. These entries describe what the registry reports in its adverse-event fields; they do not by themselves establish that no adverse events occurred in all patients or that harms were comprehensively captured.

The record does not report the definitions used for serious or other adverse events, how events were collected, the observation period for these fields, or whether reporting was complete. It also does not list serious-event terms. The primary endpoint separately names leak, obstruction, and abscess, but the registry does not explain how those complications relate to the adverse-event categories. The fields should therefore be read separately rather than assumed to cover the same events.

In general, an empty event list or a zero count is a statement about recorded data, not a guarantee that the event could not occur. Safety interpretation depends on the population observed, the duration and completeness of monitoring, and consistent event definitions. Since the registry does not describe these details, its zero affected counts are limited descriptive entries, not a comprehensive safety conclusion.

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

The posted result documents zero recorded duodenal-related complications among 41 participants at risk in the listed outcome field. This is a useful descriptive observation about the registry's reported group. The outcome description names leak, obstruction, and abscess, but the record does not provide details on event definitions, follow-up completeness, or the timing and severity of events.

The result does not establish that the risk of complication was zero. An observed count and its underlying event frequency are not the same thing, and the registry does not report an uncertainty interval. Without an interval or other inferential analysis, the statistical precision of the finding cannot be evaluated from the posted information.

Nor does the result establish that damage control or planned reoperation caused a particular outcome. The study is retrospective, allocation is not stated, and the record does not provide a clearly defined, comparable control group or a comparative estimate. The registry explicitly notes that the arms are not a like-for-like control comparison. Any causal interpretation would require information and analysis not reported in the registry record.

The enrollment figure is 43, whereas the primary outcome field lists 41 at risk. The record does not explain this discrepancy. The posted outcome should be described using its stated at-risk group, without inferring why the enrollment and outcome figures differ or treating them as interchangeable.

The study's historical scope and stated goal make it relevant to understanding changes in trauma practice, but the record does not show which practices changed, when they changed, or how patient characteristics varied across management groups. The results therefore do not quantify the size of a practice change or its comparative effect. They also do not settle the broader clinical controversy described in the study summary. The defensible conclusion is narrower: the registry records a zero-event entry for the specified primary outcome group, with substantial design and reporting details unavailable for judging comparative effects.

14. Why This Trial Matters Statistically

This record illustrates why the design and analysis context matters as much as an event count. A result of zero recorded complications can sound definitive if presented without its denominator, outcome definition, and observation context. The registry does give the at-risk count and names the complications of interest, but it does not report an uncertainty interval, follow-up completeness, or a detailed endpoint definition. Those omissions constrain how precisely the finding can be interpreted.

It also shows the importance of distinguishing a study's stated aim from the comparison that is actually described. The summary says the investigators intended to quantify a change in practice and report outcomes compared with previous practice. Yet the registry does not provide a clearly delineated historical comparison group or comparative estimate, and it cautions that the arms are not like-for-like controls. This distinction prevents a descriptive outcome from being mistaken for evidence of a treatment effect.

The population details also matter. The database search is described broadly, while the listed management group includes specific injury, procedure, survival, and exclusion criteria. The registry record does not explain how these descriptions connect to the outcome population or how the enrollment figure relates to the at-risk count. Clear participant flow and analysis-population definitions are central to interpreting retrospective reviews.

Finally, the record demonstrates why hypothetical sample-size planning should be kept separate from the original study design. Illustrations based on an observed effect and stated assumptions can support planning discussions, but they do not show that the study was designed prospectively to answer a comparison or that its result has a particular power. The registry does not report an original sample-size justification. Careful separation of descriptive data, causal claims, and planning calculations helps readers judge what the evidence can—and cannot—support.

15. Limitations

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

The registry record leaves important details unresolved. It describes a retrospective review based on a trauma database and medical records, but does not report how records were selected, how variables were abstracted, or how data quality was assessed. The summary names injury grade, mechanism, and Injury Severity Score among the information sought, but distributions and analysis results for these characteristics are not reported.

The outcome is not fully operationalized. Leak, obstruction, and abscess are listed, but the record does not define diagnostic criteria, explain whether outcomes were adjudicated, or state whether each participant was followed for a consistent period. Although the outcome timeframe is given as 20 years, the meaning of that timeframe for individual follow-up is not clarified. The registry also does not report the completeness of outcome capture.

There is no detailed account of comparison groups or of how prior practice was represented. Allocation is not stated, and the record warns that the arms are not like-for-like controls. The record does not report whether analyses adjusted for differences in injury severity, patient characteristics, or time period. It therefore cannot support a causal interpretation of differences between practices.

Participant accounting is also unclear: enrollment is listed as 43, while the outcome field lists 41 at risk. The reason is not reported in the registry record. The survival requirement and pancreaticoduodenectomy exclusion in the listed group may limit the population to which the result applies, but the record does not explain the rationale or its effect on the analysis.

Finally, no statistical analysis, confidence interval, missing-data plan, multiplicity approach, or original sample-size justification is reported. These gaps limit assessment of precision, robustness, and representativeness. They do not negate the posted descriptive count, but they restrict the conclusions that can be drawn from it.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Damage Control for Duodenal and Combined Duodenal-Pancreatic Injuries — NCT00937118.
• Registry record last updated 2013-08-12; status last verified 2013-05.
• Results first posted 2013-08-12.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

The registry record identifies a completed retrospective review titled “Damage Control for Duodenal and Combined Duodenal-Pancreatic Injuries,” sponsored by Oregon Health and Science University. Its listed start is 2009-07 and its primary completion is 2010-11. Enrollment is 43.

The summary describes a review of trauma laparotomy cases involving duodenal or duodenal-pancreatic injury during 1989-2009, using a trauma outcomes database and medical records. The listed management group includes patients with full-thickness duodenal laceration who underwent laparotomy and survived more than 72 hours; patients requiring pancreaticoduodenectomy were excluded.

The primary outcome is duodenal-related complications, including leak, obstruction, and abscess, expressed as a percentage of subjects over a 20-year timeframe. The registry lists zero events among 41 at risk. It does not report a comparative estimate or statistical analysis. The record also notes that allocation is not stated and that the arms are not a like-for-like control comparison.