Does Cefoxitin or Piperacillin-Tazobactam Prevent Postoperative Surgical Site Infections After Pancreatoduodenectomy?: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Does Cefoxitin or Piperacillin-Tazobactam Prevent Postoperative Surgical Site Infections After Pancreatoduodenectomy? (NCT03269994), a randomized phase 3 study. The primary outcome is Number of Participants Evaluated for the Effectiveness of Cefoxitin as a Surgical Antibiotic Prophylaxis..

2. Clinical Question

The clinical question is comparative: it concerns the relative results of two antibiotic prophylaxis strategies in the setting of pancreatoduodenectomy. The registry describes a randomized, open-label comparison and identifies surgical site infection as the primary clinical outcome. It does not report a separate primary endpoint result as a clear number of participants who experienced an infection. Instead, the posted outcome is titled as the number of participants evaluated for effectiveness, and its arm values are counts of participants.

That distinction matters for interpretation. A count of people evaluated is not itself an infection rate. Without clearly reported infection event counts, the posted values do not establish how many patients in either group had an infection or support a clinical comparison of infection risk. The displayed comparison values should therefore be read with caution rather than as evidence that the two prophylaxis strategies had the same infection outcome. Details such as patient eligibility criteria, specific antibiotic administration procedures, and any protocol-defined handling of infections are not reported in the registry record.

3. Trial Design

The registry describes a phase III, multicenter, open-label randomized controlled trial with a parallel-group design. The trial compares cefoxitin with piperacillin-tazobactam as surgical antibiotic prophylaxis for patients undergoing pancreatoduodenectomy. The study status is completed, and the registry lists enrollment as 967. The sponsor is Memorial Sloan Kettering Cancer Center.

Random allocation means assignment to the treatment groups is intended to occur by chance, which can help balance prognostic factors across groups. The registry record does not report the randomization sequence method, allocation concealment procedures, or whether randomization was stratified. Masking is listed as none, so the trial is open label; the record does not describe whether outcome assessment was independently blinded.

The primary outcome timeframe is 30 days. The primary outcome description defines overall surgical site infection as superficial incisional, deep incisional, or organ/space infection occurring within the first 30 days after the operation. It refers to the ACS NSQIP data collection operations manual for the definition. The registry reports 400.0 participants for the cefoxitin arm and 378.0 for the piperacillin-tazobactam arm in the outcome results. These are the reported outcome-group counts and should not be substituted for the overall enrollment figure.

Although the registry identifies the trial as phase III and randomized, it does not report the full protocol, detailed treatment procedures, eligibility criteria, or the rationale for the enrollment target. The primary completion date is listed as 2025-01-28. The record does not provide enough information to assess adherence to the intended allocation process or the completeness of follow-up for the primary endpoint.

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 Number of Participants Evaluated for the Effectiveness of Cefoxitin as a Surgical Antibiotic Prophylaxis., measured over 30 days.

The primary endpoint is described as overall postoperative surgical site infection within 30 days after pancreatoduodenectomy. It includes superficial incisional, deep incisional, and organ/space surgical site infections, with definitions linked to the ACS NSQIP data collection operations manual. Combining these infection categories into an overall endpoint can capture infections occurring at different anatomical depths and sites after the operation.

A binary infection endpoint is commonly expressed as the number or proportion of participants with at least one qualifying event during a defined follow-up period. This makes the outcome directly relevant to comparing event occurrence between randomized groups. The fixed 30-day timeframe also defines which postoperative events count toward the primary endpoint. These are general features of binary clinical outcomes; the registry does not provide additional endpoint adjudication or data-collection details.

There is an important difference between the endpoint definition and the posted outcome values. The primary outcome title says that it measures participants evaluated for effectiveness, and the arm values are 400.0 and 378.0 participants. Those values identify how many participants are represented in the outcome result, but they do not give the number who developed an infection. Consequently, the registry record does not clearly report the primary infection event counts or rates needed to interpret the intended clinical endpoint. The result should not be described as showing a particular number of infections in either arm.

7. Results Posted in the Registry

• Piperacillin-tazobactam: 100.0% (378/378) versus 100.0% (400/400) on control; risk difference 0.0 percentage points (95% CI 0.0 to 0.0), risk ratio 1.00 (1.00–1.00); unadjusted two-proportion z-test p = n/a.

8. Statistical Interpretation

9. Statistical Methods Explained

What does a risk difference represent?

In general, a risk difference compares the proportion of participants experiencing an event in one group with the proportion experiencing it in another. It is expressed on the same absolute scale as the event proportions. A value of zero means the two recorded proportions are equal, while a confidence interval describes a range of values compatible with the data and the method used to calculate it.

For this trial, the intended event is a surgical site infection within 30 days. However, the primary outcome values are reported as participants evaluated, not clearly as participants with infection. The comparison fields encode the arm values as 1.0 and report a risk difference of 0.0. This appears to reflect the recorded values and denominators rather than a transparent count of infections. As a result, the risk-difference interpretation is compromised: a mathematically defined comparison cannot resolve whether the input values represent the clinical event of interest.

How should the p-value be read?

In general, a p-value measures how unusual data at least as incompatible with a specified null model would be if that model and the analysis assumptions were correct. It is not the probability that the null hypothesis is true, and it does not measure clinical importance. The posted p-value is 1.0. Because the primary outcome values do not clearly report infections, this p-value cannot establish that infection risks are the same or that the treatment strategies are equivalent. The registry record does not report the statistical test or a detailed analysis plan that would clarify how the test was selected or applied.

What does randomization contribute?

As a general design principle, random assignment helps make comparison groups similar on average with respect to both measured and unmeasured factors. This supports causal interpretation when allocation is implemented as intended and when outcomes are measured and analyzed appropriately. Here, the registry identifies randomization and a parallel-group design, but it does not report the sequence generation or allocation concealment procedures. The trial is open label, and details about blinded outcome assessment are not reported in the registry record.

What can the posted sample counts tell us?

The outcome record gives 400.0 participants in the cefoxitin arm and 378.0 in the piperacillin-tazobactam arm. These counts indicate the participants represented in the reported outcome result. They do not tell how many experienced infection. A denominator is necessary to express an event rate, but a denominator without a clear event numerator cannot answer the primary clinical question. The record does not explain why these outcome counts differ from the listed enrollment figure or describe the analysis population in sufficient detail.

What about a binary endpoint?

In general, a binary endpoint records whether each participant did or did not experience a defined event during follow-up. Analyses may compare event proportions, estimate a risk difference or relative effect, and quantify uncertainty. The meaning of these estimates depends on having correctly identified events, denominators, and an analysis population. A count of participants evaluated is not interchangeable with a count of participants who experienced the endpoint.

What analysis details are missing?

The registry lists no analysis records and no posted secondary outcome results. The detailed statistical plan, the approach to missing data, any sensitivity analyses, and any interim analyses are not reported in the registry record. The record also does not state whether the primary analysis was adjusted, whether any covariates were included, or how deviations from assigned treatment were handled. These omissions prevent a fuller assessment of the analysis beyond the limited posted comparison.

General statistical interpretation depends on the alignment between the endpoint definition, the recorded data, and the analysis. In this case, the endpoint is clinically specified, but the numerical outcome is labelled as participants evaluated. The displayed risk difference and p-value should therefore not be treated as a reliable answer about comparative infection risk. A clearly reported event count by treatment group and a description of the analysis population would be needed to interpret the primary endpoint statistically.

10. Confidence Intervals and Precision

A confidence interval describes uncertainty around an estimated effect under the chosen statistical method. In general, a narrower interval indicates greater statistical precision, while a wider interval indicates that a broader range of effect values remains compatible with the data. A confidence interval is not a guarantee that the true effect lies within its limits, and its interpretation depends on the validity of the underlying data and assumptions.

The posted risk-difference interval runs from 0.0 to 0.0. Although that interval has no displayed width, it should not be read as unusually precise evidence about the difference in surgical site infection risk. The underlying outcome values are counts of participants evaluated, and the record does not clearly report infection events. The interval therefore does not communicate meaningful uncertainty about the clinical endpoint as defined.

Precision is also not the same as certainty about clinical importance. Even a well-calculated interval would need to be considered alongside the size and direction of the estimated effect and the clinical question. Here, the primary comparison is not sufficiently interpretable to assess either the magnitude of any infection-risk difference or how precisely it was estimated.

The registry does not report a confidence interval for the number of infections, because infection counts are not clearly posted. Nor does it provide details of the confidence-interval method, adjustment strategy, or analysis population. The reported interval should be understood as a property of the posted numerical comparison, not as a dependable measure of precision for the intended infection endpoint.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists one outcome measure and no posted secondary outcome results. It lists no analysis records. These entries provide limited information about the results available in the registry, but they do not establish the full set of outcomes or analyses prespecified in the study protocol. The protocol and any statistical analysis plan are not reported in the registry record.

Multiplicity arises when investigators conduct multiple statistical tests or examine multiple outcomes, time points, or subgroups. In general, testing many hypotheses can increase the chance of at least one apparently noteworthy result by chance. The registry does not report whether multiple comparisons were planned, whether any adjustment for multiplicity was used, or whether interim analyses were conducted. It is therefore not possible to assess those features from this record.

Missing-data methods are also not reported in the registry record. The outcome counts are 400.0 for cefoxitin and 378.0 for piperacillin-tazobactam, while overall enrollment is listed as 967. The record does not explain the relationship between these figures, specify the number with complete 30-day outcome ascertainment, or describe how participants with unavailable outcome data were handled. No assumptions about missing outcomes should be made from the posted counts alone.

The analysis population is not reported in the registry record. It is not stated whether the comparison includes all randomized participants, only participants with available follow-up, or another defined group. The record also does not describe treatment crossovers, protocol deviations, or how these might have been handled. These omissions matter because the population analyzed and the treatment of missing data can affect both the estimated effect and its uncertainty.

12. 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

Serious events with the highest arm-level rate: death (2.5%).

The registry reports serious adverse event figures separately for the two groups. For cefoxitin, it records 10 participants affected by serious adverse events among 400 at risk, and 10 deaths. For piperacillin-tazobactam, it records 5 participants affected by serious adverse events among 378 at risk, and 5 deaths. The record also lists zero participants with other adverse events in each group.

These are registry-reported counts, not a causal comparison. The record does not provide descriptions of the serious adverse events beyond listing Death as a top serious term with 2.5%. It does not establish whether each death was caused by the study treatment, the operation, the underlying illness, or another circumstance. The matching serious-event and death counts do not establish that every serious adverse event was a death for the same individuals; participant-level overlap is not reported.

Safety counts should be interpreted in light of event definitions, observation periods, ascertainment procedures, and the population at risk. Those details are not fully reported in the registry record. The figures describe what was recorded in the two groups and should not be treated as proof of a treatment-related difference or as a complete account of all safety outcomes.

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

The registry record describes a randomized comparison of cefoxitin and piperacillin-tazobactam as prophylaxis for patients undergoing pancreatoduodenectomy, with a primary endpoint defined as surgical site infection within 30 days. That framing identifies the clinical question and the intended outcome. However, the posted primary outcome values count participants evaluated rather than clearly reporting participants who had an infection.

The displayed risk difference, confidence interval, and p-value therefore do not provide a clear answer about comparative infection risk. In particular, the risk difference of 0.0 and p = 1.0 cannot be taken as evidence that the infection outcomes were identical, that either treatment is superior, or that the treatments are equivalent. Equivalence requires a prespecified margin and an analysis designed to assess it; such information is not reported in the registry record.

The record can be used to identify the trial design, the two prophylaxis groups, the intended 30-day outcome, the outcome-group counts, and the posted serious adverse event and death counts. It cannot establish from the primary outcome values how many participants in either group developed a surgical site infection. Nor does it provide enough detail to assess the missing-data approach, analysis population, or statistical plan.

These limits concern interpretation of the registry results, not a conclusion about the underlying trial or the treatments. A clear report of the infection event counts, denominators, analysis population, and prespecified analysis would be needed to evaluate the clinical comparison. The registry record alone does not support a conclusion about which prophylaxis strategy resulted in fewer surgical site infections.

14. Why This Trial Matters Statistically

This trial addresses a focused comparative question in a surgical setting: whether two antibiotic prophylaxis strategies differ in postoperative surgical site infection after pancreatoduodenectomy. Randomized comparisons can be statistically informative because, when implemented and analyzed appropriately, they help separate treatment effects from differences in patient characteristics. The registry identifies a phase III, multicenter, open-label randomized design and defines a 30-day infection endpoint.

The record also illustrates why statistical interpretation depends on data labels as well as numerical calculations. The stated clinical endpoint is infection, but the posted primary outcome is titled as participants evaluated, and the arm values are counts of evaluated participants. A comparison expressed as a risk difference cannot be assumed to describe infection risk unless the numerator is actually the number of infections. Numerical estimates can look complete while remaining ambiguous about the event they represent.

From a biostatistical perspective, clear reporting should connect the endpoint definition to the data analyzed: event counts, denominators, analysis population, missing-data handling, and the method used to estimate effects and uncertainty. When those elements are absent or unclear, readers cannot determine whether a calculated comparison answers the intended question, even when the registry displays a point estimate and p-value.

The trial's registry record is therefore useful for understanding the study question and design, while also demonstrating the limits of interpreting posted summaries without transparent event-level reporting. That distinction is important for clinical researchers who assess evidence across randomized trials and for statisticians responsible for communicating results in a way that preserves the meaning of the outcome.

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 primary limitation for interpreting this trial's posted results is that the outcome values are counts of participants evaluated, not clearly reported surgical site infection events. Although the registry defines the intended endpoint in detail, it does not clearly report how many participants experienced an infection in either group. This prevents a direct interpretation of the displayed risk difference as a treatment comparison for infection.

The registry record does not explain why the outcome-group counts differ from the overall enrollment figure, or specify the analysis population represented by those counts. It also does not report whether all participants had complete 30-day follow-up or how missing endpoint data were handled. These issues leave uncertainty about the denominators and the population to which the comparison applies.

The trial is open label, and the record does not describe whether outcome assessment was blinded. It does not report the randomization method, allocation concealment, treatment adherence, crossover, or protocol deviations. The protocol, statistical analysis plan, and details of any prespecified multiplicity adjustment or interim monitoring are not reported in the registry record.

The posted serious adverse event and death counts have limited accompanying detail. Event definitions, timing, participant-level overlap, and causal attribution are not reported. The record also provides no clear primary endpoint analysis beyond the posted comparison, and no secondary outcome results. These limitations restrict assessment of both efficacy and safety from the registry record; they do not establish what the complete trial report may contain.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Does Cefoxitin or Piperacillin-Tazobactam Prevent Postoperative Surgical Site Infections After Pancreatoduodenectomy? — NCT03269994.
• Registry record last updated 2025-10-21; status last verified 2025-01.
• Results first posted 2025-10-21.
• 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

NCT03269994 is a completed phase III, multicenter, open-label randomized controlled trial sponsored by Memorial Sloan Kettering Cancer Center. It compares cefoxitin with piperacillin-tazobactam as surgical antibiotic prophylaxis for patients undergoing pancreatoduodenectomy. The registry lists enrollment as 967 and describes a parallel-group design.

The primary outcome is defined as overall surgical site infection within 30 days, including superficial incisional, deep incisional, and organ/space infection. The posted outcome values are 400.0 participants evaluated for cefoxitin and 378.0 for piperacillin-tazobactam. Infection event counts are not clearly reported, so the posted risk difference and p-value do not provide an interpretable comparison of infection risk.

The registry lists serious adverse event counts of 10 and 5, and death counts of 10 and 5, respectively. Details needed to interpret causality, the analysis population, missing data, and the statistical plan are not reported in the registry record.