Negative Pressure Wound Therapy in Cesarean Section: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Negative Pressure Wound Therapy in Cesarean Section (NCT02289157), a randomized na study. The primary outcome is Number of Participants With Wound Complication.

2. Clinical Question

3. Trial Design

This was a randomized, parallel-group trial with no masking listed in the registry. The trial record is marked completed and reports enrollment of 441 participants. The start date is 2015-01-27, and the primary completion date is 2016-08-30. The registry lists the University of Texas Southwestern Medical Center as sponsor.

Participants received standard cesarean section and were assigned to one of two post-operative dressing approaches. The negative pressure wound therapy group had the Prevena system placed, while the comparison group had a standard dressing placed. The registry describes randomization but does not report the method used to generate or conceal the allocation sequence. It also does not report any masking; the masking field is listed as none.

The primary outcome was a count of participants with a wound complication, assessed from the day of surgery through 30 days postoperatively. The registry names hematoma, seroma, dehiscence, and surgical site infection among the complications. It also lists secondary outcomes covering postoperative stay, readmission stay, emergency room and clinic visits, maternal outcomes, cesarean scheduling, incision type, and membrane status.

The registry lists 14 outcome measures and 16 analyses. It reports 12 secondary outcomes. Details of eligibility criteria, trial sites, allocation procedures, follow-up completeness, and the protocol’s original sample-size rationale are not reported in the registry record. Consequently, the design can be described as randomized and parallel, but finer operational details cannot be established from this 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 for pre-specified exclusion criteria and if they delivered vaginally after consent was obtained. 850 patients were screened. 173 declined participation, 5 withdrew consent, 2 delivered at outside facility, 211 delivered vaginally, 10 did not meet inclusion criteria, and 8 not identified to be enrolled.

Overall Study

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Number of Participants With Wound Complication, measured over Day of surgery to 30 days postoperative.

The primary endpoint counts participants with a wound complication between the day of surgery and 30 days postoperatively. The registry describes the outcome as including hematoma, seroma, dehiscence, or surgical site infection. It is therefore a participant-level event count: the reported result indicates how many participants in each group had a qualifying complication, not the number of separate complications experienced by each participant.

A wound-complication endpoint is relevant to a comparison of post-cesarean dressing strategies because it focuses on events involving the surgical wound during the stated follow-up period. Combining the listed complications gives a broad summary of wound outcomes, but it does not show how many participants experienced each individual type. Separate counts by complication type are not reported in the primary outcome description.

The registry reports 37 participants with a wound complication among 222 assigned to negative pressure wound therapy, and 42 among 219 assigned to standard dressing. These are the posted group counts. They describe the observed trial outcomes over the specified period; they do not establish that every event was equally severe or that the endpoint captured all relevant aspects of recovery. The registry record does not report additional details about event adjudication, definitions beyond the listed terms, or how multiple complications in one participant were handled.

7. Results Posted in the Registry

• Negative Pressure Wound Therapy: 16.7% (37/222) versus 19.2% (42/219) on control; risk difference -2.5 percentage points (95% CI -9.7 to 4.6), risk ratio 0.87 (0.58–1.30); unadjusted two-proportion z-test p = 0.492.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is binary at the participant level: a participant either had a qualifying wound complication during the stated follow-up period or did not. The registry reports the number of participants with an event in each assigned group. A comparison of these counts can be expressed on different scales, including a risk difference, risk ratio, or odds ratio. These scales answer related but distinct questions, so their numerical estimates are not interchangeable.

What does a risk difference describe?

In general, a risk difference subtracts the event proportion in one group from the event proportion in the other. It expresses the contrast as an absolute difference in event probability. A value of zero represents no difference. A negative or positive sign depends on the order of subtraction and the group used as the reference, so the sign must be interpreted alongside the stated group order. For this trial, the recalculated unadjusted risk difference points toward fewer events in the negative pressure wound therapy group, and its confidence interval includes zero. The exact recalculated estimate and interval are not repeated because their displayed precision exceeds the limit used for this explanation.

What does a risk ratio describe?

In general, a risk ratio compares the event probability in one group with the event probability in another. A value of one represents equal probabilities. A value below one indicates a lower event probability in the numerator group relative to the reference group; a value above one indicates a higher probability. The sponsor reported a risk ratio of 0.9, with a 95% confidence interval from 0.5 to 1.4. Since that interval includes one, it is compatible with no relative difference as well as with effects in either direction.

How do odds ratios differ?

In general, an odds ratio compares event odds rather than event probabilities. An odds ratio of one represents equal odds. Odds and probabilities are related but are not the same quantity, so an odds ratio should not automatically be described as a risk ratio. The registry identifies the sponsor’s method as logistic regression and its reported parameter as a risk ratio. It does not give enough model detail to reconstruct the analysis or determine exactly how that reported parameter was obtained.

What do the p-values convey?

In general, a p-value measures how incompatible the observed data are with a specified null model, assuming that model and the analysis assumptions are correct. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of an effect. The sponsor reported p = 0.54. The simple unadjusted recalculation gives p = 0.492. These values arise from analyses described differently in the registry, so they should be attributed to their respective methods rather than treated as competing versions of one identical calculation. Both values are consistent with substantial uncertainty about the group contrast.

Why are confidence intervals useful?

In general, a confidence interval describes the uncertainty around an estimate under the statistical model used to calculate it. Wider intervals indicate less precision, while narrower intervals indicate more precision. An interval containing the null value does not prove that there is no effect; it means the data and method do not rule out the null value at the interval’s stated confidence level. The sponsor’s reported interval covers values below and above one for the risk ratio. The unadjusted risk-difference interval likewise crosses zero. Those features matter more than whether the point estimate alone lies on one side of the null.

What is the distinction between reported and recalculated analyses?

The registry’s sponsor analysis is described as logistic regression. A recalculation from the posted group counts is an unadjusted comparison that does not incorporate additional model covariates. Regression can account for covariates when they are included and specified, whereas a basic comparison of group summaries does not. The registry record does not report the covariates, model specification, or other technical details for the sponsor’s analysis. The posted unadjusted calculation is therefore a transparent summary-based comparison, not a replacement for the sponsor’s model or a reconstruction of its full analysis.

The record does not report whether the primary analysis followed intention-to-treat, per-protocol, or another analysis-population definition. It also does not report the approach to missing outcome data, the treatment of participants with incomplete follow-up, or whether any adjustment was made for multiple outcomes. These omissions limit the ability to assess all assumptions behind the reported inference. The statistical results can still be described, but their interpretation should remain tied to the stated endpoint, the observed counts, and the analysis details that are actually available.

10. Confidence Intervals and Precision

The sponsor’s reported risk ratio is 0.9, with a 95% confidence interval from 0.5 to 1.4. This interval is broad enough to include values below one, the value of one representing no relative difference, and values above one. Accordingly, the point estimate alone should not be read as a precise estimate of the direction or size of the underlying contrast.

The simple unadjusted recalculation also has an interval that includes the null value for the risk difference. In general, a confidence interval conveys the range of effects reasonably compatible with the observed data and the calculation’s assumptions. It is not a guarantee that the true effect lies in the interval, nor does it assign equal plausibility to every value within it. The registry does not report confidence limits for the group event counts themselves.

Precision depends in part on the amount of information in the data, including the number of participants and the number of events. Here, the primary comparison is based on 37 and 42 participants with complications in groups of 222 and 219. Those counts support an estimate, but uncertainty remains, as reflected in the intervals and p-values. The record does not provide a pre-specified clinically important difference against which to judge whether the interval excludes effects that investigators would consider meaningful.

In general, a statistically non-significant result is not evidence of equivalence. Equivalence or non-inferiority requires a suitable design and pre-specified margin, neither of which is reported for this comparison. The estimates therefore describe an uncertain contrast rather than a precise demonstration of sameness or a precise demonstration of benefit or harm.

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 sample-size figures are planning illustrations based on the observed effect, not the trial’s original design assumptions. For this binary outcome, the listed figures are participants per arm, not total events. The calculation assumes a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set to a stated fraction of the observed effect.

For the full observed effect, the illustration lists 3,660 participants per arm for 80% power and 4,900 participants per arm for 90% power. For 75% of the observed effect, it lists 6,597 participants per arm for 80% power and 8,831 participants per arm for 90% power. For 50% of the observed effect, it lists 15,041 participants per arm for 80% power and 20,135 participants per arm for 90% power.

In general, power is the chance that a planned study will detect an effect of a specified size under the assumptions used, if that effect is present. The 80% and 90% columns describe different planning targets: the higher target is associated with a larger required number of participants per arm under the same other assumptions. Smaller effects need more participants to detect, which is why the listed requirements rise as the assumed fraction of the observed effect becomes smaller. The accompanying planning relationship should be read in that same direction, not as a description of the results actually observed in this trial.

The registry does not report the original sample-size justification or the assumptions used to plan enrollment. These figures are not observed power and should not be interpreted as a retrospective judgment about the adequacy of the trial. They are conditional illustrations based on the observed effect and the stated planning assumptions.

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 = 0.192; treatment proportion p₁ = 0.167; average p̄ = 0.179.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.5424 = 1.0631.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.5421 = 0.4563.
4. n = (1.0631 + 0.4563)² / (-0.0251)² = 3,660.0, rounded up to 3,660 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 <- 0.191780821917808                          # control proportion
p1 <- 0.166666666666667                          # 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(2289157)
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 = 3,660 (80% power), 4,900 (90% power)
 75% of observed effect: n per arm = 6,597 (80% power), 8,831 (90% power)
 50% of observed effect: n per arm = 15,041 (80% power), 20,135 (90% power)
power.prop.test: 3660 
Simulated power at n = 3,660 per arm: 0.800 (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. Analyses Reported by the Sponsor

The sponsor’s own reported analysis is a logistic regression analysis with a risk ratio as the reported parameter. It gives a risk ratio of 0.9, a 95% confidence interval from 0.5 to 1.4, and p = 0.54. The interval includes one, the null value for a risk ratio, so the reported estimate does not resolve whether the underlying relative difference is below, at, or above the null.

The simple recalculation uses the posted group counts without regression adjustment. It gives p = 0.492 and an unadjusted risk-ratio point estimate of 0.869. The recalculation also produces a risk-difference interval that includes zero. The estimates are not identical to the sponsor’s reported numbers, and the record describes different analysis approaches. Details of the regression model, including any covariates, are not reported, so the precise reason for the numerical difference cannot be established from the registry record.

No discrepancy is flagged between the posted results and the recalculation. The sponsor’s analysis remains the registry’s reported analysis; the recalculation is a summary-based check, not a substitute for the sponsor’s model. Both approaches indicate uncertainty in the group contrast, and neither should be treated as proving that the dressing strategies are equivalent.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists 14 outcome measures, 16 analyses, and 12 posted secondary outcomes. The primary outcome concerns participants with wound complications. Secondary outcomes cover several kinds of healthcare use and delivery or incision characteristics. Because several outcomes and analyses are listed, it is important to know whether the statistical plan controlled the chance of false-positive findings across multiple comparisons. Any multiplicity adjustment or hierarchy for testing outcomes is not reported in the registry record.

The record does not report the analysis population used for the primary result. It is therefore not possible to determine from the registry whether all randomized participants were analyzed according to their assigned groups, whether the analysis excluded participants, or whether a different population definition was used. The approach to missing outcome data, incomplete follow-up, and participants without an evaluable wound assessment is also not reported.

The sponsor’s analysis is identified as logistic regression, but the model specification and any covariates are not reported. The registry does not report whether the primary analysis was pre-specified in a protocol or statistical analysis plan, how deviations from a plan were handled, or whether an interim analysis was performed. Details about outcome adjudication and the handling of participants with more than one complication are also not reported.

These gaps do not change the posted counts, but they limit a full assessment of the inferential process and the risk of selective analysis or reporting. The registry record does not establish how the secondary outcomes were prioritized or whether their analyses were confirmatory or exploratory. Interpretation should therefore focus on the stated primary outcome and treat findings across multiple outcomes cautiously, especially when the broader analysis plan is unavailable.

15. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

The adverse-event figures list zero serious adverse events and zero other adverse events in the negative pressure wound therapy group, among 222 participants at risk. The standard dressing group likewise lists zero serious adverse events and zero other adverse events, among 219 participants at risk. The death count is zero in each group.

These are the event counts recorded in the registry. They should be read as reported counts over the adverse-event reporting period and definitions used by the trial, not as proof that no adverse events could occur. The registry does not report the definitions, collection procedures, severity criteria, or detailed observation period for these adverse-event summaries. It also lists no serious adverse event terms.

Zero recorded events in both groups provide no observed event-count contrast to estimate for these categories. They do not establish that the groups have identical safety profiles, particularly without detail on event ascertainment and follow-up. The figures are descriptive safety information and should be interpreted alongside the limits of the registry’s reporting.

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

The result describes wound-complication counts among participants randomized to two post-cesarean dressing approaches. There were 37 participants with a complication in the negative pressure wound therapy group and 42 in the standard dressing group, over the period from the day of surgery through 30 days postoperatively. The sponsor’s risk-ratio estimate is 0.9, with a 95% confidence interval from 0.5 to 1.4 and p = 0.54.

The observed counts favor the negative pressure wound therapy group in direction, but the uncertainty intervals include the null. The data therefore do not provide a precise estimate of the size or direction of any underlying difference. A result that does not meet a conventional threshold for statistical significance is not proof that there is no effect. Likewise, an estimate on one side of the null does not, by itself, establish a treatment effect.

The trial addresses a specific question about high-risk obstetrical patients undergoing cesarean section, with the compared intervention and follow-up period described in the registry. It does not, from these figures alone, establish results for other patient groups, surgical procedures, devices, or durations of follow-up. The registry does not report a subgroup analysis or separate primary counts for hematoma, seroma, dehiscence, and surgical site infection, so the combined endpoint cannot identify which components may account for the observed totals.

The record also does not report an equivalence margin, a non-inferiority margin, or the original sample-size rationale. Thus, the findings should not be framed as demonstrating equivalence or excluding clinically important differences. The most defensible reading is descriptive: the posted event counts were close in direction and the estimated contrast remains uncertain under both the sponsor’s reported analysis and the simple unadjusted recalculation.

17. Why This Trial Matters Statistically

This trial is statistically useful as an example of the distinction between a point estimate and the uncertainty around it. The observed wound-complication counts are lower in the negative pressure wound therapy group, but the sponsor’s confidence interval includes the null value for a risk ratio, and the unadjusted recalculation’s risk-difference interval includes no difference. The direction of the point estimate therefore should not be mistaken for a precise conclusion.

It also illustrates why the analysis scale and method matter. A risk ratio describes a relative contrast, while a risk difference describes an absolute contrast. Logistic regression is the sponsor’s reported method, whereas the recalculation uses posted group summaries without adjustment. These analyses address the same general comparison but are not interchangeable. The registry does not provide enough model detail to fully reconcile their estimates.

The trial’s registry record also shows why transparency about the broader analysis plan is important. Several outcomes and analyses are listed, but multiplicity procedures, missing-data methods, and the analysis population are not reported. Those details affect how confidently readers can interpret results beyond the primary endpoint. The posted summaries are still informative, but they do not answer every methodological question.

Finally, the planning illustrations help separate observed results from study-design calculations. They show how sample-size requirements change under stated assumptions and different assumed effect sizes; they are not evidence about the power achieved in the completed trial. Keeping descriptive counts, inferential uncertainty, and planning assumptions distinct is central to a careful reading of clinical trial results.

18. Limitations

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

The registry describes the population as high-risk obstetrical patients undergoing cesarean section but does not report the criteria defining high risk or the full eligibility criteria. The generalizability of the results to other patient groups is therefore difficult to assess. The record names the sponsor but does not report trial-site details, so the setting and breadth of recruitment cannot be characterized from the available information.

The primary endpoint combines several wound complications, including hematoma, seroma, dehiscence, and surgical site infection. The registry does not report separate primary counts for these components, detailed definitions, or methods for event adjudication. The outcome is limited to the stated period through 30 days postoperatively; results beyond that period are not reported for the primary endpoint.

No masking is listed. The record does not describe the practical measures, if any, used to reduce the influence of treatment knowledge on outcome assessment. It also does not report details of allocation concealment, missing data, follow-up completeness, or the primary analysis population. These omissions make it harder to assess possible bias and reproduce the sponsor’s analysis.

The sponsor’s model is identified as logistic regression, but its specification is not reported. The registry does not report multiplicity adjustment, interim analyses, or the original sample-size justification. The trial’s observed counts and uncertainty can be described, but the information available does not support a complete audit of the protocol, analysis plan, or all sources of uncertainty.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Negative Pressure Wound Therapy in Cesarean Section — NCT02289157.
• Registry record last updated 2018-07-02; status last verified 2018-05.
• Results first posted 2018-07-02.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

22. Record Summary

NCT02289157 is a completed randomized, parallel-group trial of negative pressure wound therapy after cesarean section in high-risk obstetrical patients. The registry reports enrollment of 441 participants and no masking. The intervention group received the Prevena negative pressure wound therapy system after standard cesarean section; the comparison group received a standard dressing.

The primary outcome was the number of participants with a wound complication from the day of surgery through 30 days postoperatively. The posted counts were 37 among 222 participants assigned to negative pressure wound therapy and 42 among 219 assigned to standard dressing. The sponsor reported a risk ratio of 0.9 with a 95% confidence interval from 0.5 to 1.4 and p = 0.54. A simple unadjusted recalculation also leaves substantial uncertainty about the group contrast.

The registry lists secondary outcomes and adverse-event counts, but does not report key details about missing data, analysis populations, multiplicity, or the original sample-size rationale. The results describe this trial and its stated follow-up; they do not establish equivalence or settle the size and direction of any underlying difference.