Dermabond PRINEO for Total Shoulder Arthroplasty: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Dermabond PRINEO for Total Shoulder Arthroplasty (NCT03860181), a randomized na study. The primary outcome is Comparison of the Total Cost of Wound Closure When Using Dermabond PRINEO Versus the Surgeon's Control..

2. Clinical Question

3. Trial Design

This was a randomized, controlled trial with a parallel-group model and single masking. It is marked completed in the registry record, with enrollment of 89. The registry lists the study start as 2019-03-25 and primary completion as 2020-06-23. The identity of the masked party or parties is not reported in the registry record, nor are the method used to generate the random allocation sequence, allocation concealment procedures, or details about implementation of randomization.

The four arms combine two closure approaches with two surgeon labels. Surgeon 1's traditional control was Dermabond with subcuticular sutures, while Surgeon 2's control was metal staples. Each surgeon also has a Dermabond PRINEO arm. The posted mean-cost results give 21.0 participants for the Surgeon 1 control, 22.0 for the staples control, 25.0 for PRINEO with Surgeon 1, and 21.0 for PRINEO with Surgeon 2. These are the group counts reported with the primary outcome, not a description of how many people were randomized to each group before any exclusions or missing outcome data.

The registry record describes the purpose as investigating whether closure with PRINEO would be faster and improve wound healing compared with standard closure methods after total shoulder replacement. The posted primary endpoint, however, is closure cost. Secondary outcome titles cover assessments of inflammation, scar appearance, cosmesis, and overall opinion by surgeons, operating physicians, plastic surgeons, or patients. Numerical results for those named secondary outcomes are not reported in the registry record. Eligibility criteria, study locations, the number of sites, and details of follow-up beyond the primary endpoint timeframe are not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

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 Comparison of the Total Cost of Wound Closure When Using Dermabond PRINEO Versus the Surgeon's Control., measured over 3 months post-operative.

The primary endpoint is the mean total cost of wound closure, measured in dollars at 3 months post-operative. The registry description says cost was calculated from wound-closure product cost per patient, plus operating-room cost per minute multiplied by closure time per centimeter of incision length, and then by average incision length per patient. The posted results give a mean and standard deviation for each of the four groups.

This measure combines material costs with operating-room time and closure duration, while incorporating incision length. In general, a cost endpoint can help describe the resources associated with different approaches to a procedure. It is distinct from a direct measure of wound healing, scar quality, or patient experience. The registry record's stated overall aim includes speed and wound healing, but the posted primary result addresses cost; it does not itself establish how quickly wounds healed or whether healing differed.

The cost estimate depends on the components named in the registry description. Component-specific costs, closure times, incision lengths, and the detailed method for assigning operating-room cost are not reported in the registry record. Therefore, the mean cost differences describe the combined endpoint rather than identifying which component accounts for a difference. The record gives the assessment timeframe but does not describe the handling of costs or outcome measurements that may have been unavailable.

7. Results Posted in the Registry

• Metal Staples - Surgeon 2: mean 304.75 versus 627.83 on control; difference -323.08 (95% CI -366.06 to -280.10), standardized effect d = -4.56; z-test p = <0.001.
• Dermabond PRINEO - Surgeon 1: mean 568.53 versus 627.83 on control; difference -59.30 (95% CI -111.31 to -7.29), standardized effect d = -0.66; z-test p = 0.025.
• Dermabond PRINEO - Surgeon 2: mean 420.77 versus 627.83 on control; difference -207.06 (95% CI -251.60 to -162.52), standardized effect d = -2.81; z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary measure is continuous: cost in dollars. The registry reports each group's mean and standard deviation. A mean represents the average of the recorded values, while a standard deviation describes how spread out the individual values are around that average. In general, the mean is a useful summary when the question concerns average resource cost, but it does not show the full shape of the cost distribution or whether a few values are especially influential. The registry record does not provide individual cost observations or distribution plots for assessing those features.

How should the group comparisons be read?

The posted mean differences are unadjusted recalculations from the group summary statistics. Each comparison uses the traditional Dermabond plus subcuticular sutures group under Surgeon 1 as its reference. A negative difference means that the comparison group's posted mean is lower than that reference mean. The three comparisons are against different closure groups: metal staples under Surgeon 2, PRINEO under Surgeon 1, and PRINEO under Surgeon 2. Thus, the values do not represent three versions of a single comparison with the same control procedure.

The record reports p-values for these recalculated comparisons. A p-value describes how compatible the observed data are with a specified null hypothesis under the statistical model used; it is not the probability that a treatment works or that a result occurred by chance. The p-values are unadjusted. The exact test or model used to obtain them is not reported in the registry record, and the posted results do not describe adjustment for surgeon, other covariates, or multiple comparisons. The displayed recalculations should therefore be distinguished from a complete analysis plan or an adjusted analysis.

What do standard deviations and confidence intervals contribute?

A standard deviation summarizes variability among observations within a group. It is not the uncertainty in the estimated group mean itself. A confidence interval around a mean difference communicates uncertainty in that contrast: a narrower interval indicates greater precision under the calculation, while a wider interval indicates less precision. In general, intervals that include values on both sides of zero would allow a broader range of directions for the difference than intervals entirely on one side. The intervals reported for these comparisons lie below zero, but their interpretation remains subject to the unadjusted nature of the calculations and the limitations of the registry record.

Confidence intervals and p-values are related ways to describe uncertainty, but neither conveys the practical importance of the cost difference by itself. Practical interpretation depends on the costs and setting relevant to the reader, and those external considerations are not reported in the registry record. The interval procedure and confidence level are also not reported in the registry record, so no specific confidence level should be inferred from the numbers alone.

What is not specified about the analysis?

The registry record identifies the endpoint as a mean with standard deviation, but it does not report the statistical test, any transformation of cost values, model assumptions, covariate adjustment, or how departures from those assumptions were assessed. It does not report an analysis population definition, whether all randomized participants were included in each analysis, or how missing outcome values were handled. The posted outcome group counts do not establish whether any randomized participants were excluded or whether every participant contributed a cost measurement.

In general, cost data can be unevenly distributed, with a smaller number of observations far from the mean. If that occurs, the mean and standard deviation may not fully characterize the data, and a model that assumes a particular distribution may need scrutiny. The registry record does not provide information about skewness, outliers, or alternative summaries, so it is not possible to assess those issues from the posted summary statistics alone. This is a general consideration for cost outcomes, not a claim about the shape of these trial data.

The listed comparisons are recalculated from posted summaries, rather than a description of a fully specified trial analysis. No discrepancy between those recalculations and a sponsor analysis is flagged in the registry record. The record does not report the trial's original statistical analysis plan, the planned primary contrast in sufficient detail to establish how it was tested, or whether an interim analysis was planned or conducted. Those omissions limit the ability to reconstruct the analysis decisions or evaluate whether the posted comparisons were the sole analyses performed.

10. Confidence Intervals and Precision

The reported confidence intervals express uncertainty around the unadjusted recalculated mean differences. For metal staples under Surgeon 2 compared with the Surgeon 1 traditional-closure control, the estimate is -323.1 dollars, with an interval from -366.1 to -280.1. For PRINEO under Surgeon 1 compared with that control, the estimate is -59.3 dollars, with an interval from -111.3 to -7.285. For PRINEO under Surgeon 2 compared with the same reference group, the estimate is -207.1 dollars, with an interval from -251.6 to -162.5.

In general, a confidence interval gives a range of values that are reasonably compatible with the observed data under the method used to calculate it. The range is not a guarantee that the true value lies within it, and it does not show the distribution of individual participants' costs. The intervals in the registry record all lie below zero, while their widths differ. The interval around the PRINEO comparison under Surgeon 1 is wider than a single point estimate would suggest; it includes values closer to zero as well as more negative values. The other two intervals also span a range of possible mean differences rather than specifying an exact cost contrast.

The registry record does not report the confidence level, method for constructing these intervals, or any adjustment for multiple comparisons. Their precision should therefore be read as the precision of the posted unadjusted calculations, not as a complete account of uncertainty under every plausible analysis. Differences in group size and observed spread may contribute to uncertainty, but the record does not provide a model-based decomposition of that uncertainty. The intervals also do not capture uncertainty about the applicability of these results to other surgeons, settings, or cost structures.

11. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are illustrations based on the observed effect, not the trial's original design assumptions. The registry record does not report the trial's own sample-size justification or original design assumptions. These illustrations use a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. They show participants per arm for 80% power and participants per arm for 90% power. They are not observed power.

Each row represents planning for a stated fraction of the observed effect: 100%, 75%, or 50%. A row that plans for 100% uses the full observed comparison as the effect to detect. Planning for 75% or 50% uses a smaller target effect. Smaller effects need more participants to detect under the listed assumptions, which is why the requirements generally rise as the target effect becomes smaller. The 80% power column gives the illustrative number of participants per arm for a design with 80% power under those assumptions; the 90% column gives the corresponding number per arm for 90% power. Greater planned power generally requires a larger sample for the same target effect.

The figures are shown separately for each posted comparison, so the relevant planning row depends on which comparison and effect fraction a reader wants to consider. They should not be read as observed enrollment, as a claim about the study's achieved power, or as a recommendation for a future trial without reconsidering its target effect, population, outcome variability, and design. These are planning illustrations calculated from the observed effect and do not replace a design based on an independently justified effect of interest.

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, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -323.080; pooled standard deviation σ = 70.863.
2. n = 2 × (1.96 + 0.842)² × 70.863² / -323.080² = 0.8, rounded up to 1 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
delta <- -323.08                      # observed difference in means
sigma <- 70.863465078465                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^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 (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(3860181)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 1 (80% power), 2 (90% power)
 75% of observed effect: n per arm = 2 (80% power), 2 (90% power)
 50% of observed effect: n per arm = 4 (80% power), 5 (90% power)
power.t.test: 2.2 
Simulated power at n = 1 per arm: NA (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

13. Multiplicity, Missing Data and Analysis Populations

The registry record lists 11 outcome measures and 2 analyses. It names 8 secondary outcomes, including overall opinion, wound inflammation, scar assessment, wound cosmesis, and differences between patient and surgeon scar-assessment scores. Numerical findings for those named secondary outcomes are not reported in the registry record. The primary cost comparisons are presented as unadjusted recalculations, and a multiplicity adjustment is not reported in the registry record.

In general, when several outcomes or contrasts are examined, the chance of at least one apparently unusual result can increase if each is assessed separately. Adjustment methods can address some of that concern, but the appropriate approach depends on which outcomes and contrasts were planned and how they relate to one another. The registry record does not state which endpoint was designated primary in a statistical analysis plan beyond the posted primary outcome listing, how the secondary outcomes were prioritized, or whether any correction was used. The unadjusted p-values should not be interpreted as accounting for all possible comparisons in the study.

Missing-data handling is not reported in the registry record. The record does not say how incomplete cost observations or missing secondary assessments were handled, whether any participants were excluded from an analysis, or whether sensitivity analyses were conducted. The group counts attached to the primary outcome describe the posted results but do not establish an analysis population or show whether those counts match the randomized groups.

The record also does not report whether analyses followed an intention-to-treat principle, used a per-protocol population, or applied another population definition. The analysis plan, any interim analyses, and rules for handling protocol deviations are not reported in the registry record. Consequently, it is not possible to determine from the posted information whether all randomized participants were retained in the primary comparison or how analysis choices may have affected the estimates.

14. Safety Results

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

Deaths

Other (non-serious) adverse events

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

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: infection (4.8%).

The serious adverse event figures are listed separately from the primary cost outcome. In the traditional Dermabond plus subcuticular sutures group, 1 of 21 participants at risk had a serious adverse event. The metal-staples group reports 0 of 22, and the Dermabond PRINEO group reports 1 of 46. Other adverse events are listed as 0 in each group, and deaths are listed as 0 in each group. These counts should be read with their stated at-risk group sizes and should not be converted into rates by combining groups.

The PRINEO safety group is reported as one group, without a Surgeon 1 versus Surgeon 2 breakdown in the adverse-event information. The record lists Infection at 4.8% among the leading serious adverse-event terms. The registry record does not explain the denominator for that percentage or provide event-level detail linking the term to a specific group. Definitions of serious adverse events, timing, severity details, and methods for collecting adverse events are not reported in the registry record. The small counts and limited detail do not support broad conclusions about comparative safety.

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

The posted primary results describe mean closure costs for four groups in this trial. The recalculated comparisons show lower mean costs for each listed comparison group than for the traditional Dermabond plus subcuticular sutures control under Surgeon 1. The size of the estimated difference varies across the surgeon-specific PRINEO comparisons and the metal-staples comparison. The confidence intervals describe uncertainty around those unadjusted contrasts, while the p-values give another summary of compatibility with a null hypothesis under the calculation used.

These findings do not establish that one closure approach is clinically superior, that wound healing is better or worse, or that the results apply to every total shoulder arthroplasty setting. The primary endpoint is a combined cost measure, not a direct wound-healing outcome. The record's stated research aim includes speed and healing, but numerical results for the listed secondary assessments are not reported in the registry record. The primary cost results cannot substitute for those missing clinical or appearance outcomes.

The comparisons also involve surgeon-specific control methods. Traditional Dermabond plus subcuticular sutures is the control for Surgeon 1, while metal staples are the control for Surgeon 2. The registry record does not report an adjusted analysis that separates closure method from surgeon or other factors. Accordingly, the results are most directly read as the posted group means and contrasts in this study, rather than as a universal ranking of closure options.

Nor do the results by themselves establish the value of the cost difference for a particular health system or patient. That judgment would require context about local product costs, operating-room costs, and the importance assigned to other outcomes. Those details are not reported in the registry record. The data describe cost estimates and their statistical uncertainty; they do not provide medical advice or determine which closure method should be used.

16. Why This Trial Matters Statistically

This trial illustrates why the definition of an endpoint matters to statistical interpretation. The primary outcome is not simply the price of a closure product: the registry describes a combined measure that includes product costs, operating-room cost, closure time per centimeter, and average incision length. A difference in the endpoint therefore cannot identify which component accounts for the observed contrast. Component-level results would be needed to distinguish material cost from time-related contributions, and those are not reported in the registry record.

The four-arm structure also matters. PRINEO is listed with both Surgeon 1 and Surgeon 2, while the control procedure differs between them. The posted comparisons make this structure visible: the two PRINEO comparisons have different estimates, and the staples comparison uses a different surgeon-specific control. Treating all arms as if they were a single pooled comparison would conceal information about how the comparisons were defined. The registry record does not report a combined PRINEO estimate or an adjusted model for surgeon.

Finally, statistical evidence and practical relevance are not interchangeable. The recalculated estimates, confidence intervals, and unadjusted p-values describe the posted cost data, but they do not resolve questions about healing, cosmetic outcomes, or generalizability. The example underscores the need to read effect estimates alongside endpoint construction, group structure, multiplicity, and incomplete reporting. Those considerations are central to understanding what a trial result can support and what remains uncertain.

17. Limitations

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

Several limitations are specific to the posted record. First, the primary endpoint combines product cost with operating-room time, closure duration, and incision length. The record does not report the component-level values, the detailed operating-room cost rate, or the distribution of individual costs. It is therefore not possible to identify the source of differences among group means or to assess how sensitive the combined endpoint may be to its inputs.

Second, the groups are organized by surgeon as well as closure method. Surgeon 1's control is traditional Dermabond with subcuticular sutures, whereas Surgeon 2's control is metal staples. Each surgeon has a PRINEO arm, but the record does not report adjustment for surgeon or whether other procedural differences were controlled. Surgeon-specific differences could be relevant to interpreting the comparisons; the posted data do not separate these influences.

Third, only summary statistics are given for the primary endpoint. Individual observations, distributional summaries, and the detailed analysis procedure are not reported in the registry record. Missing-data methods and analysis-population definitions are also not reported. The recalculated comparisons are unadjusted, and multiplicity handling is not reported.

Finally, although secondary outcome titles are listed, their numerical findings are not reported in the registry record. The record also does not identify who was masked, give detailed eligibility criteria, or describe the trial's original sample-size justification. These reporting gaps limit assessment of the full clinical question, the analysis choices, and how broadly the cost findings may apply.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Dermabond PRINEO for Total Shoulder Arthroplasty — NCT03860181.
• Registry record last updated 2021-10-01; status last verified 2021-09.
• Results first posted 2021-08-25.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

21. Record Summary

NCT03860181 is a completed randomized, controlled, parallel-group trial of Dermabond PRINEO for wound closure after total shoulder arthroplasty. The registry record lists single masking and enrollment of 89. Its primary outcome is mean total wound-closure cost in dollars at 3 months post-operative. Four groups are reported: traditional Dermabond with subcuticular sutures under Surgeon 1, metal staples under Surgeon 2, and PRINEO under each surgeon.

The posted primary results include group means and standard deviations, with unadjusted recalculated mean differences and confidence intervals. The record also lists secondary outcome titles and adverse-event counts, but numerical secondary outcome results and several analysis details are not reported in the registry record. The comparisons describe the observed cost summaries and their uncertainty; they do not, on their own, answer the broader questions about wound healing, appearance, or generalizability.