Xenogeneic Collagen Matrix as an Alternative to Free Gingival Grafting for Oral Soft Tissue Augmentation: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Xenogeneic Collagen Matrix as an Alternative to Free Gingival Grafting for Oral Soft Tissue Augmentation (NCT01952301), a randomized na study. The primary outcome is Keratinized Tissue Width.

2. Clinical Question

3. Trial Design

The registry records a randomized, single-masked study with a parallel model, completed after enrolling 30. It lists xenogeneic collagen matrix and free gingival graft as the two device interventions. The brief summary, however, calls the study split-mouth. In a split-mouth design, a participant can receive both treatments at different sites, whereas a parallel design generally assigns participants to separate groups. These descriptions are not reconciled in the registry record, so the actual allocation unit and the relationship between the reported group counts and the number of distinct participants remain unclear.

The primary outcome was change in keratinized tissue width at six months. The registry describes a planned non-inferiority design using a paired t-test, a non-inferiority margin of 1.0 mm, a within-subject standard deviation of 1.0 mm, and a one-sided alpha of 0.05. It states that this calculation resulted in a sample size of 27 and that 30 subjects were planned to account for potential loss to follow-up. Further details of the original design and how the planned calculation was implemented are not reported in the registry record.

The primary outcome entries list 30 observations in each group. Those counts should not automatically be read as 60 separate participants, given the split-mouth description and overall enrollment of 30. The record does not explain the count convention. It also does not state the masking procedure or who was masked. The study’s completion status is recorded as completed. The registry lists one outcome measure and one analysis, with no secondary outcomes posted.

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: Sample size was determined using 80% power fans assuming a paired t-test of non-inferiority with a non-inferiority margin of 1.0 mm, a within-subject standard deviation of 1.0 mm, and a one-sided alpha of 0.05, resulting in a sample size of 27. To account for potential loss-to-follow-up, 30 subjects were enrolled in the trial.

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 Keratinized Tissue Width, measured over 6 months.

Keratinized tissue width is a continuous measurement of the width of tissue around teeth, reported in millimeters. The primary endpoint is the change in that width by six months. Measuring change focuses on how the measured width differs over the study period, rather than describing only the value at a single time. The registry gives no baseline group means or further details of the measurement procedure, so the baseline values and the exact calculation of change are not reported in the registry record.

This endpoint is relevant to the stated study objective because the study sought to generate keratinized tissue around teeth where the starting amount was insufficient, defined in the brief summary as < 2mm. A continuous measurement retains the magnitude of observed changes rather than reducing the result to a category such as improved or not improved. The registry reports the outcome as a mean with standard deviation and gives a six-month timeframe.

The posted group means are 4.42 for free gingival graft and 2.92 for xenogeneic collagen matrix. The comparison is expressed as collagen matrix minus graft, with a mean difference of -1.5 mm. Those values describe the posted outcome summaries; they do not by themselves establish whether a prespecified non-inferiority criterion was met. That assessment requires an analysis aligned with the study design and a clear comparison against the non-inferiority margin.

7. Results Posted in the Registry

• Xenogeneic Collagen Matrix: mean 2.92 versus 4.42 on control; difference -1.50 (95% CI -1.89 to -1.11), standardized effect d = -1.95; z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

For a continuous outcome such as keratinized tissue width, a mean gives the arithmetic average of the recorded measurements, while a standard deviation describes how spread out those measurements are around the mean. The posted mean values are 4.42 for free gingival graft and 2.92 for xenogeneic collagen matrix; the corresponding standard deviations are 0.64 and 0.88. These summaries do not show the individual measurements or how outcomes varied within participants.

What does the recalculated mean difference represent?

The unadjusted recalculation subtracts the free gingival graft mean from the collagen matrix mean. Its estimate is -1.5 mm, with a 95% confidence interval from -1.889 to -1.111. The sign depends on the chosen direction of subtraction: a negative value means that the collagen matrix group’s posted mean is lower. This is a summary-statistic comparison, not the sponsor’s reported analysis. It does not establish how the study’s randomization, any pairing, baseline measurements, or other analysis choices were handled.

What does a confidence interval communicate?

A confidence interval describes statistical uncertainty around an estimate under the assumptions of the method used. It helps readers consider both the estimated size and the precision of a difference. The interval from the unadjusted recalculation is entirely below zero, but that fact alone does not establish that the difference is clinically important, or answer whether a non-inferiority margin was met. In a non-inferiority analysis, the interval must be interpreted relative to the prespecified margin and the direction of the comparison. The registry does not post the sponsor’s interval for that assessment.

How does a p-value differ from an effect estimate?

A p-value describes how compatible observed data are with a specified statistical null model, given the test’s assumptions. It does not quantify the treatment difference, its practical importance, or the probability that a treatment is beneficial. The sponsor reports p < 0.001 for a Wilcoxon (Mann-Whitney) analysis. The recalculated unadjusted comparison gives p <0.001. These values come from different analyses and should not be treated as interchangeable. The sponsor’s estimate, confidence interval, and further test details are not reported in the registry record.

What does the reported Wilcoxon method test?

In general, the Wilcoxon rank-sum test, also called the Mann-Whitney test, compares the ranks of observations in two independent groups. It is not simply a test of arithmetic means. Under particular distributional conditions, it can be interpreted as a comparison of location, but the test’s general target is not identical to a mean difference. It also ordinarily assumes that observations in one group are independent of observations in the other. If sites within the same person were paired, an analysis should account for that dependence. The registry’s parallel-model entry, split-mouth summary, paired design paragraph, and Mann-Whitney analysis do not clarify how the actual data structure was handled.

Why is the non-inferiority framing important?

In general, a non-inferiority trial asks whether the intervention’s outcome is not worse than the comparator by more than a prespecified margin. The margin defines the largest difference that would still count as acceptable for that objective; it is not the same as a threshold for ordinary statistical significance. The registry’s design description gives a margin of 1.0 mm and states that the planned calculation used a paired t-test, within-subject standard deviation of 1.0 mm, and one-sided alpha of 0.05. It reports a resulting sample size of 27 and says 30 subjects were planned to allow for potential loss to follow-up. However, the registry does not include the sponsor’s estimate and confidence interval needed to inspect the result against the margin. The posted analysis label alone does not supply that missing evidence.

The planning description and the posted outcome summaries answer different questions. The design paragraph describes assumptions used to plan a non-inferiority study; the posted results give group means and standard deviations; and the sponsor’s analysis reports a rank-based p-value. Keeping these distinct avoids treating a difference test as a non-inferiority conclusion or treating the unadjusted recalculation as the sponsor’s analysis. The registry record does not report enough detail to determine whether the primary inferential analysis followed the paired plan.

10. Confidence Intervals and Precision

The unadjusted recalculated mean difference is -1.5 mm, with a 95% confidence interval from -1.889 to -1.111. This interval gives a range of values compatible with the recalculated comparison under its statistical assumptions. Since the entire interval is below zero, it indicates that the posted mean is lower in the collagen matrix group than in the free gingival graft group using the stated subtraction direction. The interval does not express the probability that the true effect falls within its bounds, and it does not measure clinical importance.

Precision depends on the amount of information in the data and on the variability and analysis structure. The posted standard deviations are 0.64 for free gingival graft and 0.88 for collagen matrix. The record lists 30 observations in each outcome group, but overall enrollment is 30 and the brief summary says split-mouth. The registry does not explain whether the group counts refer to participants, sites, or paired observations. Without that clarification, the effective information and the handling of within-person dependence cannot be assessed from the record.

The interval is from an unadjusted recalculation, not the sponsor’s Wilcoxon (Mann-Whitney) analysis. The sponsor’s confidence interval and estimate are not reported in the registry record. Moreover, a non-inferiority conclusion requires an interval assessed against the stated 1.0 mm margin, with a comparison direction and analysis consistent with the design. The recalculated interval should therefore be read as a precise summary-statistic comparison, not as a verified non-inferiority result.

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 display gives participants per arm, not total events. Its illustrations use the observed effect as a starting point and show planning for 100%, 75%, and 50% of that observed effect. At 100% of the observed effect, the display gives 5 participants per arm for 80% power and 6 participants per arm for 90% power. At 75%, it gives 8 participants per arm for 80% power and 10 for 90%. At 50%, it gives 17 participants per arm for 80% power and 23 for 90%.

In plain terms, the 80% power column gives the number of participants per arm in the planning illustration when the analysis is designed to have an 80% chance of detecting the specified effect under the stated assumptions. The 90% power column gives the corresponding number for a 90% chance. These are planning quantities, not observed power and not a claim about the chance that the trial’s result is correct.

The illustration assumes a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The target effect is the listed fraction of the observed effect. Smaller effects need more participants to detect under these assumptions, which is why the participant counts rise as the planned effect fraction gets smaller. These calculations are separate from the trial’s own design calculation. The registry’s own design paragraph describes a paired t-test, non-inferiority margin of 1.0 mm, within-subject standard deviation of 1.0 mm, and one-sided alpha of 0.05; it reports a sample size of 27 and planning for 30 subjects to account for potential loss to follow-up. The display’s assumptions should not be confused with that original design description.

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 = -1.500; pooled standard deviation σ = 0.769.
2. n = 2 × (1.96 + 0.842)² × 0.769² / -1.500² = 4.1, rounded up to 5 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 <- -1.5                      # observed difference in means
sigma <- 0.769415362466854                     # 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(1952301)
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 = 5 (80% power), 6 (90% power)
 75% of observed effect: n per arm = 8 (80% power), 10 (90% power)
 50% of observed effect: n per arm = 17 (80% power), 23 (90% power)
power.t.test: 5.3 
Simulated power at n = 5 per arm: 0.768 (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. Analyses Reported by the Sponsor

The sponsor’s posted analysis is a Wilcoxon (Mann-Whitney) test, with p < 0.001, labelled as a non-inferiority or equivalence analysis. The registry does not provide its effect estimate, confidence interval, analysis population, or further test details. The p-value alone does not show the difference’s direction or size, and it does not allow readers to check whether the result met the stated non-inferiority margin of 1.0 mm.

The separate unadjusted recalculation compares the posted means and gives a mean difference of -1.5 mm, with a 95% confidence interval from -1.889 to -1.111 and p <0.001. This calculation is based on the posted group summaries and is not the same method as the sponsor’s rank-based test. Its estimate and interval provide magnitude information that the sponsor’s posted analysis lacks, but they do not replace the sponsor’s analysis or resolve whether the appropriate paired structure was used.

The registry does not flag a discrepancy between its reported result and the recalculation. Even so, different methods can target different quantities and rely on different assumptions. The sponsor’s analysis should be understood as the trial’s reported inferential result; the recalculation is a supplementary, unadjusted comparison of summaries, not a correction or a verified non-inferiority assessment.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists one outcome measure and one analysis, with no secondary outcomes posted. This limited listing reduces the number of posted outcomes that can be inspected, but it does not establish whether the investigators measured or analyzed other outcomes that were not entered in the record. The registry does not report whether multiplicity adjustments were planned or used, or whether additional analyses were conducted.

Missing-data methods are not reported in the registry record. The design description says 30 subjects were planned to account for potential loss to follow-up, but it does not state how many participants completed the six-month assessment, whether any outcome values were missing, or how missing observations were handled. The outcome groups each list 30 observations, while overall enrollment is 30; the record does not explain how those counts relate to participants, sites, or follow-up completion. No imputation, complete-case rule, or sensitivity analysis is described.

The analysis population is also not reported in the registry record. It is not stated whether the posted analysis included all randomized participants, only those with available six-month measurements, or a subset defined by another rule. For a non-inferiority objective, analysis-population choices can matter because exclusions and missing outcomes can affect the comparison. The registry does not provide enough information to assess those choices.

The record gives a planned paired t-test for the design calculation and a reported Wilcoxon (Mann-Whitney) analysis, while the study is variously described as parallel and split-mouth. It does not state how the analysis addressed possible within-person dependence or whether the stated primary analysis was changed. These gaps limit assessment of the full statistical plan and the robustness of the reported result.

15. 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 registry lists 0 serious adverse events among 30 at risk in the free gingival graft group and 0 among 30 at risk in the xenogeneic collagen matrix group. It also lists 0 other adverse events among 30 at risk in each group. These are the posted counts for the groups and adverse-event categories recorded in the registry.

Zero recorded events means that no events were entered for those categories in the posted results; it does not establish that no events occurred outside the registry’s reporting or observation procedures. The record does not list serious-event terms. Death counts are not reported in the registry record. It also does not describe event definitions, collection procedures, observation periods for adverse events, or whether the same participants contributed to both treatment groups. Those limitations matter when interpreting the scope of the safety information.

The figures are descriptive counts, not a comparison demonstrating that one intervention has a different safety profile. With no posted events in either group, the registry entries do not provide an estimate of relative risk or a confidence interval for an event-rate difference. The counts should be read as the adverse-event information recorded, rather than as proof of absence of risk.

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

The posted data show a difference in six-month mean keratinized tissue width outcomes: 4.42 for free gingival graft and 2.92 for xenogeneic collagen matrix. The unadjusted comparison, defined as collagen matrix minus graft, is -1.5 mm, with a 95% confidence interval from -1.889 to -1.111. This describes the direction and estimated size of the group difference in the posted summaries. It is not, by itself, a conclusion about clinical acceptability or treatment choice.

The trial is described as addressing whether collagen matrix could serve as an alternative to free gingival graft, with a non-inferiority design. The registry gives a planned margin of 1.0 mm but does not post the sponsor’s estimate and confidence interval or a clear statement that the non-inferiority criterion was met. The sponsor’s Wilcoxon (Mann-Whitney) result, p < 0.001, is labelled non-inferiority or equivalence, but the p-value alone does not show whether the margin was satisfied.

The posted information also leaves the analysis structure uncertain. The registry identifies a parallel model, while the brief summary describes a split-mouth study and the design paragraph refers to a paired test. The sponsor’s reported method is ordinarily for independent groups. The record does not explain how these descriptions fit together or how the analysis handled any pairing. The group counts and total enrollment likewise cannot be reconciled from the posted information.

Thus, the available statistics allow a descriptive comparison and show the precision of an unadjusted recalculation. They do not fully establish the sponsor’s non-inferiority claim, the effect under a paired analysis, or how broadly the result applies. The registry records zero serious and zero other adverse events in each group, but these entries do not establish that no risk exists. The result should be understood within the limits of the reported outcome and incomplete methodological detail.

17. Why This Trial Matters Statistically

This trial illustrates why a treatment comparison depends on more than the difference between two means. The registry identifies a non-inferiority objective and gives a margin of 1.0 mm, yet the sponsor’s posted analysis offers a p-value without an estimate or confidence interval. A reader therefore cannot use the sponsor’s entry to inspect how the result relates to the margin. For non-inferiority questions, the direction of comparison and uncertainty relative to the margin are central to interpretation.

It also highlights the importance of matching the statistical method to the design. The registry describes the study as both parallel and split-mouth, and its design paragraph mentions a paired t-test, while the reported analysis is Wilcoxon (Mann-Whitney). The record does not explain whether treatment was assigned by participant or site, or whether observations within a person were modeled as dependent. That information affects what the analysis estimates and how its uncertainty should be calculated.

Finally, the study record shows the distinction between a sponsor-reported test and a summary-statistic recalculation. The recalculated mean difference and interval offer a direct description of the posted means, while the sponsor’s test is a rank-based analysis with a different inferential target. Neither should be substituted for the other without understanding the design and analysis plan. Clear reporting of the estimate, interval, analysis population, missing-data handling, and paired structure would make the non-inferiority question easier to assess.

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.

Several trial-specific reporting limitations affect interpretation. The registry identifies the model as parallel, but the brief summary calls the study split-mouth. The planned sample-size description refers to a paired t-test, whereas the reported analysis is Wilcoxon (Mann-Whitney). The record does not reconcile these design and analysis descriptions or explain whether the analysis accounted for pairing. This leaves uncertainty about the proper experimental unit and the relevant variance calculation.

The outcome group entries each list 30 observations, while overall enrollment is 30. The registry does not explain whether those entries count participants, treatment sites, or paired measurements. Baseline keratinized tissue width values are not reported in the registry record, nor are the detailed measurement procedures or group-specific follow-up counts. Missing-data handling and analysis-population definitions are also not reported.

The sponsor’s analysis is labelled as non-inferiority or equivalence and reports p < 0.001, but its estimate and confidence interval are not reported. The unadjusted recalculation has an estimate and interval, but it is not the sponsor’s analysis and does not establish the result under the stated paired design or verify non-inferiority against the margin. The registry does not report multiplicity methods, protocol deviations, or additional sensitivity analyses. These limitations constrain assessment of both the planned inferential question and the robustness of the posted comparison.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Xenogeneic Collagen Matrix as an Alternative to Free Gingival Grafting for Oral Soft Tissue Augmentation — NCT01952301.
• Registry record last updated 2014-06-09; status last verified 2014-05.
• Results first posted 2014-05-15.
• 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

The registry record describes a completed randomized study comparing xenogeneic collagen matrix with free gingival graft for oral soft tissue augmentation. It lists an enrollment of 30 and records the primary outcome as change in keratinized tissue width at six months. The brief summary describes a split-mouth comparison, although the model is listed as parallel.

Posted mean outcome values are 4.42 for free gingival graft and 2.92 for collagen matrix. The unadjusted recalculated comparison is -1.5 mm for collagen matrix minus graft, with a 95% confidence interval from -1.889 to -1.111. The sponsor reports a Wilcoxon (Mann-Whitney) analysis with p < 0.001, but its estimate and confidence interval are not reported in the registry record. The record lists zero serious and zero other adverse events in each group. Uncertainty about the design description, analysis structure, and non-inferiority result limits interpretation.