Comparison of a PEG Membrane and a Collagen Membrane for the Treatment of Bone Dehiscence Defects at Bone Level Implants: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Comparison of a PEG Membrane and a Collagen Membrane for the Treatment of Bone Dehiscence Defects at Bone Level Implants (NCT01012921), a randomized na study. The primary outcome is The Bone Fill Was Assessed at 6 Months After Regenerative Therapy..

2. Clinical Question

3. Trial Design

This was a randomized, parallel-group trial with no masking. The registry lists 117 participants as enrolled and the study status as completed. The study began in 2009-09, and primary completion was recorded as 2011-10. The phase is listed as NA. The sponsor was Institut Straumann AG.

Participants were assigned to MembraGel or Bio-Gide® membrane. The registry describes MembraGel as a synthetic degradable barrier membrane and Bio-Gide® as a biodegradable bilayer collagen membrane. The stated purpose was to compare their performance as barriers for guided bone regeneration around Straumann® SLActive bone-level titanium implants in bone dehiscence defects.

The primary outcome was the change in bone fill between baseline at regenerative therapy and the assessment at 6 months. The endpoint population is listed as PP, meaning the registry identifies the posted outcome as belonging to a per-protocol population; further details defining that population are not reported in the registry record. The posted group sizes for this outcome are 45 for Bio-Gide® and 40 for MembraGel. These are smaller than the overall enrollment, but participant flow and the reasons for the difference are not reported in the registry record.

The registry records one outcome measure and no posted analyses. Although the study description mentions complementary clinical parameters, no secondary outcomes are posted. The original sample-size rationale, allocation procedure details, masking rationale, and trial-specific analysis plan are not reported in the registry record. The posted comparison can therefore be interpreted as a summary of the available primary outcome data, not as a complete account of the planned trial analysis.

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: PP

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.

Population: Analysis population is PP

6. Primary Endpoint

The registry lists the primary outcome as The Bone Fill Was Assessed at 6 Months After Regenerative Therapy., measured over Assessed at 6 months reported.

The primary endpoint is mean bone fill, expressed in millimetres, assessed at 6 months after regenerative therapy. The registry defines the reported quantity as the change between the baseline measurement at regenerative therapy and the 6-month timepoint. It is therefore a measure of change over the follow-up period, rather than a measurement only of bone fill at the final visit.

A continuous endpoint preserves the measured scale of the outcome. Reporting each arm’s mean gives the average change among participants included in that arm’s endpoint population; the standard deviation describes how much individual measurements vary around that mean. The posted means are 3.2 mm for Bio-Gide® and 2.5 mm for MembraGel. Their standard deviations are 2.1 mm and 1.5 mm, respectively.

Comparing the arm means addresses the trial’s stated question about average bone fill with the two barrier membranes. The outcome record identifies the population as PP, but does not explain the detailed eligibility or handling rules for that population. It also does not report further measurement procedures or the handling of missing endpoint observations. Those details matter when judging how the endpoint was obtained and which participants contributed to the comparison.

7. Results Posted in the Registry

• MembraGel: mean 2.50 versus 3.20 on control; difference -0.70 (95% CI -1.47 to 0.07), standardized effect d = -0.38; z-test p = 0.075.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous: bone fill is measured in millimetres, and the registry reports a mean and standard deviation for each arm. The posted comparison expresses the difference as MembraGel minus Bio-Gide®. Its estimate is -0.7 mm, with a 95% confidence interval from -1.47 mm to a value slightly above zero and an unadjusted p-value of 0.075. The registry does not provide a detailed description of the statistical test or model used for the sponsor’s original analysis.

What does the mean difference represent?

A mean is the arithmetic average of the measurements in a group. The mean difference places the two group averages on the original measurement scale, so the estimate is expressed in millimetres. With the comparison ordered as MembraGel minus Bio-Gide®, a negative estimate means the MembraGel group mean is lower. The estimate does not describe every participant’s result, and it does not imply that each participant’s change was close to the group average.

The posted means are 2.5 mm for MembraGel and 3.2 mm for Bio-Gide®. Their reported standard deviations are 1.5 mm and 2.1 mm. A standard deviation describes the spread of observed measurements around a group mean. It is not the standard error, which describes uncertainty in an estimate, and it is not a confidence interval. The registry reports the outcome population as PP, but detailed rules for inclusion in that population are not reported in the registry record.

How should the confidence interval be understood?

General explanation: a confidence interval is a range of values produced by a method designed to capture the true parameter at a stated long-run frequency when its assumptions hold. For this trial, the interval is for the mean difference between the two arms. It describes uncertainty in the estimated average contrast, not the range of individual participant outcomes. Since the interval reaches across zero, it includes both no mean difference and possible differences on either side of zero. That uncertainty is important when interpreting the point estimate.

The interval’s width is affected by the variability in the measurements and by how many observations contribute to each group. The posted standard errors for the group means are 0.313 for Bio-Gide® and 0.237 for MembraGel. The registry does not report the complete method used to construct the sponsor’s interval, so the posted limits should be read as the recorded comparison rather than as evidence of a particular model or adjustment procedure.

What does the p-value say?

The unadjusted p-value is 0.075. General explanation: a p-value quantifies how often data at least as incompatible with a specified null hypothesis would arise if that hypothesis and the test assumptions were true. It does not measure the probability that the null hypothesis is true, the chance that a finding is due to chance, or the clinical importance of an estimated difference. Here it is one piece of evidence about the comparison and should be considered together with the effect estimate and its confidence interval.

The term unadjusted indicates that the posted p-value is not described as adjusted for multiple comparisons. There is only one posted outcome measure and no posted secondary outcome measures, but the registry does not describe whether other comparisons were planned or how multiplicity was addressed in the original plan. The registry also does not give a prespecified statistical significance threshold for the trial’s own analysis. The planning illustration separately lists a two-sided significance level of 0.05; that is an assumption for the illustration, not a documented original trial analysis rule.

Why might variability matter?

General explanation: greater variation among participant measurements makes a group mean less precise when the sample size is unchanged. If data are more variable than a statistical model allows, standard errors and confidence intervals can be too small, leading to overstated precision. Such extra variation is often called over-dispersion in settings where a model expects less variability. The primary outcome is continuous and the registry does not report that an over-dispersion model was used; this general issue should not be mistaken for a documented feature of this trial.

The posted standard deviations show that individual measurements vary within both arms. They do not, on their own, reveal why measurements differ or whether the statistical assumptions used in an analysis were satisfied. The registry does not report diagnostic checks, transformations, covariate adjustments, or sensitivity analyses for the primary outcome.

What analysis details remain unknown?

The record identifies randomization and a parallel design, but does not provide the random sequence method, allocation concealment details, handling of missing observations, or a full definition of the PP population. It reports no analyses and no secondary outcomes. The study’s original sample-size calculation and the trial-specific statistical analysis plan are not reported in the registry record. As a result, the posted summary permits interpretation of the arm means and comparison but not a full audit of the planned analysis.

The sample-size figures associated with the observed comparison are planning illustrations, not trial results. General explanation: power is the probability that a specified test will detect an effect of a particular size under stated assumptions if that effect is real. Power depends on the assumed effect, variability, significance level, allocation, and analysis method. The registry’s planning illustration uses the observed effect and specified assumptions, but it does not establish the trial’s original design rationale or the power actually achieved in this study.

10. Confidence Intervals and Precision

Precision concerns how narrowly the data locate the mean difference, rather than whether the point estimate is positive or negative. The posted estimate for MembraGel compared with Bio-Gide® is -0.7 mm. Its 95% confidence interval extends from -1.47 mm to a value slightly above zero. This range includes zero, so the posted data do not pin down the direction of the underlying mean difference with certainty.

General explanation: a confidence interval combines the estimated effect with the uncertainty associated with that estimate. A narrower interval indicates greater precision, while a wider interval leaves more plausible values consistent with the data and the method’s assumptions. A confidence interval is not a range containing a fixed proportion of individual participants’ outcomes. It is about uncertainty in the estimated group contrast.

The outcome group sizes are 45 for Bio-Gide® and 40 for MembraGel. The recorded standard deviations are 2.1 mm and 1.5 mm, respectively. Both group size and within-group variability influence the precision of a mean comparison. The registry does not report enough detail to assess all analysis assumptions or to reconstruct the interval method used in the original analysis.

The interval should not be interpreted as establishing that the membranes are equivalent. An equivalence conclusion would require an appropriate design and a prespecified margin defining what differences are sufficiently small; such a margin is not reported in the registry record. Nor does the interval establish the clinical importance of values within its range. The registry record does not report a threshold for a clinically meaningful difference in bone fill.

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 participants per arm, not total event counts. They are illustrations based on the observed effect, not the trial’s original sample-size calculation. The assumptions listed for this illustration are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is set to the stated fraction of the observed effect. The trial’s own design assumptions and sample-size justification are not reported in the registry record.

The 80% power column gives the number of participants per arm the illustration estimates would be needed for an 80% chance of detecting the specified effect under those assumptions, if that effect were real. The 90% power column gives the corresponding number per arm for a 90% chance. These are planning targets under a model, not counts observed in the trial. They are not observed power.

For planning at 100% of the observed effect, the figures are 109 participants per arm for 80% power and 146 participants per arm for 90% power. For planning at 75% of the observed effect, they are 194 and 259 participants per arm, respectively. For planning at 50% of the observed effect, they are 436 and 583 participants per arm, respectively. The pattern reflects a general statistical principle: smaller effects are harder to distinguish from random variation, so detecting them with the same power generally requires more participants. The curve presents this same relationship across the displayed effect assumptions; it is not a record of actual enrollment or an estimate of power achieved by the trial.

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 = -0.700; pooled standard deviation σ = 1.843.
2. n = 2 × (1.96 + 0.842)² × 1.843² / -0.700² = 108.8, rounded up to 109 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 <- -0.7                      # observed difference in means
sigma <- 1.84256892434553                     # 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(1012921)
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 = 109 (80% power), 146 (90% power)
 75% of observed effect: n per arm = 194 (80% power), 259 (90% power)
 50% of observed effect: n per arm = 436 (80% power), 583 (90% power)
power.t.test: 109.7 
Simulated power at n = 109 per arm: 0.797 (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

Multiplicity arises when an analysis considers multiple outcomes, comparisons, timepoints, or subgroups. Testing many hypotheses can increase the chance of obtaining at least one apparently unusual result even when there is no underlying effect. Statistical plans may address this through prespecified priorities or adjustments, but the appropriate approach depends on the questions and design.

The registry lists one outcome measure, the bone-fill change assessed at 6 months, and no posted secondary outcome measures. It also reports no posted analyses. The study description mentions complementary clinical parameters describing the bony and soft-tissue environments, but those measures are not posted as secondary outcomes in the record. Whether additional comparisons were planned, and how any multiplicity was handled, is not reported in the registry record. The posted p-value is unadjusted.

Missing-data handling is also not reported. The primary outcome population is identified as PP, but the registry does not define the population in detail or state how participants with missing 6-month measurements were handled. Overall enrollment is 117, while the posted primary outcome groups contain 45 and 40 participants. The safety at-risk groups have different denominators. Without a participant-flow account, it is not possible to determine from the record why those denominators differ or what happened to participants who are not represented in a given analysis.

The record does not provide a full analysis plan, explain whether the primary outcome population was prespecified, or describe sensitivity analyses for missing observations. These gaps limit assessment of the consequences of exclusions and missing measurements. They do not establish that a particular form of bias occurred; they identify details needed to evaluate the robustness and scope of the posted comparison.

14. 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.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: cerebral hemorrhagic infarction (1.7%).

The registry lists serious adverse events affecting 0 of 57 participants at risk in the Bio-Gide® group and 1 of 60 participants at risk in the MembraGel group. It lists other adverse events affecting 6 of 57 and 18 of 60 participants, respectively. These figures are counts in the safety record, with the stated at-risk denominators; they are distinct from the primary outcome sample sizes.

The serious adverse event term recorded is cerebral hemorrhagic infarction, listed as 1.7%. The registry does not provide further detail in the posted safety information about timing, event course, or how the term relates to the individual participant entry. Death counts are not reported in the registry record.

These counts describe events recorded in the registry and should not be interpreted as a complete comparison of safety without additional context. The record does not explain the event-collection period, definitions, attribution procedures, or whether the groups were followed for identical durations. Counts alone do not establish a causal relationship with a membrane. The safety denominators also differ from those for the primary endpoint, so they should not be substituted for the endpoint analysis populations.

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

The posted results show mean bone-fill changes of 3.2 mm for Bio-Gide® and 2.5 mm for MembraGel at the 6-month assessment. The estimated mean difference for MembraGel compared with Bio-Gide® is -0.7 mm. The 95% confidence interval extends from -1.47 mm to a value slightly above zero, and the unadjusted p-value is 0.075. Taken together, these statistics describe an estimated group contrast with uncertainty that includes no difference.

This does not establish that the treatments are equivalent, because the registry does not report an equivalence margin or an equivalence analysis. It also does not establish that one membrane is clinically superior. A statistical estimate and its uncertainty do not, by themselves, tell readers what size of difference would matter clinically; a threshold for clinical importance is not reported in the registry record.

The findings apply to the posted endpoint population, identified as PP, and to the recorded measure of change in bone fill at 6 months. The registry does not define the PP population in detail or explain why the endpoint group sizes are smaller than overall enrollment. As a result, the record does not establish how the comparison would look under a different analysis population or how missing observations affected the estimate.

Randomization is part of the recorded design, but the trial was unmasked, and details about allocation procedures and outcome assessment are not reported. The record also lacks a full statistical analysis plan, a sample-size justification for the original design, and posted secondary results. These limitations narrow what can be concluded from the registry entry. The statistics characterize the available posted comparison; they do not settle broader questions about membrane performance, all clinical outcomes, or outcomes beyond the recorded follow-up assessment.

16. Why This Trial Matters Statistically

This trial illustrates the difference between a point estimate and the uncertainty around it. The mean difference is -0.7 mm, but the confidence interval extends across zero. Reading the estimate alone would omit information about how precisely the comparison is determined; reading the p-value alone would omit both the size and direction of the estimated contrast.

The record also shows why the analysis population and denominators matter. Enrollment is listed as 117, while the primary outcome groups contain 45 and 40 participants and the safety at-risk groups contain 57 and 60. These are distinct reported populations. Without participant flow and analysis-population definitions, readers cannot determine why they differ or how the differences affect interpretation.

A further statistical lesson is the distinction between a trial’s original design and retrospective planning illustrations. The registry does not report the original sample-size rationale, while the planning figures are based on the observed effect under stated assumptions. Those illustrations show how planning requirements change with the effect size chosen, but they do not demonstrate the power achieved by this trial.

Finally, a randomized comparison can still be difficult to assess fully when procedural and analysis details are not posted. Masking was absent, and the registry does not describe allocation concealment, missing-data handling, or the complete analysis plan. Identifying these limitations helps keep the interpretation aligned with the actual information recorded, without treating missing reporting as evidence that a particular problem occurred.

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.

The registry record leaves several trial-specific details unreported. It does not give the original sample-size justification, random-sequence generation method, allocation concealment procedure, or detailed protocol for outcome assessment. Masking is listed as none. The absence of masking can matter when assessments involve judgment, but the record does not describe the measurement process enough to determine whether or how assessment could have been influenced.

The primary outcome population is identified as PP, but its detailed definition is not reported. Enrollment is 117, whereas the posted primary outcome groups are 45 and 40. Participant flow, reasons for exclusions, and handling of missing 6-month measurements are not reported in the registry record. Therefore, the extent to which the endpoint results represent all enrolled participants cannot be determined from the posted information.

The registry reports one outcome measure and no posted secondary outcomes or analyses, despite describing broader evaluation of complementary bony and soft-tissue parameters. It does not provide a complete statistical analysis plan, the test or model details, multiplicity procedures, or sensitivity analyses. The trial’s prespecified threshold for a clinically important difference is also not reported.

For safety, the record gives event counts and at-risk denominators but does not describe the observation period, event attribution process, or detailed event follow-up. Death counts are not reported in the registry record. These limitations affect how fully the posted outcome and safety summaries can be evaluated; they do not demonstrate that the trial’s conduct or findings were biased.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Comparison of a PEG Membrane and a Collagen Membrane for the Treatment of Bone Dehiscence Defects at Bone Level Implants — NCT01012921.
• Registry record last updated 2018-01-02; status last verified 2017-05.
• Results first posted 2018-01-02.
• 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.

21. Record Summary

The registry record describes a completed randomized, unmasked, parallel-group comparison of MembraGel and Bio-Gide® membrane for bone dehiscence defects at bone-level implants. Overall enrollment was 117. The primary outcome was mean change in bone fill from baseline at regenerative therapy to the 6-month assessment, reported for a PP population.

The posted means were 3.2 mm for Bio-Gide® and 2.5 mm for MembraGel, with a mean difference of -0.7 mm for MembraGel compared with Bio-Gide®. The 95% confidence interval includes zero, and the unadjusted p-value was 0.075. The registry lists one outcome measure, no posted analyses, and no posted secondary outcomes. The original sample-size justification and detailed statistical plan are not reported in the registry record.