Augment® Bone Graft (Formerly GEM OS™1 Bone Graft) Compared to Autologous Bone Graft in Foot and Ankle Fusions: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Augment® Bone Graft (Formerly GEM OS™1 Bone Graft) Compared to Autologous Bone Graft in Foot and Ankle Fusions (NCT00583375), a randomized na study. The primary outcome is Subjects Fused at 24 Weeks (as Determined by CT Assessment).

2. Clinical Question

The primary outcome classified a participant as fused only when all joints had at least 50% osseous bridging on CT assessment. The registry reports 85 of 137 participants in Group 1 and 159 of 260 in Group 2 as fused. Its reported risk difference compares the fusion proportion for Group 2 with that for Group 1. This comparison describes the posted outcome data; it does not by itself establish equivalence or non-inferiority.

3. Trial Design

This was a randomized, single-masked, parallel-group trial. The official title describes it as prospective, controlled, multicenter and pivotal. The registry lists the study phase as NA and identifies BioMimetic Therapeutics as the sponsor. It reports an enrollment of 414 and a status of completed. The listed start is 2007-04, and primary completion is 2010-01.

Participants were assigned to standard rigid fixation plus autologous bone graft in Group 1 or standard rigid fixation plus Augment® Bone Graft in Group 2. The allocation is recorded as randomized, but the randomization procedure, allocation concealment, and the identity of those masked are not reported in the registry record. The intervention entries identify Augment® Bone Graft as a device and Standard of Care as a procedure.

The study objective was to demonstrate equivalent clinical and radiologic outcomes to autologous bone graft in a representative clinical model of hindfoot and ankle fusions. The stated hypothesis described Augment® Bone Graft as an equivalent substitute under non-inferiority analysis. Although this indicates the broad comparison the trial was intended to address, the registry record does not report the non-inferiority margin or the original statistical design assumptions.

The primary outcome was fusion at 24 weeks, determined by CT assessment. The primary analysis population is described as mITT, with 397 patients after 17 subjects were excluded post-operatively from the 414-patient Safety group. The registry does not explain those exclusions in the outcome description. Five secondary outcomes are named: Pain on Weight Bearing, Pain at Fusion Site, Foot Function Index (FFI), AOFAS Hindfoot and Ankle Score, and SF-12 Physical Component Score. Results for these outcomes are not described in the primary comparison.

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: 17 subjects excluded post-operatively

Overall Study

Analysis population for the primary outcome, as defined in the registry: The mITT population consisted of 397 patients (414 patients in the Safety group minus 17 subjects excluded post-operatively).

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: The safety population consisted of 414 patients, 272 randomized to Augment® Bone Graft and 142 randomized to autograft. The mITT population, submitted as the primary effectiveness analysis for the radiographic evaluation of bridging bone, consisted of 397 patients (414 patients in the Safety group minus 17 subjects excluded post-operatively).

6. Primary Endpoint

The registry lists the primary outcome as Subjects Fused at 24 Weeks (as Determined by CT Assessment), measured over 24 weeks.

The primary endpoint is the count of participants classified as fused at 24 weeks, based on CT assessment. An independent radiologist examined CT images for all patients’ joints and determined whether each joint had at least 50% osseous bridging. A participant was classified as fused only when all of that participant’s joints met the threshold. Thus, the endpoint is a participant-level binary outcome: fused or not fused under the stated definition.

CT offers an imaging-based way to assess osseous bridging at the fusion site. Using a prespecified threshold makes the outcome definition explicit, while requiring all joints to meet that threshold makes the participant-level classification dependent on every joint assessed for that person. The independent radiologist assessment is part of the registry’s description of how the outcome was determined; details about reader masking or adjudication are not reported.

The endpoint is suitable for a comparison of the proportion classified as fused in each treatment group. The registry gives the assessment time as 24 weeks and reports counts of fused participants and group sizes. The result therefore concerns this particular CT-defined endpoint at that time, not every aspect of recovery or longer-term clinical outcome. Pain, function, and health status were listed as secondary outcomes, but their results are not part of the primary fusion comparison.

7. Results Posted in the Registry

• Group 2: 61.2% (159/260) versus 62.0% (85/137) on control; risk difference -0.9 percentage points (95% CI -10.9 to 9.2), risk ratio 0.99 (0.84–1.16); unadjusted two-proportion z-test p = 0.862.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is binary because each participant is classified as either fused or not fused according to the CT definition. For a binary outcome, the basic group summaries are the number of participants with the event and the number assessed in each group. The registry reports 85 fused participants among 137 in Group 1 and 159 among 260 in Group 2. The posted proportions are 0.620 and 0.612, respectively. These summaries describe the observed results in the mITT population.

The risk difference expresses the absolute contrast between the two proportions. The reported value is -0.009 for Group 2 compared with Group 1. A negative value means the observed proportion was lower in Group 2 under this ordering of the comparison; the estimate is close to zero. The risk difference is useful because it keeps the comparison on the same scale as the outcome proportions. It does not, by itself, say whether a difference is clinically acceptable. That judgment would require a prespecified criterion, and the non-inferiority margin is not reported in the registry record.

A confidence interval gives a range of values compatible with the observed data under the statistical method and assumptions used. The interval for the risk difference is -0.110 to 0.092. Because this range includes zero, the data do not rule out no difference at the corresponding confidence level. It also includes values on both sides of zero, so the result remains compatible with either direction of difference. A confidence interval is not a guarantee that the true effect lies in the interval, and it is not a probability statement about the hypothesis.

The unadjusted p-value is 0.862. In general, a p-value measures how unusual data at least as discrepant as those observed would be if a specified null hypothesis and model assumptions held. It does not measure the size or importance of an effect, and it does not establish equivalence. For a non-inferiority objective, the decision is generally tied to a prespecified margin and a confidence bound relative to that margin. Because the margin and full inferential plan are not reported, the registry results cannot be used alone to reconstruct a formal non-inferiority decision.

The relative measures also express the treatment contrast on different scales. The risk ratio is reported as 0.986, with an interval from 0.838 to 1.16. A risk ratio compares the event proportions, with a value of 1 representing equal proportions. The odds ratio is 0.963, with an interval from 0.629 to 1.474; an odds ratio compares the odds of the event, with a value of 1 representing equal odds. Both confidence intervals include their null value of 1. These measures complement the absolute risk difference, but their point estimates and intervals should not be interpreted as evidence of equivalence.

Randomization is intended, in general, to balance measured and unmeasured factors across assigned groups on average. The record identifies the allocation as randomized, but does not describe how the sequence was generated or concealed. Single masking is recorded, but the masked party or parties are not identified. These omissions limit the detail available for assessing the conduct of the trial from the registry record.

The primary endpoint was assessed using CT images reviewed by an independent radiologist. In general, independent outcome assessment can reduce the influence of treatment expectations on outcome classification, but the record does not state whether the radiologist was masked to group assignment, how disagreement was handled, or whether images were centrally adjudicated. The stated rule requires all of a participant’s joints to have at least 50% osseous bridging. That rule defines the endpoint clearly, but the number and type of joints evaluated per participant are not detailed in the posted endpoint description.

The primary population is described as mITT and consists of 397 patients, which the record defines as 414 Safety-group patients minus 17 subjects excluded post-operatively. The reasons for exclusion, handling of missing CT assessments, and any imputation methods are not reported in the registry record. General statistical practice distinguishes an analysis population from the randomized population and considers how exclusions may affect comparability; interpreting this mITT result therefore depends partly on details that are absent from the record.

The registry reports five secondary outcomes related to pain, foot and ankle function, and physical health status. Their names are listed, but their results and analysis methods are not described in the available result record. When many outcomes are tested, the chance of at least one apparently unusual finding can increase; general approaches to this issue include prespecifying a primary outcome or adjusting for multiple comparisons. Whether any multiplicity strategy was used is not reported. The registry also does not report the original sample-size justification, interim analyses, or a complete statistical analysis plan.

10. Confidence Intervals and Precision

Precision concerns how narrowly the data estimate the treatment contrast. For the primary risk difference, the point estimate is -0.009 and the 95% confidence interval extends from -0.110 to 0.092. The estimate is close to zero, but the interval covers a meaningful range of possible differences in both directions. This means the observed data do not identify the size of the group contrast precisely.

The interval includes zero, which is consistent with no difference under the measure used. It also includes negative and positive values, so it is compatible with the observed proportion in Group 2 being lower or higher than in Group 1. The interval should not be read as proof that all values within it are equally plausible, nor as a statement that there is a 95% probability that the true risk difference falls inside it. It describes uncertainty under the method’s assumptions.

The relative measures likewise have intervals that include their null value: the risk ratio is 0.986, with a confidence interval from 0.838 to 1.16, and the odds ratio is 0.963, with a confidence interval from 0.629 to 1.474. The interval widths indicate that the relative contrast is not estimated with high precision. Different effect scales answer different descriptive questions, so the absolute and relative results are complementary rather than interchangeable.

For a non-inferiority question, precision must be considered against a clinically and statistically prespecified margin. The registry record does not report that margin. Consequently, the confidence interval cannot be assessed against the trial’s intended non-inferiority criterion from the record alone. The uncertainty is therefore both about the size of the observed contrast and about how that contrast relates to the intended decision threshold.

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 events. They illustrate sample sizes under the stated planning assumptions: 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. The trial’s own original design assumptions and sample-size justification are not reported in the registry record.

For planning for 100% of the observed effect, the illustration gives 46,883 participants per arm for 80% power and 62,763 participants per arm for 90% power. For planning for 75% of the observed effect, it gives 83,257 participants per arm for 80% power and 111,457 participants per arm for 90% power. For planning for 50% of the observed effect, it gives 187,122 participants per arm for 80% power and 250,502 participants per arm for 90% power.

In general, power is the probability that a study will detect an effect of the specified size under the planning assumptions if that effect is present. The 80% and 90% columns describe different target probabilities for detecting the chosen effect. Smaller effects are harder to distinguish from random variation, so they require more participants under these assumptions. These are planning illustrations calculated from the observed effect, not observed power for the completed 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, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.620; treatment proportion p₁ = 0.612; average p̄ = 0.616.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6878 = 1.3481.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6878 = 0.5789.
4. n = (1.3481 + 0.5789)² / (-0.0089)² = 46,882.7, rounded up to 46,883 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 <- 0.62043795620438                          # control proportion
p1 <- 0.611538461538462                          # 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(583375)
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 = 46,883 (80% power), 62,763 (90% power)
 75% of observed effect: n per arm = 83,257 (80% power), 111,457 (90% power)
 50% of observed effect: n per arm = 187,122 (80% power), 250,502 (90% power)
power.prop.test: 46882.7 
Simulated power at n = 46,883 per arm: 0.806 (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. Multiplicity, Missing Data and Analysis Populations

The registry names one primary outcome and five secondary outcomes. The primary outcome is fusion at 24 weeks by CT assessment. The secondary outcomes are Pain on Weight Bearing, Pain at Fusion Site, Foot Function Index (FFI), AOFAS Hindfoot and Ankle Score, and SF-12 Physical Component Score. The record lists six outcome measures in total, but the secondary results are not described in the result information considered for the primary comparison.

When several outcomes or comparisons are examined, multiplicity can increase the chance of obtaining at least one nominally significant result by chance. In general, a statistical plan may address this through a prespecified hierarchy, an adjustment, or by treating secondary analyses as exploratory. The registry record does not report whether multiplicity was addressed, which outcomes were formally tested, or how secondary analyses were intended to be interpreted. No conclusion about the secondary outcomes should be inferred from the primary fusion result.

The record states that the mITT population consisted of 397 patients: 414 patients in the Safety group minus 17 subjects excluded post-operatively. It does not give the reasons for those exclusions, identify which participants had missing CT assessments, or state how missing endpoint data were handled. No imputation method or sensitivity analysis is reported in the registry record. These details matter because exclusions and missing outcomes can affect the analyzed group and the interpretation of a binary endpoint.

The 397-person mITT population is the stated population for the primary endpoint, while the adverse-event denominators are based on Safety-group counts. The record does not describe whether the randomized groups were analyzed as assigned for the primary result or provide a separate per-protocol analysis. The full analysis populations, rules for protocol deviations, and statistical analysis plan are not reported. The registry also does not report interim analyses or the trial’s original sample-size justification. Those omissions prevent a complete assessment of prespecification and the relationship between the posted comparison and the planned analysis.

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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: acute myocardial infarction (2.1%), deep vein thrombosis (2.1%), gastrointestinal haemorrhage (1.4%), osteomyelitis (1.4%), pulmonary embolism (1.1%).

The safety summaries report serious adverse events and other adverse events by group, using the stated at-risk and affected counts. In Group 1, 21 of 142 participants were affected by a serious adverse event, and 43 of 142 were affected by another adverse event. In Group 2, 28 of 272 participants were affected by a serious adverse event, and 100 of 272 were affected by another adverse event. The record reports 0 deaths in Group 1 and 1 in Group 2.

These are counts of participants affected, not counts of every event episode. The denominators differ between groups and are safety populations, so they should not be substituted for the primary mITT denominators. The registry lists Acute myocardial infarction and Deep vein thrombosis at 2.1% each; Gastrointestinal haemorrhage and Osteomyelitis at 1.4% each; and Pulmonary embolism and Pain in extremity at 1.1% each among the top serious-event terms. The listed terms are not attributed to a treatment group in this summary.

The summaries do not establish whether an event was caused by treatment. The registry information stated does not give event-level timing, severity beyond the serious classification, relatedness assessments, or details about the death. Safety counts should therefore be read descriptively and separately from the efficacy endpoint.

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

The posted primary result describes CT-defined fusion at 24 weeks in the mITT population. The observed fusion proportions were 0.620 in Group 1 and 0.612 in Group 2, with a reported risk difference of -0.009 for Group 2 compared with Group 1. The confidence interval, -0.110 to 0.092, indicates considerable uncertainty about the size and direction of the difference. The unadjusted p-value is 0.862.

These data show that the observed proportions were close, but closeness of point estimates is not the same as demonstrated equivalence. A finding of non-inferiority requires a prespecified margin and an analysis showing that the result meets that criterion. The registry record states a non-inferiority hypothesis but does not report the margin. Therefore, the posted numbers alone do not establish that Augment® Bone Graft is equivalent or non-inferior to autologous bone graft.

The result applies to the defined endpoint, population, and assessment time. It concerns participants classified as fused under the all-joints CT rule at 24 weeks. It does not by itself describe longer-term fusion, pain, function, physical health status, or other outcomes. Although secondary outcomes are named in the record, their results are not included in this primary comparison.

The result also does not establish that either treatment is preferable, and it does not determine the cause of the adverse events or death recorded in the safety summaries. The trial record gives descriptive safety counts, but the information does not support causal conclusions. Overall, the primary data provide an estimate of the between-group contrast with an interval expressing uncertainty; the missing non-inferiority margin and analysis details limit the strength of any formal conclusion about the trial’s stated hypothesis.

16. Why This Trial Matters Statistically

This trial illustrates why a near-zero point estimate and a non-significant p-value do not, by themselves, establish equivalence. The reported risk difference is -0.009 and the unadjusted p-value is 0.862, but the confidence interval extends from -0.110 to 0.092. That interval expresses uncertainty around the estimated contrast, rather than confirming that the treatments have the same outcome.

The distinction is especially important because the stated objective was non-inferiority. A non-inferiority analysis depends on a margin chosen before interpreting the results. The margin defines how much difference is acceptable for the study’s purpose. Without the margin, readers cannot judge whether the confidence interval excludes differences beyond that threshold. This is a general feature of non-inferiority interpretation, and the registry record does not provide the study-specific margin.

The endpoint definition also matters. Fusion was assessed at the participant level, and the participant had to meet the CT bridging threshold in all joints to be classified as fused. The resulting binary outcome condenses imaging findings into a clear group comparison, but does not cover the full range of pain, function, or health-status outcomes named in the registry.

Finally, the record distinguishes the mITT group used for the primary outcome from the Safety group used for adverse-event summaries. The stated post-operative exclusions and absence of details about missing endpoint data make analysis-population reporting important to interpretation. Together, these features show why effect estimates, uncertainty, endpoint definitions, and trial-analysis details need to be considered alongside the p-value.

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 unavailable. Although the stated hypothesis is non-inferiority, the non-inferiority margin is not reported, so the posted confidence interval cannot be evaluated against the intended decision threshold. The original sample-size justification and design assumptions are also not reported. The planning illustration based on the observed effect is not a substitute for the trial’s original design rationale.

The mITT population is described as 397 patients after 17 post-operative exclusions from the 414-person Safety group, but the reasons for exclusion and the groups from which those participants came are not stated in the endpoint description. The handling of missing CT assessments, protocol deviations, and any imputation is not reported. These details constrain assessment of how representative the primary analysis is of the randomized participants.

The registry identifies randomization and single masking but does not describe sequence generation, allocation concealment, or who was masked. Although an independent radiologist assessed CT images, the record does not state whether the radiologist was masked to treatment assignment or how assessment disagreements were handled. Such details would help readers evaluate possible sources of bias in outcome classification.

The primary comparison is unadjusted, and no additional analyses are listed in the registry record. The statistical analysis plan, any interim analyses, and any multiplicity strategy are not reported. The named secondary outcomes do not have results described in the primary comparison. Safety summaries provide affected counts and selected serious-event terms, but do not provide event-level context or establish treatment relatedness. These limitations restrict conclusions to the posted endpoint summaries and comparison.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Augment® Bone Graft (Formerly GEM OS™1 Bone Graft) Compared to Autologous Bone Graft in Foot and Ankle Fusions — NCT00583375.
• Registry record last updated 2018-04-02; status last verified 2018-03.
• Results first posted 2018-02-23.
• 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

NCT00583375 is a completed randomized, single-masked, parallel-group trial sponsored by BioMimetic Therapeutics. The official title describes a prospective, controlled, multicenter pivotal clinical trial comparing Augment® Bone Graft with autologous bone graft in foot and ankle fusions. Enrollment was 414; the listed start was 2007-04 and primary completion was 2010-01.

The primary outcome was CT-defined fusion at 24 weeks. The mITT population included 397 patients, and the reported risk difference for Group 2 versus Group 1 was -0.009, with a 95% confidence interval from -0.110 to 0.092 and an unadjusted p-value of 0.862. The record states a non-inferiority hypothesis, but does not report the margin or the full analysis plan. It also lists five secondary outcomes and provides safety summaries using separate denominators.