This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Safety and Efficacy of the Oxford Cementless Partial Knee System (NCT01473134), a randomized na study. The primary outcome is Radiographic Success.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT01473134 |
| Phase | NA |
| Condition | Osteoarthritis, Knee, Avascular Necrosis of the Medial Femoral Condyle |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Single |
| Enrollment | 320 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Zimmer Biomet |
| Results posted | Yes |
2. Clinical Question
The study asks whether radiographic outcomes differ between two ways of implanting the Oxford Partial Knee: without bone cement or with bone cement. It also records clinical assessments and patient-reported outcomes as secondary measures. The registry describes the investigation as evaluating safety and effectiveness, but the primary endpoint specified in the record is radiographic success at the knee level.
Population
Participants had osteoarthritis of the knee or avascular necrosis of the medial femoral condyle. The primary-endpoint analysis population was per protocol and required a radiographic assessment at the 2+ year follow-up interval. The analysis unit was the knee, including two separately counted knees for participants with bilateral study knees.
Intervention
The Oxford Cementless Partial Knee was implanted without bone cement. The registry lists this as a device intervention.
Comparator
The Oxford Cemented Partial Knee was applied with bone cement and served as the control treatment described in the study summary.
Primary question
At the 2-year timeframe, how many assessed knees met the registry’s radiographic-success criteria, and how did the results compare between the cementless and cemented systems? The registry record does not report the primary-endpoint counts or a statistical comparison, so the relative radiographic results cannot be determined from the posted results.
The randomized, parallel-group design addresses a comparison between these two device strategies. Randomization is intended, in general, to reduce systematic differences between treatment groups at assignment; it does not by itself establish that the observed results are similar or different. The record identifies single masking, but does not identify which people were masked. The study’s stated enrollment was 320.
3. Trial Design
NCT01473134 is a completed, randomized, single-masked, parallel-group clinical study sponsored by Zimmer Biomet. The registry lists the study phase as NA. The recorded start was 2011-12, and primary completion was 2020-11-30. Enrollment was 320. These design and status details describe the study structure, not the size or direction of any treatment effect.
The two arms differed in whether bone cement was used with the Oxford Partial Knee. The cementless device was implanted without bone cement; the cemented device was applied with bone cement. The study summary says the investigation was designed to document and compare clinical and radiographic results. It describes the cemented system as the control treatment.
The primary outcome was radiographic success, assessed at the knee level at the 2-year timeframe. The endpoint description specifies assessment at the 22+ month interval. The primary analysis population was per protocol and consisted of knees with a radiographic assessment at the 2+ year follow-up interval. Because the analysis counted knees rather than participants, bilateral participants could contribute two individual knees. This distinction matters: a knee count is not automatically a participant count, and two knees from one participant may not be statistically independent.
Listed secondary outcomes include Knee Society Assessment Score, Knee Society Score Assessment for Bilateral Subjects, Knee Society Function Score at All Timepoints, Knee Society Function Score for Bilateral Subjects, Oxford Knee Score, the number of knees meeting radiographic-success criteria, and Patient Assessment of Satisfaction. The registry lists 11 outcome measures overall. It does not report the statistical analysis details for those measures, the masking identity, or a sample-size justification. The record also does not report how the randomized allocation was generated or concealed.
The study status is completed, but completion status alone does not provide outcome results. The registry record contains no listed analyses for the primary endpoint. Consequently, the design can be described, but an estimate of the comparative radiographic effect cannot be recovered from the available endpoint information.
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: ELK1078R - bilateral subject was randomized to Investigational arm and received the Cementless device during the initial surgery. During the contralateral surgery, the cemented keel saw was used therefore, the Control device was implanted. Because this subject was randomized to Investigational and received 1 control knee, the investigational count in the participant flow is off by 1. A row was added with a note for this subject.
Overall Study
| Stage | Oxford Cementless Partial Knee | Oxford Cemented Partial Knee |
|---|---|---|
| STARTED | 207 | 113 |
| COMPLETED | 152 | 78 |
| NOT COMPLETED | 55 | 35 |
Reasons for not completing this period:
| Reason | Oxford Cementless Partial Knee | Oxford Cemented Partial Knee |
|---|---|---|
| Lost to Follow-up | 21 | 13 |
| Protocol Violation | 32 | 19 |
| Withdrawal by Subject | 1 | 2 |
| Death | 0 | 1 |
| ELK1078R: bilateral subject randomized to investigational, 2nd knee was cemented. | 1 | 0 |
Analysis population for the primary outcome, as defined in the registry: The analysis was performed on the per protocol population who had a radiographic assessment at the 2+ year follow up interval. The analysis is at the knee level, not participant level (bilateral participants included as two individual knees).
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.
| Characteristic | Unit | Oxford Cementless Partial Knee | Oxford Cemented Partial Knee | Total |
|---|---|---|---|---|
| Participants analysed | 207 | 113 | 320 | |
| Age, Continuous | years | 65.8 (SD 8.7) | 66.4 (SD 8.5) | 66.04 (SD 8.61) |
| Sex: Female, Male: Female | Participants | 81 | 54 | 135 |
| Sex: Female, Male: Male | Participants | 126 | 59 | 185 |
| Race and Ethnicity Not Collected | Participants | — | — | 0 |
| Region of Enrollment: United States | Participants | 207 | 113 | 320 |
6. Primary Endpoint
The registry lists the primary outcome as Radiographic Success, measured over 2 years.
| Element | Registry entry |
|---|---|
| Result type | Count Of Units |
| Measure of spread | not applicable |
| Unit | knees |
| Endpoint classification | not reduced to a single comparison |
Radiographic success is a binary knee-level outcome: each assessed knee is classified as a success or failure. To qualify as a success at the 22+ month interval, the knee must have no osteolysis, no migration or subsidence of any femoral or tibial component, and no component fracture. The failure definition includes progressive radiolucency greater than 3 mm at its maximum interface in two or more contiguous zones, or a progressive bony destructive lesion; migration or subsidence greater than 3 mm compared with 6 week radiographs; or a component fracture.
This endpoint focuses on radiographic signs of implant or bone change rather than directly measuring pain, mobility, or satisfaction. In general, a binary endpoint turns a set of clinical criteria into a count of units meeting or not meeting a defined threshold. It can make the outcome definition explicit, but it also compresses potentially different findings into the same success/failure classification. A knee that meets all success criteria and one that fails by different routes are represented by their category, not by a graded measure of severity.
The registry describes the timeframe as 2 years and the analysis population as per protocol knees with an assessment at the 2+ year interval. The reported primary outcome type is a count of units, with knees as the unit. The record does not report the number of successful or failed knees in either arm, nor a treatment-effect estimate or uncertainty interval. Thus, the endpoint definition is available, while the observed comparison is not reported in the registry record.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record establishes what the primary endpoint was intended to measure, but it does not provide the primary-endpoint results. Radiographic success was a knee-level binary classification at the 2-year timeframe, based on the absence of specified radiographic failures. The analysis population was per protocol and required a radiographic assessment at the 2+ year follow-up interval. No arm-specific counts of successful or failed knees are reported in the record, and no statistical comparison is listed.
For that reason, the available information does not establish whether radiographic success was more or less frequent with either device, or how precisely any difference was estimated. A numerical treatment effect, confidence interval, or p-value for this primary endpoint is not reported in the registry record. Without those results, the endpoint definition alone cannot support a conclusion about the comparative radiographic findings.
The record does contain safety counts by arm. For the Oxford Cementless Partial Knee group, 17 of 207 at-risk participants were affected by serious adverse events; 114 of 207 were affected in the other adverse-event category; and 3 deaths were recorded. For the Oxford Cemented Partial Knee group, 10 of 114 at-risk participants were affected by serious adverse events; 53 of 114 were affected in the other adverse-event category; and 9 deaths were recorded. These are descriptive counts. They do not, by themselves, show whether an event was caused by the device or whether the groups differ after accounting for follow-up, baseline characteristics, or other relevant factors. The record does not report an inferential comparison for these safety counts.
The registry also lists serious adverse-event terms with percentages, including Other Ipsilateral Knee Related Adverse Event and Tibial Fracture at 1.9% each, Bearing Dislocation/Disassociation and Pain (progressive/persistent) at 1.8% each, and Effusion/Swelling/Edema and Hematoma at 1.0% each. These percentages are presented as posted; the registry entries available in the record do not provide enough information to derive further comparisons or interpret their denominators.
Randomization is a design feature, not a substitute for results. Likewise, a completed study and a specified endpoint do not establish an effect. The posted information supports a description of the question, endpoint, and reported adverse-event counts, but not a conclusion about comparative radiographic success or overall benefit.
9. Statistical Methods Explained
What does a binary endpoint mean?
In general, a binary endpoint assigns each unit to one of two categories. For this study, the unit was the knee and the categories were radiographic success and failure. The endpoint definition specifies the findings that determine those categories: success requires absence of osteolysis, component migration or subsidence, and component fracture; failure is defined through the stated radiographic criteria. A binary classification can be straightforward to interpret when the criteria are applied consistently. It does not preserve all detail about the type or degree of a finding, because the final result is a category rather than a continuous measurement.
The outcome is knee-level, not participant-level. The registry explicitly includes bilateral participants as two individual knees. In general, observations from two knees belonging to one participant may be related. A statistical analysis can account for that relationship when appropriate, but the method used to handle bilateral knees is not reported in the registry record. Without that detail, the independence assumptions behind any unreported comparison cannot be assessed.
What does the analysis population tell the reader?
The registry says the primary analysis used the per protocol population with a radiographic assessment at the 2+ year follow-up interval. This identifies which knees were described as eligible for the analysis, but it does not explain how many randomized knees lacked an assessment, why any assessment was unavailable, or how missing observations were handled. A per protocol population is not interchangeable with every randomized participant or knee. General interpretation therefore depends on knowing how the analyzed set relates to the randomized groups, and that flow is not reported in the registry record.
The endpoint timeframe is listed as 2 years, with the success criteria assessed at the 22+ month interval. The failure definition for migration or subsidence compares component position with 6 week radiographs. These time references are part of the endpoint definition; they do not supply the outcome counts or establish the duration of observation for every knee.
How should a treatment comparison be read?
In general, a comparison of binary outcomes may report each group’s number of successes and failures, an absolute or relative effect measure, and an uncertainty interval. A confidence interval describes a range of values compatible with the data and analysis assumptions; it communicates precision as well as direction. A p-value, when used, addresses how compatible the data are with a specified null model, but does not measure clinical importance or the probability that a hypothesis is true. None of these primary-endpoint comparison statistics is reported in the registry record.
Random allocation is intended to make treatment assignment independent of measured and unmeasured baseline characteristics on average. It does not guarantee identical groups in a particular study, and it does not address missing outcome data or departures from the assigned intervention. The record identifies randomized allocation but does not report the sequence-generation or concealment methods, or a detailed analysis plan. These details should not be assumed.
What can be inferred from the safety counts?
The safety record gives counts of participants at risk and affected, as well as deaths, separately for each arm. Such counts are descriptive summaries. Comparing counts alone can be misleading when the groups differ in size or exposure time; a formal comparison would need a defined estimand and suitable methods. This is a general statistical point, not a claim that a particular adjustment was or was not warranted in this study. The registry record does not report inferential safety analyses, exposure-time summaries, or attribution methods for the listed counts.
Serious adverse-event terms are also listed with percentages. A term-level percentage is a descriptive value, not automatically a comparative risk estimate. To interpret it fully, readers would need the relevant denominator, event-counting rules, and the way repeated events or participants with multiple events were handled. Those details are not reported in the registry record as described in the available information.
What do the planning assumptions mean?
The stated planning illustration uses a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation. It is described as being based on a fraction of the observed effect. Such sample-size calculations are planning illustrations: they ask what sample size might be needed under specified assumptions. They are not observed power and do not establish whether the completed study detected an effect. The registry record does not report the trial’s own sample-size justification, the effect used in its design, or interim analyses.
The record lists 11 outcome measures but no analyses for the primary endpoint. Without reported outcome estimates and analysis details, readers cannot check model assumptions, the handling of bilateral knees, the extent of missing data, or the precision of the primary comparison. General statistical concepts can clarify what would be needed for interpretation, but cannot replace the absent result.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists 11 outcome measures, including the primary radiographic-success endpoint and several secondary clinical, functional, radiographic, and satisfaction outcomes. Testing or estimating multiple outcomes creates a multiplicity issue in general: as the number of separate statistical questions grows, the chance of at least one apparently notable result can increase if each is assessed without an appropriate strategy. The registry record does not report a multiplicity adjustment, a hierarchy among secondary outcomes, or a prespecified strategy for interpreting multiple comparisons. It is therefore not possible to determine from the record whether or how multiplicity was addressed.
The registry reports zero listed analyses. The absence of listed analyses does not show that no analyses were conducted; it means the registry record does not provide analysis results in that field. No primary-endpoint counts, effect estimate, confidence interval, or p-value are reported. A sample-size justification and interim-analysis plan are also not reported in the registry record.
For the primary endpoint, the stated analysis population was per protocol knees with a radiographic assessment at the 2+ year follow-up interval. The record does not state how many knees were randomized, assessed, or excluded from that analysis by arm, or why data were missing. It does not report an imputation method, sensitivity analysis, or other missing-data procedure. These omissions limit assessment of whether the analyzed knees represent the randomized groups.
The analysis unit was the knee, and bilateral participants contributed two knees. The registry does not report whether the analysis accounted for within-participant dependence. That method can matter when two observations come from one person. The record also does not give enough information to assess adherence to assigned treatment or the relationship between the per protocol set and the randomized population. These are reporting limitations; no specific bias or analytic failure can be inferred from the missing details alone.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Oxford Cementless Partial Knee | 17 | 207 | 8.2% |
| Oxford Cemented Partial Knee | 10 | 114 | 8.8% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Oxford Cementless Partial Knee | 3 of 207 |
| Oxford Cemented Partial Knee | 9 of 114 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Oxford Cementless Partial Knee | 114 of 207 |
| Oxford Cemented Partial Knee | 53 of 114 |
Most frequent serious adverse event terms
| Event term | Organ system | Oxford Cementless Partial Knee | Oxford Cemented Partial Knee |
|---|---|---|---|
| Other Ipsilateral Knee Related Adverse Event | Injury, poisoning and procedural complications | 4 of 207 | 1 of 114 |
| Bearing Dislocation/Disassociation | Injury, poisoning and procedural complications | 2 of 207 | 2 of 114 |
| Tibial Fracture | Injury, poisoning and procedural complications | 4 of 207 | 0 of 114 |
| Pain (progressive/persistent) | Injury, poisoning and procedural complications | 1 of 207 | 2 of 114 |
| Subsidence | Injury, poisoning and procedural complications | 2 of 207 | 1 of 114 |
| Effusion/Swelling/Edema | Musculoskeletal and connective tissue disorders | 2 of 207 | 0 of 114 |
| Hematoma | Vascular disorders | 2 of 207 | 0 of 114 |
| Avascular Necrosis | Musculoskeletal and connective tissue disorders | 1 of 207 | 0 of 114 |
| Cellulitis/Redness/Blistering | Skin and subcutaneous tissue disorders | 0 of 207 | 1 of 114 |
| Deep Wound Infection | Infections and infestations | 1 of 207 | 0 of 114 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Oxford Cementless Partial Knee | Oxford Cemented Partial Knee |
|---|---|---|---|
| Pain (progressive/persistent) | General disorders | 50 of 207 | 22 of 114 |
| Other Ipsilateral Knee Related Adverse Event | Injury, poisoning and procedural complications | 36 of 207 | 18 of 114 |
| Effusion/Swelling/Edema | Musculoskeletal and connective tissue disorders | 29 of 207 | 9 of 114 |
| Traumatic Injury (Study Knee) | Injury, poisoning and procedural complications | 16 of 207 | 8 of 114 |
| Clicking/Popping/Crepitus/Grinding | Musculoskeletal and connective tissue disorders | 13 of 207 | 4 of 114 |
Serious events with the highest arm-level rate: other ipsilateral knee related adverse event (1.9%), tibial fracture (1.9%), bearing dislocation/disassociation (1.8%), pain (progressive/persistent) (1.8%), effusion/swelling/edema (1.0%).
For the Oxford Cementless Partial Knee group, the registry reports 207 participants at risk for serious adverse events, with 17 affected; 207 at risk in the other adverse-event category, with 114 affected; and 3 deaths. For the Oxford Cemented Partial Knee group, it reports 114 at risk for serious adverse events, with 10 affected; 114 at risk in the other adverse-event category, with 53 affected; and 9 deaths. These figures are counts, not adjusted comparisons.
The registry lists the following serious adverse-event terms and percentages: Other Ipsilateral Knee Related Adverse Event, 1.9%; Tibial Fracture, 1.9%; Bearing Dislocation/Disassociation, 1.8%; Pain (progressive/persistent), 1.8%; Effusion/Swelling/Edema, 1.0%; and Hematoma, 1.0%. The record does not provide a comparative analysis for these terms. They should be read as posted descriptive values, not as evidence of a difference between arms or proof that a device caused an event.
In general, adverse-event counts depend on how events were defined, collected, attributed, and followed over time. The registry information does not report those details or an inferential safety analysis. Counts from groups of different sizes should not be treated as directly comparable rates without an appropriate analysis.
12. What the Result Does, and Does Not, Mean
The registry record defines a randomized comparison between a cementless and a cemented Oxford Partial Knee system and specifies a radiographic primary endpoint. It also reports descriptive adverse-event counts by arm. Those details are useful for understanding the study question and the types of outcomes recorded, but they do not provide the primary result needed to assess the radiographic comparison.
In particular, no arm-specific number of knees meeting radiographic-success criteria is reported, and no comparative effect estimate or uncertainty interval is available in the registry record. The posted information therefore cannot establish whether one group had a higher or lower radiographic-success count, or how precise a difference might be. An endpoint definition is not an outcome result.
The adverse-event figures likewise describe counts and listed terms, rather than a complete comparative safety analysis. They do not, on their own, establish causation, account for differences in group size or follow-up, or quantify uncertainty. The listed deaths should be read as registry counts; the record does not report an analysis attributing them to either device.
More generally, randomization strengthens a study’s design for comparing assigned groups, but interpretation still depends on the actual outcome data, the analysis population, missing-data handling, and the analysis method. The registry identifies the primary analysis population as per protocol knees with a radiographic assessment at the 2+ year interval. It does not report the number missing from that assessment or the relationship of the analyzed set to all randomized knees.
The results described in the registry do not support a conclusion about the devices’ comparative effectiveness or safety. That is not evidence that the devices are equivalent, nor evidence that one is superior. It reflects that the relevant primary-endpoint results and inferential comparisons are not reported in the registry record.
13. Why This Trial Matters Statistically
This study illustrates why a clearly specified endpoint and a reported result are both necessary for statistical interpretation. The primary endpoint has a concrete definition: each knee is classified as a radiographic success or failure according to specified findings at the stated follow-up interval. That definition makes the intended outcome understandable, but the lack of arm-specific counts means readers cannot evaluate the comparison itself.
The knee-level unit also highlights a design and analysis issue common to studies where participants can contribute more than one joint. Bilateral participants were counted as two knees. In general, multiple observations from one participant can be correlated, so the analytical treatment of that dependence may affect uncertainty estimates. The registry does not report whether or how this was handled. Recognizing the unit of analysis is therefore important when interpreting both the endpoint and the absent analysis details.
The record also separates a primary radiographic measure from secondary measures such as Knee Society scores, Oxford Knee Score, and patient satisfaction. Multiple outcomes can answer different questions, but they also make transparent reporting of outcome priorities and multiplicity methods important. Those methods are not reported in the registry record.
Finally, the adverse-event counts show why descriptive safety data and comparative inference are not the same. Counts communicate how many affected participants were recorded in each group, while a statistical comparison also requires a defined approach to group size, follow-up, event definitions, and uncertainty. The registry provides some descriptive counts but no inferential safety analysis in the information available.
Statistically, the central lesson is that design details, endpoint definitions, and outcome estimates have distinct roles. Randomization describes assignment; the endpoint defines what is measured; and reported results quantify what was observed and with what precision. The first two are documented, while the primary comparison is not reported in the registry record.
14. 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 most important trial-specific limitation for interpretation is that the registry record does not report the primary radiographic-success counts or a statistical comparison. The endpoint definition and analysis population are stated, but there is no arm-specific result, effect estimate, confidence interval, or p-value for that endpoint. The comparative radiographic outcome therefore cannot be assessed from the posted information.
The record identifies the primary analysis as per protocol and restricted to knees with a radiographic assessment at the 2+ year follow-up interval. It does not report the number of knees assessed or excluded by arm, reasons for missing assessments, or methods used to address missing data. Without that information, the relationship between the analyzed population and the randomized population cannot be evaluated.
The endpoint is analyzed at the knee level, with bilateral participants represented by two knees. The registry does not state whether the statistical analysis accounted for correlation between knees belonging to the same participant. That omission prevents assessment of the method used to estimate uncertainty for the primary endpoint.
Other relevant methods are not reported in the registry record, including the sample-size justification, detailed randomization procedures, identity of the masked party, interim analyses, and a detailed statistical analysis plan. The record lists 11 outcome measures and does not describe how multiplicity was handled. These reporting gaps do not establish that the methods were inadequate; they limit what can be independently evaluated from the registry record.
Safety information is also limited to reported counts and selected serious adverse-event terms with percentages. No inferential safety comparison, exposure-time summary, or event-attribution method is reported in the information available. Counts alone cannot resolve whether any observed group difference reflects treatment, follow-up, or other factors.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Safety and Efficacy of the Oxford Cementless Partial Knee System — NCT01473134.
- Registry record last updated 2025-08-11; status last verified 2024-09.
- Results first posted 2024-09-19.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
NCT01473134 is a completed, randomized, single-masked, parallel-group study sponsored by Zimmer Biomet. It compared the Oxford Cementless Partial Knee, implanted without bone cement, with the Oxford Cemented Partial Knee, applied with bone cement. Enrollment was 320; the recorded start was 2011-12, and primary completion was 2020-11-30. The registry lists the phase as NA.
The primary endpoint was knee-level radiographic success at the 2-year timeframe, assessed at the 22+ month interval. The stated per protocol analysis population included knees with a radiographic assessment at the 2+ year follow-up interval; bilateral participants contributed two knees. The registry lists 11 outcome measures and no analyses.
Arm-specific adverse-event counts and selected serious adverse-event terms are reported. However, the primary endpoint counts, comparative treatment estimate, uncertainty interval, and inferential safety analysis are not reported in the registry record. As a result, the record describes the study design and endpoint but does not provide the primary comparative result.