CORMET HIP RESURFACING SYSTEM

P050016 · Corin USA · NXT · Jul 3, 2007 · Orthopedic

Device Facts

• Record ID: P050016
• Device Name: CORMET HIP RESURFACING SYSTEM
• Applicant: Corin USA
• Product Code: NXT · Orthopedic
• Decision Date: Jul 3, 2007
• Decision: APPR
• Device Class: Class 3
• Attributes: Therapeutic, Expedited Review

Intended Use

The Cormet Hip Resurfacing System is a single use device intended for hybrid fixation: cemented femoral head and cementless acetabular component. The Cormet Hip Resurfacing System is intended for use in resurfacing hip arthroplasty for reduction or relief of pain and/or improved hip function in skeletally mature patients having the following conditions: 1. non-inflammatory degenerative arthritis such as osteoarthritis and avascular necrosis; 2. inflammatory arthritis such as rheumatoid arthritis. The Cormet Hip Resurfacing System is intended for patients who, due to their relatively younger age or increased activity level, may not be suitable for traditional total hip arthroplasty due to an increased possibility of requiring ipsilateral hip joint revision.

Device Story

The Cormet Hip Resurfacing System is a metal-on-metal hip resurfacing prosthesis consisting of a stemmed femoral head (cemented) and an acetabular cup (cementless, plasma-sprayed titanium/hydroxyapatite coating). It is used in orthopedic surgery to resurface the hip joint, preserving more bone than traditional total hip arthroplasty. The device is implanted by an orthopedic surgeon in a clinical/hospital setting. The system provides an articulating bearing surface to reduce pain and restore function. Clinical outcomes are assessed via Harris Hip Scores and radiographic monitoring. The device is intended for younger, active patients who may require future revision, offering an alternative to total hip replacement. Potential benefits include improved hip function and pain relief; however, risks include femoral neck fracture, component loosening, and metal ion release.

Clinical Evidence

Prospective, multi-center IDE study (1148 procedures) compared to historical ceramic-on-ceramic total hip control (349 procedures). Primary endpoint: Composite Clinical Success (CCS) at 24 months (HHS ≥80, radiographic success, no revisions, no device-related AEs). Results: 86.0% CCS for Cormet vs 87.5% for control (difference -0.015, 95% CI lower bound -0.063, meeting non-inferiority delta of -0.08). Revision rate 7.9% (pivotal unilateral) vs 1.9% (control). Risk factors for revision identified: female gender, small component size (40/44mm), non-OA diagnosis, leg length discrepancy ≥1cm, low baseline HHS.

Technological Characteristics

Metal-on-metal bearing system. Femoral head: Cobalt Chromium alloy (ASTM F75). Acetabular component: Cobalt Chromium alloy (ASTM F75) with plasma-sprayed unalloyed titanium (ISO 5832-2) and hydroxyapatite (ASTM F1185) coating. Fixation: Cemented femoral, cementless acetabular. Sterilization: Gamma irradiation (25kGy minimum).

Indications for Use

Indicated for skeletally mature patients with non-inflammatory degenerative arthritis (osteoarthritis, avascular necrosis) or inflammatory arthritis (rheumatoid arthritis) requiring hip resurfacing for pain relief or improved function, particularly younger or active patients at risk for revision with traditional total hip arthroplasty. Contraindicated in patients with active/suspected hip infection, skeletal immaturity, inadequate bone stock (severe osteopenia/osteoporosis, AVN >50% femoral head, cysts >1cm), vascular insufficiency, muscular atrophy, neuromuscular disease, renal insufficiency, immunosuppression, severe obesity, metal sensitivity, or females of childbearing age.

Regulatory Classification

Identification

Prosthesis, hip, semi-constrained, metal/metal, resurfacing device relieves hip pain and improves hip function by replacing the parts of the hip that have been severely damaged by degenerative joint diseases. This device is indicated for patients with degenerative joint diseases including osteoarthritis, rheumatoid arthritis, traumatic arthritis, dysplasia, or avascular necrosis and for patients who, due to their relatively younger age or increased activity level, may not be suitable for traditional total hip replacement due to an increased possibility of requiring future hip joint revision.

Reference Devices

• Alumina ceramic total hip prosthesis (PMA approved)

Submission Summary (Full Text)

{0}

# SUMMARY OF SAFETY AND EFFECTIVENESS

## I. GENERAL INFORMATION

| Device Generic Name: | Prosthesis, Hip, Semi-constrained, Resurfacing Metal/Metal hybrid fixation |
| --- | --- |
| Device Trade Name: | Cormet Hip Resurfacing System |
| Applicant Name and Address: | Corin USA
10500 University Center Drive, Suite 190
Tampa, FL 33612 |
| Premarket Approval Number: | P050016 |
| Date of Panel Recommendation: | February 22, 2007 |
| Date of Notice of Approval to Applicant: | July 3, 2007 |

## II. INDICATIONS FOR USE

The Cormet Hip Resurfacing System is a single use device intended for hybrid fixation: cemented femoral head and cementless acetabular component. The Cormet Hip Resurfacing System is intended for use in resurfacing hip arthroplasty for reduction or relief of pain and/or improved hip function in skeletally mature patients having the following conditions:

1. non-inflammatory degenerative arthritis such as osteoarthritis and avascular necrosis;
2. inflammatory arthritis such as rheumatoid arthritis.

The Cormet Hip Resurfacing System is intended for patients who, due to their relatively younger age or increased activity level, may not be suitable for traditional total hip arthroplasty due to an increased possibility of requiring ipsilateral hip joint revision.

## III. CONTRAINDICATIONS

1. Patients with active or suspected infection in or around the hip joint;
2. Patients who are skeletally immature;
3. Patients with bone stock inadequate to support the device including:

- Patients with severe osteopenia should not receive the Cormet Hip Resurfacing System procedure. Patients with a family history of severe osteoporosis or severe osteopenia.
- Patients with osteonecrosis or avascular necrosis (AVN) with >50% involvement of the femoral head (regardless of Ficat Grade) should not receive a Cormet Hip Resurfacing device.
- Patients with multiple cysts of the femoral head (>1cm) should not receive a Cormet Hip Resurfacing device.
- Note – In cases of questionable bone stock, a Dual Energy X-Ray Absorptiometry (DEXA) scan may be necessary to assess inadequate bone stock.

{1}

4. Patients with any vascular insufficiency, muscular atrophy, or neuromuscular disease severe enough to compromise implant stability or postoperative recovery.
5. Females of child bearing age due to unknown effects on the fetus of metal ion release.
6. Patients with known moderate or severe renal insufficiency.
7. Patients who are immunosuppressed with diseases such as AIDS or persons receiving high doses of corticosteroids.
8. Patients who are severely overweight.
9. Patients with known or suspected metal sensitivity (e.g., jewelry).

## IV. WARNINGS AND PRECAUTIONS

1. Patients on medications (such as high-dose or chronic amino glycoside treatment) or with comorbidities (such as diabetes) that increase the risk of future, significant renal impairment should be advised of the possibility of increase in systemic metal ion concentration. Preoperative and postoperative monitoring of renal function (such as creatinine, Glomerular Filtration Rate (GFR), Blood Urea Nitrogen (BUN)) will be necessary.
2. Currently, Corin does not have a commercially available modular femoral head for use with the Cormet resurfacing shell. If the Cormet resurfacing head must be revised to a total hip arthroplasty, the acetabular shell should also be revised even if it is well fixed.
3. Based on the analysis of a multicenter prospective study of 1030 patients in 14 centers the following were identified as risk factors for revision: Patients who are female, who receive a smaller component size (i.e. 40 or 44mm), who have a diagnosis other than osteoarthritis (i.e. avascular necrosis, rheumatoid arthritis), a leg length discrepancy greater than or equal to 1 cm, or low baseline Harris Hip Score (HHS) have a greater risk of revision than other patients. The more risk factors a patient has, the greater the risk of procedure failure requiring a revision to the hip. Please see Tables 21 and 23 for revision rates for each risk factor group.

Please see the complete list of *Warnings and Precautions* in the Instructions for Use for the Cormet Hip Resurfacing System.

## V. DEVICE DESCRIPTION

### General Overview

The Cormet Hip Resurfacing System is a metal-on-metal hip resurfacing system. The system consists of a stemmed resurfacing femoral head component designed for cemented fixation and an acetabular component designed for cementless fixation. The acetabular component has a bicoating™ of plasma sprayed titanium and hydroxyapatite (HA).

{2}

# Materials

Table 1: Materials

| Component | Material | Standard |
| --- | --- | --- |
| Femoral Resurfacing Head | Cobalt Chromium Alloy | ASTM F75^{1} |
| Acetabular Component | Cobalt Chromium Alloy
Unalloyed Pure Titanium (coating)
Hydroxyapatite powder (coating) | ASTM F75
ISO 5832 Part 2^{2}
ASTM F1185^{3} |

# Sizing and System Compatibility

Each femoral head component is compatible with two acetabular components with the exception of the 56mm diameter head, which is only compatible with the 62mm nominal outside diameter (OD) acetabular cup.

Table 2: Description of Components

| Femoral Head (Nominal Outside Diameter) | Acetabular Component
(Nominal Inside Diameter of cup x Nominal Outside Diameter of cup) |
| --- | --- |
| 40mm | 40 x 46mm, 40 x 48mm |
| 44mm | 44 x 50mm, 44 x 52mm |
| 48mm | 48 x 54mm, 48 x 56mm |
| 52mm | 52 x 58mm, 52 x 60mm |
| 56mm | 56 x 62mm |

# VI. ALTERNATE PRACTICES AND PROCEDURES

1. Non-surgical treatment (e.g., reduced activity, medications, physical therapy) or no treatment at all;
2. Other commercially available total hip replacement devices. Commonly used implant bearing materials for total hip arthroplasty include metal on ultra-high molecular weight polyethylene (UHMWPE), ceramic on UHMWPE, metal on metal, and ceramic on ceramic.
3. Rotational osteotomy;
4. Hip fusion.

# VII. MARKETING HISTORY

The Cormet Hip Resurfacing System was launched in Europe in 1997. It has been distributed in the countries listed in Table 3. The Cormet Hip Resurfacing System has not been withdrawn from marketing for any reason relating to the safety and effectiveness of the device.

3

{3}

Table 3: Worldwide Marketing History

| Argentina | Ecuador | Italy | Spain |
| --- | --- | --- | --- |
| Australia | Egypt | Japan | Sri Lanka |
| Belgium | Finland | Lebanon | Sweden |
| Brazil | France | Malta | Switzerland |
| Canada | Germany | Mexico | Syria |
| Chile | Greece | Pakistan | Taiwan |
| China | Holland | Portugal | Turkey |
| Colombia | India | Qatar | U.A.E. |
| Croatia | Iran | Saudi Arabia | U.K. |
| Cyprus | Israel | South Africa | Venezuela |

VIII. POTENTIAL ADVERSE EFFECTS OF THE DEVICE ON HEALTH

Reported Device Related Adverse Effects

The most commonly reported Cormet Hip Resurfacing device related adverse events are:
- femoral neck fracture
- femoral component migration/loosening
- acetabular component migration/loosening
- femoral subsidence
- dislocation
- greater trochanter fracture
- lesser trochanter fracture

A complete list of the frequency and rate of complications and adverse events identified in the clinical study are provided in the Summary of Clinical Studies, Tables 18-20.

Potential Adverse Effects

The following adverse effects may occur in association with hip replacement surgery including the Cormet Hip Resurfacing System:

1. Device failure because the components cannot be expected to indefinitely withstand the activity level and loads of normal healthy bone.
2. Dislocation of the hip resurfacing prosthesis can occur due to inappropriate patient activity, trauma or other biomechanical considerations.
3. Loosening of hip resurfacing components can occur. Early mechanical loosening may result from inadequate initial fixation, latent infection, premature loading of the prosthesis or trauma. Late loosening may result from trauma, infection, biological complications, including osteolysis, or mechanical problems, with the subsequent possibility of bone erosion and/or pain.
4. Fatigue fracture of the implants as a result of excessive loading, malalignment, or trauma.
5. Peripheral neuropathies, nerve damage, circulatory compromise and heterotopic bone formation may occur.
6. Surgical complications including, but not limited to: genitourinary disorders; gastrointestinal disorders; vascular disorders, including thrombus; bronchopulmonary disorders, including emboli; myocardial infarction or death.
7. A sudden, pronounced, intraoperative blood pressure decrease due to the use of bone cement.

4

{4}

8. Hematoma or damage to blood vessels resulting in large blood loss.
9. Delayed wound healing.
10. Superficial or deep infection. Infections may occur months to years after surgery and these infections are difficult to treat and may require reoperation with removal surgery and later replacement at another time.
11. Increased hip pain and/or reduced hip function.
12. Metal sensitivity reactions or allergic reactions or metallosis.
13. Adverse effects may necessitate reoperation, revision, arthrodesis of the involved joint, Girdlestone and/or amputation of the limb. Surgeons should advise patients of these potential adverse effects.
14. Bone perforation or fracture (occurring either intraoperatively or occurring postoperatively as a result of trauma, excessive loading, osteolysis or osteoporosis).
15. Wear deformation of the articular surface (as a result of excessive loading or implant malalignment).
16. Limb length discrepancy.
17. Osteolysis and/or other periprosthetic bone loss.

Any of these adverse effects may require medical or surgical intervention. Rarely, these adverse effects may lead to death.

## IX. SUMMARY OF PRECLINICAL STUDIES

### General Overview

The following preclinical studies were carried out on the Cormet Hip Resurfacing System: wear, frictional torque, fatigue strength testing, surface coating characterization, range of motion (ROM), luxation wear, metal ion analysis and sterilization/shelf-life validation.

### Wear Testing

#### Worst Case Design:

The applicant performed wear testing on both extremes of the range (40mm and 56mm) in order to explore the potential worst case scenarios.

#### Acceptance Criteria:

The amount of wear particles produced was compared to the wear generated by a 28mm bearing couple (control), which is the standard size for a total hip replacement.

#### Methods:

Three wear testing studies were completed.

1. Three variables were tested to investigate which parameters had the most effect on wear: Sphericity, Diametrical Clearance and Metallurgy. The test compared the 'heat-treated' (Hot Isostatically Pressed and Solution Annealed) Cormet device to the previously manufactured 'as-cast' type device. Two 48mm diameter Cormet devices and one 'as-cast' type device were run to two million cycles at 2.1kN maximum load (3x body weight) establishing steady-state conditions with diametrical bearing clearances of 81μm, 291μm and 300μm respectively.
2. Another wear study evaluated the effect of metallurgy ('as cast' vs. 'heat-treated' high carbon 40mm diameter Co-Cr-Mo bearings) on wear. Four 28mm and four 56mm diameter 'double heat-treated' metal-on-metal bearings were run to six million

{5}

cycles under normal gait conditions. Radial clearance of the bearings was controlled to $110\mu \mathrm{m}$, initial surface finish to $0.01\mu \mathrm{m}$ and sphericity to between 4 to $8\mu \mathrm{m}$. The loading cycle was based on the 'Paul' cycle applying between 50N and 2450N at 1Hz.

3. The final study investigated the effects of heat treatment on wear rates in metal-on-metal bearings. Four 40mm diameter 'as-cast' and four 40mm diameter 'heat-treated' metal-on-metal bearings were tested under standard and 'severe' gait (fast-jogging) conditions up to six million cycles. Diametrical clearances were a mean of $214\mu \mathrm{m}$ in the 'heat-treated' group and $258\mu \mathrm{m}$ in the 'as cast' group and sphericity was controlled to $<10\mu \mathrm{m}$ in all the samples. All components were subject to three million cycles of 'normal walking' (standard gait) with a maximum load of 2450N at 1Hz.

## Results:

1. There is no difference between the 'heat-treated' Cormet and 'as cast' devices. However, the Cormet devices, with improved sphericity, did show improved wear performance over versions of previously manufactured devices.
2. The $28\mathrm{mm}$ diameter bearings indicated the highest steady-state wear rate with the largest running-in wear occurring in the $56\mathrm{mm}$ bearings. The $40\mathrm{mm}$ group had lower running-in and steady-state wear compared to the $28\mathrm{mm}$ coupling. The $56\mathrm{mm}$ bearings produced the lowest steady-state wear of all the groups.
3. The steady-state wear rates $(0.4\mathrm{mm}^3 /10^6$ cycles) found during 'normal walking' were similar to those for $36\mathrm{mm}$ diameter metal-on-metal bearings reported in the literature. When 'normal walking' was resumed after the 'severe' wear, then the steady state wear rates returned to the level found prior to the 'severe' test regime.

## Frictional Torque

## Worst Case Design:

Size 56mm bearing samples were assumed 'worst-case' since torque is proportional to head diameter. These samples had the minimal diametrical clearances of $150\mu \mathrm{m}$, which provides the maximum initial contact area.

## Acceptance Criteria:

Andersson et al.$^{1}$ suggested the torque required to remove a well cemented acetabular cup from a cadaveric socket is $100\mathrm{Nm}$.

## Methods:

1. In December 2005, five 56mm heads were paired with five 62mm cups to give the specified diametrical clearance (150μm). The frictional torque of each bearing pair was recorded independently in flexion-extension and internal-external rotation under a joint load of 2.45kN and 4.5kN.
2. Two 48mm diameter Cormet heads were studied with diametrical clearances of 81 and 291 microns.
3. New and worn components from previous wear studies were tested (40mm and 56mm heads). Flexion-extension and rotational torques were measured. Tests were performed at normal walking loads (2.45kN) and at extreme load cycles (4.5kN).

## Results:

1. The maximum absolute torque was recorded at an average of $11.9\mathrm{Nm}$ and $22.4\mathrm{Nm}$ under a test load of $2.45\mathrm{kN}$ and $4.5\mathrm{kN}$, respectively. The maximum absolute torque for internal-

6

{6}

external rotation was found to be an average of $3.3\mathrm{Nm}$ and $3.9\mathrm{Nm}$ under a test load of $2.45\mathrm{kN}$ and $4.5\mathrm{kN}$, respectively.

2. The $81\mu \mathrm{m}$ and $291\mu \mathrm{m}$ samples exhibited maximum torques (during the initial running-in period) of $18.5\mathrm{Nm}$ and $8.5\mathrm{Nm}$, reducing to averages of $11\mathrm{Nm}$ and $5.3\mathrm{Nm}$, respectively, during steady-state wear.

3. The maximum torque was found to occur during flexion-extension motion. The maximum absolute (modulus) torque was recorded in the 'as-new' condition for the 56mm bearings at an average of 7.40 and $10.65\mathrm{Nm}$ for the loads of $2.45\mathrm{kN}$ and $4.5\mathrm{kN}$, respectively.

## Fatigue Strength Testing

### Worst case:

The distance from the center of rotation of the spherical head to the point of contact between the stem and the pre-drilled hole is the maximum for the 56mm head. The force that is transmitted through the center of the resurfacing head, therefore, creates a maximum bending moment in the 56mm device.

### Acceptance Criteria:

ISO 7206-8 “Implants for Surgery - Partial and Total Hip Joint Prostheses - Part 8: Endurance Performance of Stemmed Femoral Components with Application of Torsion” for total hip replacements.

### Methods:

The test method was configured to simulate the fault condition of the resurfacing head similar to ISO 7206-4 “Implants for Surgery - Partial and Total Hip Joint Prostheses - Part 4: Determination of Endurance Properties of Stemmed Femoral Components.” The short stem was fixed $25\mathrm{mm}$ below the underside of the head. Five static tests were performed. The failure point was identified as the point on the load/extension graphs where the elastic region ended (became nonlinear).

Five samples were then dynamically tested at $3\mathrm{kN}$ (approximately $50\%$ of the mean static failure load) at $30\mathrm{Hz}$ to five million cycles.

### Results:

The mean static failure load was $6.28 \pm 0.55 \mathrm{kN}$. All samples went on to survive higher loads without catastrophic failure, but with permanent deformation of the femoral stem. Five samples were then dynamically tested at $3\mathrm{kN}$ at $30\mathrm{Hz}$ to five million cycles without failure in the same test configuration.

7

{7}

8
# Surface Coating Characterization

The acetabular component is coated with a plasma sprayed unalloyed titanium and hydroxyapatite (HA) coating.

## Plasma Spray

### Acceptance Criteria:

The static shear strength of surface/substrate interface should exceed 20 MPa for porous surface coatings as tested per ASTM F1044 “Standard Test Method for Shear Testing of Calcium Phosphate Coatings and Metallic Coatings.” The static tensile strength of the surface/substrate shall exceed 20 MPa for porous surface coatings. Shear fatigue strength testing per ASTM F1160 “Standard Test Method for Shear and Bending, Fatigue of Calcium Phosphate and Metallic Medical and Composite Calcium Phosphate/Metallic Coatings” should exceed 10 million cycles at a stress of 10MPa.

### Methods:

The static shear strength of surface/substrate was tested per ASTM F1044. The static tensile strength of the surface/substrate was evaluated per ASTM F1147 “Standard Test Method for Tension Testing of Calcium Phosphate and Metal Coatings” and the shear fatigue strength was evaluated per ASTM F1160.

### Results:

Table 4 summarizes the results of the plasma spray coating testing.

| Test | N | Results | S.D. |
| --- | --- | --- | --- |
| Static Shear (ASTM F1044) | 5 | 20.9 MPa | 4.1 |
| Static Tension (ASTM F1147) | 3 | 35.9 MPa | 2.8 |
| Abrasion Strength | 6 | 54.1mg weight loss | 6.4 |
| Surface Roughness | 6 | Ra 25.7 microns | 7.2 |

Shear fatigue strength testing per ASTM F1160 was completed on six samples for 10 million cycles at a stress of 10MPa with no failures.

## Hydroxyapatite (HA) Coating

### Acceptance Criteria:

The acceptance criteria are described in FDA’s “510(k) Information needed for Hydroxyapatite Coated Orthopaedic Implants dated March 10, 1995 (revised 2/20/97).”

### Methods:

The HA coating was characterized with regard to density, particle size, porosity thickness, Ca/P ratio, solubility/dissolution, bonding strength and crystallinity.

### Results:

Table 5 summarizes the results of the HA characterization.

{8}

Table 5: HA Coating Characterization

| Chemical Composition | 2(Ca₅(PO₄)₃OH) |
| --- | --- |
| Trace Elements | As <1 ppm
Cd <1 ppm
Hg <1 ppm
Pb <1 ppm
Total Heavy Metals <50 |
| Ca/P ratio | Powder 1.697 (1.667±0.03)
Coating 1.655 (1.667±0.02) |
| Crystallinity | Coating: 62% |
| Crystalline Phases | %HAP
Powder: >97%
Coating: >70% |
| | % Alpha Tricalcium Phosphate (TCP)
Powder: 0
Coating: <4% |
| | % Beta TCP
Powder: 0
Coating: <6% |
| | %TCPM
Powder: 0
Coating: <7% |
| | %CaO
Powder: 0.7%
Coating: <1% |
| Density | 3.096 g/cm³ |
| --- | --- |
| Grain Size | 10% <17 μm
90% <83 μm |
| Porosity | Global porosity: 27%
Pore medium size: 38.92 μm
Standard deviation: 31.71 μm |
| Thickness | 119 μm |
| Solubility | 2x10⁻⁵⁶ |
| Tensile Strength | 33.31 MPa
(S.D. 6.5 MPa) |
| Adhesive Strength | 14.9 MPa |

## Range of Motion

### Worst Case Design:

A cylindrical “femoral neck” results in the smallest angles of articulation. Therefore, a cylindrical “femoral neck” was utilized to detect impingement between the acetabular cup and femoral neck. The 56mm diameter Cormet resurfacing head bearing surface subtends the smallest angle in the size range. The 62mm and 64mm acetabular cups’ bearing surfaces subtend the largest angle in the size range (64mm not available in US). Therefore, the 56mm diameter head paired with the 62/64mm cup coupling produced the smallest articular angle before impingement.

### Acceptance Criteria:

As outlined in ISO 21535:2002 “Specific requirements for hip-joint replacement implants” the minimum allowable angle of flexion/extension is 80°, abduction/adduction is 60° and internal/external rotation is 90°.

### Methods:

Range of motion was evaluated per ISO 21535, which is intended for stemmed total hip replacements with diaphyseal fixation. The test protocol was modified to consider the proximal bone preserving nature of hip resurfacing, by molding a cylindrical femoral neck around the stemmed component. Flexion/extension, abduction/adduction and internal/external rotation were measured by identifying the angle at which impingement occurs.

{9}

10

# Results:

The flexion/extension angle at which impingement occurred with the worst case components was 83°, the abduction/adduction angle at which impingement occurred was 70° and the internal/external degree of rotation at which impingement occurred was 111°.

# Luxation Wear

## Worst Case Design:

Five 40mm and five 56mm bearings were tested. Bearing clearances were controlled to 400µm, the maximum specified in the manufacturing tolerances.

## Acceptance Criteria:

Komistek et al.² used fluoroscopy to demonstrate that small diameter total hip replacement devices (metal-on-polyethylene) subluxed several millimeters during each gait cycle. With metal-on-metal bearings they were not able to detect any subluxation up to the resolution of the fluoroscope, which is 750 microns.

## Methods:

Five 40mm and five 56mm bearings were tested in Ringers solution at 37° according to the procedure ‘Determination of Resistance to Luxations and Repositions of Total Hip Joint Prostheses’ by Kaddick et al.³ A horizontal preload of 1kN was used. The forces required to cause luxation of the bearings during the first cycle were then recorded and the displacement noted. The cups were examined and then repeated luxations were performed (a further 999 cycles per bearing couple). A displacement of 10mm was pre-set since this was greater than the displacement required to cause luxation in the 40mm and 56mm diameter bearings found during the first luxation.

## Results:

After the first luxation cycle, a small decrease in luxation force occurred for all bearing couples, which was thought to be due to rounding of the cup rim. Thereafter, a steady increase in maximum luxation force was noted until steady-state was achieved. This increase was consistent with increased surface roughening of both the head and cup bearing surfaces. Forces to cause luxation were in excess of 2kN for both the 40mm and 56mm bearings.

# Metal Ion Analysis

## Description of Study Population

A metal ion study was conducted at Coventry &amp; Warwickshire Hospital, United Kingdom outside of the applicant’s US IDE study. A series of 29 patients who underwent a unilateral metal-on-metal hip resurfacing procedure were prospectively followed over a seven-year period. Seven of the 29 patients underwent a metal-on-metal hip resurfacing procedure on the contralateral hip during the course of the study. These seven patients along with four other patients who had a previous hip resurfacing on the other hip had their metal ion levels assessed over time to determine the effect of bilateral hip resurfacing on metal ion levels.

## Implant Identification

Both the Corin-McMinn device and the Cormet device (the subject device) used in this study have cups with a HA coated back on a plasma sprayed titanium layer over the CoCr substrate with a supero-medial peg. Heads are similar in both devices, for use with cement and uncoated in the cement/implant contact areas. The significant differences between the Corin-McMinn and the Cormet are that the former has a splined supero-medial cup peg (Cormet is parallel and non-splined), two bearing surface integral introducer holes (Cormet has no holes) and two opposing stippled pads on the cup back (Cormet has low profile locating splines and no pads).

{10}

11
11

## Measurement Techniques

Blood samples were taken from each patient. A plastic intravenous cannula was inserted, the metal needle removed and 5 mls of blood withdrawn and discarded. Samples were then taken and placed in 2 ml Heparin tubes, which had been tested for cobalt and chromium contamination. The blood was centrifuged and the plasma transferred into trace-metal free polycarbonate tubes. Cobalt and chromium levels were determined.

## Metal Ion Levels

Metal ion levels are raised and remain elevated following metal-on-metal hip resurfacing; however, it is unclear if the levels to which they are raised are of any clinical significance.

## Summary of Data

For patients with one resurfacing device, results of the study indicate that metal ion levels for cobalt and chromium initially increased following a metal-on-metal hip resurfacing but plateaued and started to decrease between one and two years post-implantation. The levels remained below their peak, but did not return to preoperative levels throughout the seven-year follow-up reported in this study. Implantation of a contralateral metal-on-metal resurfacing system further raised the metal ion levels, more notably cobalt ions compared to chromium ions. The cobalt levels did not return to normal following bilateral hip resurfacing and remained higher than patients with unilateral hip resurfacing over four years. Chromium levels following bilateral surgery do not return to normal, but are only slightly higher when compared to levels of a unilateral resurfacing. Appropriate Contraindications and Warnings have been added to the labeling due to increased metal ion levels.

## Sterilization and Shelf Life Validation

Femoral head and acetabular components of the Cormet Hip Resurfacing System are sterilized by gamma irradiation delivered from a cobalt₆₀ source. The sterilization process has been validated to achieve a sterility assurance level of 10⁻⁶ at a minimum dose of 25kGy in compliance with the requirements of EN 556-1:2001 Sterilization of medical devices - Requirements for medical devices to be designated "STERILE" - Part 1: Requirements for terminally sterilized medical devices and AAMI TIR 27, 2001, Sterilization of Health Care Products - Radiation Sterilization - Substantiation Of 25 kGy As a Sterilization Dose - Method Vd Max. The irradiation process is carried out by a subcontractor, Isotron PLC, Reading, UK. The product is not labeled "pyrogen free". The Cormet Head devices are packaged in double-peel pouch packages and the Cormet Cup devices are packaged in double-blister packages to maintain sterility. Shelf life testing on both package configurations was performed to verify package integrity equivalent to five years.

## X. SUMMARY OF CLINICAL STUDIES

### Purpose of the Investigation

The purpose of this investigation was to test the hypothesis that the Cormet Hip Resurfacing System is as effective as conventional total hip arthroplasty. The Cormet Hip Resurfacing System was the investigational treatment and a conventional total hip arthroplasty system served as the control group. Effectiveness was measured via a composite endpoint described below. Safety was determined by collection of the incidence of perioperative and postoperative complications.

{11}

# Study Design

A prospective, multi-center, IDE study was conducted utilizing components of the Cormet Hip Resurfacing System in the United States.

## Control Group

The control group was comprised of total hip arthroplasty patients from an alumina ceramic total hip prosthesis. These ceramic total hip prostheses were approved via PMA. Table 6 compares the investigational and study parameters.

Table 6: Protocol Comparisons

| Protocol Element | Cormet IDE Study | Ceramic Total Hip (Control) |
| --- | --- | --- |
| Type of Study | IDE – Hip Resurfacing | IDE – Total Hip Arthroplasty |
| Bearing Type | Metal-on-Metal | Ceramic-on-Ceramic |
| Study Design | Prospective, non-randomized, historical control | Prospective, randomized |
| Number of centers | 14 | 16 |
| Dates of enrollment | 5/17/2001- 8/5/2003 (pivotal)
Continued access through July 2006 (ongoing) | 10/29/1996 – 10/20/1998 |
| Number of procedures | 1148 | 349 |
| Follow-Up Intervals | Preoperative, 6 weeks, 6, 12, 24 and 24+ months* | Preoperative, 6 weeks, 6, 12, 24 and 24+ months* |
| Outcome Measures | Harris Hip Score
Adverse Events
Radiographs
Questionnaire | Harris Hip Score
Adverse Events
Radiographs
Questionnaire |

* 24+ month evaluations include all 24 month evaluations completed, as well as data from a later visit, if the 24 month evaluation was not available.

The core data collected from these studies was the same. In addition, the follow-up time-points and the intervals around these time-points were very similar as illustrated in Table 7.

Table 7: Follow-up Intervals Comparison

| | Cormet Approved protocol | Cormet PMA submission | Ceramic Total Hip Control |
| --- | --- | --- | --- |
| 6 weeks | ±2 weeks | ±2 weeks/+expanded | ±3 weeks |
| 6 months | ±1 month | ±1 month + expanded | ±1 month |
| 1 year | ±2 months | ±2 months/+expanded | ±2 months |
| 2 years | ±2 months | ±2 months/+ expanded | ±2 months |
| 2+ years | | Any evaluation 22+ months=24+ months | |

The patient populations recruited into both studies were similar. A side-by-side comparison of the inclusion/exclusion criteria between the studies is presented in Table 8.

{12}

Table 8: Inclusion/Exclusion Criteria Comparison

| Inclusion/Exclusion | Cormet Approved Protocol | Control Group Study |
| --- | --- | --- |
| Is skeletally mature | X | X |
| Is mentally capable of follow-up | X | X |
| Will be available for 2 yr follow-up | X | X |
| Deemed candidate by diagnosis of investigator | X | X |
| No active infection | X* | X |
| No severe osteoporosis | X* | X |
| Not a prisoner | X | X |
| Is not pregnant | X | X |
| Is not morbidly obese | X* | X* |
| No ipsilateral previous surgery | X | X |
| No extensive deformity of femoral head | X* | Not applicable |
| No known allergies to implants | X | None included in study |
| No neoplastic disease | X* | None included in study |
| No above the knee amputation either extremity | X | None included in study |
| No previous Girdlestone procedure | X | Information Not Available to Applicant |
| No previous hip fusion | X | Information Not Available to Applicant |
| No above the knee amputation (AKA) of either extremity | X | Information Not Available to Applicant |
| Does not require structural bone graft | X | Information Not Available to Applicant |
| No previous ipsilateral hemi-resurfacing, total resurfacing, total bipolar, total unipolar, or total hip replacement | X | Information Not Available |
| No nonunion or malunion of the femur | X | Information Not Available |
| Has preoperative Harris Hip Score (HHS) < 70 points | X | No limits |
| No Congenital Dysplasia of the Hip (CDH) | X | Included in study |
| Age | No specified limits | 21-75 |
| Inflammatory Arthritis | Included in study | None included in study |

*Primary Investigator (PI) discretion

The majority of the eligibility criteria were consistent for both studies. Comparisons of the populations enrolled found the groups to be similar on age, diagnosis and preoperative Harris Hip Score (HHS).

Both groups have a hard-on-hard bearing surface without the risk of issues associated with polyethylene debris.

## Composite Clinical Success Endpoints

A patient is defined as a Composite Clinical Success (CCS) if at 24 months all of the following criteria outlined in Table 9 are met.

{13}

Table 9: Summary of the Composite Clinical Success

| Composite Clinical Success Criteria |
| --- |
| Harris Hip Score ≥80 at Month 24+ |
| No revisions/pending revisions |
| Radiographic Success Criteria |
| Acetabular Migration (vertical/horizontal): <5mm |
| Acetabular Migration (varus/valgus): <5° |
| Acetabular Radiolucencies: not in all zones |
| Femoral Subsidence (axis femoral canal) < 5mm or Femoral Tilt varus/valgus <1° |
| Femoral Radiolucencies: not in all zones |
| Absence of device related Adverse Events |

The primary efficacy objective of this study was to demonstrate clinical non-inferiority with regard to the likelihood of clinical success at Month 24 relative to the control. To achieve this goal, it was necessary to determine clinical success status in as large a percentage of procedures as possible. This was necessary to insure that bias arising from losses to follow-up was small enough to permit valid inference. For procedures in the Pivotal Study Unilateral Cohort for which Month 24 Harris Hip Total score was missing, a later Harris Hip Total score was used. The analysis demonstrated that these later scores were an accurate predictor of Month 24 score, and as such minimized bias by imputing missing Month 24 values with these later values. Additional statistical analyses were performed in order to demonstrate that the rollback imputation as well as the out of window procedures had no effect on the overall results.

Month 24+ CCS Actual$^{B}$ Pivotal Study Unilateral Patients (see Table 16) was defined to be the primary comparison.

## Study Modifications

The data presented is from a study designed to be a prospective, non-randomized (yet concurrently controlled) clinical study used to evaluate the Cormet Hip Resurfacing System. However, multiple modifications were made to the study design throughout the course of the study and subsequent data analyses. These changes were discussed by the Orthopaedic and Rehabilitation Devices Panel (the Panel) on Thursday, February 22, 2007.

### 1. Control Group History

The original study compared the investigational device group, Group I, and two non-concurrent control groups, Group II and Group III. Group I was the investigational group treated with the Cormet Hip Resurfacing System. Patients were to be sequentially enrolled. Group II was the control group treated with a marketed metal-on-metal total hip replacement. Group III was the control group treated with any marketed metal-on-polyethylene total hip replacement. Each investigational site was to generate data for Groups I and II, or data for Groups I and III. However, as enrollment into the study progressed, no control patients were actually enrolled for utilization in the analysis of data for this study. To address the absence of a control group, the applicant proposed and explored multiple historical control alternatives (metal-on-metal then ceramic-on-ceramic) as they began to analyze their data. The ceramic total hip approved via PMA was finally selected as the control.

### 2. Data Analysis

The composite clinical success analysis included all of the originally approved and proposed study endpoints; however, the radiographic endpoint success criterion was different and appeared to be less stringent.

14

{14}

a. Changes to data collection (radiographic) techniques

The IDE approved protocol included a radiographic measurement technique as outlined in Table 10.

Table 10: Comparison of Measurement Techniques

| Radiographic Analysis | Original IDE Protocol dated March 20, 2003 | Original PMA Submission March 30, 2005 | Actual Technique PMA Amendments 8 and 13 |
| --- | --- | --- | --- |
| Acetabular Migration vertical/ horizontal | Reference inferior teardrops | Reference bottom of pelvis | Same as Original Protocol |
| Acetabular Migration varus/valgus | Angle between a line joining edges of the cup and a line joining tear drops | Angle between a line joining edges of the cup and a line joining bottom of pelvis | Same as Original Protocol |
| Acetabular Radiolucencies | Serial | Same as Original Protocol | Same as Original Protocol |
| Femoral Subsidence Axis Femoral Canal | Line to lateral femoral cortex | Same as Original Protocol | Line from head center to top of greater trochanter |
| Femoral Tilt Varus/Valgus | Lines through femur midpoint and stem | Same as Original Protocol | Same as Original Protocol |
| Femoral Radiolucencies | Serial | Same as Original Protocol | Same as Original Protocol |

b. Changes to radiographic analyses

The final data analysis used a revised measurement technique and then applied revised success criteria to evaluate radiographic success. Please see Table 11 for a summary of the radiographic success criteria. The sponsor has combined the femoral subsidence and femoral tilt endpoint into one. Therefore, the femoral component need only meet one of its endpoints to be a success. In addition, the original criteria indicated a radiolucency in any zone was considered a failure, as is common in hip prosthesis studies. However, the final proposed analysis indicated radiolucencies not in all zones to be a success.

{15}

16

Table 11: Comparison of Radiographic Success Criteria

| Radiographic Success Criteria | Original IDE Protocol | Original PMA Submission | Actual Criteria Used |
| --- | --- | --- | --- |
| Acetabular Migration vertical/ horizontal | < 5mm | Same as Original Protocol | Same as Original Protocol |
| Acetabular migration varus/valgus | < 5 degrees | Same as Original Protocol | Same as Original Protocol |
| Acetabular Radiolucencies | None in any zone | Not Evaluated | Not in all zones |
| Femoral subsidence axis femoral canal | < 5mm | Same as Original Protocol | Combined < 5mm and < 1 degree (must have both for failure) |
| Femoral tilt varus/valgus | < 1 degree | Same as Original Protocol | |
| Femoral Radiolucencies | None in any zone | Not Evaluated | Not in all zones |

Metal-on-metal hip resurfacing devices are relatively new to the US orthopaedic community and the radiographic evaluation criteria of these devices have not been uniformly accepted.

c. Proposed primary effectiveness and safety endpoints definitions

This section outlines the composite clinical success outcomes used in the approved IDE Protocol, which differ from the endpoints described in the previous section.

Approved IDE Protocol

The original IDE success definition was outlined as “...at 24 months a patient is defined as a success, if all four of the following are met:

1. Harris Hip Score (HHS) ≥ 20 point improvement
2. Has not had and is not planning a revision surgery.
3. Radiographic Success:

a. Acetabular component
- Migration &lt;5mm vertical or horizontal
- Migration &lt;5° in varus/valgus
- No new or progressive radiolucencies &gt;1mm in any zones

b. Femoral component
- Subsidence &lt;5mm
- Tilting &lt;1° in varus/valgus
- No new or progressive radiolucencies &gt;2mm in any zones

4. No device related complications—an AE due to the design and/or material composition of the implant and/or implant instrumentation; the relationship to the device will be determined by the investigator.

Any patient who does not meet all of the above criteria during any evaluation time point out to two years will be considered a failure.”

This summary outlines the changes in study design and analysis that have been implemented throughout the course of the study and PMA review. Please see Section XII Panel Recommendation for additional information.

{16}

# Study Objectives and Assessments

## Study Population

Corin collected US IDE clinical trial data on 1154 cases implanted with the Cormet Hip Resurfacing System. Six procedures involved use of a pegged acetabular component, not part of the IDE. These cases were analyzed separately. Study data were therefore presented on 1148 study cases. There were no major protocol deviations reported during a comparable timeframe in the control group, however, there was one approved deviation for inflammatory arthritis. These data are not included in this submission. Eight investigational procedures involved enrollment under the compassionate use provisions. The study populations are identified in Table 12.

Table 12: Study Cohort Definitions

| Cohort Name | Definition | Procedures/patients |
| --- | --- | --- |
| All Enrolled | All patients enrolled in either the pivotal study or continued access. | 1148/1030 |
| Pivotal Study Unilateral | Unilateral patients enrolled in the pivotal study. Includes patients who had second side replaced after two years of follow-up (730 days). | 337/337 |
| Pivotal Study Bilateral | Patients with first implant in the pivotal study who had their second hip replaced within 730 days of the index procedure. Four incidences where second hip was not included in this study group because of use of pegged cups. | 105/55 |
| Continued Access | Patients implanted after the pivotal study close (Aug 6, 2003) under continued access provision. Note: seven of these patients are also included in the pivotal study bilateral patient population above. | 698/640 |
| Compassionate Use | Implanted with investigational device under compassionate use between end of pivotal IDE study and beginning of continued access approval. | 8/7 |

## Baseline Characteristics of Investigational and Control Groups

The demographics for the pivotal unilateral study as compared to the control are identified in Table 13.

Table 13: Comparison of Pivotal Study to Control

| Population | Investigational Pivotal Study Unilateral Patients | Ceramic Total Hip Control Pivotal Study Unilateral Patients | Wilcoxon (continuous) or Chi-squared (discrete) P values |
| --- | --- | --- | --- |
| Number of procedures | 337 | 266 | |
| Number of patients | 337 | 266 | |
| Mean Age | 50.1 | 53.3 | <0.01 |
| Gender M/F | 67.7%/32.3% | 62%/38% | 0.150 |
| Mean weight (lbs) | 190.4 | 188.7 | 0.692 |
| Diagnosis | 85.8% OA, 1.2% RA, 13.1% AVN | 83.7% OA, 16.3% AVN | For Diagnosis=OA
p=0.135 |
| Preoperative HHS mean total score | 50.1, SD=11.6 | 49.7, SD=11.3 | 0.233 |

{17}

Baseline demographic and clinical characteristics between Cormet Hip Resurfacing System patients in Pivotal Study Unilateral group and Ceramic-on-Ceramic Unilateral Control group patients were compared. There was no statistically significant difference in patient gender. Approximately two-thirds of both cohorts were male (67.7% versus 62.0%). There were no statistically significant differences in the distributions of diagnoses with approximately 85% of both cohorts presenting with a primary diagnosis of osteoarthritis, (85.8% versus 83.7%). Mean weights were similar between these two groups (190.4, SD=40.7 lbs versus 188.7 lbs; SD=39.7). Height was not obtained during the Cormet Hip Resurfacing System study precluding comparisons involving BMI.

Baseline overall function as reflected in mean Harris Hip Total scores also was very similar between the Pivotal Study Unilateral investigational and control groups. In contrast, the difference in mean ages was statistically significant between groups (50.1, SD=9.6 versus 53.3, SD=11.1, Wilcoxon rank sum p&lt;0.01). Although statistically significant, a difference in mean age of just three years is unlikely to be of clinical significance.

Propensity score analyses were performed to assess the magnitude of and to adjust for potential selection bias. The propensity model included age, weight, baseline HHS, gender, and pre-surgery presence of marked pain. The mean propensity scores were very similar in the Cormet and ceramic total hip control groups, [0.589 (0.088 SD) and 0.550 (0.103 SD), respectively]. The values were very similar in magnitude suggesting that, taken as a set, these variables had relatively little impact on characterizing what kind of patients received the Cormet implant relative to what kind of patients received the ceramic total hip control device. This implies that any between group differences in patient populations for the covariates included in the model did not affect the conclusion of the non-inferiority for CCS.

The demographics for all other populations are identified in Table 14.

Table 14: Demographics for Other Populations

| Population | Investigational Pivotal Study Bilateral | Continued Access | All Enrolled |
| --- | --- | --- | --- |
| Number of procedures | 105 | 698 | 1148 |
| Number of patients | 55 | 640 | 1030 |
| Mean Age | 47.7 | 52.3 | 51.2 |
| Gender M/F | 71.4%/28.6% | 74.2%/25.8% | 71.9%/28.1% |
| Mean weight (lbs) | 195.1 | 194.9 | 193.8 |
| Diagnosis | 81.0% OA, 1.9% RA, 17.1% AVN | 92.1% OA, 0.3% RA 7.6% AVN | 89.1% OA, 0.8% RA 10.1% AVN |
| Preoperative HHS mean total score | 48.7, SD=11.8 | 50.1, SD=11.4 | 50.0, SD=11.5 |

## Patient Accounting

*Pivotal Study Unilateral Patient Accountability*

Table 15 presents an overview of the data available for the pivotal study cohort.

{18}

Table 15: Pivotal Study Unilateral Patient Accountability

| Status at Month 24+ | Number of Subjects |
| --- | --- |
| Pivotal study group enrollment | 337 |
| Patients with complete CCS score | 292 |
| Patient died before month 24+ | 1 |
| Patients not evaluated for CCS | 44 |
| Died after 24 month interval | 2 |
| Complete HHS data only | 9 |
| Complete radiographic data only | 5 |
| Patients with no Month 24+ data; Potential lost to follow-up | 28 |

The availability of follow-up evaluation for the investigational and control Pivotal Study Unilateral group is provided in Table 16.

Table 16: Procedure Accounting and Follow-up Compliance Table Pivotal Study Unilateral Patients and Controls

| As of Date of Database Closure | Pre-Op | | Week 6 | | Month 6 | | Month 12 | | Month 24 | | Month 24+ | | Month 36 | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| | I | C | I | C | I | C | I | C | I | C | I | C | I | C |
| (1) Theoretical follow-up | 337 | 266 | 337 | 266 | 337 | 266 | 337 | 266 | 337 | 266 | 337 | 266 | 314 | 266 |
| (2) Cumulative deaths including non-theoretically due | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 1 | 2 | 1 | 2 | 4 | 2 |
| (3) Cumulative revisions including non-theoretically due | 0 | 0 | 2 | 1 | 5 | 1 | 7 | 3 | 16 | 3 | 16 | 3 | 24 | 3 |
| (4) - Not Yet Overdue | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 32 | 0 |
| (5) - Deaths+revisions among theoretical due | 0 | 0 | 2 | 1 | 5 | 1 | 7 | 4 | 17 | 5 | 17 | 5 | 26 | 5 |
| (6) = Expected due for clinic visit | 337 | 266 | 335 | 265 | 332 | 265 | 330 | 262 | 320 | 261 | 320 | 261 | 256 | 261 |
| (7) = Expected due+revisions among theoretical due | 337 | 266 | 337 | 266 | 337 | 266 | 337 | 265 | 336 | 264 | 336 | 264 | 280 | 264 |
| All Evaluated Accounting (Actual^{B}) Among Expected Due Procedures^{1} | | | | | | | | | | | | | | |
| | I | C | I | C | I | C | I | C | I | C | I | C | I | C |
| (8) All Evaluated Visit Compliance (%) | 100.0% | 100.0% | 99.1% | 99.2% | 90.4% | 94.0% | 89.7% | 98.1% | 85.6% | 97.7% | 91.3% | 98.5% | 39.8% | 73.2% |
| (9) Harris Hip Total Score | 337 | 252 | 328 | 245 | 288 | 238 | 285 | 245 | 263 | 246 | 283 | 252 | 77 | 186 |
| (10) Radiographic evaluation | | | 313 | | 232 | | 234 | | 259 | | 291 | | 53 | |
| (11) CCS at Mos. 24, 24+ or HHS+radio. Otherwise | | | 332 | 245 | 297 | 238 | 294 | 245 | 243 | 250 | 292 | 256 | 97 | 186 |
| (12) Actual^{B} % Follow-up for CCS or HHS+radio. CCS | | | 99.1% | 92.5% | 89.5% | 89.8% | 89.1% | 93.5% | 72.3% | 94.7% | 88.9% | 97.0% | 37.9% | 71.3% |
| Within Window Accounting (Actual^{A}) Among Expected Due^{1} | | | | | | | | | | | | | | |
| | I | C | I | C | I | C | I | C | I | C | I | C | I | C |
| (13) Harris Hip Total Score | 337 | 252 | 277 | 221 | 161 | 183 | 192 | 215 | 200 | 206 | 281 | 251 | 22 | 156 |
| (14) Radiographic evaluation | | | 277 | | 161 | | 192 | | 202 | | 283 | | 22 | |
| (15) CCS at Mos. 24, 24+ or HHS+radio otherwise | | | 277 | 221 | 161 | 183 | 192 | 215 | 202 | 209 | 285 | 254 | 22 | 156 |
| (16) Actual^{A} % Follow-up for CCS or HHS+radio. CCS | | | 82.7% | 83.4% | 48.5% | 69.1% | 58.2% | 82.1% | 60.1% | 79.2% | 84.8% | 96.2% | 8.6% | 59.8% |
| 1 Actual A: Patients contributing all endpoint data that were evaluated within the protocol defined window.
Actual B: Patients contributing any data that were evaluated at a visit regardless of whether the visit was within the follow-up windows (not overlapping other protocol defined visit intervals) | | | | | | | | | | | | | | |

{19}

20

## Pivotal Study Unilateral

The follow-up rate at Month 24+ for patients with complete information to determine safety and effectiveness was 84.8% (285/336) for the investigational group and 96.2% (254/264) for the control group.

The following follow-up rates are also of interest:

## Pivotal Study Bilateral

At Month 24+, the follow-up rate is 56.1% (55/98) in comparison to 95% (79/83) for the bilateral control cohort.

## Continued Access

At Month 24+, 54.9% (134/244) of subjects due for evaluation have complete Harris Hip Scores and 6.1% (15/244) of subjects have complete radiographic data. Many subjects have not reached the Month 24 endpoint.

## All Enrolled Patient Accountability

At Month 24+, the follow-up rate is 50.7% (348/686) in comparison to 96.5% (335/347) for the control “all enrolled” cohort. In addition, although there have been 1,148 procedures completed to date, many of the patients have not yet reached the Month 24+ endpoint in the continued access study.

## Control

At Month 24+, the control follow-up rate was 96.2% (254/264).

## Patient Discontinuation

### Pivotal Study Unilateral Patients

#### Investigational Group (N=337)

All 337 patients in the Pivotal Study Unilateral group were theoretically due for Month 24+ follow-up evaluation. Of these 320 patients were expected due for the Month 24+ follow-up evaluation. One patient died prior to the Month 24+ follow-up and 16 patients had one or more of the components of the Cormet Hip Resurfacing System revised or removed prior to the Month 24+ visit. A total of 292 of the 336 patients (86.9%) who did not die prior to Month 24 were included in the CCS analysis. Of the 44 patients not included, two died after Month 24, nine had complete HHS (all success) only, and five had complete radiographs (all success) only. Twenty-eight patients did not return for Month 24+ follow-up.

#### Control Group (N=266)

Of the 266 patients in the control group, 261 patients were due for Month 24+ follow-up. One patient had one or more components of the ceramic total hip system revised, two patients died prior to the Month 24 follow-up visit. A total of 256 of a possible 264 patients (97.0%) were included in the CCS analysis. Of the nine patients not included, five had incomplete HHS scores and four patients were considered lost to follow-up.

### Pivotal Study Bilateral Procedures

#### Investigational Group (N=105)

Of the 105 procedures (55 patients) in the bilateral group, 99 procedures were theoretically due for the Month 24+ follow-up evaluation including one (1) procedure not yet overdue. Three patients (three hips) had one or more of the components of the Cormet Hip Resurfacing System revised or removed prior to the Month 24 visit and one patient is not yet overdue for the Month

{20}

24+ visit. Of the 95 procedures expected to be seen, 73 procedures (76.8%) had at least some clinical follow-up at Month 24 follow-up. Patients representing 22 procedures did not complete Month 24 follow-up.

## Control Group (N=83)

For the 52 patients (83 procedures) in the bilateral ceramic total hip control group, 82 hips were evaluated at Month 24+. No patients died or were revised. One patient was lost to follow-up. This patient was last seen at the Month 6 evaluation at which time a HHS of 97 points was reported.

## Continued Access (N=698)

The 640 patients representing 698 hips continue to be followed in the Continued Access study. Two hundred thirty-six procedures are expected due for Month 24+ evaluation and 32 procedures are not yet overdue for the Month 24+ evaluation. The Sponsor continues to enroll and follow patients according to the study protocol. To date, 18 patients have discontinued in the study; two patients died and 16 patients had revision to one or more components of the Cormet Hip Resurfacing System.

## Patient complaints

### Investigational Group

There were no specific complaints related to the use of the Cormet Hip Resurfacing System from patients involved in the IDE. The Short Musculoskeletal Functional Assessment (SMFA) is a 46-item assessment of patient function and a subset of 12 questions, the "bother index," is an assessment of how patients are bothered by functional problems. Study patients completed the SMFA during the course of the investigation. Overall there were large improvements in scores over time, which may be an indicator of general satisfaction.

Eight Product Experience Reports were received by the Sponsor for product marketed outside the US from December 2003 through November 10, 2005. There were no complaints by patients reported in this series.

### Control Group

Study results indicate that no patients were reported to specifically complain about their total hip replacement. However, this study asked the question: "Are you satisfied with the results of your surgery?" at each follow-up interval. Six patients (2.2%) answered "no" to this question at the Month 24 follow-up interval. Reasons included two incidences of patients involved in traumatic events, one post revision of femoral stem, one patient with multiple medical problems and two patients with pain at or near the operative site.

## Safety

### Operative Site Adverse Events

The safety of the Cormet Hip Resurfacing System was evaluated on the basis of adverse events (AEs) which were defined as any untoward medical occurrence during the course of the investigation including any unintended sign, symptom, or disease related to the device use. The All Enrolled cohort of 1148 procedures is used as the denominator for safety considerations of the investigational group and 349 for the control procedures even though 24 month follow-up data is not available on all patients.

21

{21}

# Intraoperative Events

Thirty-one cases (2.7%) reported intraoperative adverse events in the Cormet IDE study compared to 52 cases (14.9%) reported for the control population.

# Postoperative Events

Two hundred eight cases (18.1%) with postoperative hip related events were reported for the Cormet IDE population compared to 79 (22.6%) for the control population.

Fifty-eight cases (5.1%) with postoperative device related events were reported for the Cormet IDE population compared to 19 (5.4%) for the control population.

Table 17 provides a summary of intra-operative and postoperative site complications for the 1148 Cormet Hip Resurfacing System procedures and the ceramic total hip control.

Table 17: Summary of Complication Comparisons between All Enrolled Investigational and Control Devices

| | Investigational | | | Control | | | Exact p-value |
| --- | --- | --- | --- | --- | --- | --- | --- |
| | n^{8} | N^{9} | % | n^{8} | N^{9} | % | |
| Any complication | 427 | 1148 | 37.2% | 229 | 349 | 65.6% | 0.000 |
| Any hip-related complication | 219 | 1148 | 19.1% | 97 | 349 | 27.8% | 0.001 |
| Any device-related complication | 58 | 1148 | 5.1% | 27 | 349 | 7.7% | 0.064 |
| Any operative complication^{1} | 31 | 1148 | 2.7% | 52 | 349 | 14.9% | 0.000 |
| Any post operative complication^{2} | 412 | 1148 | 35.9% | 212 | 349 | 60.7% | 0.000 |
| Any post operative hip-related complication^{3} | 208 | 1148 | 18.1% | 79 | 349 | 22.6% | 0.063 |
| Any post operative device-related complication^{4} | 58 | 1148 | 5.1% | 19 | 349 | 5.4% | 0.782 |
| Any post operative serious complication^{5} | 104 | 1148 | 9.1% | | | | |
| Any post operative serious hip-related complication^{6} | 18 | 1148 | 1.6% | | | | |
| Any post operative serious device-related complication^{7} | 49 | 1148 | 4.3% | | | | |
| Deaths | 6 | 1148 | 0.5% | 5 | 349 | 1.4% | 0.142 |
| Notes:
1 Complications occurring during implant procedure as reported on Complication Form.
2 Complications occurring after implant procedure as reported on Complication Form.
3 Includes any post operative hip-related complication.
4 Includes any post operative complication.
5 Includes any post operative complication meeting the criteria for a serious complication as assessed by the investigator.
6 Includes any post operative complication hip-related complication assessed by the investigator as serious.
7 Includes any post operative complication.
8 Number of procedures with at least one of the specific types of complications.
9 Total number of procedures in this cohort of patients. | | | | | | | |

{22}

# Systemic Events

Systemic adverse events were those reported events that did not relate directly to the operation or the operative site/device. An analysis of the types of events reported and their time course showed expected rates of other body system complications for this type of population.

The control group’s higher rates of systemic AEs were distributed over time. These trends may be attributable to the greater maturity of the control database. In the control database, all patients had greater than 24 months of experience reported whereas only a fraction of the investigational group has experience past Month 24.

Table 18 provides a summary of systemic complications for the 1148 Cormet Hip Resurfacing System procedures and the ceramic total hip control.

Table 18: Systemic Adverse Events for All Enrolled Investigational and Control Devices

| | Investigational | | | Control | | | Exact |
| --- | --- | --- | --- | --- | --- | --- | --- |
| | n¹ | N² | % | n¹ | N² | % | p-value³ |
| Arrhythmia (operative) | 1 | 1148 | 0.1% | 0 | 349 | 0.0% | 1.000 |
| Bronchopulmonary | 2 | 1148 | 0.2% | 12 | 349 | 3.4% | <0.001 |
| Carcinoma | 4 | 1148 | 0.3% | 18 | 349 | 5.2% | <0.001 |
| Cardiovascular | 14 | 1148 | 1.2% | 33 | 349 | 9.5% | <0.001 |
| Death unrelated to device | 6 | 1148 | 0.5% | 5 | 349 | 1.4% | 0.142 |
| Deep Venous Thrombosis (DVT) | 9 | 1148 | 0.8% | 0 | 349 | 0.0% | 0.128 |
| Gastrointestinal | 8 | 1148 | 0.7% | 19 | 349 | 5.4% | <0.001 |
| Genitourinary | 8 | 1148 | 0.7% | 20 | 349 | 5.7% | <0.001 |
| Infection remote location | 10 | 1148 | 0.9% | 4 | 349 | 1.1% | 0.750 |
| Lack of nutrition | 1 | 1148 | 0.1% | 0 | 349 | 0.0% | 1.000 |
| Low hemoglobin/hematocrit | 3 | 1148 | 0.3% | 0 | 349 | 0.0% | 1.000 |
| Neuropathy | 1 | 1148 | 0.1% | 0 | 349 | 0.0% | 1.000 |
| Neurosensory | 8 | 1148 | 0.7% | 32 | 349 | 9.2% | <0.001 |
| Nosebleed | 1 | 1148 | 0.1% | 0 | 349 | 0.0% | 1.000 |
| Pulmonary Embolism (PE) | 4 | 1148 | 0.3% | 1 | 349 | 0.3% | 1.000 |
| Rash | 8 | 1148 | 0.7% | 10 | 349 | 2.9% | 0.003 |
| Thrombophlebitis | 0 | 1148 | 0.0% | 3 | 349 | 0.9% | 0.013 |
| Trauma (non-hip related) | 10 | 1148 | 0.9% | 30 | 349 | 8.6% | <0.001 |
| Varicose veins | 1 | 1148 | 0.1% | 0 | 349 | 0.0% | 1.000 |
| Other | 218 | 1148 | 19.0% | 102 | 349 | 29.2% | <0.001 |
| ¹ Number of procedures experiencing this type of complication
² Total population number, 24+ Month data only available on 532 procedures
³ Two-sided Fisher’s Exact tests. Comparisons were not performed for femoral neck notched (operative), greater trochanter notching (operative), ceramic insert chip (operative), and femoral neck fracture since both devices were not exposed to these types of events. Also, p-values are not reported when there were no events in either group. | | | | | | | |

# Hip Related Events

Hip related events were the most reported postoperative complications concerning the hip or operative site. Table 19 gives a breakdown of the rates of hip related complications for the investigational group and the corresponding control group.

Reviewing these events as they occurred over time, the rates of hip related complications seem to be higher in the control group in the immediate postoperative (up to Week 6) timeframe. However, there were more hip related complications in the investigational group at Week 6 to Month 6. The majority of events seem to be related to postoperative hip pain (bursitis, tendonitis and muscle weakness) at the operative site.

{23}

Table 19: Hip Related Adverse Events by Time Occurrence All Enrolled Procedures

| | Intra-operative | | Post Surgery to Week 6 | | Week 6 To Month 6 | | Month 6 to Month 12 | | Month 12 to Month 24 | | Post Month 24 | | Total | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| | I | C | I | C | I | C | I | C | I | C | I | C | I | C |
| Acetabular crack (operative) | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 |
| Acetabular malpositioned (operative) | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 3 | 0 | 0 | 0 | 4 | 0 |
| Broken drill bit | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 |
| Bursitis | 0 | 0 | 0 | 0 | 14 | 5 | 10 | 4 | 5 | 4 | 4 | 3 | 33 | 16 |
| Deep Infection | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 2 | 0 | 1 | 0 | 3 | 1 |
| Elevated metal ion level | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 |
| Femoral Crack (operative) | 0 | 12 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 12 |
| Femoral neck notched (operative) | 6 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 6 | 0 |
| Femoral radiolucency | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 6 | 0 | 5 | 0 | 12 | 0 |
| Greater Trochanter Notching (operative) | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 |
| Hematoma | 1 | 0 | 3 | 3 | 2 | 0 | 1 | 0 | 1 | 0 | 0 | 0 | 8 | 3 |
| Heterotopic Bone Formation | 0 | 0 | 2 | 6 | 4 | 5 | 0 | 0 | 7 | 1 | 0 | 1 | 13 | 13 |
| Hip Pain (operative side) | 0 | 1 | 15 | 2 | 17 | 3 | 10 | 1 | 12 | 1 | 7 | 1 | 61 | 9 |
| Leg Length Discrepancy | 1 | 0 | 7 | 0 | 8 | 0 | 1 | 0 | 3 | 0 | 2 | 0 | 22 | 0 |
| Limp | 0 | 0 | 7 | 0 | 5 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 13 | 0 |
| Loose Body | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 |
| Muscle Weakness | 2 | 0 | 2 | 0 | 5 | 1 | 0 | 0 | 1 | 0 | 0 | 0 | 10 | 1 |
| Myositis ossificans | 0 | 0 | 1 | 0 | 3 | 0 | 1 | 0 | 1 | 0 | 0 | 0 | 6 | 0 |
| Nerve palsy | 1 | 2 | 1 | 3 | 2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 4 | 5 |
| Skin split | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 |
| Soft tissue trauma | 0 | 0 | 1 | 0 | 0 | 2 | 0 | 2 | 0 | 6 | 1 | 4 | 2 | 14 |
| Squeaking implant/clicking | 0 | 0 | 2 | 0 | 10 | 0 | 4 | 0 | 4 | 1 | 0 | 1 | 20 | 2 |
| Subchondral cyst | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 1 | 0 |
| Subluxation | 0 | 0 | 1 | 0 | 1 | 0 | 1 | 0 | 0 | 0 | 3 | 0 | 6 | 0 |
| Superficial infection | 0 | 0 | 4 | 5 | 2 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 7 | 5 |
| Tendonitis | 0 | 0 | 1 | 1 | 7 | 1 | 3 | 3 | 6 | 1 | 3 | 0 | 20 | 6 |
| Trochanteric Crack (operative) | 0 | 7 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 7 |
| Wound Related (non-infected) | 0 | 0 | 17 | 16 | 1 | 0 | 2 | 1 | 2 | 0 | 0 | 0 | 22 | 17 |
| Other | 0 | 4 | 0 | 1 | 2 | 0 | 2 | 1 | 1 | 1 | 0 | 1 | 5 | 8 |

## Device Related Events Among Patients

The protocol definition of device related adverse event was “an adverse event that occurs due to the design and/or material composition of the implant and/or implant instrumentation.” From this definition, we have further refined the category to include:

- Bone breakage around the implanted components;
- Aseptic loosening of the components, including complete radiolucency around the stem or evidence of AVN under the femoral head;
- Breakage of the device components (stem fracture, acetabular liner fracture, etc.);
- Movement of the components in situ;
- Dislocation of the hip.

Table 20 gives a breakdown of the rates of device related AEs for the investigational group and the corresponding control group.

24

{24}

Table 20: Device Related Adverse Events by Time Occurrence All Enrolled Procedures

| | Intra-Operative | | Post Surgery to Week 6 | | Week 6 To Month 6 | | Month 6 to Month 12 | | Month 12 to Month 24 | | Post Month 24 | | Total | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| | I | C | I | C | I | C | I | C | I | C | I | C | I | C |
| Acetabular fracture | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 |
| Acetabular loosening | 0 | 0 | 3 | 0 | 3 | 0 | 0 | 0 | 3 | 0 | 2 | 0 | 11 | 0 |
| Avulsed lesser trochanter | 0 | 0 | 0 | 1 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 1 |
| Ceramic Insert Chip (operative) | 0 | 8 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 8 |
| Dislocation | 0 | 0 | 1 | 8 | 0 | 2 | 0 | 0 | 1 | 0 | 0 | 0 | 2 | 10 |
| Femoral fracture (operative) | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 |
| Femoral fracture (post-op) | 0 | 0 | 0 | 4 | 0 | 2 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 7 |
| Femoral loosening | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 7 | 0 | 6 | 0 | 14 | 0 |
| Femoral neck fracture | 0 | 0 | 3 | 0 | 12 | 0 | 5 | 0 | 5 | 0 | 1 | 0 | 26 | 0 |
| Femoral subsidence | 0 | 0 | 0 | 0 | 1 | 0 | 1 | 1 | 1 | 1 | 1 | 0 | 4 | 2 |
| Trochanter (greater) fracture | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 1 | 0 |

## Pivotal Study Bilateral Patients

Six device related events (5.7%) were reported for the Pivotal Study Bilateral Procedures in the investigational group at Month 24+ follow-up. Reasons for device related events included: acetabular loosening (one patient, 1.0%), femoral loosening (one patient, 1.0%), femoral neck fracture (two patients, 1.9%), and femoral subsidence (two patients, 1.9%).

There were no events of acetabular fractures, avulsed lesser trochanter, dislocation, intraoperative or postoperative femoral fractures, or greater trochanter fracture in the bilateral investigational group.

There were six device related events (7.2%) during a comparable time frame for the Pivotal Study Bilateral Procedures in the control group. Reasons for device related events included two events of operative ceramic insert chip (2.4%), three events of dislocation (3.6%), and one event of postoperative femoral fracture (1.2%).

There were no events of acetabular fracture, acetabular loosening, avulsed lesser trochanter, operative femoral fracture, femoral loosening, femoral neck fracture, femoral subsidence, and greater trochanter fracture in the bilateral control group.

## Continued Access Patients

Twenty device related events (2.9%) were reported for the Continued Access Cohort at Month 24+ follow-up. Reasons for device related events included: acetabular loosening (five procedures, 0.7%), dislocation (one procedure, 0.1%), femoral neck fracture (thirteen procedures, 1.9%), and femoral subsidence (one procedure, 0.1%).

There were no events of acetabular fracture, avulsed lesser trochanter, intraoperative femoral fracture, postoperative femoral fracture, femoral loosening or greater trochanter fracture in this cohort of procedures.

## All Enrolled Patients

Fifty-nine device related events among 58 procedures were reported for the All Enrolled Cohort at Month 24+ follow-up. Reasons for device related events included: acetabular loosening (eleven procedures, 1.0%), avulsed lesser trochanter (one procedure, 0.1%), dislocation (two procedures, 0.2%), femoral loosening (14 procedures, 1.2%), femoral neck fracture (26

25

{25}

procedure, 2.3%s), femoral subsidence (four procedures, 0.3%), and trochanter (greater) fracture (one procedure, 0.1%).

There were no events of acetabular fracture, intraoperative femoral fracture, postoperative femoral fracture in this cohort of procedures.

## Revisions

### Investigational Group (N=1148)

A revision is defined as an adverse event necessitating removal or replacement of the original surgical device. A revision is considered to be the most severe adverse event as it indicates total failure of the surgical procedure or device.

Twenty-four (24) revisions were noted in the Pivotal Unilateral group and forty-four (44) revisions were observed in the entire investigational group. Tables 21 and 22 identify the study cohort and reason for revision and or removal of study components.

Table 21: Revisions in Pivotal Unilateral, Pivotal Bilateral, Continued Access, All Enrolled and Control Procedures

| | Pivotal Unilateral (all procedures) | Pivotal Unilateral with Month 24+ Follow-up | Pivotal Bilateral (all procedures) | Continued Access (all procedures)* | Compassionate Use (all procedures) | All Enrolled (all procedures) | All Enrolled with Month 24+ Follow-up | Control All Enrolled with Month 24+ Follow-up |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Revisions | 24 | 24 | 4 | 16 | 0 | 44 | 44 | 5 |
| N | 337 | 302 | 105 | 698 | 8 | 1148 | 532 | 266 |
| % | 7.1% | 7.9% | 3.9% | 2.3% | 0.0% | 3.8% | 8.3% | 1.9% |

* Most continued access procedures have not been followed for 24+ Months.

Table 22: Reasons for Revision per Cohort

| | Pivotal Study Unilateral | Pivotal Study Bilateral | Continued Access | Compassionate Use | Total |
| --- | --- | --- | --- | --- | --- |
| Number | 337 | 105 | 698 | 8 | 1148 |
| Femoral Neck Fracture | 8 | 2 | 11 | 0 | 21 |
| Acetabular Component Loosening | 4 | 0 | 4 | 0 | 8 |
| Femoral Component Loosening | 11 | 0 | 0 | 0 | 11 |
| Deep Joint Infection | 0 | 1 | 1 | 0 | 2 |
| Dislocation | 1 | 0 | 0 | 0 | 1 |
| Femoral Subsidence | 0 | 1 | 0 | 0 | 1 |
| Total | 24 | 4 | 16 | 0 | 44 |

Considering the denominator of the entire pivotal group as 337 procedures, the revision rate is 7.1%. However, only 302 of the pivotal group procedures had Month 24+ follow-up available, making the revision rate for the pivotal unilateral group 7.9% (24/302).

Considering all enrolled procedures, the estimate of 3.8% based on 1148 enrolled procedures is a best case scenario because many of the continued access subjects were not yet due for their 2 year follow-up, yet they are considered revision free. Only 532 of the All Enrolled procedures had Month 24+ follow-up available, making the revision rate for the All Enrolled Group 8.3% (44/532).

26

{26}

27
27

# Control Group (N=349)

Five patients (1.4%) were reported to have revision of one or more components of the ceramic total hip system. These patients were among the 266 unilateral patients. No patient in the bilateral group was revised; the reasons for revision are as follows:

- For one patient, the femoral component and alumina head were revised. The patient fell approximately one month post surgery and sustained a periprosthetic fracture, which was treated with Open Reduction Internal Fixation (ORIF) and a cast. The patient went on to non-union and the femoral component and head were revised nine months post original surgery.
- For one patient, the acetabular component, insert and femoral head were revised due to recurrent anterior dislocation five days post original surgery.
- One patient had all components revised due to deep joint infection approximately 10 months post implantation. The patient was reported to have a girdlestone procedure performed after the components were removed.
- One patient had all components removed due to hip pain approximately three years postoperatively. Sepsis was suspected, but not confirmed.
- One patient had femoral stem and head revised 4.5 years postoperative due to stem loosening post a traumatic event that took place 15 months after the index procedure.

# Survival Analysis

The primary study cohort for safety is considered the All Enrolled Procedures group. The distribution of study device failure over time was summarized by constructing a life-table that indicated the number of failures and the number of at-risk procedures over time. The distribution of failures over time was further assessed using Kaplan-Meier survival curves in order to provide graphical representations of survivorship over time.

The primary survival analysis defined device failure as any revision, no matter the model of failure, censoring only at death. The groups exhibit a Month 24 device survivorship of 99% and 96% for the control and investigational groups respectively. This was statistically significant in favor of the control (p&lt;0.01). A survival curve on the 337 subjects in the Pivotal Study unilateral cohort showed that the survival at 24 months was 95%, virtually identical to that of the All Enrolled cohort.

{27}

FIGURE 1
KAPLAN-MEIER SURVIVAL CURVE:
ALL ENROLLED INVESTIGATIONAL AND CONTROL DEVICES

# Risk Factor Analysis

The revision rate is 8.3% (44/532) if only procedures with Month 24+ follow-up are taken into consideration. A post-hoc subgroup analysis (Table 23) showed that within this patient cohort certain patients were at greater risk of experiencing a revision than other patients. Males had a lower revision rate than females (6.5% vs. 12.9%). Further, patients in whom a smaller component was implanted (40 or 44mm), patients with a diagnosis other than Osteoarthritis (OA, i.e., Avascular Necrosis, Rheumatoid Arthritis), patients with significant leg length discrepancy (≥ 1cm) and baseline HHS in the lowest quartile of function all had revision rates greater than the overall average of 7.9% for the Pivotal Unilateral group or 8.3% for the All Enrolled group with a Month 24+ follow-up.

Table 23 provides the risk of revision in the Pivotal Unilateral group and for the All Enrolled patients who received an investigational device.

{28}

Table 23: Risk of Revision in Pivotal Unilateral Cohort and All Enrolled Procedures

| | | Pivotal Unilateral (all procedures) | Pivotal Unilateral with Month 24+ Follow-up | All Enrolled (all procedures) | All Enrolled with Month 24+ Follow-up |
| --- | --- | --- | --- | --- | --- |
| Gender | Female | 11.9% (13/109) | 12.8% (13/102) | 6.5% (21/323) | 12.4% (21/170) |
| | Male | 4.8% (11/228) | 5.5% (11/200) | 2.8% (28/825) | 6.4% (23/362) |
| Small Component Size | 40/44 mm | 16.7% (13/178) | 17.3% (13/75) | 7.4% (22/296) | 15.2% (22/145) |
| | >40/44 mm | 4.3% (11/259) | 4.9% (11/227) | 2.6% (22/843) | 5.7% (22/387) |
| Non Osteoarthritis Diagnosis | AVN | 13.6% (6/44) | 15.8% (6/38) | 6.9% (8/116) | 12.7% (8/63) |
| | RA | 25.0% (1/4) | 25.0% (1/4) | 11.1% (1/9) | 14.4% (1/7) |
| | Osteoarthritis | 5.9% (17/289) | 6.5% (17/260) | 3.4% (35/1023) | 7.6% (35/462) |
| Leg Length Discrepancy greater than or equal to 1 cm | ≥ 1 cm | 13.0% (12/92) | 14.5% (12/83) | 6.1% (18/296) | 14.0% (18/129) |
| | <1 cm | 4.9% (12/245) | 5.5% (12/219) | 3.1% (26/849) | 6.5% (26/403) |
| Baseline lowest quartile of function (HHS) | < 42.58 | 17.7% (15/85) | 20.3% (15/74) | 6.4% (18/283) | 13.1% (18/137) |
| | ≥ 42.58 | 3.6% (9/252) | 4.0% (9/228) | 3.1% (26/846) | 6.7% (26/391) |
| Among 1^{st} 25 procedures within a specific site | First 25 | 8.2% (12/147) | 8.9% (12/135) | 6.8% (16/234) | 8.3% (16/192) |
| | After 1^{st} 25 | 6.3% (12/190) | 7.2% (12/167) | 3.1% (28/914) | 8.2% (28/340) |

Twenty-one revisions occurred among the 323 procedures for female patients (6.5%) compared to 23 revisions for the 825 procedures for male patients (2.8%; p=0.015). Crude (single predictor variable) survival analyses were performed in order to account for unequal follow-up. Analyses revealed a hazard ratio for risk of revision equal to 2.1 (95% CI 1.2 to 3.9; p=0.01) comparing females to males. There were 22 revisions of patients with size 40mm or 44mm femoral components (296 procedures; 7.4%) compared to 22 revisions for patients implanted with sizes 48mm, 52mm or 56mm (843 procedures; 2.6%) [Note: implant size was unavailable for nine patients]. The hazard ratio for risk of revision was equal to 2.8 (95% CI 1.6 to 5.1; p=0.0006) comparing size 40mm or 44mm to larger sizes. When the effects of gender and size of component are simultaneously estimated, component size (p=0.02) but not gender (p=0.81) retains statistical significance. Reduced baseline function was associated with increased revision risk. The hazard ratio comparing patients in the lowest quartile of HHS scores (&lt;43) to those with higher scores was 2.0 (95% CI 1.1 to 3.6; p=0.03). Having a preoperative leg length discrepancy ≥ 1cm was also associated with increased risk of revision. The crude hazard ratio was 2.2 (95% CI 1.2 to 3.9; p=0.01).

As an additional post-hoc analysis, the initial twenty-five (25) procedures at each center were evaluated to determine whether a learning curve could explain the number of revisions noted in the study. However, evaluation of the procedures with adequate follow-up data did not reveal revision rates to be significantly affected by a learning curve.

Two investigative sites (site 5 and site 10) had higher revision rates. These sites accounted for 17 of the 44 (38.6%) revisions in the study but only 250 of 1148 total procedures (21.8%). There were 27 revisions among the remaining 930 procedures (2.9% revision rate). When site 5 and site 10 were excluded, the small component size and female gender hazard ratios were 3.3 (95% CI 1.6 to 7.0, p=0.002) and 2.5 (95% CI = 1.2 to 5.4; p=0.02), respectively. Therefore, removing these two sites from the analysis did not affect the clinical or statistical significance of these two risk factors.

{29}

Patient age, diagnosis and preoperative weight were associated with increased risk for revision when all sites were included in the analysis. However, when site 5 (15 of the 44 revisions) was removed from the analysis, diagnosis other than osteoarthritis, emerged as a statistically significant risk factor in both the pivotal study unilateral patients and all enrolled procedures. The hazard ratio was 2.8 (95% CI 1.3 to 6.2) comparing procedures with diagnoses other than osteoarthritis (i.e., AVN and rheumatoid arthritis) to those with osteoarthritis after excluding site 5. A diagnosis other than osteoarthritis was a significant risk factor in the Pivotal Study Unilateral cohort (hazard ratio=2.5, 95% CI = 1.0 to 5.9). All of the other risk factors noted above were also statistically significant in the Pivotal Study Unilateral cohort. Table 24 provides the prevalence of risk factors for the Pivotal Unilateral group and for the All Enrolled patients with and without Site 5.

Table 24: Prevalence of Risk Factors for Pivotal Unilateral and All Enrolled Patients with and without excluding Site 5

| | | Pivotal Unilateral | Pivotal Unilateral Excluding Site 5 | Pivotal Unilateral Site 5 Only | All Enrolled | All Enrolled Excluding Site 5 | All Enrolled Site 5 Only |
| --- | --- | --- | --- | --- | --- | --- | --- |
| Small Component Size (40 or 44mm) | % | 23.1% | 22.7% | 26.3% | 26.0% | 24.0% | 35.6% |
| | n | 78 | 68 | 10 | 296 | 227 | 69 |
| | N | 337 | 299 | 38 | 1139 | 945 | 194 |
| Non Osteoarthritis Diagnosis | % | 14.2% | 15.7% | 2.6% | 10.9% | 12.2% | 4.6% |
| | n | 48 | 47 | 1 | 125 | 116 | 9 |
| | N | 337 | 299 | 38 | 1148 | 954 | 194 |
| Leg Length discrepancy ≥ 1cm | % | 27.3% | 19.7% | 86.8% | 25.9% | 13.8% | 85.1% |
| | n | 92 | 59 | 33 | 296 | 131 | 165 |
| | N | 337 | 299 | 38 | 1145 | 951 | 194 |
| Baseline lowest quartile of function (HHS) | % | 25.2% | 21.4% | 55.3% | 25.1% | 24.8% | 26.3% |
| | n | 85 | 64 | 21 | 283 | 233 | 50 |
| | N | 337 | 299 | 38 | 1129 | 939 | 190 |
| Among 1^{st} 25 procedures within a specific site | % | 43.6% | 41.8% | 57.9% | 20.4% | 21.9% | 12.9% |
| |…