Cement technique changes improved hip resurfacing longevity: implant retrieval findings.
Abstract: Background: Most designs of metal-on-metal hip resurfacing utilize cement for femoral fixation, but the amount, application, and distribution of cement varies considerably according to implant design and surgeon preference. In one type of hip resurfacing system (Conserve[R] Plus), the objective was to achieve a 1-mm cement mantle and several millimeters of penetration. In early cases of the senior investigator's (HCA) series, cement fixation failures were noted, and this prompted changes in femoral head preparation and cement application techniques. Survivorship improved following implementation of these changes. The aim of the current study was to examine revised femoral components for the cement distribution, especially in cases where the improved techniques were subsequently applied.

Method: Fifteen resurfacing femoral components were sectioned and the slices were radiographed and photographed, and the amount and distribution of cement were examined and measured. Cases representative of the evolving cementing techniques were examined in detail.

Results: There was considerable variation observed in the amount and distribution of cement, partly as a consequence of variable bone quality in this "all-comers" included series. The cement analyses showed that the newer cementing techniques helped to reduce over-penetration while providing better cement interdigitation. The use of extra fixation holes and cementing the stem in cases with poor bone quality were associated with improved cement-to-bone contact area.

Conclusion: Meticulous femoral head preparation was helpful in providing durable cement fixation and is essential in cases with poor bone quality.
Subject: Implants, Artificial (Methods)
Prosthesis (Methods)
Surgeons (Methods)
Authors: Campbell, Pat
Takamura, Karren
Lundergan, William
Esposito, Christina
Amstutz, Harlan C.
Pub Date: 04/01/2009
Publication: Name: Bulletin of the NYU Hospital for Joint Diseases Publisher: J. Michael Ryan Publishing Co. Audience: Academic Format: Magazine/Journal Subject: Health Copyright: COPYRIGHT 2009 J. Michael Ryan Publishing Co. ISSN: 1936-9719
Issue: Date: April, 2009 Source Volume: 67 Source Issue: 2
Product: SIC Code: 3842 Surgical appliances and supplies
Accession Number: 247449427
Full Text: Current designs of hip resurfacing prostheses mostly rely on bone cement for femoral fixation, but they will differ in the choice of cement type, volume, degree of penetration, and clearance between the implant and bone (cement mantle). Generally, there are two schools of thought regarding femoral cement fixation: one school advocates minimal amounts of cement and a few millimeters of penetration, similar to that used in total knee component fixation, while the other school advocates abundant cement and deep penetration into the femoral head bone. Some designs, such as the Conserve[R] Plus (Wright Medical Technology, Arlington, Tennessee), allow for a 1 -mm cement mantle to facilitate a more uniform pressurization and penetration of doughy cement, which is applied using manual pressure into the reamed areas of bone.

The systems that have a tighter fit of the femoral component to the prepared bone permit a very thin cement mantle, if at all; this requires the use of low viscosity cement applied via the component as it is pushed into the reamed bone. The Birmingham Hip Resurfacing (BHR) prosthesis (Smith and Nephew, Arlington, Tennessee) uses this technique. Bench studies have been performed to assess the adequacy of fixation achieved by these dissimilar cementing approaches, and it has been noted that differing cement application techniques result in variable degrees of cement penetration. (1)

Implant retrieval studies have emphasized the variability of the degree of cement penetration regardless of implant design and the degree of intended cement penetration. (2) Retrieval studies have suggested as well that cement technique plays an important role in component longevity. Components that were incompletely seated were noted to be at risk for fracture; inadequate cement fixation could lead to component loosening and excessive cementing could also contribute to loosening by thermal necrosis-induced interface membrane formation. The risk for thermal necrosis has been verified by thermal probe studies performed during hip resurfacing, (3) but its role in long-term survivorship is controversial.

In the early series of hip resurfacing performed by the senior investigator (HCA), the cohort included patients with poor quality femoral head bone, including severe cystic degeneration from advanced osteoarthritis and osteonecrosis. (4) These circumstances were due to the procedure largely being done on an "all comers are welcome" basis. Analysis of the first 400 cases showed that femoral fixation failures were an early cause of failure. This was in contrast to failures reported by other resurfacing centers where short-term femoral neck fractures were responsible for many failures. (5) In response to the retrieval and histological findings of loosened femoral components, which included apparent areas with very little penetration and interfacial membranes, (2) technique changes were instituted by the investigator. These included improvements in the preparation of the femoral head by using more meticulous debridement of all soft tissue with a high speed burr, cleaning with selective lavage, employing dome and trochanteric suction, and using carbon dioxide blow drying (CarboJet[R] Kinamed, California) to further clean and dry the bone. (6-8) The goal of fine tuning the technique was to eliminate any tissue or moisture that could deter optimal cement interdigitation with the femoral head bone. These changes were found to improve outcomes, and femoral fixation failures declined sharply. (7)

Several retrieval specimens from patients whose components were inserted using these improvements became available through our "Willed Joint" program and provided the opportunity to examine the cementing technique in explanted specimens. The aims of this study were to describe the retrieval findings of hip resurfacing retrieval specimens, to compare the old and new femoral preparation and cement techniques, and to provide surgeons with cementing guidelines that may be helpful in providing long-term durable fixation.

Materials and Methods

The Conserve[R] Plus (Wright Medical Technology, Arlington, Tennessee) surgical technique and follow-up regimen have been described in detail elsewhere. (8,9) All of the cases included in the following analyses were performed through the posterior approach and utilized Simplex P regular viscosity cement. All of the femoral heads were photographed after they had been prepared prior to cementation, which provided a means of comparing the effectiveness of the evolving femoral preparation technique changes.

Retrievals

At the time of revision surgery, the femoral components were resected with a portion of the femoral neck where possible and immediately fixed in 10% buffered formalin. They then were sent to a retrieval laboratory where the femoral components were sectioned using a water-cooled band saw into at least three 3-mm thick coronal sections through the anterior, middle, and posterior segments. The sections were radiographed, photographed, and submitted for decalcified histological analysis.

Twenty-six Conserve[R] Plus surface arthroplasty specimens (implanted between 1996 and 2006 from our series of 1000) were submitted to our implant retrieval laboratory between 1997 and 2008. Fifteen specimens had coronal sections that were suitable for cement analysis; the others were incomplete specimens or were cut transversely, and were not suitable. The demographics of these patients are shown in Table 1. This group included cases that had been implanted with the cement technique used in the first 300 implantations: Case 1, described below; one patient with two components, Case 2, which had staged bilateral resurfacings using improved preparation (second-generation technique that included improved bone preparation, a single suction in the dome, and sponge drying); and two patients, Cases 3 and 4, both of whom received the improved third-generation technique, including the use of the CarboJet[R] and intertrochanteric as well as dome suction All four patients were male, with an average age of 60 years (range, 49 to 64 years). These cases are detailed below. In addition, one BHR implant, Case 5, that was revised by the senior investigator was sectioned to provide a comparison to the Conserve[R] Plus specimens. This case is also detailed below.

Cement Analysis

Cut sections were immersed in 70% ethanol for approximately 48 hours to enhance the visual differentiation of the cement and bone. The cut section photographs and microradiographs were scanned into a computer. An image analysis software program (MetaMorph, Version 4.6,

Universal Imaging Corp., Downingtown, Pennsylvania) was used to measure the total area of cement within each cut section (Fig. 1). This included the cement mantle, penetrated bone, and any cement-filled fixation holes or cysts. The cement mantle was defined as the cement layer between the metal and the outer edge of the prepared surface of the bone; in the Conserve[R] Plus, this is designed to be 1 mm. The depth of cement penetration into cancellous bone was considered to be the interdigitation of cement into cancellous bone, starting from the outer edge of the prepared (cut) bone.

[FIGURE 1 OMITTED]

In the hips used to demonstrate the evolution of the femoral head preparation technique, the thickness of the cement mantle, and the depth of penetration into the bone, were measured at 10 sites in each section, including the wall, chamfer, and dome (Fig. 1). To assess whether the pegs and the cementation of the femoral stem added to the fixation area, the depth of penetration from the edges of the peg or the cemented stem were measured, along with the perimeter of cement in contact with bone.

Results

There was considerable variation in the amount and distribution of cement within and between the specimens. Typically, there was more cement in the chamfers and dome areas (thicker mantles and deeper penetration) and the least cement at the edge of the components (Table 2). The average area of the bone sections occupied by cement for the entire group ranged from 23% to 63% (data not shown). This variability was associated often with the location of the cut sections, since the outermost cuts frequently included the edges of the femoral head where there was more irregularity in the shape of the femoral head, resulting in focally thick cement. The middle sections of the heads tended to have less cement. Thick areas of cement often included air bubbles.

There were insufficient specimens to allow a statistically robust comparison of the early, improved, and third-generation cementing techniques, and it must be emphasized that many factors, including bone quality, the number and size of cement-filled pegs, and defects, will affect the cement measurements. Thus, the following data should be considered qualitative at best. The thickness of the cement mantle along the walls of components applied with older techniques was typically greater than in those procedures done with improved techniques. The degree of seating at the dome was better in the newer technique specimens, averaging 1.4 mm, while in the older technique specimens, the gap between the implant and the bone averaged 3.4 mm. The average depth of cement penetration was comparable within all of these specimens (i.e., combining the measurements from all locations within the slices and in each of the three slices), but varied according to location within the head (i.e., the domes and chamfers having deeper penetrated cement than edges, as would be expected). The average area occupied by cement was greater in the older technique (44% ) compared with the newer technique group (31%), possibly reflecting better seating and less excess cement at the dome.

Examination of the pegs and the cemented stems confirmed that they increased the depth of cement penetration by an average of 1.1 mm and increased the overall contact between cement and bone. There did not appear to be any fractured pegs, despite their relatively small (3 mm) diameter.

Case Studies Case 1

A 31-year-old male required treatment for osteoarthritis secondary to Perthes disease and had a simultaneous bilateral hip resurfacing, despite concerns about femoral head quality due to severe, superior femoral head erosion and cystic degeneration. This patient was treated early in our series, and at that time femoral head preparation consisted of manual curettage of the cystic bone, some water lavage, dome suction alone, and sponge drying (Fig. 2A). In addition, stems also were not cemented at that point in time, despite large defects. The results for this case were disappointing, even though the patient returned to an active lifestyle and continued to mountain bike race for 3 years postoperatively when he developed some right hip pain after heavy activity. The radiographic follow-up indicated suboptimal fixation, with circumferential stem radiolucencies of both hips and eventual right hip femoral component migration.

[FIGURE 2 OMITTED]

[FIGURE 3 OMITTED]

The femoral components were revised to bilateral conventional total hip arthroplasties after 55 months. The left hip, which had the least femoral head erosion of the two hips at surgery and was the least symptomatic at the time of revision, had the best femoral specimen for study. The cut sections of the head clearly showed that there was often minimal penetration, despite large amounts of cement that filled the erosion defect (average area, 51%; Table 2) (Figs. 2B and C). Thick interfacial membranes were present and the bone was undergoing resorption, consistent with loosening.

Case 2

This case represents a second-generation technique that included better bone preparation, using a high-speed burr, single dome suction, sponge drying, and the application of stem cement for additional fixation. The patient was a 57-year-old male, who had bilateral hip resurfacing implants 4 months apart for osteoarthritis secondary to Perthes disease. Bone quality at the time of surgery was thought to be adequate, as shown in intraoperative photographs (Fig. 3A, right hip; Fig. 4A, left hip), but the stems were cemented to help improve fixation. A 52-mm femoral component was used on the left femoral head, and a 48-mm component was used on the right. The patient died 7 years postoperative and donated his hips to our Willed Joint program.

The components were sectioned and analyzed, as described above. Figure 3 shows the gross and radiographic appearance of the middle of the resurfaced left head, while Figure 4 shows the right side. The results of cement measurements are provided in Table 2. Areas of thick cement and defects within these thick areas are clearly seen. It appears that the right femoral head contains more cement than the left, and that there is a proximal necrotic segment partly separated from the underlying bone in the anterior section, with more normal bone in the middle and posterior sections. Histology confirmed that the bone and marrow below the cement in the dome were necrotic for several millimeters in the anterior segment. Below this dead area, there was a zone of fibrosis, consistent with "walling-off," but beyond the fibrotic zone, the bone was vascular and viable. The cemented stem interface consisted of a thin, fibrous membrane that contained histiocytes and occasional giant cells with cement debris. The bone had remodeled along the stem, often following the contours of the cement, and was noticeably thickened inferiorly and distally. Thick new bone also surrounded some of the cement pegs.

In the left femoral head, the bone-cement interface varied between giant-cell lined fibrotic marrow in the middle section and thin fibrous membrane and necrosis underlying the cement, which was more limited. The bone quality overall was more normal and the cement penetration was noticeably less. The cemented stem interface also had a thin fibrous membrane and a surrounding, thickened bone layer following the contour of the cement.

The next two cases were performed after changes to the cementing technique had been incorporated into the operative procedure, including more extensive lavage, drying with dome and intertrochanteric suction, and the use of stem hole and exterior femoral head carbon dioxide blow drying with the CarboJet[R].

Case 3

A 43-year-old male received a hip resurfacing for osteoarthritis. The head was reamed to accept a 48-mm femoral component and was judged intraoperatively to have good bone quality (Fig. 5A). The clean, dry appearance of the bone prior to cementing can be seen in the intraoperative photograph. Unfortunately, a revision was required after 1 year for hematogenous sepsis after a failed debridement for Actinomyces infection. The sectioned femoral component revealed a well-seated component, good bone quality, and well-integrated cement interfaces (Figs. 5B and C). Histological examination verified the absence of fibrous interface membranes and showed that bone had remodeled close to the cement. Cement measurements are shown in Table 2.

[FIGURE 4 OMITTED]

[FIGURE 5 OMITTED]

Case 4

Case 4 was one of the first patients to receive a prosthesis with a technique similar to that used by surgeons using a tight European fit. This European technique was designed to provide a mantle greater than 0.5 mm, and it was used in this study with the aim of evaluating the safety and efficacy relative to this method in improving cement penetration through increased pressurization of a tighter fit. The patient was a male and 61 years of age at the time he received a 50-mm femoral component for osteoarthritis. The bone quality of the femoral head was excellent, with no cystic disease. Multiple drill holes were placed into the dome and chamfered areas (Fig. 6A). The cement was applied very early in the dough stage. Unlike our standard technique with a 1-mm cement mantle, where hand pressurization is sufficient, hammer blows were necessary to seat the component fully.

[FIGURE 6 OMITTED]

Unfortunately, the neck of the left hip fractured at 2 months after only minor trauma. The fracture occurred within the component, and sectioning and histology revealed a large proximal necrotic segment occupying most of the bone remaining in the component (Figs. 6B and C). Near the fracture line, there was viable granulation tissue, edema, active osteoclastic remodeling, and new bone formation on dead bone remnants, consistent with fracture repair. The area occupied by cement was relatively low at 26%; therefore, over-penetration was not thought to have contributed to the necrosis or the fracture, and the cause is still unclear. Cement penetration was not noticeably increased, despite the tightfit design. This unexpected fracture, along with one other of the 74 hips implanted with this technique, prompted the investigator to return to the standard 1-mm mantle design.

Case 5

The senior investigator did not implant this 42-mm BHR prosthesis, but the patient, a 63-year-old female, sought a revision for pain associated with acetabular loosening at 16 months following her surgery. The femoral head sections showed the deep penetration of cement that is typical of the BHR technique with a tight fit and low viscosity cement (Figs. 7A and B), but interfacial membranes indicated that loosening was in progress. This was confirmed histologically. The bone and marrow closest to the cement were necrotic, and a thick membrane was present; however, beyond the necrosis was the presence of ongoing, appositional new bone formation with consolidation and thickening. There was osteoclast activity along the membrane-bone interface. The stem tip was encased in reactive, woven thick bone, which was lined by a membrane and focal pockets of bone debris consistent with loosening. In both the anterior and posterior sections, there was a disassociation of extensively cement penetrated bone from the more distal cylindrically reamed bone.

[FIGURE 7 OMITTED]

Discussion

Cementing technique, bone quality, and femoral head preparation are important factors that will determine cement penetration, durable fixation, and femoral component survivorship in hip resurfacing. Our previous retrieval studies, performed on a range of hip resurfacing designs, noted extensive variability, which suggested that the control of cement mantle thickness and penetration may be difficult. (2) A review of the failed cases within the senior investigator's patient cohort support the clinical findings that technique changes have led to more uniform and better cement penetration, with reduction in membrane formation and improved overall fixation.

The illustrated cases presented above emphasize the importance of intraoperative photographs of the head prior to cementation so that more accurate comparisons and interpretation of the retrieval analysis can be made. Our histological studies also demonstrate the importance of performing multiple sections (three minimum) for every case to evaluate the changes and variability of bone remodeling processes. The actual measurement of cement penetration was often difficult because of the lack of color and contrast between cement and yellow marrow or cancellous bone. The well-oriented decalcified bone sections helped to verify the extent of cement penetration and the immediate interface reaction to the cement.

The desire to provide a more uniform cement penetration into cylindrical reamed areas emphasizes the importance of bone preparation (removal of all tissue, lavage, and drying) and the hand pressurization of doughy cement to optimize fixation. Inadequate preparation of the femoral head was associated with a high risk for loosening, as seen in early failures in the series (2,4) and illustrated by Case 1. Since the technique improvements were adopted (generations two and three), this type of cementing has worked well in over 1000 cases in the senior investigator's series. The improved efficacy of generation three (CarboJet[R] plus cementing the stem for defects greater than 1 cm) has not yet been established, but the improvements in cement interdigitation shown in Cases 3 and 4 bodes well for long-term durability.

This cement method is very different from the BHR technique, in which a large volume of low viscosity cement is applied early and a smaller reamed chamfer area is prepared. Various studies have examined these two disparate methods to provide surgeons with cement technique guidelines. Bitsch and colleagues (10) used two cement viscosities, six different cementing techniques, and two different porosities of open-cell foam specimens to test the efficacy of uniform cement penetration. They concluded that higher viscosity (later cement application) resulted in greater penetration of the outer wall, mid-stem, interior area, and distal stem; whereas, filling half of the component with low and regular cement viscosity led to incomplete seating, deep penetration, and a decreased outer wall penetration. Howald and associates (11) used nine fresh frozen, paired whole cadaver femora and Durom (Zimmer, Warsaw, Indiana) resurfacing components (which have a 1-mm mantle like the Conserve[R] Plus) to compare the penetration achieved with two low viscosity bone cements, using standing times of 1.5 minutes and 3 minutes. They found that cement penetration increased as the cement standing increased from 1.5 to 3 minutes. However, both would be considered to be in the early low viscosity stage, so that a slight increase in viscosity aided penetration. Beaule and coworkers,1 using paired cadavers, compared cement penetration of five different resurfacing components, employing the manufacturers' recommended cementing techniques and then switching to low viscosity for the Conserve[R] Plus and higher viscosity for the BHR. They found greater penetration in the BHR, using both low and regular viscosity cement in the dome area. Both Beaule and colleagues1 and Bitsch and associates10 found more uniform penetration with a 1-mm cement mantle technique.

These studies also raise the question of whether or not thermal necrosis does occur and, if so, is the amount of cement penetration a factor? There is an apparent paradox with the early to mid-term clinical success of the BHR, despite the reported risk of thermal necrosis and membrane formation, or fracture associated with overpenetration of cement. (2,10) Extensive interfacial thermal necrosis and thick membranes in two of the hips studied here (Case 2) did not lead to failure by loosening, as judged by the findings in the 7-year postoperative Willed Joint program specimens. They had second-generation bone preparation and cementing technique, but the bone quality was adequate after preparation and, in addition, the stems were cemented in. This decision to cement the stems was made as part of our plan to evaluate the safety and efficacy of stem cementation.

The BHR (Case 5) illustrates the potentially negative effects of cement over-penetration in the dome. Retrieval analysis demonstrated disassociation of the dome bone and the distal head in the anterior and posterior sections. Although somewhat tenuous, the neck had not fractured because of the integrity of the neck in the middle section. In Case 1, where there was poor quality bone and inadequate first-generation technique, the resurfacing may have benefited if the stem had been cemented at the time of surgery.

Conclusion

We feel there is sufficient evidence to show that poor femoral head preparation and lack of uniform cement penetration places certain resurfacing femoral components at risk for loosening. It is not possible to prescribe precisely the "perfect" depth, but we believe that 2 to 4 mm is sufficient and greater than approximately 5 mm is probably excessive. The senior investigator's current recommendations (thirdgeneration technique) include removal of all cystic and soft tissue with a high speed burr, jet lavage with low pressure if the bone is osteopenic to avoid marrow washout, CarboJet[R] blow drying (again, adjusting the pressure according to bone density), and cementing the stem in those patients with small components (less than 48 mm) and in those with cystic degeneration more than 1 cm. Continued implant retrieval studies are recommended to allow the usefulness of evolving techniques to be evaluated.

Acknowledgment

These studies were funded by the Orthopaedic Hospital Foundation (Los Angeles, CA) and Wright Medical Technology, Inc. (Arlington, TN).

Disclosure Statement

None of the authors have a financial or proprietary interest in the subject matter or materials discussed, including, but not limited to, employment, consultancies, stock ownership, honoraria, and paid expert testimony.

References

(1.) Beaule PE, Matar WY, Poitras P, et al. 2008 Otto Aufranc Award: component design and technique affect cement penetration in hip resurfacing. Clin Orthop Rel Res. 2009;(467):84 93.

(2.) Campbell P, Beaule P, Ebramzadeh E, et al. A study of implant failure in metal-on-metal surface arthroplasties. Clin Orthop Rel Res. 2006;(453):35-46.

(3.) Gill HS, Campbell PA, Murray DW, De Smet KA. Reduction of the potential for thermal damage during hip resurfacing. J Bone Joint Surg Br. 2007;89:16-20.

(4.) Amstutz HC, Beaule PE, Dorey FJ, et al. Metal-on-metal hybrid surface arthroplasty: two to six-year follow-up study. J Bone and Joint Surg Am. 2004;86:28-39.

(5.) Shimmin AJ, Back D. Femoral neck fractures following Birmingham hip resurfacing: a national review of 50 cases. J Bone and Joint Surg Br. 2005;87:463-4.

(6.) Amstutz HC, Le Duff MJ. Cementing the Metaphyseal Stem in Metal-on-Metal Resurfacing: When and Why. Clin Orthop Relat Res. 2009 Jan;467(1):79-83. Epub 2008 Oct 30.

(7.) Amstutz HC, Le Duff MJ. Eleven years of experience with metal-on-metal hybrid hip resurfacing: a review of 1000 conserve plus. J Arthroplasty. 2008;23:36-43.

(8.) Amstutz HC. Surgical technique. In: Amstutz HC (ed): Hip Resurfacing Principles, Indications, Technique and Results. Philadelphia: Saunders Elsevier, 2008, pp. 77-94.

(9.) Amstutz HC, Beaule PE, Dorey FJ, et al. Metal-on-Metal Hybrid Surface Arthroplasty. Surgical Technique. J Bone Joint Surg Am. 2006 Sep;88(Suppl 1 Pt 2):234-49.

(10.) Bitsch RG, Heisel C, Silva M, Schmalzried TP. Femoral cementing technique for hip resurfacing arthroplasty. J Orthop Res. 2007;25:423-31.

(11.) Howald R, Kesteris U, Klabunde R, Krevolin J. Factors affecting the cement penetration of a hip resurfacing implant: an in vitro study. Hip Int. 2006;16:S82-9.

Pat Campbell, Ph.D., Karren Takamura, B.A., William Lundergan, B.A., Christina Esposito, B.A., and Harlan C. Amstutz, M.D. Pat Campbell, Ph.D., Karren Takamura, B.A., William Lundergan, B.A., and Christina Esposito, B.A., are from the J. Vernon Luck Orthopaedic Research Center, the UCLA and Orthopaedic Hospital Department of Orthopaedic Surgery, Los Angeles, California. Harlan C. Amstutz, M.D., is from the Joint Replacement Institute, St. Vincent Medical Center, Los Angeles, California.

Correspondence: Harlan C. Amstutz, M.D., Medical Director, Joint Replacement Institute, The S. Mark Taper Building, 2200 West Third Street, Suite 400, Los Angeles, California 90057; harlanamstutz@dochs.org.
Table 1 Patient Demographics

No. of Patients            N = 15
Gender                     F = 8, M = 7
Average time to revision   55 months (2-99)
Failure mode               n = 1 Acetabular loosening
                           n = 2 Autopsy
                           n = 5 Femoral neck fracture
                           n = 1 Infection
                           n = 1 Lysis
                           n = 5 Femoral loosening
Average ball size          F = 41.4 (36-46)
                           M = 48.6 (46-52)

Table 2 Results of Cement Analyses

                             Average       Average       Average
                            Total Area      Mantle      Mantle at
Case                        of Cement    Thickness at    Chamfer
Cement Era                     (%)       Walls, (mm)      (mm)

#1 Earliest       Average      50.8          3.58           4
                  Range       48-52        1.6-9.1       2.2-8.3
2nd Generation
#2 Right          Average      50.7          1.96         3.38
                  Range       50-51        0.9-4.3       2.7-4.3
    #2 Left       Average      37.7          1.96         4.87
                  Range       29-53        0.4-5.2       0.2-7.6
3rd Generation
#3                Average       33           1.75         3.27
                  Range       26-46        0.7-3.1       2.2-5.4

#4 European fit   Average       26           1.29         3.02
                  Range       23-29        0.5-2.0       2.8-3.3

BHR               Average      44.3          0.16         0.72
                  Range       33-58        0.0-0.73       0-2.1

                                Average       Average       Average
                   Average    Penetration   Penetration   Penetration
Case              Mantle at    at Walls     at Chamfer      at Dome
Cement Era        Dome (mm)      (mm)          (mm)          (mm)

#1 Earliest         3.88         0.67          2.44          2.12
                   1.0-8.0      0.1-1.9       0.1-6.2       0.1-4.3
2nd Generation
#2 Right             2.4         1.97           5.5           8.8
                    0.7-5       0.9-3.1      2.3-11.9      6.3-11.7
    #2 Left         3.62         1.29          2.17           1.9
                   0.3-7.3       0-3.7         0-5.4         0-5.0
3rd Generation
#3                  1.28         2.49           3.9           4.2
                   0.6-2.1       0-9.6        0-10.6       0.5-15.3

#4 European fit     1.07         0.63          2.17          5.88
                   0.6-1.8       0-1.5        0.8-4.0      1.3-13.4

BHR                 1.13          4.1            8            8.9
                   0.1-2.5     0.21-8.1      5.5-10.0      3.6-11.5
Gale Copyright: Copyright 2009 Gale, Cengage Learning. All rights reserved.