AUGUST 2015 | Volume 38 • Number 8

n Feature Article

abstract

Thin-Walled Cross-Linked Acetabular Liners Need Not Exhibit Reduced Locking StrengthAndrew S. MurthA, Md; MArcel e. roy, Phd; leo A. whiteSide, Md; dAvid S. tilden, BS; KryStAl l. SchMitt, BS

Use of larger diameter femoral heads has emerged as a promising strategy to reduce the risk of dislocation after total hip arthroplasty, but thinning the walls of cross-linked ultra-high-molecular-weight polyethylene (UHMWPE) acetabular liners to accommodate these larger heads may compromise the locking mechanism of the liner. The purpose of this study was to test the mechanical integrity of the locking mechanism in cross-linked and re-melted UHMWPE acetabular components with reduced wall thickness. The locking mechanism of cross-linked (100 kGy/re-melted) acetabular liners in sizes 50/28, 50/36, and 52/36 mm of 1 design was evaluated by lever-out tests and torsion tests. Torsion tests were performed at 2 angles to isolate the liner’s locking tabs independent of the contribution of its central post. Lever-out testing demonstrated nominally reduced failure strength in 50/36-mm liners (13.3 N · m) compared with 50/28-mm liners (12.3 N · m; P=.0502), whereas the lever-out strength of 52/36-mm liners was 12.2±0.94 N · m. Failure torques were similar between 50/28- and 50/36-mm liners at 45° and 90°, but the failure torque of size 52/36-mm liners was significantly higher at each angle. The use of larger diameter femoral heads does not compromise the locking mechanism of thinned MicroSeal (Signal Medical Corp, Marysville, Michigan) acetabular liners. Use of a cross-linked UHMWPE acetabular liner, with a locking mechanism that is not compromised when the liner is thinned to a thickness of at least 2.86 mm, appears to be a bio-mechanically sound construct when articulated with large diameter femoral heads. [Orthopedics. 2015; 38(8):e727-e732.]

The authors are from San Antonio Military Medical Center (ASM), San Antonio, Texas; and the Missouri Bone & Joint Research Foundation (MER, LAW, DST) and Signal Medical Corp (LAW, KLS), St Louis, Missouri.

Dr Murtha, Mr Tilden, and Ms Schmitt have no relevant financial relationships to disclose. Dr Roy has served as an unpaid consultant for Signal Medical Corp. Dr Whiteside is a paid consultant for and holds stock in Signal Medical Corp and receives royalties from Smith & Nephew.

This study was funded by the Missouri Bone & Joint Research Foundation.Correspondence should be addressed to: Marcel E. Roy, PhD, Missouri Bone & Joint Research

Foundation, 1000 Des Peres Rd, Ste 150, St Louis, MO 63131 (mroy@mobojo.org).Received: May 20, 2014; Accepted: November 24, 2014.doi: 10.3928/01477447-20150804-62

e727

Copyright © SLACK inCorporAted

n Feature Article

In the past decade, the use of larger di-ameter femoral heads (≥32 mm) has been shown to reduce the incidence of

dislocation following total hip arthroplasty (THA).1-6 Initially, concerns were voiced about the increased volumetric wear of these larger components2,7-9; however, the intro-duction of highly cross-linked and re-melted ultra-high-molecular-weight polyethylene (UHMWPE) as a bearing surface in THA has markedly improved the wear resistance of acetabular liners, with excellent reported results when used in conjunction with larg-er femoral heads.10-15

Although cross-linking improves the wear resistance of UHMWPE, it has been shown to reduce its tensile strength and resistance to fatigue crack propaga-tion,10,16-18 and studies have indicated that subsequent thermal processing such as re-melting further compromises these properties.19,20 As a liner is thinned to ac-commodate the increased diameter of the larger femoral head, one of the elements most prone to failure is its locking mecha-nism, where stresses are concentrated.21-23 Rim fracture and subsequent failure at the locking mechanism, resulting in liner dissociation, have been documented in a variety of modular hip systems.24-27 Re-cent case reports of rim fractures in cross-linked UHMWPE liners suggest that this is also a potential problem with current designs that use thin polyethylene liners or designs where the locking mechanism is insufficiently supported.27,28 Evalua-tion of the fracture surface with scanning electron microscopy in these case reports revealed damage consistent with fatigue cracking.28 Given these concerns, it is rea-sonable to presume that thinning the poly-ethylene through the rim could increase the risk of failure through the peripheral locking mechanism.

The purpose of this study was to test the mechanical integrity of the locking mechanism in cross-linked and re-melted UHMWPE acetabular components of one THA design wherein the liner’s wall thick-ness was reduced to increase its inner di-

ameter and accommodate large-diameter femoral heads. The locking mechanism of the system tested comprises a plural-ity of pegs and ridges that protrude from the polyethylene surface to engage slots in the metal shell. It does not have grooves or recesses that penetrate the polyethyl-ene around the peripheral rim. Therefore, the authors hypothesized that the locking mechanism would not be affected by in-creasing the inner diameter of the liner and that the liner would maintain its lever-out strength and torsional load to failure.

Materials and MethodsStock 50/28-mm acetabular liners and

custom 50/36- and 52/36-mm liners of the same design (MicroSeal; Signal Medical Corp, Marysville, Michigan) were ma-chined from GUR 1050 medical-grade UHMWPE that was cross-linked to 100 kGy, re-melted, and then donated for test-ing in this study. The minimum nominal rim thickness varied from 2.86 mm (size 50/36 mm) to 7.04 mm (size 50/28 mm) (Table 1). All liners had stock backside surfaces and locking mechanisms, which consisted of 12 locking tabs around the rim and a central post at the dome (Figure 1A). Eight liners of each size designated for lever-out testing had an additional hole drilled in the inside surface, allowing a bar to serve as a lever to lift it out of the shell (Figure 1B). Six to 9 liners of each size used for torsion tests had radial and con-centric grooves created on the interior sur-face (Figure 1C) to facilitate a mechanical

bond with the acrylic cement used to affix femoral head fixtures within the liner.

Lever-out tests followed a previous protocol29 and used a hydraulic testing platform (Instron 8501; Instron, Nor-wood, Massachusetts) to displace the free end of a steel bar that rested on a hard ful-crum with the other end mounted inside a hole in the test liner. The actuator was lowered at a speed of 2.9 mm/s, and a pair of pillow bearings allowed the free end of the lever to trace an arc at approximately 1.3 rad/min. Failure was defined as disso-ciation of the liner from its shell.

Torsion tests were performed in rota-tion control (1°/s ramp, up to 20°) on the same hydraulic testing platform. Liners were driven into an acetabular shell fix-

Table 1

Summary of Liner Dimensionsa

Nominal Thickness, mm

Liner Size, OD/ID, mm Dome

Rim (Min)

Rim (Max)

50/28 6.93 7.04 8.2

50/36 4.11 2.86 4.2

52/36 5.94 4.14 5.21

Abbreviations: OD/ID, outer diameter/inner diameter; Max, maximum; Min, minimum. aDome thickness does not include the central post, whereas the difference between minimum and maximum rim thickness represents the thickness of the liner’s locking tab.

Figure 1: Photographs showing that all acetabular liners used in this study featured a stock backside and locking mechanism, including 12 locking tabs and a central post (A), whereas liners for lever-out testing had an additional hole in the inside surface (B) and liners used for torsion tests had radial and concentric grooves on the interior surface (C).

e728

AUGUST 2015 | Volume 38 • Number 8

n Feature Article

ture, and then femoral head fixtures were cemented into the liners at 90° (perpen-dicular) using Fastray acrylic dental ce-ment (Harry J. Bosworth Co, Skokie, Illinois). Although this setup does not rep-resent physiological loading conditions, it measures the relative strength of the 12 locking tabs independent of the liner’s central post. After the acrylic cement was dry, femoral head fixtures were connected to the actuator of the hydraulic testing platform using a steel, 2-piece, clamp-on coupling and the complete specimen was lowered into a “yoke” mounted into com-bined load/torsion cell, which mated with a fin on the backside of the acetabular shell fixture and thus resisted torque when the actuator was rotated (Figure 2). These specimens were tested with a compressive load of 334 N to prevent dissociation, with failure defined as yielding or complete avulsion of the locking tabs.

A second round of torsion tests had a similar setup, except that the yoke was fixed to a sine plate at 45° and the femo-ral head was likewise cemented at 45° to simulate a more physiological neck-shaft angle of 135°, which is within the range of reported anatomical findings.30-32 In these tests, the acetabular shell fixture was low-ered to within 0.5 mm of the yoke, with no compressive load applied, which allowed the more clinically relevant failure mode of dissociation of the liner from the acetabular shell fixture (Figure 3).

For each test, data from the 50/28- and 52/36-mm groups were compared with data from the 50/36-mm group using a heteroscedastic 2-tailed t test (Excel; Mi-crosoft, Redmond, Washington) with P<.05 for significance. Post hoc statisti-cal power was evaluated by Power on X (MMISoftware, Newcastle upon Tyne, United Kingdom), with power greater than 0.8 deemed sufficient to avoid making type II errors.

resultsThe lever-out data revealed similar dis-

sociation torques among acetabular liner

sizes, although the failure strength of the 50/28-mm liners (13.3±1.2 N · m) was slightly, but not statistically, greater than that of the 50/36-mm liners (12.3±0.34 N · m; P=.0502). The lever-out strength of 52/36-mm liners (12.2±0.94 N m) was not different from that of 50/36-mm liners (Ta-ble 2). Damage to the acetabular liners fol-lowing lever-out tests was observed mainly at the liner’s central post (Figure 4).

All acetabular liners mounted at 90° exhibited severe damage to their locking tabs when loaded by torsion with yielding

Figure 2: Photograph showing torsion setup at 90°. The femoral head fixture was clamped to the actuator of the hydraulic testing platform and fixed in the acetabular liner using acrylic cement. The shell fixture was lowered into a “yoke” mounted on the load/torsion cell.

Figure 3: Photograph showing torsion setup at 45°. The femoral head was fixed in the acetabular liner at a 45° angle. The “yoke” was also mounted at 45° to a sine plate producing a neck-shaft angle of 135°.

Table 2

Summary of Lever-out Dataa and Comparison With Corresponding Data From 50/36-mm Liners

Liner Size, OD/ID, mmFailure Torque, Average±SD,

N · m Different From 50/36 mm?

50/28 13.3±1.2 No; P=.0502, power=0.56

50/36 12.3±0.34 N/A

52/36 12.2±0.94 No; P=.73, power=0.05

Abbreviations: N/A, not applicable; OD/ID, outer diameter/inner diameter. aFor each size, n=8.

Figure 4: Photograph showing acetabular liner with minor damage to the central post (highlight-ed) following lever-out testing.

e729

Copyright © SLACK inCorporAted

n Feature Article

or complete avulsion of the tabs (Figure 5A). All liners tested at 45° exhibited moderate damage to their tabs and to the central post before they dissociated from the acetabular shell (Figure 5B). The tor-sion data (Figure 6, Table 3) revealed that load to failure decreased by about one

third when the liners were angled at 45° and allowed to dissociate from their ac-etabular shell. Size 50/28- and 50/36-mm liners exhibited similar failure torques at each mounting angle, whereas 52/36-mm liners exhibited significantly higher fail-ure torques at each angle tested.

discussionThis study found that the overall

strength of this type of acetabular liner locking mechanism did not significantly change when the inner diameter was in-creased from 28 to 36 mm. This 59% re-duction in liner thickness from 50/28 to 50/36 mm corresponded with an 8% reduc-tion in lever-out strength. This reduction is not likely to be significant in a clinical context given its small magnitude.29 In-creasing the outer diameter to 52 mm did not increase lever-out torque. Torsion tests likewise revealed no reduction in locking mechanism failure strength when com-paring the 50/28- and 50/36-mm liners, regardless of the angle of the test. The sig-nificant increase in failure torque observed from 52/28-mm liners is as expected due to the larger moment arm provided by the increased outer diameter of the liner.

The current study has several limita-tions. First, as a biomechanical study, several aspects of the physiological en-vironment were not considered to avoid introducing confounding variables; lin-ers were tested in air (instead of inside a bovine serum bath) and at room tempera-ture. However, given the consideration for fluid uptake in load-soak control specimens in hip simulator studies,33 it could be reasonable to presume that such conditions may in fact ensure a tighter fit between the liner and shell, thus enhanc-ing the performance of the liner’s locking mechanism. In addition, this study only

Figure 5: Photographs showing acetabular liners mounted and tested at 90º exhibited severe damage to their locking tabs after testing (A), whereas liners tested at 45º exhibited moderate damage to their tabs and central post (highlighted) (B).

Figure 6: Line graph showing sample torsion- rotation curves for 50/36-mm liners tested at 90° (dark blue plot; locking tabs completely sheared off at 68.1 N · m) and 45° (light gray plot; liner dissoci-ated from shell at 44.7 N · m).

Table 3

Summary of Torsion Data for Liners Mounted at 90º and 45º, and Comparison With Corresponding Data From 50/36-mm

LinersLiners Mounted at 90º Liners Mounted at 45º

Liner Size, OD/ID, mm No.

Failure Torque, Average ±SD,

N · m

Different From 50/36

mm? No.

Failure Torque, Average±SD,

N · m

Different From 50/36

mm?

50/28 9 68.6±4.1 No; P=.11, power=0.35

8 44.0±2.4 No; P=.76, power=0.05

50/36 9 65.4±4.0 N/A 7 44.3±2.2 N/A

52/36 6 80.4±7.4 Yes, P=.0027

6 52.0±4.3 Yes, P=.0053

Abbreviations: N/A, not applicable; OD/ID, outer diameter/inner diameter.

e730

AUGUST 2015 | Volume 38 • Number 8

n Feature Article

considered one design of acetabular liner (MicroSeal), whereas other liners and locking mechanisms may behave differ-ently as their walls are thinned. Finally, it should be noted that the etiology of liner dissociation is a multifactorial issue in-cluding causes such as poor intraopera-tive component positioning and oxidative degeneration, which were beyond the scope of this study.24,34

The lever-out data in the current study were within the range of previously re-ported values for 52/32-mm conventional, non-cross-linked, UHMWPE modular systems.29 Among the acetabular liners ex-amined by Tradonsky et al,29 the lever-out strength of MicroSeal’s locking mecha-nism was most comparable to that of the HGP II (Zimmer, Inc, Warsaw, Indiana). Using a similar but not identical protocol, Hoffmann et al35 used 49.2/32-mm Du-rasul liners (Zimmer, Inc) as a control in a study investigating the cementation of the UHMWPE liners into acetabular shells. However, it is difficult to compare our data as theirs expressed in N, rather than in N · m without providing a lever arm length for conversion.

The data from the current custom torque fixtures clearly illustrated that the overall strength of the acetabular liner’s locking mechanism was not significantly reduced when the inner diameter was expanded to accept 36-mm heads. When mounted at 90°, each of the locking tabs resisted approximately the same amount of torque and were viscoelastically de-formed until failure. When mounted at 45°, the central post could also resist torque, but the tabs were not loaded equal-ly and the liner could not resist as much torque before dissociation of the liner from its shell. In the current study, the 90° torsion findings for the 50/28- and 50/36-mm liners were comparable with previ-ously reported results from the 49.2/32-mm Durasul inserts,35 and the 52/36-mm liners performed considerably better still. The torsion data support previously re-ported finite element studies, which sug-

gest that increased contact stresses should not limit the use of large-diameter femo-ral heads with thinner liners.15,36 Finally, early results from 2 studies using a differ-ent design and thin acetabular liners made of sequentially cross-linked/annealed UHMWPE have been promising. Say-eed et al37 reported no clinical failures of 3.8-mm thick liners after a minimum 2-year follow-up, and Johnson et al38 found no evidence of liner failure due to fracture and cracking after 2.4 million cy-cles in a hip wear simulator, even in liners that were just 1.9-mm thick. Because the current study used liners with a minimum thickness of 2.86 mm (for size 50/36 mm), the relative durability of 1.9-mm thick lin-ers is encouraging.

conclusionUse of a cross-linked UHMWPE ac-

etabular liner, with a locking mechanism that is not compromised when the liner is thinned to a thickness of at least 2.86 mm, appears to be a biomechanically sound construct when articulated with large di-ameter femoral heads.

references 1. Amstutz HC, Le Duff MJ, Beaulé PE. Preven-

tion and treatment of dislocation after total hip replacement using large diameter balls. Clin Orthop Relat Res. 2004; (429):108-116.

2. Beaulé PE, Schmalzried TP, Udomkiat P, Amstutz HC. Jumbo femoral head for the treatment of recurrent dislocation following total hip replacement. J Bone Joint Surg Am. 2002; 84(2):256-263.

3. Berry DJ, von Knoch M, Schleck CD, Harm-sen WS. Effect of femoral head diameter and operative approach on risk of dislocation after primary total hip arthroplasty. J Bone Joint Surg Am. 2005; 87(11):2456-2463.

4. Bistolfi A, Crova M, Rosso F, Titolo P, Ven-tura S, Massazza G. Dislocation rate after hip arthroplasty within the first postoperative year: 36 mm versus 28 mm femoral heads. Hip Int. 2011; 21(5):559-564.

5. Jameson SS, Lees D, James P, et al. Lower rates of dislocation with increased femoral head size after primary total hip replacement: a five-year analysis of NHS patients in Eng-land. J Bone Joint Surg Br. 2011; 93(7):876-880.

6. Muratoglu OK, Bragdon CR, O’Connor D,

et al. Larger diameter femoral heads used in conjunction with a highly cross-linked ultra-high molecular weight polyethylene: a new concept. J Arthroplasty. 2001; 16(8)(suppl 1):24-30.

7. Atkins GJ, Haynes DR, Howie DW, Findlay DM. Role of polyethylene particles in peri-prosthetic osteolysis: a review. World J Or-thop. 2011; 2(10):93-101.

8. Kosashvili Y, Backstein D, Safir O, Lakstein D, Gross AE. Dislocation and infection af-ter revision total hip arthroplasty: compari-son between the first and multiply revised total hip arthroplasty. J Arthroplasty. 2011; 26(8):1170-1175.

9. Livermore J, Ilstrup D, Morrey B. Effect of femoral head size on wear of the polyethyl-ene acetabular component. J Bone Joint Surg Am. 1990; 72(4):518-528.

10. Bragdon CR, Kwon YM, Geller JA, et al. Minimum 6-year followup of highly cross-linked polyethylene in THA. Clin Orthop Relat Res. 2007; 465:122-127.

11. Digas G, Kärrholm J, Thanner J, Malchau H, Herberts P. The Otto Aufranc Award. Highly cross-linked polyethylene in total hip arthro-plasty: randomized evaluation of penetration rate in cemented and uncemented sockets us-ing radiostereometric analysis. Clin Orthop Relat Res. 2004; (429):6-16.

12. Geerdink CH, Grimm B, Vencken W, Hey-ligers IC, Tonino AJ. Cross-linked compared with historical polyethylene in THA: an 8-year clinical study. Clin Orthop Relat Res. 2009; 467(4):979-984.

13. Kurtz SM, Gawel HA, Patel JD. History and systematic review of wear and osteolysis out-comes for first-generation highly crosslinked polyethylene. Clin Orthop Relat Res. 2011; 469(8):2262-2277.

14. McCalden RW, MacDonald SJ, Rorabeck CH, Bourne RB, Chess DG, Charron KD. Wear rate of highly cross-linked polyethyl-ene in total hip arthroplasty: a randomized controlled trial. J Bone Joint Surg Am. 2009; 91(4):773-782.

15. Shen FW, Lu Z, McKellop HA. Wear versus thickness and other features of 5-Mrad cross-linked UHMWPE acetabular liners. Clin Or-thop Relat Res. 2011; 469(2):395-404.

16. Gencur SJ, Rimnac CM, Kurtz SM. Failure micromechanisms during uniaxial tensile fracture of conventional and highly cross-linked ultra-high molecular weight polyeth-ylenes used in total joint replacements. Bio-materials. 2003; 24(22):3947-3954.

17. Lewis G. Properties of crosslinked ultra-high-molecular-weight polyethylene. Bioma-terials. 2001; 22(4):371-401.

18. Pruitt LA. Deformation, yielding, fracture and fatigue behavior of conventional and highly cross-linked ultra high molecular weight polyethylene. Biomaterials. 2005; 26(8):905-915.

e731

Copyright © SLACK inCorporAted

n Feature Article

19. Gencur SJ, Rimnac CM, Kurtz SM. Fatigue crack propagation resistance of virgin and highly crosslinked, thermally treated ultra-high molecular weight polyethylene. Bioma-terials. 2006; 27(8):1550-1557.

20. Medel FJ, Peña P, Cegoñino J, Gómez- Barrena E, Puértolas JA. Comparative fatigue behavior and toughness of remelted and an-nealed highly crosslinked polyethylenes. J Biomed Mater Res B Appl Biomater. 2007; 83(2):380-390.

21. Furmanski J, Anderson M, Bal S, et al. Clini-cal fracture of cross-linked UHMWPE acetab-ular liners. Biomaterials. 2009; 30(29):5572-5582.

22. Furmanski J, Kraay MJ, Rimnac CM. Crack initiation in retrieved cross-linked highly cross-linked ultrahigh-molecular-weight poly-ethylene acetabular liners: an investigation of 9 cases. J Arthroplasty. 2011; 26(5):796-801.

23. Patel PD, Potts A, Froimson MI. The dis-locating hip arthroplasty: prevention and treatment. J Arthroplasty. 2007; 22(4)(suppl 1):86-90.

24. French K, Moore R, Gawel H, et al. Retrieval analysis of Harris-Galante I and II acetabular liners in situ for more than 10 years. Acta Or-thop. 2012; 83(4):366-373.

25. Mesko JW. Acute liner disassociation of a Pinnacle acetabular component. J Arthro-plasty. 2009; 24(5):815-818.

26. Tower SS, Currier JH, Currier BH, Lyford KA, Van Citters DW, Mayor MB. Rim crack-ing of the cross-linked longevity polyethylene acetabular liner after total hip arthroplasty. J Bone Joint Surg Am. 2007; 89(10):2212-2217.

27. Waewsawangwong W, Goodman SB. Unex-pected failure of highly cross-linked polyeth-ylene acetabular liner. J Arthroplasty. 2012; 27(2):e321-e324.

28. Blumenfeld TJ, McKellop HA, Schmalzried TP, Billi F. Fracture of a cross-linked poly-ethylene liner: a multifactorial issue. J Ar-throplasty. 2011; 26(4):e665-e668.

29. Tradonsky S, Postak PD, Froimson AI, Greenwald AS. A comparison of the disas-sociation strength of modular acetabular components. Clin Orthop Relat Res. 1993; (296):154-160.

30. Noble PC, Alexander JW, Lindahl LJ, Yew DT, Granberry WM, Tullos HS. The anatom-ic basis of femoral component design. Clin Orthop Relat Res. 1988; (235):148-165.

31. Noble PC, Box GG, Kamaric E, Fink MJ, Alexander JW, Tullos HS. The effect of ag-ing on the shape of the proximal femur. Clin Orthop Relat Res. 1995; (316):31-44.

32. Sugano N, Noble PC, Kamaric E. Predicting the position of the femoral head center. J Ar-throplasty. 1999; 14(1):102-107.

33. American Society for Testing and Materials.

ASTM F1714-96(2008), Standard Guide for Gravimetric Wear Assessment of Prosthetic Hip Designs in Simulator Devices. West Con-shohocken, PA: ASTM International; 2008.

34. Gray CF, Moore RE, Lee GC. Spontaneous dissociation of offset, face-changing poly-ethylene liners from the acetabular shell: a report of four cases. J Bone Joint Surg Am. 2012; 94(9):841-845.

35. Hofmann AA, Prince EJ, Drake FT, Hunt KJ. Cementation of a polyethylene liner into a metal acetabular shell: a biomechanical study. J Arthroplasty. 2009; 24(5):775-782.

36. Plank GR, Estok DM II, Muratoglu OK, O’Connor DO, Burroughs BR, Harris WH. Contact stress assessment of conventional and highly crosslinked ultra high molecular weight polyethylene acetabular liners with finite element analysis and pressure sensitive film. J Biomed Mater Res B Appl Biomater. 2007; 80(1):1-10.

37. Sayeed SA, Mont MA, Costa CR, et al. Early outcomes of sequentially cross-linked thin polyethylene liners with large diameter fem-oral heads in total hip arthroplasty. Bull NYU Hosp Jt Dis. 2011; 69(suppl 1):S90-S94.

38. Johnson AJ, Loving L, Herrera L, Delanois RE, Wang A, Mont MA. Short-term wear evaluation of thin acetabular liners on 36-mm femoral heads. Clin Orthop Relat Res. 2014; 472(2):624-629.

e732