Cementless femoral components were introduced in an effort to avert the high rates of loosening and osteolysis reported with cemented femoral components using early cementing techniques . However, cementless femoral components did not eliminate either of these complications and improvements in cementing techniques have resulted in significantly longer survivorship of cemented implants [11-13]. As a result, the decade of the 1980s was considered by many hip surgeons to be the decade of cement versus cementless THA . Since then, the implant designs and fixation strategies have varied widely between countries. Recent reports consistently show lower revision rates for cemented stems [5, 6, 15, 17]. Despite these well-documented and publicized results, cementless fixation use for the oldest age group has continued to increase in many countries [9, 24].
The recent Australian Orthopaedic Association National Joint Replacement Registry (AOANJRR) annual report indicated that in patients 75 years or older, cementless stem fixation had a higher rate of revision compared with cemented stem fixation and that this difference varied with time . However, these data represented pooling of 167 cementless and 89 cemented stems. Thus, it is possible that the poorer results seen with cementless stems may simply represent the number, volume, and performance of the individual stems used rather than the mode of fixation . These differences in stem performance cast doubt on the reason why the AOANJRR and other registry results show that patients older than 75 years with cemented stems have a lower risk of revision [5, 6, 15, 17, 24].
We therefore explored the AOANJRR data to discern which stems provided the best long-term results, defined as the lowest cumulative percent revision at 10 years. We aimed to answer the following questions: (1) Is the risk of revision higher in patients older than 75 years of age who receive one of the three cementless stems with the highest overall survivorship in the registry than in those of that age who received one of the three best-performing cemented stems? If so, is there a difference in risk of early revision versus late revision, defined as revision within 1 month after index surgery? (2) Are there any diagnoses (such as osteoarthritis [OA] or femoral neck hip fracture) in which the three best-performing cementless stems had better survivorship than one of the three best-performing cementless stems? (3) Do these findings change when evaluated by patient sex?
Patients and Methods
At our request, the AOANJRR approved and performed an analysis comparing the best three cemented femoral stems with the best three cementless femoral stems in patients 75 years or older undergoing primary THA. For purposes of this study, we defined the “best three” as the three cemented and cementless femoral stems in which each was used in > 1000 procedures and with the lowest cumulative percent revision (CPR) at 10 years, regardless of primary diagnosis. All procedures were performed between September 1, 1999, and December 31, 2015. Only THAs with crosslinked polyethylene were included in this study to ensure uniformity of bearing surface between groups and essentially excluding osteolysis as a reason for revision. Of the 214,800 primary THAs with polyethylene in the registry during this time period, 174,409 had crosslinked polyethylene and their CPR rate at 15 years is 5.6% .
The registry includes > 98% of joint replacement procedures undertaken in Australia . On initial submission of forms from participating hospitals, the registry’s capture rate is 95.9%. After verification against health department data, checking of unmatched data, and subsequent retrieval of unreported procedures, the registry is able to obtain an almost complete data set relating to hip replacement in Australia.
The AOANJRR identified 89 different cemented femoral stems used during the study period. The three stems in common use with the lowest risk of revision at 10 years were the polished, tapered MS-30® stem (Zimmer-Biomet, Warsaw, IN, USA), the composite I beam Omnifit® stem (Stryker, Mahwah, NJ, USA), and the polished, tapered Exeter V40TM (Stryker) stem. The cumulative revision rate of these three femoral stems ranged from 4% to 5% at 10 years. These cemented stems were used in 31,635 cases and represent 56% of all the surgeries with a cemented stem during this time. There were 167 cementless femoral stems used and the three best were the proximally porous-coated, double wedge SynergyTM stem (Smith & Nephew, London, UK), the proximally hydroxyapatite-coated, tapered round Secur-FitTM stem (Stryker), and the tapered rectangular SL-PlusTM (Smith & Nephew). The cumulative revision rate of these three stems ranged from 4.5% to 6.1% at 10 years. They were used in 5023 cases (14%). The primary diagnosis was OA in 31,150 cases with 26,627 (86%) being in the cemented group and 4523 (14%) in the cementless group. In patients whose primary diagnosis was OA, 65% were female in the best three cemented femoral stem group and 54% were female in the best three cementless femoral stem group. THAs were performed for a femoral neck fracture in 3337 cases with the majority being cemented (92%). One of the three best cementless stems was used in 74 of the 819 THAs in men and 206 of the 2518 THAs in females with a femoral neck fracture.
The CPR of the combined results of the three best-performing prostheses for each fixation type was calculated using the Kaplan-Meier estimates of survivorship. Isolated acetabular revisions were excluded from the analysis. The 95% confidence interval (CI) was calculated using pointwise Greenwood estimates. The CPR analysis was determined for all diagnoses, for primary OA, and for femoral neck fracture. Hazard ratios (HRs) from Cox proportional hazard models, adjusted for age and sex, were used to compare the combined rate of revision of the best three femoral stems in the corresponding group. HRs were determined for the first month postoperatively and over the entire followup period. All statistical tests were two-tailed at the 5% level of significance. A secondary analysis was done to evaluate the CPR of primary total conventional hip replacement in patients 75 years or older by femoral stem fixation and sex.
In aggregate, among patients older than 75 years of age, the CPR in the first 3 months postoperatively was lower among those treated with one of the three best-performing cemented stems than those treated with one of the three best-performing cementless stems (HR for best three cementless versus best three cemented = 3.47 [95% CI, 1.60-7.53], p = 0.001; Fig. 1). Early revision was 9.14 times more common in the best three cementless stems than in the best three cemented stems (95% CI, 5.54-15.06, p = 0.001). In the entire cohort (Fig. 2), revision surgery for fracture and loosening in the best three cementless stems was at least double that for the best three cemented stems (Table 1). There was no difference in the CPR between the two stem fixation groups at any time after 3 months up to the final followup at 13 years (cemented CPR = 2.2%, cementless CPR = 3.7%, HR = 1.08 [95% CI, 0.80-1.46], p = 0.615).
Likewise, among patients with OA (Fig. 3) and femoral neck fracture (Fig. 4) who were older than 75 years of age, the CPR was consistently higher at 1 month postoperatively among those treated with one of the three best-performing cementless stems than those treated with one of the three best-performing cementless stems (OA: HR for best three cementless versus best three cemented = 8.82 [95% CI, 5.08-15.31], p < 0.001; hip fracture: HR for best three cementless versus best three cemented = 27.78 [95% CI, 1.39-143.3], p < 0.001). This difference was maintained for the first 3 months in the patients with OA (HR for best three cementless versus best three cemented = 3.31 [95% CI, 1.39-7.90], p < 0.006), but not in patients undergoing THA for a fracture of the femoral neck. The higher early CPR rate with the cementless stems in OA (Fig. 5) was the result of the higher revisions for fracture and loosening (Table 2), whereas fracture, loosening, and dislocation resulted in a higher CPR in femoral neck fractures treated with THA (Table 3). There was no difference in the CPR between the two stem fixation groups used to treat OA or a femoral neck fracture at final followup (OA at 13 years: cemented CPR = 2.0%, cementless CPR = 3.7%, HR = 1.13 [95% CI, 0.82-1.57], p = 0.450; hip fracture at 8 years: cemented CPR = 2.8%, cementless CPR = 4.8%, HR = 1.35 [95% CI, 0.57-3.19], p = 0.498).
Overall, the CPR was lower in the three best cemented stem group than the three best cementless stem group for both males and females at 1 month postoperatively (male: HR = 0.42 [95% CI, 0.20-0.92], p = 0.030; female: HR = 0.06 [95% CI, 0.03-0.10], p < 0.001) and for females at 3 months postoperatively (HR = 0.15 [95% CI, 0.06-0.33], p < 0.001) after which there was no difference in the CPR (Fig. 6). For a THA with one of the best three cementless stems, males had a lower CPR than females at all time periods (HR at 13 years = 1.78 [95% CI, 1.10-2.88], p = 0.019; male CPR = 2.8, female CPR = 4.3). However, the CPR for a THA with one of the best three cemented stems was lower in females than in males at all time periods (HR at 13 years = 2.09 [95% CI, 1.66-2.62], p < 0.001; male CPR = 2.9, female CPR = 1.8). Comparable findings were noted in those patients who underwent THA for the diagnosis of OA (Fig. 7).
Registry data have consistently found lower revision rates for cemented stems than cementless stems in patients older than 75 years of age undergoing primary THA . However, it is clear that not all implants are of equal reliability. As a result, the registry data reflect a combination of both well- and poor-performing implants, in which the volume and number of poor-performing implants can significantly skew the outcomes. In this registry study of patients older than 75 years of age undergoing primary THA, we compared the three best-performing cemented and cementless femoral stems in the AOANJRR. This approach allowed us to eliminate the potential confounding effect of a large number of stems with poor performance and focus primarily on the effect of the mode of femoral stem fixation on revision rates. Using this methodology, we found that overall, the CPR was lower among those treated with one of the three best-performing cemented stems than those treated with one of the three best-performing cementless stems for only the first 3 months postoperatively, after which there was no difference in the CPR. In patients with a diagnosis of a femoral neck fracture, the lower CPR was seen in the first month and in OA, it was seen in the first 3 months, independent of the patient’s sex.
This study has several limitations. First, the endpoint is revision and therefore does not include other postoperative problems including radiographically loose prostheses that have not been revised. This may have its greatest effect in the cementless group for femoral neck fracture, where frail, elderly, and sick patients may not be medically well enough to have revision surgery for a loose cementless femoral stem. It is not possible to estimate the effect that this may have had on the overall data. Second, the data are not controlled for patient, surgeon, and hospital factors that are known to affect revision rates after THA [11, 21, 25]. However, registry data are reflective of the general health care of the country because it includes population-level data from a large number of patients, from surgeons with all levels of experience, and with practically no exclusions. Third, the AOANJRR reports loosening and fracture rates for THA, which includes the femoral and the acetabular components. However, in this study, we were able to exclude all revisions related solely to the acetabulum. Ideally, the cementless and cemented femoral stems would be matched with the same cementless cup. This was not done in this study so that the number of patients would not be severely limited. By eliminating all revisions related directly to the acetabular component, this limitation has been minimized. Finally, these registry data provide a national overview of the risk of revision by mode of fixation in patients 75 years or older, but it may not be a valid measure of revision rates that can be obtained in expert single centers.
Recently, Troelsen et al.  analyzed the annual reports of hip arthroplasty registries and have questioned the paradox of the increasing use of cementless femoral components in older patients in light of their inferior registry results compared with cemented components. In all four of the registries that provided age-stratified risk estimates of revision when comparing THA with different fixation techniques (Australia, Denmark, England-Wales, and New Zealand), cemented fixation resulted in statistically significantly lower revision rates than uncemented fixation in the oldest age groups (> 75 years; except England-Wales, > 65 years). However, raw data from the registries cannot be assessed, thereby preventing supplemental analysis, to clarify any confounders that may have led to their findings. In particular, the registries do not follow specific implant designs to observe their performance individually. By comparing the three best-performing cementless stems with the three best-performing cemented stems, we eliminated the potential influence of implant design. In this study, cemented stems had a lower risk of revision than cementless stems only in the first 3 months postoperatively and then demonstrated no difference in the CPR up to 13 years after the primary THA. These early revisions were mainly attributable to the risk of having revision for fracture or loosening being at least double that in the cementless group compared with the cemented group. Therefore, by selecting cementless and cemented implants with similar 10-year survivorship, this AOANJRR study contradicts the previous registry analysis by Troelsen and demonstrates that after an early increased failure rate of cementless implants, the long-term cumulative revision rate of cementless stems is no different than that of cemented stems in patients older than 75 years of age.
In femoral neck fractures, the CPR was consistently higher at 1 month postoperatively among those treated with one of the three best-performing cementless stems than those treated with one of the three best-performing cementless stems. The early failures in the cementless group were mainly a result of the increase in the revision rate for fracture, loosening, and dislocation compared with the cemented group. This is not unexpected because hip fractures are osteoporotic fractures and older age is a surrogate for proximal femoral bone quality and morphologic features. Osteoporotic bones are very brittle and much more susceptible to intraoperative fracture than normal bone, although biomechanical studies have shown a protective effect of cement in terms of load to failure in osteoporotic bones [16, 23]. Proximally coated cementless stems place increased stresses on the proximal femur and increase the risk of fracture. Therefore, it is not surprising that in the older patients investigated in our study, the fracture rate resulting in early revision in the cementless group was at least double that seen in the cemented group for femoral neck fracture. However, the absence of a difference in the two groups in this study undergoing THA for all diagnoses after 3 months suggests that there can be a role for cementless implants in selected cases, depending on the surgeon’s expertise and the quality and shape of the proximal femoral bone. Nonetheless, in patients with severe proximal femoral osteoporosis, the surgeon who is unfamiliar with the nuances of a particular cementless stem probably is better off using cemented fixation to achieve stem stability and to reduce the risk of fracture and loosening.
This study demonstrated that sex did influence the CPR of cementless stems. Specifically, males with a THA had a lower CPR than females at all time periods. This correlates with the anatomic changes that occur in each sex with age. The shape and bone mass of the osteoporotic proximal femur is different than that of the nonosteoporotic femur and these changes are age- and sex-specific . In males, there are substantial changes in the femoral neck, but there is no change in shaft cortical thickness or medullary width with age [19, 20]. In females, there is no change in the metaphyseal width, but there is thinning of the diaphyseal cortices and expansion of the canal resulting in a decrease in the canal flare index . A Dorr Type C stovepipe femur is consistent with severe osteoporosis, and this shape makes it more difficult to obtain cementless stem fixation. A radiostereometric analysis study that evaluated patients with severe osteoporosis undergoing cementless hip replacement found that the cementless stems had a higher subsidence of the stem during the first 3 months after surgery . Aro et al.  concluded that a low bone mineral density, changes in intraosseous dimensions of the proximal femur, and aging adversely affected initial stability and delayed osseointegration of cementless stems in females. Like their study, the registry data in our study indicate that during the early postoperative period, there is a high rate of loosening with cementless stems in patients 75 years or older. We found the revision rate for loosening in the cementless group was four times that of the cemented group during the first 3 months postoperatively in patients with all diagnoses and during the first month in patients with a femoral neck fracture, respectively.
Although the majority of femoral stems used in some parts of the world are cementless and their use throughout the world continues to increase [9, 24], it is apparent from this study that surgeons performing arthroplasties need to be able to cement a femoral stem to achieve the highest possible survivorship in patients 75 years or older [4, 18]. The results of cemented femoral stems are technique-dependent and a constellation of surgical techniques can help surgeons consistently provide a high-quality cement mantle around a femoral stem . We believe that as the use of cementless femoral implants has increased, it has resulted in decreased training of residents or registrars in proper cementing techniques. Based on this review of registry data and our knowledge of the morphologic features and biomechanical properties of older bone, it would seem important to ensure the proper training of contemporary cementing techniques for the next generation of arthroplasty surgeons so they are able to use this option when required. Cementless femoral stem fixation in older patients has a higher earlier rate of revision when only the best-performing prostheses in this age group are compared. Implant choice alone is not responsible for these findings and clearly indicates that the mode of fixation has a substantial effect on the rate of revision in this age group. However, the absence of a difference in the two groups undergoing THA after 3 months suggests that there can be a role for cementless implants in selected cases, depending on the surgeon’s expertise and the quality and shape of the proximal femoral bone.
1. Anand R, Graves SE, de Steiger RN, Davidson DC, Ryan P, Miller LN, Cashman K. What is the benefit of introducing new hip and knee prostheses? J Bone Joint Surg Am. 2011;93(Suppl 3):51–54.
2. Aro HT, Alm JJ, Moritz N, Makinen TJ, Lankinen P. Low BMD affects initial stability and delays stem osseointegration in cementless total hip arthroplasty in women: a 2-year RSA study of 39 patients. Acta Orthop. 2012;83:107–114.
3. Australian Orthopaedic Association National Joint Replacement Registry. Hip, Knee & Shoulder Arthroplasty Annual Report 2016. Available at: https://aoanjrr.sahmri.com/en/annual-reports-2016
. Accessed February 12, 2018.
4. Ballard TW, Callaghan JJ, Sullivan PM, Johnston RC. The results of improved cementing techniques for total hip arthroplasty in patients less than fifty years old. A ten-year follow-up study. J Bone Joint Surg Am. 1994;76:959–964.
5. Corbett KL, Losina E, Nti AA, Prokopetz JJ, Katz JN. Population-based rates of revision of primary total hip arthroplasty: a systematic review. PloS One. 2010;5:e13520.
6. Hailer NP, Garellick G, Karrholm J. Uncemented and cemented primary total hip arthroplasty in the Swedish Hip Arthroplasty Register. Acta Orthop. 2010;81:34–41.
7. Harris WH. Hybrid total hip replacement: rationale and intermediate clinical results. Review. Clin Orthop Relat Res. 1996;333:155–164.
8. Hartman CW, Gilbert BJ, Paprosky WG. Gender issues in total hip arthroplasty: length, offset, and osteoporosis. Semin Arthroplasty. 2009;20:62–65.
9. Huo MH, Dumont GD, Knight JR, Mont MA. What’s new in total hip arthroplasty. J Bone Joint Surg Am. 2011;93:1944–1950.
10. Jones LC, Hungerford DS. Cement disease. Clin Orthop Relat Res. 1987;225:192–206.
11. Losina E, Barrett J, Mahomed NN, Baron JA, Katz JN. Early failures of total hip replacement: effect of surgeon volume. Arthritis Rheum. 2004;50:1338–1343.
12. Malcolm AJ. Pathology of cemented low-friction arthroplasties in autopsy specimens. In: Older J, ed. Implant Bone Interface. New York, NY, USA: Springer; 1990:77–82.
13. Maloney WJ, Harris WH. Comparison of a hybrid with an uncemented total hip replacement. A retrospective matched-pair study. J Bone Joint Surg Am. 1990;72:1349–1352.
14. Maloney WJ, Jasty M, Burke DW, O'Connor DO, Zalenski EB, Bragdon C, Harris WH. Biomechanical and histologic investigation of cemented total hip arthroplasties. A study of autopsy-retrieved femurs after in vivo cycling. Clin Orthop Relat Res. 1989:249:129–140.
15. McMinn DJ, Snell KI, Daniel J, Treacy RB, Pynsent PB, Riley RD. Mortality and implant revision rates of hip arthroplasty in patients with osteoarthritis: registry based cohort study. BMJ.2012; 344:e3319.
16. Mears SC. Management of severe osteoporosis in primary total hip arthroplasty. Current Translational Geriatrics and Experimental Gerontology Reports. 2013;2:99–104.
17. Morshed S, Bozic KJ, Ries MD, Malchau H, Colford JM Jr. Comparison of cemented and uncemented fixation in total hip replacement: a meta-analysis. Acta Orthop. 2007;78:315–326.
18. Mulroy RD Jr, Harris WH. The effect of improved cementing techniques on component loosening in total hip replacement. An 11-year radiographic review. J Bone Joint Surg Br. 1990;72:757–760.
19. Noble PC, Box GG, Kamaric E, Fink MJ, Alexander JW, Tullos HS. The effect of aging on the shape of the proximal femur. Clin Orthop Relat Res. 1995;316:31–44.
20. Peacock M, Liu G, Carey M, Ambrosius W, Turner CH, Hui S, Johnston CC Jr. Bone mass and structure at the hip in men and women over the age of 60 years. Osteoporos Int. 1998;8:231–239.
21. Schurman DJ, Bloch DA, Segal MR, Tanner CM. Conventional cemented total hip arthroplasty. Assessment of clinical factors associated with revision for mechanical failure. Clin Orthop Relat Res. 1989;240:173–180.
22. Stas VA, Templeton JE, Paprosky WG. Implications of female sex/osteoporosis on femoral implant selection and restoration of biomechanics of the hip. In: Glassman AH, Lachiewicz PF, Tanzer M, eds. Orthopaedic Knowledge Update 4. Hip and Knee Reconstruction. 4th
ed. Rosemont, IL, USA: AAOS; 2011:219–232.
23. Thomsen MN, Jakubowitz E, Seeger JB, Lee C, Kretzer JP, Clarius M. Fracture load for periprosthetic femoral fractures in cemented versus uncemented hip stems: an experimental in vitro study. Orthopedics. 2008;31:653.
24. Troelsen A, Malchau E, Sillesen N, Malchau H. A review of current fixation use and registry outcomes in total hip arthroplasty: the uncemented paradox. Clin Orthop Relat Res. 2013;471:2052–2059.
25. Young NL, Cheah D, Waddell JP, Wright JG. Patient characteristics that affect the outcome of total hip arthroplasty: a review. Can J Surg. 1998;41:188–195.