- Email: [email protected]
Periprosthetic femoral bone loss after total knee arthroplasty: 1-year follow-up study of 69 patients
The Knee 11 (2004) 297–302
Periprosthetic femoral bone loss after total knee arthroplasty: 1-year follow-up study of 69 patients Tarja A. Soininvaaraa,b,*, Hannu J.A. Miettinenb, Jukka S. Jurvelinc, Olavi T. Suomalainenb, ¨ b,d Esko M. Alhavab, Heikki P.J. Kroger a
Department of Surgery, Savonlinna Central Hospital, Savonlinna, Finland b Department of Surgery, Kuopio University Hospital, Kuopio, Finland c Department of Clinical Physiology and Nuclear Medicine, Kuopio University Hospital and Department of Applied Physics, University of Kuopio, Kuopio, Finland d Bone and Cartilage Research Unit (BCRU), University of Kuopio, Kuopio, Finland Received 1 September 2003; accepted 30 September 2003
Abstract The clinical survival of joint arthroplasties is related to the quality of the surrounding bone environment. Bone mineral density (BMD) is an important measure of bone strength and quality. The aim of this prospective study was to measure the quantitative changes in BMD in the distal femur after cemented total knee arthroplasty (TKA) in osteoarthrotic knee joints. Sixty-nine patients with TKA were scanned postoperatively using dual-energy X-ray absorptiometry (DXA) within a week of surgery, and at 3-, 6-, and 12-month follow-ups. An average decrease in bone density of 17.1% (mean range of 12.1–22.8%) was measured adjacent to the prosthesis at the 12-month follow-up (repeated measures ANOVA P-0.0005). Bone loss was most rapid during the first 3 months after TKA. The clinical status and function parameters of the knee joint, evaluated by the American Knee Society (AKS) score, had improved significantly on the preoperative values at the three- and 12-month follow-ups (P-0.0005). However, improvement in the AKS score was not associated with periprosthetic BMD change (Ps0.204), whereas age (Ps0.067) and body mass index (Ps0.019) correlated with BMD loss for the total metaphyseal region of interest (ROI), by repeated measures ANOVA. We suggest that the observed periprosthetic bone loss was mainly the result of prosthesis-related stress-shielding. 䊚 2003 Elsevier B.V. All rights reserved. Keywords: Bone loss; BMD; Total knee arthroplasty; Follow-up
1. Introduction Total knee arthroplasty (TKA) alters the mechanical loading of the knee joint and causes adaptive bone remodelling. The bone surrounding the TKA adjusts its mineral density and structure to meet the new mechanical demands. The ‘cup-like’ shape of the femoral component results in the protection from stress of the distal femoral bone and protection from shear loading at the attachment of the anterior and posterior flange surfaces w1x. Moreover, the absence of compressive loading of the patella on the surface of the distal femur may result in stress-shielding of the patellofemoral region w1–10x. Van Lenthe et al. w4x introduced a theory *Corresponding author. Present address: Puijonsarventie 11 D 30, 70260 Kuopio, Finland. Tel.: q358-17-3647-654. E-mail address: [email protected] (T.A. Soininvaara).
for the long-term prediction of adaptive remodelling, based on a finite numerical element model, and reported severe bone loss in the anterior and mid-distal femur behind the anterior flange. Several studies describe a significant decrease in postoperative bone mineral density (BMD) of up to 44% adjacent to the implants after TKA w1,2,4–6,9– 12x. Prosthesis-related bone loss is thought to occur mainly as a result of stress-shielding, although immobilization in combination with local bone and tissue reactions to operative trauma have separate effects on bone loss w2,4,10,12x. The loss of bone in the distal anterior femur after TKA has been cited as a risk factor for supracondylar fractures of the femur. Although periprosthetic fractures are not common after TKA, they present a treatment dilemma. Bone loss in the distal anterior femur can also
0968-0160/04/$ - see front matter 䊚 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.knee.2003.09.006
T.A. Soininvaara et al. / The Knee 11 (2004) 297–302
298
Table 1 General baseline characteristics and AKS (American Knee Society) score values at follow-up Baseline characteristics
Male
Female
No of patients Age (years) (S.D.) Weight (kg) (S.D.) Height (cm) (S.D.) Body mass index (S.D.)
20 67 85 174 28
49 67 78 161 30
AKS score values (S.D.) At baseline At 3 months At 12 months
105 (22.0) 167 (21.2) 176 (23.3)
(8.2) (8.9) (6.4) (3.0)
(6.2) (14.5) (6.3) (5.1)
93 (29.5) 164 (26.2) 173 (32.5)
lead to loosening of the component and cause difficulties during a revision knee arthroplasty. Loosening of the femoral component may result from condylar osteoporosis w2,4,5,7–10,13–16x. Traditionally, the results of TKA have been evaluated from the postoperative clinical status (knee function, stability, range of motion, painfulness) and plain radiographs. Plain radiographs can be used to assess the position of a prosthesis and the alignment of the knee joint, to evaluate the bone–prosthesis and the bone– cement interfaces, and to provide evidence of infection, loosening, or subsidence w2,8,14x. The quantitative evaluation of periprosthetic bone density is unreliable when made with plain radiographs, which detect only the presence or absence of bone loss. Dual-energy X-ray absorptiometry (DXA) measurements, on the other hand, correlate strongly with the ash content of the bone and are seven times more accurate than visual evaluation w14x. Changes in bone density must exceed 20–50% to be visually distinguishable on standard radiographs w2,3,5–9,11,14x. Computer processing of radiographs allows the detection of losses of 8% and greater, whereas DXA can detect bone losses below 8%. DXA can provide reproducible, high-quality measurements of periprosthetic BMD. Commercial software allows the measurement of bone density adjacent to metal implants w6,8,9,11,14,17x, as confirmed in our previous study, with an average precision error of 1.3–3.1% in the femoral regions of interest (ROI). The aim of the present study was to quantify early BMD changes in the distal femur after cemented TKA, using DXA measurements. We focused particular attention on a 1-year follow-up, the period during which it is expected that early rapid postoperative stress-shielding bone loss will occur. We were also interested in collecting more information on the behaviour of BMD in the presence of cemented total knee prostheses, and to elucidate the roles of other possible factors involved in BMD changes. These issues have been inadequately characterised in previously published DXA-based studies, which have included rather small numbers of patients.
2. Materials and methods Sixty-nine patients were recruited from the orthopedic department at Kuopio University Hospital, Kuopio, Finland, between May 6, 1997 and May 11, 2000. The study population comprised patients with primary (ns 65) or posttraumatic (ns4) knee osteoarthrosis. Seventy-one per cent of the patients were female (ns49). A previous joint replacement of either knee was an exclusion criterion. Patients were free from diseases and all medications known to influence bone-mineral metabolism throughout the follow-up period. The mean age of the patients at the time of the operation was 67 (S.D. 6.8) years, with mean weight 80 (S.D. 13.5) kg and mean body mass index (BMI) 29.5 (S.D. 4.7) kgym2 . The baseline characteristics of patients, together with the American Knee Society (AKS) scores w18x, are presented in Table 1. This study protocol was approved by the Kuopio University Local Research Ethics Committee. All patients provided written informed consent. Bone mineral density in the distal femur was measured postoperatively within one week of TKA and at 3-, 6- and 12-month follow-ups, using a fan-beam dual X-ray absorptiometer (Expert XL, Lunar Co., Madison, WI). The measurement areas (ROI) were both metaphyseal (close to the prosthesis) and diaphyseal (above the implant) (Fig. 1). Cemented TKA was performed on these patients using modular prostheses from Duracon (ns37; Howmedica
Fig. 1. Periprosthetic regions of interest (ROIs) in the distal femur: metaphyseal anterior (ROI 1), central (ROI 2), posterior (ROI 3), diaphyseal (ROI 4), and total metaphyseal (ROI 5).
T.A. Soininvaara et al. / The Knee 11 (2004) 297–302
299
Table 2 Mean (S.D.) femoral periprosthetic BMD (gycm2) values at 0, 3, 6 and 12 months follow-up of 69 patients ROI
0 months
Anterior metaphyseal Central metaphyseal Posterior metaphyseal Total metaphyseal Diaphyseal
1.26 1.41 1.74 1.47 1.36
*
(0.23) (0.34) (0.31) (0.27) (0.24)
3 months 1.12 1.20 1.53 1.28 1.30
(0.24) (0.34) (0.34) (0.28) (0.25)
6 months 1.08 1.16 1.52 1.25 1.27
(0.26) (0.36) (0.31) (0.29) (0.25)
12 months 1.05 1.10 1.52 1.22 1.25
(0.24)*-0.0005 (0.33)*-0.0005 (0.31)*-0.0005 (0.27)*-0.0005 (0.24)*-0.0005
Significant BMD changes during the follow-up (epsilon corrected repeated measures ANOVA).
Inc., Rutherford, NJyInternational Division of Pfizer), Nexgen (ns23; Zimmer, Warsaw, IN), or AMK (ns9; DePuy, Division of Boehringer Mannheim Corporationy Depuy, Warsaw, IN). All patients were operated on by senior orthopedic surgeons. Full weight-bearing was allowed immediately after the operation. The AKS score was used to evaluate both the knee status and function during the daily activities of the study subjects. The AKS score was registered preoperatively and at each follow-up visit by the orthopedic surgeon. At follow-up visits, a long-standing X-ray was taken to measure the tibiofemoral angle necessary for AKS scoring. The maximum AKS score value is 200, consisting of knee status (100) and knee function (100) w18x. During an extra postoperative appointment with a study-nurse at 6 months, BMD was measured and overall rehabilitation was evaluated with a questionnaire. All patients received prophylactic antibiotic (cefuroxime) and antithrombotic (enoxaparinydalteparin) medication. Two patients received subsequent antibiotics for superficial wound irritation, without any positive bacterial culture. Both patients were cured. No complications were encountered and all patients attended scheduled follow-up examinations regularly. The 95% confidence intervals (CI) were calculated for the changes in BMD. For statistical analyses, we used SPSS software, version 10.0 (SPSS Inc., Chicago, IL). Data from each follow-up visit were confirmed to be normally distributed by the Shapiro–Wilk test. Comparisons of the changes in BMD and AKS scores were performed with epsilon-corrected repeated measures ANOVA. The influence of patient age, gender, BMI, and changes in AKS score was also considered. 3. Results The highest periprosthetic bone-loss rate was observed during the first 3 months after TKA: the average bonedensity loss was 4.5% in the diaphyseal ROI and ranged from 11 to 15.6% in the metaphyseal ROIs. At the 12month follow-up, the mean bone-density loss was 8.4% in the diaphyseal ROI, whereas the highest bone losses were seen in the central metaphyseal ROI (22.8%) and total metaphyseal ROI (17.1%). The bone loss was in the posterior metaphyseal ROI (12.1%). All BMD
reductions were significant by the epsilon-corrected repeated measures ANOVA. There were no significant differences in bone loss between genders or prosthesis models used in any of the measured ROIs. The bone loss rate diminished after 6 months, but bone loss continued (Table 2). The magnitude of the BMD change was not significantly effected by the baseline BMD values. By multivariate analysis, the higher BMI was correlated with the smaller bone loss, and there was a tendency for age to predict BMD decrease (e.g. in the metaphyseal total ROI, Ps0.019 and Ps0.067 for BMI and age, respectively). The mean AKS score was 95.6 (S.D. 28.0) before surgery, 165 (S.D. 24.7) at 3 months, and 174 (29) at 12 months. Gender (Ps0.749) and change in AKS score (Ps0.204) were not associated with BMD changes. The clinical and functional status of the knee improved steadily during the follow-up period. 4. Discussion This prospective 1-year DXA study of 69 patients revealed a tendency for major bone loss to occur in the distal femur during the first 3 months after cemented TKA. Reductions in BMD continued up to 12 months, although after 6 months the changes were minor. Gender, the prosthesis model used, and changes in the AKS score did not account for the rate of bone loss in any ROI. Only BMI correlated with the changes in BMD. Average bone-density loss ranged from 11 to 15.6% during the first 3 postoperative months in all metaphyseal ROIs examined. Our results are similar to those of Spittelhouse et al. w9x, who reported the greatest BMD decrease (16%) in the distal anterior femur over the first 6-month postoperative period in 16 patients with uncemented knee prostheses. These reductions in BMD were most significant during the early postoperative phase, which might be related to postoperative stressshielding w2,5,9,10x. Total knee arthroplasty particularly eliminates the compression between the patella and the distal femur. The normal patellofemoral contact pressure force has been estimated to reach 4600 N (6.5 times the normal body weight) w1–3,6,9–11x, and arthroplasty inevitably results in a stress-shielded region. Early bone loss might be accelerated by operative trauma and
300
T.A. Soininvaara et al. / The Knee 11 (2004) 297–302
postoperative immobilization, as well as by the altered mechanical situation. After 12 months, the mean bone-density loss ranged from 12.1 to 22.8%. The highest bone losses were seen in the central metaphyseal ROI. This relatively high bone loss might be explained by the location of the ROI in the center of the distal femur, where the intramedullary guiding rod is introduced during surgery. This measured change may reflect also resorption of the used bone plug used to fill the insertion hole. The overall bone loss was less pronounced in the posterior metaphyseal ROI than in the other metaphyseal ROIs, being limited to the first 3 months. Previously, Petersen et al. w5x found a BMD loss after 2 years of 36% behind the anterior flange of a femoral component in eight patients with uncemented TKA. Petersen et al. w10x subsequently showed significant (19–44%) bone loss in the distal femur after 1 year, using DXA analysis of 29 uncemented femoral components. Seki et al. retrospectively studied the radiographs of 114 well-functioning knees with four different prosthesis models, and showed a decrease in bone density up to 19% and 41% with uncemented and cemented components, respectively, in the distal femurs after 1 year w1x. Karbowski et al. compared postoperative and 9 months follow-up BMD values for periprosthetic distal femurs in 12 patients after TKA w11x. Their study revealed an average bone loss of 21.5% in the distal femurs with an uncemented implant. Similarly, Van Loon et al. showed a median decrease in BMD of 22% in the region behind the anterior flange of a cemented femoral component in 12 patients at the 1-year follow-up w12x. Our results after 1 year are in close agreement with the results of Karbowski et al. and van Loon et al. w11,12x. However, it has been suggested that the higher levels of bone loss associated with cemented components, which have been reported in several studies, might result from a more rigid fixation leading to more impressive stress-shielding compared with that provided by the more flexible bone ingrowth fixation. However, our bone-loss rate is equal to that found in some studies of cementless prostheses w1,11x. Therefore, it is hard to evaluate the factors contributing to the success of different fixation methods. Our reference ROI was situated 1 cm above the implant (Fig. 1), where stress-shielding is assumed to be negligible. In this diaphyseal ROI, a mean bone loss of 8.4% was recorded at 1 year. Our result is in close agreement with that of van Loon et al. who showed a median decrease of 8% just above the femoral component at the 1-year follow-up w12x. Seki et al. found diaphyseal BMD to be relatively stable, with 1.2–1.7% bone loss during the follow-up period (Fig. 2). Their reference zone was smaller and higher above the prosthesis than ours. We suggest that the less pronounced bone loss detected in the diaphyseal ROI represents both operation-related and postoperative immobilization-
induced bone loss, because age-related bone loss is minor w7,19x. After the first 3 months the physical activity, particularly walking increased, as demonstrated by the improvement in AKS score. Furthermore, bone loss continued in this ROI also in the period from 6 to 12 months at a very similar rate to that of total metaphyseal ROI. This seems to partly be a consequence of arthroplasty, though not direct stress-shielding, and needs to be noted weaker bone quality (supracondylar fracture risk). The clinical and functional status of the knee joint, evaluated with the AKS score w18x, improved steadily and statistically significantly during the follow-up w12x. Paradoxically, the major improvement in the clinical and functional status of the knee was insufficient to overcome the overriding postoperative early-phase stressshielding phenomenon. We could not find association between initial BMD and subsequent BMD changes. However, a higher BMI was associated with less bone loss. The positive correlation between BMI and BMD has been reported previously w20x. From a clinical point of view, high BMD is considered to give a better support for bone implant fixation. A number of cementing techniques, as well as uncemented arthroplasties, and numerous implant modifications have been introduced to improve the quality of the primary arthroplasty w2,3,9,10,12x. Bone loss occurs with all designs of TKA, and the prevalence of bone loss seems highly independent of the fixation method w2,4,5, 9,10,12x. Loosening is associated with bone resorption at the implant surface, and possible causes include wear particles, instability, hydrostatic pressure, and poor bone quality w3x. Theoretically, component loosening or increased risk of periprosthetic fractures could occur in patients with greater postoperative bone loss. Moreover, the regions of low bone density adjacent to the femoral component could lead to difficulties in revision surgery because of the absence of solid bone mass w1,3, 5,8,10,12,14,16x. We have described a rapid decrease in periprosthetic metaphyseal bone density after cemented TKA, using a precise and high-resolution DXA method w3,5,8,10, 11,14x. The profound rapid early-phase stress-shielding phenomenon is sustained during the first 3 postoperative months. The strength of the phenomenon is too great to be explained by factors such as gender or the prosthesis model used. Not even a significant improvement in AKS score values, which was one of the key issues we wished to evaluate with this analysis, affected bone loss. Only a higher BMI was found to correlate (negatively) with bone loss. Because the average metaphyseal bone loss (up to 22.8%) observed is of a significant magnitude and cannot be ignored, it is necessary to pursue followup analyses to study the possible determinants and mechanisms of periprosthetic bone loss after cemented TKA. Furthermore, a study period of 1 year is probably
T.A. Soininvaara et al. / The Knee 11 (2004) 297–302
301
Fig. 2. Mean relative BMD change (%) in the anterior (a), central (b), posterior (c), total metaphyseal (d) ROIs, and the diaphyseal (e) ROI in the distal femur, with different prosthesis models.
insufficient to trace the meaning and significance of periprosthetic femoral bone loss in osteolysis or aseptic loosening. Acknowledgments The authors thank Riitta Toroi, R.N. and Eila Koski, R.N. for their technical assistance, and Pirjo Halonen, biostatistician, M.Sc., for her statistical assistance. References w1x Seki T, Omori G, Koga Y, Suzuki Y, Ishii Y, Takahashi HE. Is bone density in the distal femur affected by use of cement and by femoral component design in total knee arthroplasty? J Orthop Sci 1999;4:180 –186.
w2x Mintzer CM, Robertson DD, Rackemann S, Ewald FC, Scott RD, Spector M. Bone loss in the distal anterior femur after total knee arthroplasty. Clin Orthop November 1990;260:135 – 143. w3x Lewis PL, Brewster NT, Graves SE. The pathogenesis of bone loss following total knee arthroplasty. Orthop Clin North Am April 1998;29:187 –197. w4x Van Lenthe GH, de Waal Malefijt MC, Huiskes R. Stress shielding after total knee replacement may cause bone resorption in the distal femur. J Bone Jt Surg Br January 1997;79:117 –122. w5x Petersen MM, Olsen C, Lauritzen JB, Lund B. Changes in bone mineral density of the distal femur following uncemented total knee arthroplasty. J Arthroplasty February 1995a;10:7 – 11. w6x Trevisan C, Bigoni M, Denti M, Marinoni EC, Ortolani S. Bone assessment after total knee arthroplasty by dual-energy
302
w7 x
w8 x
w9 x
w10x
w11x
w12x
T.A. Soininvaara et al. / The Knee 11 (2004) 297–302 X-ray absorptiometry: analysis protocol and reproducibility. Calcif Tissue Int April 1998;62:359 –361. Trevisan C, Ortolani S. Periprostetic bone mineral density and other orthopedic applications. In: Genant HK, Gugliemi G, Jegars Me, editors. Bone densitometry and osteoporosis. New York: Springer, 1998. p. 541 –582. Soininvaara T, Kroger H, Jurvelin JS, Miettinen H, Suomalainen O, Alhava E. Measurement of bone density around total knee arthroplasty using fan-beam dual energy X-ray absorptiometry. Calcif Tissue Int September 2000;67:267 –272. Spittlehouse AJ, Getty CJ, Eastell R. Measurement of bone mineral density by dual-energy X-ray absorptiometry around an uncemented knee prosthesis. J Arthroplasty December 1999;14:957 –963. Petersen MM, Lauritzen JB, Pedersen JG, Lund B. Decreased bone density of the distal femur after uncemented knee arthroplasty. A 1-year follow-up of 29 knees. Acta Orthop Scand August 1996;67:339 –344. Karbowski A, Schwitalle M, Eckardt A, Heine J. Periprosthetic bone remodelling after total knee arthroplasty: early assessment by dual energy X-ray absorptiometry. Arch Orthop Trauma Surg 1999;119:324 –326. van Loon CJ, Oyen WJ, de Waal Malefijt MC, Verdonschot N. Distal femoral bone mineral density after total knee arthroplasty: a comparison with general bone mineral density. Arch Orthop Trauma Surg May 2001;121:282 –285.
w13x Levitz CL, Lotke PA, Karp JS. Long-term changes in bone mineral density following total knee replacement. Clin Orthop December 1995;321:68 –72. w14x Robertson DD, Mintzer CM, Weissman BN, Ewald FC, LeBoff M, Spector M. Distal loss of femoral bone following total knee arthroplasty. Measurement with visual and computer-processing of roentgenograms and dual-energy X-ray absorptiometry. J Bone Jt Surg Am January 1994;76:66 –76. w15x Rorabeck CH, Taylor JW. Periprosthetic fractures of the femur complicating total knee arthroplasty. Orthop Clin North Am April 1999;30:265 –277. w16x DiGioia AM 3rd, Rubash HE. Periprosthetic fractures of the femur after total knee arthroplasty. A literature review and treatment algorithm. Clin Orthop October 1991;271:135 –142. w17x Kroger H, Venesmaa P, Jurvelin J, Miettinen H, Suomalainen O, Alhava E. Bone density at the proximal femur after total hip arthroplasty. Clin Orthop July 1998;352:66 –74. w18x Insall JN, Dorr LD, Scott RD, Scott WN. Rationale of the knee society clinical rating system. Clin Orthop November 1989;248:13 –14. w19x Burger H, van Daele PL, Algra D, van den Ouweland FA, Grobbee DE, Hofman A, et al. The association between age and bone mineral density in men and women aged 55 years and over: the Rotterdam study. Bone Miner April 1994;25:1 – 13. w20x Kroger H, Tuppurainen M, Honkanen R, Alhava E, Saarikoski E. Bone mineral density and risk factors for osteoporosis – a population-based study of 1600 perimenopausal women. Calcif Tissue Int July 1994;55:1 –7.
Periprosthetic femoral bone loss after total knee arthroplasty: 1-year follow-up study of 69 patients Tarja A. Soininvaaraa,b,*, Hannu J.A. Miettinenb, Jukka S. Jurvelinc, Olavi T. Suomalainenb, ¨ b,d Esko M. Alhavab, Heikki P.J. Kroger a
Department of Surgery, Savonlinna Central Hospital, Savonlinna, Finland b Department of Surgery, Kuopio University Hospital, Kuopio, Finland c Department of Clinical Physiology and Nuclear Medicine, Kuopio University Hospital and Department of Applied Physics, University of Kuopio, Kuopio, Finland d Bone and Cartilage Research Unit (BCRU), University of Kuopio, Kuopio, Finland Received 1 September 2003; accepted 30 September 2003
Abstract The clinical survival of joint arthroplasties is related to the quality of the surrounding bone environment. Bone mineral density (BMD) is an important measure of bone strength and quality. The aim of this prospective study was to measure the quantitative changes in BMD in the distal femur after cemented total knee arthroplasty (TKA) in osteoarthrotic knee joints. Sixty-nine patients with TKA were scanned postoperatively using dual-energy X-ray absorptiometry (DXA) within a week of surgery, and at 3-, 6-, and 12-month follow-ups. An average decrease in bone density of 17.1% (mean range of 12.1–22.8%) was measured adjacent to the prosthesis at the 12-month follow-up (repeated measures ANOVA P-0.0005). Bone loss was most rapid during the first 3 months after TKA. The clinical status and function parameters of the knee joint, evaluated by the American Knee Society (AKS) score, had improved significantly on the preoperative values at the three- and 12-month follow-ups (P-0.0005). However, improvement in the AKS score was not associated with periprosthetic BMD change (Ps0.204), whereas age (Ps0.067) and body mass index (Ps0.019) correlated with BMD loss for the total metaphyseal region of interest (ROI), by repeated measures ANOVA. We suggest that the observed periprosthetic bone loss was mainly the result of prosthesis-related stress-shielding. 䊚 2003 Elsevier B.V. All rights reserved. Keywords: Bone loss; BMD; Total knee arthroplasty; Follow-up
1. Introduction Total knee arthroplasty (TKA) alters the mechanical loading of the knee joint and causes adaptive bone remodelling. The bone surrounding the TKA adjusts its mineral density and structure to meet the new mechanical demands. The ‘cup-like’ shape of the femoral component results in the protection from stress of the distal femoral bone and protection from shear loading at the attachment of the anterior and posterior flange surfaces w1x. Moreover, the absence of compressive loading of the patella on the surface of the distal femur may result in stress-shielding of the patellofemoral region w1–10x. Van Lenthe et al. w4x introduced a theory *Corresponding author. Present address: Puijonsarventie 11 D 30, 70260 Kuopio, Finland. Tel.: q358-17-3647-654. E-mail address: [email protected] (T.A. Soininvaara).
for the long-term prediction of adaptive remodelling, based on a finite numerical element model, and reported severe bone loss in the anterior and mid-distal femur behind the anterior flange. Several studies describe a significant decrease in postoperative bone mineral density (BMD) of up to 44% adjacent to the implants after TKA w1,2,4–6,9– 12x. Prosthesis-related bone loss is thought to occur mainly as a result of stress-shielding, although immobilization in combination with local bone and tissue reactions to operative trauma have separate effects on bone loss w2,4,10,12x. The loss of bone in the distal anterior femur after TKA has been cited as a risk factor for supracondylar fractures of the femur. Although periprosthetic fractures are not common after TKA, they present a treatment dilemma. Bone loss in the distal anterior femur can also
0968-0160/04/$ - see front matter 䊚 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.knee.2003.09.006
T.A. Soininvaara et al. / The Knee 11 (2004) 297–302
298
Table 1 General baseline characteristics and AKS (American Knee Society) score values at follow-up Baseline characteristics
Male
Female
No of patients Age (years) (S.D.) Weight (kg) (S.D.) Height (cm) (S.D.) Body mass index (S.D.)
20 67 85 174 28
49 67 78 161 30
AKS score values (S.D.) At baseline At 3 months At 12 months
105 (22.0) 167 (21.2) 176 (23.3)
(8.2) (8.9) (6.4) (3.0)
(6.2) (14.5) (6.3) (5.1)
93 (29.5) 164 (26.2) 173 (32.5)
lead to loosening of the component and cause difficulties during a revision knee arthroplasty. Loosening of the femoral component may result from condylar osteoporosis w2,4,5,7–10,13–16x. Traditionally, the results of TKA have been evaluated from the postoperative clinical status (knee function, stability, range of motion, painfulness) and plain radiographs. Plain radiographs can be used to assess the position of a prosthesis and the alignment of the knee joint, to evaluate the bone–prosthesis and the bone– cement interfaces, and to provide evidence of infection, loosening, or subsidence w2,8,14x. The quantitative evaluation of periprosthetic bone density is unreliable when made with plain radiographs, which detect only the presence or absence of bone loss. Dual-energy X-ray absorptiometry (DXA) measurements, on the other hand, correlate strongly with the ash content of the bone and are seven times more accurate than visual evaluation w14x. Changes in bone density must exceed 20–50% to be visually distinguishable on standard radiographs w2,3,5–9,11,14x. Computer processing of radiographs allows the detection of losses of 8% and greater, whereas DXA can detect bone losses below 8%. DXA can provide reproducible, high-quality measurements of periprosthetic BMD. Commercial software allows the measurement of bone density adjacent to metal implants w6,8,9,11,14,17x, as confirmed in our previous study, with an average precision error of 1.3–3.1% in the femoral regions of interest (ROI). The aim of the present study was to quantify early BMD changes in the distal femur after cemented TKA, using DXA measurements. We focused particular attention on a 1-year follow-up, the period during which it is expected that early rapid postoperative stress-shielding bone loss will occur. We were also interested in collecting more information on the behaviour of BMD in the presence of cemented total knee prostheses, and to elucidate the roles of other possible factors involved in BMD changes. These issues have been inadequately characterised in previously published DXA-based studies, which have included rather small numbers of patients.
2. Materials and methods Sixty-nine patients were recruited from the orthopedic department at Kuopio University Hospital, Kuopio, Finland, between May 6, 1997 and May 11, 2000. The study population comprised patients with primary (ns 65) or posttraumatic (ns4) knee osteoarthrosis. Seventy-one per cent of the patients were female (ns49). A previous joint replacement of either knee was an exclusion criterion. Patients were free from diseases and all medications known to influence bone-mineral metabolism throughout the follow-up period. The mean age of the patients at the time of the operation was 67 (S.D. 6.8) years, with mean weight 80 (S.D. 13.5) kg and mean body mass index (BMI) 29.5 (S.D. 4.7) kgym2 . The baseline characteristics of patients, together with the American Knee Society (AKS) scores w18x, are presented in Table 1. This study protocol was approved by the Kuopio University Local Research Ethics Committee. All patients provided written informed consent. Bone mineral density in the distal femur was measured postoperatively within one week of TKA and at 3-, 6- and 12-month follow-ups, using a fan-beam dual X-ray absorptiometer (Expert XL, Lunar Co., Madison, WI). The measurement areas (ROI) were both metaphyseal (close to the prosthesis) and diaphyseal (above the implant) (Fig. 1). Cemented TKA was performed on these patients using modular prostheses from Duracon (ns37; Howmedica
Fig. 1. Periprosthetic regions of interest (ROIs) in the distal femur: metaphyseal anterior (ROI 1), central (ROI 2), posterior (ROI 3), diaphyseal (ROI 4), and total metaphyseal (ROI 5).
T.A. Soininvaara et al. / The Knee 11 (2004) 297–302
299
Table 2 Mean (S.D.) femoral periprosthetic BMD (gycm2) values at 0, 3, 6 and 12 months follow-up of 69 patients ROI
0 months
Anterior metaphyseal Central metaphyseal Posterior metaphyseal Total metaphyseal Diaphyseal
1.26 1.41 1.74 1.47 1.36
*
(0.23) (0.34) (0.31) (0.27) (0.24)
3 months 1.12 1.20 1.53 1.28 1.30
(0.24) (0.34) (0.34) (0.28) (0.25)
6 months 1.08 1.16 1.52 1.25 1.27
(0.26) (0.36) (0.31) (0.29) (0.25)
12 months 1.05 1.10 1.52 1.22 1.25
(0.24)*-0.0005 (0.33)*-0.0005 (0.31)*-0.0005 (0.27)*-0.0005 (0.24)*-0.0005
Significant BMD changes during the follow-up (epsilon corrected repeated measures ANOVA).
Inc., Rutherford, NJyInternational Division of Pfizer), Nexgen (ns23; Zimmer, Warsaw, IN), or AMK (ns9; DePuy, Division of Boehringer Mannheim Corporationy Depuy, Warsaw, IN). All patients were operated on by senior orthopedic surgeons. Full weight-bearing was allowed immediately after the operation. The AKS score was used to evaluate both the knee status and function during the daily activities of the study subjects. The AKS score was registered preoperatively and at each follow-up visit by the orthopedic surgeon. At follow-up visits, a long-standing X-ray was taken to measure the tibiofemoral angle necessary for AKS scoring. The maximum AKS score value is 200, consisting of knee status (100) and knee function (100) w18x. During an extra postoperative appointment with a study-nurse at 6 months, BMD was measured and overall rehabilitation was evaluated with a questionnaire. All patients received prophylactic antibiotic (cefuroxime) and antithrombotic (enoxaparinydalteparin) medication. Two patients received subsequent antibiotics for superficial wound irritation, without any positive bacterial culture. Both patients were cured. No complications were encountered and all patients attended scheduled follow-up examinations regularly. The 95% confidence intervals (CI) were calculated for the changes in BMD. For statistical analyses, we used SPSS software, version 10.0 (SPSS Inc., Chicago, IL). Data from each follow-up visit were confirmed to be normally distributed by the Shapiro–Wilk test. Comparisons of the changes in BMD and AKS scores were performed with epsilon-corrected repeated measures ANOVA. The influence of patient age, gender, BMI, and changes in AKS score was also considered. 3. Results The highest periprosthetic bone-loss rate was observed during the first 3 months after TKA: the average bonedensity loss was 4.5% in the diaphyseal ROI and ranged from 11 to 15.6% in the metaphyseal ROIs. At the 12month follow-up, the mean bone-density loss was 8.4% in the diaphyseal ROI, whereas the highest bone losses were seen in the central metaphyseal ROI (22.8%) and total metaphyseal ROI (17.1%). The bone loss was in the posterior metaphyseal ROI (12.1%). All BMD
reductions were significant by the epsilon-corrected repeated measures ANOVA. There were no significant differences in bone loss between genders or prosthesis models used in any of the measured ROIs. The bone loss rate diminished after 6 months, but bone loss continued (Table 2). The magnitude of the BMD change was not significantly effected by the baseline BMD values. By multivariate analysis, the higher BMI was correlated with the smaller bone loss, and there was a tendency for age to predict BMD decrease (e.g. in the metaphyseal total ROI, Ps0.019 and Ps0.067 for BMI and age, respectively). The mean AKS score was 95.6 (S.D. 28.0) before surgery, 165 (S.D. 24.7) at 3 months, and 174 (29) at 12 months. Gender (Ps0.749) and change in AKS score (Ps0.204) were not associated with BMD changes. The clinical and functional status of the knee improved steadily during the follow-up period. 4. Discussion This prospective 1-year DXA study of 69 patients revealed a tendency for major bone loss to occur in the distal femur during the first 3 months after cemented TKA. Reductions in BMD continued up to 12 months, although after 6 months the changes were minor. Gender, the prosthesis model used, and changes in the AKS score did not account for the rate of bone loss in any ROI. Only BMI correlated with the changes in BMD. Average bone-density loss ranged from 11 to 15.6% during the first 3 postoperative months in all metaphyseal ROIs examined. Our results are similar to those of Spittelhouse et al. w9x, who reported the greatest BMD decrease (16%) in the distal anterior femur over the first 6-month postoperative period in 16 patients with uncemented knee prostheses. These reductions in BMD were most significant during the early postoperative phase, which might be related to postoperative stressshielding w2,5,9,10x. Total knee arthroplasty particularly eliminates the compression between the patella and the distal femur. The normal patellofemoral contact pressure force has been estimated to reach 4600 N (6.5 times the normal body weight) w1–3,6,9–11x, and arthroplasty inevitably results in a stress-shielded region. Early bone loss might be accelerated by operative trauma and
300
T.A. Soininvaara et al. / The Knee 11 (2004) 297–302
postoperative immobilization, as well as by the altered mechanical situation. After 12 months, the mean bone-density loss ranged from 12.1 to 22.8%. The highest bone losses were seen in the central metaphyseal ROI. This relatively high bone loss might be explained by the location of the ROI in the center of the distal femur, where the intramedullary guiding rod is introduced during surgery. This measured change may reflect also resorption of the used bone plug used to fill the insertion hole. The overall bone loss was less pronounced in the posterior metaphyseal ROI than in the other metaphyseal ROIs, being limited to the first 3 months. Previously, Petersen et al. w5x found a BMD loss after 2 years of 36% behind the anterior flange of a femoral component in eight patients with uncemented TKA. Petersen et al. w10x subsequently showed significant (19–44%) bone loss in the distal femur after 1 year, using DXA analysis of 29 uncemented femoral components. Seki et al. retrospectively studied the radiographs of 114 well-functioning knees with four different prosthesis models, and showed a decrease in bone density up to 19% and 41% with uncemented and cemented components, respectively, in the distal femurs after 1 year w1x. Karbowski et al. compared postoperative and 9 months follow-up BMD values for periprosthetic distal femurs in 12 patients after TKA w11x. Their study revealed an average bone loss of 21.5% in the distal femurs with an uncemented implant. Similarly, Van Loon et al. showed a median decrease in BMD of 22% in the region behind the anterior flange of a cemented femoral component in 12 patients at the 1-year follow-up w12x. Our results after 1 year are in close agreement with the results of Karbowski et al. and van Loon et al. w11,12x. However, it has been suggested that the higher levels of bone loss associated with cemented components, which have been reported in several studies, might result from a more rigid fixation leading to more impressive stress-shielding compared with that provided by the more flexible bone ingrowth fixation. However, our bone-loss rate is equal to that found in some studies of cementless prostheses w1,11x. Therefore, it is hard to evaluate the factors contributing to the success of different fixation methods. Our reference ROI was situated 1 cm above the implant (Fig. 1), where stress-shielding is assumed to be negligible. In this diaphyseal ROI, a mean bone loss of 8.4% was recorded at 1 year. Our result is in close agreement with that of van Loon et al. who showed a median decrease of 8% just above the femoral component at the 1-year follow-up w12x. Seki et al. found diaphyseal BMD to be relatively stable, with 1.2–1.7% bone loss during the follow-up period (Fig. 2). Their reference zone was smaller and higher above the prosthesis than ours. We suggest that the less pronounced bone loss detected in the diaphyseal ROI represents both operation-related and postoperative immobilization-
induced bone loss, because age-related bone loss is minor w7,19x. After the first 3 months the physical activity, particularly walking increased, as demonstrated by the improvement in AKS score. Furthermore, bone loss continued in this ROI also in the period from 6 to 12 months at a very similar rate to that of total metaphyseal ROI. This seems to partly be a consequence of arthroplasty, though not direct stress-shielding, and needs to be noted weaker bone quality (supracondylar fracture risk). The clinical and functional status of the knee joint, evaluated with the AKS score w18x, improved steadily and statistically significantly during the follow-up w12x. Paradoxically, the major improvement in the clinical and functional status of the knee was insufficient to overcome the overriding postoperative early-phase stressshielding phenomenon. We could not find association between initial BMD and subsequent BMD changes. However, a higher BMI was associated with less bone loss. The positive correlation between BMI and BMD has been reported previously w20x. From a clinical point of view, high BMD is considered to give a better support for bone implant fixation. A number of cementing techniques, as well as uncemented arthroplasties, and numerous implant modifications have been introduced to improve the quality of the primary arthroplasty w2,3,9,10,12x. Bone loss occurs with all designs of TKA, and the prevalence of bone loss seems highly independent of the fixation method w2,4,5, 9,10,12x. Loosening is associated with bone resorption at the implant surface, and possible causes include wear particles, instability, hydrostatic pressure, and poor bone quality w3x. Theoretically, component loosening or increased risk of periprosthetic fractures could occur in patients with greater postoperative bone loss. Moreover, the regions of low bone density adjacent to the femoral component could lead to difficulties in revision surgery because of the absence of solid bone mass w1,3, 5,8,10,12,14,16x. We have described a rapid decrease in periprosthetic metaphyseal bone density after cemented TKA, using a precise and high-resolution DXA method w3,5,8,10, 11,14x. The profound rapid early-phase stress-shielding phenomenon is sustained during the first 3 postoperative months. The strength of the phenomenon is too great to be explained by factors such as gender or the prosthesis model used. Not even a significant improvement in AKS score values, which was one of the key issues we wished to evaluate with this analysis, affected bone loss. Only a higher BMI was found to correlate (negatively) with bone loss. Because the average metaphyseal bone loss (up to 22.8%) observed is of a significant magnitude and cannot be ignored, it is necessary to pursue followup analyses to study the possible determinants and mechanisms of periprosthetic bone loss after cemented TKA. Furthermore, a study period of 1 year is probably
T.A. Soininvaara et al. / The Knee 11 (2004) 297–302
301
Fig. 2. Mean relative BMD change (%) in the anterior (a), central (b), posterior (c), total metaphyseal (d) ROIs, and the diaphyseal (e) ROI in the distal femur, with different prosthesis models.
insufficient to trace the meaning and significance of periprosthetic femoral bone loss in osteolysis or aseptic loosening. Acknowledgments The authors thank Riitta Toroi, R.N. and Eila Koski, R.N. for their technical assistance, and Pirjo Halonen, biostatistician, M.Sc., for her statistical assistance. References w1x Seki T, Omori G, Koga Y, Suzuki Y, Ishii Y, Takahashi HE. Is bone density in the distal femur affected by use of cement and by femoral component design in total knee arthroplasty? J Orthop Sci 1999;4:180 –186.
w2x Mintzer CM, Robertson DD, Rackemann S, Ewald FC, Scott RD, Spector M. Bone loss in the distal anterior femur after total knee arthroplasty. Clin Orthop November 1990;260:135 – 143. w3x Lewis PL, Brewster NT, Graves SE. The pathogenesis of bone loss following total knee arthroplasty. Orthop Clin North Am April 1998;29:187 –197. w4x Van Lenthe GH, de Waal Malefijt MC, Huiskes R. Stress shielding after total knee replacement may cause bone resorption in the distal femur. J Bone Jt Surg Br January 1997;79:117 –122. w5x Petersen MM, Olsen C, Lauritzen JB, Lund B. Changes in bone mineral density of the distal femur following uncemented total knee arthroplasty. J Arthroplasty February 1995a;10:7 – 11. w6x Trevisan C, Bigoni M, Denti M, Marinoni EC, Ortolani S. Bone assessment after total knee arthroplasty by dual-energy
302
w7 x
w8 x
w9 x
w10x
w11x
w12x
T.A. Soininvaara et al. / The Knee 11 (2004) 297–302 X-ray absorptiometry: analysis protocol and reproducibility. Calcif Tissue Int April 1998;62:359 –361. Trevisan C, Ortolani S. Periprostetic bone mineral density and other orthopedic applications. In: Genant HK, Gugliemi G, Jegars Me, editors. Bone densitometry and osteoporosis. New York: Springer, 1998. p. 541 –582. Soininvaara T, Kroger H, Jurvelin JS, Miettinen H, Suomalainen O, Alhava E. Measurement of bone density around total knee arthroplasty using fan-beam dual energy X-ray absorptiometry. Calcif Tissue Int September 2000;67:267 –272. Spittlehouse AJ, Getty CJ, Eastell R. Measurement of bone mineral density by dual-energy X-ray absorptiometry around an uncemented knee prosthesis. J Arthroplasty December 1999;14:957 –963. Petersen MM, Lauritzen JB, Pedersen JG, Lund B. Decreased bone density of the distal femur after uncemented knee arthroplasty. A 1-year follow-up of 29 knees. Acta Orthop Scand August 1996;67:339 –344. Karbowski A, Schwitalle M, Eckardt A, Heine J. Periprosthetic bone remodelling after total knee arthroplasty: early assessment by dual energy X-ray absorptiometry. Arch Orthop Trauma Surg 1999;119:324 –326. van Loon CJ, Oyen WJ, de Waal Malefijt MC, Verdonschot N. Distal femoral bone mineral density after total knee arthroplasty: a comparison with general bone mineral density. Arch Orthop Trauma Surg May 2001;121:282 –285.
w13x Levitz CL, Lotke PA, Karp JS. Long-term changes in bone mineral density following total knee replacement. Clin Orthop December 1995;321:68 –72. w14x Robertson DD, Mintzer CM, Weissman BN, Ewald FC, LeBoff M, Spector M. Distal loss of femoral bone following total knee arthroplasty. Measurement with visual and computer-processing of roentgenograms and dual-energy X-ray absorptiometry. J Bone Jt Surg Am January 1994;76:66 –76. w15x Rorabeck CH, Taylor JW. Periprosthetic fractures of the femur complicating total knee arthroplasty. Orthop Clin North Am April 1999;30:265 –277. w16x DiGioia AM 3rd, Rubash HE. Periprosthetic fractures of the femur after total knee arthroplasty. A literature review and treatment algorithm. Clin Orthop October 1991;271:135 –142. w17x Kroger H, Venesmaa P, Jurvelin J, Miettinen H, Suomalainen O, Alhava E. Bone density at the proximal femur after total hip arthroplasty. Clin Orthop July 1998;352:66 –74. w18x Insall JN, Dorr LD, Scott RD, Scott WN. Rationale of the knee society clinical rating system. Clin Orthop November 1989;248:13 –14. w19x Burger H, van Daele PL, Algra D, van den Ouweland FA, Grobbee DE, Hofman A, et al. The association between age and bone mineral density in men and women aged 55 years and over: the Rotterdam study. Bone Miner April 1994;25:1 – 13. w20x Kroger H, Tuppurainen M, Honkanen R, Alhava E, Saarikoski E. Bone mineral density and risk factors for osteoporosis – a population-based study of 1600 perimenopausal women. Calcif Tissue Int July 1994;55:1 –7.