Bone Mineral Density of the Femur in Autopsy Retrieved Total Knee Arthroplasties

YARTH-53892; No of Pages 6 The Journal of Arthroplasty xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org 1

2 3

Q1 4 5

Bone Mineral Density of the Femur in Autopsy Retrieved Total Knee Arthroplasties Carlos J. Lavernia, MD a, Jose A. Rodriguez, MD b, David A. Iacobelli, MD a, c, David S. Hungerford, MD d, Kenneth A. Krackow, MD e

6 7 8 9 10

a

11

a r t i c l e

12 13 14 15 16 18 17 19 20 21 22 23 24 25 26

Center for Advanced Orthopedics at Larkin, Miami, Florida Fawcett Memorial Hospital Radio, Charlotte, Florida Arthritis Surgery Research Foundation, Miami, Florida d Johns Hopkins Orthopaedics at Good Samaritan Hospital, Baltimore, Maryland e Department of Orthopaedic Surgery, Kaleida/Buffalo General Hospital, New York b c

i n f o

Article history: Received 19 September 2013 Accepted 11 March 2014 Available online xxxx Keywords: DEXA bone mineral density distal femur stress shielding total knee arthroplasty knee arthroplasty

a b s t r a c t Bone mineral density (BMD), as measured by DEXA, can vary depending on bone rotation and fat content of soft tissues. We performed DEXA measurements, under controlled positioning, on 24 autopsy-retrieved femora from patients who had fully functional and asymptomatic successful TKA to determine periprosthetic BMD changes and compared results to 24 normal cadaveric femora. In TKA specimens, BMD was affected by gender, preoperative diagnosis, and zone under analysis. The lowest mean BMD was in the anterior femoral condylar zone. Males had higher mean BMD at all zones while patients with preoperative diagnosis of osteoarthritis had higher BMD in the posterior condylar zone. The mean BMD in the anterior femoral condylar zone in TKA specimens was significantly lower than in normal specimens without arthroplasties, most likely due to stress shielding. © 2014 Published by Elsevier Inc.

40

27 28 29 30 31 32 33 34 35 36 38 37

39 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

More than 700,000 total knee arthroplasty (TKA) procedures are performed annually in the United States alone [1]. Every year, the frequency of these procedures continues to rise [2]. Currently, an increasing number of procedures are being performed in younger patients [3,4]; TKA survivorship has been reported to be around 95% at mid-term to long-term follow-up periods according to large US joint registries [5,6]. A growing number of TKA revisions are expected in the future. Implants may fail for different reasons: loosening, infection, instability, arthrofibrosis, extensor mechanism deficiency, component malpositioning, and/or periprosthetic fracture [7–9]. To date, mechanical loosening and infection remain the top concerns of failure. Loosening of both the femoral and tibial components in TKA may result from bone quality changes at the bone–implant interface [10,11]. In hip arthroplasties, low preoperative BMD has been found to be a predictor of delayed osseointegration and component loosening [12,13]. The femoral periprosthetic bone quality around a TKA is directly affected by several factors such as type of degenerative joint disease (DJD) [14–17], quantity and quality of preexistent bone matrix [18–21], and bone remodeling following the procedure [22,23]. Bone remodeling is influenced by the orientation and magnitude of The Conflict of Interest statement associated with this article can be found at http:// dx.doi.org/10.1016/j.arth.2014.03.010. Reprint requests: Carlos J. Lavernia, MD, The Center for Advanced Orthopedics at Larkin, Miami, FL 33143.

the functional strains at the periprosthetic bone interface. An implant, in direct contact with bone, will shield certain areas from the mechanical forces needed for its normal maintenance and remodeling, causing it to become osteopenic and decreasing its strength [22,24], which may lead to long term implant failure and even complicate revision surgeries due to poor remaining bone mass around the implant [25,26]. Previous authors have reported their findings on the periprosthetic bone density changes around the components of TKA [27–37]. Currently the most precise and accurate method to quantitatively assess BMD has been achieved by dual energy x-ray absorptiometry (DEXA) [38,39]. It has been successfully utilized to measure periprosthetic bone density around femoral components in total knee arthroplasty [33–37]; however, DEXA scanning is not as accurate as previously thought [40]. The measured BMD can vary depending on the rotation of the femora [39,41–44], and the fat content of soft tissues surrounding the bone of in-vivo patients or in-situ cadavers [44,45]. Our main objective was to accurately determine the BMD in the femoral bone of clinically successful autopsy retrieved TKA specimens, and compare the ratios of the measurements to those in similar zones in normal cadaveric specimens. As secondary objectives we describe the differences in BMD related to patient’s age at surgery, gender, weight, length of implantation, method of fixation, the preoperative diagnosis that led to DJD, and roentgenographic findings on TKA specimens.

0883-5403/© 2014 Published by Elsevier Inc. http://dx.doi.org/10.1016/j.arth.2014.03.010

Please cite this article as: Lavernia CJ, et al, Bone Mineral Density of the Femur in Autopsy Retrieved Total Knee Arthroplasties, J Arthroplasty (2014), http://dx.doi.org/10.1016/j.arth.2014.03.010

61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85

2

C.J. Lavernia et al. / The Journal of Arthroplasty xxx (2014) xxx–xxx

86

Materials and Methods

87

Study Design

88

124 125

Forty-eight knee specimens were studied. Twenty-four were autopsy retrieved distal femurs that had previous TKA procedures performed by the senior authors (DSH, KAK). All the prostheses were primary porous coated anatomic (PCA) from Howmedica (Rutherford, NJ). The mean age at surgery for the TKA specimens was 72.6 ± 2.0 (SE) years (range, 53 to 87), sixteen were from females; nineteen were from patients with osteoarthritis (OA) and five from patients with rheumatoid arthritis (RA). Of the TKA specimens, 19 components were uncemented (12 women) and 5 were cemented (4 women). The average length of implantation was 75.6 ± 8.2 (range, 11 to 135) months. For the normal 24 knees (15 women), the mean age was 77.2 ± 2.9 (range, 41 to 91) years. Specimens were obtained at time of death from patients with fully functional asymptomatic TKA based upon quantitative knee scoring systems and roentgenographic reviews. All specimens were collected by trained personnel, wrapped in saline soaked towels, and immediately frozen at − 140 °C. In addition, twenty-four distal femurs with no known history of arthritis were obtained from human cadavers. These specimens were formalin fixed and upon dissection were stored at room temperature in air-tight bags. The rationale that we used in order to compare frozen samples against formalin-fixed knees, was based on the findings by Lochmuller et al, [44] who showed no difference in BMD measurements between non-fixed and formalin-fixed bones using DEXA scanning. Specimens were obtained as part of an IRB approved retrieval program instituted by one of the authors (DSH). Femoral rotation can affect the BMD measured with DEXA [39,41– 44]. Therbo et al [43] demonstrated these variations when the knee was rotated while measuring distal femoral TKA periprosthetic BMD in a lateral view, using a DEXA machine similar to ours (Norland XR-26 mark II, Norland Corporation, WI). Furthermore, fat contained within soft tissues of in-vivo or cadaveric in-situ bones can influence the BMD obtained with DEXA scan [40,44,45]. To reduce the variability of the BMD measurements within our study, an aparatus was developed in order to accommodate the knees and obtain a true lateral view. Additionally, the femora were stripped from their soft tissues. Thus, while performing the DEXA scanning, we guaranteed neither rotation nor superimposition of the femoral component nor influence of soft tissues, obtaining a more accurate BMD measurement.

126

Roentgenographic Analysis

127 128

The TKA specimens were thawed using a bath of warm normal saline and unwrapped for roentgenographic analysis and DEXA

89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123

scanning, as previous studies have shown that frozen samples yield a higher BMD when using DEXA scanning [46]. The normal specimens were removed from their sealed bags, analyzed and resealed for storage. In order to avoid the inherent error of soft tissues and fat composition that is introduced during scanning [40,44,45], soft tissue was removed from each knee to expose the femoral shaft and the femoral component of the prosthesis. Antero-posterior (AP) and lateral roentgenographs of all the specimens were obtained. A foam positioner was utilized to align the specimens in a true AP position. The specimens were then rotated and tilted until a true lateral roentgenograph view was also obtained. For the TKA specimens the presence of loose beads of the component in the femur was recorded from previous roentgenographs available.

129

DEXA Scanning

142

DEXA scanning evaluated the BMD of all the 48 specimens. The specimens were placed using the same pre-designed foam positioner in order to obtain a true lateral image. This device ensured precise positioning of successive specimens, and thus reduced the influence of the rotation of the anisotropic bone on the measured BMD [39,41–44]. DEXA measurements were obtained using a Norland XR-26 densitometer (Norland Corporation, WI) and analyzed using Norland software version 2.2.2. Bone mineral density was measured in g/cm 2. To ensure accuracy of the DEXA scanner, daily calibrations were performed employing a dual material standard, as recommended by the manufacturer. One investigator completed all BMD measurements for the TKA and normal knee specimens. Scan acquisition was started approximately 40 mm above the proximal end of the retrieved femoral component and continued until approximately 30 mm below the distal end of the femoral component. The scans were obtained using a pixel size of 1.5 mm × 1.5 mm and a scan speed of 60 mm/s. Average scan time duration was 4.5 min. The Norland software provided a subroutine to measure the density close to the bone–implant interface. Five zones on the lateral view were selected for bone density measurements in both sets of specimens (Fig. 1). Zones 1 and 2 referred to the anterior proximal and distal parts of the femoral condylar area, respectively (Fig. 1). Zone 3 and 4 referred to the distal and proximal mid part of the femoral condylar area, respectively (Fig. 1). Zone 5 referred to the posterior part of the femoral condylar area (Fig. 1). Each zone was referenced from specific locations on the implant or anatomical landmarks, thus allowing for the duplication of their relative size and location among all specimens. Using tools available in the Norland software package, the operator defined each zone manually on all the specimens. A mean BMD value was also calculated for the five zones in each of the specimens.

143 144

Fig. 1. Representation of the five zones used for BMD measurements in the knee specimens.

Please cite this article as: Lavernia CJ, et al, Bone Mineral Density of the Femur in Autopsy Retrieved Total Knee Arthroplasties, J Arthroplasty (2014), http://dx.doi.org/10.1016/j.arth.2014.03.010

130 131 132 133 134 135 136 137 138 139 140 141

145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173

C.J. Lavernia et al. / The Journal of Arthroplasty xxx (2014) xxx–xxx

174

Statistical Analysis

175

180 181

Linear regression analysis, Pearson correlations, and a mixed effects model were used. The later model was implemented using the PROC MIXED procedure in SAS (SAS Institute, Cary, NC), and was done to evaluate the relationships between BMD and the zone measured, patient age at surgery, gender, weight, length of implantation, method of fixation, preoperative diagnosis, and roentgenographic evidence of loose beads.

182

Results

183

TKA & Normal Specimens

184 185

213

Periprosthetic osteolysis and radiolucencies were not observed in any of the TKA specimens upon roentgenographic review. The mean BMD values (± SD) for each of the zones in the TKA and normal knee specimens, as well as for uncemented and cemented knee arthroplasty specimens are listed in Table 1 and graphed in Fig. 2. Correlation coefficients (r) among the different zones for the TKA and normal samples ranged from 0.52 to 0.94 (all P’s b 0.01) and 0.88 to 0.97 (all P’s b 0.001), respectively (Table 2). For the uncemented TKA knees “r” ranged from 0.56 to 0.95 (all P’s b 0.05) and from − 0.19 to 0.94 for the cemented knees (Table 2). For the TKA specimens, the mean BMD between the anteroproximal and antero-distal femur (zones 1 and 2; Table 1, Fig. 2) was not statistically different, and the mean BMD among mid and posterior femur (zones 3, 4, and 5; Table 1, Fig. 2) was not statistically different either. However, mean BMD of the anterior femur (zones 1 and 2; Table 1, Fig. 2) was significantly lower than that of the mid and posterior femur (zones 3, 4, and 5; Table 1, Fig. 2) (P = 0.0001). For the normal specimens, mean BMD between the antero-distal, mid and posterior femur (zones 2, 3, 4, and 5; Table 1, Fig. 2) was not statistically different, while the mean BMD of the antero-proximal femur (zone 1) was significantly lower than the rest of the femoral zones (Table 1, Fig. 2) (P = 0.0001). The antero-distal femur (zone 2), had the lowest BMD value for the TKA specimens and the highest for the normal specimens (0.74 vs. 1.22 g/cm 2; Table 1, Fig. 2), and the difference was statistically significant (P = 0.0001). Bone densities in the other zones were not significantly different between the TKA and normal knees (Table 1, Fig. 2).

214

Age

215

In normal knee samples, age was inversely correlated (r = 0.694) with BMD at all zones (P b 0.001), meaning that as age increased the BMD decreased. Further analysis of age influence on BMD of TKA specimens was carried out with the mixed effects model described below.

176 177 178 179

186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212

216 217 218 219 t1:1

Table 1 Average BMD ± SD (g/cm2) Zone Values for TKA and Normal Knee Specimens.

t1:2 t1:3

Zone

t1:4 t1:5 t1:6 t1:7 t1:8

1 2 3 4 5

t1:9 t1:10

a b

Normal (n = 24) 0.87 1.22 1.21 1.14 1.02

± ± ± ± ±

0.32 0.48a 0.45 0.40 0.36

TKA (n = 24) 0.83 0.74 1.05 1.10 1.13

± ± ± ± ±

0.23 0.33 0.34 0.39 0.25

Cemented (n = 5) 0.92 0.89 1.22 1.20 1.15

± ± ± ± ±

0.14 0.30b 0.25b 0.35b 0.20

Uncemented (n = 19) 0.81 0.70 1.00 1.07 1.13

Significantly different than BMD of all TKA (P = 0.0001). Significantly different than BMD of uncemented TKA (P’s ≤ 0.026).

± ± ± ± ±

0.24 0.33 0.36 0.40 0.26

3

Fig. 2. Average BMD zone values for normal knee, all TKA, cemented TKA, and uncemented TKA specimens. *Statistical difference between normal and all TKA specimens in Zone 2 (P = 0.0001). †Statistical difference between cemented and uncemented TKA specimens in Zones 2, 3 and 4 (P’s ≤ 0.026).

Gender

236 235 234 233 232 231 230 229 228 227 226 225 224 223 222 221

Zone by gender interaction was analyzed by examining the BMD estimates at each combination of zone and gender for all TKA specimens, for all normal knee specimens, and for the uncemented TKA specimens (Figs. 3, 4, and 5, respectively). It was seen that in each zone, the BMD for males was higher than for females (Figs. 3 to 5) (P b 0.001). Between genders, the largest difference for all the TKA specimens and the uncemented TKA specimens occurred in the midproximal femur (zone 4. Fig. 1) as seen in the bar graphs (Figs. 3, and 5, respectively), while for the normal specimens it was seen in the antero-distal femur (zone 2. Fig. 1) as seen in the bar graph (Fig. 4). The cemented specimens could not be sub-analyzed due to the small sample size (4 female vs. 1 male).

237

Uncemented Vs. Cemented

249

When comparing uncemented against cemented TKA specimens, it was found that the mean BMD among zones was higher for the cemented specimens (Table 1, Fig. 2); this difference in BMD was significant in the antero-distal, mid-proximal and mid-distal femur (zones 2, 3, and 4; Table 1, Fig. 2) (P = 0.016, 0.010, and 0.026, respectively).

250 251

Table 2 Correlation Coefficients Between BMD Zones for TKA and Normal Knee Specimens.

t2:1

Normal

1 vs. 2 1 vs. 3 1 vs. 4

0.974 0.947 0.941

1 vs. 5 2 vs. 3 2 vs. 4 2 vs. 5

0.908 0.976 (highest) 0.942 0.915

3 vs. 4 3 vs. 5

0.952 0.950

4 vs. 5

0.879 (lowest)

b c

240 241 242 243 244 245 246 247 248

252 253 254 255

t2:2 t2:3

Zones Compared

a

238 239

a

TKA

b

0.810 0.806 0.936 (highest) 0.768 0.744 0.704 0.521 (lowest) 0.865 0.585 0.740

Cemented 0.626 0.653 0.860 0.626 −0.139 (lowest) 0.150 −0.188 0.885c 0.887c 0.940c (highest)

Uncemented

c

0.832 0.813 0.953 (highest) 0.790 0.855

t2:4 t2:5 t2:6

0.793 0.640

t2:9 t2:10

0.867 0.562 (lowest) 0.715

t2:11 t2:12

P’s b 0.001. P’s b 0.01. P’s b 0.05.

Please cite this article as: Lavernia CJ, et al, Bone Mineral Density of the Femur in Autopsy Retrieved Total Knee Arthroplasties, J Arthroplasty (2014), http://dx.doi.org/10.1016/j.arth.2014.03.010

t2:7 t2:8

t2:13 t2:14 t2:15 t2:16

4

C.J. Lavernia et al. / The Journal of Arthroplasty xxx (2014) xxx–xxx

Fig. 3. Average BMD (± SEM values presented over the bars) zone values for all TKA specimens comparing males and females.

256

All TKA Specimens — Mixed Effects Model

257 258

272

In the model used to evaluate the effects of the variables interacting with the periprosthetic BMD measurements, few parameters were found to be influencing BMD at all zones. As previously described those of male gender had a higher BMD than females (1.32 vs. 0.85 g/cm 2, P = 0.0001); also regarding the method of fixation, BMD was higher in the cemented TKA when compared to the uncemented specimens (1.17 vs. 1.01 g/cm 2, P = 0.024). Additionally, a difference in BMD at specific zones was seen depending on the preoperative diagnosis, as mean BMD in OA knees was higher than in RA knees in the posterior femur (zone 5, Fig. 1) (1.25 vs. 1.05 g/cm 2, P = 0.026). Age at surgery, weight, and length of implantation were not correlated with BMD differences. Roentgenographic loose beads were found in eleven knees, however there was no correlation with BMD. The mixed effects model could not be utilized to sub-analyze the cemented specimens because of the small sample size.

273

Uncemented TKA Specimens — Mixed Effects Model

274

When analyzing only the uncemented specimens, again male gender was related to higher BMD than females (1.26 vs. 0.60 g/cm 2, P = 0.0001). Age at surgery, weight, length of implantation and preoperative diagnosis were not correlated with bone density. The presence of loose beads (found in ten patients) was not correlated with BMD either.

259 260 261 262 263 264 265 266 267 268 269 270 271

275 276 277 278 279

Fig. 4. Average BMD (± SEM values presented over the bars) zone values for normal knee specimens comparing males and females.

Fig. 5. Average BMD (± SEM values presented over the bars) zone values for uncemented TKA specimens comparing males and females.

Discussion

280

The effects of stress shielding on bone mass have been previously reported for uncemented and cemented femoral components in knee arthroplasty surgery [22,23]. After TKA the physiological load applied by the patella to the distal femur is mostly carried by the anterior flange of the femoral implant, shielding the bone about this area from any stress. This change in the pattern of strain leads to remodeling of the bone beneath the flange with loss of its mineral density, as it has been reported in clinical studies using DEXA scanning [33–36]. The aim of our study was to determine any difference between the BMD of the distal femur between disease free anatomic knees and knees that underwent arthroplasty and were clinically functional without roentgenographic signs of loosening. Bone density from the twenty-four autopsy retrieved femoral components was measured in five different zones, therefore 120 observations were available for analysis. Since the BMDs in different zones of the same patients were not independent (“r” range, 0.52 to 0.93), ordinary analysis of variance (ANOVA) was not used. Instead, mixed effects model, which assumes BMDs from the different zones were correlated, was used to analyze this dataset. Additionally we estimated the influence of several variables on the BMD of each zone for the TKA specimens. Observations of bone loss in the distal anterior femur after TKA have been reported [30–36]. Cameron and Cameron [30], and Mintzer et al, [31] evaluations reported bone loss; however, their assessment was based on qualitative observations of roentgenographically measurable bone loss, which requires anywhere from 20%–30% BMD change for detection. With a different method, Petersen et al [32] were able to quantify an average decrease of 36% in BMD behind the anterior flange of the femoral prosthesis by using dual photonabsorptiometry. With the development of DEXA, the measurement of BMD in specific areas around metal implants has improved, and we were able to quantify these regional differences as previous authors did [33–36]. However, we attempted to control additional factors to increase the accuracy of our measured BMD. The effects of stress shielding on the reduction of bone mass can be supported by various findings in our study. We observed a difference in the correlation coefficient of the antero-distal and posterior femur (zones 2 vs. 5. Table 2) of the normal specimens as compared to the TKA specimens (r = 0.915 vs. 0.521), which for normal knees yielded an r 2 of 0.84. This means that in this group the posterior femur contributes to 84% of the changes measured in the antero-distal femur (P b 0.001). However, in the TKA specimens changes measured in the posterior femur only reflect 27% of the change in the antero-distal femur (P b 0.01). Also, in the TKA specimens alone, there were regional differences in BMD values,

281 282

Please cite this article as: Lavernia CJ, et al, Bone Mineral Density of the Femur in Autopsy Retrieved Total Knee Arthroplasties, J Arthroplasty (2014), http://dx.doi.org/10.1016/j.arth.2014.03.010

283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324

C.J. Lavernia et al. / The Journal of Arthroplasty xxx (2014) xxx–xxx

325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390

with the anterior femur showing significantly lower BMD than the mid and posterior femur. Moreover, TKA specimens had a significantly lower BMD in the antero-distal femur than normal knees did (0.74 vs. 1.22. P = 0.0001). Thus, we concluded that a large proportion of the lower BMD measured in the antero-distal femur of TKA specimens is due to stress shielding. However, the clinical significance of such a decrease in BMD was not determined. As reported by numerous other authors, after adjusting for other variables, age is a significant determinant of BMD [19]. We found a strong inverse correlation (r = 0.694, r 2 = 0.48) with age and BMD in the normal knee specimens, meaning that as age increased, BMD decreased (P b 0.001). In the TKA group however, there was no correlation between age and bone density. This is probably due to the large amount of stress shielding observed in these long term explants as well. The significance of the effect of other variables on the differences in BMD measurements was also examined. We found that BMD was higher for males than females throughout all zones in both the normal and TKA knees. Gender effects on BMD have been studied in the past and total body bone mineral has been reported to be lower in females than males. Kelly et al [18] compared premenopausal women to an age-matched group of males and found no significant differences in BMD between the genders at the femoral neck; however, females had greater BMD at the lumbar spine. However, following menopause, the effect of bone loss contributes significantly to the gender difference in the incidence of osteoporosis, which, in turn, translates to lower BMD at the most likely age that patients will be receiving a TKA. The effect of length of implantation on BMD has been properly described as after TKA the BMD decreases during the first year [35] and progresses until up to 2 years when it reaches a steady state [32]. Most of our TKA specimens had been implanted for more than 2 years, and the one with the shortest length of implantation in our cohort had almost 1 year (11 months), thus the BMD changes had probably already occurred in the specimens we studied and for this reason we were not able to show any correlation between the length of implantation and the bone density. Our results showed that the cemented TKA had significantly higher BMD than uncemented implants at the antero-distal and mid femur, however its clinical significance was not determined in our study. There are mixed reports regarding the clinical results depending on the method of fixation. Barrack et al in 2004 [47] reported a high failure rate at 2-years in a group of 73 uncemented Low Contact Stress (LCS) mobile-bearing knees (DePuy, Warsaw, IN) with an 8% revision rate of the tibial components when compared to no revisions in a matched-group of 66 cemented TKA using the same design. Further, a meta-analysis comparing the fixation method of knee prosthesis published in 2012, found that uncemented TKAs have a higher chance of failing due to aseptic loosening (odds ratio: 4.2, 95% CI: 2.7 to 6.5) than cemented knees [48]. These clinical results outweigh the potential benefits of the more costly uncemented fixation. However, Abu-Rajab et al in 2006 [33] demonstrated no difference in stressshielding at 2-year follow-up between 20 uncemented and 18 cemented LCS mobile-bearing knees (DePuy, Warsaw, IN). More recently, a series of 471 TKA published in 2013, using the uncemented Active system (Advanced Surgical Design and Manufacture Pty Ltd, Sydney, Australia), reported a survivorship of 94.5% when the endpoint was revision for any reason at 18-year follow-up [49], which is a result similar to recent reports in US joint registries [5,6], including both methods of fixation but with a predilection for the cemented one (N85%–90%) [1]. We did not study the effect of different kinematics within a knee design on BMD; however, Minoda et al [37] showed in cemented femoral components that 32 fixed-bearing TKAs (NexGen LPS-flex; Zimmer, Warsaw, Indiana) had a significantly higher BMD loss in the antero-distal femur at 2-year follow-up, when compared to 28 matched mobile-bearing TKAs (PFC Sigma RP, Depuy, Warsaw, IN), with the

5

latter actually gaining BMD relative to their 2-week postoperative values. Nonetheless, this gain was not statistically significant. The effect of the different pathologies leading to knee arthroplasty is also important to consider. As reported by others, we observed bone density differences between TKA specimens that had OA or RA. Research has suggested a direct relationship between BMD and OA [14,15], though some controversy exists as studies have shown no specific association [16,17]. Rather than higher BMD and OA coexisting, several authors have reported that the higher BMD exists prior to the development of OA [14,15,21]. Specker et al [21] showed a genetic influence in the descendants of Hutterites in eastern South Dakota (an isolated community) that underwent hip and knee arthroplasty due to OA; these descendants had higher BMD than the control group. On the other hand, rheumatoid arthritis patients show atrophic bone changes. These changes include osteoporosis, along with bony erosions, which yield a lower BMD profile [30]. Arthroplasty of the bearing surface with a metal implant subjected all specimens to significant stress shielding in areas adjacent to the prosthesis with certain zones affected more than others. We think that the preoperative DJD that led to knee arthroplasty accounts for the difference seen in our study, as osteoarthritic knees had greater bone density values than rheumatoid arthritis specimens in the posterior condyles of the femur. There are limitations to this study. The BMD measurements in TKA specimens and normal specimens were obtained from different populations. These cohorts had different ages, and different distributions of gender. The ideal situation would have been to have obtained the BMD for normal specimens on the patients prior to TKA or on their contralateral knee. Some of the specimens used in this study were from bilateral arthroplasties, precluding us from a direct comparison. Since our study is cross sectional, it is difficult to conclude from the data that the differences in bone density depended only on the above factors. Only longitudinal measurements of the bone density would be appropriate for studying the causality over bone remodeling. Further, we evaluated BMD changes of the femur with an early TKA model no longer in use, however, the data are still useful in understanding that bone stress shielding occurs over time in specific areas of the distal femur after resurfacing with metal components. Finally, while using the mixed model effects some correlations were found to be statistically significant but as they were too weak they were not even reported, we think that the study might have been underpowered to identify the real magnitude of the correlation of these variables with BMD (i.e., age and length of implantation). There is a direct relationship between BMD and bone strength. After TKA, decreases in bone density may lead to bone “weakening” and will most likely complicate revision surgery. Bone stock at the time of revision arthroplasty is an important factor in the reconstruction of a joint. Understanding periprosthetic bone mineral density may aid in total knee device design. Newer generation TKA designs should show different patterns of bone loss.

391

Acknowledgments

441

392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440

We thank Jesus M. Villa, MD, Marta L. Villarraga, PhD and Roy D. 442 Altman, MD for their assistance during this work. 443 References 1. Mendenhall Associates Inc. 2013 Hip and knee implant review. Orthop Netw News 2013;24(3):1. 2. Cram P, Lu X, Kates SL, et al. Total knee arthroplasty volume, utilization, and outcomes among Medicare beneficiaries, 1991–2010. JAMA 2012;308(12):1227. 3. Khatod M, Inacio M, Paxton EW, et al. Knee replacement: epidemiology, outcomes, and trends in Southern California: 17,080 replacements from 1995 through 2004. Acta Orthop 2008;79(6):812. 4. Weinstein AM, Rome BN, Reichmann WM, et al. Estimating the burden of total knee replacement in the United States. J Bone Joint Surg Am 2013;95(5):385.

Please cite this article as: Lavernia CJ, et al, Bone Mineral Density of the Femur in Autopsy Retrieved Total Knee Arthroplasties, J Arthroplasty (2014), http://dx.doi.org/10.1016/j.arth.2014.03.010

444 445 446 447 448 449 450 451 452 453

6

454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 574

C.J. Lavernia et al. / The Journal of Arthroplasty xxx (2014) xxx–xxx

5. Kurtz S, Ong K, Lau E, et al. Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030. J Bone Joint Surg Am 2007;89(4):780. 6. Paxton EW, Inacio MC, Kiley ML. The Kaiser Permanente implant registries: effect on patient safety, quality improvement, cost effectiveness, and research opportunities. Perm J 2012;16(2):36. 7. Callaghan JJ, O'Rourke MR, Saleh KJ. Why knees fail: lessons learned. J Arthroplasty 2004;19(4 Suppl 1):31. 8. Kurtz SM, Medel FJ, MacDonald DW, et al. Reasons for revision of first-generation highly cross-linked polyethylenes. J Arthroplasty 2010;25(6 Suppl):67. 9. Sharkey PF, Hozack WJ, Rothman RH, et al. Insall Award paper. Why are total knee arthroplasties failing today? Clin Orthop Relat Res 2002(404):7. 10. Sculco T, Martucci E, Lombardi P, Miric A, Sculco T: Mechanical loosening of total knee arthroplasty. p. 171. In Knee arthroplasty. Springer Vienna; 2001: 171. 11. Windsor RE, Scuderi GR, Moran MC, et al. Mechanisms of failure of the femoral and tibial components in total knee arthroplasty. Clin Orthop Relat Res 1989(248):15. 12. Aro HT, Alm JJ, Moritz N, et al. 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(2):107. 13. Nixon M, Taylor G, Sheldon P, et al. Does bone quality predict loosening of cemented total hip replacements? J Bone Joint Surg Br 2007;89(10):1303. 14. Hochberg MC, Lethbridge-Cejku M, Tobin JD. Bone mineral density and osteoarthritis: data from the Baltimore Longitudinal Study of Aging. Osteoarthritis Cartilage 2004;12(Suppl A):S45. 15. Stewart A, Black AJ. Bone mineral density in osteoarthritis. Curr Opin Rheumatol 2000;12(5):464. 16. Arokoski JP, Arokoski MH, Jurvelin JS, et al. Increased bone mineral content and bone size in the femoral neck of men with hip osteoarthritis. Ann Rheum Dis 2002;61(2):145. 17. Madsen OR, Brot C, Petersen MM, et al. Body composition and muscle strength in women scheduled for a knee or hip replacement. A comparative study of two groups of osteoarthritic women. Clin Rheumatol 1997;16(1):39. 18. Kelly PJ, Twomey L, Sambrook PN, et al. Sex differences in peak adult bone mineral density. J Bone Min Res 1990;5(11):1169. 19. Riggs BL, Wahner HW, Dunn WL, et al. Differential changes in bone mineral density of the appendicular and axial skeleton with aging. J Clin Invest 1981;67:328. 20. Urabe K, Mahoney OM, Mabuchi K, et al. Morphologic differences of the distal femur between Caucasian and Japanese women. J Orthop Surg (Hong Kong) 2008;16(3):312. 21. Specker BL, Wey HE, Binkley TL, et al. Higher BMC and areal BMD in children and grandchildren of individuals with hip or knee replacement. Bone 2010; 46(4):1000. 22. Tissakht M, Ahmed AM, Chan KC. Calculated stress-shielding in the distal femur after total knee replacement corresponds to the reported location of bone loss. J Orthop Res 1996;14(5):778. 23. 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 J Surg 1997;79B(1):117. 24. Tonino AJ, Davidson CL, Klopper PJ, et al. Protection from stress in bone and its effects. experiments with stainless steel and plastic plates in dogs. J Bone J Surg 1976;58B:107. 25. Sierra RJ, Cooney WPt, Pagnano MW, et al. Reoperations after 3200 revision TKAs: rates, etiology, and lessons learned. Clin Orthop Relat Res 2004(425):200. 26. Whittaker JP, Dharmarajan R, Toms AD. The management of bone loss in revision total knee replacement. J Bone Joint Surg Br 2008;90(8):981. 27. Lonner JH, Klotz M, Levitz C, et al. Changes in bone density after cemented total knee arthroplasty: influence of stem design. J Arthroplasty 2001;16(1):107. 28. Regner LR, Carlsson LV, Karrholm JN, et al. Bone mineral and migratory patterns in uncemented total knee arthroplasties: a randomized 5-year follow-up study of 38 knees. Acta Orthop Scand 1999;70(6):603.

29. Seki T, Omori G, Koga Y, et al. 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(3):180. 30. Cameron HU, Cameron G. Stress-relief osteoporosis of the anterior femoral condyles in total knee replacement. Orthop Rev 1987;16:449. 31. Mintzer CM, Robertson DD, Rackemann S, et al. Bone loss in the distal anterior femur after total knee arthroplasty. Clin Orthop Relat Res 1990;260:135. 32. Petersen MM, Olsen C, Lauritzen JB, et al. Change in bone mineral density of the distal femur following uncemented total knee arthroplasty. J Arthroplasty 1995;10(1):7. 33. Abu-Rajab RB, Watson WS, Walker B, et al. Peri-prosthetic bone mineral density after total knee arthroplasty. Cemented versus cementless fixation. J Bone Joint Surg Br 2006;88(5):606. 34. Robertson DD, Mintzer CM, Weissman BN, et al. Distal loss of femoral bone following total knee arthroplasty. Measurement with visual computer-processing of roentgenograms and dual-energy x-ray absorptiometry. J Bone J Surg 1994;76A:66. 35. Soininvaara TA, Miettinen HJ, Jurvelin JS, et al. Periprosthetic femoral bone loss after total knee arthroplasty: 1-year follow-up study of 69 patients. Knee 2004;11(4):297. 36. van Loon CJ, Oyen WJ, de Waal Malefijt MC, et al. Distal femoral bone mineral density after total knee arthroplasty: a comparison with general bone mineral density. Arch Orthop Trauma Surg 2001;121(5):282. 37. Minoda Y, Ikebuchi M, Kobayashi A, et al. A cemented mobile-bearing total knee replacement prevents periprosthetic loss of bone mineral density around the femoral component: a matched cohort study. J Bone Joint Surg Br 2010;92(6):794. 38. Orwoll ES, Oviatt SK, Group. NBS. Longitudinal precision of dual-energy x-ray absorptiometry in a multicenter study. J Bone Min Res 1991;6:191. 39. Cohen B, Rushton N. Accuracy of DEXA measurement of bone mineral density after total hip arhtroplasty. J Bone J Surg 1995;77B:479. 40. Bolotin HH. DXA in vivo BMD methodology: an erroneous and misleading research and clinical gauge of bone mineral status, bone fragility, and bone remodelling. Bone 2007;41(1):138. 41. Goh JC, Low SL, Bose K. Effect of femoral rotation on bone mineral density measurements with dual energy X-ray absorptiometry. Calcif Tissue Int 1995;57(5):340. 42. Mortimer ES, Rosenthall L, Paterson I, et al. Effect of rotation on periprosthetic bone mineral measurements in a hip phantom. Clin Orthop Relat Res 1996(324):269. 43. Therbo M, Petersen MM, Schroder HM, et al. The precision and influence of rotation for measurements of bone mineral density of the distal femur following total knee arthroplasty: a methodological study using DEXA. Acta Orthop Scand 2003;74(6):677. 44. Lochmuller EM, Krefting N, Burklein D, et al. Effect of fixation, soft-tissues, and scan projection on bone mineral measurements with dual energy X-ray absorptiometry (DXA). Calcif Tissue Int 2001;68(3):140. 45. Svendsen OL, Hassager C, Skodt V, et al. Impact of soft tissue on in vivo accuracy of bone mineral measurements in the spine, hip, and forearm: a human cadaver study. J Bone Miner Res 1995;10(6):868. 46. Wahnert D, Hoffmeier KL, Lehmann G, et al. Temperature influence on DXA measurements: bone mineral density acquisition in frozen and thawed human femora. BMC Musculoskelet Disord 2009;10:25. 47. Barrack RL, Nakamura SJ, Hopkins SG, et al. Winner of the 2003 James A. Rand Young Investigator's Award. Early failure of cementless mobile-bearing total knee arthroplasty. J Arthroplasty 2004;19(7 Suppl 2):101. 48. Gandhi R, Tsvetkov D, Davey JR, et al. Survival and clinical function of cemented and uncemented prostheses in total knee replacement: a meta-analysis. J Bone Joint Surg Br 2009;91(7):889. 49. Melton JT, Mayahi R, Baxter SE, et al. Long-term outcome in an uncemented, hydroxyapatite-coated total knee replacement: a 15- to 18-year survivorship analysis. J Bone Joint Surg Br 2012;94(8):1067.

Please cite this article as: Lavernia CJ, et al, Bone Mineral Density of the Femur in Autopsy Retrieved Total Knee Arthroplasties, J Arthroplasty (2014), http://dx.doi.org/10.1016/j.arth.2014.03.010

514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573