Periprosthetic bone mineral density changes after unicondylar knee arthroplasty

The Knee 20 (2013) 120–127

Contents lists available at SciVerse ScienceDirect

The Knee

Periprosthetic bone mineral density changes after unicondylar knee arthroplasty Tarja A. Soininvaara a,⁎, Kristiina A.L. Harju b, Hannu J.A. Miettinen c, Heikki P.J. Kröger a, d a

North Karelia Central Hospital, Joensuu Finland University of Eastern Finland, Medical Faculty, Kuopio, Finland c Department of Orthopaedics and Traumatology and Hand Surgery, Kuopio University Hospital, Kuopio, Finland d Bone and Cartillage Research Unit, University of Eastern Finland; Kuopio, Finland b

a r t i c l e

i n f o

Article history: Received 15 August 2011 Received in revised form 8 October 2012 Accepted 9 October 2012 Keywords: Bone loss Bone mineral density Unicondylar knee arthroplasty Dual-energy X-ray absorptiometry Prospective follow-up

a b s t r a c t Background: Unicompartmental knee arthroplasty (UKA) has received renewed interest in the last decade. UKA involves minor injury to soft tissues, limited removal of bone and delicate preservation of knee anatomy and geometry. In theory, UKA provides an opportunity to restore post-surgical knee kinematics to near normal. Hypothesis: UKA leaves patellofemoral joint free to meet high mechanical forces with no stress-shielding and therefore might preserve bone mineral density (BMD). Patients and methods: We studied 21 patients with osteoarthritis(OA), who had received medial compartment UKA at Kuopio University Hospital between October 1997 and September 2000. BMD was measured by dual-energy X-ray absorptiometry (DEXA), at baseline (within a week after surgery) and at intervals until 7 years. Results: DEXA results were reproducible. The highest rate of periprosthetic bone loss occurred during the first 3 months after UKA. The average loss in BMD was 4.4% (p=0.039) in the femoral diaphysis and it ranged from 11.2% (pb 0.001) to 11.9% (p=0.002) in the distal femoral metaphysis; however, BMD changes in these regions, from 2 years to 7 years, were nonsignificant. At the 1-year follow-up, the BMD of the medial tibial metaphysis had increased by 8.9% (p=0.02), whereas those in the lateral tibial metaphysial (–2.4%) and diaphysial regions (–2.0%) did not change significantly. Interpretations: UKA did not preserve periprosthetic BMD in the distal femoral metaphysis, whereas BMD changes in the tibial metaphysis were minor, consistent with a mechanical balance between the medial and lateral tibial compartments. Level of Evidence 2b: Prospective case control study. © 2012 Elsevier B.V. All rights reserved.

1. Introduction Over the last decade, there has been increased interest in unicompartmental knee arthroplasty (UKA) for medial osteoarthritis (OA). Symptomatic and advanced OA can be treated by total knee arthroplasty (TKA), with or without patellar surfacing. Whereas knee OA can affect any one, or all three compartments of the joint, about 30% of patients have predominantly medial unicompartmental OA [1]. Surgical interventions for medial unicompartmental OA include high tibial osteotomy, UKA or TKA. However, there is a considerable overlap in indications for these three options [1–7]. Traditionally, UKA has been used in older non-obese patients with a sedentary lifestyle. UKA requires an intact anterior cruciate ligament, a good range of motion, minor axial malalignment, which can be corrected to the neutral and with flexion contracture of less than 5° [1,3,8–10]. The increasing popularity of UKA is based on the advantages of replacing only the severely damaged compartment, the preservation of bone stock and the associated fast recovery times. Relative to TKA, ⁎ Corresponding author. Tel.: +358 050 351 8646. E-mail address: Tarja.Soininvaara@uef.fi (T.A. Soininvaara). 0968-0160/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.knee.2012.10.004

UKA involves less damage to soft tissue and an improved preservation of knee anatomy and geometry. In addition, improvements in surgical technique, implant materials and prosthetic design have made UKAs more durable and reliable even resulting in excellent prosthesis performance such that a 98% survivorship can be achieved for more than 10 years [2] Currently the 10-year survivorship is approximately 73%–90% and the revision rate is about twice that of TKAs [4]. Ten-year survivorship data for freely mobile-bearing unicompartmental implants are available from multiple centres, and these show survivorship rates ranging from 85% to 98%, results that are comparable to rates for total knee replacements [11]. While the newer meniscal-bearing prostheses may provide 15 to 20 years of good implant function [9], the biological quality and integrity of the surrounding host bone becomes more important. With these implants, bone removal is limited to the minimum amount required for prosthetic insertion, so that the components largely interface with metabolically active trabecular bone. Essentially, all knee prostheses markedly alter the mechanical loading of both femur and tibia, and this result in an adjustment of the BMD and bone architecture, in order to meet new mechanical demands.

T.A. Soininvaara et al. / The Knee 20 (2013) 120–127

AKS Score Points / Max 100

Several studies have described a decrease of up to 44% in bone mineral density (BMD) adjacent to TKA implants [12–19]. It is a commonly held idea that such loss of BMD is largely due to stress shielding by the implant; however, periods of reduced mobility postoperatively and the reaction of local tissues to surgical trauma to the bone and soft tissue are known to have separate effects on bone turnover [15,19–21]. Dual-energy X-ray absorptiometry (DEXA) can provide highly reproducible measurements of periprosthetic BMD. In addition, dedicated software now allows for the measurement of BMD adjacent to metal implants [22–26]. Not surprisingly, prosthesis design has been reported to be one determinant of BMD changes around implants. For example, mobilebearing implants appear to prevent bone loss around the femoral component after TKA [27], and a porous tantalum tibial component was found to better maintain the periprosthetic BMD, when compared with cemented tibial components [28]. In addition, the loss in BMD, which occurs in the distal femur after knee arthroplasty, is greater with posteriorly stabilized inserts [16]. Since UKA implants are designed to replace only one condyle, they are expected to result in more reduced stress shielding than TKA implants. Theoretically, UKA has the potential to normalize knee kinematics, which in turn may result in a more normal use of the knee. UKA also leaves most of the patellofemoral joint available to meet the mechanical demands on this joint. This prospective follow-up study was designed to determine whether UKA preserves periprosthetic BMD, particularly in the femoral regions.

121

100 90 80 70 60 50 40 30 20 10 0 0 3 months months

1 year

2 years 4 years 7 years

Follow-up period Knee score

Function score

Fig. 1. The mean AKS Knee and Function scores at different periods after surgery for all patients evaluated.

the Ethics Committee and Institutional Review Board of Kuopio University and all patients provided written informed consent. BMD was measured in multiple regions of interest (ROIs) with a fan-beam DEXA (Lunar Expert; Lunar Co., Madison, WA, USA) [22] and the distance between measurements was automatically adjusted with the forearm software (version 1.63) protocol of Lunar Expert XL (Lunar Co.). All femoral and tibial diaphysial ROIs are shown on a lateral scan, and the tibial metaphysial ROIs are shown on an anterior–posterior scan (Figs. 2 and 3, respectively).

2. Materials and methods 2.2. Statistical methods

2.1. Patients The study included 21 patients (eight men and 13 women), with a mean age of 65.2 years (SD, 9.6 years, range 50–91 years). All had been diagnosed with medial compartment OA, and unilateral UKAs were implanted at Kuopio University Hospital between October 1997 and September 2000. Their mean weight was 79.3 kg (SD, 12.0 kg) and their mean body mass index (BMI) was 29.2 (SD, 3.1). The unicondylar prostheses used were seven Duracon unicondylar (Howmedica International Inc., Division of Pfizer Hospital Product Group, Shannon Industrial Estate, Ireland) and 14 Miller-Galante (Zimmer, Warsaw, IN, USA). All prostheses were implanted with bone cement using standard techniques and full weight bearing was allowed immediately after surgery. BMD was measured at intervals by dual-energy X-ray absorptiometry (DEXA), at the baseline (within a week of surgery) and at 3 months, 6 months, 1 year, 2 years, 4 years and 7 years post-operation. The prosthesis-free contralateral knee of each patient was used as a DEXA control, and BMD for both hips was also measured and analysed. Inclusion criteria included a willingness to participate in a 5- to 7-year study, the absence of any bone-related disease or use of any medication known to influence bone mineral metabolism, and women were required to be postmenopausal. Both the Knee and Functional parameters of the American Knee Society Score (AKS) [29] were used to evaluate daily activities in the study subjects. Patient characteristics at baseline and changes in the AKS score are presented in Table 1 and Fig. 1. The study protocol was approved by

Four experienced technicians performed 17 double scans between the femoral ROIs and 14 double scans between the tibial ROIs. The reproducibility of the ROIs was calculated as the root-mean-square average value of the coefficient of variation (CV%). This was calculated by using the following formula derived from Glüer et al. [30]: sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi n X 2 1 di 2n CVð% Þ ¼ n

i¼1

n  X 1 x1;i þ x2;i 2 i¼1

;

where n is number of paired observations and d is the difference between the paired DEXA measurements, x1 and x2. Friedman's test was used to analyse the non-parametric BMD changes of measured ROIs at the seven time intervals for operated knees and five time intervals for control knees and both hips. The results are given as means with standard deviations. Differences at baseline and follow-ups between operated and control hip BMDs and also between medial and lateral tibial metaphyses were tested with paired samples t-test. P-values less than 0.05 were considered significant. For statistical analyses, SPSS software (version 17.0; SPSS Inc., Chicago, IL, USA) was used. 3. Results

Table 1 General characteristics of the patients at baseline.

3.1. Reproducibility of the measurements Value (SD)

Number of patients Male/female Age (years) Weight (kg) Body mass index (kg/cm2)

21 8/13 65 (9.2) 79.3 (12.0) 29.2 (3.1)

Double scans between ROIs were reproducible, with an average CV ranging from 3.8% to 6.7% for femoral metaphysial ROIs and 4.4% for the total femoral metaphysis. Tibial metaphysial CVs were 3.2% for medial ROIs and 4.2% for lateral ROIs, whereas diaphysial measurements had relatively low CVs, being 1.5% for the femur and 2.8% for the tibia. BMD assays in the present study of periprosthetic changes after UKA were performed in locations consistent with previously published work on periprosthetic changes after TKA [21].

122

T.A. Soininvaara et al. / The Knee 20 (2013) 120–127

11

2

2

1

A

B

Fig. 2. The location of ROIs in antero-posterior DEXA scans. A: Operated knee. B: Unoperated control knee. A diaphysial midline was used to divide the medial (ROI 1) and lateral (ROI 2) ROIs.

3.2. Periprosthetic distal femoral bone changes after UKA The highest rate of periprosthetic bone loss was observed in the femur during the first 3 postoperative months after UKA. The average loss in BMD was 4.4% (p=0.039) in the diaphysis and ranged from 11.2% (pb 0.001) to 11.9% (p=0.002) in the metaphysis. At

the 12-month follow-up, the mean loss was 8.03% (p=0.012) in the diaphysis, whereas the highest bone losses were seen in the posterior metaphysis (15.8%) (pb 0.001) and the total metaphysis (15.5%) (pb 0.001). At the 7-year follow-up, bone loss from post-operative baseline was 18.4% in the posterior metaphysis and 17.4% in the total metaphysis (Table 2) (Fig. 4). However, the bone density changes between 2 years and

8

7

8

6

5 10

10 5

6

7

9

9

A

B

Fig. 3. The location of ROIs in lateral projection DEXA scans. A: Operated knee. B: Unoperated control knee. There are both diaphysial (tibial and femoral) and metaphysial (femoral) measurement areas. Diaphysial ROIs (8, 9) were situated at the same distance from the joint line in each patient. In the metaphysial area, the midlongitudinal axis of the femur was used as a boundary between anterior (5) and central ROIs (6). The posterior cortex of the femur indicates the boundary between the central and posterior ROIs (7). The large distal metaphysial ROI enclosed the area of anterior (5), central (6), and posterior (7) ROIs, but was separately measured (ROI 11). The patellar was measured as a separate ROI (10).

T.A. Soininvaara et al. / The Knee 20 (2013) 120–127

123

Table 2 Bone mineral density (g/cm2) (SD) changes after unicompartmental arthroplasty of the operated knees. Region of interest

Tibial medial (1) Tibial lateral (2) Tibial diaphysial (9) Femoral anterior(5) Femoral central (6) Femoral posterior(7) Femoral total (11) Femoral diaphysial (8) Patella (10)

0 months

3 months

6 months

1 year

2 years

4 years

7 years

n = 21

n= 20

n= 19

n= 21

n= 19

n= 15

n = 13

1.15 1.00 1.18 1.26 1.50 1.66 1.47 1.43 1.13

(0.16) (0.16) (0.23) (0.22) (0.32) (0.21) (0.22) (0.19) (0.19)

1.23 1.02 1.16 1.12 1.35 1.47 1.30 1.37 1.00

(0.21) (0.20) (0.23) (0.21) (0.30) (0.23) (0.21) (0.20) (0.18)

1.26 1.01 1.19 1.12 1.33 1.49 1.30 1.37 1.01

(0.22) (0.18) (0.26) (0.18) (0.30) (0.25) (0.22) (0.19) (0.19)

1.25 0.97 1.16 1.05 1.29 1.40 1.24 1.31 1.00

(0.23) (0.20) (0.25) (0.18) (0.29) (0.26) (0.21) (0.20) (0.28)

1.19 0.97 1.17 1.05 1.23 1.50 1.25 1.31 0.99

(0.27) (0.24) (0.25) (0.18) (0.32) (0.29) (0.23) (0.19) (0.18)

1.21 0.94 1.12 1.04 1.20 1.41 1.21 1.26 1.03

(0.28) (0.24) (0.24) (0.20) (0.28) (0.31) (0.24) (0.20) (0.31)

1.20 0.91 1.04 0.99 1.13 1.30 1.14 1.18 0.92

p-value

(0.24) (0.20) (0.21) (0.19) (0.29) (0.22) (0.21) (0.20) (0.14)

0.012 0.496 0.795 0.000 0.000 0.000 0.000 0.002 0.001

P-values by Friedman's test.

7 years were not significant, i.e., for a comparison of BMD in posterior metaphysis (p = 0.178) and in the total metaphysis (p = 0.761). 3.3. Periprosthetic tibial bone changes after UKA The greatest change in periprosthetic BMD in the tibia was a 7.3% increase in the tibial medial metaphysis, and this occurred during the first 3 months after UKA. In addition, there was a minor increase (+0.3%) in the lateral tibial metaphysial compartment and a minor loss (1.7%) in the diaphysis (Table 2) (Fig. 5). At the 1-year follow-up, the increase in the medial metaphysis was 8.9% (p=0.02), whereas in contrast, the BMD changes in the lateral metaphysis (–2.4%) and diaphysis (–2.0%) were not significant (Fig. 5). At 7 years, the tibial medial metaphysial bone gain (7.3% at 3 months) remained significant (p=0.012), whereas the lateral tibial metaphysial and diaphysial changes were not (Table 2; Friedman's test). Further, the medial and lateral metaphysial bone densities were statistically different (pb 0.001) at baseline and at each follow-up period. 3.4. Differences between implants There were seven Duracon and 14 Miller-Galante prostheses implanted. During the study period, one patient with a Duracon prosthesis died, four Duracon UKAs were converted to TKAs, and two of the Miller-Galante prostheses were revised. Regarding BMD changes, the two prosthesis types behaved similarly for the diaphysial regions and for the lateral tibial and femoral metaphysial areas for up to 4 years. There were minor BMD differences in favour of the Duracon prosthesis in anterior femoral and medial tibial areas, but these were not statistically different. 3.5. Bone mineral density changes in the contralateral knee after UKA The mean BMD values of the contralateral knee are presented in Table 3 and BMD behaviour in Figs. 6 and 7. The posterior femoral metaphysis (p = 0.020), total metaphysis (p = 0.002), and lateral tibial metaphysis (p = 0.030) showed significant BMD decreases at the 7-year follow-up.

BMD change in distal femur of operated knees

3.6. Bone mineral density in the proximal femur after UKA The mean postoperative BMD of the proximal femur on the operated side was significantly lower than that of the contralateral side in all ROIs (p b 0.001). This difference remained at 1 year, 2 years and to some degree at 4 years, whereas at 7 years there were no significant differences between the sides. The BMD changes of the ipsilateral proximal femur (hip) were minor during the follow-up, ranging from –1.0% to 5.15%; however, the BMD was significantly reduced in wards (p = 0.039), shaft (p = 0.016) and the total hip (p = 0.014). The BMD change of the contralateral hip varied from +0.33% to –7.08% (p = 0.019-p = 0.001) during follow-up, showing significant changes in all measured ROIs. BMD changes are shown in Tables 4 and 5 and Figs. 8 and 9. 3.7. Patients with incomplete analysis The BMD analysis of some patients was incomplete. One patient died after 2 years from causes unrelated to knee surgery, and one patient moved abroad. Six patients had revision of UKA, one for a loose tibial component at 1 year. Clinical and X-ray based progression of OA was the reason for revisions at 2 years, 3 years and 5 months, 5 years and 5 months, 6 years and 3 months, and 6 years and 6 months. One patient had UKA on the contralateral knee at 1 year. Three patients had TKA of their contralateral knee, one at 2 years, one at 4 years and 6 months, and one at 6 years and 6 months.

4. Discussion In terms of achieving joint stability and near-normal knee kinematics, the outcome of UKAs has generally been described as good to excellent [2,10]. On the downside, UKAs are known to have a slightly higher revision rate than TKAs, although some surgeons describe only minor difficulties with UKA revisions. While some report a diminished bone stock in UKA revisions [3], we are unaware of any publications describing BMD changes after UKA. Because the number of UKAs appears to be increasing, we decided to study BMD changes in the peri-implant areas of two commonly used UKA devices. All patients had medial compartment OA as an indication for UKA. BMD was measured with DEXA, and data were collected from multiple ROIs for each patient at several

0 -2

BMD change in tibia of operated knees

-6 -8

% BMD Change

% BMD change

-4

-10 -12 -14 -16 -18 -20 0 3 6 months months months

1 year

2 years 4 years 7 years

Follow-up period Femoral anterior

Femoral central

Femoral total

12 10 8 6 4 2 0 -2 -4 -6 -8

0 3 6 months months months

Femoral posterior

Femoral diaphysial

Fig. 4. The mean BMD change (% of baseline value) with time in distal femoral ROIs of operated knees.

1 year

2 years 4 years 7 years

Follow-up period Tibial medial

Tibial lateral

Tbial diaphysial

Fig. 5. The mean tibial BMD change (% of baseline value) with time in the ROIs of operated knees.

124

T.A. Soininvaara et al. / The Knee 20 (2013) 120–127

Table 3 Bone mineral density (g/cm2) (SD) changes in the contralateral knees. Region of interest

Tibial medial (1) Tibial lateral (2) Tibial diaphysial (9) Femoral anterior (5) Femoral central (6) Femoral posterior (7) Femoral total (11) Femoral diaphysial (8) Patella (10)

0 months

1 year

2 years

4 years

7 years

n = 21

n = 21

n = 19

n = 15

n=9

1.08 1.08 1.19 1.30 1.48 1.69 1.49 1.46 1.16

(0.14) (0.17) (0.25) (0.22) (0.31) (0.25) (0.24) (0.20) (0.19)

1.07 1.06 1.17 1.28 1.44 1.69 1.48 1.46 1.15

(0.14) (0.17) (0.21) (0.20) (0.29) (0.23) (0.22) (0.20) (0.19)

1.08 1.07 1.21 1.28 1.43 1.73 1.49 1.43 1.15

(0.14) (0.18) (0.22) (0.19) (0.29) (0.23) (0.22) (0.21) (0.16)

1.10 1.06 1.21 1.23 1.39 1.72 1.46 1.42 1.15

(0.19) (0.17) (0.21) (0.20) (0.31) (0.27) (0.23) (0.21) (0.17)

1.03 1.01 1.20 1.16 1.28 1.60 1.37 1.35 1.02

p-value

(0.15) (0.19) (0.21) (0.24) (0.29) (0.32) (0.26) (0.23) (0.29)

0.471 0.031 0.182 0.551 0.291 0.020 0.002 0.145 0.287

P-values by Friedman's test.

defined intervals during the first seven postoperative years. The chosen ROIs were reproducible. The highest femoral periprosthetic bone loss rate was observed during the first 3 months after UKA; however, BMD changes from 2 years to 7 years were not significant. In particular, after UKA there was a significant loss of BMD from distal femoral sites, whereas BMD changes were minor in tibial metaphysial regions, consistent with a mechanical axis balance between the medial and lateral sides of the tibia. Clinical outcome improved during follow-up and AKS score behaved in a predictable way (Fig. 1) and in accordance with results in TKAs [17,27]. Reduction of the BMD did not affect the patients' functional performance. There was a high reproducibility of BMD measurements with all ROIs. Both diaphysial areas had low CVs, being 1.5% for the femoral diaphysis and 2.8% for the tibial diaphysis. We chose these diaphysial ROIs as reference measurement areas because 1) they were easy to define, 2) neither the joint configuration nor the presence of a prosthesis would be expected to affect measurements, and 3) there should be minimal changes in diaphysial BMD due to stress shielding by the implant. Tibial metaphysial CVs ranged from 3.2% on the medial side to 4.2% on the lateral side. A comparison between three relatively small ROIs in the distal femur showed higher CVs (3.8%–6.7%) compared with measurements within the total area (4.4%), which enclosed the area of all three ROIs. These CV values are similar to our data with TKAs and also with previous and recent reproducibility studies [22–26,31]. Differences in BMD between ROIs can result, at least partly, from variation in the proportion of cortical and cancellous bone at different sites, which in turn can be affected by both the precise anatomy of the knee joint and the presence or absence of a prosthesis. For example, with a slight rotational change in the position of a knee

BMD change in distal femur of contralateral knees

during DEXA measurements the prosthesis can shadow either the patella or distal femoral bone and potentially diminish reproducibility between lateral measurements, especially when comparing anterior ROIs (CV = 6.7%). Thus, we found higher reproducibility between posterior ROIs on the distal femur (CV = 4.0%). Considering the relatively low CV values obtained broadly in this study, we feel confident that the results obtained are reliable and can be used for further studies on the effects of UKA on BMD at specific locations around the joint. Because of the implant design, UKA leaves most of the patellofemoral joint available to deal with the mechanical strains in that joint. Because of the relatively low stress shielding expected, we predicted that UKA would preserve bone mass in the distal femur. However, major periprosthetic femoral BMD decreases (up to 15.8%) were seen during the first year after UKA, and after 2 years, the BMD stabilized at the reduced level. These BMD findings are in accordance with previous shortand long-term studies after TKA [12,15,16,32–35] and also with dynamic models simulating the response of the distal femur, as well as finite element analysis of the effects of TKAs [21,36–38]. For example, in a study of 14 patients with patellofemoral arthroplasty for localized OA [18], Van Jonbergen et al. also found a major decrease (–15%) in BMD of the distal femur at 1 year after surgery. However, Minoda et al. reported that a mobile-bearing insert appears to prevent bone loss around the femoral component after TKA [27]. In addition, porous tantalum tibial components maintained better periprosthetic BMD compared with cemented tibial implants [28] and the BMD decrease in the distal femur was found to be greater with posteriorly stabilized inserts [16]. The extent of the BMD losses in the distal femoral metaphysis after UKA, which we have described here, was greater than we expected. Indeed, our data do not support the rationale that UKA should be preferred over TKA because of superior BMD preservation. These findings challenge our current assumptions regarding the origins of distal femoral bone strains and the nature of the effects of stress shielding.

2

BMD change in tibia of contralateral knees

-2

2 -4

% BMD change

% BMD change

0

-6 -8 -10 0 months

1 year

2 years

4 years

7 years

Femoral central

Femoral total

-4 -6 -8 -10 -12 0 months

Follow-up period Femoral anterior

0 -2

1 year

Femoral diaphysial

Fig. 6. The mean BMD change (% of baseline value) with time in distal femoral ROIs of contralateral knees.

2 years

4 years

7 years

Follow-up period

Femoral posterior Tibial medial

Tibial lateral

Tibial diaphysial

Fig. 7. The mean tibial BMD change (% of baseline value) with time in the ROIs of contralateral knees.

T.A. Soininvaara et al. / The Knee 20 (2013) 120–127

125

Table 4 Bone mineral density (g/cm2) (SD) changes in the operated side hips. Region of interest

Neck Wards Troch Shaft Total

0 months

1 year

2 years

4 years

7 years

n = 20

n= 20

n= 18

n= 15

n = 12

0.99 0.76 0.87 1.24 1.05

(0.14) (0.18) (0.13) (0.19) (0.15)

0.96 0.71 0.84 1.20 1.01

(0.14) (0.17) (0.13) (0.18) (0.14)

0.96 0.72 0.85 1.21 1.02

(0.14) (0.14) (0.14) (0.18) (0.15)

0.92 0.67 0.82 1.16 0.98

(0.13) (0.13) (0.14) (0.17) (0.14)

0.92 0.68 0.83 1.15 0.98

p-value

(0.16) (0.17) (0.16) (0.19) (0.17)

0.171 0.039 0.270 0.016 0.014

P-values by Friedman's test.

Table 5 Bone mineral density (g/cm2) (SD) changes in the contralateral side hips. Region of interest

0 months

1 year

n = 20 Neck Wards Troch Shaft Total

1.02 0.77 0.91 1.25 1.07

2 years

n = 20 (0.16) (0.18) (0.16) (0.20) (0.17)

1.00 0.75 0.91 1.25 1.07

4 years

n = 19 (0.16) (0.18) (0.15) (0.18) (0.17)

1.03 0.78 0.92 1.26 1.08

7 years

n = 15 (0.16) (0.18) (0.16) (0.20) (0.17)

0.92 0.69 0.86 1.17 1.01

p-value

n = 12 (0.11) (0.14) (0.16) (0.19) (0.16)

0.96 0.71 0.83 1.16 0.99

(0.17) (0.17) (0.17) (0.20) (0.18)

0.007 0.004 0.004 0.019 0.001

P-values by Friedman's test.

Bone analysis of patients with both mild and severe knee OA [39–42] has shown that the BMD of the medial compartment of the tibia is significantly higher than that of the lateral compartment. Our present results fully support these findings. Thus, we found that the baseline BMD of the ROI on the medial metaphysis was significantly higher than for the lateral side, and also that this difference remained during follow-up after UKA. It should be noted that as a result of UKA, any preoperative varus malalignment is left somewhat under-corrected in terms of the mechanical axis. This suggests that, even after UKA, more of the loading still passes through the medial metaphysical compartment. In fact, this is precisely what was found from our series of BMD analyses following UKA. Thus, the medial side showed significant increases in BMD up to 2 years after surgery, and a minor decrease after 2 years, possibly due to OA-related changes. In contrast, the BMD on the lateral side remained essentially constant over this period. It is interesting that in our patient series, the baseline hip BMD of the operated side was significantly lower than that of the contralateral side. One explanation for this is a pain-limited immobility of the OA limb accompanied by compensatory overuse of the contralateral limb [43–45]. Further, normalization of limb usage by UKA is consistent with our finding (Table 5 and Fig. 9) that the contralateral hip showed significant decreases in BMD for all ROIs during follow-up.

In the contralateral knee, the femoral metaphysial total and posterior ROIs and also the lateral tibial ROIs showed significant BMD decreases at the 7-year follow-up. These results suggest that the metaphysial bone of the femur and tibia at the knee joint has the capacity to respond to factors such as ageing, increased usage and incipient OA (Figs. 6 and 7). Indeed the BMD decrease noted in the lateral tibial region may result from OA progression. The explanation for this is that as the medial joint space decreases it causes the mechanical axis of the knee to move towards the medial compartment and so the lateral compartment bears less of the body weight. Limitations of this study are the small number of enrolled patients and the loss of follow-up data due to the introduction of confounding variables such as revision surgeries. We were not able to evaluate the role of joint alignment in BMD changes. However, with UKA we tend to under-correct the alignment, which means that the changes in mechanical axis angles are minor and most probably would not change the BMD results comparing the medial and lateral tibial metaphyses. At the 7-year follow-up, there remained 13 operated knees and nine control knees to be analysed. However, we were still able to describe BMD changes, which were consistent with those found in earlier studies with TKAs. Despite the progressive loss of controlled patients, there remained 19 UKA knees for evaluation at 2 years, until when the most marked BMD changes usually occur.

BMD change in the contralateral hip

BMD change in ipsilateral hip 2

2

1

% BMD changes

% BMD change

1 0 -1 -2 -3 -4

-2 -3 -4 -5 -6 -7

-5 -6

0 -1

-8 0 0

1 year

2 years

4 years

Follow-up period Neck

Wards

Trochanter

2 years

4 years

7 years

Follow-up period Neck

Shaft

1 year

7 years Wards

Trochanter

Shaft

Total

Total

Fig. 8. The mean BMD change (% of baseline value) with time in the ROIs of ipsilateral hips.

Fig. 9. The mean BMD change (% of baseline value) with time in the ROIs of contralateral hips.

126

T.A. Soininvaara et al. / The Knee 20 (2013) 120–127

The interest in UKA as a treatment option for select cases of medial compartment tibio-femoral OA appears to be increasing [2,46–48]. It has been suggested that the major cause of periprosthetic bone loss is an adaptive stress shielding by the implant. We have shown that following UKA, BMD changes in tibial metaphysial regions are minor and appear to be shifting towards a balance between the medial and lateral sides. However, UKA did not preserve BMD in distal femoral periprosthetic regions, a finding contrary to our study hypothesis. Conflict of interest statement None of the authors have any financial or personal relationships with other people or organizations that could inappropriately influence (bias) their work in this study. Acknowledgements We give special thanks to our study nurses, particularly Eila Koski for her technical help with ROI localization, and Riitta Toroi and Elina Jalava for their help in providing access to patient records and X-ray images. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.knee.2012.10.004. References [1] Griffin T, Rowden N, Morgan D, Atkinson R, Woodruff P, Maddern G. Unicompartmental knee arthroplasty for the treatment of unicompartmental osteoarthritis: a systematic study. ANZ J Surg 2007;77(4):214-21. [2] Berger RA, Meneghini RM, Jacobs JJ, Sheinkop MB, Della Valle CJ, Rosenberg AG, et al. Results of unicompartmental knee arthroplasty at a minimum of ten years of follow-up. J Bone Joint Surg Am 2005;87(5):999–1006. [3] Dettoni F, Bonasia DE, Castoldi F, Bruzzone M, Blonna D, Rossi R. High tibial osteotomy versus unicompartmental knee arthroplasty for medial compartment arthrosis of the knee: a review of the literature. Iowa Orthop J 2010;30:131-40. [4] Hang JR, Stanford TE, Graves SE, Davidson DC, de Steiger RN, Miller LN. Outcome of revision of unicompartmental knee replacement. Acta Orthop 2010;81(1): 95-8. [5] Koskinen E, Eskelinen A, Paavolainen P, Pulkkinen P, Remes V. Comparison of survival and cost-effectiveness between unicondylar arthroplasty and total knee arthroplasty in patients with primary osteoarthritis: a follow-up study of 50,493 knee replacements from the Finnish Arthroplasty Register. Acta Orthop 2008;79(4): 499-507. [6] Lombardi Jr AV, Berend KR, Walter CA, Aziz-Jacobo J, Cheney NA. Is recovery faster for mobile-bearing unicompartmental than total knee arthroplasty? Clin Orthop Relat Res 2009;467(6):1450-7. [7] Takeuchi R, Umemoto Y, Aratake M, Bito H, Saito I, Kumagai K, et al. A mid term comparison of open wedge high tibial osteotomy vs unicompartmental knee arthroplasty for medial compartment osteoarthritis of the knee. J Orthop Surg Res 2010;5(1):65. [8] Collier MB, Eickmann TH, Sukezaki F, McAuley JP, Engh GA. Patient, implant, and alignment factors associated with revision of medial compartment unicondylar arthroplasty. J Arthroplasty 2006;21(6 Suppl. 2):108-15. [9] Keene GCR, Forster MC. Modern unicompartmental knee replacement. Curr Orthop 2005;19(6):428-45. [10] Patil S, Colwell Jr CW, Ezzet KA, D'Lima DD. Can normal knee kinematics be restored with unicompartmental knee replacement? J Bone Joint Surg Am 2005;87(2):332-8. [11] Price AJ, Svard U. A second decade lifetable survival analysis of the Oxford unicompartmental knee arthroplasty. Clin Orthop Relat Res 2011;469(1):174-9. [12] Abu-Rajab RB, Watson WS, Walker B, Roberts J, Gallacher SJ, Meek RM. Peri-prosthetic bone mineral density after total knee arthroplasty. Cemented versus cementless fixation. J Bone Joint Surg Br 2006;88(5):606-13. [13] 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(5–6):324-6. [14] Liu TK, Yang RS, Chieng PU, Shee BW. Periprosthetic bone mineral density of the distal femur after total knee arthroplasty. Int Orthop 1995;19(6):346-51. [15] 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 1996;67(4):339-44. [16] Saari T, Uvehammer J, Carlsson LV, Regner L, Karrholm J. Posterior stabilized component increased femoral bone loss after total knee replacement. 5-year follow-up of 47 knees using dual energy X-ray absorptiometry. Knee 2006;13(6):435-9.

[17] Soininvaara TA, Miettinen HJ, Jurvelin JS, Suomalainen OT, Alhava EM, Kroger HP. Periprosthetic femoral bone loss after total knee arthroplasty: 1-year follow-up study of 69 patients. Knee 2004;11(4):297-302. [18] van Jonbergen HP, Koster K, Labey L, Innocenti B, van Kampen A. Distal femoral bone mineral density decreases following patellofemoral arthroplasty: 1-year follow-up study of 14 patients. BMC Musculoskelet Disord 2010;11:74. [19] 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 2001;121(5):282-5. [20] 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 Relat Res 1990;260:135-43. [21] 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 Joint Surg Br 1997;79(1):117-22. [22] 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 2000;67(3):267-72. [23] Spittlehouse AJ, Getty CJ, Eastell R. Measurement of bone mineral density by dualenergy X-ray absorptiometry around an uncemented knee prosthesis. J Arthroplasty 1999;14(8):957-63. [24] Therbo M, Petersen MM, Schroder HM, Nielsen PK, Zerahn B, Lund B. 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):677-82. [25] Tjornild M, Soballe K, Bender T, Stilling M. Reproducibility of BMD measurements in the prosthetic knee comparing knee-specific software to traditional DXA software. A clinical validation. J Clin Densitom 2011;14(2):138-48. [26] Trevisan C, Bigoni M, Denti M, Marinoni EC, Ortolani S. Bone assessment after total knee arthroplasty by dual-energy X-ray absorptiometry: analysis protocol and reproducibility. Calcif Tissue Int 1998;62(4):359-61. [27] Minoda Y, Ikebuchi M, Kobayashi A, Iwaki H, Inori F, Nakamura H. 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 Jun 2010;92(6):794-8. [28] Minoda Y, Kobayashi A, Iwaki H, Ikebuchi M, Inori F, Takaoka K. Comparison of bone mineral density between porous tantalum and cemented tibial total knee arthroplasty components. J Bone Joint Surg Am 2010;92(3):700-6. [29] Insall JN, Dorr LD, Scott RD, Scott WN. Rationale of the Knee Society clinical rating system. Clin Orthop Relat Res 1989;248:13-4. [30] Gluer CC, Blake G, Lu Y, Blunt BA, Jergas M, Genant HK. Accurate assessment of precision errors: how to measure the reproducibility of bone densitometry techniques. Osteoporos Int 1995;5(4):262-70. [31] Stilling M, Soballe K, Larsen K, Andersen NT, Rahbek O. Knee flexion influences periprosthetic BMD measurement in the tibia. Suggestions for a reproducible clinical scan protocol. Acta Orthop 2010;81(4):463-70. [32] Petersen MM, Jensen NC, Gehrchen PM, Nielsen PK, Nielsen PT. The relation between trabecular bone strength and bone mineral density assessed by dual photon and dual energy X-ray absorptiometry in the proximal tibia. Calcif Tissue Int 1996;59(4): 311-4. [33] Petersen MM, Nielsen PT, Lebech A, Toksvig-Larsen S, Lund B. Preoperative bone mineral density of the proximal tibia and migration of the tibial component after uncemented total knee arthroplasty. J Arthroplasty 1999;14(1):77-81. [34] Petersen MM, Olsen C, Lauritzen JB, Lund B, Hede A. Late changes in bone mineral density of the proximal tibia following total or partial medial meniscectomy. A randomized study. J Orthop Res 1996;14(1):16-21. [35] Saari T, Uvehammer J, Carlsson L, Regner L, Karrholm J. Joint area constraint had no influence on bone loss in proximal tibia 5 years after total knee replacement. J Orthop Res 2007;25(6):798-803. [36] Completo A, Fonseca F, Simoes JA. Experimental validation of intact and implanted distal femur finite element models. J Biomech 2007;40(11):2467-76. [37] Completo A, Fonseca F, Simoes JA. Strain shielding in proximal tibia of stemmed knee prosthesis: experimental study. J Biomech 2008;41(3):560-6. [38] 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):778-85. [39] Finlay JB, Bourne RB, Kraemer WJ, Moroz TK, Rorabeck CH. Stiffness of bone underlying the tibial plateaus of osteoarthritic and normal knees. Clin Orthop Relat Res 1989;247: 193-201. [40] Hulet C, Sabatier JP, Souquet D, Locker B, Marcelli C, Vielpeau C. Distribution of bone mineral density at the proximal tibia in knee osteoarthritis. Calcif Tissue Int 2002;71(4):315-22. [41] Madsen OR, Schaadt O, Bliddal H, Egsmose C, Sylvest J. Bone mineral distribution of the proximal tibia in gonarthrosis assessed in vivo by photon absorption. Osteoarthritis Cartilage 1994;2(2):141-7. [42] Wada M, Maezawa Y, Baba H, Shimada S, Sasaki S, Nose Y. Relationships among bone mineral densities, static alignment and dynamic load in patients with medial compartment knee osteoarthritis. Rheumatology (Oxford) 2001;40(5): 499-505. [43] Zhang Y, Hannan MT, Chaisson CE, McAlindon TE, Evans SR, Aliabadi P, et al. Bone mineral density and risk of incident and progressive radiographic knee osteoarthritis in women: the Framingham Study. J Rheumatol 2000;27(4):1032-7. [44] Bruyere O, Dardenne C, Lejeune E, Zegels B, Pahaut A, Richy F, et al. Subchondral tibial bone mineral density predicts future joint space narrowing at the medial femoro-tibial compartment in patients with knee osteoarthritis. Bone 2003;32(5): 541-5.

T.A. Soininvaara et al. / The Knee 20 (2013) 120–127 [45] Hart DJ, Cronin C, Daniels M, Worthy T, Doyle DV, Spector TD. The relationship of bone density and fracture to incident and progressive radiographic osteoarthritis of the knee: the Chingford Study. Arthritis Rheum 2002;46(1):92-9. [46] Jeer PJ, Keene GC, Gill P. Unicompartmental knee arthroplasty: an intermediate report of survivorship after the introduction of a new system with analysis of failures. Knee 2004;11(5):369-74.

127

[47] Lastad Lygre SH, Espehaug B, Havelin LI, Furnes O, Vollset SE. Pain and function in patients after primary unicompartmental and total knee arthroplasty. J Bone Joint Surg Am 2010;92(18):2890-7. [48] W-Dahl A, Robertsson O, Lidgren L, Miller L, Davidson D, Graves S. Unicompartmental knee arthroplasty in patients aged less than 65. Acta Orthop 2010;81(1):90-4.