HIP
The influence of femoral offset on healthrelated quality of life after total hip replacement T. R. Liebs, L. Nasser, W. Herzberg, W. Rüther, J. Hassenpflug From Department of Orthopaedic Surgery, University of Schleswig-Holstein Medical Centre, Kiel, Germany
T. R. Liebs, MD, Consultant Orthopaedic Surgeon L. Nasser, MD, Foundation Doctor J. Hassenpflug, MD, PhD, Professor, Chairman University of SchleswigHolstein Medical Centre, Kiel Campus, Department of Orthopaedic Surgery, Michaelisstr. 1, 24105 Kiel, Germany. W. Herzberg, MD, Chief Orthopaedic Surgeon Asklepios Westklinikum Hamburg, Department of Orthopaedic Surgery, Suurheid 20, 22559 Hamburg, Germany. W. Rüther, MD, PhD, Professor, Chairman Rheumaklinik Bad Bramstedt, Department of Orthopaedic Surgery, Oskar-Alexander-Str. 26, 24576 Bad Bramstedt, Germany. Correspondence should be sent to Dr T. R. Liebs; e-mail: [email protected] ©2014 The British Editorial Society of Bone & Joint Surgery doi:10.1302/0301-620X.96B1. 31530 $2.00 Bone Joint J 2014;96-B:36–42. Received 28 December 2012; Accepted after revision 5 September 2013
36
Several factors have been implicated in unsatisfactory results after total hip replacement (THR). We examined whether femoral offset, as measured on digitised post-operative radiographs, was associated with pain after THR. The routine post-operative radiographs of 362 patients (230 women and 132 men, mean age 70.0 years (35.2 to 90.5)) who received primary unilateral THRs of varying designs were measured after calibration. The femoral offset was calculated using the known dimensions of the implants to control for femoral rotation. Femoral offset was categorised into three groups: normal offset (within 5 mm of the height-adjusted femoral offset), low offset and high offset. We determined the associations to the absolute final score and the improvement in the mean Western Ontario and McMaster Universities osteoarthritis index (WOMAC) pain subscale scores at three, six, 12 and 24 months, adjusting for confounding variables. The amount of femoral offset was associated with the mean WOMAC pain subscale score at all points of follow-up, with the low-offset group reporting less WOMAC pain than the normal or high-offset groups (six months: 7.01 (SD 11.69) vs 12.26 (SD 15.10) vs 13.10 (SD 16.20), p = 0.006; 12 months: 6.55 (SD 11.09) vs 9.73 (SD 13.76) vs 13.46 (SD 18.39), p = 0.010; 24 months: 5.84 (SD 10.23) vs 9.60 (SD 14.43) vs 13.12 (SD 17.43), p = 0.004). When adjusting for confounding variables, including age and gender, the greatest improvement was seen in the low-offset group, with the normal-offset group demonstrating more improvement than the high-offset group. Cite this article: Bone Joint J 2014;96-B:36–42.
Although total hip replacement (THR) is a standard procedure in the treatment of symptomatic osteoarthritis (OA) of the hip,1 not all patients are satisfied with the results, although the clinical and radiological examinations appear normal.2,3 The aim of THR is to restore the presumed pre-pathological centre of rotation, restore leg length and obtain sufficient stability in order to avoid post-operative dislocation.4,5 In order to aid the surgeon in this respect, many manufacturers offer implants with different degrees of femoral head offset. Femoral offset is defined as the distance of the centre of the femoral head from the long axis of the femur5,6 and it is influenced by the design of the implant, the diameter of the head and the valgus/varus positioning of the stem within the femoral canal. Several studies have examined the relationship of femoral offset to post-operative function, such as range of movement,7-9 abductor strength5,8 and polyethylene wear,10 finding that each aspect is improved with a higher degree of offset. However, there is only limited information on the relationship between femoral offset and
post-operative health-related quality of life, especially pain. In a study that retrospectively analysed patients who have received a highoffset femoral stem, 15% of these patients reported pain,11 but there was no control group. Only one evaluation has explored the association between femoral offset with measures of health-related quality of life.6 In that study, the radiographs of 249 patients with a THR were categorised into one of three groups: decreased offset (≤ 5 mm compared with the contralateral hip), normal offset (between -5 mm and +5 mm) and increased offset (≥ 5 mm). The decreased-offset group demonstrated Western Ontario and McMaster Universities osteoarthritis index (WOMAC)12 physical function scores that were lower than those found in the other groups, and concluded that reducing the patients’ natural femoral offset led to inferior functional outcome scores.6 However, they also showed that the decreasedoffset group had the greatest improvement in WOMAC pain scores. We have investigated whether the femoral offset, as measured on routine post-operative THE BONE & JOINT JOURNAL
THE INFLUENCE OF FEMORAL OFFSET ON HEALTH-RELATED QUALITY OF LIFE AFTER TOTAL HIP REPLACEMENT
Fig. 1 Sample anteroposterior radiograph showing the digitised points.
radiographs of patients, is associated with their WOMAC score up to 24 months after THR.
Patients and Methods Our patients underwent unilateral THR for primary OA and had been recruited to multicentre randomised controlled trials evaluating different methods of rehabilitation, which have already been published in part.13,14 The study protocol was approved by the local ethics committee. Exclusion criteria were intra-operative complications, and inability to complete the questionnaires because of cognitive or language difficulties. Implant selection was based on the surgeon’s preference and the final decision on implant size was made during surgery. The participating centres comprised five German hospitals and 704 patients were recruited between 1 January 2003 and 30 April 2006. Postoperatively patients received a standardised program of rehabilitation with daily physiotherapy and analgesics according to a standard scheme. Routine post-operative anteroposterior (AP) radiographs were performed with a source-to-film distance of 1.05 m. The radiological system was calibrated at regular intervals VOL. 96-B, No. 1, JANUARY 2014
37
in order to correct distortion. On a weekly basis a study nurse visited each participating centre with a mobile X-ray scanner (Sierra Medical Film Digitiser; Vidar Systems, Herndon, Virginia) to digitise the radiographs.15 All patients were required to answer a questionnaire at the time of hospital admission, with compliance encouraged by the study nurse. After three, six, 12 and 24 months, participants were sent a questionnaire by post. Nonresponding participants received postal reminders up to three times at intervals of two weeks. Those who still failed to reply were contacted by telephone to determine the reason for not responding. Of the 704 patients recruited, 654 (93%) completed the postal questionnaire at three months. The follow-up rate dropped to 90% (n = 632) at six months’ follow-up, 87.2% (n = 612) at 12 months and 81% (n = 569) at 24 months. There were no distinctive features in the baseline characteristics of either responders or non-responders. Details of these patients have been published.13,14,16 In all, 341 patients were excluded for reasons including no scan of the radiographs (n = 74), defective or poor-quality scans (n = 50), inability to calibrate the scan owing to incomplete operative documentation (n = 199), and because five patients received resurfacing implants and 13 received proximal implants rather than conventional THRs. This left 363 radiographs for analysis, but we had no information on the height of one patient, which reduced the number to 362 for final analysis. In all, 19 different stem and acetabular component combinations were implanted. The primary outcome was self-reported pain which was measured with a validated translated version of the WOMAC scale.12 For the WOMAC items, responses were recorded on a visual analogue scale (VAS) with terminal descriptors. Scores were added for each category and standardised to a score of 0 to100, with higher scores indicating more pain, more stiffness or more dysfunction. We also report the outcomes for WOMAC function and stiffness, and of the Short-Form 36 (SF-36) in a validated, translated version.17 Data analysis. The images of the scanned radiographs were imported into ImageJ Software (National Institutes of Health, Bethesda, Maryland),18 where an author (LN), who was unaware of the clinical outcome, identified predefined anatomical structures (Fig. 1). The x/y coordinates of these structures were exported into a database, where various axes and distances were calculated in pixels. These included the axis of the stem, the axis of the femoral shaft, and the projected femoral and acetabular offsets. Using the known true dimensions of the circular components a calibration factor was derived for each radiograph. This factor was used to convert all pixel distances to millimetres. Projected femoral offset was defined as the perpendicular distance of the femoral head centre from the axis of the femur. In order to adjust for possible malrotation while the radiograph was taken, we compared the known true mediolateral width of the femoral component (from the centre of the
15
38
O O
O O O O O O O OOO O O O O O O O O O O O O O O O O O O O OO O O OO O O O OO O O O O OOOO O O OO O O O O O O O O OO O O O O O O O OO O O OO O O OOO OO OO O O O O O O O O O O O O O O O O O O O OO O OO O O O O O OO O O O O O O O O O O O O OO O OO OO OOO OO OO O O O O O O O O O O O O O O OOOO O O O OOOO O O OO O O O O OO O O O O O O O OO OO O O OO O OOO O OO O O O O O O O O O OO O OO O O O O O O O O O O O O O O OOOO O O OO OO O O O OO O O O O O O O O O O O OO O O O O O O OO OO O O OO O O O O O O O O O O O O O OOO O O O O O OO O O O O O O O O O O O O
5
O
O
+ 2SD = 2.49
25
30
35
40
45
50
55
Mean Fig. 2 Bland–Altman plot for difference between original femoral offset measurement and repeated measurement.
Femoral offset (mm)
0
50
− 2SD = −3.13
−15 −10 −5
Difference
10
Femoral offset (mm) = -6.96 + 0.28 × height O OO R2 = 0.13 O
O O O O O
40 O
OO
O O
O O
O
30 O
O
150
head to the lateral border of the femoral component) with the projected mediolateral width of the femoral component. The ratio of these values was used to calculate an adjusting factor to identify the actual femoral offset, which was subsequently used for analysis. We calculated the association between height and femoral offset across all radiographs, as the femoral neck length is correlated with height in European women19 and in both elderly men and elderly women.20 From this we calculated an expected femoral offset for each height. The difference between the expected and the actual femoral offset was used to define three groups: the normal-offset group (within 5 mm of the height-adjusted femoral offset); a lowoffset group with a femoral offset < 5 mm compared with the height-adjusted femoral offset; and a high-offset group with a femoral offset > 5 mm compared with the heightadjusted femoral offset.6 We calculated the associations between these femoral offset groups in relation to WOMAC pain scores at three, six, 12 and 24 months’ follow-up. Only the scores for the 362 patients whose radiographs have been included were analysed. Since the femoral offset is unknown for the other patients, they were excluded. Statistical analysis. For an analysis of intra-observer reliability, the analysis of femoral offset for 27 radiographs (nine randomly chosen from each group) was repeated. These results are demonstrated in a Bland–Altman plot and Pearson’s correlation coefficient was calculated for the original measurements in comparison to the repeated measurement. Examination of the data for normality of distribution using the Kolmogorov–Smirnov test with Lilliefors significance correction, found only age, height, weight and the WOMAC improvement (change) scores to be normally distributed. Accordingly the non-parametric Jonckheere– Terpstra test21 was applied to test for differences between the three groups.
160
O
O
170
180
190
Height (cm) Fig. 3 Scatter plot of femoral offset by height with the formula for the line of best fit.
In order to adjust for potential confounding variables, multifactorial analysis of variance models were developed in which the improvement of the WOMAC pain score served as the dependent variable and the femoral offset groups served as another independent variable, along with age, weight, gender, number of comorbidities and acetabular offset as confounding variables. The type III sum-ofsquares option was used to calculate the adjusted effect of femoral offset on the improvement in WOMAC pain for each of the four follow-up periods. In order to test for an association between dislocations and offset-group we used the Fisher’s exact test. All p-values were two tailed and no corrections were made for multiple comparisons. Statistical analysis was performed using SPSS (SPSS Inc., Chicago, Illinois) with significance set at 5%.
Results The Bland–Altman plot for the analysis of intra-observer reliability in the measurement of femoral offset is shown in Figure 2. Pearson’s correlation coefficient was 0.988, indicating excellent reproducibility. The overall mean femoral offset was 41.0 mm (SD 6.7; 25.0 to 58.1). Based on the association between body height and femoral offset and using the formula obtained during the regression analysis (femoral offset = 6.96 + 0.28 × height) (Fig. 3) to calculate the difference between measured femoral offset and expected height-adjusted offset, there were 195 patients (53.9%) with a normal offset, 75 (20.7%) with low offset and 92 (25.4%) with a high offset. Aside from age and gender, there were no statistically significant differences between the three groups at baseline (Table I). THE BONE & JOINT JOURNAL
THE INFLUENCE OF FEMORAL OFFSET ON HEALTH-RELATED QUALITY OF LIFE AFTER TOTAL HIP REPLACEMENT
39
Table I. Baseline characteristics of 362 patients for those with low (< 5 mm), normal or high (> 5 mm) height-adjusted offset Femoral offset Characteristic
*
Mean (SD) age (yrs) Female (n, %) Mean (SD) body mass index (kg/m2) Mean (SD) WOMAC score Physical function Pain Stiffness Mean (SD) SF-36 score Physical component summary Mental component summary Comorbidities (n, %) 0 1 ≥2 Additional limitation due to... (n, %) ‡ Contralateral same great joint Ipsilateral adjacent great joint Contralateral adjacent great joint Low back pain Upper limbs Feet
Low offset (n = 75)
Normal offset (n = 195)
High offset (n = 92)
p-value†
67.8 (7.8) 55 (73.3) 27.5 (4.7)
69.3 (8.1) 131 (67.2) 27.1 (4.4)
73.0 (7.2) 44 (47.8) 26.6 (3.2)
< 0.001 0.001 0.356
58.3 (24.4) 55.8 (24.0) 56.9 (27.0)
57.3 (24.0) 55.3 (25.0) 53.8 (29.1)
53.1 (22.6) 54.7 (23.4) 53.3 (26.4)
0.074 0.656 0.432
27.1 (8.7) 48.9 (12.1)
26.9 (7.3) 48.7 (12.4)
27.8 (6.9) 49.1 (11.3)
0.406 0.837 0.793
7 (9.3) 21 (28.0) 47 (62.7)
21 (10.8) 60 (30.8) 114 (58.5)
11 (12.0) 22 (23.9) 59 (64.1)
22 (26.2) 16 (22.9) 8 (16.7) 28 (21.9) 5 (15.2) 16 (41.0)
43 (51.2) 40 (57.1) 22 (45.8) 76 (59.4) 20 (60.6) 12 (30.8)
19 (22.6) 14 (20.0) 18 (37.5) 24 (18.8) 8 (24.2) 11(28.2)
0.288 0.450 0.136 0.071 0.666 0.255
* WOMAC, Western Ontario and McMaster Universities osteoarthritis index; SF-36, Short-Form 36 † Jonckheere–Terpstra test ‡ patients were allowed to mark more than one body region that caused them additional limitations
There was a statistically significant association between femoral offset and the WOMAC pain subscale at six, 12 and 24 months, with the low-offset group reporting less pain than the normal and the high-offset groups (six months: p = 0.006; 12 months: p = 0.010; 24 months: p = 0.004; all Jonckheere–Terpstra test) (Table II, Fig. 4). In order to adjust for confounding variables, which included age, weight, gender, number of comorbidities and acetabular offset, we calculated multivariate models with the improvement of WOMAC pain subscale at all followup periods as the dependent variable. The greatest improvement was seen in the low-offset group, with the high-offset group demonstrating less improvement than the normaloffset group (Table III) (Fig. 5). The mean acetabular offset was 33.9 mm (SD 4.75). There were 13 dislocations, four of which occurred in the low-offset group, seven in the normal group and two in the high-offset group (Fisher’s exact test: p = 0.679). This suggests that a particular offset is not associated with a higher dislocation rate.
Discussion In our series we found less pain reported on the WOMAC pain subscale in the low femoral offset group compared to the normal or high-offset groups at all intervals of followup. In addition, after adjusting for multiple confounding variables, the greatest improvement in the pain subscale was seen in the low-offset group, with the high-offset group having less improvement than the normal-offset group. VOL. 96-B, No. 1, JANUARY 2014
These results must be viewed in the light of several limitations. First, the examined radiographs were routine postoperative exposures and had not been prepared specifically for this analysis. Nevertheless, the results of our intraobserver reliability analysis demonstrate excellent reproducibility and the observer was blinded to the outcome. In addition, adjustment was made for possible malpositioning and overall, the mean femoral offset observed in our study compares well with the mean femoral offset of 42.2 mm reported in an analysis of CT scans.22 Second, the contralateral hip could have been used for reference, but a common finding of OA in the opposite hip23 would have introduced a further confounding variable. Given no pathology in the opposite hip some studies have used the contralateral hip6 on the assumption that both hip joints are likely to have similar morphology.23 However, that assumption may be incorrect, as it has been demonstrated that the offset of a single hip will differ from that of the opposite hip by 4.6 mm.24 The few clinical studies examining the association of femoral offset with measures of function have not analysed the effect of femoral rotation5,6,8,10 and have ignored its influence on the radiological appearance of the proximal femur, resulting under-estimation of femoral offset.22,25-29 For these reasons we preferred to use femoral offset adjusted for height as a more appropriate reference measure, especially as this approach enables us to account for femoral rotation. Finally, it is possible that other factors not measured in our study, such as surgical approach or surgical experience,
40
Table II. Western Ontario and McMaster Universities osteoarthritis index (WOMAC) subscale outcomes and Short-Form (SF)-36 at three, six, 12 and 24 months by height-adjusted femoral offset Mean score (SD; SEM) Score WOMAC Pain 3-month 6-month 12-month 24-month Physical function 3-month 6-month 12-month 24-month Stiffness 3-month 6-month 12-month 24-month SF-36 Physical Component summary 3-month 6-month 12-month 24-month Mental Component Summary 3-month 6-month 12-month 24-month
Low femoral offset (n = 75)
Normal femoral offset (n = 195)
High femoral offset (n = 92)
p-value*
12.69 (17.93; 2.14) 7.01 (11.69; 1.43) 6.55 (11.09; 1.38) 5.84 (10.23; 1.31)
13.91 (15.24; 1.12) 12.26 (15.10; 1.13) 9.73 (13.76; 1.05) 9.60 (14.43; 1.13)
14.00 (13.50; 1.44) 13.10 (16.20; 1.74) 13.46 (18.93; 2.04) 13.12 (17.43; 2.01)
0.089 0.006 0.010 0.004
20.49 (18.71; 2.25) 15.18 (14.91; 1.85) 15.08 (16.56; 2.16) 14.29 (14.37; 1.89)
20.31 (16.67; 1.25) 17.29 (15.75; 1.18) 14.64 (15.23; 1.17) 13.55 (16.12; 1.27)
12.44 (17.37; 1.87) 18.26 (17.87; 1.96) 17.81 (16.69; 2.10) 18.03 (18.70; 2.17)
0.485 0.389 0.327 0.161
26.00 (26.62; 3.18) 19.85 (22.55; 2.78) 19.21 (22.45; 2.83) 17.13 (19.67; 2.52)
25.05 (21.09; 1.55) 23.89 (21.44; 1.60) 18.03 (18.84; 1.44) 16.01 (19.02; 1.49)
27.47 (23.06; 2.47) 22.38 (22.05; 2.38) 20.94 (23.07; 2.58) 20.14 (19.78; 2.30)
0.296 0.410 0.389 0.131
38.75 (9.24; 1.12) 42.83 (9.38; 1.12) 44.46 (9.93; 1.25) 45.96 (10.01; 1.29)
38.62 (8.99; 0.67) 42.57 (9.40; 0.71) 44.19 (9.91; 0.77) 45.71 (8.90; 0.71)
38.10 (9.48; 1.02) 41.77 (9.49; 1.03) 43.16 (10.44; 1.15) 43.04 (10.64; 1.24)
0.579 0.481 0.421 0.114
51.77 (11.49; 1.39) 53.57 (10.02; 1.22) 53.54 (8.91; 1.12) 51.26 (10.83; 1.40)
52.22 (10.54; 0.78) 52.33 (9.41; 0.71) 51.85 (10.04; 0.78) 51.46 (10.11; 0.80)
53.27 (10.47; 1.13) 51.89 (9.87; 1.08) 52.10 (9.63; 1.06) 51.94 (10.44; 1.22)
0.502 0.238 0.556 0.863
* Jonckheere–Terpstra test
0
WOMAC pain
10 20 30
Low femoral offset (< 5.0 mm than mean height-adjusted femoral offset)
40
Normal femoral offset (within 5 mm of mean height-adjusted femoral offset) High femoral offset (> 5.0 mm than mean height-adjusted femoral offset)
50 60 Baseline 3-month 6-month
12-month
24-month
Fig. 4 Mean Western Ontario and McMaster Universities osteoarthritis index (WOMAC) pain subscale at baseline, three, six, 12 and 24 months’ follow-up by femoral offset (low, normal and high). Lower scores represent a higher quality of life. The error bars denote the standard error.
are associated with the health-related quality of life outcomes, and these effects might exhibit an even greaterinfluence than femoral offset, but we are unaware of any such study.30 We accept that our study does not establish a cause and effect relationship.
The overall hip offset consists of both the acetabular and the femoral offsets. The mean acetabular offset in our study was 33.9 mm (SD 4.75). Loughead et al31 identified a mean acetabular offset of 37 mm but used the medial border of the teardrop for reference, whereas we used the most distal THE BONE & JOINT JOURNAL
THE INFLUENCE OF FEMORAL OFFSET ON HEALTH-RELATED QUALITY OF LIFE AFTER TOTAL HIP REPLACEMENT
41
Table III. Adjusted estimates for the effect of femoral offset on improvement of Western Ontario and McMaster Universities osteoarthritis index (WOMAC) pain subscale at three, six, 12 and 24 months. Estimates are adjusted for age, weight, gender, number of comorbidities and acetabular offset (CI, confidence interval) Improvement in WOMAC pain subscale
Femoral offset
Estimated mean (SE; 95% CI) effect of height-adjusted femoral offset
3 months
Low (< 5.0 mm of expected) Normal (± 5.0 mm of expected) High (> 5.0 mm of expected)
44.51 (3.39; 37.83 to 51.18) 40.98 (2.02; 37.02 to 44.95) 43.56 (3.17; 37.33 to 49.79)
6 months
Low (< 5.0 mm of expected) Normal (± 5.0 mm of expected) High (> 5.0mm of expected)
49.43 (3.47; 42.60 to 56.26) 43.11 (2.06; 39.05 to 47.17) 44.34 (3.24; 37.96 to 50.71)
12 months
Low (< 5.0 mm of expected) Normal (± 5.0 mm of expected) High (> 5.0 mm of expected)
49.52 (3.37; 42.89 to 56.16) 45.00 (2.00; 41.05 to 48.94) 43.54 (3.15; 37.35 to 49.74)
24 months
Low (< 5.0 mm of expected) Normal (± 5.0 mm of expected) High (> 5.0 mm of expected)
50.02 (3.44; 43.24 to 56.80) 44.45 (2.05; 40.42 to 48.49) 42.81 (3.22; 36.48 to 49.14)
Improvement in WOMAC pain
60 50 40 30
Low femoral offset (< 5.0 mm than mean height-adjusted femoral offset)
20
Normal femoral offset (within 5 mm of mean height-adjusted femoral offset) High femoral offset (> 5.0 mm than mean height-adjusted femoral offset)
10 0 Baseline 3-month 6-month
12-month
24-month
Fig. 5 Adjusted estimates for mean improvements in Western Ontario and McMaster Universities osteoarthritis index (WOMAC) pain subscale at three, six, 12 and 24 months’ follow-up by femoral offset (low, normal and high), adjusted for age, weight, gender, number of comorbidities and acetabular offset. Higher scores represent a greater improvement in quality of life. The error bars denote the standard error.
point of the teardrop. We used the acetabular offset only as a confounding variable for the multivariate model, as our aim was to examine the association of pain with femoral offset alone; the surgeon has far more influence on the amount of femoral than acetabular offset. As statistically significant differences for age and gender existed in our study groups at baseline, we performed a multivariate analysis in which we adjusted for these and several other factors, such as acetabular offset. In that analysis, the differences between groups lost statistical significance, but the main result of the univariate analysis was unchanged. Although we identified an association between femoral offset and pain, we did not identify an association with VOL. 96-B, No. 1, JANUARY 2014
functional outcome measures. Our results appear to contradict the conclusion drawn by Cassidy et al.6 However, the mean pre-operative WOMAC pain subscale score was different between groups in Cassidy’s study (29.7 points in the decreased-offset group compared with 43.4 in the normal-offset group). At one year post-operatively the mean WOMAC pain subscale scores were not statistically significant between the groups in their study (decreased offset 86.5; normal offset 91.7), which suggests that the decreased-offset group found by Cassidy et al6 obtained greater improvement than the normal-offset group, which would be in keeping with our study. It has been suggested that the longer lever arm in the circumstances of a high-offset could result in increased tension in the abductor muscles
42
or iliotibial band6,11 and a greater bending moment at the bone–implant interface,8 which might account for the greater pain observed in this group. In conclusion, our low-offset group reported less pain than our normal and high-offset groups up to 24 months’ follow-up. It would appear that high offsets in THR are best avoided in order to avoid disappointment for patients through failure to satisfactorily relieve a significant component of their pain. Supplementary material A table detailing the combinations of implants in the study is available alongside the electronic version of this article on our website www.bjj.boneandjoint.org.uk No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. This article was primary edited by G. Scott and first-proof edited by D. Rowley.
References 1. Learmonth ID, Young C, Rorabeck C. The operation of the century: total hip replacement. Lancet 2007;370:1508–1519. 2. Nikolajsen L, Brandsborg B, Lucht U, Jensen TS, Kehlet H. Chronic pain following total hip arthroplasty: a nationwide questionnaire study. Acta Anaesthesiol Scand 2006;50:495–500. 3. Beswick AD, Wylde V, Gooberman-Hill R, Blom A, Dieppe P. What proportion of patients report long-term pain after total hip or knee replacement for osteoarthritis?: a systematic review of prospective studies in unselected patients. BMJ Open 2012;2:000435. 4. Bourne R, Rorabeck CH. Soft tissue balancing: the hip. J Arthroplasty 2002;17(Suppl 1):17–22. 5. Asayama I, Chamnongkich S, Simpson KJ, Kinsey TL, Mahoney OM. Reconstructed hip joint position and abductor muscle strength after total hip arthroplasty. J Arthroplasty 2005;20:414–420. 6. Cassidy KA, Noticewala MS, Macaulay W, Lee JH, Geller JA. Effect of femoral offset on pain and function after total hip arthroplasty. J Arthroplasty 2012;27:1863– 1869. 7. Malik A, Maheshwari A, Dorr LD. Impingement with total hip replacement. J Bone Joint Surg [Am] 2007;89-A:1832–1842. 8. McGrory, Morrey BF, Cahalan TD, An KN, Cabanela ME. Effect of femoral offset on range of motion and abductor muscle strength after total hip arthroplasty. J Bone Joint Surg [Br] 1995;77-B:865–869. 9. Matsushita A, Nakashima Y, Jingushi S, et al. Effects of the femoral offset and the head size on the safe range of motion in total hip arthroplasty. J Arthroplasty 2009;24:646–651. 10. Sakalkale DP, Sharkey PF, Eng K, Hozack WJ, Rothman RH. Effect of femoral component offset on polyethylene wear in total hip arthroplasty. Clin Orthop Relat Res 2001;125–134. 11. Incavo SJ, Havener T, Benson E, et al. Efforts to improve cementless femoral stems in THR: 2- to 5-year follow-up of a high-offset femoral stem with distal stem modification (Secur-Fit Plus). J Arthroplasty 2004;19:61–67.
12. Bellamy N, Buchanan WW, Goldsmith CH, Campbell J, Stitt LW. Validation study of WOMAC: a health status instrument for measuring clinically important patient relevant outcomes to antirheumatic drug therapy in patients with osteoarthritis of the hip or knee. J Rheumatol 1988;15:1833–1840. 13. Liebs TR, Herzberg W, Rüther W, et al. Ergometer cycling after hip or knee replacement surgery: a randomized controlled trial. J Bone Joint Surg [Am] 2010;92A:814–822. 14. Liebs TR, Herzberg W, Rüther W, et al. Multicenter randomized controlled trial comparing early versus late aquatic therapy after total hip or knee arthroplasty. Arch Phys Med Rehabil 2012;93:192–199. 15. Liebs TR, Kloos SA, Herzberg W, Rüther W, Hassenpflug J. The significance of an asymmetric extension gap on routine radiographs after total knee replacement: a new sign and its clinical significance. Bone Joint J 2013;95-B:472–477. 16. Liebs TR, Herzberg W, Gluth J, et al. Using the patient's perspective to develop function short forms specific to total hip and knee replacement based on WOMAC function items. Bone Joint J 2013;95-B:239–243. 17. Bullinger M. German translation and psychometric testing of the SF-36 Health Survey: preliminary results from the IQOLA Project: International Quality of Life Assessment. Soc Sci Med 1995;41:1359–1366. 18. Abramoff MD, Magelhaes PJ, Ram SJ. Image Processing with ImageJ. Biophotonics International 2004;11:36–42. 19. Chin K, Evans MC, Cornish J, Cundy T, Reid IR. Differences in hip axis and femoral neck length in premenopausal women of Polynesian, Asian and European origin. Osteoporos Int 1997;7:344–347. 20. Center JR, Nguyen TV, Pocock NA, et al. Femoral neck axis length, height loss and risk of hip fracture in males and females. Osteoporos Int 1998;8:75–81. 21. Jonckheere AR. A test for significance for the relation between m rankings and k ranked categories. Br J Clin Psychol 1954;93-100. 22. Sariali E, Mouttet A, Pasquier G, Durante E. Three-dimensional hip anatomy in osteoarthritis: analysis of the femoral offset. J Arthroplasty 2009;24:990–997. 23. Hodge WA, Andriacchi TP, Galante JO. A relationship between stem orientation and function following total hip arthroplasty. J Arthroplasty 1991;6:229–235. 24. Krishnan SP, Carrington RW, Mohiyaddin S, Garlick N. Common misconceptions of normal hip joint relations on pelvic radiographs. J Arthroplasty 2006;21:409– 412. 25. Eckrich SG, Noble PC, Tullos HS. Effect of rotation on the radiographic appearance of the femoral canal. J Arthroplasty 1994;9:419–426. 26. Hananouchi T, Sugano N, Nakamura N, et al. Preoperative templating of femoral components on plain X-rays: rotational evaluation with synthetic X-rays on ORTHODOC. Arch Orthop Trauma Surg 2007;127:381–385. 27. Scheerlinck T. Primary hip arthroplasty templating on standard radiographs: a stepwise approach. Acta Orthop Belg 2010;76:432–442. 28. Engh CA. Recent advances in cementless total hip arthroplasty using the AML prosthesis. Tech Orthop 1991;6:59–72. 29. Husmann O, Rubin PJ, Leyvraz PF, de Roguin B, Argenson JN. Three-dimensional morphology of the proximal femur. J Arthroplasty 1997;12:444–450. 30. Ethgen O, Bruyère O, Richy F, Dardennes C, Reginster JY. Health-related quality of life in total hip and total knee arthroplasty: a qualitative and systematic review of the literature. J Bone Joint Surg [Am] 2004;86-A:963–974. 31. Loughead JM, Chesney D, Holland JP, McCaskie AW. Comparison of offset in Birmingham hip resurfacing and hybrid total hip arthroplasty. J Bone Joint Surg [Br] 2005;87-B:163–166.
THE BONE & JOINT JOURNAL
The influence of femoral offset on healthrelated quality of life after total hip replacement T. R. Liebs, L. Nasser, W. Herzberg, W. Rüther, J. Hassenpflug From Department of Orthopaedic Surgery, University of Schleswig-Holstein Medical Centre, Kiel, Germany
T. R. Liebs, MD, Consultant Orthopaedic Surgeon L. Nasser, MD, Foundation Doctor J. Hassenpflug, MD, PhD, Professor, Chairman University of SchleswigHolstein Medical Centre, Kiel Campus, Department of Orthopaedic Surgery, Michaelisstr. 1, 24105 Kiel, Germany. W. Herzberg, MD, Chief Orthopaedic Surgeon Asklepios Westklinikum Hamburg, Department of Orthopaedic Surgery, Suurheid 20, 22559 Hamburg, Germany. W. Rüther, MD, PhD, Professor, Chairman Rheumaklinik Bad Bramstedt, Department of Orthopaedic Surgery, Oskar-Alexander-Str. 26, 24576 Bad Bramstedt, Germany. Correspondence should be sent to Dr T. R. Liebs; e-mail: [email protected] ©2014 The British Editorial Society of Bone & Joint Surgery doi:10.1302/0301-620X.96B1. 31530 $2.00 Bone Joint J 2014;96-B:36–42. Received 28 December 2012; Accepted after revision 5 September 2013
36
Several factors have been implicated in unsatisfactory results after total hip replacement (THR). We examined whether femoral offset, as measured on digitised post-operative radiographs, was associated with pain after THR. The routine post-operative radiographs of 362 patients (230 women and 132 men, mean age 70.0 years (35.2 to 90.5)) who received primary unilateral THRs of varying designs were measured after calibration. The femoral offset was calculated using the known dimensions of the implants to control for femoral rotation. Femoral offset was categorised into three groups: normal offset (within 5 mm of the height-adjusted femoral offset), low offset and high offset. We determined the associations to the absolute final score and the improvement in the mean Western Ontario and McMaster Universities osteoarthritis index (WOMAC) pain subscale scores at three, six, 12 and 24 months, adjusting for confounding variables. The amount of femoral offset was associated with the mean WOMAC pain subscale score at all points of follow-up, with the low-offset group reporting less WOMAC pain than the normal or high-offset groups (six months: 7.01 (SD 11.69) vs 12.26 (SD 15.10) vs 13.10 (SD 16.20), p = 0.006; 12 months: 6.55 (SD 11.09) vs 9.73 (SD 13.76) vs 13.46 (SD 18.39), p = 0.010; 24 months: 5.84 (SD 10.23) vs 9.60 (SD 14.43) vs 13.12 (SD 17.43), p = 0.004). When adjusting for confounding variables, including age and gender, the greatest improvement was seen in the low-offset group, with the normal-offset group demonstrating more improvement than the high-offset group. Cite this article: Bone Joint J 2014;96-B:36–42.
Although total hip replacement (THR) is a standard procedure in the treatment of symptomatic osteoarthritis (OA) of the hip,1 not all patients are satisfied with the results, although the clinical and radiological examinations appear normal.2,3 The aim of THR is to restore the presumed pre-pathological centre of rotation, restore leg length and obtain sufficient stability in order to avoid post-operative dislocation.4,5 In order to aid the surgeon in this respect, many manufacturers offer implants with different degrees of femoral head offset. Femoral offset is defined as the distance of the centre of the femoral head from the long axis of the femur5,6 and it is influenced by the design of the implant, the diameter of the head and the valgus/varus positioning of the stem within the femoral canal. Several studies have examined the relationship of femoral offset to post-operative function, such as range of movement,7-9 abductor strength5,8 and polyethylene wear,10 finding that each aspect is improved with a higher degree of offset. However, there is only limited information on the relationship between femoral offset and
post-operative health-related quality of life, especially pain. In a study that retrospectively analysed patients who have received a highoffset femoral stem, 15% of these patients reported pain,11 but there was no control group. Only one evaluation has explored the association between femoral offset with measures of health-related quality of life.6 In that study, the radiographs of 249 patients with a THR were categorised into one of three groups: decreased offset (≤ 5 mm compared with the contralateral hip), normal offset (between -5 mm and +5 mm) and increased offset (≥ 5 mm). The decreased-offset group demonstrated Western Ontario and McMaster Universities osteoarthritis index (WOMAC)12 physical function scores that were lower than those found in the other groups, and concluded that reducing the patients’ natural femoral offset led to inferior functional outcome scores.6 However, they also showed that the decreasedoffset group had the greatest improvement in WOMAC pain scores. We have investigated whether the femoral offset, as measured on routine post-operative THE BONE & JOINT JOURNAL
THE INFLUENCE OF FEMORAL OFFSET ON HEALTH-RELATED QUALITY OF LIFE AFTER TOTAL HIP REPLACEMENT
Fig. 1 Sample anteroposterior radiograph showing the digitised points.
radiographs of patients, is associated with their WOMAC score up to 24 months after THR.
Patients and Methods Our patients underwent unilateral THR for primary OA and had been recruited to multicentre randomised controlled trials evaluating different methods of rehabilitation, which have already been published in part.13,14 The study protocol was approved by the local ethics committee. Exclusion criteria were intra-operative complications, and inability to complete the questionnaires because of cognitive or language difficulties. Implant selection was based on the surgeon’s preference and the final decision on implant size was made during surgery. The participating centres comprised five German hospitals and 704 patients were recruited between 1 January 2003 and 30 April 2006. Postoperatively patients received a standardised program of rehabilitation with daily physiotherapy and analgesics according to a standard scheme. Routine post-operative anteroposterior (AP) radiographs were performed with a source-to-film distance of 1.05 m. The radiological system was calibrated at regular intervals VOL. 96-B, No. 1, JANUARY 2014
37
in order to correct distortion. On a weekly basis a study nurse visited each participating centre with a mobile X-ray scanner (Sierra Medical Film Digitiser; Vidar Systems, Herndon, Virginia) to digitise the radiographs.15 All patients were required to answer a questionnaire at the time of hospital admission, with compliance encouraged by the study nurse. After three, six, 12 and 24 months, participants were sent a questionnaire by post. Nonresponding participants received postal reminders up to three times at intervals of two weeks. Those who still failed to reply were contacted by telephone to determine the reason for not responding. Of the 704 patients recruited, 654 (93%) completed the postal questionnaire at three months. The follow-up rate dropped to 90% (n = 632) at six months’ follow-up, 87.2% (n = 612) at 12 months and 81% (n = 569) at 24 months. There were no distinctive features in the baseline characteristics of either responders or non-responders. Details of these patients have been published.13,14,16 In all, 341 patients were excluded for reasons including no scan of the radiographs (n = 74), defective or poor-quality scans (n = 50), inability to calibrate the scan owing to incomplete operative documentation (n = 199), and because five patients received resurfacing implants and 13 received proximal implants rather than conventional THRs. This left 363 radiographs for analysis, but we had no information on the height of one patient, which reduced the number to 362 for final analysis. In all, 19 different stem and acetabular component combinations were implanted. The primary outcome was self-reported pain which was measured with a validated translated version of the WOMAC scale.12 For the WOMAC items, responses were recorded on a visual analogue scale (VAS) with terminal descriptors. Scores were added for each category and standardised to a score of 0 to100, with higher scores indicating more pain, more stiffness or more dysfunction. We also report the outcomes for WOMAC function and stiffness, and of the Short-Form 36 (SF-36) in a validated, translated version.17 Data analysis. The images of the scanned radiographs were imported into ImageJ Software (National Institutes of Health, Bethesda, Maryland),18 where an author (LN), who was unaware of the clinical outcome, identified predefined anatomical structures (Fig. 1). The x/y coordinates of these structures were exported into a database, where various axes and distances were calculated in pixels. These included the axis of the stem, the axis of the femoral shaft, and the projected femoral and acetabular offsets. Using the known true dimensions of the circular components a calibration factor was derived for each radiograph. This factor was used to convert all pixel distances to millimetres. Projected femoral offset was defined as the perpendicular distance of the femoral head centre from the axis of the femur. In order to adjust for possible malrotation while the radiograph was taken, we compared the known true mediolateral width of the femoral component (from the centre of the
15
38
O O
O O O O O O O OOO O O O O O O O O O O O O O O O O O O O OO O O OO O O O OO O O O O OOOO O O OO O O O O O O O O OO O O O O O O O OO O O OO O O OOO OO OO O O O O O O O O O O O O O O O O O O O OO O OO O O O O O OO O O O O O O O O O O O O OO O OO OO OOO OO OO O O O O O O O O O O O O O O OOOO O O O OOOO O O OO O O O O OO O O O O O O O OO OO O O OO O OOO O OO O O O O O O O O O OO O OO O O O O O O O O O O O O O O OOOO O O OO OO O O O OO O O O O O O O O O O O OO O O O O O O OO OO O O OO O O O O O O O O O O O O O OOO O O O O O OO O O O O O O O O O O O O
5
O
O
+ 2SD = 2.49
25
30
35
40
45
50
55
Mean Fig. 2 Bland–Altman plot for difference between original femoral offset measurement and repeated measurement.
Femoral offset (mm)
0
50
− 2SD = −3.13
−15 −10 −5
Difference
10
Femoral offset (mm) = -6.96 + 0.28 × height O OO R2 = 0.13 O
O O O O O
40 O
OO
O O
O O
O
30 O
O
150
head to the lateral border of the femoral component) with the projected mediolateral width of the femoral component. The ratio of these values was used to calculate an adjusting factor to identify the actual femoral offset, which was subsequently used for analysis. We calculated the association between height and femoral offset across all radiographs, as the femoral neck length is correlated with height in European women19 and in both elderly men and elderly women.20 From this we calculated an expected femoral offset for each height. The difference between the expected and the actual femoral offset was used to define three groups: the normal-offset group (within 5 mm of the height-adjusted femoral offset); a lowoffset group with a femoral offset < 5 mm compared with the height-adjusted femoral offset; and a high-offset group with a femoral offset > 5 mm compared with the heightadjusted femoral offset.6 We calculated the associations between these femoral offset groups in relation to WOMAC pain scores at three, six, 12 and 24 months’ follow-up. Only the scores for the 362 patients whose radiographs have been included were analysed. Since the femoral offset is unknown for the other patients, they were excluded. Statistical analysis. For an analysis of intra-observer reliability, the analysis of femoral offset for 27 radiographs (nine randomly chosen from each group) was repeated. These results are demonstrated in a Bland–Altman plot and Pearson’s correlation coefficient was calculated for the original measurements in comparison to the repeated measurement. Examination of the data for normality of distribution using the Kolmogorov–Smirnov test with Lilliefors significance correction, found only age, height, weight and the WOMAC improvement (change) scores to be normally distributed. Accordingly the non-parametric Jonckheere– Terpstra test21 was applied to test for differences between the three groups.
160
O
O
170
180
190
Height (cm) Fig. 3 Scatter plot of femoral offset by height with the formula for the line of best fit.
In order to adjust for potential confounding variables, multifactorial analysis of variance models were developed in which the improvement of the WOMAC pain score served as the dependent variable and the femoral offset groups served as another independent variable, along with age, weight, gender, number of comorbidities and acetabular offset as confounding variables. The type III sum-ofsquares option was used to calculate the adjusted effect of femoral offset on the improvement in WOMAC pain for each of the four follow-up periods. In order to test for an association between dislocations and offset-group we used the Fisher’s exact test. All p-values were two tailed and no corrections were made for multiple comparisons. Statistical analysis was performed using SPSS (SPSS Inc., Chicago, Illinois) with significance set at 5%.
Results The Bland–Altman plot for the analysis of intra-observer reliability in the measurement of femoral offset is shown in Figure 2. Pearson’s correlation coefficient was 0.988, indicating excellent reproducibility. The overall mean femoral offset was 41.0 mm (SD 6.7; 25.0 to 58.1). Based on the association between body height and femoral offset and using the formula obtained during the regression analysis (femoral offset = 6.96 + 0.28 × height) (Fig. 3) to calculate the difference between measured femoral offset and expected height-adjusted offset, there were 195 patients (53.9%) with a normal offset, 75 (20.7%) with low offset and 92 (25.4%) with a high offset. Aside from age and gender, there were no statistically significant differences between the three groups at baseline (Table I). THE BONE & JOINT JOURNAL
THE INFLUENCE OF FEMORAL OFFSET ON HEALTH-RELATED QUALITY OF LIFE AFTER TOTAL HIP REPLACEMENT
39
Table I. Baseline characteristics of 362 patients for those with low (< 5 mm), normal or high (> 5 mm) height-adjusted offset Femoral offset Characteristic
*
Mean (SD) age (yrs) Female (n, %) Mean (SD) body mass index (kg/m2) Mean (SD) WOMAC score Physical function Pain Stiffness Mean (SD) SF-36 score Physical component summary Mental component summary Comorbidities (n, %) 0 1 ≥2 Additional limitation due to... (n, %) ‡ Contralateral same great joint Ipsilateral adjacent great joint Contralateral adjacent great joint Low back pain Upper limbs Feet
Low offset (n = 75)
Normal offset (n = 195)
High offset (n = 92)
p-value†
67.8 (7.8) 55 (73.3) 27.5 (4.7)
69.3 (8.1) 131 (67.2) 27.1 (4.4)
73.0 (7.2) 44 (47.8) 26.6 (3.2)
< 0.001 0.001 0.356
58.3 (24.4) 55.8 (24.0) 56.9 (27.0)
57.3 (24.0) 55.3 (25.0) 53.8 (29.1)
53.1 (22.6) 54.7 (23.4) 53.3 (26.4)
0.074 0.656 0.432
27.1 (8.7) 48.9 (12.1)
26.9 (7.3) 48.7 (12.4)
27.8 (6.9) 49.1 (11.3)
0.406 0.837 0.793
7 (9.3) 21 (28.0) 47 (62.7)
21 (10.8) 60 (30.8) 114 (58.5)
11 (12.0) 22 (23.9) 59 (64.1)
22 (26.2) 16 (22.9) 8 (16.7) 28 (21.9) 5 (15.2) 16 (41.0)
43 (51.2) 40 (57.1) 22 (45.8) 76 (59.4) 20 (60.6) 12 (30.8)
19 (22.6) 14 (20.0) 18 (37.5) 24 (18.8) 8 (24.2) 11(28.2)
0.288 0.450 0.136 0.071 0.666 0.255
* WOMAC, Western Ontario and McMaster Universities osteoarthritis index; SF-36, Short-Form 36 † Jonckheere–Terpstra test ‡ patients were allowed to mark more than one body region that caused them additional limitations
There was a statistically significant association between femoral offset and the WOMAC pain subscale at six, 12 and 24 months, with the low-offset group reporting less pain than the normal and the high-offset groups (six months: p = 0.006; 12 months: p = 0.010; 24 months: p = 0.004; all Jonckheere–Terpstra test) (Table II, Fig. 4). In order to adjust for confounding variables, which included age, weight, gender, number of comorbidities and acetabular offset, we calculated multivariate models with the improvement of WOMAC pain subscale at all followup periods as the dependent variable. The greatest improvement was seen in the low-offset group, with the high-offset group demonstrating less improvement than the normaloffset group (Table III) (Fig. 5). The mean acetabular offset was 33.9 mm (SD 4.75). There were 13 dislocations, four of which occurred in the low-offset group, seven in the normal group and two in the high-offset group (Fisher’s exact test: p = 0.679). This suggests that a particular offset is not associated with a higher dislocation rate.
Discussion In our series we found less pain reported on the WOMAC pain subscale in the low femoral offset group compared to the normal or high-offset groups at all intervals of followup. In addition, after adjusting for multiple confounding variables, the greatest improvement in the pain subscale was seen in the low-offset group, with the high-offset group having less improvement than the normal-offset group. VOL. 96-B, No. 1, JANUARY 2014
These results must be viewed in the light of several limitations. First, the examined radiographs were routine postoperative exposures and had not been prepared specifically for this analysis. Nevertheless, the results of our intraobserver reliability analysis demonstrate excellent reproducibility and the observer was blinded to the outcome. In addition, adjustment was made for possible malpositioning and overall, the mean femoral offset observed in our study compares well with the mean femoral offset of 42.2 mm reported in an analysis of CT scans.22 Second, the contralateral hip could have been used for reference, but a common finding of OA in the opposite hip23 would have introduced a further confounding variable. Given no pathology in the opposite hip some studies have used the contralateral hip6 on the assumption that both hip joints are likely to have similar morphology.23 However, that assumption may be incorrect, as it has been demonstrated that the offset of a single hip will differ from that of the opposite hip by 4.6 mm.24 The few clinical studies examining the association of femoral offset with measures of function have not analysed the effect of femoral rotation5,6,8,10 and have ignored its influence on the radiological appearance of the proximal femur, resulting under-estimation of femoral offset.22,25-29 For these reasons we preferred to use femoral offset adjusted for height as a more appropriate reference measure, especially as this approach enables us to account for femoral rotation. Finally, it is possible that other factors not measured in our study, such as surgical approach or surgical experience,
40
Table II. Western Ontario and McMaster Universities osteoarthritis index (WOMAC) subscale outcomes and Short-Form (SF)-36 at three, six, 12 and 24 months by height-adjusted femoral offset Mean score (SD; SEM) Score WOMAC Pain 3-month 6-month 12-month 24-month Physical function 3-month 6-month 12-month 24-month Stiffness 3-month 6-month 12-month 24-month SF-36 Physical Component summary 3-month 6-month 12-month 24-month Mental Component Summary 3-month 6-month 12-month 24-month
Low femoral offset (n = 75)
Normal femoral offset (n = 195)
High femoral offset (n = 92)
p-value*
12.69 (17.93; 2.14) 7.01 (11.69; 1.43) 6.55 (11.09; 1.38) 5.84 (10.23; 1.31)
13.91 (15.24; 1.12) 12.26 (15.10; 1.13) 9.73 (13.76; 1.05) 9.60 (14.43; 1.13)
14.00 (13.50; 1.44) 13.10 (16.20; 1.74) 13.46 (18.93; 2.04) 13.12 (17.43; 2.01)
0.089 0.006 0.010 0.004
20.49 (18.71; 2.25) 15.18 (14.91; 1.85) 15.08 (16.56; 2.16) 14.29 (14.37; 1.89)
20.31 (16.67; 1.25) 17.29 (15.75; 1.18) 14.64 (15.23; 1.17) 13.55 (16.12; 1.27)
12.44 (17.37; 1.87) 18.26 (17.87; 1.96) 17.81 (16.69; 2.10) 18.03 (18.70; 2.17)
0.485 0.389 0.327 0.161
26.00 (26.62; 3.18) 19.85 (22.55; 2.78) 19.21 (22.45; 2.83) 17.13 (19.67; 2.52)
25.05 (21.09; 1.55) 23.89 (21.44; 1.60) 18.03 (18.84; 1.44) 16.01 (19.02; 1.49)
27.47 (23.06; 2.47) 22.38 (22.05; 2.38) 20.94 (23.07; 2.58) 20.14 (19.78; 2.30)
0.296 0.410 0.389 0.131
38.75 (9.24; 1.12) 42.83 (9.38; 1.12) 44.46 (9.93; 1.25) 45.96 (10.01; 1.29)
38.62 (8.99; 0.67) 42.57 (9.40; 0.71) 44.19 (9.91; 0.77) 45.71 (8.90; 0.71)
38.10 (9.48; 1.02) 41.77 (9.49; 1.03) 43.16 (10.44; 1.15) 43.04 (10.64; 1.24)
0.579 0.481 0.421 0.114
51.77 (11.49; 1.39) 53.57 (10.02; 1.22) 53.54 (8.91; 1.12) 51.26 (10.83; 1.40)
52.22 (10.54; 0.78) 52.33 (9.41; 0.71) 51.85 (10.04; 0.78) 51.46 (10.11; 0.80)
53.27 (10.47; 1.13) 51.89 (9.87; 1.08) 52.10 (9.63; 1.06) 51.94 (10.44; 1.22)
0.502 0.238 0.556 0.863
* Jonckheere–Terpstra test
0
WOMAC pain
10 20 30
Low femoral offset (< 5.0 mm than mean height-adjusted femoral offset)
40
Normal femoral offset (within 5 mm of mean height-adjusted femoral offset) High femoral offset (> 5.0 mm than mean height-adjusted femoral offset)
50 60 Baseline 3-month 6-month
12-month
24-month
Fig. 4 Mean Western Ontario and McMaster Universities osteoarthritis index (WOMAC) pain subscale at baseline, three, six, 12 and 24 months’ follow-up by femoral offset (low, normal and high). Lower scores represent a higher quality of life. The error bars denote the standard error.
are associated with the health-related quality of life outcomes, and these effects might exhibit an even greaterinfluence than femoral offset, but we are unaware of any such study.30 We accept that our study does not establish a cause and effect relationship.
The overall hip offset consists of both the acetabular and the femoral offsets. The mean acetabular offset in our study was 33.9 mm (SD 4.75). Loughead et al31 identified a mean acetabular offset of 37 mm but used the medial border of the teardrop for reference, whereas we used the most distal THE BONE & JOINT JOURNAL
THE INFLUENCE OF FEMORAL OFFSET ON HEALTH-RELATED QUALITY OF LIFE AFTER TOTAL HIP REPLACEMENT
41
Table III. Adjusted estimates for the effect of femoral offset on improvement of Western Ontario and McMaster Universities osteoarthritis index (WOMAC) pain subscale at three, six, 12 and 24 months. Estimates are adjusted for age, weight, gender, number of comorbidities and acetabular offset (CI, confidence interval) Improvement in WOMAC pain subscale
Femoral offset
Estimated mean (SE; 95% CI) effect of height-adjusted femoral offset
3 months
Low (< 5.0 mm of expected) Normal (± 5.0 mm of expected) High (> 5.0 mm of expected)
44.51 (3.39; 37.83 to 51.18) 40.98 (2.02; 37.02 to 44.95) 43.56 (3.17; 37.33 to 49.79)
6 months
Low (< 5.0 mm of expected) Normal (± 5.0 mm of expected) High (> 5.0mm of expected)
49.43 (3.47; 42.60 to 56.26) 43.11 (2.06; 39.05 to 47.17) 44.34 (3.24; 37.96 to 50.71)
12 months
Low (< 5.0 mm of expected) Normal (± 5.0 mm of expected) High (> 5.0 mm of expected)
49.52 (3.37; 42.89 to 56.16) 45.00 (2.00; 41.05 to 48.94) 43.54 (3.15; 37.35 to 49.74)
24 months
Low (< 5.0 mm of expected) Normal (± 5.0 mm of expected) High (> 5.0 mm of expected)
50.02 (3.44; 43.24 to 56.80) 44.45 (2.05; 40.42 to 48.49) 42.81 (3.22; 36.48 to 49.14)
Improvement in WOMAC pain
60 50 40 30
Low femoral offset (< 5.0 mm than mean height-adjusted femoral offset)
20
Normal femoral offset (within 5 mm of mean height-adjusted femoral offset) High femoral offset (> 5.0 mm than mean height-adjusted femoral offset)
10 0 Baseline 3-month 6-month
12-month
24-month
Fig. 5 Adjusted estimates for mean improvements in Western Ontario and McMaster Universities osteoarthritis index (WOMAC) pain subscale at three, six, 12 and 24 months’ follow-up by femoral offset (low, normal and high), adjusted for age, weight, gender, number of comorbidities and acetabular offset. Higher scores represent a greater improvement in quality of life. The error bars denote the standard error.
point of the teardrop. We used the acetabular offset only as a confounding variable for the multivariate model, as our aim was to examine the association of pain with femoral offset alone; the surgeon has far more influence on the amount of femoral than acetabular offset. As statistically significant differences for age and gender existed in our study groups at baseline, we performed a multivariate analysis in which we adjusted for these and several other factors, such as acetabular offset. In that analysis, the differences between groups lost statistical significance, but the main result of the univariate analysis was unchanged. Although we identified an association between femoral offset and pain, we did not identify an association with VOL. 96-B, No. 1, JANUARY 2014
functional outcome measures. Our results appear to contradict the conclusion drawn by Cassidy et al.6 However, the mean pre-operative WOMAC pain subscale score was different between groups in Cassidy’s study (29.7 points in the decreased-offset group compared with 43.4 in the normal-offset group). At one year post-operatively the mean WOMAC pain subscale scores were not statistically significant between the groups in their study (decreased offset 86.5; normal offset 91.7), which suggests that the decreased-offset group found by Cassidy et al6 obtained greater improvement than the normal-offset group, which would be in keeping with our study. It has been suggested that the longer lever arm in the circumstances of a high-offset could result in increased tension in the abductor muscles
42
or iliotibial band6,11 and a greater bending moment at the bone–implant interface,8 which might account for the greater pain observed in this group. In conclusion, our low-offset group reported less pain than our normal and high-offset groups up to 24 months’ follow-up. It would appear that high offsets in THR are best avoided in order to avoid disappointment for patients through failure to satisfactorily relieve a significant component of their pain. Supplementary material A table detailing the combinations of implants in the study is available alongside the electronic version of this article on our website www.bjj.boneandjoint.org.uk No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. This article was primary edited by G. Scott and first-proof edited by D. Rowley.
References 1. Learmonth ID, Young C, Rorabeck C. The operation of the century: total hip replacement. Lancet 2007;370:1508–1519. 2. Nikolajsen L, Brandsborg B, Lucht U, Jensen TS, Kehlet H. Chronic pain following total hip arthroplasty: a nationwide questionnaire study. Acta Anaesthesiol Scand 2006;50:495–500. 3. Beswick AD, Wylde V, Gooberman-Hill R, Blom A, Dieppe P. What proportion of patients report long-term pain after total hip or knee replacement for osteoarthritis?: a systematic review of prospective studies in unselected patients. BMJ Open 2012;2:000435. 4. Bourne R, Rorabeck CH. Soft tissue balancing: the hip. J Arthroplasty 2002;17(Suppl 1):17–22. 5. Asayama I, Chamnongkich S, Simpson KJ, Kinsey TL, Mahoney OM. Reconstructed hip joint position and abductor muscle strength after total hip arthroplasty. J Arthroplasty 2005;20:414–420. 6. Cassidy KA, Noticewala MS, Macaulay W, Lee JH, Geller JA. Effect of femoral offset on pain and function after total hip arthroplasty. J Arthroplasty 2012;27:1863– 1869. 7. Malik A, Maheshwari A, Dorr LD. Impingement with total hip replacement. J Bone Joint Surg [Am] 2007;89-A:1832–1842. 8. McGrory, Morrey BF, Cahalan TD, An KN, Cabanela ME. Effect of femoral offset on range of motion and abductor muscle strength after total hip arthroplasty. J Bone Joint Surg [Br] 1995;77-B:865–869. 9. Matsushita A, Nakashima Y, Jingushi S, et al. Effects of the femoral offset and the head size on the safe range of motion in total hip arthroplasty. J Arthroplasty 2009;24:646–651. 10. Sakalkale DP, Sharkey PF, Eng K, Hozack WJ, Rothman RH. Effect of femoral component offset on polyethylene wear in total hip arthroplasty. Clin Orthop Relat Res 2001;125–134. 11. Incavo SJ, Havener T, Benson E, et al. Efforts to improve cementless femoral stems in THR: 2- to 5-year follow-up of a high-offset femoral stem with distal stem modification (Secur-Fit Plus). J Arthroplasty 2004;19:61–67.
12. Bellamy N, Buchanan WW, Goldsmith CH, Campbell J, Stitt LW. Validation study of WOMAC: a health status instrument for measuring clinically important patient relevant outcomes to antirheumatic drug therapy in patients with osteoarthritis of the hip or knee. J Rheumatol 1988;15:1833–1840. 13. Liebs TR, Herzberg W, Rüther W, et al. Ergometer cycling after hip or knee replacement surgery: a randomized controlled trial. J Bone Joint Surg [Am] 2010;92A:814–822. 14. Liebs TR, Herzberg W, Rüther W, et al. Multicenter randomized controlled trial comparing early versus late aquatic therapy after total hip or knee arthroplasty. Arch Phys Med Rehabil 2012;93:192–199. 15. Liebs TR, Kloos SA, Herzberg W, Rüther W, Hassenpflug J. The significance of an asymmetric extension gap on routine radiographs after total knee replacement: a new sign and its clinical significance. Bone Joint J 2013;95-B:472–477. 16. Liebs TR, Herzberg W, Gluth J, et al. Using the patient's perspective to develop function short forms specific to total hip and knee replacement based on WOMAC function items. Bone Joint J 2013;95-B:239–243. 17. Bullinger M. German translation and psychometric testing of the SF-36 Health Survey: preliminary results from the IQOLA Project: International Quality of Life Assessment. Soc Sci Med 1995;41:1359–1366. 18. Abramoff MD, Magelhaes PJ, Ram SJ. Image Processing with ImageJ. Biophotonics International 2004;11:36–42. 19. Chin K, Evans MC, Cornish J, Cundy T, Reid IR. Differences in hip axis and femoral neck length in premenopausal women of Polynesian, Asian and European origin. Osteoporos Int 1997;7:344–347. 20. Center JR, Nguyen TV, Pocock NA, et al. Femoral neck axis length, height loss and risk of hip fracture in males and females. Osteoporos Int 1998;8:75–81. 21. Jonckheere AR. A test for significance for the relation between m rankings and k ranked categories. Br J Clin Psychol 1954;93-100. 22. Sariali E, Mouttet A, Pasquier G, Durante E. Three-dimensional hip anatomy in osteoarthritis: analysis of the femoral offset. J Arthroplasty 2009;24:990–997. 23. Hodge WA, Andriacchi TP, Galante JO. A relationship between stem orientation and function following total hip arthroplasty. J Arthroplasty 1991;6:229–235. 24. Krishnan SP, Carrington RW, Mohiyaddin S, Garlick N. Common misconceptions of normal hip joint relations on pelvic radiographs. J Arthroplasty 2006;21:409– 412. 25. Eckrich SG, Noble PC, Tullos HS. Effect of rotation on the radiographic appearance of the femoral canal. J Arthroplasty 1994;9:419–426. 26. Hananouchi T, Sugano N, Nakamura N, et al. Preoperative templating of femoral components on plain X-rays: rotational evaluation with synthetic X-rays on ORTHODOC. Arch Orthop Trauma Surg 2007;127:381–385. 27. Scheerlinck T. Primary hip arthroplasty templating on standard radiographs: a stepwise approach. Acta Orthop Belg 2010;76:432–442. 28. Engh CA. Recent advances in cementless total hip arthroplasty using the AML prosthesis. Tech Orthop 1991;6:59–72. 29. Husmann O, Rubin PJ, Leyvraz PF, de Roguin B, Argenson JN. Three-dimensional morphology of the proximal femur. J Arthroplasty 1997;12:444–450. 30. Ethgen O, Bruyère O, Richy F, Dardennes C, Reginster JY. Health-related quality of life in total hip and total knee arthroplasty: a qualitative and systematic review of the literature. J Bone Joint Surg [Am] 2004;86-A:963–974. 31. Loughead JM, Chesney D, Holland JP, McCaskie AW. Comparison of offset in Birmingham hip resurfacing and hybrid total hip arthroplasty. J Bone Joint Surg [Br] 2005;87-B:163–166.
THE BONE & JOINT JOURNAL
