The Knee 21 (2014) 257–263
Contents lists available at ScienceDirect
Does ﬂexion of the femoral implant in total knee arthroplasty increase knee ﬂexion: A randomised controlled trial Michael Murphy a,⁎, Simon Journeaux b, Julie Hides c, Trevor Russell a a b c
Mater Medical Research Institute, Mater Health Services, Raymond Terrace, South Brisbane, Australia, 4101 School of Health and Rehabilitation Science, University of Queensland, St Lucia, Australia, 4076 Faculty of Health Sciences, Australian Catholic University, Banyo, Australia, 4014
a r t i c l e
i n f o
Article history: Received 4 April 2012 Received in revised form 15 October 2012 Accepted 28 October 2012 Keywords: Total knee arthroplasty Implant ﬂexion Knee ﬂexion Randomised controlled trial
a b s t r a c t Introduction: Prosthetic and operative modiﬁcations in total knee arthroplasty (TKA) have been proposed to maximise post-operative knee ﬂexion as it is essential in routine functional activities. Methods: We performed a double blind randomised controlled trial to compare clinical outcomes of primary cruciate-retaining TKA for osteoarthritis with the femoral component implanted in either 4° ﬂexion in the sagittal plane (F) or in a neutral position (C). The primary outcome of knee ﬂexion and secondary outcomes knee extension, quadriceps strength, WOMAC, SF-12v2, timed stand test, stair climb test and satisfaction were assessed at 1 year. Knee ﬂexion and extension were also assessed intra-operatively. Implant ﬂexion was measured from true lateral radiographs. Results: Thirty-nine participants (40 knees) were recruited, 20 knees per group. Three subjects from the control group and two from the ﬂexed group were lost to 1 year follow-up but numbers were sufﬁcient to satisfy the sample size calculation. Signiﬁcant differences were found between the groups in knee ﬂexion (F: 113.6±8.8° pre-operative, 122.4±6.0° intra-operative, 110.2±7.5° 1 year, C: 117.4±11.7°, 117.4±7.6°, 103.5±10.7°. p =0.031) and mental component score of the SF12-v2 (F 53.3±13.2, C 61.1±7.3, p =0.009) but there were no signiﬁcant differences in other outcomes and patients were equally satisﬁed. Conclusion: Flexing the femoral implant in this cruciate retaining TKA system provided a signiﬁcant difference in knee ﬂexion compared to a neutral position. The improvement appears to occur predominantly at surgery and was not associated with a clinical or functional beneﬁt at 1 year. (ACTRN12606000325505). Level of evidence: Level 1; randomised controlled trial. © 2012 Elsevier B.V. All rights reserved.
1. Introduction Knee ﬂexion is traditionally one of the key outcomes in determining success of total knee arthroplasty (TKA) as it is critical to many routine daily activities. Up to 93° of knee ﬂexion is required to rise from a chair, 117° to negotiate stairs or pick up an object off the ﬂoor, and 135° to get in and out of a bath [1,2]. Additionally, non-Western populations typically pursue activities such as squatting or cross-legged sitting for cultural or religious purposes which require knee ﬂexion up to 165° . Following conventional TKA however patients seldom exceed 110 to 115° [4–6]. Many variables affect knee ﬂexion after TKA including patient factors such as gender, body mass index, age and pre-operative ﬂexion [5–7]. Surgically modiﬁable variables such as implant sizing, ligament balancing, and osteophyte removal have also been well documented as important in optimising post-operative ﬂexion . Attention has turned recently to implant design with the emergence of “high-ﬂex” prostheses but there is to date little consensus on their efﬁcacy  ⁎ Corresponding author. Tel.: +61 7 3163 8787; fax: +61 7 3163 1671. E-mail address: [email protected]
(M. Murphy). 0968-0160/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.knee.2012.10.028
and some concern with early loosening [10,11]. Moreover many studies investigating these designs which have demonstrated improvements in ﬂexion were unable to show associated functional beneﬁts [12,13]. Posterior condylar offset (PCO), ﬁrst deﬁned by Bellemans et al.  as the maximal sagittal plane thickness of the posterior femoral condyle relative to the posterior femoral cortex, is another surgical variable which has been reported to affect knee ﬂexion. Thickened posterior femoral condyles are a common design feature in contemporary “high-ﬂex” knee arthroplasty implants. Computer modelling , radiographic templating studies  and ﬂuoroscopic studies  have shown PCO to be correlated with knee ﬂexion, though there are some reports that it may only be helpful in offsetting the paradoxical anterior femoral translation reported in cruciate-retaining implants [17,18]. PCO can also be increased by ﬂexing the femoral component in the sagittal plane . Flexion of the femoral implant is an option with some knee systems to address ﬂexion instability during surgery which can be caused by an inadequate restoration of PCO. Without this feature, surgeons may need to use a larger femoral component potentially leading to “overstufﬁng” of the patello-femoral joint and quadriceps mechanism tensioning, or medio-lateral overhang which may irritate soft tissue . Only one previous trial was identiﬁed
M. Murphy et al. / The Knee 21 (2014) 257–263
which has attempted to investigate the impact of implant position on post-operative ROM. In this retrospective study, Faris et al.  reviewed the sagittal plane orientation of 623 cruciate-retaining knee replacements and found no correlation between implant position and knee ROM. We are not aware of any studies evaluating the effect of electively ﬂexing the femoral implant. This study is a prospective randomised controlled trial of the clinical outcomes following cruciate-retaining TKA with a 4 degree ﬂexed femoral component compared with a neutral resection. We hypothesised that subjects undergoing TKA with a ﬂexed femoral implant would achieve signiﬁcantly greater knee ﬂexion compared to those with a neutral femoral implant.
The study was granted approval by the institutional human ethical review board and was registered in the Australian New Zealand Clinical Trials Registry (trial number ACTRN12606000325505). All participants provided written informed consent.
(0 degree or neutral resection). Randomisation was performed by an independent research support service attached to the institution using a computer generated randomisation code (balanced block design of size 4) and sealed in opaque envelopes. The surgeon sequentially selected an envelope to reveal the grouping immediately prior to surgery. Participants and all other investigators were blinded to group allocation. All aspects of the surgery other than the intervention followed an identical procedure with a medial parapatellar approach, standard ligament balancing techniques and osteophyte removal. The patella was not resurfaced and the tibial slope resection was 0° for both groups. If any deviation from the standardised surgical procedure was required the participant was excluded. Reinfusion wound drains were routinely used and removed within 24 h of surgery. The post-operative rehabilitation protocol for both groups was identical, following a standardised clinical pathway. Mobilisation without weight restriction, active and passive knee ﬂexion and extension exercises and quadriceps strengthening commenced on post-operative day 1. Patients were discharged directly home when they were independently ambulating and were provided with a home exercise programme.
Participants were recruited from a single surgeon's outpatient clinic by the surgeon or principal investigator between June 2006 and May 2009. Participants were eligible if they were scheduled for primary TKA with a diagnosis of osteoarthritis and were available for follow-up at 1 year but were excluded if they had undergone lower limb arthroplasty in the previous 12 months or were scheduled for bilateral TKA.
The primary clinical outcome of the study was knee ﬂexion. Secondary outcomes included, knee extension, quadriceps strength, Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) , the Short Form Health Survey (SF-12v2) , physical function tests (stair climb test , timed stands test ) and patient satisfaction. Radiographic measurement of the femoral implant ﬂexion in the sagittal plane was essential to verify two distinct groups were available for analysis. Clinical evaluations were conducted pre-operatively and at 1 year. Knee ﬂexion and extension was also measured at the completion of surgery whilst the patient was still anaesthetised. A ﬂow diagram for the study, in accordance with CONSORT guidelines  is presented in Fig. 2. Knee ﬂexion, extension and quadriceps lag were assessed with digital photographs using a 2-dimensional kinematic software analysis application adapted from telerehabilitation research . This method has demonstrated validity (limits of agreement with universal goniometer measurements of −1.66 to 1.76°), intra-tester (ICC= 0.97 to > 0.99) and inter-tester reliability (ICC = 0.97 to> 0.99) . Photographs were taken with the subject lying supine with the lens of the camera visually aligned with the horizontal axis of the knee. The measurements were performed by a blinded assessor. Knee ﬂexion was
2. Materials and methods
2.2. Procedure All subjects underwent TKA with the Proﬁx Total Knee System (Smith & Nephew, Memphis, TN) cruciate retaining implant. This device shares features common in “high-ﬂex” designs such as a smaller radius of curvature in shorter and thickened posterior femoral condyles which aim to minimise edge loading and improve articular contact with deeper ﬂexion. The instrumentation with this system allows the anterior femoral resection to be made at an angle of 0 or 4° ﬂexion relative to the intramedullary axis in the sagittal plane. Flexion of the femoral component increases the PCO compared to a neutral resection (Fig. 1). Subjects were randomised on the day of surgery to ﬂexed (4 degree ﬂexed femoral component) or control
Fig. 1. Diagram representing the lateral view of the control (left) and ﬂexed (right) groups. The posterior condylar offset is larger with the ﬂexed group's 4° resection (P) than the control's 0° resection (p) accommodating greater knee ﬂexion before tibial component impingement on the posterior cortex of the femur (Adapted from Bellemans, Banks et al. ).
M. Murphy et al. / The Knee 21 (2014) 257–263
Fig. 2. Flow diagram of participants in accordance to the CONSORT guidelines .
assessed with the hip ﬂexed to 90° and the knee ﬂexed passively as described by Lee et al. , and knee extension with the ankle supported to avoid calf contact with the examination table. Knee extension was recorded as a ﬁxed ﬂexion deformity (FFD) as this positive value is more commonly observed. Hyperextension, if observed, was recorded with a negative value. At the pre and post-operative evaluation photographs of a straight leg raise taken immediately after the leg lifted from the ankle support were taken to ascertain “quadriceps lag” or loss of terminal extension. Quadriceps lag was deﬁned as the difference between the subject's passive knee extension and that when attempting a straight leg raise. In addition to the quadriceps lag test, maximal isometric quadriceps strength was assessed at 90° knee ﬂexion with a hand-held dynamometer (Chatillon, Ametek Inc. USA). The dynamometer was placed immediately proximal to the ankle joint line and after a submaximal practice and a 30 second rest, subjects were asked to extend their knee with maximal effort for 3 s. This was repeated and the higher of the two attempts recorded. WOMAC pain, stiffness and function components normalised to scores of 100  and SF-12v2 summary scales (the normalised physical component summary (PCS) and mental component summary (MCS))  were calculated. Higher WOMAC scores indicate worse outcomes, while higher SF-12v2 scores indicate better health status. The physical functional tests were selected as they incorporate functional knee ﬂexion. The stair climb test (SCT) has demonstrated reliability in lower limb osteoarthritis and TKA populations [23,28] and the timed stand test (TST) is a simple, reliable and valid test of lower limb function . In the SCT the subjects were timed ascending, turning and descending a set of ﬁve standard steps (step height 15 cm) using reciprocal stepping. In the TST subjects were timed as they rose ﬁve times from a chair, which for this study was normalised to a height 5 cm below the subject's popliteal crease. Participants
were asked to rate their satisfaction with their knee arthroplasty in the domains of pain, physical function and overall, on a scale of 0 (completely unsatisﬁed) to 10 (completely satisﬁed) at 1 year. The angle of ﬂexion of the femoral component in the sagittal plane was measured from true lateral radiographs using OrthoView™ orthopaedic digital imaging software (Meridian Technique Limited, Hampshire, UK) and is described in Fig. 3. The radiographic measurement involved a similar methodology to that used by Minoda et al.  where a line tangential to a pre-deﬁned point on the anterior cortex of the femur was used as the reference for the implant ﬂexion. The point, 80 mm proximal to the most distal aspect of the femoral component, was selected as the reference point as this exceeded the 70 mm length of the largest femoral component used in the study. 2.4. Statistical methods A power analysis indicated that a sample size of 17 participants per group would provide 80% statistical power for detecting a 10 degree difference in knee ﬂexion between the groups. This assumes a p value of b0.05, and a standard deviation of 10° as reported in previous TKA literature [4,30]. In the absence of any published clinically important difference in knee ﬂexion following TKA the authors believed a 10 degree improvement would represent a meaningful clinical improvement. Allowing for a 15% loss to follow-up, 20 knees were required in each group. A total of 49 patients (50 knees) were approached and 39 participants (40 knees) were subsequently randomised. One participant was included with a TKA on the other knee 21 months after their ﬁrst. Baseline characteristics of the groups were compared using general linear model univariate analysis of variance for continuous data. A linear mixed model statistic was used to compare groups on both primary and secondary outcomes. The statistic was computed with
M. Murphy et al. / The Knee 21 (2014) 257–263 point is presented graphically in Fig. 4. Univariate analysis of variance of patient satisfaction scores revealed no signiﬁcant differences in any domain; overall satisfaction (control 8.8±1.5, ﬂexed 9.1±1.4, p=0.644), satisfaction with reduction in pain (control 8.8±2.0, ﬂexed 9.1±1.0, p=0.517) or satisfaction with physical activity (control 8.3±1.2, ﬂexed 8.3±1.6, p=0.934). Superﬁcial wound infections in one subject from each group required intravenous antibiotics and in the ﬂexed group one subject developed a deep venous thrombosis. All cases resolved without further complication. Three cases of anterior femoral notching occurred; two in the control group and one in the ﬂexed group. Only the subject from the ﬂexed group was excluded from the study as they required revision to a stemmed prosthesis with neutral sagittal orientation. Baseline data for this subject was used in analysis. Data from three subjects from the control group and two from the ﬂexed group was not available for analysis at one year (Fig. 2).
Fig. 3. A true lateral radiograph demonstrating measurement of the angle of femoral implant ﬂexion in the sagittal plane using OrthoView orthopaedic digital imaging software (Meridian Technique Limited, Hampshire, UK). A 25 mm spherical marker (A) was used to minimise magniﬁcation error in measuring the reference point on the anterior cortex (B). A line (CD) tangential to B was moved to the most proximal point of the implant (E) (the software allows movement of lines in parallel). The femoral component angle is displayed as the angle between the line DE and a line EF along the inner surface of the femoral ﬂange.
observed outcomes as the dependent variables and with ﬁxed factors of treatment group (control and ﬂexed) and assessment point (pre-operative, intra-operative and one year). The outcome of primary interest was the interaction between group and time. Fixed predicted values and residuals from these analyses were used for data inspection purposes. Outliers, deﬁned as data points greater than three standard deviations from the mean, were removed for the analyses. Femoral component angle and participant satisfaction were compared descriptively and with a general linear model univariate analysis of variance. A p value of b 0.05 was accepted as signiﬁcant and the calculations were performed using SPSS version 15.0 (SPSS Inc., Chicago, Illinois).
This randomised controlled trial compared the outcomes of cruciate-retaining TKA with a femoral implant ﬂexed 4° in the sagittal plane to those with a neutral femoral implant. Signiﬁcant differences between the groups were found in knee ﬂexion and the mental component score of the SF12v2 survey but not in any other clinical or functional outcome. The cause of the difference in the mental component scores, with the ﬂexed group appearing to fare worse, is not clear and not particularly concerning, as both groups exceed the norm-based mean of 50 (SD = 10) . Inspection of the data suggests the difference in ﬂexion may be due to a combination of an improvement in the ﬂexed group at the intra-operative evaluation and an overall loss of ﬂexion in the control group. The results of this study contrast with those of Faris et al. , the only other study of which we are aware that assessed the effect of sagittal orientation of the femoral implant on ROM. They retrospectively reviewed 623 lateral radiographs of cruciate-retaining knee arthroplasty patients and found no correlation with postoperative knee ROM. Comparison of our results to this study is difﬁcult as they provided scant detail on statistical analysis or the ROM measurement technique used. A further difference between the two studies was the addition of the intra-operative evaluation point in our study. Assessment at this time allows a measure of knee ﬂexion without the confounding inﬂuence of pain, swelling and psychosocial factors [27,31] providing a more direct evaluation of a surgical technique or implant design. The improvement in knee ﬂexion seen in the ﬂexed group at the intra-operative time point (Fig. 4) demonstrated that ﬂexing the femoral implant appears to provide an immediate beneﬁt of increased knee ﬂexion when compared to the neutral orientation. Interestingly, following surgery both groups lost knee Table 1 Baseline characteristics of the groups (mean ± standard deviation) from the general linear model univariate analysis of variance for continuous data. FFD = ﬁxed ﬂexion deformity; WOMAC = Western Ontario McMaster Universities Osteoarthritis Index. SF-12v2 = 12-item Short Form Health Survey; PCS = Physical Component Summary, MCS = Mental Component Summary.
3. Results Twenty knees were available for baseline pre-operative analysis in each group; 11 females and 9 males received the ﬂexed implant intervention and 14 females and 6 males were in the control group. Prior to statistical analysis, the data was assessed for normality and residual outliers. Three data outliers were removed; all three were from one subject from the ﬂexed group in functional tests (pre-operative and 1 year SCT and 1 year TST) who was substantially impaired by severe hip co-morbidity. Pre-operative demographic data and outcome measures including femoral implant ﬂexion angle are shown in Table 1. There were no signiﬁcant differences between any of the outcomes (all p > 0.05) although there were slight trends towards the ﬂexed group reporting more pain and having better mental component scores in the SF-12v12. Radiographic evaluation of the femoral component ﬂexion in the sagittal plane conﬁrmed that the instrumentation provided two signiﬁcantly different groups (p b 0.001); control (mean ± standard deviation, 1.5 ± 1.8°), and ﬂexed (5.3 ± 1.8°). There were no signiﬁcant differences in length of hospital stay (control 6.1±1.9 days, ﬂexed 5.6±1.3, p=0.323). The linear mixed model analysis of outcomes between the two groups is presented in Table 2. Between the groups there were signiﬁcant differences in knee ﬂexion (p=0.031) and the MCS of the SF-12v2 (p=0.009). Mean knee ﬂexion for the groups at each time
Age Gender BMI Knee Flexion Knee FFD Isometric Quadriceps Quadriceps Lag Timed Stand test Stair Climb test WOMAC
SF-12v2 Femoral implant ﬂexion angle
(years) (kg/m2) (degrees) (degrees) (Newtons) (degrees) (seconds) (seconds) Pain Stiffness Function PCS MCS (degrees)
n = 20
n = 20
73.1 ± 10.0 6 M/14 F 29.4 ± 7.7 117.4 ± 11.7 9.5 ± 6.0 166.0 ± 62.6 0.9 ± 3.2 19.2 ± 6.4 14.6 ± 5.8 43.5 ± 25.5 58.4 ± 31.4 51.8 ± 24.0 28.6 ± 7.6 49.9 ± 11.5 1.5 ± 1.8
68.4 ± 9.5 9 M/11 F 31.6 ± 7.1 113.6 ± 8.8 11.1 ± 7.1 175.2 ± 82.6 0.3 ± 3.4 20.1 ± 5.8 14.0 ± 4.7 56.4 ± 23.2 64.7 ± 27.4 61.2 ± 23.2 26.5 ± 7.1 56.5 ± 12.2 5.3 ± 1.8
2.34 – 0.84 1.25 0.63 0.16 0.28 0.22 0.13 2.81 0.47 1.58 0.81 2.92 34.49
0.134 – 0.365 0.271 0.432 0.692 0.603 0.643 0.723 0.102 0.500 0.217 0.374 0.096 b0.001
M. Murphy et al. / The Knee 21 (2014) 257–263
Table 2 Results (mean ± standard deviation) from the Linear Mixed Model analysis of the group x time interaction effect. Outcome
Knee Flexion (degrees) Knee FFD (degrees) Isometric Quadriceps (Newtons) Quadriceps Lag (degrees) Timed Stand test (seconds) Stair Climb test (seconds) WOMAC Pain WOMAC Stiffness WOMAC Function SF12v2 PCS SF12v2 MCS
Control Flexed Control Flexed Control Flexed Control Flexed Control Flexed Control Flexed Control Flexed Control Flexed Control Flexed Control Flexed Control Flexed
Time point Pre-operative
117.4 ± 11.7 113.6 ± 8.8 9.5 ± 6.0 11.1 ± 7.1 166.0 ± 62.6 175.2 ± 82.6 0.9 ± 3.2 0.3 ± 3.4 19.2 ± 6.4 20.1 ± 5.8 14.6 ± 5.8 14.0 ± 4.7 43.5 ± 25.5 56.4 ± 23.2 58.4 ± 31.4 64.7 ± 27.4 51.8 ± 24.0 61.2 ± 23.2 28.6 ± 7.6 26.5 ± 7.1 49.9 ± 11.5 56.5 ± 12.2
117.4 ± 7.6 122.4 ± 6.0 4.6 ± 5.8 4.5 ± 5.3 – – – – – – – – – – – – – – – – – –
103.5 ± 10.7 110.2 ± 7.5 6.0 ± 5.6 5.9 ± 6.3 215.0 ± 81.5 206.9 ± 87.6 1.4 ± 2.6 1.4 ± 2.8 13.6 ± 3.9 15.5 ± 6.2 10.6 ± 4.3 10.8 ± 4.0 14.9 ± 14.85 19.5 ± 22.0 20.4 ± 18.8 28.4 ± 27.5 20.2 ± 16.8 28.2 ± 25.3 34.9 ± 6.1 32.2 ± 6.4 61.1 ± 7.3 53.3 ± 13.2
⁎ pb 0.05.
ﬂexion to the one year time point, a trend frequently reported in the literature [4,32,33]. Flexion of the femoral implant has theoretical advantages and disadvantages. One of the beneﬁts is thought to be a reduction in the risk of femoral notching with the anterior bone cut . Though the instrumentation system in this study was able to consistently guide femoral resection in the required sagittal plane orientation there were three cases of notching; two in the control group and one in the ﬂexed group. Accurate component positioning, is reliant on the accuracy of bone cuts and the type of instrumentation system used [35,36] but is also affected by normal variability in distal femoral bowing [20,37]. This variability may have contributed to the case of
Fig. 4. Mean knee ﬂexion with 95% conﬁdence intervals for each group at each evaluation time point.
notching seen in the ﬂexed group, though a larger sample size may be required to determine whether ﬂexing the implant signiﬁcantly reduces the frequency of notching. A disadvantage of implant ﬂexion may be compromised knee extension. In our study however the mean ﬂexion contracture in the ﬂexed group was comparable to that seen in the control group, and though slightly greater than 5°, which is reported to be normal post-knee arthroplasty , was not concerning. The goal of innovations to improve knee ﬂexion in TKA is to improve the ability to perform routine domestic or occupational activities which require deep ﬂexion  though few studies have thoroughly explored this relationship. Ritter et al.  and Meneghini et al.  assessed the functional components of Knee Society Scores and reported improved stair climbing and walking, with knee ﬂexion greater than 125 and 128° respectively. In contrast Miner et al.  and Park et al.  found only weak associations between knee ﬂexion and WOMAC function and Naylor et al.  reported Oxford Knee Scores improved regardless of the extent of knee ﬂexion gains. We were unable to demonstrate any relationship between knee ﬂexion and the WOMAC or functional tests (timed stand or stair climb tests). This could indicate that these validated functional measures are not sufﬁciently sensitive to detect a difference between the two groups, that the minimum clinically important difference for knee ﬂexion is greater than the 10° we postulated, or perhaps that a strong association between deep ﬂexion and function does not exist. Another reason may be related to quadriceps weakness. Previous studies have demonstrated that TKA patients tend not to utilise all available knee ﬂexion [44,45] suggesting they have either become accustomed to avoiding deep weight bearing knee ﬂexion  or may not have sufﬁcient quadriceps strength to control the large moments in this activity , or both. Quadriceps strength in both groups in our study was below age-matched normative values . This is commonly reported in the literature with long term strength deﬁcits between 30 and 48% compared to age-matched normal populations . The beneﬁt of any improvement in knee ﬂexion therefore may not be fully realised without adequate quadriceps strengthening and for this usually elderly population intensive strengthening programmes may not be a priority. Several studies have shown that patients, even in non-western populations, are not particularly concerned about an inability to achieve deep ﬂexion activities post-operatively  and cite pain relief as a more important goal . Pain relief is
M. Murphy et al. / The Knee 21 (2014) 257–263
reported to be a stronger correlate for patient satisfaction than functional outcome [41,51] or knee ﬂexion . Moreover patients in “high-ﬂex” studies have reported equal satisfaction whether or not their knee ﬂexion was improved [45,52]. Similarly, in our study satisfaction of subjects in the control group, though losing knee ﬂexion at 1 year from their pre-operative baseline was not signiﬁcantly different to those in the ﬂexed group whose mean ﬂexion was essentially unchanged. This study has a number of limitations which should be considered. While an increase in knee ﬂexion as a result of ﬂexing the femoral implant is theoretically derived from an increase in PCO, we did not directly measure PCO. Angular radiographic measurements of implant position are not susceptible to the magniﬁcation error associated with the linear measurements used in PCO studies. Moreover a recent study found that post-operative PCO measurements taken from computerised tomography scans did not signiﬁcantly correlate with lateral radiographs  and few researchers account for the contribution of the translucency of the articular cartilage in pre to post-operative comparative studies . We therefore cannot comment on the speciﬁc mechanism by which increased knee ﬂexion was achieved. Future studies should therefore explore the mechanism by which a ﬂexed femoral implant exerts its inﬂuence. The sample size of 20 per group may also be construed as a study limitation. However, this number was obtained through a sample size calculation to determine whether the intervention would provide a clinically important difference of 10°. In the absence of any established minimum clinically important difference in ﬂexion we believed an improvement of 10° or more would be necessary to translate to functional improvements. That the difference between the groups reached signiﬁcance justiﬁes this sample size. Finally, as this study was conducted with only one knee system; the Proﬁx Total Knee System (Smith & Nephew, Memphis, TN) with posterior cruciate ligament retention, surgeons must be cautious when applying these ﬁndings to other implant designs. 5. Conclusion Flexing the femoral implant in cruciate retaining TKA provides an immediate increase in knee ﬂexion but does not translate to a meaningful clinical or functional beneﬁt at 1 year. Further research is recommended to explore the association between quadriceps strength and improving function with deep knee ﬂexion following TKA. 6. Conﬂict of interest statement Beneﬁts have been received from Smith & Nephew Inc. in the form of funding of a part-time research ofﬁcer in the Department of Orthopaedics, Mater Health Services. Smith & Nephew Inc. did not in any way, initiate, direct or assist with the study's design, the collection, analysis and interpretation of data; the writing of the manuscript; or in the decision to submit the manuscript for publication. References  Rowe PJ, Myles CM, Walker C, Nutton R. Knee joint kinematics in gait and other functional activities measured using ﬂexible electrogoniometry: how much knee motion is sufﬁcient for normal daily life? Gait Posture 2000;12:143-55.  Laubenthal KN, Smidt GL, Kettelkamp DB. A quantitative analysis of knee motion during activities of daily living. Phys Ther 1972;52:34-43.  Mulholland SJ, Wyss UP. Activities of daily living in non-Western cultures: range of motion requirements for hip and knee joint implants. Int J Rehabil Res 2001;24:191-8.  Anouchi YS, McShane M, Kelly F, Elting J, Stiehl J. Range of motion in total knee replacement. Clin Orthop Relat Res 1996;331:87-92.  Ritter MA, Harty LD, Davis KE, Meding JB, Berend ME. Predicting range of motion after total knee arthroplasty. Clustering, log-linear regression, and regression tree analysis. J Bone Joint Surg Am 2003;85A:1278-85.  Schurman DJ, Rojer DE. Total knee arthroplasty: range of motion across ﬁve systems. Clin Orthop Relat Res 2005;430:132-7.
 Rajgopal V, Bourne RB, Chesworth BM, MacDonald SJ, McCalden RW, Rorabeck CH. The impact of morbid obesity on patient outcomes after total knee arthroplasty. J Arthroplasty 2008;23:795-800.  Dennis DA, Komistek RD, Scuderi GR, Zingde S. Factors affecting ﬂexion after total knee arthroplasty. Clin Orthop Relat Res 2007;464:53-60.  Luo SX, Su W, Zhao JM, Sha K, Wei QJ, Li XF. High-ﬂexion vs conventional prostheses total knee arthroplasty: a meta-analysis. J Arthroplasty 2011;26:847-54.  Han HS, Kang SB, Yoon KS. High incidence of loosening of the femoral component in legacy posterior stabilised-ﬂex total knee replacement. J Bone Joint Surg Br 2007;89:1457-61.  Cho SD, Youm YS, Park KB. Three- to six-year follow-up results after high-ﬂexion total knee arthroplasty: can we allow passive deep knee bending? Knee Surg Sports Traumatol Arthrosc 2011;19:899-903.  Gandhi R, Tso P, Davey JR, Mahomed NN. High-ﬂexion implants in primary total knee arthroplasty: a meta-analysis. Knee 2009;16:14-7.  Murphy M, Journeaux S, Russell T. High-ﬂexion total knee arthroplasty: a systematic review. Int Orthop 2009;33:887-93.  Bellemans J, Banks S, Victor J, Vandenneucker H, Moemans A. Fluoroscopic analysis of the kinematics of deep ﬂexion in total knee arthroplasty—inﬂuence of posterior condylar offset. J Bone Joint Surg Br 2002;84B:50-3.  Mizu-uchi H, Colwell Jr CW, Matsuda S, Flores-Hernandez C, Iwamoto Y, D'Lima DD. Effect of total knee arthroplasty implant position on ﬂexion angle before implant-bone impingement. J Arthroplasty 2011;26:721-7.  Massin P, Gournay A. Optimization of the posterior condylar offset, tibial slope, and condylar roll-back in total knee arthroplasty. J Arthroplasty 2006;21:889-96.  Hanratty BM, Thompson NW, Wilson RK, Beverland DE. The inﬂuence of posterior condylar offset on knee ﬂexion after total knee replacement using a cruciate-sacriﬁcing mobile-bearing implant. J Bone Joint Surg Br 2007;89:915-8.  Bauer T, Biau D, Colmar M, Poux X, Hardy P, Lortat-Jacob A. Inﬂuence of posterior condylar offset on knee ﬂexion after cruciate-sacriﬁcing mobile-bearing total knee replacement: a prospective analysis of 410 consecutive cases. Knee 2010;17:375-80.  Bellemans J, Ries MD, Victor JMK. Total Knee Arthroplasty: A Guide to Get Better Performance. Berlin: Springer; 2005.  Faris PM, Ritter MA, Keating EM. Sagittal plane positioning of the femoral component in total knee arthroplasty. J Arthroplasty 1988;3:355-8.  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;15:1833-40.  Ware Jr J, Kosinski M, Keller SD. A 12-Item Short-Form Health Survey: construction of scales and preliminary tests of reliability and validity. Med Care 1996;34:220-33.  Rejeski WJ, Ettinger Jr WH, Schumaker S, James P, Burns R, Elam JT. Assessing performance-related disability in patients with knee osteoarthritis. Osteoarthritis Cartilage 1995;3:157-67.  Newcomer KL, Krug HE, Mahowald ML. Validity and reliability of the timedstands test for patients with rheumatoid arthritis and other chronic diseases. J Rheumatol 1993;20:21-7.  Schulz KF, Altman DG, Moher D. CONSORT 2010 statement: updated guidelines for reporting parallel group randomised trials. BMJ 2010;340:c332.  Russell TG. Goniometry via the internet. Aust J Physiother 2007;53:136.  Lee DC, Kim DH, Scott RD, Suthers K. Intraoperative ﬂexion against gravity as an indication of ultimate range of motion in individual cases after total knee arthroplasty. J Arthroplasty 1998;13:500-3.  Kennedy DM, Stratford PW, Wessel J, Gollish JD, Penney D. Assessing stability and change of four performance measures: a longitudinal study evaluating outcome following total hip and knee arthroplasty. BMC Musculoskelet Disord 2005;6:3.  Minoda Y, Kobayashi A, Iwaki H, Mitsuhiko I, Kadoya Y, Ohashi H, et al. The risk of notching the anterior femoral cortex with the use of navigation systems in total knee arthroplasty. Knee Surg Sports Traumatol Arthrosc 2010;18:718-22.  Gupta SK, Ranawat AS, Shah V, Zikria BA, Zikria JF, Ranawat CS. The P.F.C. sigma RP-F TKA designed for improved performance: a matched-pair study. Orthopedics 2006;29(9 Suppl.):49-52.  Bennett D, Hanratty B, Thompson N, Beverland DE. The inﬂuence of pain on knee motion in patients with osteoarthritis undergoing total knee arthroplasty. Orthopedics 2009;32:4 [http://www.healio.com/orthopedics/knee/journals/ORTHO/ %7BE8898EC3-8312-4165-80B0-1EEF26A53BD3%7D/The-Inﬂuence-of-Pain-on-KneeMotion-in-Patients-With-Osteoarthritis————Undergoing-Total-Knee-Arthroplasty? full=1].  Naylor JM, Yeo AE, Mittal R, Ko VW, Harris IA. Improvements in knee range and symptomatic and functional behavior after knee arthroplasty based on preoperative restriction in range. J Arthroplasty 2012;27:1100-5.  Chiu KY, Ng TP, Tang WM, Yau WP. Review article: knee ﬂexion after total knee arthroplasty. J Orthop Surg (Hong Kong) 2002;10:194-202.  Zalzal P, Backstein D, Gross AE, Papini M. Notching of the anterior femoral cortex during total knee arthroplasty characteristics that increase local stresses. J Arthroplasty 2006;21:737-43.  Shakespeare D. Conventional instruments in total knee replacement: what should we do with them? Knee 2006;13:1-6.  Mihalko WM, Boyle J, Clark LD, Krackow KA. The variability of intramedullary alignment of the femoral component during total knee arthroplasty. J Arthroplasty 2005;20:25-8.  Tang WM, Chiu KY, Kwan MF, Ng TP, Yau WP. Sagittal bowing of the distal femur in Chinese patients who require total knee arthroplasty. J Orthop Res 2005;23:41-5.  Ritter MA, Lutgring JD, Davis KE, Berend ME, Pierson JL, Meneghini RM. The role of ﬂexion contracture on outcomes in primary total knee arthroplasty. J Arthroplasty 2007;22:1092-6.
M. Murphy et al. / The Knee 21 (2014) 257–263  Huddleston JI, Scarborough DM, Goldvasser D, Freiberg AA, Malchau H. 2009 Marshall Urist Young Investigator Award: how often do patients with high-ﬂex total knee arthroplasty use high ﬂexion? Clin Orthop Relat Res 2009;467:1898-906.  Ritter MA, Lutgring JD, Davis KE, Berend ME. The effect of postoperative range of motion on functional activities after posterior cruciate-retaining total knee arthroplasty. J Bone Joint Surg Am 2008;90:777-84.  Meneghini RM, Pierson JL, Bagsby D, Ziemba-Davis M, Berend ME, Ritter MA. Is there a functional beneﬁt to obtaining high ﬂexion after total knee arthroplasty? J Arthroplasty 2007;22:43-6.  Miner AL, Lingard EA, Wright EA, Sledge CB, Katz JN. Knee range of motion after total knee arthroplasty: how important is this as an outcome measure? J Arthroplasty 2003;18:286-94.  Park KK, Chang CB, Kang YG, Seong SC, Kim TK. Correlation of maximum ﬂexion with clinical outcome after total knee replacement in Asian patients. J Bone Joint Surg Br 2007;89:604-8.  Myles CM, Rowe PJ, Walker CR, Nutton RW. Knee joint functional range of movement prior to and following total knee arthroplasty measured using ﬂexible electrogoniometry. Gait Posture 2002;16:46-54.  Huang HT, Su JY, Wang GJ. The early results of high-ﬂex total knee arthroplasty: a minimum of 2 years of follow-up. J Arthroplasty 2005;20:674-9.  Kim TK, Chang CB, Kang YG, Kim SJ, Seong SC. Causes and predictors of patient's dissatisfaction after uncomplicated total knee arthroplasty. J Arthroplasty 2009;24: 263-71.
 Nagura T, Dyrby CO, Alexander EJ, Andriacchi TP. Mechanical loads at the knee joint during deep ﬂexion. J Orthop Res 2002;20:881-6.  Andrews AW, Thomas MW, Bohannon RW. Normative values for isometric muscle force measurements obtained with hand-held dynamometers. Phys Ther 1996;76: 248-59.  Meier W, Mizner RL, Marcus RL, Dibble LE, Peters C, Lastayo PC. Total knee arthroplasty: muscle impairments, functional limitations, and recommended rehabilitation approaches. J Orthop Sports Phys Ther 2008;38:246-56.  Park KK, Shin KS, Chang CB, Kim SJ, Kim TK. Functional disabilities and issues of concern in female Asian patients before TKA. Clin Orthop Relat Res 2007;461: 143-52.  Wylde V, Dieppe P, Hewlett S, Learmonth ID. Total knee replacement: is it really an effective procedure for all? Knee 2007;14:417-23.  Kim Y, Sohn K, Kim J. Range of motion of standard and high-ﬂexion posterior stabilized total knee prostheses: a prospective, randomized study. J Bone Joint Surg Am 2005;87A:1470-5.  Ishii Y, Noguchi H, Takeda M, Ishii H, Toyabe S. Changes in the medial and lateral posterior condylar offset in total knee arthroplasty. J Arthroplasty 2010;26:255-9.  Clarke HD, Hentz JG. Restoration of femoral anatomy in TKA with unisex and gender-speciﬁc components. Clin Orthop Relat Res 2008;466:2711-6.