Abstract
Background
High tibial osteotomy (HTO) is a well-established and commonly utilized technique in medial knee osteoarthritis secondary to varus malalignment. Accurate measurement of the preoperative limb alignment, and the amount of correction required are essential when planning limb realignment surgery. The hip-knee-ankle angle (HKA) measured on a full length weightbearing (FLWB) X-ray in the standing position is considered the gold standard, since it allows for reliable and accurate measurement of the mechanical axis of the whole lower extremity. In general practice, alignment is often evaluated on standard anteroposterior weightbearing (APWB) X-rays, as the angle between the femur and tibial anatomic axis (TFa). It is, therefore, of value to establish if measuring the anatomical axis from limited APWB is an effective measure of knee alignment especially in patients undergoing osteotomy about the knee.
Methods
Three independent observers measured preoperative and postoperative FTa with standard method (FTa1) and with circles method (FTa2) on APWB X-ray and the HKA on FLWB X-ray at three different time-points separated by a two-week period. Intra-observer and inter-observer reliabilities and the comparison and relationship between anatomical and mechanical alignment were calculated.
Results
Intra- and interclass coefficients for all the three methods indicated excellent reliability, having all the values above 0.80. Using the mean of paired t-student test, the comparison of HKA versus TFa1 and TFa2 showed a statistically significant difference (p<.0001) both for the pre-operative and post-operative sets of values. The correlation between the HKA and FTal was found poor for the preoperative set (R=0.26) and fair for the postoperative one (R=0.53), while the new circles method showed a higher correlation in both the preoperative (R=0.71) and postoperative sets (R=0.79).
Conclusions
Intra-observer reliability was high for HKA, FTal and FTa2 on APWB x-rays in the pre- and post-operative setting. Inter-rater reliability was higher for HKA and TFa2 compared to FTal. The femoro-tibial angle as measured on APWB with the traditional method (FTal) has a weak correlation with the HKA, and based on these findings, should not be used in everyday practice. The FTa2 showed better correlation with the HKA, although not excellent
Level of Evidence
Level III, Retrospective study.
Introduction
Medial knee osteoarthritis is a common disease frequently caused by additional load on the medial compartment due to varus deformity1. It has been shown that varus alignment increases the risk of medial OA progression, and valgus alignment increases the risk of lateral OA progression, with the severity of malalignment predicting the decline in physical function2. Young active patients with medial compartment osteoarthritis of the knee combined with a varus deformity can be treated with a high tibial osteotomy (HTO), a well- established and commonly used technique1,3,4,5. HTO is performed to reduce pain, diminish the progression of OA perhaps to postpone or avoid total knee arthroplasty3. Accurate measurement of limb alignment, and the amount of correction required are essential when planning a limb realignment surgery6,4. As suggested by Miniaci et al7, Noyes et al.8, and Dugdale et al.9 the optimal outcome would be achieved transferring the weight bearing line to a point around 62.5% of tibial plateau width from medial to lateral (usually just lateral to the lateral tibial spine), aiming for a mechanical axis comprised between 3° and 5° of valgus. Significant under- or overcorrection of the mechanical axis may lead to disappointing clinical results1. Possible factors affecting the outcomes after an HTO are inaccurate preoperative planning of the desired correction, inadequate correction during surgery and loss of correction in the post-operative period. The hip-knee-ankle angle (HKA) measured on a full length weightbearing (FLWB) X-ray made in standing position is considered the gold standard, since allows for reliable and accurate measurement of the mechanical axis of the whole lower extremity. This method is however time consuming, requires special equipment, involves significant radiation exposure and generates extra costs10. In general practice, alignment is often evaluated on a standard anteroposterior weightbearing (APWB) X-rays6, as the angle between the femur and tibial anatomic axis (TFa). It is, therefore, of value to establish if measuring anatomical axis from limited APWB is an effective measure of knee alignment especially in patients undergoing to HTO. We asked two research questions: (1) Is the measurement of limb alignment reliable in both FLWB and APWB? (2) Do the HKA and TFa correlate in patients with medial knee OA secondary to varus alignment?
Materials and Methods
Film series of thirty-six consecutive patients, undergoing isolated HTO by the same experienced surgeon (AA.) from 2010 to 2013, were retrospectively reviewed. The indication for surgery for all the subjects was symptomatic medial compartmental knee osteoarthritis with varus deformity. The surgical technique was performed according to the former description by the senior author11. Each patient underwent to a standard anteroposterior weightbearing (APWB) X-ray and full length weightbearing (FLWB) X-ray, both preoperatively and postoperatively. Full-limb radiographs were obtained using a long length vertical cassette holder containing four 14×17 inches graduated cassettes. The X-ray beam was centered at the knee at a distance of 94 inches. The beam was parallel to the floor and the machine's settings were 100 to 200 mA-s and a kilo voltage of 90, depending on limb size and tissue characteristics. The APWB films were obtained on a 17 × 17 inch cassette. The X-ray beam centered at the knee at a distance of 45 inches with the patients having the back of their knees in contact with the vertical cassette. The beam was parallel to the floor and the machine's settings were 3.2 mA-s and a kilo voltage of 55. For both the X-ray types, the subjects were asked to stand without footwear, with tibial tubercles facing forward. Both limbs were radiographed. The same protocol was applied in performing both types of radiological exams in all the patients. Both knees and both legs were scanned in full extension and with full weightbearing for every examination. In FLWB X-rays the alignment was measured as the angle between femoral and tibial mechanical axes. Femoral and tibial mechanical axes run from the midpoint of tibial spines, respectively to the center of hip and ankle joint (HKA)2. In APWB x-rays, alignment was assessed by measuring the angle between femoral and tibial anatomical axes (femur-tibial angle, FTa). Two different methods were used to measure the FTa. The first one (TFa1), has been previously used by many authors12,13,14,15,16,17,18,19,6: a line running from the midpoint of tibial spines to a point located midway between medial and lateral cortical bone surface, 10 cm above the tibial spine for the femur and 10 cm below for the tibia17 (Figure 1). The second one (TFa2) is based on a method recently described by Veljkovic et al.20 to define the position of the talus as it relates to the anatomic tibial axis. In this method20, on the same type of X-ray, two different circles were fit between medial and lateral cortices of femur and tibia, 5 and 10 cm far from the joint line (Figure 2 Point A and B). A joint extended, mid-diaphyseal line was drawn for femur and tibia at intersection of the center of the circles at points A and B. Fta2 was obtained at the intersection of the two diaphyseal lines. In order to assess the reliability of the above mentioned methods, three independent observers measured preoperative and postoperative APWB and FLWB views at three different time-points separated by a two-week period, with dedicated software. The measures were recorded in degrees, and angles lower than 180° were considered as varus alignment and those higher than 180° valgus alignment.
Figure 1. Tibio-femoral angle measurement on APWB x-ray: traditional method (TFa1).
Figure 2. Tibio-femoral angle measurement on APWB x-rays: circles method (TFa2).
Table 2.
Tibio-femoral angle as measured on APWB x-rays: intra- and inter-tester reliability of die traditional method (TFa1)
| TFa1 | |||
| Intra-tester reliability | 1 | 2 | 3 |
| preop | 0.91 | 0.95 | 0.93 |
| postop | 0.92 | 0.95 | 0.96 |
| Inter-tester reliability | |||
| preop | 0.81 | ||
| postop | 0.84 | ||
Statistical Analysis
The intra-observer and inter-observer reliabilities for continuous measurements were calculated with the Intra-Class Correlation Coefficient ICC (3, 1). Since a formula to calculate the sample size based on ICC (3,1) is not available, the ICC (2,1)-based sample size formula described by Doros and Lew was used as a reference. Assuming that ICC is at least 0.7, then 32 (or more) subjects used by each of the three raters would be sufficient to assure an ICC 95% confidence interval smaller than 0.3. The comparison and relationship between anatomical and mechanical alignment were calculated by Paired t test and Pearson's correlation/linear regression.
Results
All the patients had varus alignment. The preoperative average HKA was 172.9°, while FTa1 was 175.7° and FTa2 was 175.3°. The postoperative average HKA was 181.3°, while the FTa1 was 183.3° and FTa2 was 182.7°. Intra- and intertester intraclass coefficients for all the three methods indicated excellent reliability, having all the values above 0.80 (Tables 1–3). The comparison of HKA versus TFa1 and TFa2 showed, by mean of paired t-student test, a statistically significant difference (p<.0001) both for the pre-operative and post-operative values (Table 4 and 5). The correlation between the HKA and FTal was found poor for the preoperative set (R=0.26) and fair for the postoperative one (R=0.53), while the new method showed a higher correlation either in the preoperative set (R=0.71) and in the postoperative one (R=0.79) (Scattered plots, Fig 3 – 6).
Table 1.
Hip-Knee-Ankle angle (HKA) intra- and inter-tester reliability
| HKA | |||
| Intra-tester reliability | 1 | 2 | 3 |
| preop | 0.92 | 0.95 | 0.97 |
| postop | 0.94 | 0.96 | 0.99 |
| Inter-tester reliability | |||
| preop | 0.89 | ||
| postop | 0.95 | ||
Table 3.
Tibio-femoral angle as measured on APWB x-rays: intra- and inter-tester reliability of the circles method (TFa2)
| TFa2 | |||
| Intra-tester reliability | 1 | 2 | 3 |
| preop | 0.89 | 0.90 | 0.89 |
| postop | 0.97 | 0.93 | 0.97 |
| Inter-tester reliability | |||
| preop | 0.90 | ||
| postop | 0.97 | ||
Table 4.
Comparison of pre-operative and post-operative Mechanical axis (HKA) versus Anatomical axis (FTa1). Mean difference between the values and standard deviation are given in degrees.
| HKA VS TFa1: PAIRED T-TEST BETWEEN THE AVERAGE OF THE HKA and TFa1 MEASUREMENTS | ||
|---|---|---|
| Mean difference ± SD | ||
| Preop | 2.85°±3.05° | |
| p<.0001 | ||
| Postop | 1.99°±3.2° | |
Table 5.
Comparison of pre-operative and post-operative Mechanical axis (HKA) versus Anatomical axis (FTa2). Mean difference between the values and standard deviation are given in degrees.
| HKA VS TFa2: PAIRED T-TEST BETWEEN THE AVERAGE OF THE HKA and TFa2 MEASUREMENTS | ||
|---|---|---|
| Mean difference ± SD | ||
| Preop | 2.48°±2.73° | |
| p<.0001 | ||
| Postop | 2.01°±1.81° | |
Fig.3. Scattered Plot with linear regression showing the correlation between preoperative mechanical axis (full length weight bearing x-rays, HKA) and tibio-femoral angle (as measured on standard AP weight bearing x-rays, TFa1). R=0.26.
Fig.6. Scattered Plot with linear regression showing the correlation between postoperative mechanical axis (full length weight bearing x-rays) and tibio-femoral angle (as measured on standard AP weight bearing x-rays, TFa2). R=0.79.
Fig.4. Scattered Plot with linear regression showing the correlation between postoperative mechanical axis (full length weight bearing x-rays, HKA) and tibio-femoral angle (as measured on standard AP weight bearing x-rays, TFa1). R=0.53.
Fig.5. Scattered Plot with linear regression showing the correlation between preoperative mechanical axis (full length weight bearing x-rays) and tibio-femoral angle (as measured on standard AP weight bearing x-rays, TFa2). R=0.71.
Discussion
The gold standard for measuring knee alignment is represented by mechanical axis of the lower limb, measured on FLWB X-ray. This is not routinely performed on patients in the pre and postoperative clinical setting, because of greater cost and increased radiation exposure. FLWB X-ray involves exposure of the pelvis to radiation, with effective radiation from one film at 73-fold higher than a standard APWB x-ray21. The first research question asked whether the two methods (HKA and TFa) were reliable. Results from intra- and inter-tester reliability shows as both methods have acceptable intra-tester reliability. The inter-tester reliability was satisfactory for the FLWB X-ray both in the pre- and post-operative sets of x-rays, while it was just fair for APWB x-ray. These outcomes are in agreement with the current available literature19. The second research question asked is if the HKA and TFa correlate in a population affected by medial knee OA secondary to varus deformity. The results from this study show that there is a very low correlation with statistically significant differences between the HKA and the TFa1. Compared to our results, previous studies showed a better correlation between mechanical and anatomical axes (r= 0.66 to 0.93)12,15,13,14,16,17,18,19,6. One other study had comparable results to this study. Van Raaij6 found high intra- and inter-observer reliability for FTa but low correlation with the HKA angle (r = 0.34) in cohort of 68 patients with painful medial knee OA. A possible explanation for such conflicting results may be related to the differences in the methods used to measure the TFa in the study populations in the various studies. Most of the previous studies used the same FLWB x-rays to measures both the HKA and the TFa, allowing any variance in patient positioning in terms of rotation and weight bearing to be non-existent. Two studies with a dedicated standard AP x-ray in flexion to measure the TFa, used a specific leg holder (Synaflex)13,14. This is another factor that could help in reducing the variability of patient positioning. In addition, a clear description of the patients' population in terms of OA severity was not clearly stated in all the previous studies. It has been shown that the difference between mechanical and anatomical axes varies with gender and presence of advanced OA12. In this study patients were indicated and underwent to HTO for symptomatic medial knee OA. Kraus et al.14 identified a mean offset for the anatomical axis of 4.21° valgus from the mechanical axis (3.5° in women, 6.4° in men) using fixed flexion posteroanterior (PA) knee x-rays. We used fully extended knee radiographs and found no evidence of an offset angle or constant relationship. One possible explanation for these results is the crossing point of the tibial and femoral anatomical angles. The FT angle or anatomical axis does not pass through the center of the knee joint due to the physiological valgus of femur, intersecting the knee joint line slightly medial to the joint center22,23. The knee joint center is regarded as a landmark for FTa measurement6. Thus, accepting the joint center as the landmark for the FTa measurement might diminish our ability to find the true anatomical axis and its relationship with the HKA angle. Sheeny et al.18 showed as the longer were the landmark on the femur and tibia to calculate the FTa, the higher was the correlation with the HKA, therefore suggesting that a longer x-rays would be necessary to have a closer measurement of the actual FTa, though this would defeat the advantages of standard AP knee x-rays. The same authors18 showed that the offset of FTa measurements from standard AP x-rays varies depending on direction and degree of knee deformity. They found that in varus limbs the shorter the shaft length for measuring FTa, the wider was the offset magnitude between HKA and FTa values. The opposite goes for valgus alignment. The patients in our study also changed from a varus alignment to a neutral or valgus alignment after HTO surgery, this could explain why the correlation between HKA and TFa improved from pre- to post-operative assessment. Since the TFa1 showed a low correlation with the HKA, in order to find and alternative method to measure the lower limb alignment, we studied in addition the reliability and the correlation of a new method, the TFa2, developed on the basis of a previously described method to define the position of the talus as it relates to the anatomic tibial axis. This new method showed high reliability for both intra- and intertester intraclass coefficient and it showed a good correlation with the HKA (R up to 0.79). The authors explanation is that when using the circles, the femoral and tibial anatomical axis are measured independently from the joint center, therefore allowing for a closer measurement to the true anatomical axis, as opposite to the traditional method (TFa1). Though the correlation was improved compared to the traditional method, it's still not excellent. The same factors that influenced negatively the correlation between HKA and TFa1 (varus alignment of the study population, variability in patient positioning from FLWB to APWB) could play a detrimental role in the relationship between HKA and TFa2. The weaknesses of this study includes the following: it is a retrospective study; the FLWB and APWB x-rays were not always taken the same day, thus increasing the variability in patient positioning; we did not evaluate the relationship between the two angles separately for men and women, since its known that the offset varies depending on sex16,14. However this study does assess the reliability of APWB x-rays, and the correlation between HKA and FTa in a cohort of patients with varus alignment of the knee that underwent a HTO in a common clinical setting.
Conclusion
In summary it was found that intra-observer reliability is high for HKA, FTal and FTa2 on APWB x-rays in the pre- and post-operative setting. Inter-rater reliability was higher for HKA and TFa2 compared to FTal. The femoro-tibial angle as measured on APWB with the traditional method (FTal) has a weak correlation with the HKA, and based on these findings, should not be used in everyday practice. The FTa2 showed higher correlation with the HKA, though not excellent. Further studies are necessary to better understand the correlation between HKA and the TFa as measured by the new method. Thus, at the time being, substituting full leg x-rays with standard AP may cause inaccuracy, which is undesirable in patients undergoing corrective knee surgery.
References
- 1.Hankemeier S, et al. Accuracy of high tibial osteotomy: comparison between open- and closed-wedge technique. Knee Surg Sports Traumatol Arthrosc. 2010;18(10):1328–33. doi: 10.1007/s00167-009-1020-9. [DOI] [PubMed] [Google Scholar]
- 2.Sharma L, et al. The role of knee alignment in disease progression and functional decline in knee osteoarthritis. JAMA. 2001;286(2):188–95. doi: 10.1001/jama.286.2.188. [DOI] [PubMed] [Google Scholar]
- 3.Gaasbeek RD, et al. Correction accuracy and collateral laxity in open versus closed wedge high tibial osteotomy. A one-year randomised controlled study. Int Orthop. 2010;34(2):201–7. doi: 10.1007/s00264-009-0861-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Schroter S, et al. Digital planning of high tibial osteotomy. Interrater reliability by using two different software. Knee Surg Sports Traumatol Arthrosc. 2013;21(1):189–96. doi: 10.1007/s00167-012-2114-3. [DOI] [PubMed] [Google Scholar]
- 5.Amendola A. Unicompartmental osteoarthritis in the active patient: the role of high tibial osteotomy. Arthroscopy. 2003;19(Suppl 1):109–16. doi: 10.1016/j.arthro.2003.09.048. [DOI] [PubMed] [Google Scholar]
- 6.van Raaij TM, et al. Conventional knee films hamper accurate knee alignment determination in patients with varus osteoarthritis of the knee. Knee. 2009;16(2):109–11. doi: 10.1016/j.knee.2008.10.003. [DOI] [PubMed] [Google Scholar]
- 7.Miniaci A, et al. Proximal tibial osteotomy. A new fixation device. Clin Orthop Relat Res. 1989;(246):250–9. [PubMed] [Google Scholar]
- 8.Noyes FR, Barber SD, Simon R. High tibial osteotomy and ligament reconstruction in varus angulated, anterior cruciate ligament-deficient knees. A two- to seven-year follow-up study. Am J Sports Med. 1993;21(1):2–12. doi: 10.1177/036354659302100102. [DOI] [PubMed] [Google Scholar]
- 9.Dugdale TW, Noyes FR, Styer D. Preoperative planning for high tibial osteotomy. The effect of lateral tibiofemoral separation and tibiofemoral length. Clin Orthop Relat Res. 1992;(274):248–64. [PubMed] [Google Scholar]
- 10.Brouwer RW, et al. The whole leg radiograph: standing versus supine for determining axial alignment. Acta Orthop Scand. 2003;74(5):565–8. doi: 10.1080/00016470310017965. [DOI] [PubMed] [Google Scholar]
- 11.Rossi R, Bonasia DE, Amendola A. The role of high tibial osteotomy in the varus knee. J Am Acad Orthop Surg. 2011;19(10):590–9. doi: 10.5435/00124635-201110000-00003. [DOI] [PubMed] [Google Scholar]
- 12.Chang CB, et al. What should be considered in using standard knee radiographs to estimate mechanical alignment of the knee? Osteoarthritis Cartilage. 2010;18(4):530–8. doi: 10.1016/j.joca.2009.12.004. [DOI] [PubMed] [Google Scholar]
- 13.Felson DT, et al. Can anatomic alignment measured from a knee radiograph substitute for mechanical alignment from full limb films? Osteoarthritis Cartilage. 2009;17(11):1448–52. doi: 10.1016/j.joca.2009.05.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Kraus VB, et al. A comparative assessment of alignment angle of the knee by radiographic and physical examination methods. Arthritis Rheum. 2005;52(6):1730–5. doi: 10.1002/art.21100. [DOI] [PubMed] [Google Scholar]
- 15.Navali AM, Bahari LA, Nazari B. A comparative assessment of alternatives to the full-leg radiograph for determining knee joint alignment. Sports Med Arthrosc Rehabil Ther Technol. 2012;4(1):40. doi: 10.1186/1758-2555-4-40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Issa SN, et al. Full-limb and knee radiography assessments of varus-valgus alignment and their relationship to osteoarthritis disease features by magnetic resonance imaging. Arthritis Rheum. 2007;57(3):398–406. doi: 10.1002/art.22618. [DOI] [PubMed] [Google Scholar]
- 17.Hinman RS, May RL, Crossley KM. Is there an alternative to the full-leg radiograph for determining knee joint alignment in osteoarthritis? Arthritis Rheum. 2006;55(2):306–13. doi: 10.1002/art.21836. [DOI] [PubMed] [Google Scholar]
- 18.Sheehy L, et al. Does measurement of the anatomic axis consistently predict hip-knee-ankle angle (HKA) for knee alignment studies in osteoarthritis? Analysis of long limb radiographs from the multicenter osteoarthritis (MOST) study. Osteoarthritis Cartilage. 2011;19(1):58–64. doi: 10.1016/j.joca.2010.09.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Colebatch AN, et al. Effective measurement of knee alignment using AP knee radiographs. Knee. 2009;16(1):42–5. doi: 10.1016/j.knee.2008.07.007. [DOI] [PubMed] [Google Scholar]
- 20.Veljkovic A, et al. Lateral talar station: a clinically reproducible measure of sagittal talar position. Foot Ankle Int. 2013;34(12):1669–76. doi: 10.1177/1071100713500489. [DOI] [PubMed] [Google Scholar]
- 21.Hart DWB. Radiation exposure of the UK population from medical and dental X-ray examinations. Chilton, UK: National Radiological Protection Board; 2002. [Google Scholar]
- 22.Moreland JR, Bassett LW, Hanker GJ. Radiographic analysis of the axial alignment of the lower extremity. J Bone Joint Surg Am. 1987;69(5):745–9. [PubMed] [Google Scholar]
- 23.Ja J. Accuracy of intramedullary femoral alignment in total knee replacement: intraoperative assessment of alignment rod position. The Knee. 1999;6:211–5. [Google Scholar]