This article has Open Peer Review reports available.
How accurate is anatomic limb alignment in predicting mechanical limb alignment after total knee arthroplasty?
© Lee et al. 2015
Received: 15 June 2015
Accepted: 5 October 2015
Published: 27 October 2015
Anatomic limb alignment often differs from mechanical limb alignment after total knee arthroplasty (TKA). We sought to assess the accuracy, specificity, and sensitivity for each of three commonly used ranges for anatomic limb alignment (3-9°, 5-10° and 2-10°) in predicting an acceptable range (neutral ± 3°) for mechanical limb alignment after TKA. We also assessed whether the accuracy of anatomic limb alignment was affected by anatomic variation.
This retrospective study included 314 primary TKAs. The alignment of the limb was measured with both anatomic and mechanical methods of measurement. We also measured anatomic variation, including the femoral bowing angle, tibial bowing angle, and neck-shaft angle of the femur. All angles were measured on the same full-length standing anteroposterior radiographs. The accuracy, specificity, and sensitivity for each range of anatomic limb alignment were calculated and compared using mechanical limb alignment as the reference standard. The associations between the accuracy of anatomic limb alignment and anatomic variation were also determined.
The range of 2-10° for anatomic limb alignment showed the highest accuracy, but it was only 73 % (3-9°, 65 %; 5-10°, 67 %). The specificity of the 2-10° range was 81 %, which was higher than that of the other ranges (3-9°, 69 %; 5-10°, 67 %). However, the sensitivity of the 2-10° range to predict varus malalignment was only 16 % (3-9°, 35 %; 5-10°, 68 %). In addition, the sensitivity of the 2-10° range to predict valgus malalignment was only 43 % (3-9°, 71 %; 5-10°, 43 %). The accuracy of anatomical limb alignment was lower for knees with greater femoral (odds ratio = 1.2) and tibial (odds ratio = 1.2) bowing.
Anatomic limb alignment did not accurately predict mechanical limb alignment after TKA, and its accuracy was affected by anatomic variation. Thus, alignment after TKA should be assessed by measuring mechanical alignment rather than anatomic alignment.
Coronal alignment of the lower limb is a major determinants of successful total knee arthroplasty (TKA) [1–3], and mechanical limb alignment is considered the gold standard in the assessment of coronal alignment after TKA [4–7]. Many recent studies have used mechanical limb alignment to assess radiographic outcomes after TKA. However, the measurement of mechanical limb alignment requires special equipment to check the full-length standing anteroposterior (AP) radiographs. In contrast, anatomic limb alignment can be measured on standard (14 × 17 inch) knee radiographs, which are readily available in most clinics. Thus, a number of large, multicenter studies with long-term follow-up periods have used anatomic limb alignment to assess radiographic outcomes [8–10]. However, anatomic limb alignment often differs from mechanical limb alignment, which can make it difficult to compare radiographic outcomes between studies that used different methods of measurement.
The alignment of the limb after TKA is often assessed by using an acceptable range for neutral alignment and using categorical analyses to determine the radiographic outcome. Previous studies have found that knees within an acceptable range for mechanical limb alignment (neutral ± 3°) show better clinical outcomes after TKA than knees for which the coronal alignment was out of this range . Despite recent disagreement regarding the usefulness of this range [11, 12], mechanical limb alignment within ±3° of neutral is most frequently used as an acceptable range to assess the alignment of the lower limb after TKA. In contrast, there is no representative acceptable range for anatomic limb alignment. Given the physiological difference of 6° between the mechanical and anatomic axes of the femur, 6 ± 3° (i.e. 3–9°) may be reasonable [13, 14]. In contrast, the Knee Society Score (KSS) uses 5–10° as the acceptable range for anatomic limb alignment . In addition, the new KSS, which has recently been devised, adopted 2–10° as the acceptable range for anatomic limb alignment [16, 17]. Nonetheless, there is a lack of information regarding which range for anatomic limb alignment can best predict the acceptable range for neutral mechanical limb alignment (neutral ± 3°) with the highest accuracy, specificity, and sensitivity.
The difference in alignment assessment between anatomic and mechanical alignments may be caused by deformities of the femur and/or the tibia . For the femur, mechanical alignment is determined by measuring the line joining the center of the femoral head and the center of the femoral notch. Thus, mechanical alignment is not affected by anatomic variation . In contrast, anatomic alignment uses the line bisecting the distal shaft of the femur. Thus, the degree of femoral bowing can influence the difference between the two methods. Tibia bowing can also affect the accuracy with which anatomic alignment predicts mechanical alignment . Furthermore, the differences between the two types of alignment can be exaggerated by varus orientation of the femoral neck because the center of the femoral head is more medially located in varus deformity of the femoral neck.
We sought to assess the accuracy, specificity, and sensitivity of each of the three commonly used ranges for anatomic limb alignment (3–9°, 5–10° and 2–10°) in predicting an acceptable range (neutral ± 3°) for mechanical limb alignment after TKA. We also investigated whether the accuracy of anatomic limb alignment was affected by anatomic variation, such as the degree of femoral bowing, tibial bowing, and varus orientation of the femoral neck. We hypothesized that anatomic limb alignment would not accurately predict mechanical limb alignment for most knees and that the acceptable range for anatomic limb alignment in the new KSS (i.e., 2–10°) would show the highest accuracy. We also hypothesized that anatomic limb alignment would be less accurate in knees that had greater femoral bowing, tibial bowing, or varus orientation of the femoral neck.
This retrospective study included 314 primary TKAs. From January to July 2011, 284 primary TKAs were performed at our institution. Because the vast majority of TKA candidates in Korea are women, we extended the review of medical records to include 87 knees from men who underwent primary TKA from August 2011 to December 2012. Thus, in total, 371 knees were considered for inclusion in this study. Of these, 57 knees were excluded for the following reasons: 1) 50 (13 %) knees had poor image quality in terms of rotation of the limb, 2) 4 (1 %) knees did not have full-length standing AP radiographs, and 3) 3 (1 %) knees underwent revision surgery within 1 year of the primary TKA. No patient had flexion contracture greater than 20° at 1-year follow-up. Finally, 314 primary TKAs in 212 patients were included in this study. Most knees (312; 99 %) had TKA due to osteoarthritis. The remaining 2 knees had TKA due to rheumatoid arthritis. There were 204 bilateral TKAs (65 %; 102 patients) and 110 unilateral TKAs (35 %; 110 patients). There were 150 (71 %) women and 62 (29 %) men with a mean age of 68 years (range, 52 to 84 years). The mean weight was 65 kg (range, 46 to 95 kg), and the mean height was 157 cm (range, 140 to 178 cm). The mean body mass index (BMI) was 26.6 kg/m2 (range, 18.4 to 39 kg/m2). The patients and/or their families were informed that data from the case would be submitted for publication, and gave their consent. All participants gave their informed consent to assessing and using their data. The study protocols were approved by the ethics committee of the Samsung Medical Center.
Mechanical tibiofemoral angle (°)
Anatomic tibiofemoral angle (°)
Femoral bowing angle (°)
Tibial bowing angle (°)
Neck-shaft angle (°)
Both anatomic and mechanical alignments of the lower limb were measured on the same full-length standing AP radiographs taken 1 year after surgery. When the radiographs were checked, a reference template on the platform of radiography machine was used to control limb rotation, and the patient was asked to stand with the feet shoulder length apart. Radiographic measurements were performed with a picture archiving and communication system (PACS) (General Electric Medical systems, Milwaukee, WI). Alignment was measured to the nearest 0.1 mm for length measurements and 0.1° for angular measurements. The intra- and interobserver reliabilities of all measurements were determined by selecting 20 knees and measuring all angles twice (two weeks apart) by two observers (two of the authors). The reliability of the measurements was assessed with intraclass correlation coefficients (ICC). The ICCs for the intra- and interobserver reliability of the measurements were almost perfect (>0.9).
Statistical analyses were performed with SAS version 9.3 (SAS Institute, Cary, NC). For mechanical limb alignment, neutral ± 3° was considered to be the acceptable range. Applying this range to the knees in the present study revealed that mechanical limb alignment was within the acceptable range for 270 (86 %) knees. For anatomic limb alignment, three acceptable ranges were considered for this study: 3–9°, 5–10° and 2–10°. The accuracy of each range was calculated and compared with the angles for mechanical limb alignment, which were used as a reference standard. To determine the specificity and sensitivity for each range of anatomic limb alignment, varus or valgus malalignment using mechanical limb alignment were set as positive findings. In contrast, knees within an acceptable range of mechanical limb alignment were set as negative findings. Then, we calculated the specificity of each range of anatomic limb alignment. In addition, the sensitivity for varus or valgus malalignment was calculated separately. The results are presented as percentages and 95 % confidence intervals (CI). Statistical significance was determined with McNemar tests with Bonferroni corrections. To determine the associations between the accuracy of anatomic limb alignment and femoral bowing, tibial bowing, or varus orientation of the neck of the femur, only the 2–10° range was used as a dependent variable because it is the most recently recommended range of the Knee Society. Statistical significance was determined with multiple logistic regression analyses, and the results are presented as odds ratios (OR) and 95 % CI.
Accuracy of three commonly used ranges for anatomic limb alignment to predict the acceptable range (neutral ± 3°) for mechanical limb alignment
95 % CI
Specificity of three commonly used ranges for anatomic limb alignment to predict the acceptable range (neutral ± 3°) for mechanical limb alignment*
95 % CI
Sensitivity of three commonly used ranges for anatomic limb alignment in predicting knees with varus malalignment using the mechanical limb alignment method*
95 % CI
Sensitivity of three commonly used ranges for anatomic limb alignment in predicting knees with valgus malalignment using the mechanical limb alignment method*
95 % CI
The accuracy of anatomic limb alignment was affected by the degree of femoral and tibial bowing, but not by the degree of varus orientation of the femoral neck. The accuracy of anatomic limb alignment was reduced in knees with greater femoral bowing (p < 0.001, OR 1.2, 95 % CI [1.1, 1.3]) and tibial bowing (p < 0.001, OR 1.2, 95 % CI [1.1, 1.3]). For each 1° increase in femoral or tibial bowing, the odds of inaccuracy were 1.2 times greater.
Coronal alignment of the lower limb is an important radiographic outcome variable after TKA [1, 3]. To determine coronal limb alignment, both anatomic limb alignment and mechanical alignment have been used. However, these two alignments often differ. Furthermore, no consensus exists regarding the acceptable range for anatomic limb alignment for the prediction of an acceptable range of mechanical limb alignment (neutral ± 3°). If anatomic alignment cannot accurately predict mechanical alignment, the results of clinical studies that use anatomical alignment are likely to be inaccurate. Thus, we sought to assess the accuracy, specificity and sensitivity of each of three commonly used ranges for anatomic limb alignment (3–9°, 5–10° and 2–10°) in predicting an acceptable range (neutral ± 3°) of mechanical limb alignment after TKA. We also aimed to determine whether the accuracy of anatomic limb alignment was affected by anatomic variation.
This study has several limitations. First, we only included patients from one Asian country. Thus, this study cannot provide information on the accuracy of anatomic alignment after TKA for other ethnicities. A previous study found that the relative difference between anatomic and mechanical alignment depends on the study population . Furthermore, Asian patients are more likely to have femoral or tibial bowing than are Caucasians [4, 14]. Thus, the accuracy of anatomic limb alignment can differ according to ethnicity. However, considering that an increasing number of TKAs are being performed in Asian countries, we believe the present study provides valuable information to a broad readership. Second, 71 % of the subjects included in this study were women. The characteristics of bone geometry can differ between the sexes, and thus, caution should be used when applying our results to other populations with different sex ratios. However, we did attempt to enroll more men despite the extreme predominance of female TKA patients in our country [4, 22]. Third, this study only included radiographic results without clinical data. Thus, we do not know how the differences between the two methods affect clinical outcomes. We focused on determining the degree of difference and its characteristics between the radiographic data measured with the anatomic and mechanical limb alignments. In addtion, this study used two-dimensional assessment with conventional radiographs even though femoral and/or tibial bowing may also be affected by sagittal bowing of the bones and rational shapes. Similarly, flexion contracture of the knee joint can also affect the results of the two-dimensional study. Finally, our results may have been different if we had used a different range of anatomic limb alignments. However, we assessed the ranges proposed in the KSS, both the new and the old, which are the most popular scoring system in TKA [15, 17], so we believe that we chose the most appropriate ranges for our analyses.
Our findings support the hypothesis that anatomic limb alignment does not accurately predict mechanical limb alignment in most knees. Some previous studies have assessed the correlation between anatomic and mechanical limb alignment. These studies found moderate to excellent correlations (r = 0.65 to 0.86) and thus proposed that anatomic limb alignment can be used as a proxy for mechanical alignment [13, 23, 24]. However, the offset angles between anatomic and mechanical limb alignments were reported to have large variations (0.1 to 4.21°) in previous studies [5, 23–25]. In addition, the offset angles differed according to sex . Therefore, even if moderate to excellent correlations exist between anatomic and mechanical limb alignments, the absolute values can differ considerably between the two methods. The inaccuracy of the anatomic alignment measurements was probably caused by the mismatch between the acceptable ranges for the two methods. Surgeons typically use femoral bushings with 5–6° of valgus during TKA on the assumption that the distal femoral mechanical-anatomical angles are 5–6°. On the basis of this assumption, an angle of 6 ± 3° is a reasonable range for acceptable anatomic limb alignment [13, 14]. However, a significant number of patients (28.6 %) have distal femoral mechanical-anatomical angles that are outside of the range of 5 ± 2° (range, 2.0 - 9.6°) . Furthermore, the acceptable ranges for anatomic limb alignment used in previous studies have shown large variability [15–17, 20]. The desired anatomic limb alignment is defined as 2–10° in the new KSS score [16, 17]. Compared to the range of 5–10° in the old KSS , the range of 2–10° had substantially higher accuracy and specificity in the current study.
Anatomic limb alignment did not accurately predict mechanical limb alignment after TKA, and its accuracy was affected by anatomic variation. Thus, alignment after TKA should be assessed by measuring mechanical alignment rather than anatomic alignment. In addition, our findings should be considered when interpreting radiographic results on alignment of the limb after TKA.
The authors would like to thank Ka Young Kim, a clinical investigator, for assisting with data collection.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Jeffery RS, Morris RW, Denham RA. Coronal alignment after total knee replacement. J Bone Joint Surg. 1991;73:709–14.Google Scholar
- Matsuda S, Miura H, Nagamine R, Urabe K, Harimaya K, Matsunobu T, et al. Changes in knee alignment after total knee arthroplasty. J Arthroplasty. 1999;14:566–70.View ArticlePubMedGoogle Scholar
- Ritter MA, Faris PM, Keating EM, Meding JB. Postoperative alignment of total knee replacement. Its effect on survival. Clin Orthop Relat Res. 1994;299:153–6.PubMedGoogle Scholar
- Chang CB, Choi JY, Koh IJ, Seo ES, Seong SC, Kim TK. What should be considered in using standard knee radiographs to estimate mechanical alignment of the knee? Osteoarthritis Cartilage. 2010;18:530–8.View ArticlePubMedGoogle Scholar
- 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;55:306–13.View ArticlePubMedGoogle Scholar
- Jessup DE, Worland RL, Clelland C, Arredondo J. Restoration of limb alignment in total knee arthroplasty: evaluation and methods. J South Orthop Assoc. 1997;6:37–47.PubMedGoogle Scholar
- Rauh MA, Boyle J, Mihalko WM, Phillips MJ, Bayers-Thering M, Krackow KA. Reliability of measuring long-standing lower extremity radiographs. Orthopedics. 2007;30:299–303.PubMedGoogle Scholar
- Hunter DJ, Niu J, Felson DT, Harvey WF, Gross KD, McCree P, et al. Knee alignment does not predict incident osteoarthritis: the Framingham osteoarthritis study. Arthritis Rheum. 2007;56:1212–8.View ArticlePubMedGoogle Scholar
- Ritter MA, Davis KE, Davis P, Farris A, Malinzak RA, Berend ME, et al. Preoperative malalignment increases risk of failure after total knee arthroplasty. J Bone Joint Surg Am. 2013;95:126–31.View ArticlePubMedGoogle Scholar
- Ritter MA, Davis KE, Meding JB, Pierson JL, Berend ME, Malinzak RA. The effect of alignment and BMI on failure of total knee replacement. J Bone Joint Surg Am. 2011;93:1588–96.View ArticlePubMedGoogle Scholar
- Bellemans J, Colyn W, Vandenneucker H, Victor J. The Chitranjan Ranawat award: is neutral mechanical alignment normal for all patients? The concept of constitutional varus. Clin Orthop Relat Res. 2012;470:45–53.View ArticlePubMedGoogle Scholar
- Magnussen RA, Weppe F, Demey G, Servien E, Lustig S. Residual varus alignment does not compromise results of TKAs in patients with preoperative varus. Clin Orthop Relat Res. 2011;469:3443–50.View ArticlePubMedPubMed CentralGoogle Scholar
- Skytta ET, Lohman M, Tallroth K, Remes V. Comparison of standard anteroposterior knee and hip-to-ankle radiographs in determining the lower limb and implant alignment after total knee arthroplasty. Scand J Surg. 2009;98:250–3.PubMedGoogle Scholar
- Tang WM, Zhu YH, Chiu KY. Axial alignment of the lower extremity in Chinese adults. J Bone Joint Surg Am. 2000;82-A:1603–8.PubMedGoogle Scholar
- Insall JN, Dorr LD, Scott RD, Scott WN. Rationale of the Knee Society clinical rating system. Clin Orthop Relat Res. 1989;13–14.Google Scholar
- Noble PC, Scuderi GR, Brekke AC, Sikorskii A, Benjamin JB, Lonner JH, et al. Development of a new Knee Society scoring system. Clin Orthop Relat Res. 2012;470:20–32.View ArticlePubMedGoogle Scholar
- Scuderi GR, Bourne RB, Noble PC, Benjamin JB, Lonner JH, Scott WN. The new Knee Society Knee Scoring System. Clin Orthop Relat Res. 2012;470:3–19.View ArticlePubMedGoogle Scholar
- Yau WP, Chiu KY, Tang WM, Ng TP. Coronal bowing of the femur and tibia in Chinese: its incidence and effects on total knee arthroplasty planning. J Orthop Surg (Hong Kong). 2007;15:32–6.Google Scholar
- Nagamine R, Inoue S, Miura H, Matsuda S, Iwamoto Y. Femoral shaft bowing influences the correction angle for high tibial osteotomy. J Orthop Sci. 2007;12:214–8.View ArticlePubMedGoogle Scholar
- Lasam MP, Lee KJ, Chang CB, Kang YG, Kim TK. Femoral lateral bowing and varus condylar orientation are prevalent and affect axial alignment of TKA in Koreans. Clin Orthop Relat Res. 2013;471:1472–83.View ArticlePubMedGoogle Scholar
- Matsumoto T, Hashimura M, Takayama K, Ishida K, Kawakami Y, Matsuzaki T, et al. A radiographic analysis of alignment of the lower extremities--initiation and progression of varus-type knee osteoarthritis. Osteoarthritis Cartilage. 2015;23:217–23.View ArticlePubMedGoogle Scholar
- Koh IJ, Kim TK, Chang CB, Cho HJ, In Y. Trends in Use of Total Knee Arthroplasty in Korea From 2001 to 2010. Clin Orthop Relat Res. 2012;417(5):1441–50.Google Scholar
- Issa SN, Dunlop D, Chang A, Song J, Prasad PV, Guermazi A, 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;57:398–406.View ArticlePubMedGoogle Scholar
- Colebatch AN, Hart DJ, Zhai G, Williams FM, Spector TD, Arden NK. Effective measurement of knee alignment using AP knee radiographs. Knee. 2009;16:42–5.View ArticlePubMedGoogle Scholar
- Kraus VB, Vail TP, Worrell T, McDaniel G. A comparative assessment of alignment angle of the knee by radiographic and physical examination methods. Arthritis Rheum. 2005;52:1730–5.View ArticlePubMedGoogle Scholar
- Nam D, Maher PA, Robles A, McLawhorn AS, Mayman DJ. Variability in the relationship between the distal femoral mechanical and anatomical axes in patients undergoing primary total knee arthroplasty. J Arthroplasty. 2013;28:798–801.View ArticlePubMedGoogle Scholar