Chinese Medical Journal 2013;126(20):3840-3844
Anatomical references for tibial sagittal alignment in total knee arthroplasty: A comparison of three anatomical axes based on 3D reconstructed CT images

SHAO Jun-jie, Thomas Parker Vail, WANG Qiao-jie, SHEN Hao, CHEN Yun-su, WANG Qi, JIANG Yao and ZHANG Xian-long

Keywords
knee arthroplasty; fibula; tibia; axis; computer-assisted image reconstruction
Abstract
Background This study was designed to analyze three tibial axis reference lines including the anterior tibial cortex (ATC) line, the fibular line (FL), and the anatomical axis of tibia (AAT) line, to determine which line most closely parallels the mechanical axis (MA) of the tibia in the sagittal plane. The clinical relevance of the study is that through finding a reliable landmark on the leg, a surgeon may minimize posterior tibial slope measurement errors thereby and improving the technique for assuring proper alignment of total knee arthroplasty.
Methods The material for this study included CT scans of the tibia from 85 consecutive patients and 168 knees (78 without osteoarthritis (OA) and 90 knees with OA). Measurements of the angles between the tibial mechanical axis and each of three reference lines in the sagittal plane were carried out using 3D imaging software.
Results Mean angles of 168 knees were as follows: aMT (3.96±0.85)º, aMF (0.70±0.58)º, and aMA (1.40±0.66)º, (aMT: an angle between MA and ATC, aMF: an angle between MA and FL, aMA: an angle between MA and AAT. All above-mentioned angles were measured in the sagittal plane of tibia) and the aMF was significantly smaller than the others (P <0.0001). The mean value of the medial tibial slope angle vs. the MA was (9.19±3.97)º, and this was significantly larger than the mean lateral slope angle of (6.62±4.23)º (P <0.0001). The difference between aMF without OA and with OA was not statistically significant (P=0.5015) and the association between the aMT and aMA was strong (r=0.82, P <0.01).
Conclusions FL was more closely parallel to the MA of tibia, and more showed less variation between OA and non-OA controls than ATC and AAT lines. Furthermore, the amount of posterior slope in medial plateau was greater than that in lateral plateau. The findings of this analysis suggest that when using the anterior tibial cortex line as is commonly done with extramedullary tibial resection guides, the tibial resection should be sloped approximately four degrees more posteriorly.
The long-term success of total knee arthroplasty (TKA) depends in part upon the mechanical alignment of the lower affected limb and position the prosthetic components relative to the mechanical axis.1,2 In contemporary knee replacement systems, the tibial component is positioned perpendicular to mechanical axis (MA) of tibia in the coronal plane and some degree of the posterior tilt in the sagittal plane.3,4 Posterior tibial slope (PTS) contributes to antero-posterior stability. Variation in the posterior tibial slope may cause the flexion gap to be too tight or too loose, affecting the range of motion, anteroposterior stability, and contact pressure within the tibiofemoral joint.5,6 Both extramedullary and intramedullary (IM) alignment guides are commonly used to prepare the tibia. The extramedullary (EM) alignment guide requires that the guide be parallel to the coronal and sagittal mechanical axes of the bone in order to cut the tibial slope precisely. While the position of the extramedullary guide is defined according to the tibial anatomical landmarks, there is no consensus point of view concerning the optimal reference for tibial sagittal alignment.7,8 An IM guide has direct access to tibial canal, with the goal being alignment of the guide rod parallel to the anatomical axis of tibia (AAT).9Although available literature shows no statistical difference in the accuracy of alignment of the tibial component when using the IM or extramedullary guides, the accuracy of either technique may be influenced by any intervening diaphyseal tibial deformity and the bowing of the diaphysis.10
The MA of tibia is defined as a weight-transmission vector from the knee to the ankle joint. The anatomic structures that define this axis are not visible to the surgeon during the operation. The surgeon must find a visible reference line to achieve the desired tibial alignment. The extramedullary rod may be positioned either parallel to the anterior tibial cortex (ATC) or to the fibular line (FL) in the sagittal plane, while the IM rod represents the AAT. This study was to explore these three reference lines above-mentioned, to determine which line is closest to parallel to the MA of tibia in the tibial sagittal plane, to determine the smallest angle between the tibial MA and each of three reference lines in the sagittal plane. Throughfinding a reliable landmark on the leg, surgeon may minimize posterior tibial slope measurement errors and improve the clinical technique of TKA.
METHODS
From January 2012 to July2012, we selected 85 consecutive patients with knee or hip osteoarthritis (OA) and evaluated 168 knees (78 normal and 90 OA). Two knees were excluded due to prior TKA surgery. Patients with a history of previous knee fracture, rheumatoid or Crowe III-IV of developmental dysplasia of the hip (DDH) were also excluded. The average patient age was (63.6±8.6) years. The average body height, weight, and body mass index was (159.2±8.2) cm, (66.15±10.88) kg, and (26.09
3.73)
kg/m2, respectively. Patientsdata of general description are seen in Table 1. All CT scans (Siemens Somatom Definition Dual Source 128 Slice CT, Germany) of whole tibia, including knee and ankle joint, were preoperatively obtained using a 1.0 mm section thickness for all subjects. DICOM fomat documents of images were digitally acquired and saved in disc and measurements were carried out using 3D imaging software (SuperImage1.26, Cybermed, Shanghai, China).
The following steps were used to conduct all measurements (Figure 1): The sagittal plane of tibia (SPT): Three points including the center of tibial proximal plateau, the center of tibial distal plafond and the medial 1/3 of the tibial tubercle to form a plane; MA: A straight line from the center of tibial proximal plateau to the center of tibial distal plafond. The centre of the proximal tibia was approximated at midpoint of the tibial spines using the coronal, sagittal and axial sections at the level of the articular surface; ATC: A line connecting the anterior border of the tibia at upper and lower 1/4 of tibia, respectively; FL: A line connecting the prominence of fibular head and lateral crest of distal fibula, which are easy tobe palpable on the leg; AAT: A line connecting the midpoints of the shaft at upper and lower 1/4 of tibia, respectively; medial tibial plateau line (MTPL): the line connecting the anterior and posterior edgeat the midportion of medial tibial plateau; Lateral tibial plateau line (LTPL): the line connecting the anterior and posterior edge at the midportion of lateral tibial plateau.
All above-mentioned lines were projected on SPT, and measure the angles with MA (Figures 2 and 3). aMT: an angle between MA and ATC; aMF: an angle between MA and FL; aMA: an angle between MA and AAT; Medial tibial slope (MTS): 90° minus the angle between MTPL and MA in sagittal plane; Lateral tibial slope (LTS): 90° minus the angle between LTPL and MA in sagittal plane.
All measurements were repeated three times by one orthopaedic surgeon at intervals of one week in order to assess the reliability of the measurements using the references mentioned above. Statistical analysis was performed with SAS version 9.2 (SAS Institute Inc, Cary, USA), and a P <0.05 in a 2-tailed test was considered as statistically significant. All continuous variables were presented as means ± standard deviation (SD), and all categorical variables were presented as numbers (propotion). Comparisons between groups were performed with or without an adjustment for a covariate, using analysis of variance (ANOVA) for continuous variables and Chi-square test for categorical variables, respectively. Paired samples t-test was used to compare the differences between the variables. Pearson correlation was used to test the association between the two variables and using one-way ANOVA to evaluate the reliability of data.
RESULTS
Measurement reliability was accepted (r, 0.822–0.905) for the all angles (Tables 2 and 3). Mean angles of 168 knees were as follows: aMT(3.96±0.85)º,aMF(0.70±0.58)º,
and
aMA(1.40±0.66)º, and the aMF was significantly smaller than the others (P<0.0001) (Table 2). The mean value of the medial tibial slope angle vs. the MA was (9.19±3.97, and this was significantly larger than the mean lateral slope angle of (6.62±4.23(P <0.0001) (Table 3). All angles were evaluated in 78 normal knees and 90 OA knees. The mean values of aMT, aMF, aMA in normal knees and OA knees are seen in Table 2. The values of aMT and aMA with OA were less than without OA, and differences were statistically significant (P <0.001). However, the difference between aMF without OA and with OA was not statistically significant (P=0.5015). The mean tibial posterior slope in the medial plateau was (9.41±3.55
(range:
2.1º16.7º) in the normal knees and (9.01±4.31
(range:
0.2º22.6º) in the OA knees. The mean tibial posterior slope in the lateral plateau was (6.35±4.41
(range: –5.1º– 16.9º), negativity means a posterior inclination with respect to the MA in the normal knees and (6.85±4.08(range: –0.5º23.5º) in the OA knees (Table 3). These differences were not statistically significant through age and BMI adjusted. The association between the aMT and aMA was strong (r=0.82, P <0.01).
DISCUSSION
In TKA, the position of tibial prosthesis in the sagittal plane is decided by the angle of posterior tibia slope. The position of the implant can affect the knee flexion stability, stress distribution on the polyethylene insert, and ultimately influences the knee function and survivorship.5,11 The ideal slope angle for a given tibial component depends on the requirements of the implant design and on the soft tissue balancing performed during the operation. Despite the importance of the proximal tibial cut, there is not consensus regarding the best anatomic reference axis for the posterior tibial slope.12 The tibial MA is an imaginary line from the midpoint of the tibial plateau to the mid-point of tibial distal plafond. Although it is not visible, tibial MA is not changed by the variation of the tibia anatomy. Therefore, we considered the MA of tibia might be an ideal reference for posterior tibial slope. In this study, the aMT, aMF, aMA were (3.96±0.85)º,
(
0.70±0.58)º, (1.40±0.66)º respectively in totality and the differences were statistically significant and the similar statistically significant differences were gotten in OA group. The present study demonstrates that the FL was closest parallel to the MA of tibia. Moreover, we suggest that tibial resection should be sloped approximately 4 degrees more posteriorly when using ATC or 12 degrees more using IM guide. For example, if PTS of 3 degrees referring to the MA of tibia is desired, the surgeon can use different cutting blocks (giving 3°–7° slope) combing with different anatomical references to reach, 3 degrees cutting block with EM alignment rod parallel to FL, 7 degrees cutting block with EM alignment rod parallel to ACT and 5 degrees cutting block with IM alignment rod parrallel to AAT.
This conclusion is different from a similar study by Han et al13 from Korea who found the aMT, aMF, aMA were (2.2±0.92)º, (2.1±0.98)º , (0.9±0.67)º respectively and concluded the AAT was the closest to the MA. Inconsistencies between the two studies maybe come from the definition of the line of the fibula which is defined from the center of fibular neck and lateral crest of distal fibula by Han; however, in this study, it is defined from the prominence of fibular head and lateral crest of distal fibula which are easy tobe palpable on the leg in surgery and clearly pointed out in three-dimensional model. In this study, we used the FL rather than the axis of the fibula because the FL is not same as the fibula axis which may be difficult to identify during a TKA procedure.
This study has used a non-OA group as a control group. The values of aMT and aMA with OA were less than without OA, and differences were statistically significant (P <0.001). However, the difference between aMF without OA and with OA was not statistically significant (P=0.5015). The findings suggestthat the FL as the reference of the tibial MA is more consistent than the ATC and AAT in OAof the knee. The association between the aMT and aMA was strong (r=0.82, P <0.01), suggesting that ATC and AAT changed with tibia shape in knee OA, while this was not the case with the FL. This phenomenon may be related to the tibia remodeling in OA of the knee. Although the fibula morphological changes occur in OA process of the knee, the FL is not changed in OA. For OA of the knee, many studies have focused on the pathological changes about lower limb alignment, articular cartilage, synovium, ligaments and so on. However, the morphological remodeling changes of the tibia and fibula during the OA process has been ignored by scholars for a long time which maybe a very interesting research direction and we will consider to further explore in future studies.
Chiu et al7 recommended that the anterior tibial cortex be used in the visual inspection and radiographic assessment as a reference for tibial cutting because surgeons commonly used the anterior tibial crest to align the cutting jig at the time of the operation. But he also proposed the difference between the use of extramedullary and IM alignment lines to determine the posterior tibial slope was found to be >3º. If the surgeon used the same tibial cutting jig for extramedullary and IM guides, there would be a 3º difference in the posterior slope. In other words, if the extramedullary guide was placed parallel to ATC, the tibial cut would be 3º less sloped posteriorly compared with when an IM guide was used. This result is consistent with our findings of this study that AAT and ATC are not parallel, approximately 2.5º–3º of angle on the SPT. However, we advise that the MA of tibia, not the ATC or AAT, is used as the reference. If an extramedullary alignment guide is used, the fibula serves as a reliable guide and the alignment rod must be drawn distally, away from the ankle, to parallel to the fibula line (Figure 4).
Despite many studies focused on the PTS, it is difficult to compare the results due to different references, different techniques, and different groups of people. Yoo et al14 found the average posterior slope was 10.6° (range, 1.9°–19.6°) with reference to the MA on radiograph. Chiu et al7 found the posterior slope of the medial plateau was 14.8º (range, 5°–25°), and the posterior slope of the lateral plateau was 11.8º (range, 4°–23°) on cadaveric tibial bones with reference to the ATC. Matsuda et al15showed the mean medial slope was 10.7º (range: 5º–15.5º) and the mean lateral slope 6º (range: 1º–13º) on MR images using the tibial proximal anatomical axis in normal knee. In this study, the mean tibial posterior slope in the medial plateau was (9.19±3.97)º (range: 0.2º22.6º) compared with (6.62±4.23)º (range: –5.1º23.5º) of the lateral plateau for the whole population and the difference was statistically significant. it is the consistent result of the different studies that the PTS in medial plateau was more inclined than that in lateral plateau and there was a large range in the posterior tibial slope in all groups.
There are some limitations of this study. Firstly, it is difficult to get normal group as a control without different age and BMI from OA group. Therefore, normal knee of the patients with hip OA were chosen into the study because of the patients without obvious lower alignment deformity and very close to the OA group in the age and BMI. Secondly, the data were obtained from Chinese population. There may be racial differences among and between Asian populations and Caucasian populations leading to subtle errors. However, the main contribution of the present study is to describe the relationship between ACT, FL, AAT and MA in the sagittal plane of tibia and to provide a scientific basis for clinical practice.
In summary, this study demonstrated that the FL was more parallel to the MA of tibia and more consistent than ATC and AAT. Moreover, given these findings we suggest that tibial resection should be sloped approximately four degrees more posteriorly when using ATC or 1-2 degrees more using an intramedullary tibial guide. The amount of posterior slope in medial plateau was more inclined than that in lateral plateau and there was a large range in the posterior tibial slope in knees with and without OA.
REFERENCES
  • 1.

    Callaghan JJ, O’rourke MR, Saleh KJ. Why knees fail: lessons learned. J Arthroplasty 2004; 19 (4 Suppl 1): S31-S34.

  • 2.

    Sharkey PF, Hozack WJ, Rothman RH, Shastri S, Jacoby SM. Insall Award paper. Why are total knee arthroplasties failing today? Clin Orthop Relat Res 2002; (404): 7-13.

  • 3.

    Vail TP, Lang JE. Surgical techniques and instrumentation in total knee arthroplasty. In: Scott WN, Scott WN, ed. Surgery of the knee. Philadelphia: Churchill Livingstone; 2006: 1455.

  • 4.

    Shao JJ, Wang Q, Yuan JB, Zhang XL. Surgical guidance system and computer navigation for simultaneous femoral osteotomy and total knee arthroplasty for treatment of osteoarthritis associated with severe extra-articular deformity. Chin Med J 2012; 125: 4145-4148.

  • 5.

    Jojima H, Whiteside LA, Ogata K. Effect of tibial slope or posterior cruciate ligament release on knee kinematics. Clin Orthop Relat Res 2004; (426): 194-198.

  • 6.

    Whiteside LA, Amador DD. The effect of posterior tibial slope on knee stability after Ortholoc total knee arthroplasty. J Arthroplasty 1988; 3 Suppl: S51-S57.

  • 7.

    Chiu KY, Zhang SD, Zhang GH. Posterior slope of tibial plateau in Chinese. J Arthroplasty 2000; 15: 224-227.

  • 8.

    Hungerford DS. Alignment in total knee replacement. Instructional course lectures 1995; 44: 455-468.

  • 9.

    Shakespeare D. Conventional instruments in total knee replacement: what should we do with them? Knee 2006; 13: 1-6.

  • 10.

    Rottman SJ, Dvorkin M, Gold D. Extramedullary versus intramedullary tibial alignment guides for total knee arthroplasty. Orthopedics 2005; 28: 1445-1448.

  • 11.

    Bai B, Baez J, Testa N, Kummer FJ. Effect of posterior cut angle on tibial component loading. J Arthroplasty 2000; 15: 916-920.

  • 12.

    Bae DK, Song SJ, Yoon KH, Noh JH, Moon SC. Comparative study of tibial posterior slope angle following cruciate-retaining total knee arthroplasty using one of three implants. Int Orthop 2012; 36: 755-760.

  • 13.

    Han HS, Chang CB, Seong SC, Lee S, Lee MC. Evaluation of anatomic references for tibial sagittal alignment in total knee arthroplasty. Knee Surg Sports Traumatol Arthrosc 2008; 16: 373-377.

  • 14.

    Yoo JH, Chang CB, Shin KS, Seong SC, Kim TK. Yoo JH, et al. Anatomical references to assess the posterior tibial slope in total knee arthroplasty: a comparison of 5 anatomical axes. J Arthroplasty 2008; 23: 586-892.

  • 15.

    Matsuda S, Miura H, Nagamine R, Urabe K, Ikenoue T, Okazaki K, et al. Posterior tibial slope in the normal and varus knee. Am J Knee Surg 1999; 12: 165-168.

(Received July 10, 2013)
Edited by GUO Li-shao
Table 1. Patients data of general description
Variables
Total
(n=170)
OA status (n=168)*
Without OA
(n=78)
With OA
(n=90)
P values
Male (n (%)) 54 (31.76) 29 (37.18) 24 (26.67) 0.1436
Age (years) 63.6±8.6 60.3±9.8 66.3±6.3 <0.01
BMI (kg/m2) 26.09±3.72 25.05±3.54 27.19±3.61 <0.01
Data were mean±SD or numbers (percentage). *Two participants without OA status were excluded from this analysis.
Figure 1. Three points formed a sagittal plane of tibia (SPT) (A). There were anatomical points used for all reference axes (BD), including the center of tibial proximal plateau (M1), the center of tibial distal plafond (M2), the anterior border of the tibia at upper 1/4 (T1) and lower 1/4 of tibia (T2), the prominence of fibular head (F1) and lateral crest of distal fibula (F2), the midpoints of the shaft at upper 1/4 (A1) and lower 1/4 of tibia (A2).
Figure 2.The angle between the sagittal MA and 3 anatomical references was measured (2A): aMT, an angle between MA and ATC (2B), aMF, an angle between MA and FL (2C) and aMA-an angle between MA and AAT (2D).
Figure 3.Medial tibial slope (3A) and lateral tibial slope (3B) were measured at the midportion of each tibial plateau.
Figure 4. If an extramedullary alignment guide is used, the fibula serves as a reliable guide to parellel to the MA of tibia and the alignment rod must be drawn distally, away from the ankle, to parallel to the fibula line.