1 Introduction

Knee osteoarthritis (OA) is the most common form of arthritis, affecting around 30 % of adults over 60.¹ The causes of knee OA have traditionally been divided into biological and mechanical factors, with lower limb malalignment representing one of the mechanical factors.² Total knee arthroplasty (TKA) is the widely accepted treatment for severe cases of OA, aimed to decrease pain, improve ROM, and enhance quality of life, with an 85 %–90 % satisfaction rate. Despite these facts, a high dissatisfaction rate of 15 %–25 % is reported in TKA patients.³

Implant positioning and limb alignment are the key factors for patient satisfaction and durability in TKA.⁴ Alignment parameters based on two-dimensional (2D) coronal radiography, such as the Mikulicz line and femorotibial angle (FTA), have been widely used to diagnose and plan, including preoperative and postoperative assessments in knee osteoarthritis.⁵,⁶ Various alignment philosophies have evolved over the years to improve patient satisfaction rates, and recently, personalized (functional) alignment has become popular to restore patients' pre-arthritic or constitutional knee alignment.⁷ Based on this constitutional alignment of the lower limb by hip–knee–ankle angle (HKA) and the joint line respective to the floor by joint line obliquity (JLO), MacDessi et al. proposed Coronal Plane Alignment of the Knee (CPAK) classification system, which categorizes knees into nine coronal plane phenotypes using the lateral distal femoral angle (LDFA) and medial proximal tibial angle (MPTA) obtained from 2D radiographs. We can use this classification as an indicator for adjusting lower limb alignment to physiological alignment during total knee arthroplasty (TKA). CPAK has increasingly gained acceptance as a well-defined, simple, and universal classification system of the knee.⁸

Previously, Victor et al. focused on joint line orientation relative to the floor in a standing position, showing that joint line orientations of the normal and constitutional varus lower limb alignment were parallel to the floor, whereas varus alignment with advanced medial knee OA exhibited divergence of the joint line orientation from parallel to the floor.⁹ However, these studies were based on 2D coronal radiography, in which tibial rotation and knee flexion angle may be difficult to control, particularly for patients with knee OA, and are not considered. Thus, the effects of tibial rotation and knee flexion angle on 2D coronal joint line orientation remain unknown.¹⁰

Moreover, the sagittal aspect is not considered, in which the posterior tibial slope (PTS) plays a crucial role in the biomechanics and cruciate ligament tension in the normal and the prosthetic knee. In knee arthroplasty, PTS remarkably influences the range of motion (ROM) and flexion stability, especially in cruciate retaining designs, by altering the posterior cruciate ligament tension, adversely affecting the femoral rollback.¹¹,¹² Its potential role in the subsidence and tibial component loosening has also been reported.¹³,¹⁴ It is also suggested that the negative tibial slope affects patellar instability due to pseudo patella alta.¹⁵ It should be noted that PTS is a 2D measurement of inclination in the sagittal plane; subsequently, tibial rotation considerably affects its measurement in the X-ray.¹¹ True lateral projections of X-rays are critical, and rotations in radiographs shall result in assessment errors as well as poor reproducibility, as they may appear superimposed,¹⁶ making the 2-dimensional radiographs unreliable.¹⁷,¹⁸

In addition, an X-ray taken in a standing position would reflect the joint line angles as a projection in the coronal plane rather than a cross-sectional image.¹⁹ While CT can give a cross-sectional image with a 3D rotational profile, as the rotation can be controlled using 3D software tools, making it a superior modality²⁰ to measure the PTS, providing good reproducibility as well.²¹ This review aims to critically elucidate the limitations of the CPAK classification system, emphasize the importance of three-dimensional assessment of joint line orientation and limb alignment, describe current technologies for measuring 3D joint line parameters, and highlight the future clinical need for upright 3D-CT imaging in osteoarthritis evaluation and total knee arthroplasty planning.

2 CPAK classification and alignment

Due to the common belief since ages that, mechanical alignment (MA) provides the best biomechanics for the knee and for the prosthetic durability which invariably aligns all the knees to neutral axis irrespective of the individual's native alignment, MacDessi et al. proposed a new classification system for the coronal plane alignment of the knee (CPAK) to give precedence to patient's (prearthritic) constitutional alignment which results in better soft tissue balance²²,²³ With more natural movements of the knee and improved patient satisfaction than mechanical alignment.²⁴,²⁵ The nine observed traits of the CPAK classification are based on medial proximal tibial angle (MPTA), Lateral distal femoral angle (LDFA), and joint line obliquity (JLO), which are coronal parameters obtained from 2D radiographs ⁸, completely disregarding the sagittal plane parameters. Furthermore, the 2D radiograph is a projection image in a given position in a given plane (uniplanar), and it does not necessarily reflect the actual joint orientation, given that it is unable to directly visualize OA-associated damage in articular and periarticular non-osseous joint structures.²⁶ and it is subject to variations by many factors, including tibial rotation, knee flexion, and technical errors in getting an accurate view.²⁷

2.1

2.1 The sagittal plane and its parameters

Posterior tibial slope (PTS) is the primary determinant of the sagittal plane alignment of the knee, which is typically ignored in the assessment and execution of the more popular personalized alignment TKA, prioritising the coronal plane.²⁸ PTS have a significant impact on the biomechanics of the normal knee and in TKA.¹¹ Increased PTS causes increased anterior tibial translation (ATT) affecting the strain on the cruciates in the normal knee and overloads the collateral ligaments in TKA, altering the contact areas, creating abnormal forces in the implant bone interface, causing excessive wear and potential instability.²⁹,³⁰ Similarly, a decrease in PTS may affect the femoral rollback in cruciate retaining TKAs^12,31 and also compromise knee stability by impairing the PCL's ability to maintain posterior instability.³²,³³

The increase in the difference between the medial and lateral PTS contributes significantly to tibial torsion. Especially when the lateral PTS is more than the medial PTS, there is more anterior translation of the lateral tibia (Fig. 1), eventually resulting in torsion, making lateral PTS a more significant risk factor for torsional injuries³⁴,³⁵ and PTS as the primary contributor of tibial internal rotation and large ATT.³⁶,³⁷ Specifically, with every 10° increase in PTS, there is a 6 mm translation of the tibia anteriorly, increasing the ACL strain in the normal knee and excess contact pressure and wear in the prosthetic knee.³⁶ Recently, it's been found that the failure in kinematic alignment TKA is due to increased slope of the tibial component and subsequent subsidence.³⁸,³⁹

Close

Fig. 1 Illustrated axial view of the tibia depicting the mechanics of the native posterior tibial slope (PTS) ∗ATT- Anterior tibial translation.

Export to PPT

The impact of reduced PTS is equally significant as it elevates the patella, resulting in pseudo patella alta¹⁵ and warrants consideration as it leads to patellar maltracking, as the higher riding patella engages less effectively within the trochlear groove during extension, increasing susceptibility to subluxation or dislocation and eventual instability. Moreover, patellar maltracking generates increased stress on the patellofemoral joint, which ultimately results in pain and subsequent dysfunction.⁴⁰,⁴¹ It is also essential to observe that, unlike coronal alignment, the sagittal parameters are not affected by weight-bearing positions or by soft tissue and are therefore considered to remain unchanged. The changes incurred during surgery are unforgivable.²⁸ To comprehend the scale, it is worth noting that reducing PTS from 15° to 0° reclassified nearly 45 % of normal subjects into ‘patella alta.⁴² Thus, it becomes imperative to equally emphasize and evaluate the sagittal parameters in lower limb alignment to assess and plan osteoarthritis and TKA.

3 Evolution of 3D-CT

Since its inception, computed tomography (CT) has enabled cross sectional imaging of the entire body, including soft tissues, typically with the patient in supine position and because supine position is helpful for the evaluation of organic diseases, such as infectious diseases, cancer, and atherosclerotic disease but for the evaluation of many functional diseases cross sectional imaging in standing position is necessary. Hence, there were plans to develop this modality in the standing position; however, it was not amenable due to challenges in maintaining the patients in the standing position for prolonged imaging periods, which also produced motion artifacts and increased exposure to radiation at that time.⁵² Recently, Fujii et al. demonstrated three-dimensional alignment changes in limbs affected by osteoarthritis of the knee in the standing position and suggested that it is essential to evaluate 3D whole leg alignment under weight-bearing in assessing and planning treatment for OA⁵³,⁵² Currently, 0.5-s scanning per rotation is clinically available in CT, which enables 3-dimensional imaging of the entire torso in less than 15 s using 64-row or 80-row detector CT. In addition, with the reemergence of reconstruction techniques in 2009, the radiation dose necessary for CT has dramatically decreased.⁵⁴ Thus, it is reported that the effective dose estimate for the body trunk was 9.3 ± 2.2 mSv for supine CT and 8.9 ± 2.0 mSv for upright CT.

Cone beam CT (CBCT): can provide cross-sectional images of joints with superior test-retest reliability for the three-dimensional (3D) measurement of joint space width in the standing (weight bearing) OA knee, with intraclass correlation coefficients of 0.95–0.97 and 0.90–0.97 for the lateral and medial compartment, respectively.⁵⁵,⁵⁶ CBCT uses a pyramid-shaped X-ray beam with a flat panel detector that rotates 216.5° around the patient, capturing volumetric data in a single sweep, and multiple low-dose X-ray projections are acquired and reconstructed using advanced algorithms into a high-resolution 3D image. In knee imaging, both non-weight-bearing and weight-bearing scans can be performed, and the entire scan is completed in seconds, minimizing motion artifacts and improving patient comfort. The major advantages of CBCT is that it is less prone to metal artifacts and delivers low radiation dose with high-resolution bone visualization.²⁶ However, the drawbacks include less soft tissue information, which is insufficient due to its low contrast⁵⁷,⁵⁸ limiting its use to high contrast tissues such as dentition and extremities.^59,60 The scan range is often limited due to narrow bore size, making the early changes associated with OA less discernible even in 3D. Also, Hirschmann et al. reported that the fields used in CBCT were not sufficiently wide to scan the entire leg or evaluate the lower limb alignment while standing, consequently limiting its use in the 3D analysis of the tibiofemoral geometry.⁶¹

4 Upright CT

As the use of CT has progressed significantly, resulting in shorter scanning times with less radiation dose, an upright CT scanner has been developed to enable whole body cross-sectional imaging in less than 15 s to examine the effects of weight bearing and gravity. Specialized support structures are employed to hold the frail and elderly, like acrylic walls encircling the body, including poles and velcro to fasten the body. Upright CT scan can reveal the alignment and deformity of the entire lower limb, which are closely related to the symptoms in a patient with osteoarthritis of the knee, and enabled objective diagnosis and grading in a manner not apparent when using conventional X-rays and CTs in high spatial resolution.¹⁰

Recently, tibial alignment was shown to exert a greater influence on implant survival than femoral alignment,⁶² and coronal outliers are more frequent in tibia than femur, due to inconsistency in identifying the tibial geometry in axial, coronal, and midsagittal planes⁶³,⁶⁴ using proximal antero-posterior (AP) and the distal transmalleolar axis (TMA). Identifying the midsagittal plane based on TMA is difficult since the malleolus are difficult to palpate in obese and moreover, the TMA rotates externally, so, we refer the midsagittal plane indirectly in accordance with the 2nd metatarsus, which can be affected by foot deformities.⁶⁵

5 Midsagittal plane tibial coordinate

Using upright 3D CT, it is possible to get the midsagittal plane coordinate applying highly consistent bony landmarks that are relatively unaffected by osteoarthritis as defined by Enomoto et al.⁶⁵, Two bony landmarks were used as proximal references to designate the AP axis, the center of the posterior intercondylar facet proximal to the enthesis of the posterior cruciate ligament (PCL) and the proximal medial edge of the tibial tubercle (see Fig. 2). The apex of the tibial plafond, which was defined as the ‘midpoint of the plafond axis’, where the plafond axis (PLA) is the line connecting each midpoint of the medial and lateral talocrural facet margins, was selected as a distal reference point. Thus, the mid-sagittal plane tibial coordinate was determined by straight line from the midpoint of the medial and lateral tibial eminences (the origin) to the apex of the tibial plafond represented as y-axis and the normal vector of the sagittal plane (mediolateral plane) from ‘the origin’ was assigned the x-axis and the final anteroposterior z-axis is determined as the cross product of the y-axis and the x-axis as shown in Fig. 2.

Close

Fig. 2 Two proximal bony landmarks, (A) the center of the posterior intercondylar facet near the PCL enthesis and (B) the proximal medial edge of the tibial tubercle, were used to define the AP axis. The apex of the tibial plafond (C), the midpoint of the line connecting the medial and lateral talocrural facet margins (D, plafond axis), served as the distal reference. The mid-sagittal plane (E1) was set from the midpoint of the tibial eminences (the origin) to the plafond apex (y-axis). The x-axis was the mediolateral normal vector (E2) to this plane, and the z-axis (E2, AP direction) was defined as the cross product of the y- and x-axes.

Export to PPT

Posterior tibial slope (PTS) measurement: Precise PTS measurements is challenging due to high interobserver variability and complex anatomical shapes of tibial plateau that hardly resemble a plane with a single slope,⁶⁶ moreover measurements based on lateral images report tibial slopes ranging from 3° anterior to 25° posterior⁶⁷,⁶⁸ The variability in measuring the posterior tibial slope (PTS) on lateral images arises primarily from the challenge of accurately establishing the tibial axis.⁶⁹ In 3D CT, this was achieved by aligning the tibia in the anatomical axis both in the sagittal and coronal planes by defining the centre of the ankle and the centre of the proximal tibia, its axis being a line that connected these two points (Fig. 3). The centre of the ankle was located by a best fit sphere algorithm to the combined articular surface of the tibial plafond and the inner surfaces of the medial and lateral malleoli of the ankle. The center of the proximal tibia was identified by plotting a series of center points in a 6 cm shaft segment distal to the tibial tubercle. The points are connected and projected proximally to the surface of the tibial plateau of the knee as a single point, which defines the centre of the proximal tibia as shown in Fig. 3B. Then, the tibia was rotationally aligned to the tibial plateau using the tibial centroid axis (TCAx), and PTS measurement was made. The TCAx was determined by marking the medial and lateral surfaces of the tibial plateau using a virtual brush, and the software calculates the respective centroids (geometrical center). The TCAx is a line connecting these 2 centroids (Fig. 3C1 and C2), and the tibia is rotationally aligned by ensuring the TCAx is parallel to the medio-lateral axis in the axial view. The tibial slope was derived from planes which are created by best-fit algorithm to the combined marked surfaces of the medial and lateral tibial plateaus (Fig. 3D). The medial and lateral PTSs (MPTS and LPTS) were also individually derived using the same method from individually marked surfaces of the tibial plateau. Rotational tibial alignment using TCAx yields more accurate medial and lateral PTS measurements than using transmalleolar axis (TMAx), with statistically significant results.¹¹

Close

Fig. 3 Schematic representation of the tibia showing the method of measuring tibial slope from the anatomical axis. A determination of ‘center of ankle’ by best fit sphere method (through combined articular surface of the tibial plafond and the inner surfaces of the medial and lateral malleoli), B the center of proximal tibia, identified by plotting a series of center points in a 6 cm shaft segment distal to the tibial tubercle and by connecting the points to project it proximally to the surface of the tibial plateau. C showing the method of rotational alignment of the tibia using tibial centroid axis (TCAx) C1 by aligning it parallel to the medial-lateral axis of the tibia (C2) in axial view. D shows the PTS measurement by deriving the best fit plane with the combined marked surfaces of the medial and lateral tibial plateau and measuring the angle between the plane and the anatomical (y) axis.

Export to PPT

Patellofemoral alignment: Assessment of patellar tracking and quantifying the patellar indices (patellofemoral bisect offset, lateral offset, congruence angle, lateral patellofemoral angle, and patellar tilt) enables surgeons to anticipate intraoperative malpositioning and tracking issues that helps to determine the need for component adjustment and soft tissue balancing.⁷⁰,⁷¹ This is critical in decision-making as it can prevent complications like anterior knee pain, lateral subluxation, and early revision rates, which might occur due to component (femur/tibia) malrotation and consequent increase in patellofemoral pressure.⁷²,⁷³ X-rays and CT-scans are commonly employed for these assessments; however, static imaging often falls short in capturing the dynamic nature of patellar tracking, hindering precise assessment and diagnosis.⁴² In order to better assess patellofemoral indices, dynamic imaging techniques, such as dynamic computed tomography (CT) procedures, have been used to overcome these diagnostic limitations,⁴² and especially in varying positions of flexion, Williams et al.³² noted that the patellar indices including tibial tuberosity-trochlear groove (TT-TG) distance were highest in positions closer to extension and Tanaka et al.³³ stated that those patellar tracking types with more than two quadrants of lateral translation strongly associated with instability symptoms. Also, recently it's been suggested that maximum quadriceps contraction during dynamic CT may reveal even the subtle abnormalities, possibly by replicating the dynamic forces during functional activities.⁴² In this scenario, Upright weightbearing 3D-CT reflects true patellar tracking under muscle tone and axial load providing improved biomechanical realism, particularly decreasing the patellar tilt and TT-TG distance on comparing supine CT (from 15.6° supine to 12.5° upright and 13.8 mm in supine CT to about 10.5 mm in upright respectively) demonstrating more realistic personalized alignment and maltracking of patella.⁶¹

5.1

5.1 Intrinsic tibial torsion/rotation

Torsion or rotation deformity of the tibia can influence the position of the tibial component and might result in malalignment during TKA. To avoid this malalignment, precise measurement of the intrinsic tibial rotation stays pivotal, since these angles alter in association with osteoarthritic deformity. The proximal mediolateral axis (x-axis in Fig. 3C2) and the posterior tibial condylar axis (PCA), which is defined as the posterior tangential line of the cross-sectional area at a level 10 mm below the lateral tibial condylar joint surface (Fig. 4), are used as proximal tibial reference axes. The transmalleolar axis (TMA), connecting the apex of both the malleoli and the plafond axis (PLA) as shown in Fig. 2D, are used as distal tibial reference axes. The torsional angles between the proximal and any of the distal reference axes represent the intrinsic torsional angle of the tibia.

Close

Fig. 4 Illustrated view of the posterior tibia with section taken at 10 mm below the lateral tibial condylar joint surface with a line tangential to the posterior cortex depicting the posterior tibial condylar axis (PCA).

Export to PPT

Since significant correlation of the torsional angle based on the PLA with that of the TMA was confirmed, the PLA can be considered to be as clinically relevant axis as the TMA. Despite its significant correlation, the discrepancy of the measured angles with reference to the PLA and the TMA ranged from 2.0° to 19.9° among the patients. Thus, the PLA is considered to imply its individualistic role, which might be independent of the TMA, and is worth being an additional reference for distal tibia.⁶⁵

6 Tibial internal rotation

Weight bearing causes changes in the 3D alignment of the osteoarthritic knee, and eventually, greater 3D deformities are seen in cases of end-stage OA knees. The most important finding by Kellegren and Lawrence et al. was that weight bearing produces more tibial internal rotation in relation to the femur, which is appreciable on comparing KL grade 1 and 2 OA knees.⁴³ In 2D radiograph, the KL grade 2 OA knee shows only subtle joint space narrowing with minimal osteophyte formation, while weight-bearing CT clearly detects early change occurring in the knee joint, which is increased internal rotation with weight-bearing.⁵² Similarly, Hirschmann et al.⁶¹ demonstrated that the tibial internal rotation increased in the weight-bearing knee by using CBCT, where they studied both OA knees and also normal knees. Moreover, the supine position only demonstrated joint flexion and varus in the high KL grade of OA on comparing low KL grade, but the standing position, in addition, demonstrates clear internal rotation in the high KL grade, where the tibia was originally externally rotated in the low KL grade of OA.

The tibial rotations and their difference in CT examination between supine and standing positions are calculated in relation to the pelvis. The angle between the perpendicular of the line connecting both the anterior superior iliac spines (ASIS) of the pelvis and the AP axis (Z-axis) of the tibia represents the tibial torsion or rotation angle (Fig. 5).⁴³

Close

Fig. 5 Illustrated image showing the method of measurement of tibial rotation based on the pelvis by A the angle between the perpendicular of the line connecting both the anterior superior iliac spines (ASIS) of the pelvis, and B the Z axis of the tibia.

Export to PPT

7 Tibial alignment

Tibial alignment was found to have a more significant impact on implant survival than femoral alignment.⁶² The incidence of coronal malpositioning was found to be significantly higher in the tibia compared to the femur, particularly due to the usage of extramedullary alignment guide for proximal tibial osteotomy in TKA, rather than an intramedullary guide.^63,64 This is mainly due to the usage of indirect references for the tibial coordinate axis, like second metatarsus⁶⁵ and tendons of extensor hallucis longus and tibialis anterior⁷⁴,⁷⁵ which are not consistently reliable and discernible, especially in the obese patients. Given the precise depiction of the axes in the 3D CT, specifically in upright CT, the incidence of tibial component malpositioning can be substantially decreased. Using the Euler rotation sequence of the anatomical axis of the proximal bone, the alignment of the tibia relative to the femur can be quantified in three axes X, Y, and Z (Fig. 2E2), where it represents varus/valgus, flexion/extension, and external/internal rotations, respectively.⁶⁵

8 Measurement of 3D joint surface orientation

The three-dimensional joint surface measured by CT is a complete depiction and a parameter that is unaltered due to the innate changes of the limb in osteoarthritic disease, such that its orientation relative to the floor is also unchangeable, particularly on weight bearing,¹⁹ validating the need and significance of upright CT in the evaluation of limb alignment for TKA. Upright CT images were acquired from the pelvis to the ankle and ground; the patients were made to stand with feet shoulder-width apart, which prevents rotations in the lower limb. Patients were advised only to extend the knees as much as possible, and the data of the entire femur, tibia, including the floor, were extracted from CT DICOM data using 3D visualization software (AVIZO 6.4; Thermo Fisher Scientific, Tokyo, Japan).

The 3D joint surface orientation in relation to the floor is calculated by first establishing a best-fit plane for the tibial plateau (as shown in Fig. 3D, similar to the measurement of PTS). The tibial plateau was marked, excluding the spurs and both intercondylar eminences, ensuring the complete marking of the articular area. From this marked articular area, a best-fit plane was created using the best-fit algorithm. The joint vector (joint surface angle) is calculated as the angle between this best-fit plane and a vertical line. The floor vector is the angle between the plane of the floor and a vertical line, calculated using 3D surface floor data (Fig. 6). Ultimately, the 3D angle between the joint and the floor vectors was determined using the “inner product” and defined as the 3D joint surface-floor angle. The “inner (dot) product” is used to determine the similarity or relationship between two geometrically aligned vectors in a three-dimensional space (in this case, ‘the joint vector’ and ‘the floor vector’) expressed as a single (scalar) variable. It can be extended to any number of vectors in three or more dimensions. Thus, the determined angle (3D joint surface floor angle) becomes unchangeable and scalar by virtue of the inner product.⁷⁶

Close

Fig. 6 Illustrated image of the tibia showing the best-fit plane created by marking the medial and lateral centroids of the tibial plateau (inset). The joint surface angle (Joint vector) is the angle between the best-fit plane and the y-axis. The floor vector is the angle between the plane of the floor and the y-axis.

Export to PPT

9 Discussion

The current scientific discussions and trends are inclined towards personalized alignment of the knee in TKA to improve patient satisfaction using minimal soft tissue release,⁷⁷ yet the evaluation of the knee endures with primitive methods, making use of 2D radiographs, including Mac Dessi in the classification of coronal plane alignment of the knee²² (CPAK). Though the CPAK classification is well received, it is obvious and clear that it fails to account the sagittal plane and its parameters,¹⁹ and the 2D radiograph nor describes the cross-sectional characteristics of the knee,¹⁰ besides causing inaccurate measurements and observer variability due to deformities involving tibial rotation and knee flexion in osteoarthritis that are non-manipulable^16. Biomechanically, a recent study by Hess et al. found that among the CPAK phenotypes, rotational alignment variations can be similar in two or more types and suggested an extended classification of CPAK including rotational alignment parameters.⁷⁸ Moreover, Graichen et al. reported that not all knees maintain an equal gap mediolaterally throughout the entire range of motion, and this gap is not the same in extension and flexion. They described that it is due to the presence of laxity in the surrounding soft tissues and may vary “within varus and valgus” and stated there is “no one varus or one valgus among knees” which again warrants further consideration of laxity patterns⁷⁹,⁸⁰ challenging the framework of CPAK classification.

KR Kraus et al. in their review of 2427 primary TKAs stated that maintaining a patient's native coronal alignment (based on CPAK) postoperatively did not yield better patient-reported outcome measures (PROM).⁸¹ Similarly, Bertugli et al. in their study of 201 robotic-assisted primary TKAs demonstrated that there is no statistically significant association in mean satisfaction between patients changing and maintaining their CPAK classification preoperatively to postoperatively, measured using Likert 5-level scale and Forgotten Joint Score-12(FJS-12).⁸²

Variable outcomes have been widely reported regarding the impact of CPAK on results following TKA. Leaving knees in residual varus was not associated with improved outcomes or pain, as stated by Meneghini et al.⁸³ while Ritter et al. found that the TKA knees should be aligned in neutral or minimally in valgus for better durability.⁸⁴ In a study comparing 237 TKAs performed using personalized alignment (FA, n = 121) and Mechanical alignment (MA, n = 116) principles, the overall clinical outcomes, PROMs and FJS-12 were found to be similar at the end of 2 years, yet a greater percentage of FA were satisfied at 1 year, 87 % versus 81 % and pronounced to “recommend” the procedure at the end of 2 years (94 % versus 82 %), however, the clinical significance of these differences remains unclear.⁸⁵ Comparably, Paratte et al. compared 40 FA to 40 MA TKAs and found no difference in outcome scores at 1 year, besides demonstrating slight faster recovery.⁸⁶Farooq et al., by studying 1311 TKAs, reported that MCIDs for PROMs were achieved more often by maintaining the preoperative alignment in coronal plane in addition to maintaining native posterior tibial slope and incorporating a moderate femoral component flexion in the sagittal plane.⁸⁷

Thus, it becomes indispensable to progress towards the evaluation of the limb in coronal, sagittal, and axial planes to better comprehend the alignment with the respective parameters in the planning of TKA. The 2D radiograph in this regard is suboptimal and inconsistent,⁶⁶ while 3D CT becomes convenient and versatile, especially in the measurements of tibial slope,¹¹ patellar alignment with tracking (sagittal)^42, and tibial torsion and rotation⁶⁵ (axial), besides coronal parameters. Although 3D CT in supine position is excellent in visualizing the three-dimensional joint structure and cross-sectional area, it lacks information about the functional joint orientation and limb alignment, where upright CT becomes beneficial. The effect of weight bearing and gravity plays a vital role in internally rotating the tibia on the femur in osteoarthritis, and it is proportional to the grade of involvement,⁵² which is possibly because on weight bearing, the surfaces (of the femur and tibia) come in contact and gets aligned based on the sectoral congruency of the arthritic surfaces in the maximally stable position (Fig. 7A and A1) tensioning the ligaments, and when the gravity is eliminated the relaxed joint makes it to realign in a position where the ligaments are naturally relaxed (Fig. 7B and B1) but not necessarily opposed in congruency.

Close

Fig. 7 Schematic representation of the effect of weight bearing and gravity in the tibia with respect to femur in the propagation of osteoarthritis. A, On weight bearing the femur and tibial surfaces comes in contact and gets aligned based on the congruency of the arthritic knee(A1), that in varus produces internal rotation in the tibia, creating a moment of torque in the adjacent ligaments (shown in black circles i & ii, tensioning the ligaments) and B when the gravity is eliminated the relaxed joint realigns in a position where the ligaments are naturally relaxed (iii & iv) but the joint surfaces may not necessarily be opposed in congruency (B1). [ NAL – Neutral axis of ligaments, Tm - Moment of torque (created due to weight bearing and gravity), NT – Neutral axis of tibia, V – Vertical, AT - Articular surface of tibia, AF – Articular surface of femur].

Export to PPT

It must be remembered that, with the ability to precisely adjudge the orientation of deformity in the limb using upright CT, optimal alignment must be achieved for a biomechanically stable and better patient satisfaction, including long-term prosthetic survivorship.¹⁹ Numerous alignment strategies evolved, emphasizing improved patient satisfaction, shifting away from the gold standard and time-tested mechanical alignment. It should be agreed upon that there is large variability in patients’ anatomy and every knee is unique with its bone, soft tissues, and biomechanics, which influences the surgical technique, thereby favoring personalized alignment for individual patients. Though many schemes were set to discussion including kinematic alignment, restricted kinematic alignment, inverse restricted kinematic alignment and Personalized alignment, the long-term efficacy of these alternative techniques remains uncertain Similarly, the coronal plane alignment of the knee by Mac Dessi, that builds on personalized alignment though novel, it is recognized widely for its systematic phenotyping in establishing the native knee alignment. Given the lapse and inherent limitations in CPAK, which are apparent, it should be highlighted that improved PROMs in TKA are multifactorial^78, including accurate bone cuts, component design, component positioning in all three planes, soft tissue balance, and patient expectations. The surgeon attempting to improve patient satisfaction with Personalized alignment and CPAK must realize that the surgical technique is only a single factor influencing patient satisfaction and should not compromise on longevity. Thus, continued research is needed in establishing a thorough typology involving three-dimensional parameters, and until there is established evidence in long term and compelling rationale to change, mechanical alignment should remain the universal choice for all extremes of knee deformity for consistently proving effective and successful for more than 40 years.⁷⁷

10 Conclusion

The Coronal Plane Alignment of the Knee (CPAK) classification provides a framework for understanding alignment but may oversimplify the complexities of knee anatomy. Incorporating three-dimensional (3D) computed tomography into clinical practice can enhance diagnostic accuracy and better account for factors like tibial rotation and knee flexion. Ultimately, adopting 3D imaging techniques may lead to improved surgical outcomes and higher patient satisfaction following TKA. Though the future of TKA lies in personalized alignment driven by robust 3D-based classification systems, given the current state of knowledge, it is advisable to rely on mechanical alignment, a time-tested and reliable strategy, pending further validation.

Ethical statement

Not applicable for this article.

Author statement

Contribution by authors: Ram Sudhan, Gopinathan P, Sharat Balemane, and Sibin Surendran devised the protocol for the study. Raju Vaishya and Gopinathan P contributed the idea and patient identification with literature search. Sharat Balemane and Ram Sudhan S designed and drafted the figures. Raju Vaishya and Ram Sudhan S prepared the initial manuscript draft. Sibin Surendran and Raju Balemane helped with the language and did the proof reading of the article. All authors contributed to, reviewed, and approved the final draft of the paper.

Guardian/patient’s consent

Not applicable for this article.

Funding statement

This original research or any of its authors were not been funded by any sponsors or corporates.