Optimal component alignment is crucial for successful outcome of primary total knee arthroplasty (TKA). Femoral component rotation is one of the most important factors in TKA, as rotational misalignment affects the flexion stability as well as tibiofemoral and patellofemoral kinematics. Flexion malalignment is a known cause of pain, stiffness, patellar and flexion instability 1-8.
In the classical mechanical alignment concept, the femur must be implanted parallel to the surgical transepicondylar axis (STEA) 9. The surgical TEA is thought to best approximate the flexion/extension axis of the knee, however it can be difficult to palpate and reference intraoperatively 10-14. The surgical epicondylar axis and posterior condylar axis form the posterior condylar angle (PCA), which is on average of 3 degrees of external rotation. Although many factors such as gender, condylar hypoplasia, and coronal alignment can disturb the rotation of the distal femur and change the angle between the posterior condylar line (PCL) and transepicondylar axis. Many studies have demonstrated that the angle between PCL and surgical TEA may range from 3° of internal rotation to 10° of external rotation 13, 33, 34, 64.
Aglietti et al. studied preoperative knee CT scans, and developed a simplified formula for the relation between distal femur rotation and frontal alignment of the knee which increases the PCA by 1° per 10° of coronal deformity increments from varus to valgus, resulting for instance in 2° external rotation for a 20° varus knee and 5° external rotation for a 20° valgus knee. Based on this concept and without the use of the preoperative CT scan, rotational accuracy is within ± 2° of TEA in 80% of the cases 65. We can accept, that standard external rotation for about 80% of the knees are 3° for varus knees and 5° for valgus knees 21.
The rationale for the TEA method is derived from the observation that the normal tibial joint line is between 3° and 5° of varus relative to the long axis of the tibia. If the tibial resection is made in 3° of varus, an equal symmetrical posterior condylar resection will result in a rectangular flexion gap. If the tibial resection is 90° to the long axis of the tibia, 3° to 5° of external rotation will be necessary in order to recreate a rectangular gap 29.
Determination of the surgical TEA is known to be difficult; therefore most systems use the default posterior reference and PCL landmark to determine the femoral component ER. Hungerford 15 introduced the concept of 3 degrees of external rotation (ER) relatively to the posterior condylar line (PCL) for the femoral component.
Femoral component rotation may be determined by several techniques; these include gap balancing techniques, dependent on ligament tension, and currently used by most surgeons measured resection techniques based on anatomical landmarks, including:
– Trochlear anterior-posterior axis (TAPA) – Whiteside’s line (WL)
– Sulcus Line (SL)
– Anatomical Transepicondylar Axis (ATEA)
– Surgical Transepicondylar Axis (STEA)
– Femoral Transverse Axis (FTA)
– Posterior Condylar Line (PCL)
Alternative methods exist that involve patient-specific instrumentation (PSI) and computer navigation.
Fig. 1 – Bony landmarks for femoral rotation
Measured resection – “femur first” or “bony landmarks” technique
The measured resection method was developed by Hungerford for use in cruciate-retaining TKA 15. In this technique, the bone resections are performed according to the bony landmarks followed by soft tissue balancing, in which the femoral component is aligned with respect to the epicodylar line which best approximates the flexion/extention gap. This approach involves referencing the rotation off several possible bony landmarks summarized in Figure 1. No gold standard for rotational reference has yet been generally accepted.
Posterior condylar line (PCL)
The posterior condylar line is defined as the tangent line to the most posterior part of the femoral condyles withe the femur viewed along its mechanical axis 22. Referencing the femoral rotation off the posterior condylar line can be used in most current systems and is relatively easy. Unfortunately, it has been postulated that this reference may be unreliable and give rise to many errors, such as in the case of posterior condylar defects in varus and valgus knees 17. The PCL is in relative internal rotation to the femoral rotation due to the larger size of the posteromedial femoral condyle. Victor reported a literature review analysis of different studies describing the angular relationship between the different axis of the distal femur in the axial plane. The posterior condylar line has a mean internal rotation of 3 degrees relative to the surgical TEA, 5 degrees relative to the anatomical TEA, and 4 degrees relative to the trochlear AP axis, respectively 21. In the valgus knee, the PCL internal rotation tendency is even greater because of the hypoplastic lateral femoral condyle 23. Three degrees of femoral component rotation is the value routinely used in varus, and five degrees – in valgus malalignment, respectively 27. For every 1 mm of asymmetry in condylar cartilage loss, the femoral rotation measurement with the use of PCL changes by 1 degree 16. PCL can be in the range of 1 to 9 degrees of internal rotation relatively to the anatomical TEA 24. The PCL referencing should be therefore used with caution.
Surgical transepicondylar axis (STEA) and anatomical transepicondylar axis (ATEA).
The transepicondylar axis reference method is reliable as the TEA approximates the flexion-extension axis of the knee. Placing the femoral component parallel to the TEA allows to obtain a rectangular resection gap in over 90% of cases. In some cases the TEA is easier to identify intraoperatively compared to the identification of the posterior condylar angle or the posterior condylar line especially in revision cases. However, the primary disadvantage of this technique is the difficulty in defining the TEA in obese patients. The epicondylar eminences are often poorly visualized and can be overlied by the collateral ligaments, everted patella, and particularly by fat tissue. There are studies suggesting that over 50% of malalignment errors are due to difficulty in identifying the epicondylar eminences 19. In the Yoshino et al. study of 48 patients with osteoarthritis, the medial sulcus could only be determined in 30% of the knees. The difficulty in locating the sulcus increased with the severity of arthritic changes 28. Removal of the soft tissue improves the identification of the condyles, assessment of the TEA and reduces the risk of femoral component malrotation.
Two tranepicondylar lines can be identified, the anatomical (ATEA) and the surgical (STEA) .
The TEA is classically defined as a transverse line drawn between the most prominent points on the epicondyles and is also known as the anatomical epicondylar axis (AEA) 24. According to a study by Akagi, the angle between ATEA and PCL was found to be 6.8° on preoperative CT scans 2.
The STEA refers to a line drawn from the medial sulcus to the lateral epicondyle. It is a secondary anatomical axis, useful for determination of rotational orientation of the femoral component when the posterior condylar surfaces cannot be used. Berger et al. used surgical transepicondylar axis (STEA) to determine the posterior condylar angle subtended as the angle between this axis and the PCL. Measurement of the posterior condylar angle referenced from the surgical epicondylar axis yielded a mean posterior condylar angle of 3.5 degrees (+/- 1.2 degrees) of internal rotation in males, and a mean posterior condylar angle of 0.3 degrees (+/- 1.2 degrees) of internal rotation in females. 22. The surgical epicondylar axis provides a visual rotational alignment reference during primary arthroplasty and may improve femoral component alignment in revision.
An angle between ATEA and STEA was reported by Yoshino et al. to be 3.2°±1°, with ATEA being more externally rotated 28. Another study by Victor reported that the mean angle between the anatomical and surgical TEA is 2 degrees 21.
Trochlear anterior-posterior axis (TAPA) – Whiteside’s line (WL) and the sulcus line (SL)
The trochlear anterior-posterior axis – Whiteside’s line, is a line connecting the deepest point of the trochlea to the center of the intercondylar notch 23. The anteroposterior axis indicates the direction of the trochlea in healthy knees and is perpendicular to the ATEA 25. Femoral component rotation is oriented perpendicular to the TAP axis. This reference is reliable and can be applied in patients with distorted condylar anatomy – such as condylar hypoplasia or defect. TAPA is less reliable than TEA in the valgus knee and in trochlear dysplasia 26. The line perpendicular to the AP axis is externally rotated by 3.5° relative to the PCL in normal knees. The internal rotation angle of the line perpendicular to the anteroposterior axis relative to the epicondylar axis is 0.1° ± 3.3° (medial femorotibial arthritis), 1.3° ± 3.3° (patellofemoral arthritis), and 2.3° ± 3.1° (normal knees) 32.
The main disadvantage of this reference lies in the difficulty of defining the trochlea AP axis in trochlear dysplasia and in patients with destructive arthritis of the anterior compartment 18, and with significant varus or valgus deformity 26,32. The AP line is variable and therefore its isolated use to determine the femoral component rotation in patients with destructive arthritis may result in malrotation of the femoral component and should not be used as a single landmark 21 but complementary to other reference axis.
An alternative to determination of the TAPA is the sulcus line reference. In this technique, the trochlear groove is perceived as a three-dimensional structure; multiple points in the trochlear groove (forming usually an arc) are connected then reoriented to achieve a straight line along the coronal aspect of the trochlear groove. This technique is believed to reduce the parallax error compared to TAPA because there is only one true coronal alignment axis. The TAPA relies on the accuracy of determination of the anterior point in the proximal section of the trochlea; this point is frequently affected by osteoarthritis and even though both axes reference the trochlear groove, the SL has geometrical advantages that make it a more accurate landmark 61. Accuracy of the SL approach was measured using postoperative CT scans. Chao et al. in their study compared the SL to SEA and PCL which showed a mean 0.7° of internal rotation (5.5 ° internal to 4.6° external rotation) and a mean of 1.6° of external rotation (7.6° internal rotation and 9.3° of external rotation) respectively 62.
Summary of “bony landmarks” technique
There is no consensus yet as to which is the best rotational reference method for proper femoral component rotation and to which all other parameters can be compared. Each landmark is affected by various factors as mentioned previously. To increase accuracy, it is recommended to cross-check at least two landmarks and use multiple references whenever possible to reduce errors.
Gap balancing or “tibia first” technique
Another technique of determining the femoral component rotation is the gap balancing method, relying on ligament balancing to establish a symmetrical and rectangular flexion and extension gaps prior to definite bone resection and component placement. Spacers of different types are used to achieve correct ligament tension. These devices rely on force applied manually by the surgeon or may include various sensors and tensor tools.
This technique is possible in knees with moderate degenerative changes and small deformities not requiring extensive soft tissue release. The extension gap is balanced first with appropriate medial release in varus knees. After balancing the knee in extension, it is flexed to 90° and some form of tension measuring device is applied across the medial and lateral compartments. The femoral component is then rotated in order to achieve flexion gap symmetry. When this method is used, 90% of knees are implanted in 5° of external rotation relative to the posterior femoral condylar line 29,30. Boldt et al. found that with this technique the posterior condylar angle was within 3° of the surgical TEA in 90% of knees 31. The flexion gap balancing method is characterized by excellent reliability with the prerequisition of intact collateral ligaments. The technique does not involve identification of anatomic landmarks and closely approximates the knee flexion axis. For a rectangular flexion gap to be created, it is paramount to perform an accurate proximal tibial cut.
Summary of “gap balancing” technique
There is no gold standard for assessing the resection quality. Various devices have been developed for this purpose, e.g. spacer blocks, laminar spreaders, and tension jigs. In a study comparing balancing methods: the gap balancing, AP trochlear axis, and TEA, Katz et al. demonstrated that the gap balancing method may be superior in accuracy and reliability; this possibly results from the fact, that the technique does not involve identification of poorly visible bone landmarks 20.
Computer-assisted navigation was introduced to supplement TKA surgery with the potential to improve positioning and alignment of TKA components. Several meta-analyses 35-37 have demonstrated that although the average coronal plane alignment after computer-assisted navigation TKA was not different from conventional TKA, the variability in the outcome was reduced. A number of studies showed that navigation-assisted TKA improves alignment in TKA more predictably than conventional jig-based surgery, while decreasing blood loss and enabling faster post-operative recovery 35, 38-41. There is conflicting evidence as to whether computer navigation improves the accuracy of component rotation. The technology proved to be effective in reducing outliers in the coronal and sagittal planes, but to has failed to improve the rotational alignment. To date, navigation has not provided an efficient solution for optimizing the rotational alignment of femoral component 42-49. Siston et al. also showed that navigation systems that rely on directly digitizing the femoral epicondyles to establish alignment axis did not provide a more reliable means of establishing femoral rotational alignment than traditional techniques did.
New technologies are continually emerging in arthroplasty. Patient-specific instrumentation (PSI) systems are interactive computer planning tools that use preoperative imaging techniques such as computed tomography (CT), magnetic resonance imaging (MRI) and full-length radiography with rapid prototyping technology for preoperative determination of bony resections and implant sizing. Computer-generated models are used to manufacture disposable cutting blocks that are thought to help the surgeon reproduce the preoperative plan during surgery. Such systems aim to improve three-dimensional implant positioning while reducing overall costs of instrumentation and implants 52-53. The literature is inconclusive in terms of superiority of PSI over conventional instrumentation (CI). There are only a few studies reporting the accuracy of femoral component rotation using PSI 54,55. Other studies report that PSI does not improve femoral rotation in TKA 56, 57. Fu et al. (58) conducted a meta-analysis and found no obvious statistical difference between PSI and CI in the postoperative mechanical limb axis or femoral component placement. However, in other studies PSI was found to be effective in significantly reducing outliers in femoral component rotation 54, 55. As with all technologies, PSI has its disadvantages, including delay in surgery and considerable costs for preoperative scans and manufacture of cutting guides, radiation exposure associated with CT prototyping (59,60), as well as the learning curve.
Functional outcomes of TKA, both short and long-term, are highly dependent on correct rotational alignment of prosthetic components. There are many studies discussing individual advantages and potential problems with methods used for referencing the rotation alignment. No gold standard has been universally agreed upon to date; therefore surgeons should familiarize themselves with a variety of references and methods to establish the correct femoral component rotation. To reduce the rate of femoral component malrotation, cross-checking of at least two references should be performed during the TKA procedure.