Abstract

Introduction: Rotational alignment of femoral component in TKA affect the clinical results of long-term follow up (Stiehl). Improper alignment may lead to unstable femoro-tibial joint, to wear or loosening of tibial component, and is associated with the subluxation or dislocation of the patella by patella-femoral mal-tracking (Pascal 1996). The precise setting of femoral component is important for the smooth patella tracking and good ligament balancing in TKA.

Previously orientation of rotation of the femoral component has been set by equal resection of the posterior condyle (Hungerford 1985, Laskin 1989). The anteroposterior axis of the distal femur that was defined by a line through the deepest part of the patellar groove anteriorly and the center of the intercondylar notch posteriorly, was an easy and reliable landmark of the rotational alignment of the femoral component (Whiteside, Arima).

The posterior condylar line (PC line) that connects the posterior condyle of the femur is widely used as a landmark for the cutting of the posterior condyle. Also, 3°external rotation off the posterior femoral condyle has been commonly used as a intraoperative landmark (Laskin1995).

The anatomical and functional axis of the femur has been studied so far (Poilvache.Yoshioka1987). Transepicondylar axis (TEA) as the origin of collateral ligament is valuable axis for the parallel cut of the posterior condyle (Berger, Miller). TEA was found to be a reliable landmark to proper rotation of the femoral component, measuring the angle between the axis and the posterior condylar line to orient the femoral component is very important.

However, intra-operative manual palpation of the TEA was not reproducible because most prominent point was covered with soft tissue (Jenny). It is sometimes difficult to identify the sulcus of the medial epicondyle accurately with palpation even during surgery ().

Therefore, it is crucial to measure and evaluate the TEA as the preoperative planning. The posterior condylar line (PC line) that connects the posterior condyle of the femur is also used for the landmark of the cutting of the posterior condyle. The methods of examining the angle between TEA and PC line are computed tomography (CT) and kneeling view that was simple radiographic technique by Takai et al. Posterior condyle of deformed side makes inaccurate decision of the angle for TEA and PC line because thickness of cartilage and bone are different between medial and lateral condyle. PCA is not applicable in MIS-TKA because it is very difficult to visualize the posterior condyle in the lateral side by the medial approach.

Alternative landmark of the angle between TEA and anterior trochlear line of lateral and medial femoral condyles for the determination of rotational positioning of the femoral component may be considered. We have improved the simple radiographic view of evaluating the TEA and PC line but also anterior trochlear line for the assessment of rotational alignment of the distal femur in TKA. The purpose of this study was to measure these angles and to evaluate the reliabilities in compared with 3D-CT.

Subjects and methods Our new radiograph we describe is the antero-posterior view of looked-up distal femur. The patient lies on the supine position and flexes the knee about 130 degrees as much as possible. X-ray is applied to the knee at the right angle to the front of the skin from 20 degrees bottom (Figure 1).

We pointed out the location of the anterior surface of the condyles, medial epicondyle and lateral epicondyle. We marked the medial and lateral epicondyle of anterior surface of condyles, and posterior condyles as the indivisual reference points in these views. We defined the anterior intercondylar line (trochlear line) as the most axial projections of the medial and lateral femoral condyles. We defined PC line as a line connecting the surfaces of the subarticular cortex of the medial and lateral posterior femoral condyles likewise. We used to obtain clinical TEA that was defined by drawing the most prominent points of the medial and lateral epicondyles. We measured the external rotational angle between PC line and clinical TEA (condylar twist angle), and the internal rotational angle between clinical TEA and trochlear line (Figure 2).

Reproducibility of our radiographic technique We examined the reproducibility of our new radiographical technique by 20 healthy volunteers. They included ? males and ? females and the average age of the patients was # years (# ~ # years). No knees in volunteers showed remarkable deformities. We photographed at the flexion angle from 110 to 140 degrees every 10 degrees, at the incident angle of 20 and 30 degrees.

The anterior trochlear line, PC line and clinical TEA were drawn on the images and measured condylar twist angle and the internal rotation angle between clinical TEA and trochlear line. The differences of their measurements were quantified using paired t-test.

Comparison with our view and reconstruction images of 3-dimensional helical CT system The CT images of 35 knee joints in 28 patients had been taken at full extension of the knee using 512 × 512 pixel matrix, in addition of plain X-ray. From the data of CT images, two different images were acquired such as the composition images and the reconstruction images of 3D.

The composition images were obtained by putting a photograph with slices of every 2 mm on top of one another. The CT slices (Shimazu Co Ltd, Kyoto, Japan) obtained from the proximal edge of the patella to the joint line of the knee. We added anterior surface of condyles, medial epicondyle, lateral epicondyle and posterior condyles on tracing paper every slice in the same place. Then we drawn trochlear line, PC line and clinical TEA, and measured the external rotation angle between PC line and clinical TEA (condylar twist angle) and the internal rotation angle between clinical TEA and trochlear line.

The reconstruction images were obtained by the distal femoral view looked-up from distal aspect and reconstructed with 3-dimensional helical CT system. We have drawn trochlear line, PC line, clinical TEA, and measured the external rotation angle between PC line and clinical TEA (condylar twist angle), and the internal rotation angle between clinical TEA and trochlear line from three methods mentioned above and had compared them. The differences of their measurements were compared with three groups.

This study involved 122 knees in 82 patients including 22 males and 80 females with osteoarthritis of the knee. The average age of the patients was 67.3 years from 37 to 89 years. We classified by Kellgren and Lawrence classification (K-L grade). They consisted of grade 1; 12 knees, grade 2; 37 knees, grade 3; 34 knees, and grade 4; 39 knees. Tibiofemoral angle (TFA) on long-leg radiography at the standing position were ranged from 164°to 197°; mean, 180.2°±6.7°. We examined the correlation between condylar twist angle and gender, TFA, height and weight.

Informed consent

Statistics: Statistical analysis was performed on a personal computer using a statistical software of Statview (SAS institute, Chicago, IL). P values of less than.05 was considered as statistically significant. We used the t-test, which does not need the raw data, to compare our results with those published. We were able to obtain information on the number of subjects (to calculate the degree of freedom) and the mean from the publications.

Results: The external rotation angle between PC line and clinical TEA (PC-TEA), that is condylar twist angle (CTA) was 5.6°±2.8°(mean±s.d). The internal rotational angle between clinical TEA and anterior trochlear line (trochlear-TEA) was 5.7°±3.2°. K-L grade was negatively correlated with these rotational angles using Kruskal Wallis test (Table 1). These angles of female was larger than those of male (Table 2).

The varus angle was negatively correlated with the CTA (R=−0.30) and positively correlated with the internal rotation angle of trochlear-TEA (R=0.376) (Figure 3).

The external rotation angle between PC line and clinical TEA was 5.3°±2.4° at our view, and 5.5°±2.3° at reconstruction images from 3-dimensional helical CT system. The difference of condylar twist angle between plain X-ray and 3D-CT was shown in Figure 4.

The internal rotation angle between clinical TEA and anterior trochlear line was 5.3°±2.4° at our view, °and 5.7°±2.3° at reconstruction images from 3-dimentional helical CT system. The difference of the internal rotation angle between clinical TEA and anterior trochlear line between plain X-ray and 3D-CT was shown in Figure 5.

Regarding the reproducibility about the flexion angle of the knee and the incident angle, correlation coefficients were ? for the flexion angle of the knee, ? for the incident angle. All cases were within 5° variations of the external rotation angle between PC line and clinical TEA, and 4° variations of the internal rotation angle between clinical TEA and trochlear line, respectively. The case of at 110° flexion and 30° incident angle, however, tends to be more variable than the other cases due to unclear PC line (SD 3.3°; range 3–16°).

Discussion: Two kinds of TEAs are used for the reference of femoral rotation on the surgical TEA and the clinical TEA. Surgical TEA is a line connecting the sulcus of the medial epicondyle and the lateral epicondylar prominence (Berger 1993). The posterior condylar angle (PCA) is the angular measurement subtended by the surgical TEA and the posterior condylar line (PCL). The clinical TEA is a line connecting the medial and lateral epicondylar prominence. Clinical twist angle (CTA) is the angular measurement subtended by the clinical TEA and the PCL. The most prominent point of the medial epicondyle that is a landmark of CTA was much better identifiable than the medial sulcus for the landmark of PCA (Suter 2006). The sulcus of the medial epicondyle could only be identified in 53 % of the CT images, however, the most prominent point of the medial epicondyle in CT images was clearly discernible in all knees (Suter 2006). Yoshino et al. recommended the use of the CTA in planning for TKA. Even during surgery the determination of the sulcus is difficult by the palpation (). Medial sulcus become obscures in the severely deformed osteoarthritic knee (Yoshino).

Intra-operative palpation of the trans-epicondylar axis involved a mean of 5° intra- and inter-observer variations (Jenny 2004). Some authors reported that CTA was 3.6° ± 2.02, 3.58° in male and 3.62° in female during TKA (Poilvache), however, in CT study

Arima reported that CTA was 5.7°± 1.7, in cadaver study 4.4° in male and 6.4° in female. In our study, there was significant difference in gender of CTA.

There has been only a few reports regarding the angle between the TEA and anterior trochlear line of the lateral and medial femoral condyles (trochleo-epicondylar angle). The line between the most anterior projections of the lateral and medial femoral condyle was called as trochlear line, was measured (Poilvache 1996), trochleo-epicondylar (surgical) angle was 4.95° ± 2.15, 4.4° in male, 5.38° in female during TKA. The mean value of the trochleo-epicondylar angle in CT view was 8.0°± 1.76 of internal rotation in all subjects, 8.8° in male, 7.3° in female, there was significant gender difference (Won). Our developed view is the first method of showing the trochleoepicondylar angle in plain radiography. Our results demonstrated trochleo-epicondylar angle using clinical epicondylar axis was 5.6° ± 2.85 of internal rotation in all subjects, 5.27° in male, 5.77° in female, there was no significant gender difference.

Line drawing of posterior condylar line between medial and lateral condyle in osteoarthritic knee sometimes make error of the angle measurement because thickness of cartilage and wear of subchondral bone is not equal in the both side of the condyle. Our view is the first method that is able to examine both the CTA and trochleo-epicondylar angle simultaneously, simple, need not to use special instrument, and reveal reproducible.

A minimally invasive operative method in TKA is reported to be effective and recommended in primary OA. However, the reference guide of the angle between PCA and TEA is sometimes difficult to set properly with the full contact of both condyles in the limited view of the non-open side, especially MIS TKA.

In contrast, it is easy to set the guide or template properly for the trochlear line angle during the surgery because the anterior trochlear is completely visible. Surgeons should not use only one method of femoral rotational alignment and make appropriate adjustment in TKA (Olcott 2000). Then, we focused on the angle between the anterior trochlear line and TEA,

And we developed the simple method of the radiographic view that is able to evaluate the trochlear line and clinical epicondylar axis as the preoperative surgical planning.

From our data, the trochlear line angle with a landmark of the anterior femoral condyle by our radiographic view was reproducible. Our method may be a possible one for determining the rotational alignment of the femoral component in total knee arthroplasty.

Regarding the study of variability of these angles in several kinds of flexion angle of the knee and Therefore, we are able to measure and evaluate both angles, and reduce the measurement error by double-checking the conventional CTA and trochlear line angle.