Differences between the Craig's test and computed tomography in measuring femoral anteversion in patients with anterior cruciate ligament injuries.

[Purpose] This study aimed to compare the Craig's test and computed tomography (CT) in measuring the femoral anteversion angle (FAA) in patients with anterior cruciate ligament (ACL) injuries. The relationship between the FAA measured on CT, and the range of axial rotation of the hip joint and muscle tightness around the hip joint was also investigated. [Participants and Methods] Twenty-six patients who received CT examination within 3 months after ACL reconstruction were enrolled in this study. The Craig's test, internal and external rotation of the hip, the Ely test, Ober's test, and FAA on the CT were assessed.
[Results] The FAA on the Craig's test and CT in female patients was 24.3 ± 3.9° and 23.0 ± 10.3°, respectively on the uninjured side and 25.0 ± 5.2° and 20.3 ± 11.2°, respectively on the injured side, indicating no significant correlation between the 2 measurement techniques. In contrast, the FAA on the CT was significantly correlated with the range of internal rotation of the hip, which was 40.5 ± 6.1° on the uninjured side and 37.8 ± 5.6° on the injured side.
[Conclusion] The results suggest that measuring the internal hip rotation range rather than the Craig's test, provides more reliable estimates of the magnitude of FAA, and may help to evaluate the risk of ACL injuries in female athletes. 2020©by the Society of Physical Therapy Science. Published by IPEC Inc.

INTRODUCTION

The major mechanisms for anterior cruciate ligament (ACL) injuries have been reported to be forceful valgus and internal rotation of the knee incited by high-impact motions, especially in non-contact injuries that are common in basketball and handball1). Kaneko et al.2) reported that increased femoral anteversion angle (FAA) results in lower hip flexion angle, higher knee valgus alignment, and greater rectus femoris muscle activity, leading to during single-leg landing. Since the congruity of the hip joint improved with excessive internal rotation of the hip and valgus of the knee, increased FAA might be a risk factor for non-contact ACL injury. Various measuring methods and equipment have been developed to assess the FAA. While magnetic resonance imaging and computed tomography (CT) can be useful for an accurate quantitative analysis, a physical examination is required in daily clinical practice to determine femoral anteversion. Craig’s test is the most commonly used physical examination technique for measuring femoral anteversion. However, to our knowledge, only a few studies reported the comparison of FAA between Craig’s test and CT has not been reported.

This study aimed to compare the FAA measured by Craig’s test and CT in patients with ACL injury. In addition, we investigated the relationship between the range of axial rotation of the hip joint and muscle tightness around the hip joint and the FAA measured using CT.

PARTICIPANTS AND METHODS

This study was a retrospective case-control study. All participants visited our hospital from May 1st to June 30th in 2016. Patients who underwent ACL reconstruction surgery more than 3 months previously in the department of orthopedic surgery of our institute were enrolled in this study. The inclusion criteria were patients who received CT examination of whole legs within 3 months after the surgery. The exclusion criteria were skeletally immature with open physes around the knee joint and a history of trauma to the lower extremities, other than ACL injuries. Twenty-six patients (15 male and 11 female patients) were consequently included. This study was approved by the Institutional Review Board of our institute (2016-002), and a written informed consent was obtained from all patients.

For the Craig’s test,3) the patient was placed in the prone position with the knee flexion of 90° on an examination table. The examiner palpated the greater trochanter of tested side while passively rotating the hip until the most prominent portion of the greater trochanter reached its most lateral position. The FAA, comprising the angle between the shaft of the tibia and a line perpendicular to the table, was measured using a digital goniometer (Fig. 1). One physical therapist records the value on the digital goniometer as the FAA, while another physical therapist maintained the tibia in place. Internal and external rotation angles of the hip joint were measured using the digital goniometer by 2 different methods as follows: in the supine position with 90° of hip and knee flexion and in the prone position with hip extension and 90° of knee flexion. In addition, the muscle tightness of the quadriceps and tensor fasciae latae in both lower extremities were evaluated by the Ely test4) and Ober’s test5), respectively.

Fig. 1.

Measurement of femoral anteversion with a digital goniometer.

The FAA on CT images was defined as a crossing angle between a line drawn from the center of the femoral head to the center of the base of the femoral neck (A to B) and a line connecting the exact posterior aspects of the lateral and medial condyles (A to C) (Fig. 2). All measurements were performed by a single physical therapist (I.I.) who was not aware of the results of Craig’s test. The therapist has over 15 years of clinical experience and greatly familiar with functional anatomy of musculoskeletal systems and rehabilitation for orthopaedic disorders.

Fig. 2.

Measurement of femoral anteversion angle (FAA) on computed tomography (CT).

a: Line drawn from the center of the femoral head to the center of the base of the femoral neck (A to B).

b: Line connecting the exact posterior aspects of the lateral and medial condyles (A to C).

The FAA was defined as a crossing angle between A to B and A to C.

The involved and uninvolved sides were separately analyzed in the male and female patients. The Spearman’s correlation coefficient was determined for the correlation between the FAA in Craig’s test and that on CT. The correlation between the internal rotation angle of the hip and the FAA was determined by a linear regression analysis with the FAA on CT as the dependent variable and that on Craig’s test as the evaluation item. Regression analysis was performed using these dependent and independent variables for the internal and external rotation angles of the hip in the supine and prone positions. The regression models were corrected for age and the Ely test and Ober’s test results as factors that can potentially affect the range of motion. Differences were considered to be statistically significant at p<0.05. All statistical analyses were performed using SPSS software ver. 12.0 J (SPSS Inc., Chicago, IL, USA).

RESULTS

In the 26 patients who were evaluated, the mean body weight, height and age were 67.9 ± 11.5 kg, 172.1 ± 4.3 cm, and 32.2 ± 15.5 years in male patients, and 61.1 ± 11.9 kg, 160.2 ± 4.9 cm and 24.8 ± 10.4 years in female patients. Twenty-one patients underwent primary ACL reconstruction, 3 patients underwent revision reconstruction, and 2 patients underwent bilateral reconstruction. The transplanted autograft was the bone-patellar-bone tendon in 7 patients, and the semitendinosus tendon in 19 patients. Concomitant meniscectomy in 2 knees and meniscal repair in 8 knees were performed at the ACL reconstruction.

The measurement values of FAA, hip rotation angle, and the positive rates on the Ely test and Ober’s test are shown in Table 1. There was no significant difference in the FAA between the involved and uninvolved sides, measured with either Craig’s test or CT. There was no significant difference in the FAA between Craig’s test and CT. However, no significant correlation of the FAA was found between the Craig’s test and CT measurements on both the involved (r=0.12, p=0.59) and uninvolved sides (r=0.04, p=0.84). Meanwhile, the linear regression analysis showed that significant positive correlation between the FAA on CT and the internal rotation angle, especially in the prone position, in female patients (Table 2). However, this correlation was not found in male patients. In both male and female patients, the FAA on CT showed no significant correlation with external rotation of the hip in the supine and prone positions, Ely test, or Ober’s test.

Table 1.

Comparison between the involved side and uninvolved side

	Male		Female
	Involved	Uninvolved	Involved	Uninvolved
Craig’s test	22.9 ± 5.4°	20.5 ± 5.2°	25.0 ± 5.2°	24.3 ± 3.9°
CT anteversion	10.3 ± 6.1°	14.1 ± 8.1°	20.3 ± 11.2°	23.0 ± 10.3°
IR in prone	40.5 ± 6.1°	37.8 ± 5.6°	45.2 ± 12.0°	46.4 ± 9.5°
IR in supine	28.5 ± 5.8°	28.3 ± 6.8°	38.3 ± 13.1°	40.9 ± 10.3°
ER in prone	29.7 ± 6.8°	30.9 ± 5.6°	25.6 ± 5.4°	29.5 ± 9.3°
ER in supine	39.2 ± 5.3°	37.5 ± 5.8°	34.5 ± 11.8°	33.1 ± 9.2°
Ely test	40%	33%	18%	9%
Ober’s test	67%	60%	73%	64%

CT: computed tomography; IR: internal rotation of the hip; ER: external rotation of the hip.

Table 2.

Correlation of internal rotation of the hip and femoral anteversion angle on CT*

	Involved					Uninvolved
	Male		Female			Male		Female
	β	p-value	β	p-value		β	p-value	β	p-value
Prone	−0.28	0.39	1.32	0.01		0.2	0.61	1.18	0.01
Supine	0.12	0.77	1.32	0.01		0.38	0.24	0.81	0.02

Dependent variable: FAA on CT.

Correction factors: age, Ely test, Ober’s test.

CT: computed tomography; β: standardized partial regression coefficient.

DISCUSSION

The most important result of this study was that the FAA on CT correlated with the range of internal rotation of the hip in female patients. In general, increased FAA has been reported to cause larger internal rotation of the hip. Kunita et al.6) reported that players with increased FAA might excessively rotate the hip internally to improve the congruity of the hip joint during landing or cutting motion in patients of ACL injury. Kaneko et al.2, 7) reported that female basketball players with increased FAA on Craig’s test showed excessive knee valgus during single-leg landing. Therefore, increased FAA could be one of the possible risk factors for ACL injury. Yasuda et al.8) reported that decreased range of external and increased range of internal rotation of the hip were associated with the incidence of ACL injuries in female handball players. There was a significant association of decreased range of external rotation of the hip and dominance of the internal rotation of the hip with knee valgus induced by an ACL injury screening test8). Female athletes with decreased external rotation of the hip due to greater FAA may hold the femoral head within the acetabular cavity by internal rotation of the hip. Such compensatory changes in femoral internal rotation could lead to knee valgus.

On the other hand, Tainaka et al.9) reported that a narrow range of hip rotation has a significant relationship with an increased risk of a non-contact ACL injury. The study analyzing a non-contact ACL injury mechanism using a model-based image matching technique showed that the hip joint was fixed in a flexion and internally rotation position at the injury onset10). The hip joint with smaller internal rotation range could easily reach to maximum internal rotation and be frequently fixed during sports activities, increasing risk of non-contact ACL injury.

The results of this study showed that the FAA on CT was not correlated with that using Craig’s test. Tamari et al.11) reported that the use of the greater trochanter could potentially cause a palpation error in Craig’s test. The large and rough bony prominence of the greater trochanter makes it difficult to determine the rotation angle at which the greater trochanter reaches its most lateral position accurately and consistently. Another probable reason for the measurement error in Craig’s test may be associated with the use of tibial inclination as a guide for hip rotation. Ruwe et al.3) reported that the increased ligament laxity of the knee joint might affect the results of Craig’s test. Yoon et al.12) reported that any potential increase in the joint space of the knee could affect the angle between the vertical line and the tibial crest during Craig’s test. Direct measurement of the transcondylar line of the distal femur, if possible, may be more reliable. The patients in this study showed stable varus and valgus rotation in uninvolved and involved knees after ACL reconstruction and no osteoarthritic changes were observed on plain radiographs or CT images. Therefore, our results of no statistical correlation between the FAA on Craig’s test and CT might be affected more by palpation errors of the greater trochanter, rather than soft tissue laxity or joint space opening.

In contrast with the results in female patients, male patients showed no significant correlation between FAA on CT and the range of internal rotation of the hip. It was suggested that the smaller standard deviation in internal rotation allowed much more negative effect of measurement error on analyzing correlation with FAA. Also, the smaller standard deviation in FAA might make other factors (i.e. anatomy of acetabulum, muscular tightness and lower limb alignment) affect measurements of internal rotation more greatly.

There were several limitations in this study. The first limitation was the small sample size including only those patients who had undergone ACL reconstruction. Another limitation was that Craig’s test was performed by one examiner, and the FAA on CT was measured by a different examiner. Although the examiner for Craig’s test had much experience of orthopaedic rehabilitation, relatively less experience of Craig’s test in a daily clinical practice might induce measurement error. Future studies with a larger sample size are required to validate the findings of our study.

In conclusion, no significant correlation was found between the FAA using Craig’s test and that on CT in patients who underwent ACL reconstruction. In female patients, a significant positive correlation was observed between the internal rotation angle of hip and the FAA on CT. During clinical examination, measurement of the internal rotation of hip might be more reliable than Craig’s test to estimate the magnitude of FAA, and have much more benefit to evaluate the potential risk of ACL injury in female athlete.

Conflict of interest

None.