Keywords
adult hip dysplasia - acetabular dysplasia - radiography - pelvic radiograph - Tönnis
angle - computed tomography
Introduction
Adult dysplasia of the hip (ADH) encompasses a spectrum of disordered development
of the hip joint resulting in a shallow acetabulum with a lack of anterior and lateral
coverage of the femoral head.[1] The incidence of hip dysplasia is reported to range from 3 to 5% with a female preponderance
and it is a well-recognized risk factor for early osteoarthritis. Hip osteoarthritis
in up to 20 to 40% of patients is thought to be secondary to dysplasia.[2]
[3]
[4] ADH typically presents with hip pain in the groin area or lateral aspect with or
without a limp. This is often exacerbated with activity and may be associated with
a sensation of catching or popping. Radiographs and cross-sectional imaging including
computed tomography (CT) and magnetic resonance imaging (MRI) play a key role in assessing
coverage of the femoral head, acetabular deficiency, and any associated soft tissue
damage of the cartilage, ligamentum teres, and labrum.[5]
[6]
[7]
[8] Radiographs tend to be the preliminary investigation for diagnosis in hip dysplasia
with a spectrum of useful radiographic measurements including Tönnis angle (acetabular
index), lateral center edge angle of Wiberg, femoral neck-shaft angle, and delta angle
to evaluate deficient coverage of the femoral head or assess the slope of the acetabulum[6]
[9]
[10] ([Table 1]; [Figs. 1] and [2]). However, the aforementioned radiographic calculations are based on angular and
are susceptible to variability resulting from the position of the pelvis. Furthermore,
research has demonstrated poor interobserver and intraobserver reliability of these
radiographic measurements in the estimation of hip dysplasia.[11]
[12]
Table 1
Traditionally used radiological indices to assess and measure adult dysplasia of the
hip (ADH) on plain radiography
|
Radiological measurement
|
Radiological technique calculation method
|
Measures
|
Normal values
|
Abnormal values
|
|
1
|
Lateral center edge angle (LCEA) of Wiberg
|
Anteroposterior plain pelvis radiograph
-Angle between a vertical, perpendicular line through the center of femoral head of
interest (Line A) and a tangential line to lateral margin of the acetabulum (Line
B)
|
Coverage of femoral head
|
Between 25 and 35 degrees
|
Dysplasia <20 degrees
Severe < 5 degrees
|
|
2
|
Tönnis angle
(acetabular index)
|
Anteroposterior plain pelvis radiograph
-Angle between a horizontal line at the level medial edge of “sourcil” (Line A) and
a line tangential to medial and lateral edges of sourcil (Line B) ([Fig. 1])
|
Slope of acetabulum socket
|
Between 3 and 13 degrees
|
Dysplasia > 13 degrees
|
|
3
|
Femoral neck-shaft angle
|
Anteroposterior plain pelvis radiograph
-Angle between a line along the femoral neck axis through the center of head and intersecting
line drawn along the femoral shaft axis
|
Angle of Inclination
|
Between 120 and 135 degrees
|
Coxa Valga > 135 degrees
Coxa Vara < 120 degrees
|
|
4
|
Delta angle
|
Anteroposterior plain pelvis radiograph
-Angle between a line through the femoral head center to medial edge of sourcil (Line
A) and superior edge of Fovea capitis (Line B)
|
Sphericity of femoral head
|
More than > 10 degrees
|
Fovea Alta ≤ 10 degrees
|
Fig. 1 Anteroposterior radiograph (A) and schematic (B) showing measurement of Tönnis angle.
Fig. 2 Schematic showing measurement of lateral center edge angle. The angle between a vertical
line from the center of the femoral head (A) and the line connecting the center of the femoral head and the lateral edge of the
acetabulum (B).
CT allows a clinician to evaluate hip dysplasia more thoroughly in three-dimensional
(3D) and is unaffected by pelvis position[13] ([Table 2]).
Table 2
Computed tomography (CT) imaging measurements of adult dysplasia of the hip (ADH)
|
CT scan measurement
|
Imaging technique calculation method
|
Measures
|
Normal values
|
Abnormal values
|
|
1
|
Anterior acetabular sector angle (AASA)
|
Axial CT scan taken one cut above greater trochanter
Angle between lines through centers of both femoral heads and another line tangential
to the anterior lip of the acetabulum
|
Anterior coverage
|
>50 degrees
|
Dysplasia ≤ 50 degrees
|
|
2
|
Posterior acetabular sector angle (PASA)
|
Axial CT scan taken one cut above greater trochanter
Angle between lines through centers of both femoral heads and another line tangential
to the posterior lip of the acetabulum
|
Posterior coverage
|
>90 degrees
|
Dysplasia ≤ 90 degrees
|
|
3
|
Horizontal acetabular sector angle (HASA)
|
Axial CT scan taken one cut above greater trochanter
Angle between lines from anterior lip of acetabulum going through the center of the
femoral head and the posterior lip of the acetabulum
|
|
>140 degrees
|
Dysplasia ≤ 140 degrees
|
We have designed and investigated the use of a new ancillary measure of ADH on cross-sectional
CT imaging, the Birmingham Royal Orthopaedic Hospital (BROH) femoral offset. The BROH
femoral offset is a linear measure, easy to calculate on axial-CT, and allows evaluation
of the anterior femoral head coverage in patients with ADH.
Patients and Methods
Study Design and Patients
Following local hospital committee approval as a service evaluation, a retrospective
evaluation of our Radiology Information System and Picture Archiving and Communication
System was performed to identify 100 consecutive patients with suspected ADH and have
both anteroposterior radiograph and CT of the pelvis. Patients with trauma, previous
surgery and with hip replacements were excluded. The imaging protocol included anteroposterior
radiographs of the pelvis. All the CTs were obtained in the axial plane (1 mm slice
thickness) and multiplanar reformats were created (64 slice, Siemens Somatom Sensation
AS [Siemens Medical Systems; Erlangen, Germany]).
Image Analysis
The radiological images of all the patients were reviewed by two independent experienced
clinicians. Tönnis angle was calculated on an anteroposterior radiograph and BROH
femoral offset on axial CT images. One reader repeated measurement after 2 weeks to
assess interobserver reliability.
Calculation of the Tönnis angle: Tönnis angle is calculated by measuring the angle
between a horizontal line at the level of the medial edge of “sourcil” (Line A) and
a line tangential to the medial and lateral edges of sourcil (Line B) ([Fig. 1]).[6]
[14]
Calculation of BROH Femoral Offset
This was calculated on the axial CT images. The cranial most slice of the hip at the
level of the tip of the greater trochanter was identified. A line was drawn from the
anterior-most tip of the greater trochanter to the anterior-most part of the acetabulum.
(Line A). A line was drawn perpendicular to Line “A” to the anterior most part of
the femoral head that is termed the BROH femoral offset (Line B; [Fig. 3]).
Fig. 3 Axial computed tomography (A) and schematic (B) showing BROH femoral offset measurement.
Data Collection
We recorded patient demographics details, Tönnis angle, and BROH femoral offset for
each hip for every patient in the cohort. Data collection was undertaken using Microsoft
Excel data spreadsheet and analyzed using SPSS 24.0 software (SPSS Inc.; Chicago,
Illinois, United States).
Statistical Analysis
Mean, standard deviation or median (range), and standard error of mean were used to
summarize data. Analysis of variance (ANOVA) test was performed between the three
cohorts: normal Tönnis angle and those with reduced or increased Tönnis angle. Student's
t-test was used to assess significance between normal and abnormal Tönnis angle cohort.
Intraclass correlation coefficient (ICC) analysis was undertaken to assess reliability.
The ICC is a value between 0 and 1, in which score of excellent reliability is over
0.9, good reliability of 0.75 to 0/9, moderate reliability of 0.5 to 0.75 moderate,
and poor reliability less than 0.5.[15] A p-value of less than 0.05 was considered significant in this study.
Results
There was a total of 100 patients (128 hips) included in the study (60 had normal
Tönnis angle, 53 had dysplasia, and 15 had decreased Tönnis angle). The average age
among all three cohorts was comparable and there was a female predominance in all
three cohorts. The average BROH femoral offset in the dysplastic cohort was increased
in comparison to the normal cohort and this was statistically significant with a p-value of 0.0001 (Student's t-test; [Figs. 4], [5]) ANOVA was performed between all three cohorts, and this was also statistically
significant with a p-Value of 0.00031. The BROH femoral offset was shown to be reliable with good intra-
and interobserver reliability of 0.9 and 0.9, respectively. Bland–Altman plots also
demonstrated good intra- and interobserver reliability ([Figs. 6] and [7]). Demographic details, BROH femoral offset, and Tönnis angle measurements in the
study group with descriptive statistics are depicted in [Tables 3] and [4].
Table 3
Demographics of the study group of 100 patients (n = 128 hips)
|
Age in years
|
|
Tönnis angle
|
|
>10 degrees
|
0–10 degrees
|
<10 degrees
|
|
Male
|
21
|
16
|
6
|
|
Female
|
32
|
44
|
9
|
|
Mean
|
29.3
|
25.6
|
28.5
|
|
Maximum
|
69
|
52
|
16
|
|
Minimum
|
16
|
11
|
14
|
Table 4
BROH femoral offset and Tönnis angle measurements in the study group of 100 patients
(n = 128 hips)
|
Tönnis angle
|
|
>10 degrees
|
0–10 degrees
|
<10 degrees
|
|
Femoral offset in centimeters (cm)
|
|
Mean
|
1.2
|
0.78
|
0.96
|
|
SD
|
0.65
|
0.41
|
0.45
|
|
SEM
|
0.09
|
0.05
|
0.11
|
|
Number
|
53
|
60
|
15
|
|
Median
|
1.1
|
0.76
|
0.85
|
|
95%CI
|
1.02–1.37
|
0.68–0.89
|
0.71–1.2
|
Abbreviations: CI, confidence interval; SD, standard deviation; SEM, standard error
of mean.
Fig. 4 Anteroposterior radiographs of the right hip (A) of dysplastic hips with increased Tönnis angle with corresponding computed tomography
(B) showing increased femoral offset.
Fig. 5 Anteroposterior radiographs of the right hip (A) of dysplastic hips with increased Tönnis angle with corresponding computed tomography
(B) showing increased femoral offset.
Fig. 6 Bland–Altman plot shows correlation between two readers with good interobserver reliability.
Fig. 7 Bland–Altman plot showing good intraobserver reliability.
Discussion
ADH is characterized by morphological abnormality of congruity of the hip joint, ranging
from a shallow acetabulum, uncovering of the femoral head to a completely dislocated
hip. It is a well-recognized cause of adult hip pain.[16] Untreated, ADH has been reported to contribute to the development of symptomatic
hip osteoarthritis in 20 to 40% of patients with osteoarthritis of the hip requiring
surgical intervention such as total hip replacement.[3]
[6] Despite the attempts of widespread screening for hip dysplasia at birth or during
infancy, a considerable number of patients go undetected until adulthood due to the
later onset of symptoms or delayed diagnosis.[17] Since periacetabular osteotomy and hip preservation procedures at an early age when
acetabular remodeling is possible may prevent the need for hip replacement surgery,
early detection, and treatment of acetabular dysplasia have been emphasized.[18]
Radiological imaging plays a key role in the diagnosis, monitoring, and deciding management
strategies in patients with ADH.[3]
[5]
[6]
[7]
[13]
Plain radiography is the preliminary investigation in the assessment of hip dysplasia.
An anteroposterior radiograph of the pelvis, a lateral “false profile” view, and an
abduction view of the hip is commonly performed views.[5]
[6] Plain radiography evaluation has the advantage of affordability, better accessibility,
and universal availability. Traditional measurements undertaken include the LCEA of
Wiberg, Tönnis angle (acetabular index), femoral neck-shaft angle, and the delta angle.[6]
[9]
[10] These angles can assess the coverage of the femoral head, the slope of the acetabulum,
and the femoral version to guide the management of ADH. However, these angular measurements
are prone to measurement errors with variable reliability.[11]
[12] Plain radiographs also lack the sensitivity to detect early osteoarthritis of the
hip.
The LCEA of Wiberg remains the oldest, widely used measurement in the evaluation of
hip dysplasia.[19] It measures the coverage of the femoral head; however, it has had modifications
undertaken over the years to improve its accuracy. Ogata et al suggest their “refined”
LCEA can more accurately determine head coverage, especially in younger children with
dysplastic hips.[20] Though there is a general agreement that an LCEA more tahn25 degrees is normal and
less than 20 degrees is consistent with dysplasia, debate remains about the characterization
of hips with LCEA between 20 and 25 degrees and its application in patients undergoing
periacetabular osteotomy.[21]
[22]
The Tönnis angle (acetabular index) or horizontal “toit externe angle” measures the
slope of the weight-bearing surface of the acetabulum or “sourcil”.[14] A normal hip has values between 3–13 and more than 10 degrees suggestive of a dysplastic
hip.[6] However, Tönnis angle is unable to evaluate anterior femoral coverage and cannot
be measured if the medial edge of the acetabular sourcil on the pelvic radiograph
is blurred.[12]
ADH has been established to be a 3D deformity and hence radiographic evaluation has
its inherent limitation bearing this concept in mind. There is an increased use of
cross-sectional imaging (CT and MRI) in the management of hip dysplasia. This involves
assessment of the coverage of the femoral head, cartilage, labrum, femoral and acetabular
version, and periarticular soft tissues. CT or MRI also provides 3D assessment of
the hip joint allowing clarification of size, shape, and orientations of the acetabular
deficiencies if present in patients with ADH.[13] It is thus helpful in surgical planning and preoperative assessment of the dysplastic
hip.[9] Various studies have highlighted the utility of cross-sectional imaging in evaluating
coverage of the femoral head and analysis of acetabular deficiencies in patients with
ADH.[6]
[9]
[13]
[23] The angles mentioned in the CT evaluation of hip dysplasia include anterior acetabular
sector angle (AASA), posterior acetabular sector angle (PASA), and the horizontal
acetabular sector angle. Some of the radiographic measurements crucial in the diagnosis
of dysplasia mentioned before can be used to some extent in cross-sectional imaging.
However, the main limiting factor is the identification of the exact slice to measure
the angles. AASA and PASA are anterior and posterior coverage angles of the femoral
head that have been described on cross-sectional imaging and are assessed on axial
images (normal AASA <50 degrees and PASA <90 degrees).[24]
[25] The measurements of these angles for hip dysplasia on MRI and CT are, however, found
to be comparable.[26]
There is an intrinsic advantage of cross-sectional imaging in the assessment of ADH
due to its versatility, 3D characterization of the hip abnormality, and use in surgical
planning. The BROH femoral offset is a linear measurement that is measured on axial
images at the tip of the greater trochanter and found to increase proportionally with
the degree of hip dysplasia. We measured these on CT but believe that this can be
analyzed on MRI too. The BROH femoral offset can be used in the diagnosis and management
of hip dysplasia. Based on our results, we propose a normal BROH femoral offset to
be less than 1 cm. BROH femoral offset of more than 1cm should be considered as dysplasia.
Limitations of the Study
There were a few limitations to this study. This was a small retrospective study.
However, the application of ICC analysis has allowed us to reinforce the reliability
of the BROH femoral offset. Also, we used only the Tönnis angle (acetabular index)
for the assessment of hip dysplasia. Further larger, prospective, cohort studies comparing
other described angular measurements will be useful to strengthen our findings and
applicability of the BROH femoral offset in the assessment of ADH.
Conclusion
The BROH femoral offset is a new, ancillary measure on cross-sectional imaging that
can be used to supplement existing indices to assess ADH. Since it is a linear measure,
it is easy to calculate and has the potential of being included as another useful
parameter in the CT scan evaluation of ADH. It has shown good intra- and interobserver
reliability in the current study in measuring anterior coverage of the femoral in
ADH. Further studies are needed to understand how the BROH femoral offset can be validated
as an additional measure in the diagnosis and monitoring of patients with ADH.
