Abstract
Background
The anatomy of the posterior wall of the acetabulum is important for the hip stability. We wanted to know whether differences can be observed.
Material and methods
On lateral 3D pelvic CT reconstruction of the pelvis two types were identified. On axial CT-images acetabular angles were determined.
Results
We observed 35 type I acetabular and 61 type II acetabular. Posterior acetabular sector angle was 114° in type I and 94° in type II acetabular (p < 0.01). At an cut-off angle of 100° we could predict the type of acetabulum.
Conclusions
We could describe reference values for the posterior wall to distinguish two morphological types.
Keywords: Acetabulum, Posterior wall, Anteversion acetabular angle, Posterior sector acetabular angle
1. Introduction
Acetabulum morphology and version has been described thoroughly. Reference values according to Tönnis and Heinecke for the acetabulum anteversion are between 15 and 20°.1 Further analysis of the acetabular shape have been performed by Valera et al. who described the anterior and posterior acetabular coverage in healthy patients younger than 55 years without higher degrees of osteoarthritis.2 Both, acetabular version and posterior wall are known to be associated with impingement syndrome and the development of osteoarthritis.1, 2, 3, 4, 5
Circumferential osseous extension in the acetabular rim occurs regardless of acetabular coverage and the degree of osteoarthritis and are maximal between the 11 and 1 o'clock positions.6,7 The mechanism is described by Seldes et al.8 as a result of labral ossification due to recurrent mechanical compression in case of pincer impingement. Acetabular overcoverage, defined as pincer impingement, has been reported to be associated with femoroacetabular impingement and osteoarthritis.4,9,10 However, this is often related to the anterior lip of the acetabulum limiting the hip flexion and not the posterior wall.4
Important information for understanding the bone structure and cortical thickness of the pelvis in relation to the acetabulum was gained by Krebs et al.3 However, as these values were gained by cadaveric studies the clinical comparison remains difficult.
So far, a CT based method to describe the anatomical variances of the acetabular posterior wall in a representative collective without regard to clinical symptoms has not been described. This is especially important as besides to the labral stabilization of the hip the osseous anteversion and shape of the posterior wall is shown to be heterogeneous, believed to be influence the rate of arthritis and might be important for patient's hip stability after total hip arthroplasty or dual mobility arthroplasty.
We therefore wanted to examine I) whether differences in posterior wall configuration can be observed in 3D reconstructions and II) whether these differences can be seen in axial CT scan images of the acetabulum.
2. Methods
We retrospectively analyzed 58 consecutive pelvic CTs and excluded all hips with an acetabular or pelvic ring fracture. All CTs were performed using a Somatom Force (Siemens Healthcare GmbH, Erlangen, Germany). Image slices were 0,6 mm. A 3D reconstruction was done using the Visage 7.1.11 software (Visage Imaging GmbH, Berlin, Germany).
First, we used an exact lateral view of a 3D reconstruction. This was defined by projecting both anterior superior iliac spine, obturator foramen and the greater sciatic notch above each other. We than inspected the posterior wall and determined three types by the feasibility to look onto the posterior articular surface. We distinguished between Type I: defined as invisible posterior acetabular joint surface and Type II: with visible posterior acetabular joint surface (Fig. 1).
Fig. 1.
Posterior wall classification.
3D-reconstrucion of pelvic CT-scan of the pelvis with true lateral view defined as projection of anterior superior iliac spine, obturator foramen and the greater sciatic notch above each other. Type I: defined as a non-visible posterior acetabular surface with roofing of the posterior acetabular wall, Type II: defined as a visible posterior acetabular surface without roofing of the posterior acetabular wall.
We then used strict axial images of the pelvis. This was controlled in all three planes and if necessary corrected. An intercapital center line (ICL) was drawn on true axial images through both femoral heads at the point of biggest diameter in all three planes. Afterwards an orthogonal line to the ICL, the ICL90 was drawn. A line between the anterior and posterior acetabular lip of the acetabulum was drawn, the anteversion line (AVL). The angle between the ICL90 and AL was measured to determine the acetabular anteversion angle (AAA°). The posterior sector acetabular angle (PASA°) as described by Valera et al. was determined by measuring the angle between the ICL and a line from the femoral head center to the lateral edge of the posterior wall.2 A new posterior wall angle (PWA), to describe the posterior wall formation in the axial plane, was measured by using the angle between the ICL90 and the tangent to the posterior articular surface area (Fig. 2). The femoral head coverage was determined using the ICL and the AVL. The part within the acetabulum to the AVL was divided by the whole femoral head diameter.
Fig. 2.
Description of CT based wall measurements.
A: Acetabular roofing %: An intercapital centre line (ICL) was drawn on true axial images. A line between the anterior and posterior acetabular lip of the acetabulum was drawn and the distance of the covered part of the femoral head was divided by the whole diameter of the femoral head. B: AAA: anteversion acetabular angle. The ICL was drawn on true axial images. A line between the anterior and posterior acetabular lip of the acetabulum was drawn and the angle to the orthogonal line to the ICL (ICL90) was measured. C: PASA: the posterior acetabular sector angle was determined my measuring the angle between the ICL and a line from the femoral head centre to the lateral edge of the posterior wall. PWA: posterior wall angle. The ICL was drawn on true axial images. An orthogonal line to the ICL (ICL90) as described by Valera et al. was added. The posterior joint line (PJL) was determined as a tangent to the posterior articular surface area and the angle between ICL90 and PJL was measured.
2.1. Statistical analysis
Data were tested for normal distribution using the Shapiro-Wilk test. For comparative statistics, in case of normal distribution T-Test was used. If data were not normal distributed Mann-Whitney U-Test was used. Data are presented as median with 25% and 75% quantile or if normal distributed as mean ± standard deviation. A receiver operating characteristic (ROC) analysis, including the area under the curve (AUC) calculation, was performed for all angles to predict type I the acetabulum. An optimal cut point for the two best predictors was calculated. Sensitivity and specificity were calculated for the two best predictors at the optimal cut points. Statistical significance was considered at p < 0.05.
3. Institutional review board approval (IRB approval)
An IRB Approval is not required, because the investigator conducting this research obtained I) no data through interventional interaction and II) no identifiable private information.
4. Results
We included 58 acetabular. Mean patients age was 63 years (range: 16–96), AAA° 21° (17–27), PASA° 100° (92–112), PWA° 72° (66–79) and acetabular roofing was 42° (33–50). We observed 35 type I acetabular with a mean age of 74 years (20–92) and 61 type II acetabular with a mean age of 58 years (16–96). PASA° was 114° (103–121) in type I and 94° (90–102) in type II acetabular (p < 0.01). PWA° was 80° (76–90) in type I and 70° (63–73) in type II acetabular (p < 0.01). Out of 96 hips 41 and 55 were from female and male patients respectively. (Table 1, Fig. 3).
Table 1.
Measured angles on axial CT scans: Data are given as median with 25% and 75% interquartile range. AAA°: anteversion acetabular angle. PASA°: posterior acetabular sector angle, PWA°: posterior wall angle, Acetabular roofing %: Femoral head coverage by the acetabulum. Type I acetabular showed significant higher values than type II acetabular.
| Total (n = 96) | Type I (n = 35) | Type II (n = 61) | P-Value Type I vs. Type II | |
|---|---|---|---|---|
| AAA (°) | 21 (17–27) | 27 (21–30) | 20 (15–24) | p < 0,01 |
| PASA (°) | 100 (92–112) | 114 (103–121) | 94 (90–102) | p < 0,01 |
| PWA (°) | 72 (66–79) | 80 (76–90) | 70 (63–73) | p < 0,01 |
| Acetabula roofing (%) | 42 (33–50) | 51 (45–55) | 36 (32–43) | p < 0,01 |
Fig. 3.
Acetabular angle in type I and II acetabular.
grey box: Type I, white box: Type II, **: Significant difference p < 0.01, AAA: anteversion acetabular angle. PASA: posterior acetabular sector angle, PWA: posterior wall angle, acetabular roofing (%): Femoral head coverage by the acetabulum. Type I acetabular show significant higher values for PWA°, PASA°, AAA° and acetabular roofing than type II acetabular (p < 0.01).
Mean age in female patients was 66 ± 25 years and in male patients 62 ± 22 years (p < 0.01). AAA° was 25° (20–28) in female and 20° (15–25) in male patients (p < 0.01). PASA° was 110° (97–117) in female and 95° (90–106) in male patients (p < 0.01). PWA° was 76 (71–84) in female and 70 (63–77) in male patients (p < 0.01). Acetabular roofing was 48% (37–52) in female and 41% (33–50) in male patients (p < 0.01). Acetabular angle are higher in case of higher age (p < 0.01) (Fig. 5, Fig. 7).
Fig. 5.
Acetabular angles according to sex.
Acetabular angles according to sex. AAA: anteversion acetabular angle. PASA: posterior acetabular sector angle, PWA: posterior wall angle, Acetabular roofing %: Femoral head coverage by the acetabulum. Reverence line at 70°. **: Female acetabular showed significant higher values for PWA°, PASA°, AAA° and femoral head coverage than male acetabular (p < 0.01).
Fig. 7.
Distribution of acetabular angles in three age groups.
Distribution of acetabular angles in three age groups. AAA: anteversion acetabular angle. PASA: posterior acetabular sector angle, PWA: posterior wall angle, Acetabular roofing %: Femoral head coverage by the acetabulum. Means of femoral head coverage, AAA°, PASA° and PWA° are higher by rising age (p < 0.01).
Signs of osteoarthritis on axial CT images were seen in 70 acetabular (age: 73 ± 17) and no signs in 26 acetabular (age: 40 ± 21). If signs of osteoarthritis were present AAA°, PASA°, PWA° and femoral head coverage were higher than in acetabular without signs of osteoarthritis (p < 0.01 each) (Table 2, Fig. 4; Fig. 5).
Table 2.
Data are given as median with 25% and 75% interquartile range. AAA: anteversion acetabular angle. PASA: posterior sector acetabular angle, PWA: posterior wall angle, Acetabular roofing (%): Femoral head coverage by the acetabulum. All measured angles showed significant higher values in case of osteoarthritis.
| Osteoarthritis (n = 70) | No Osteoarthritis (n = 26) | P-Value Type I vs. Type II | |
|---|---|---|---|
| AAA (°) | 22 (19–27) | 17 (14–22) | p < 0,01 |
| PASA (°) | 105 (95–115) | 91 (87–97) | p < 0,01 |
| PWA (°) | 74 (70–80) | 69 (63–74) | p < 0,01 |
| Acetabular roofing (%) | 47 (40–52) | 33 (28–41) | p < 0,01 |
Fig. 4.
Acetabular angle in acetabular with- and without signs of osteoarthritis.
Acetabular angle in acetabular with- and without signs of osteoarthritis. AAA: anteversion acetabular angle. PASA: posterior sector acetabular angle, PWA: posterior wall angle, Acetabula roofing %: Femoral head coverage by the acetabulum. Reverence line at 70°. **: Acetabular with signs of osteoarthritis show significant higher values for PWA°, PASA°, AAA° and femoral head coverage than acetabula without signs of osteoarthritis (p < 0.01).
PASA° and PWA° were highly correlated (81%, p < 0.01). Femoral head coverage showed a high correlation with PASA° and PWA (73% and 63 respectively, p < 0.01). Age was correlated with PWA°, PASA°, AAA° and for femoral head coverage (p < 0.01) (Fig. 7).
At an angle of 100° the PASA had a 94,3% sensitivity and a specificity of 73,8% to distinguish between type I and II acetabular. PWA° bigger than 71,9° has a sensitivity of 94,3% and a specificity of 63,9% to distinguish between type I and II acetabular (Fig. 6). Neither AAA° nor acetabular roofing showed good sensitivity or specify to distinguish between type I and II acetabular (Fig. 8).
Fig. 6.
ROC Curve of acetabular angles.
ROC Curve of acetabular angles. Blue: Roofing of femoral head in the acetabulum in %, AAA: anteversion acetabular angle. PASA: posterior acetabular sector angle, PWA: posterior wall angle. At an angle of 100° the PASA had a 94,3% sensitivity and a specificity of 73,8% to distinguish between type I and II acetabular. PWA° bigger than 71,9° has a sensitivity of 94,3% and a specificity of 63,9% to distinguish between type I and II acetabular. Neither AAA° nor acetabular roofing showed good sensitivity or specify to distinguish between type I and II acetabular.
Fig. 8.
Correlation of acetabular angles.
Correlation of acetabula angles. AAA°: anteversion acetabular angle. PASA°: posterior acetabular sector angle, PWA°: posterior wall angle, Acetabular roofing %°: Femoral head coverage by the acetabulum. All angles showed high correlations.
A power analysis was performed for AAA°, PASA° and PWA°. For PASA the actual power between Type I and II was 87% for AAA°, 100% for PASA° and 99% for PWA° given alpha error of 0.05%, and the given sample size and calculated effect size.
5. Discussion
The given data show reference values for the posterior wall anatomy in a heterogeneous patient collective, randomly chosen, without regard to prior complains of any kind of hip pain. We could distinguish two kinds of posterior wall anatomy which can be described either by a simple axial few on the acetabulum or by measuring acetabular roofing, AAA°, PASA° and PWA°. A higher degree of acetabular roofing, AAA°, PASA° and PWA° indicate type I acetabular. An PASA° of 100° indicates type I acetabular with the highest sensitivity of 94.3% and good specificity of 73.8°. The clinical relevance needs to be evaluated in further analysis.
Our data are in line with the data shown by Tönnis and Heinecke for the acetabular anteversion showing that we shoes a representative collective.1 As Valera et al. only included patients younger than 55 years of age and without higher degrees of osteoarthritis we could have expected different values. Within 57 healthy control patients they observed a mean AAA° of 21 ± 18° and a mean PASA of 96 ± 8°. Reference values from the literature were given as 24 ± 5 for AAA and 103 ± 9 for PASA°.11,12 However, these reference values have been obtained in a cohort of patients with cam impingement by Barton et al. (12). Interestingly our whole study collective showed similar values as observed in the reference group of healthy patients by Valera et al. whereas the reference values obtained by Barton et al. in patients with impingement syndrome are more similar to those in type I from our study collective.
Both, acetabular version and posterior wall are known to be associated with impingement syndrome and the development of osteoarthritis.1, 2, 3, 4, 5 Interestingly patients with type II acetabular, with a mean age of 58 years, were significantly younger than patients with type I acetabular with a mean age of 74 years. Furthermore, of any signs of osteoarthritis like subchondral sclerosis was observed, femoral head coverage and PWA°, PASA°, AAA° were higher than in acetabular without signs of osteoarthitis. Therefore, we believe that in some part the difference between both groups can be explained by a higher degree of osteoarthritis. However, young patients without any sign of osteoarthritis and with high femoral head coverage, PWA°, PASA° and AAA° show that type I acetabular are independent of osteoarthritis. Clear data are missing and a correlation for instance with conventional x-rays classified by Kellgren-Lowrence Score could help to elucidate this observation. Due to the exploratory, retrospective nature we do not have any clinical data to correlate the acetabular morphology. However, as shown by Beck et al. circumferential osseous extension in the acetabular rim occurs regardless of acetabular coverage and the degree of osteoarthritis.7 Nevertheless, we assume a reduced range of motion in type I acetabular compared to type II acetabular. Acetabular overcoverage as a morphological part of the pincer impingement is higher in type I acetabular than in type II acetabular and we therefore assume a higher degree amount of pincer impingement in type I acetabular.
As the posterior wall and acetabular version might be a protective parameter against hip dislocation the clinical relevance has to be evaluated by further clinical analysis.
We conclude that we could describe the posterior wall anatomy giving reference values in a representative study collective. We could define two posterior wall types with different amount of femoral head coverage and higher PASA° and PWA°, AAA° and acetabular roofing in type I acetabular than in type II acetabular. The clinical relevance i.e. for impingement or femoral head stability needs to be analyzed in further studies.
Limitations of this study are the retrospective nature of this study. Whereas many studies concentrated on general acetabular morphology this is the first study that presents comprehensive, detailed data on the posterior wall morphology and the correlation of PASA°, PWA°, AAA° and acetabular roofing with age, gender, and osteoarthritis.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Declaration of competing interest
The Authors declare that there is no conflict of interest.
Contributor Information
Tilman Graulich, Email: graulich.tilman@mh-hannover.de, graulich.tilman@mh-hannover.de.
Pascal Graeff, Email: graeff.pascal@mh-hannover.de.
Stine Nicolaides, Email: nicolaides@web.de.
Marco Haertle, Email: haertle.marco@mh-hannover.de.
Mohamed Omar, Email: omar.mohamed@mh-hannover.de.
Christian Krettek, Email: krettek.christian@mh-hannover.de.
Emmanouil Liodakis, Email: liodakis.emmanouil@mh-hannover.de.
References
- 1.Tönnis M., Heinecke E. Diminished femoral antetorsion syndrome. A cause of pain and osteoarthritis. J Pediatr Orthop. 1991 Jul-Aug;11(4):419–431. doi: 10.1097/01241398-199107000-00001. [DOI] [PubMed] [Google Scholar]
- 2.Valera M., Ibáñez N., Sancho R., Llauger J., Gich I. Acetabular overcoverage in the horizontal plane: an underdiagnosed trigger of early hip arthritis. A CT scan study in young adults. Arch Orthop Trauma Surg. 2018;138:73–82. doi: 10.1007/s00402-017-2811-y. [DOI] [PubMed] [Google Scholar]
- 3.Krebs V., Incavo S.J., Bs W.H.S. The anatomy of the acetabulum what is normal? Clin Orthop Relat Res. 2009 Apr;467(4):868–875. doi: 10.1007/s11999-008-0317-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Genovese E., Spiga S., Vinci V. Femoroacetabular impingement: role of imaging. Musculoskel Surg. 2013 Aug;97(Suppl 2):S117–S126. doi: 10.1007/s12306-013-0283-y. [DOI] [PubMed] [Google Scholar]
- 5.García-Rey E., García-Cimbrelo E., Cruz-Pardos A. Cup press fit in uncemented THA depends on sex, acetabular shape, and surgical technique. Clin Orthop Relat Res. 2012 Nov;470(11) doi: 10.1007/s11999-012-2381-9. 3014-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Watarai K., Kimura F., Kadono Y. Complete circumferential osseous extension in the acetabular rim occurs regardless of acetabular coverage. Clin Orthop Relat Res. 2017;475:2074–2080. doi: 10.1007/s11999-017-5381-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Beck M., Kalhor M., Leunig M., Ganz R. Hip morphology influences the pattern of damage to the acetabular cartilage: femoroacetabular impingement as a cause of early osteoarthritis of the hip. J Bone Joint Surg Br. 2005 Jul;87(7):1012–1018. doi: 10.1302/0301-620X.87B7.15203. [DOI] [PubMed] [Google Scholar]
- 8.Seldes R.M., Tan V., Hunt J., Katz M., Winiarsky R., Fitzgerald R.H., Jr. Anatomy, histologic features, and vascularity of the adult acetabular labrum. Clin Orthop Relat Res. 2001 Jan;(382):232–240. doi: 10.1097/00003086-200101000-00031. [DOI] [PubMed] [Google Scholar]
- 9.Hack K., Di Primio G., Rakhra K., Beaulé P.E. Prevalence of cam-type femoroacetabular impingement morphology in asymptomatic volunteers. J Bone Joint Surg Am. 2010 Oct 20;92(14):2436–2444. doi: 10.2106/JBJS.J.01280. [DOI] [PubMed] [Google Scholar]
- 10.Ganz R., Leunig M., Leunig-Ganz K., Harris W.H. The etiology of osteoarthritis of the hip: an integrated mechanical concept. Clin Orthop Relat Res. 2008 Feb;466(2):264–272. doi: 10.1007/s11999-007-0060-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Anda S., Svenningsen S., Dale L.G., Benum P. The acetabular sector angle of the adult hip determined by computed tomography. Acta Radiol Diagn. 1986 Jul-Aug;27(4):443–447. doi: 10.1177/028418518602700415. [DOI] [PubMed] [Google Scholar]
- 12.Barton C., Salineros M.J., Rakhra K.S., Beaulé P.E. Validity of the alpha angle measurement on plain radiographs in the evaluation of cam-type femoroacetabular impingement. Clin Orthop Relat Res. 2011 Feb;469(2):464–469. doi: 10.1007/s11999-010-1624-x. [DOI] [PMC free article] [PubMed] [Google Scholar]