Femoroacetabular impingement (FAI) is a pathological mechanical process within the hip joint resulting from either acquired or developmental hip morphology. It is a common source of hip pain in young and active patients, with a prevalence between 10% and 15%.1,2 The following three forms of FAI exist: pincer impingement, which is characterized by overgrowth of the acetabulum, cam impingement, which is characterized by abnormal sphericity of the femoral head, and a mixture of both types of FAI, which can lead to limited motion, joint damage, pain, and early osteoarthritis.1-6 FAI has been noted as a factor related to early onset degenerative osteoarthritis.3,5 Early diagnosis of FAI has been suggested as potentially important because it may allow for early intervention to correct biomechanical factors that are related to irreversible chondral damage.1,2
Radiographic evidence of FAI can assist in clinical decision-making, but multiple reports have highlighted asymptomatic populations with radiographic evidence of FAI.7,8 A systematic review evaluating 2,114 asymptomatic hips found that 67% had radiographic evidence of a pincer deformity, and 37% had evidence of a cam deformity.9 This review by Frank et al, along with other reviews, mainly focused on adult populations.1 Only a handful of reports have evaluated radiographic evidence of FAI in asymptomatic adolescent populations.10-12 More importantly, it is assumed that the radiographic abnormalities precede the onset of hip pain. The clinical dilemma is the determination of those specific individuals whose radiographs have parameters indicating FAI and will develop early onset degeneration. No previous study has followed asymptomatic adolescent patients to determine whether the radiographic abnormalities, present at the baseline asymptomatic visit, increase the risk of future hip pain with subsequent cartilage degeneration.
The aim of this cross-sectional study was to establish the prevalence of radiographic abnormalities indicative of FAI among asymptomatic adolescents and young adults. Using a cohort study design, we also aimed to determine whether the radiographic abnormalities obtained from the cross-sectional portion of the study predicted the development of future hip pain.
After institutional review board approval, we retrospectively reviewed the pelvic portion of standing spinal radiographs obtained from individuals referred for potential idiopathic scoliosis (IS) at a single quaternary-referral children's hospital from January 2006 through September 2009. We included all radiographs obtained from patients 12 to 25 years of age with no previous hip complaints, no known hip deformity, or diagnosis of a neuromuscular disorder. Patients with pelvic radiographs with open triradiate physes were excluded.
Radiographs were independently reviewed by a fellowship-trained pediatric orthopaedic surgeon with experience in hip preservation. The AP pelvis portions of the images were analyzed to confirm that the pelvis was centered and lacked any significant rotation in all planes (including flexion/extension). Lack of rotation was confirmed through correct alignment of the sacrum with the pubic symphysis in the coronal plane and a distance of 1 to 4.5 cm measured between the sacrococcygeal joint to the superior border of the pubic symphysis.2
Images were reviewed using (PACSWEB) the digital measurement software tools to evaluate the following five radiographic abnormalities described as potential indicators of FAI: lateral center-edge (LCE) angle of Wiberg >40° or <25°, Tönnis angle < 0° or >10°, acetabular retroversion (crossover sign with a posterior wall sign), acetabular overcoverage (crossover sign without a posterior wall sign), and anterior offset alpha angle (AOAA) (calculated using alpha angle of Nötzli >50°).13
A cohort study design was used to determine whether patients in the cross-sectional analysis developed hip pain after the baseline radiographic evaluation. Only patients with one or more clinical follow-up visits were included in the cohort analysis. Patient charts were retrospectively reviewed to identify the incidence of hip pain, defined as a report of pain in the groin during routine daily activity that may or may not increase with higher-intensity activities such as sports. Patients were followed for an average of 3.1 years (±2.8, range, 0.1 to 11.8 years) after their baseline radiograph.
Descriptive statistics were used to estimate the prevalence of the five hip radiographic abnormalities according to the number of hips included in the data set (hip level prevalence) and the number of patients included in the data set (patient level prevalence). In the cohort analysis, Cox-proportional hazard models were used to test the association between radiographic abnormalities at baseline and the incidence of hip pain. Correlation because of the inclusion of multiple hips per patient was accounted for in the statistical models. All radiographic abnormalities were tested in separate models. Radiographic variables significantly associated with onset of hip pain were tested together in the final, multivariable model. Age and sex were included as covariates in all statistical models.
A total of 233 patients (466 hips) met the inclusion criteria. Consistent with the demographics among patients referred for treatment of adolescent IS at our institution, 78% were women (181/233) and 22% were men (52/233). The mean age at the baseline radiograph was 14.7 years (±2.0, range, 11 to 20 years old). The average LCE angle was 31.9° (range, 16.0 to 51.0), and the average Tönnis angle was 4.8° (range, −12.0 to 18.7°) among all hips.
Prevalence Among Hips
Most hips included in the final analysis, 60% (281/466), presented with at least one and as many as four radiographic abnormalities (Table 1). Among hips with one or more radiographic abnormalities (n = 281, Table 1), 31%, 86/281, of the hips were associated multiple radiographic abnormalities, whereas the remaining 69% (195/281) were associated with a single abnormality.
Table 1 -
Hip and Patient Level Prevalence
||Hip Level Prevalence (n = 466), n (%)
||Patient Level Prevalence (n = 233), n (%)
| LCE angle <25° or >40°
| Tönnis <0° or >10°
| Acetabular retroversion
| Acetabular overcoverage
| Anterior offset alpha angle
|No. of parameters
LCE = lateral center edge
An LCE angle <25° or >40° was the most common abnormality, prevalent in 27% (127/466) of the hips (Table 1). A Tönnis angle <0° or >10° was the least common FAI abnormality, prevalent in only 9% (43/466) of the hips. The radiographic abnormalities also tended to present in patterns (Table 2). An LCE <25° or >40° was the most commonly isolated abnormality.
Table 2 -
Patient Level Prevalence Patterns
||Absent, n (%)
||Present in One Hip Only, n (%)
||Present in Both Hips, n (%)
|LCE angle <25° or >40°
|Tönnis angle <0° or >10°
|Anterior offset alpha angle
LCE = lateral center-edge
We also estimated the most frequent radiographic abnormality pairings. Acetabular overcoverage was most likely to present in combination with a AOAA (5%, 25/466); acetabular retroversion was most likely to present with AOAA (4%, 20/466), and a Tönnis angle <0° or >10° was most likely to present with an LCE angle <25° or >40° (3%, 16/466).
Prevalence Among Patients
At the patient level, one or more radiographic abnormalities were present in at least one hip in 72% of the patients (Table 1). The abnormalities tended to be bilateral in nature, indicating a high level of correlation between hips (Table 2).
Incidence of Future Hip Pain
The hips were followed for an average of 3.1 years after the initial baseline, radiograph. Among these hips, 13 (3%) developed hip pain. The median time to onset of hip pain was 3.3 years (range, 2.1 to 6.7 years). We tested the association between radiographic abnormalities at baseline and the hazard of hip pain (Table 3). After adjusting for age at radiograph and sex, a decreasing Tönnis angle (hazard ratio per 1-degree decrease: 1.25, 95% confidence interval, 1.10–1.41, P = 0.006) and presence of acetabular retroversion (hazard ratio: 3.55, 95% confidence interval, 1.15–10.95, P = 0.0272) were the only radiographic abnormalities significantly associated with the incidence of hip pain.
Table 3 -
Hazard of Hip Pain
|Tönnis angle: per 1° decrease
|Tönnis angle: abnormal vs normala
|No. of (+) signs
|LCE angle: per 1° increase
|Anterior offset alpha angle
|LCE angle: abnormal vs normalb
HR = hazard ratio, LCE = lateral center edge, LCL = lower control limit, UCL = upper control limit
aAbnormal >10° or <0°.
bAbnormal <25° or >40°. The p-values listed indicate the significance of these statistics.
In a large cohort of 233 adolescent individuals without hip symptomatology, 60% of hips had at least one radiographic abnormality indicative of FAI. Among these hips, 69% of hips were associated with a single abnormality, whereas 31% of hips were associated with multiple radiographic abnormalities. Midterm follow-up revealed that a decreased Tönnis angle and the presence of acetabular retroversion were predictive of future hip pain.
Our study represents a large cohort of initially asymptomatic adolescent and young adult individuals with standard pelvis radiographs (466 hips). Schmitz et al11 evaluated hip morphology (cam, crossover, coxa profunda, ischial spine sign, lateral center edge, and alpha angle) in 90 asymptomatic patients (180 hips) using an EOS Imaging technique.11 Coxa profunda was present in 81.7% of the hips. Nepple et al14 have discouraged the use of coxa profunda to diagnose pincer impingement because of its high prevalence in asymptomatic hips. Based on these factors, we did not include coxa profunda in the current study. Schmitz et al11 also observed that acetabular overcoverage (as measured by the lateral center edge angle) was present in 15% of asymptomatic hips. Acetabular overcoverage, defined in our study as the presence of the crossover sign without a posterior wall sign, was present in 13% of the hips in our study. Acetabular over or under coverage, LCE <25° or >40°, was present in 27% of the hips in our study. The current study corroborates the high prevalence of radiographic FAI abnormalities among asymptomatic individuals.
Our results also demonstrate a high level of within patient correlation, as indicated by the high prevalence of radiographic abnormalities present bilaterally (Table 2). Correlation between affected and unaffected hips, as represented by bilateral presentation of FAI signs, has also been noted in related studies.8,14 Specific radiographic abnormalities also tended to be correlated. Acetabular overcoverage and an AOAA were the most prevalent combination of abnormalities, present in 5% of hips.
A major limitation of the previous cross-sectional designs used to estimate the prevalence of FAI abnormalities in asymptomatic hips has been the lack of follow-up data to determine whether abnormalities are predictive of future hip pain. Assuming that anatomic variants contribute to future hip pain, the anatomic abnormalities must precede the onset of hip pain. FAI abnormalities could represent latent hip pain. Therefore, we followed patients to determine whether patients sought care for hip pain after the baseline radiograph. In the final multivariable analysis, after adjusting for age and sex, a decreasing Tönnis angle and the presence of acetabular retroversion were significantly associated with the onset of hip pain. Although not significant, hips that developed pain were associated with a higher number of FAI abnormalities, 1.2, compared with hips that did not develop pain, 0.8. These findings suggest that FAI radiographic abnormalities may have utility in the overall clinical and radiographic assessment and prediction of future hip pain.
This study includes several limitations. The cross-sectional analysis was based on AP spinal radiographs among individuals referred for potential IS, which may not represent a true cross-section of asymptomatic subjects. In addition, the review of radiographic imaging was limited to AP radiographs, which is known to be limited in the evaluation of the complex three-dimensional anatomy of the pelvis and proximal femur. Specific to cam-type morphology, previous studies have demonstrated that evaluation of sphericity of the femoral head is assessed best using the Dunn view.15 We may be underreporting the prevalence of AOAA in this study cohort using only the AP pelvic images; however, it should be noted that our prevalence of AOAA is similar to other studies that used advanced imaging such as MRI and EOS.7,11 Finally, it should be noted that coexisting hip and spine pathologies have been reported in the literature, suggesting that patients with IS may have a predisposition to the development of hip issues.16,17
We observed a high prevalence of radiographic abnormalities potentially indicative of FAI in asymptomatic adolescents, further demonstrating that that the observed radiographic abnormalities are associated with a low level of specificity for FAI (not effective at ruling out the presence of hip pathology among asymptomatic hips). An LCE angle <25° or >40° was the most common abnormality, present in greater than one-quarter of the of hips. The results do not support sole reliance on radiographic abnormalities. However, radiographic abnormalities may be useful in predicting future hip pain. In our study, an abnormal Tonnis angle and acetabular retroversion were significantly associated with the development of future hip pain. These results support the judicious use of both clinical and radiographic factors in the treatment of FAI to avoid overtreatment.
1. Yamasaki T, Yasunaga Y, Shoji T, et al.: Inclusion and exclusion criteria in the diagnosis of femoroacetabular impingement. Arthroscopy 2015;31:1403-1410.
2. Tannast M, Siebenrock KA, Anderson SE: Femoroacetabular impingement: Radiographic diagnosis—What the radiologist should know. AJR Am J Roentgenol 2007;188:1540-1552.
3. Ganz R, Parvizi J, Beck M, et al.: Femoroacetabular impingement: A cause for osteoarthritis of the hip. Clin Orthop Relat Res 2003;417:112-120.
4. Murphy S, Tannast M, Kim YJ, et al.: Debridement of the adult hip for femoroacetabular impingement: Indications and preliminary clinical results. Clin Orthop Relat Res 2004;429:178-181.
5. Tanzer M, Noiseux N: Osseous abnormalities and early osteoarthritis: The role of hip impingement. Clin Orthop Relat Res 2004;429:170-177.
6. Jager M, Wild A, Westhoff B, et al.: Femoroacetabular impingement caused by a femoral osseous head-neck bump deformity: Clinical, radiological, and experimental results. J Orthop Sci 2004;9:256-263.
7. Hack K, Di Primio G, Rakhra K, et al.: Prevalence of cam-type femoroacetabular impingement morphology in asymptomatic volunteers. J Bone Joint Surg Am 2010;92:2436-2444.
8. Laborie LB, Lehmann TG, Engesaeter IO, et al.: Prevalence of radiographic findings thought to be associated with femoroacetabular impingement in a population-based cohort of 2081 healthy young adults. Radiology 2011;260:494-502.
9. Frank JM, Harris JD, Erickson BJ, et al.: Prevalence of femoroacetabular impingement imaging findings in asymptomatic volunteers: A systematic review. Arthroscopy 2015;31:1199-1204.
10. Philippon MJ, Ho CP, Briggs KK, et al.: Prevalence of increased alpha angles as a measure of cam-type femoroacetabular impingement in youth ice hockey players. Am J Sports Med 2013;41:1357-1362.
11. Schmitz MR, Bittersohl B, Zaps D, et al.: Spectrum of radiographic femoroacetabular impingement morphology in adolescents and young adults: An EOS-based double-cohort study. J Bone Joint Surg Am 2013;95:e90.
12. Siebenrock KA, Ferner F, Noble PC, et al.: The cam-type deformity of the proximal femur arises in childhood in response to vigorous sporting activity. Clin Orthop Relat Res 2011;469:3229-3240.
13. Meyer DC, Beck M, Ellis T, et al.: Comparison of six radiographic projections to assess femoral head/neck asphericity. Clin Orthop Relat Res 2006;445:181-185.
14. Nepple JJ, Lehmann CL, Ross JR, et al.: Coxa profunda is not a useful radiographic parameter for diagnosing pincer-type femoroacetabular impingement. J Bone Joint Surg Am 2013;95:417-423.
15. Meyer DC, Beck M, Ellis T, et al.: Comparison of six radiographic projections to assess femoral head/neck asphericity. Clin Orthop Relat Res 2006;445:181-185.
16. Segreto FA, Vasquez-Montes D, Brown AE, et al.: Incidence, trends, and associated risks of developmental hip dysplasia in patients with early onset and adolescent idiopathic scoliosis. J Orthop 2018;15:874-877.
17. Banerjee P, Mclean CR: Femoroacetabular impingement: A review of diagnosis and management. Curr Rev Musculoskelet Med 2011;4:23.