INFLUENCE OF AGE ON PARAMETERS FOR FEMOROACETABULAR IMPINGEMENT AND HIP DYSPLASIA IN X-RAYS

ABSTRACT Objective: While several radiographic parameters have been established to describe the geometry and pathology of the hip, their reference values and clinical significance remain a matter of dispute. The present study tests the hypothesis that age has a relevant impact on radiographic hip parameters. Method: Pelvic antero-posterior views were measured for CE angle, Sharp’s angle, acetabular depth-to-width ratio, femoral head extrusion index, roof obliquity, caput-collum-diaphyseal (CCD) angle, and Murray’s femoral head ratio, and the values obtained were correlated with age. Results: Significant weak and moderate linear correlations (all Ps<0.001) were observed between age and CE angle (ρ=0.31), Sharp’s angle (ρ=-0.38), extrusion index (ρ=-0.22), CCD angle (ρ=-0.15), depth-to-width ratio (ρ=-0.38), and roof obliquity (ρ=-0.19), while Murray’s femoral head ratio (ρ=0.05; P=0.274) was not associated with age. Interestingly, the parameters describing the acetabulum all showed a relevant increase in coverage with age, leading to CE-angles well beyond 40° and a Sharp’s angle below 35° in a large portion of asymptomatic older adults. Conclusion: While a decrease in CCD angle with age is described in most orthopedic textbooks, the changes observed with age in acetabular geometry far exceed those measured at the femoral head-neck junction. We recommend considering these alterations that may be attributable to age when formulating a radiographic diagnosis. Level of Evidence III, Diagnostic Studies - Investigating a Diagnostic Test.


INTRODUCTION
Wiberg center-edge (CE) angle <25°, femoral head extrusion index >25%, Sharp's angle >40°, acetabular roof obliquity angle >10°, and acetabular roof obliquity angle >10° have been established as a classic sign of hip dysplasia, 1,2 a predominant prearthrotic deformity. In recent years, other changes in hip geometry have been added to describe pathological anatomy of the hip joint. Terms such as excessive overcoverage, acetabular retroversion, and abnormal head-neck junction ("pistol-grip deformity") are now the focus of scientific interest. These anatomical changes can cause two forms of femoroacetabular impingement (FAI) leading to early hip pain and osteoarthritis (OA): 3-6 increased acetabular coverage ("pincer impingement") causes damage at the acetabular rim, and an enlarged femoral neck ("cam impingement") destroys the antero-superior area of the acetabulum. 4,7 After detailed physical examination, a diagnosis of hip dysplasia or FAI is largely based on appropriate imaging. Different radiographic parameters in pelvic antero-posterior views have been established to detect these pathologies. Some have become quite popular over the past few years, such as the "pistol-grip deformity," which is quantified as Murray's femoral head ratio. [8][9][10] In these conditions, over time progressive degenerative changes lead to osteophytes, narrowing of the joint space, subchondral sclerosis, and deformity of the bone ends, which in turn have a negative impact on the different radiographic parameters themselves. 11 When evaluating radiographic parameters, however, age also needs to be considered as a factor. The caput-collum-diaphyseal (CCD) angle, for example, is known to decrease significantly with increasing age. 12 Moreover, aged cartilage usually shows non-progressive changes: decreased cellularity, reduced proteoglycan concentration, and reduced mechanical properties. The present study was performed to investigate the hypothesis that not only OA, but also age itself has an impact on different radiographic parameters used to describe hip dysplasia or FAI.

MATERIAL AND METHODS
We analyzed our data bank for all pelvic views performed in our institution between January 1, 2006 and December 31, 2011. The images were analyzed by a specialist registrar from our Department of Radiology. To avoid the negative influence of pelvic tilt and rotation on radiographic parameters, we included only standardized pelvic antero-posterior radiographs in the measurements. The mean distance between the tip of the coccyx and the middle of the symphysis was 32 mm for men and 47 mm for women, and the teardrop sign appeared to be symmetrical. 3 To evaluate the severity of OA of the hip, we used the classification by Kellgren and Lawrence. Only radiographs with no signs of OA of the hip (Kellgren and Lawrence 0) were included in the present study. The pelvic antero-posterior views were measured for CE angle, Sharp's angle, acetabular depth-to-width ratio, femoral head extrusion index, acetabular roof obliquity angle, CCD angle, and Murray's femoral head ratio 8,10,13-17 (for details concerning the measurement of these parameters, see Figure 1). Alpha-angle was not evaluated since recent studies have shown only a limited reliability for conventional radiographs and recommended it for 3D imaging techniques instead (reviewed by Sutter et al. 18 ). In cases with unilateral total hip arthroplasty, fracture, or dysplasia (Crowe II-IV), 19 only the contralateral side was measured. Full approval was obtained from the departmental, institutional, and ethical review boards (project number 025/2014R) before the study began. Due to the retrospective character of the study, no informed consent was obtained.

Statistical analysis
In order to account for repeated measurements of both hips on the patient level, we conducted analyses by using summarized values for both sides. Categorical variables are presented as frequencies and percentages, and continuous variables as means and standard deviations. The strength of linear associations between age and radiographic parameters was assessed by Pearson's correlation coefficient (ρ). Linear or logistic regression analyses were conducted to describe the influence of age on the radiographic measurements. We considered the influence of possible confounding factors by calculating the corresponding odds ratio (OR) and 95% confidence interval (CI). Because of the descriptive character of the study, no alpha adjustment was performed with a two-sided significance level of 0.05. Statistical analyses were conducted using SPSS version 21 (IBM Corp., Armonk, NY) and R software, version 3.1.0 (R Foundation for Statistical Computing, Vienna).

RESULTS
Of all the pelvic radiographs performed in our institution between January 1, 2006 and December 31, 2011, those of 525 patients met all inclusion criteria. In 245 patients both sides were measured, and in 280 patients only the left (n=122) or right side  The present study was performed to assess the impact of age itself on various radiographic parameters used to diagnose FAI and hip dysplasia. We showed that age has a significant influence on many of these measurements. With respect to the strength of the correlations with age, it is noteworthy that the decrease in CCD angle with age is described in most orthopedic textbooks.

DISCUSSION
Hip dysplasia with reduced CE angle, decreased depth-to-width ratio, and increased extrusion index is widely accepted as the main reason for OA in young adults. 11,13,20 Recently, other changes in acetabular geometry with excessive local or global overcoverage and reduced head-neck offset have been detected as further major causes for progressive hip pain and early OA of the hip. However, the predictability of these findings for early OA onset remains a matter of debate. On the one hand, abnormal hip morphology with either "classic" acetabular dysplasia or impingement due to excessive overcoverage of the femoral head, acetabular retroversion, or an abnormal head-neck junction has been reported in approximately 51%-97% of all cases of hip OA. 9,13 On the other hand, Laborie et al. 9 demonstrated a prevalence of radiographic findings for FAI in the majority of a cohort of 2081 healthy adults. De Bruin et al. 21 found only 58 hip radiographs devoid of signs for FAI in a sample of 522 hips not clinically suspected of FAI. This leaves room to discuss the extent to which these parameters can be used as predictive factors for OA. Some of these findings may also be a consequence of early OA onset. For example, whether the posterior head tilt in osteoarthritic hips should be considered an acquired deformity created by the formation of osteophytes is currently under discussion. This has led to the suggestion that radiographic signs to detect FAI and hip dysplasia should be used carefully in patients with OA of the hip. 11  Nevertheless, except for the non-correlating Murray's head-neck ratio, CCD angle showed the weakest correlation of all parameters analyzed (ρ=-0.15) in this study. It is noteworthy that while CCD angle and Murray's head-neck ratio describe femoral changes, the parameters with much stronger correlation are all linked to the shape of the acetabulum. So far it is well known that Wiberg's CE angle increases during skeletal growth. In adults, however, only a very weak increase has been described. 22 The increase in acetabular coverage might seem accidental when just looking at one single parameter. The values observed are, however, all coherent. Just as Wiberg's angle increases due to better coverage, Sharp's angle, the acetabular roof obliquity angle, and the femoral head extrusion index decrease. And even though clinical experience might indicate that higher depth-to-width ratios are observed in older people (as in an osteoarthritic coxa profunda, for example), this conjecture is deceiving, since it only applies to osteoarthritic hip joints. With an increase in acetabular coverage, the resulting increase in width of the fossa exceeds the increase in depth, leading to a decrease of the ratio. How the radiological increase in acetabular coverage is produced still needs to be clarified. One possible explanation might be ossification of the labral base 23 leading to false interpretation of the actual acetabular rim. It is more likely, however, that in the zone of maximum biomechanical stress the pelvis reacts over time by strengthening the apical zone. It is essential to realize that this acetabular increase leads to CE-angles well beyond 40° and a Sharp's angle below 35° in a large portion of elderly people. These radiographic angles would usually be considered as FAI, 7,16 but remain asymptomatic in many cases. Since the long-term outcome of surgical resurfacing of the head-neck junction still varies widely, we suggest considering age-related alterations before formulating a radiographic diagnosis from measured values.

Study limitations
Some degenerative changes can almost always be observed in pelvic radiographs of older patients with hip pain. It is consequently difficult to make a clear distinction between whether these changes are attributable to ageing of the joint or to degeneration. In this study, however, only radiographs of patients not diagnosed with or treated for OA were included to minimize this effect. We did not test intra-or inter-observer reliability, although strong inter-and intra-observer discrepancies are known to occur in radiographic measurements to diagnose dysplasia and FAI. Even so, the lack of this evaluation should not have affected our results, since the only observer was blinded to patient age and considered all cases equally. Certain hip parameters may also be affected by a change in posture in elderly patients, but such a change in postural pelvic orientation would also affect joint function. Possible differences due to different projection angles in the radiographs may be solved in the future by using three-dimensional imaging techniques with volume renderings.

CONCLUSION
Patient age has a relevant impact on various radiographic parameters to detect FAI and hip dysplasia. While femoral CCD angle decreases only marginally, acetabular coverage increases considerably over time. Although these changes are in many cases negligible, especially when borderline values are found, alterations that may be attributable to age should be considered.