Dysplastic Hip Joint Conguration After Bernese Periacetabular Osteotomy

Background: Residual hip dysplasia is one of the factors contributing to early hip joint osteoarthritis. The main problems caused by residual dysplasia are pain and instability of the hip joint caused by the lack of sucient bony covering of the femoral head. The aim of this work was to radiologically assess the conguration change of a dysplastic hip joint after surgical treatment using the Bernese periacetabular osteotomy procedure. Methods: We assessed the radiological parameters of patients with hip dysplasia treated by Bernense periacetabular osteotomy by performing a digital antero-posterior pelvis X-ray: central edge angle and femoral head coverage, medialization, distalization and ilio-ischial angle parameters. For normally distributed parameters, we used Student’s t-test; for parameters without a normal distribution, we used the Wilcoxon signed-rank test. Correlations were assessed according to a normal distribution using the Pearson and Spearman method. Results: For all parameters, we observed statistically signicant differences in the measurements of dysplastic hip joints before and after the surgery. We also observed a statistically signicant difference between the structure of dysplastic hip joints prior to the surgery and healthy hip joints from the control group based on all radiological parameters. The resulting medialization was 2.68 mm, distalization was 3.65 mm, and the ilio-ischial angle was changed by 2.62°. There was also an improvement in the femoral head bony covering: CEA by 17.61° and FHC by 16.46%. Conclusions: Based on all the radiological parameters, we presented the difference between healthy and dysplastic hip joints. Learning the parameter values that are used to describe dysplastic hip joints will allow us to improve the imaging of the condition and will also allow for better planning and proper qualication of patients for surgical treatment of hip joint dysplasia.

Results: For all parameters, we observed statistically signi cant differences in the measurements of dysplastic hip joints before and after the surgery. We also observed a statistically signi cant difference between the structure of dysplastic hip joints prior to the surgery and healthy hip joints from the control group based on all radiological parameters. The resulting medialization was 2.68 mm, distalization was 3.65 mm, and the ilio-ischial angle was changed by 2.62°. There was also an improvement in the femoral head bony covering: CEA by 17.61° and FHC by 16.46%.
Conclusions: Based on all the radiological parameters, we presented the difference between healthy and dysplastic hip joints. Learning the parameter values that are used to describe dysplastic hip joints will allow us to improve the imaging of the condition and will also allow for better planning and proper quali cation of patients for surgical treatment of hip joint dysplasia.
Trial registration: Consent of the bioethics commission Medical Centre of Postgraduate Education 83/PB/2015 18.11.2015 Warsaw Background Residual hip joint dysplasia is a sequela of incorrect development of the hip joint acetabulum during fetal development or a result of incorrect treatment during infancy. It is estimated that in among young patients, up to 80-90% of cases of hip joint osteoarthritis develop as a result of hip joint dysplasia [1,2,3].
Bernese periacetabular osteotomy is one of the methods used to surgically "save" the patient's hip joint [4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19]. The procedure involves a periacetabular cut of the pubic, ischium, and ilium in an attempt to change the con guration of the joint socket and achieve the stability of the dysplastic hip joint. This osteotomy technique makes it possible to save the back column of the socket, which in turn guarantees the stability of the pelvic ring after the procedure and allows patients to quickly resume walking [8,9,19]. The correction of the socket part of the hip joint improves the bony covering of the femoral head, as well as the medialization and distalization of the joint. The abovementioned parameters are important because of the repositioning of the femoral head and bringing it closer to the ischium. This position offers favorable conditions for improving the dynamic stability of the joint and extends the lever arm for gluteal muscles. It increases the moment of hip joint stabilizing force and decreases the arm of body weight force acting on the joint [1,8,9,15,20,21,22]. The main assumption of this procedure is stopping or slowing down, as well as preventing, the development of hip joint osteoarthritis, which might potentially result in the joint being replaced by endoprosthesis. Many authors emphasize the fact that computed tomography (CT) is the basic test enabling the accurate assessment of dysplastic hip joints and surrounding anatomical structures [23,24,25]. On the other hand, a CT scan is characterized by a much higher dose of ionizing radiation and higher costs of the examination [24,25]. This research is also necessary to determine an easy, cheap, repeatable and nonburdensome method to con gure dysplastic hip joints. Bearing in mind the patient's well-being and awareness of symptomatic dysplastic disease, mainly in young women, we decided to work on the basis of digital X-ray in the anterior-posterior (AP) view of the pelvis. The aim of this work was to radiologically assess the con guration change of a dysplastic hip joint after surgical treatment using the Bernese periacetabular osteotomy procedure.

Materials And Methods
The retrospective study group consisted of patients with residual dysplasia who quali ed for surgical treatment using the abovementioned method in our department and met the following inclusion and exclusion criteria. The control group for the retrospective study consisted of people not diagnosed with hip joint dysplasia who were treated in our hospital. Ninety patients (90 hip joints) were analyzed in the control group, with an average age of 34 years (19-62 years old). There were 74 women and 16 men.
The material for retrospective analysis consisted of digital AP pelvis X-rays of patients with residual hip dysplasia who underwent surgical procedure of Bernese periacetabular osteotomy, as well as digital AP pelvis X-rays of patients not diagnosed with hip dysplasia who served as the control group. The radiological assessment included the measurements of the femoral head coverage: Wiberg angle (CEAcentral edge angle), Heyman's and Herndon's acetabular-head factor (FHC-femoral head coverage), as well as parameters of medialization (line connecting the body midline and edge of femoral head closest to the midline) (Figure 1), distalization (line connecting the baseline, which connects the tuber ischii, and the lowest edge of femoral head) ( Figure 2) and the reorientation of the acetabulum measured as the ilioischial angle (angle between the ilio-ischium line and the line connecting the tuber ischii) (Figure 3) based on the radiograms in AP digital X-ray projection. We expected that CEA and FHC would be increased after surgery and that medialization, distalization and ilio-ischial angle would be decreased after surgery. Three independent studies with various levels of experience also conducted an analysis verifying the accuracy and repeatability of radiological measurements. To assess the level of residual hip joint dysplasia, we used Crowe's Classi cation [26].
This study was approved by the institutional review board. Informed consent was obtained from patients for participation in the study.

Statistical analysis:
We checked the normal distribution using the Shapiro-Wilk test. For normally distributed parameters, we used Student's t-test; for parameters without a normal distribution, we used the Wilcoxon signed-rank test.
The signi cance criterion was p<0.05. Correlations were assessed according to a normal distribution using the Pearson and Spearman analyses. Measurement accuracy: angles -0.5 degrees, distances -0.5 mm (all measurements were done in CareStream Solution software). The main measurements were made by one surgical researcher.

Results
According to Crowe's Classi cation, there were type I -62 hip joints (52%); type II -50 hip joints (42%); and type III -7 hip joints (6%). Mean follow-up was 36 months (12-48 months). For all parameters, we observed statistically signi cant (p<0.05) measurement differences in dysplastic hip joints before and after the surgery (Table 1). We also observed a statistically signi cant (p<0.05) difference in the anatomy of dysplastic hip joints before the surgery and healthy hip joints from the control group based on all radiological parameters (Table 2). We obtained a strong positive Pearson correlation (0.756) between CEA and FHC ( Figure 4) and a negative Pearson correlation (-0.522) between CEA and the ilio-ischial angle ( Figure 5). We also observed a statistically signi cant negative Spearman correlation (-0,517) between CEA and distalization parameters ( Figure 6). The analysis verifying the correctness and repeatability of radiological parameters did not reveal any statistically signi cant differences between studies of various levels of experience.
The retrospective study of radiological measurements we presented showed statistically signi cant differences in the structure of healthy and dysplastic hip joints. The medialization, distalization, and ilioischial angle parameters describe the hip joint in a statistically signi cant way. The ilio-ischial angle and distalization parameters seem to be of high importance, statistically signi cantly correlating with other parameters established in the literature.

Discussion
There are many documented and widely accepted parameters for assessing the con guration of hip joints. However, they are insu cient to assess the con guration change of the acetabulum subjected to acetabular osteotomy. Zhang et al. described the LCEA value as a key radiological parameter of clinical and prognostic importance in the surgical treatment of hip dysplasia using the PAO method [17]. Two of these parameters are the so-called medialization and distalization of the acetabular fragment. Medialization is the acetabular fragment's displacement toward the body's midline and, as a result, a displacement of the entire hip joint toward the body's central axis. Distalization is a parameter describing the correction of a subluxed hip joint in the frontal plane toward the patient's extremities. The literature available on that topic lacks any reports regarding the assessment of hip joint distalization after PAO. Hip joint medialization and distalization are important parameters due to the repositioning of the femoral head and bringing it closer to the ischium. This position offers favorable conditions for improving the joint's dynamic stabilization and lengthens the arm of gluteal muscles. It increases the moment of force stabilizing the hip joints and reduces the arm of body weight force acting on the joint [2,7,9,13,20,27,28]. In a retrospective study, we determined that the differences shown are noticeable not only in the structure of the hip acetabulum itself but also in the entire pelvis, covered by residual hip dysplasia. This is evidenced by a statistically signi cant postoperative change in the iliosischial angle and its difference in healthy hips. The difference between healthy and dysplastic hip joints was 8.3 degrees. It shows a more vertical con guration of the structure of the iliac wing and ischium bone as a result of the steep and shallow dysplastic acetabulum of the hip joint. It seems that hip dysplasia does not concern only the hip acetabulum but its entire pelvis spatial con guration. We can de ne such a pelvis as a dysplastic model in the course of residual hip dysplasia.

Mavcic et al. mathematically calculated the tension distribution in healthy and dysplastic hip joints and
proved that the forces acting on a dysplastic hip joint are twice as strong as those acting on a healthy joint [29]. Delp  PAO by measuring the distance between the medial edge of the femoral head and the ilio-ischium line (Kohler's line) [7,14,16]. Based on our observations, we believe that the medialization measurements performed using the methods of the abovementioned authors can be distorted by the con guration change of the whole acetabular fragment, which results in the change of location and con guration of the "Kochler's tears" and consequently the ilio-ischium line (Kochler's line). This is proof of the statistically signi cant change in the ilio-ischial angle parameter that we introduced. Mimura et al. measured the center of the femoral head lateralization index (HLI -Head Lateralization Index) after triple pelvis osteotomy [1]. Tanaka et al. assessed medialization based on DBSPFH measurements and the head lateralization index (HLI) using 3D computer tomography. The authors pointed out the di culties in achieving repeatability in determining the middle of the femoral head, which is very important in millimeter measurements [22]. Jae Suk Chang et al. measured the distance between the pubic symphysis and femoral head (DBSPFH) in patients who underwent PAO with hip joint dysplasia and femoral head deformation [31]. Goronzy et al. assessed the predictors of coxarthrosis after PAO on the basis of AP Xray and MRI. The authors described that only posteriorly deformed femoral heads had an in uence on the functional outcome without alternating degeneration in the 5-year interval [18].
The justi cation for conducting the research was an attempt to nd the abovementioned hip joint medialization and distalization parameters. There is a need for objective assessment of parameters before and after surgery. It seems that they are often intuitively assessed as correct during mid-surgery control under an X-ray monitor by the surgeon performing the procedure.

Conclusion
Learning about the values of parameters describing a dysplastic hip joint allowed us to improve the imaging of the condition and proved to be helpful in planning for surgical treatment of residual hip joint dysplasia. The measurement methods of dysplastic hip joints presented in this manuscript are suggestions for further observation and open discussion of the dysplastic hip joint con guration. In the future, we plan to conduct a study based on prospective data that includes radiological measurements in correlation with clinical data.     Pearson correlation between CEA and FHC.

Figure 5
Pearson correlation between CEA and the ilio-ischial angle.

Figure 6
Spearman correlation between CEA and distalization.