Minimally Invasive Radiographic Evaluation of the Dysplastic Hip Joint Configuration in Terms of Surgical Treatment with Hip Preservation by PAO

Abstract

Introduction

Residual developmental dysplasia of hip (RDDH) is a factor of early osteoarthritis of the hip. The main problems are pain and instability of the hip joint due to inadequate coverage of the femoral head by the acetabulum. The purpose of this study was to radiologically evaluate RDDH after Bernese periacetabular osteotomy (PAO) and to compare RDDH to healthy hips.

Materials and Methods

The radiological parameters of RDDH treated by PAO were retrospectively evaluated. Digital AP pelvic radiographs were taken, including parameters of central edge angle and femoral head coverage, medialization, distalization, and ilio-ischial angle. Clinical assessment is based on the VAS scale. The study group consisted of patients with RDDH, and the control group consisted of patients without RDDH.

Results

After PAO radiological parameters decreased: medialization by 2.68 mm, distalization by 3.65 mm, and ilio-ischial angle by 2.62°. However, there was an increase in the parameters: CEA by 17.61° and FHC by 16.46%. There was a mean 3 point decrease in pain on the VAS scale. There was also a statistically significant radiological difference in the structure of dysplastic hip joints before surgery and healthy hip joints of the control group.

Conclusions

Radiological studies confirmed the effectiveness of the PAO method in the treatment of RDDH. Based on all radiological parameters, differences between healthy and dysplastic hip joints were demonstrated. We believe that a thorough understanding of the values of radiological parameters used to describe dysplastic hip joints will allow us to improve the imaging diagnosis of this condition.

Introduction

Residual developmental dysplasia of hip (RDDH) 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 among young patients, up to 80–90% of cases of hip joint osteoarthritis develop as a result of hip joint dysplasia [1–3]. The Bernese periacetabular osteotomy (PAO) is one of the treatment options for the hip joint preservation surgery [4–17]. The procedure involves a periacetabular cut of the pubic, ischium, and ilium to change the configuration of the joint acetabulum and achieve the stability of the dysplastic hip joint. Our surgical technique uses a modified Smith–Petersen approach (bikini approach) to the hip joint. Then, we perform an osteotomy of the pubic bone and under fluoroscopy guidance an osteotomy of the ischium bone. As the last, we perform an osteotomy of the iliac bone. This is a two-stage osteotomy: we incise the iliac bone (with an oscillating saw) in the supra-acetabular region and then cut (with a curved osteotome) the iliac and ischium cutting off the acetabular fragment from the posterior column of the pelvis (Fig. 1). This osteotomy technique makes it possible to save the back column of the acetabulum, which guarantees the stability of the pelvic ring [8, 19]. The correction of the acetabulum of the hip joint improves the bony covering of the femoral head, and allows medialization and distalization of the joint. It increases the moment of hip joint stabilizing force and decreases the arm of body weight force acting on the joint [1, 8, 13, 19, 20]. 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 that computed tomography (CT) is the basic test enabling the accurate assessment of dysplastic hip joints and surrounding anatomical structures [21–23]. On the other hand, a CT scan is characterized by a much higher dose of ionizing radiation and higher costs [22, 23]. This study is aimed at determining an easy, cheap, repeatable, and non-burdensome method to configure dysplastic hip joints. Bearing in mind the patient's well-being and the awareness of the symptomatic dysplastic disease, mainly in young women, we decided to use digital X-ray in the pelvis’s antero-posterior (AP) view. This work aimed to radiologically assess the configuration change of a dysplastic hip joint after surgical treatment using the PAO procedure. We also performed a radiological evaluation of healthy and dysplastic hip joints.

Fig. 1.

Images show bone cut lines during in-vitro PAO: a pubic and ischial bones; b iliac and ischial bones with preservation of the posterior column of the pelvis; c pubic, iliac and ischial bones

Materials and Methods

The retrospective study group consisted of patients with residual dysplasia who qualified for surgical treatment using the above-mentioned method in our department and met the following inclusion and exclusion criteria (Table 1).

Table 1.

Demographic data and clinical outcome of VAS score

	RDDH patient (hip)	Healthy patient (hip)
Number	99 (119)	90 (90)
Age	29 (14–55)	34 (19–62)
sex	F: 89 M: 10	F: 74 M: 61
Side	L: 61 R: 58	L: 52 R: 38
VAS	Pre-op: 7 (6–9) Post-op: 4 (1–5)	0

Inclusion criteria:

• Hip joints were not treated surgically in early childhood

• Symptomatic residual hip joint dysplasia

• Full and correct radiological documentation of the entire treatment period

• Completed treatment process (removal of screw fixation)

• Bone fusion in the osteotomy area

• No degenerative changes.

  Exclusion criteria:

• Secondary hip acetabulum

• Patients have previously undergone hip joint acetabulum surgery

• Patients with femoral head deformation.

The study included 99 patients (119 hips), whose mean age at the time of surgery was 29 years (14–55 years). There were 61 left and 58 right hip joints, and there were 89 women and 10 men.

The control group consisted of people treated in our hospital without a diagnosis of hip dysplasia, for example, patients with back or knee pain. In the control group, 90 patients (90 hips) were examined. The mean age at study entry was 34 years (19–62 years). There were 52 left and 38 right hip joints, and there were 74 women and 16 men.

The material for retrospective analysis consisted of digital AP X-ray pelvis of patients with RDDH who underwent surgical procedure of PAO, and digital AP X-ray pelvis 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 (CEA- central edge angle), Heyman’s and Herndon’s acetabular-head factor (FHC—femoral head coverage), parameters of medialization (line connecting the body midline and edge of femoral head closest to the midline) (Fig. 2), distalization (line connecting the baseline, which connects the tuber ischii, and the lowest edge of femoral head) (Fig. 3), and the reorientation of the acetabulum measured as the ilio-ischial angle (angle between the ilio-ischial line and the line connecting the tuber ischii) (Fig. 4) based on the radiograms in AP digital X-ray projection. CEA and FHC parameters will increase after surgery and medialization, distalization, and ilio-ischial angle will decrease after PAO surgery (Fig. 5). Postoperative acetabular version was assessed on the basis of crossover sign symptom also on AP digital X-ray. Three independent studies with various levels of experience also conducted an analysis verifying the accuracy and reliability of the radiological measurements. We used the Crowe’s classification to assess the level of residual hip joint dysplasia [24]. Clinical assessment was undertaken using the VAS pain score with radiological evaluation, just prior to surgery by the admitting physician, and approximately 12 months postoperatively by a physician from our department.

Fig. 2.

Hip joint medialization (x) (line connecting the body midline and medial edge of femoral head); a preoperation (x—83 mm); b postoperation (x—81 mm)

Fig. 3.

Hip joint distalization (y) (line connecting the baseline and the lowest edge of the femoral head); a preoperation (y—57 mm); b postoperation (y—53 mm)

Fig. 4.

Ilio-ischial angle (k)—angle between the ilio-ischial line and the line connecting the ischiatic tuber; a preoperation (k—88 deg); b postoperation (k—84 deg)

Fig. 5.

a–d Radiographic measurements of CEA (10 deg.), x-medialization (83 mm), y-distalization (57 mm), k-ilio-ischial angle (87 deg.) before surgery; e–h radiographic measurements of CEA (28 deg.), x-medialization (81 mm), y-distalization (53 mm), k-ilio-ischial angle (83 deg.) after surgery

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 normality of the distribution using the Shapiro–Wilk test. For normally distributed parameters, we used the Student’s t test; for parameters without a normal distribution, we used the Wilcoxon signed-rank test. The significance criterion was p < 0.05. The correlations were assessed using Pearson and Spearman analyses as appropriate, according to the normality of the distribution. SPSS Statistics v15.0 software (SPSS Inc., Chicago, IL, USA) was used for statistical analysis. The measurement accuracy was as follows: angles—0.5° and distances—0.5 mm (all measurements were done in CareStream Solution software). The main measurements were performed by one surgical researcher.

Results

According to Crowe’s classification, there were 62 type I hip joints (52%), 50 type II hip joints (42%), and 7 type III hip joints (6%). The mean follow-up was 36 months (12–48 months). We observed statistically significant (p < 0.05) measurement differences in dysplastic hip joints before and after the surgery. The PAO results were as follows: a decreased medialization by 2.68 mm (range 2–3), decreased distalization by 3.65 mm (range 3–4), and decreased ilio-ischial angle by 2.62° (range 2–3). There was also an improvement in the femoral head bony covering: CEA by 17.61° (range 16–18) and FHC by 16.46% (range 15–19) (Table 2). We observed a statistically significant negative correlation according to the Pearson’s linear correlation between the CEA parameter recognized in the literature and the ilio-ischial angle described in the study (r = − 0.522) (Fig. 6). This correlation indicates that the ilio-ischial angle decreases linearly with increasing CEA. We also observed a statistically significant negative correlation, according to Pearson’s linear correlation between the CEA parameter recognized in the literature and the distalization parameter described in the study (r = − 0.517) (Fig. 7). This correlation indicates that distalization decreases linearly with increasing CEA. As a result of PAO, the increase in femoral head coverage is associated with an increase in CEA and FHC, decreased ilio-ischial angle, and a slight distalization. Based on the crossover sign symptom on AP digital X-ray, we did not observe acetabular retroversion after PAO in any of the patients.

Table 2.

Results of pre- and postoperative changes in the hip joint configuration based on radiological measurements

	Preoperative (95% CI) n = 119	Postoperative (95% CI) n = 119	Difference	p value
CEA (°)	3.48 ± 10.58 (4.94–7.76)	21.09 ± 8.1 (19.5–22.5)	17.61	< 0.0001
Medialization (mm)	83.72 ± 5.75 (82.68–84.76)	81.04 ± 6.33 (79.9–82.2)	− 2.68	0.0001
Distalization (mm)	58.05 ± 9.66 (56.30–59.81)	54.4 ± 10.02 (52.6–56.4)	− 3.65	< 0.0001
FHC (%)	53.07 ± 11.82 (50.93–55.22)	69.53 ± 7.53 (68.6–71.5)	16.46	< 0.0001
Ilio-ischial angle (°)	86.5 ± 3.10 (85.94–87.07)	83.88 ± 3.02 (83.4–84.6)	− 2.62	< 0.0001

mm millimeter

Fig. 6.

Graph of Pearson's linear correlation between CEA and ilio-ischial angle (r = − 0.522)

Fig. 7.

Graph of Pearson’s linear correlation between CEA and distalization (r = − 0.517)

We also observed a statistically significant (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. According to our observations and measurements, the proximalization of the dysplastic hip joint qualified for PAO is approximately 10.5 mm. The lateralization is approximately 5 mm to the healthy hip joint. There were also noticeable and statistically significant differences in bone coverage by the acetabulum of the femoral head. The difference in the coverage of the femoral head was 25.9° for the CEA parameter and 23.5% for the FHC. The measurements of the ilio-ischial angle turned out to be very interesting, as it describes, on the one hand, the change in the orientation of the acetabulum subjected to PAO. On the other hand, it presents a noticeable and statistically significant difference in the position and structure of the pelvis affected by hip dysplastic disease. The difference between healthy and dysplastic individuals is 8.3°. This shows a more vertical configuration of the structure of the iliac plate and the ischium because of the steep and shallow dysplastic acetabulum of the hip joint (Table 3). The analysis verifying the validity and reliability of radiological parameters did not reveal any statistically significant differences between studies of various experience levels.

Table 3.

Preoperative results compared to the control group hip joint configuration based on radiological measurements

	Preoperative (95% CI) n = 119	Control (95% CI) n = 90	Difference	p value
CEA (°)	3.48 ± 10.58 (4.94–7.76)	29.4 ± 4.43 (28.48–30.3)	25.92	< 0.0001
Medialization (mm)	83.72 ± 5.75 (82.68–84.76)	78.56 ± 4.74 (77.57–79.56)	− 5.16	0.0019
Distalization (mm)	58.05 ± 9.66 (56.30–59.81)	47.52 ± 6.89 (46.08–48.97)	− 10.53	< 0.0001
FHC (%)	53.07 ± 11.82 (50.93–55.22)	76.6 ± 4.11 (75.75–77.47)	23.53	< 0.0001
Ilio-ischial angle (°)	86.5 ± 3.10 (85.94–87.07)	78.17 ± 3.07 (77.53–78.81)	− 8.33	< 0.0001

mm millimeter

Clinically significant improvement in VAS score was observed in all patients after PAO. Decrease pain in VAS score mean of 3 points (range 1–5) was noted (p = 0.007).

Discussion

There are many documented and widely accepted parameters for assessing the configuration of hip joints. However, they are insufficient to assess the configuration change of the acetabulum subjected to acetabular osteotomy. Two of these missing 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 leading to 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 does not report on the assessment of hip joint distalization after PAO. 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 [25]. Delp and Malloney described the effects that the medialization and distalization of the center of the hip joint have on the strength of the abductor muscle. The authors proved the significance of lowering of the center of the hip joint on the force and arm of the hip abductors [26]. According to them, simultaneous medialization and distalization have the most significant impact on the reconfiguration of muscle forces acting on the joint. There have been several publications describing the medialization parameter after PAO.

Siebenrock et al., Junfeng Zhu et al., and Clohisy et al. assessed the hip joint medialization after PAO by measuring the distance between the medial edge of the femoral head and the ilio-ischial line (Kohler’s line) [7, 12, 14]. Based on our observations, we believe that the medialization measurements performed using the methods of the above-mentioned authors can be distorted by the configuration change of the whole acetabular fragment, which results in the change in location and configuration of the “Kochler’s tears” and consequently the ilio-ischial line (Kochler’s line). This is proof of the statistically significant change in the ilio-ischial angle parameter that we introduced. 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 [27]. Spiker et al. reported on the differences between dysplastic and non-dysplastic hips with femoral-acetabular impingement (FAI). The authors measured the relationship of acetabular version (AcetV) and femoral version (FV) in both groups. They compared 113 patients with dysplasia who underwent PAO with 1332 patients without dysplasia with FAI (CEA > 25). They found that dysplastic hips had statistically higher AcetV and FV values than non-dysplastic hips with FAI. Only 5% of patients with excessive FV (> 34°) required subsequent femoral derotation osteotomy, suggesting that in most patients with hip dysplasia, FV may not affect the clinical course after surgery [28]. Fowler et al. focus on the change in the iliac-ischial line during the correction of the acetabular in PAO procedure, which has a significant clinical and radiological value in the evaluation of medialization of the acetabular element after PAO. Moreover, they suggest changing the place of medialization measurement to 1/3 distal part of the femoral head [29]. On the other hand, Pun et al. take up the topic of determining the desired value of medialization to obtain a good long-term PAO result [30]. Czubak presented the correction of the Wiberg angle ranged from 12° to 55° (mean 29). Complications included urinary tract infections, ectopic bone formation, and early post-op infection of soft tissue. There was no evidence of bleeding or vascular complication, or of injury to either the sciatic or femoral nerves [31]. Goronzy et al. assessed the predictors of coxarthrosis after PAO based on AP X-ray and MRI. The authors described that only posteriorly deformed femoral heads influenced the functional outcome without alternating degeneration in the 5-year interval. According to the authors, it is also important to arrange in the appropriate acetabular version, so as not to perform a retroversion but also not to leave excessive anteversion [16]. Okano et al. presented a retrospective radiological examination of 224 hips in 112 patients with acetabular dysplasia. There were 103 women and 9 men with a mean age of 37.6 years (18–49). A total of 201 hips were placed in the acetabular dysplasia group and 23 in the normal group. The central edge angle and acetabular index were significantly lower in the hip dysplastic group (p < 0.001). The CEA difference was 25.3°, and the head acetabular index was 22%. The acetabular angle, acetabular roof angle, and roundness index were also significant in the group with acetabular dysplasia compared with the healthy group (p < 0.001) [32]. In our study, the difference in the coverage of the femoral head was 25.9° for the CEA parameter and 23.5% for the FHC. Moreover, we showed that the demonstrated differences are visible in the structure of the acetabulum itself and the entire pelvis with residual hip dysplasia. This is evidenced by the statistically significant postoperative change in the ilio-ischial angle and its difference in healthy hips. The difference between healthy and dysplastic hips was 8.3°. It shows a more vertical configuration of the structure of the iliac wing and ischium bone because of the steep and shallow dysplastic acetabulum of the hip joint. It seems that hip dysplasia does not concern only the hip acetabulum but rather its entire pelvis spatial configuration. We can define such a pelvis as a dysplastic model in RDDH. Learning about the values of parameters describing a dysplastic hip joint allowed for the improvement of the imaging and proved to be helpful in planning for surgical treatment of residual hip joint dysplasia. The methods of radiographic measurement of dysplastic hip structures presented in this work suggest further observations and are part of the ongoing discussion on dysplastic hip configurations. A limitation of the study is its retrospective study.

Conclusion

The authors believe that PAO meets all the criteria for hip-preserving surgery for RDDH, especially when the disease affects young people. It significantly reduces pain resulting from hip dysplasia. Not only lateral coverage but also complete reorientation of the acetabular component has a significant impact on the long-term surgical outcome of PAO. However, further studies in this area are desirable to clarify these results. Further studies focusing on prospective radiological and clinical studies are needed to better understand.

Abbreviations

RDDH

Residual developmental dysplasia of hip

PAO

Bernense periacetabular osteotomy

DDH

Developmental dysplasia of hip

CEA

Central edge angle

FHC

Femoral head coverage

LCEA

Lateral center edge angle

AP

Antero-posterior

CT

Computed tomography

HLI

Head lateralization index

DBSPFH

Distance between symphysis pubis and femoral head

MRI

Magnetic resonance imaging

mm

Millimeters

deg

Degrees

Funding

There was no financial support associated with preparation of this manuscript.

Declarations

Conflicts of Interest

Kamil Kołodziejczyk, Adam Czwojdziński, Andrzej Sionek, and Jarosław Czubak declare that they have no conflict of interest.

Ethics Approval

Consent of the bioethics commission Medical Centre of Postgraduate Education 83/PB/2015 18.11.2015 Warsaw. Informed consent was obtained from all patients participating in the study.

Informed consent

Informed consent was obtained from all patients to participate in the study.