Diagnosis and Management of Borderline Hip Dysplasia and Acetabular Retroversion

Abstract

Borderline hip dysplasia and acetabular retroversion are common radiographic findings in young individuals with and without hip pain. Orthopaedic surgeons should be knowledgeable about the radiographic findings, diagnosis, and appropriate nonsurgical and surgical treatment of these conditions. Borderline hip dysplasia is generally defined by a lateral center edge angle of Wiberg from 20 to 25° (some define as 18–25°) and is a cause of joint microinstability. The degree of soft tissue laxity can have significant implications for joint stability in patients with borderline hip dysplasia. The most common presenting symptoms are groin pain and lateral hip pain. Acetabular retroversion is defined by radiographic findings of crossover sign, ischial spine sign, and posterior wall sign. Individuals with symptomatic retroversion have a clinical presentation consistent with impingement, groin pain with flexion activities, and less commonly lateral hip pain. Physical therapy has been shown to improve symptoms in a subset of individuals with these conditions. There are multiple recent publications about arthroscopic treatment of patients with borderline hip dysplasia. These reports generally find that good short-term outcomes can be expected when using arthroscopic techniques that include labral preservation/repair and capsular plication. There are limited reports of periacetabular osteotomy as a treatment for borderline hip dysplasia. Publications focusing specifically on surgical treatment of acetabular retroversion are also infrequent. Periacetabular osteotomy has been shown to have superior long-term clinical outcomes to surgical hip dislocation with anterior rim trimming in patients with all three radiographic findings of retroversion. Arthroscopic treatment has been shown to have good short-term outcomes. Future work in the areas of borderline hip dysplasia and acetabular retroversion should focus on reporting long-term clinical follow-up of these surgical treatments and using computation techniques as a tool to determine appropriate surgical and nonsurgical treatment for each individual patient.

Introduction

Over the past 30 years, significant advancements in surgical techniques have changed how we manage prearthritic hip conditions in the young adult. We have long understood how these conditions adversely affect the mechanical environment in the joint and lead to the development of osteoarthritis,¹ but until recent progress in our understanding of hip anatomy and mechanics there were limited surgical options for young adults with hip pain before developing end-stage osteoarthritis. Periacetabular osteotomy (PAO), surgical hip dislocation, and hip arthroscopy are increasingly performed to treat hip pathology in the skeletally mature individual, and our understanding of conditions best managed with these procedures continues to evolve. Borderline hip dysplasia and acetabular retroversion are two hip conditions with significant controversy that has received substantial attention in recent literature.^2–7

Borderline hip dysplasia is characterized by decreased coverage of the femoral head by the acetabulum. Microinstability as a result of under coverage can lead to labral tears, cartilage injury, and clinical symptoms.⁸ The radiographic criteria for borderline hip dysplasia are variably defined in the literature.^4,6,9–11 While there is general consensus that mild and moderate/severe hip dysplasia is best treated with PAO,^12–14 the efficacy and longevity of arthroscopic procedures for the treatment of borderline hip dysplasia are controversial.^15,16 In contrast, acetabular retroversion is characterized by excess anterior coverage of the femoral head by the acetabulum resulting in femoroacetabular impingement (FAI) and subsequent joint damage.¹⁷ Recent studies have compared results of surgical hip dislocation and PAO to treat acetabular retroversion² and investigated hip arthroscopy as a treatment of impingement.¹⁷ As individuals with these conditions commonly present to orthopaedic surgeons, it is important to understand the appropriate nonsurgical and surgical options for these patients.

The goal of this review is to discuss clinical presentation, radiographic evaluation, nonsurgical treatments, operative techniques, clinical outcomes, and future directions for research to improve care of individuals with borderline hip dysplasia and acetabular retroversion.

Clinical Presentation

Intra-articular hip pathology can have a variable clinical presentation. There is significant crossover between symptoms caused by instability and impingement. Taking a careful history can help to discern the cause of hip pain. Generally, patients will have insidious onset of groin pain that can be activity related.¹⁸ Groin pain can also be worsened by deep flexion positions typical of impingement (such as sitting). It is important to be cognizant of the job requirements of the patient and athletic participation to identify positions that are exacerbating for the hip pain. This can be helpful to determine if groin pain is primarily an instability or impingement issue.

In addition to groin pain, pain in the lateral hip is often the first symptom-of intra-articular pathology. In a consecutive clinical series of patients with hip dysplasia, 72% of patients presented with groin pain and 66% presented with lateral hip pain.¹⁸ Individuals with hip dysplasia have decreased extension and increased abduction during the stance phase of the gait cycle,¹⁹ to position the femoral head deeper in the acetabulum. This gait modification likely leads to abductor fatigue and the lateral hip pain over the trochanter that is commonly experienced in patients with hip instability. Individuals with anterior impingement more commonly present with groin pain and less commonly have lateral hip pain. Eighty-six percent of female and 94% of male patients with FAI presented with groin pain and 19% of female and 10% of male patients presented with lateral trochanteric pain in a clinical series reported by Nepple et al.²⁰ Other extra-articular causes of hip pain including psoas snapping,²¹ iliotibial band snapping,²² extra-articular impingement,²³ and hernia²⁴ can also present with similar symptoms, further highlighting the importance of taking a careful history in these patients.

Physical Examination

The physical exam is important to identify the source of pain and disability in hip conditions. Careful testing of gait, muscle strength, range of motion, and instability helps to identify structural pathology. Exam findings can be used to support pertinent positives in the history, identify further pathology, and to determine if further imaging studies are needed.

The exam starts with a visual assessment of gait. As stated previously, patients with instability due to dysplasia ambulate with increased abduction and internal rotation during the stance phase of gait. Special attention is also paid to identify the presence and character of any limp during the gait cycle. Trendelenburg testing is performed to assess abductor and core strength. The patient is then positioned supine on the exam table to measure abduction, flexion, internal rotation, and external rotation. Rotation is measured with the hip in 90° of flexion and then prone with the hip extended. Motion is carefully assessed by noting when the pelvis begins to rotate in the plane of motion. Femoral anteversion has a significant impact on hip range of motion and excessive internal rotation should be evaluated radiographically.²⁵

Multiple techniques for assessing instability on physical exam have been described. All techniques describe hip extension and external rotation in varying degrees of abduction to assess for anterior instability. The abduction–hyperextension–external rotation test performed in the lateral position with the pelvis stabilized has been shown to be the most sensitive and specific for instability.²⁶ A detailed assessment of soft tissue laxity is also important when assessing an individual for joint instability. Radiographic findings of borderline hip dysplasia can have significantly different implications for individuals with hyperlaxity of their joint compared with those with decreased range of motion. The Breighton scale is commonly used to quantify the degree of soft tissue laxity in the clinical setting.²⁷ Patients with hypermobility often carry the diagnosis of Ehlers–Danlos syndrome or other conditions with increased soft tissue laxity and can have significant hip-related symptoms.²⁸

Anterior impingement is reproduced with the hip in maximal flexion, internal rotation, and adduction. Groin pain with this maneuver constitutes a positive anterior impingement test. Lateral tenderness over the greater trochanter is common. Areas of tenderness can be palpated with the patient in the lateral position. If the patient has posterior pain, tenderness over the ischium can be assessed. Ischiofemoral impingement is assessed with hip in extension and adduction while the contralateral hip is held flexed to stabilize the pelvis. Posterior pain with a noted resistance to external/internal range of motion is a positive test.

Radiographic Evaluation

Radiographic evaluation of dysplasia and retroversion relies heavily on plain radiographs. When assessing young adult hip pathology, the most common imaging study described is the standing anterior posterior pelvis radiograph. This radiograph is taken standing to capture the functional position of the pelvis when the patient is standing. An inlet standing anteroposterior (AP) pelvis radiograph can increase the apparent coverage of the anterior wall over the femoral head and cause the appearance of a crossover sign in an otherwise normal pelvis radiograph.²⁹ Patients with this finding on standing radiographs may benefit from a structured stretching program of the hip flexors and working on pelvis positioning with physical therapy. A falsely positive crossover sign can also be present caused by variable morphology of the anterior inferior iliac spine.³⁰ The lateral center edge angle (LCEA) of Wiberg is minimally altered by pelvic tilt.²⁹ Other common measurements evaluated include the Tönnis angle, extrusion index, and posterior wall sign.^29,31 The location of the crossover sign is also noted. The anterior and posterior wall coverage is quantified with wall indices.³²

Evaluation of the proximal femur on radiographs is also important because offset deformity of the femoral head–neck junction is common in both borderline hip dysplasia and acetabular retroversion.^2,33,34 A 45° Dunn lateral view provides a view of the anterolateral femoral head–neck junction, allowing proper measurement of the α angle. A frog leg lateral view of the hip provides an assessment of the anterior femoral head–neck junction. A cross table lateral view of the hip can also provide a view of the anterior femoral head–neck junction, but this is typically not taken because of higher radiation doses required for this image.³⁵

Hip dysplasia is most commonly quantified on the standing AP pelvis radiograph. There is significant variability in the values selected to diagnose hip dysplasia, but most commonly borderline hip dysplasia is defined as a LCEA of 20 to 25°,mild hip dysplasia is 15 to 20°, and moderate-to-severe hip dysplasia is less than 15°.³⁵ Other studies have defined borderline hip dysplasia as a LCEA of 18 to 25.^36,37 It is important to note that the lateral most edge of the sclerotic portion of the acetabular sourcil should be used to measure the angle to maximize reproducibility.³¹ There is often nonsubchondral bone that extends lateral to the sourcil that should not be used for this measurement. For the Tönnis angle, 0 to 10° is normal, 10 to 15° indicates borderline dysplasia, and > 15° qualifies as dysplasia.³⁵ Figure 1 demonstrates a patient with borderline hip dysplasia and instability with a LCEA of 20°.

Figure 1:

A 20-year-old female with the diagnosis of Ehlers–Danlos syndrome hypermobility type with activity related right groin and lateral trochanteric pain. Hip instability on physical exam. Lateral center edge angle 20°. Treated with an extended course of physical therapy.

Acetabular retroversion has three characteristic findings on the standing AP pelvis radiograph: the ischial spine sign, the posterior wall sign, and crossover sign.² Again, it is important to confirm that pelvic tilt is appropriate on the radiograph as this can result in a false positive for these signs. The location of the crossover sign is noted. Also, the morphology of the anterior inferior iliac spine should be considered when measuring crossover sign. Global retroversion is caused by an external rotation of the hemipelvis causing posterior wall deficiency and a prominent anterior wall. An ischial spine sign is seen with a narrow and medial obturator foramen and a narrow medial to lateral width of the pelvis. Focal retroversion is seen with a prominent, cranial anterior wall resulting in a crossover sign and anterior FAI. A low hanging anterior inferior iliac spine can falsely create the appearance of a crossover sign.^30,38 This emphasizes the importance of cross-sectional imaging to assess acetabular coverage. Clinical series reporting outcomes after treatment of retroversion have variable definitions of acetabular retroversion.

Cross-sectional imaging is particularly useful for the assessment of individuals with borderline hip dysplasia and acetabular retroversion. When considering joint preserving surgery, it is important to assess for cartilage degeneration and labral tears on magnetic resonance imaging (MRI) as they are best visualized using this imaging modality. However, labral tears are common on MRI, so clinical decisions should not be made solely based on that finding.³⁹ Low-dose computed tomography scan is useful for the assessment of the acetabular bone coverage of the femoral head. Focal over-coverage or under-coverage can often be better visualized on three dimensional (3D) reconstructions of the pelvis CT than on radiographs. Computational methods have been developed based on CT scans to determine potential impingement-free range of motion,^40–42 and cross-sectional imaging through the distal femur (CT or MRI) can be used to assess femoral version.

There have been multiple recent efforts to better radiographically predict instability in borderline hip dysplasia. Wyatt et al hypothesized that instability of the hip could be predicted by measuring the angle between the physeal scar of the femoral head and the sclerotic sourcil of the acetabulum.³ This was based on knowledge that the orientation and growth of the subcapital physis respond to the direction of forces applied across the hip joint during development. They refer to this measurement as the femoroepiphyseal acetabular roof (FEAR) index. A prediction can be made that the hip is stable when the LCEA is less than 25°, but the FEAR index is less than 5°. Haefeli et al used radiographic measurements of the iliocapsularis muscle, an important stabilizer of the anterior hip capsule, as a reference for microinstability.⁴³ They compared the ratio of the cross-sectional area, thickness, width, and circumference of the iliocapsularis muscle and the rectus muscle at the level of the maximum diameter of the femoral head in patients with hip dysplasia (LCEA < 25°) to patients with pincer impingement (LCEA >39°). Hypertrophy of iliocapsularis compared with the rectus muscle was found in patients with radiographic dysplasia. Other studies have worked to further characterize the location of acetabular deficiency on 3D imaging⁴⁴ and also by the direction of the instability⁴⁵ that can both be missed on plain radiographs.

Nonoperative Treatment

Nonoperative therapies are an important component in the treatment of symptomatic borderline acetabular dysplasia and acetabular retroversion. Directed physical therapy, activity modification, patient education, and anti-inflammatory medications are some of the conservative treatment options available to patients presenting with intra-articular hip disorders.⁴⁶ In a series of patients that were treated with a structured nonsurgical treatment program for symptomatic prearthritic hip conditions, 44% were satisfied with conservative care, but 56% chose to pursue surgical intervention after nonsurgical treatment.⁴⁶ Dysplastic deformity was not predictive of conservative treatment failure. Patients with higher activity levels had a greater likelihood of advancing to surgical treatment. These results indicate that a significant number of patients can be adequately managed with nonsurgical treatment.

The optimal treatment paradigm for borderline hip dysplasia remains unclear. There is one case report in the literature that details the utility of postural correction in the treatment of borderline hip dysplasia.⁴⁷ The subject was a 31-year-old female with hip pain. Imaging revealed a LCEA of 22° on the left and 18° on the right, with an acetabular labral tear on MRI. Based on variable radiographic definitions of borderline hip dysplasia, the left hip would be classified as borderline hip dysplasia and the right hip as mild or borderline hip dysplasia. Therapeutic intervention involved posture correction to reduce hip extension by tilting the pelvis anteriorly and contracting the abdominal muscles to reduce sway of the posterior trunk. Follow-up at 3 and 12 months revealed significant improvements in posture and patient-reported outcomes (PRO) including the modified Harris Hip score (mHHS), activities of daily livings, and quality of life scales of the Hip disability and Osteoarthritis Outcome Score. Postural positioning during common daily activities may provide benefit for individuals with mild instability and should be used a first-line treatment prior to pursuing surgical treatment.

The best current recommendations indicate that conservative treatment including focused physical therapy by an experienced therapist can benefit a significant number of patients and should be first-line treatment. Further studies are needed to more comprehensively determine the efficacy of nonsurgical methods in the treatment of borderline dysplasia and acetabular retroversion.

Operative Treatment of Borderline Hip Dysplasia

Surgical management of borderline hip dysplasia is an area of multiple recent clinical investigations.^{7,10,11,36,37} There is consensus that symptomatic mild and moderate/severe dysplasia is best treated with PAO with good clinical results.^12–14 Recent retrospective studies have evaluated clinical outcomes of hip arthroscopy in patients with borderline hip dysplasia. The definition of borderline dysplasia is variable in these studies. Most commonly borderline hip dysplasia is defined as a LCEA of 20 to 25° or 18 to 25°.

Concern for persistent instability after hip arthroscopy

There is significant concern that without reorientation of the acetabulum with PAO, stability cannot be achieved in the dysplastic hip joint. Soft tissue only procedures can have declining clinical outcomes over time or even acceleration of the degenerative process resulting in progressive instability and osteoarthritis.¹⁶ Case reports and case series have raised concerns that instability can be worsened or created with hip arthroscopic procedures. For example, Matsuda reported a case of early postoperative anterior dislocation after hip arthroscopy with rim trimming and resection of the labrum.⁴⁸ He highlighted the important osseous and soft tissue stabilizers that can be altered arthroscopically creating instability. The hip was stabilized with a mini-open anterior capsule repair. Preoperatively, the hip appeared to have borderline hip dysplasia. In a separate report, Matsuda and Khatod reported two cases of rapid progression to osteoarthritis after hip arthroscopy in two patients with hip dysplasia.⁴⁹ Mei-Dan et al reported yet another case of rapid progression to osteoarthritis after hip arthroscopy in a patient with mild hip dysplasia that underwent labral repair and partial debridement of a ligamentum teres tear, further emphasizing how alteration of stabilizing structures arthroscopically can potentially induce or worsen instability.⁵⁰ In an often cited clinical series, Parvizi et al highlighted the concern that persistent pain and progressive instability occurs after arthroscopic only treatment of individuals with hip dysplasia.¹⁶ They reviewed a selected series of 34 patients with persistent pain after hip arthroscopy, 30 of which had radiographic dysplasia and progression to osteoarthritis was common in this group. In larger, consecutive series of patients that underwent hip arthroscopy reported by Bogunovic et al, hip dysplasia and instability were identified as a cause of failure in 24% of patients (14/58).⁵¹ Ross et al reported a series of patients from the Academic Network for Conservational Hip Outcomes Research Study Group that underwent PAO after failed hip arthroscopy.⁵² They found that this group commonly had the diagnosis of dysplasia with a LCEA < 20° and 8/30 were classified as borderline hip dysplasia with LCEA of 20 to 25°. Findings of labral and cartilage pathology were common in these patients during the revision surgery. They associated female sex and mild-to-moderate hip dysplasia with persistent symptoms after hip arthroscopy in this group of patients that underwent subsequent PAO. These reports highlight concerns that hip arthroscopy in patients with hip dysplasia can fail to address the underlying structural abnormality and may potentially induce further instability.

Results of Arthroscopic Treatment for Borderline Dysplasia

Multiple recent series have reported good short-term clinical outcomes of arthroscopic treatment for individuals with borderline hip dysplasia. They highlight the importance of labral preservation and repair rather than debridement; as well as capsular closure and plication, especially in patients with instability. Table 1 highlights these series. Byrd and Jones reported an early series of individuals with dysplasia (16 with LCEA < 20°) and borderline hip dysplasia (32 with LCEA 20–25°).¹⁵ Labral debridement was performed in this early series. At an average follow-up of 27 months (range: 12–60 months), the average mHHS improved from 50 to 77 in the individuals with borderline dysplasia and 79% of the entire series improved at least 27 points on the mHHS. Ligamentum teres tears were common in this series (56%) and two patients underwent subsequent total hip arthroplasty.

Table 1.

Clinical series reporting outcomes of patients that underwent surgical treatment of borderline hip dysplasia. Abbreviations: ACEA, anterior center edge angle; HO, heterotopic ossification; LCEA, lateral center edge angle; mHHS, modified Harris Hip score; PAO, periacetabular osteotomy; THA, total hip arthroplasty.

Investigators	Number of Hips	Criteria for Borderline Hip Dysplasia	Treatment	Average Follow Up (Range)	Revision	Outcome Scores
Only Borderline Dysplasia
Domb et. al., 201810	25	LCEA 18–25°	Arthroscopy	68.8 months (60–83.8 months)	4 Arthroscopy	mHHS 70.3–85.9
Hatakeyama et. al., 201811	45	LCEA 20–25°	Arthroscopy	42.5 months (24.0–72.6 months)	11 hips (2 THA)	mHHS 72.1–100 (no revision group)
Chandrasekaran et. al., 20175	55	LCEA 18–25°	Arthroscopy	25.4 months (24–30.4 months)	6 Arthroscopy	mHHS 63.7–84.4
Cvetanovich et. al., 201736	36	LCEA 18–25°	Arthroscopy	2.6 +/- 0.6 years	1 Arthroscopy	mHHS 57.2–79.9
Ricciardi et. al., 20177	28	LCEA 18–25°	PAO	15 months (6–30 months)	HO excision	mHHS 58–86
Nawabi et. al., 201637	55	LCEA 18–25°	Arthroscopy	31.3 months (23.1–67.3 months) unrevised group and 21.6 months (4.7–40.3 months) revised group	2 Arthrosocpy	mHHS 61.7–86.2
Fukui et. al., 20156	102	LCEA 20–25°	Arthroscopy	40 months (24–97 months)	7 Arthroscopy and 5 THA	mHHS 63.5–84.9
Byrd and Jones, 200315	32	LCEA 20–25°	Arthroscopy	27 months (12–60 months)	2 THA	mHHS 50–77
McCarthy and Lee, 20029	144	LCEA 22–28°	Arthroscopy	Not reported	4 THA	Not reported
Includes Dysplastic Hips
Uchida et. al., 201656	28	LCEA <25° (average 22° success and 16° failure)	Arthroscopy	37.9 months (35–40.8 months)	6 Hips (2 RAO, 2 THA, 2 Shelf)	mHHS 61.6–94.3 (success group)
Larson et. al., 201528	88	LCEA <25° or Tönnis >10° (average LCEA 20.8° range 8.7° - 24.5°)	Arthroscopy	26.0 months (12–80 months)	32% mHHS <70, THA, or PAO	mHHS 81.3 at final f/u (average change 15.6)
Kalore and Jiranek, 201253	50	LCEA <25°, Sharp’s angle >40°, ACEA <25°	Arthroscopy	33 months (12–65 months)	15 Hips (30%)	Not reported

Highlighting the importance of labrum preservation, Kalore and Jiranek reported a series of patients with borderline dysplasia defined as LCEA < 25°, Sharp’s angle > 40°, or anterior center edge angle < 25°.⁵³ When these patients were compared with a series of patients without hip dysplasia, the revision rate was higher (30% compared with 14%) and patients who had labral debridement rather than labral repair had a significantly higher reoperation rate.

In an extension of a previously reported series,⁵⁴ Chandrasekaran et al reported a series of 55 patients with borderline hip dysplasia, defined as a LCEA between 18 and 25°, treated with hip arthroscopy.⁵ Labral repair and capsular plication were performed as a part of the surgical procedure. Intraoperatively, they found ligamentum teres tears in just over half of the cases. At minimum 2-year follow-up, mean mHHS score significantly improved from 63.7 to 84.4. Six of 55 (10.7%) patients underwent revision arthroscopy. In a later article, analysis was performed demonstrating that patients with ligamentum teres tear had a higher rate of revision surgery (25 and 5%) and conversion to total hip arthroplasty (5 and 0%) compared with patients without ligamentum teres tears.⁵⁵ In another report of this series of patients, follow-up was extended to minimum 5 years, where Domb et al reported maintenance of improved PRO and 19% (4/21 hips) that underwent revision arthroscopy.¹⁰ Figure 2 shows an arthroscopic image of capsular plication similar to the technique described in this series.

Figure 2:

Arthroscopic image demonstrating capsular plication in a patient with instability.

Nawabi et al reported a series comparing outcomes of hip arthroscopy in 55 patients with borderline hip dysplasia, defined as LCEA 18 to 25°, to a group of patients with a LCEA 25 to 40°.³⁷ Minimum follow-up was 23months. Intraoperatively, ligamentum teres tears were more common in patients with borderline hip dysplasia; however, there was no difference in reoperation or patient reported clinical outcomes between the groups. They also did not see a difference in patients that had debridement versus repair of the labral tear. Cvetanovich et al similarly defined borderline hip dysplasia as an LCEA of 18 to 25°.³⁶ They found no differences in clinical outcomes or revision rates with arthroscopic treatment when comparing patients with borderline hip dysplasia to patients with normal acetabular coverage. Among patients with borderline dysplasia, female patients had a significantly greater improvement in PRO scores.³⁶

Hatakeyama et al reported a series of 45 patients with borderline hip dysplasia defined as LCEA of 20 to 25°.¹¹ Labral repair, cam osteoplasty, and capsular plication were performed in all cases. At minimum 2-year follow-up, 11 of 45 (24%) patients had undergone revision surgery. Predictors of revision surgery included older age, broken Shenton’s line, Tönnis grade 1 osteoarthritis, Tönnis angle greater than 15°, and acetabular and femoral chondral damage. In another study that defined borderline dysplasia as LCEA 20 to 25°, Fukui et al reported a series of patients with borderline dysplasia that underwent similar arthroscopic treatment.⁶ They had 80 patients (82 hips) with a minimum 2 years of follow-up. They noted significant improvement in PRO. Seven patients (9%) underwent revision arthroscopy and five were converted to total hip arthroplasty (6%).

Larson et al compared clinical outcomes of patients with radiographic hip dysplasia treated with arthroscopy to patients without radiographic dysplasia treated with arthroscopy for FAI.⁴ The dysplastic group included all patients with a LCEA of less than 25°. The deformity was relatively mild with a mean LCEA of 21° and Tönnis angle of 11°. At minimum 1-year follow-up, patients with dysplasia had 60.9% good or excellent results with 32.2% failures. This was compared with 81.2% good/excellent results and 10.5% failures for FAI patients. When labral repair and capsular plication were performed, the results improved to 73% good/excellent results and 14% failure for patients with hip dysplasia. This study again emphasized the importance of labral preservation and capsular repair. Uchida et al also evaluated clinical outcomes in patients with hip dysplasia, defined as a LCEA < 25°.⁵⁶ The average LCEA in this series was 18.9°. At minimum 2-year follow-up, 32% of patients (9/28) had either progressed to osteoarthritis (Tӧnnis grade 2), revision surgery, or a mHHS < 80. Predictors of failure in their population included broken Shenton’s line, femoral neck shaft angle > 140°, LCEA < 19°, and body mass index > 23 kg/m².

Other series have reported outcomes in patients with LCEA < 20° with results comparable to patients without dysplasia with short-term follow-up.⁵⁷ McCarthy and Lee defined mild hip dysplasia as a LCEA between 22 and 28° and moderate hip dysplasia as a LCEA between 16 and 22°.⁹ They found a significantly higher rate of conversion to total hip arthroplasty in patients with moderate dysplasia (13/24, 54%) compared with patients with mild dysplasia (4/146, 3%).

There is significantly less literature describing outcomes of borderline hip dysplasia treated with PAO. Multiple papers report outcomes of treatment with PAO for patients with mild and moderate/severe acetabular dysplasia,^{12,13,58–60} but results after PAO treatment of borderline deformities are less well described. Ricciardi et al evaluated short-term outcomes in 27 patients with a LCEA between 18 and 25°.⁷ At an average 15 months follow-up (range: 6–30 months), they found a similar improvement in PRO and complications compared with patients with more severe dysplasia. Figure 3 demonstrates postoperative images of a patient with hip pain and symptoms of instability with borderline hip dysplasia and Ehlers–Danlos syndrome. This patient failed conservative treatment including extensive physical therapy and subsequently underwent hip arthroscopy with labral repair, capsular plication, and PAO. Larger clinical series with longer follow-up are needed to serve as comparison to outcomes with arthroscopic treatment.

Figure 3:

A 20-year-old female shown in Figure 1, 6 months after undergoing hip arthroscopy with labral repair, capsular plication, and periacetabular osteotomy.

Operative Treatment of Acetabular Retroversion

There are limited reports of clinical outcomes for surgical treatment of acetabular retroversion with PAO, surgical hip dislocation, or hip arthroscopy. Similar to borderline hip dysplasia, there are variable radiographic criteria for the diagnosis of retroversion. Most commonly studies report patients with symptoms of anterior impingement and radiographic findings including crossover sign, posterior wall deficiency, and ischial spine sign.

Results of Open Procedures to Treat Acetabular Retroversion

Siebenrock et al reported clinical results of anteverting PAO in 22 patients (29 hips) with acetabular retroversion at 10-year follow-up.⁶¹ Preoperatively, patients had radiographic acetabular retroversion with groin pain and a positive impingement test. No patients had undergone total hip arthroplasty at follow-up, though four hips had progressed from Tönnis grade 0 to grade 1. Four patients underwent revision surgery for reasons including persistent anterior impingement and excessive postoperative anteversion. In a later study, this same group compared intermediate-term outcomes of acetabular retroversion treated with either surgical hip dislocation or anteverting PAO.² After 5 years, there were no differences in outcomes, but by 10-year follow-up, failure was more common in patients that underwent surgical hip dislocation. Lower clinical outcome scores in the surgical hip dislocation cohort accounted for most of the difference. This likely highlights the importance of posterior wall deficiency in long-term clinical outcome as this is not addressed by anterior rim trimming alone.

An insightful treatment algorithm for the management of acetabular retroversion proposed by Peters et al used type of acetabular retroversion to determine treatment method.⁶² In patients with a LCEA < 20° or a clear posterior wall deficiency, an anteverting PAO was used to reduce anterior impingement and increase posterior coverage. PAO was indicated to treat patients with global retroversion. Patients without posterior wall deficiency and normal LCEA were treated with surgical hip dislocation, rim trimming, and femoral head–neck offset correction. Use of this treatment algorithm improved clinical outcomes and survivorship to 96% at average 4-year follow-up. In a later report, clinical outcomes at minimum 2-year follow-up were reported for 20 patients with isolated acetabular retroversion and compared with 10 patients with retroversion and LCEA < 20°.⁶³ Both groups demonstrated improved mHHS, though in the isolated retroversion group, two patients underwent subsequent hip arthroscopy with femoral head–neck osteochondroplasty, highlighting importance of addressing offset deformity at the time of surgery.

Results of Arthroscopic Procedures to Treat Acetabular Retroversion

Multiple previous studies have reported good short- and mid-term outcomes for patients with FAI treated with arthroscopy.^64–66 Hartigan et al reported a series of 78 patients (82 hips) with global retroversion that underwent arthroscopic treatment.¹⁷ At minimum of 22-month follow-up (average 39 months), there was a significant improvement in HHS with no worsening of Tönnis grade in 15 hips, and only 1 hip had undergone total hip arthroplasty. At short-term follow-up, hip arthroscopy was an effective treatment for acetabular retroversion.

Computational Analysis of Joint Mechanics in Dysplasia and Acetabular Retroversion

The spectrum of deformities even within the somewhat narrowed diagnoses of borderline dysplasia and acetabular retroversion make systematic study of outcomes in large groups of patients challenging. Computational analysis methods, such as finite element analysis (FEA) and discrete element analysis (DEA), are more commonly being used to improve understanding of hip deformities and their relationships to joint mechanics, patient outcomes, and long-term joint survival. Also, computational methods based on 3D imaging have been implemented to determine impingement-free range of motion in borderline dysplasia and retroversion patients,^40,41 similar to computational tools that have been used to study more traditional cases of hip dysplasia or cam/pincer FAI and surgical correction.^67–69

The goal of reorientation of the acetabulum with PAO is to reduce joint contact stress while preventing impingement and improving joint stability. Computational modeling provides the opportunity to quantify the alterations in joint mechanics accomplished through surgical treatment. Among the earliest work in this area, Hadley et al performed contact stress calculations on anatomy derived from anterior posterior pelvis radiographs of individuals treated for developmental dysplasia of the hip as a child. Higher calculated joint contact stress combined with years of exposure predicted worse outcomes at 30-year follow-up.⁷⁰

Discrete element analysis is a basic computational modeling technique that can be used to predict contact forces in dysplastic hips. DEA has shown that contact forces in dysplastic hips differ from those present in normal hips,^71–73 and these tools have been employed to try to guide surgeons to optimal surgical corrections to minimize joint contact stresses.^74–77 Two such studies attempted to determine the effects of joint contact stresses calculated using DEA on patient outcomes.^78,79 Perhaps unsurprisingly, the relationship between their particular contact stress calculations and outcomes were not as good as those found by Hadley et al,⁷⁰ likely due to differences in pathological thresholds found using different computational techniques.⁸⁰ While there have not been large series of borderline dysplasia or retroversion patients analyzed with DEA, this technique can be applied to those populations to determine changes in intraarticular contact mechanics with different treatments. For example, DEA modeling of the patient with instability and borderline hip dysplasia shown upon presentation in Figure 1 and after surgical correction in Figure 3 showed a reduction in peak contact stress, particularly in the lateral acetabulum, and an overall medialization of the joint forces after PAO (Figure 4).

Figure 4:

Discrete element analysis calculated contact stress map produced from the pre- and postoperative computed tomography scan of the patient in Figures 1 and 3 demonstrating decreased contact stress after periacetabular osteotomy.

Finite element analysis is a much more complex and mechanically comprehensive computational modeling technique that allows for direct inclusion of soft-tissue structures, sliding contact, advanced material property representation, and the full spectrum of mechanical stress and strain outcomes. FEA has been used very similarly to DEA to evaluate joint contact stresses in normal versus dysplastic hips^81–84 and to evaluate or guide surgical interventions based on reduction in contact stress.^85–90 Modeling overcorrection and residual impingement after PAO can best be performed using FEA,⁸⁶ which allows for incorporation of the mechanically complex soft tissues⁹¹ that are at highest risk of developing high stresses and strains and ultimately progressing to failure.⁸¹ A recent FEA modeling study of a small number of acetabular retroversion patients found that while PAO served to offload forces passing through the labrum during several activities of daily living and shift contact stress medially and posteriorly within the articular surface of the acetabulum, peak contact stress in these patients actually increased postoperatively.⁴² While this finding presently only adds to the controversy of how best to treat individuals with this deformity, it shows the potential for computational modeling to contribute to a better understanding of the effects of different surgical procedures in these unique patient populations.

Conclusion

Indications for surgical management of prearthritic hip conditions in the young adult continue to evolve. Careful, reproducible history, physical exam, and radiographic evaluation should be performed to diagnose these conditions. Borderline hip dysplasia, most often defined as a LCEA of 20 to 25°, has been managed with arthroscopic techniques and PAO. Short-term results of arthroscopy, with most studies averaging minimum 2-year follow-up, demonstrate good results with improved outcome scores and low revision rates; however, there are concerns that failure to preserve the labrum or inadequate capsular closure can worsen instability and lead to persistent symptoms. These short-term findings highlight the importance of capsular closure and plication. Long-term follow-up of the results of PAO has focused primarily on hip dysplasia with LCEA < 20°, which falls into the mild and moderate/severe hip dysplasia categories based on the most common definitions.

Acetabular retroversion causes symptoms of impingement and is diagnosed radiographically with the crossover sign, posterior wall sign, and the ischial spine sign. Most of the literature on the management of global retroversion focuses on open procedures including PAO and surgical hip dislocation. PAO has demonstrated improved long-term clinical benefit compared with surgical hip dislocation for global retroversion, likely because anteverting PAO both relieves impingement and improves posterior coverage. Arthroscopic techniques have also demonstrated good clinical outcomes at 2-year follow-up, though this is well defined in only one study. Future work in borderline hip dysplasia and acetabular retroversion should focus on incorporating computational modeling techniques to help determine best surgical approach and help us better understand long-term clinical outcomes.