Closed versus medial open reduction in the treatment of developmental dysplasia of the hip under age 1: A retrospective comparative study

Abstract

Objective:

This study aimed to compare the mid-term results of closed reduction (CR) versus medial open reduction (MOR) in the treatment of children with developmental dysplasia of the hip (DDH) under age 1.

Methods:

Thirty-four patients with DDH (41 hips) were included in this retrospective study with a mean follow-up of 4.2 years (range: 2-6.9 years). All hips were then divided into 2 groups based on the treatment type: the CR group (20 hips) and the MOR group (21 hips). All hips from both groups were assessed with post-spica magnetic resonance imaging (MRI) in the first 24 hours, and reinterventions were recorded. Medial dye pool width was also measured.

Results:

Age at the time of reduction was similar between the CR and MOR groups (6.6 ± 1.3 months vs. 6.7 ± 1.6 months). There was no significant difference between groups regarding avascular necrosis rate and further corrective surgery (FCS) requirement (P = .454, .697). The appropriate reduction was seen at 38/41 hips. Three hips in the CR group had revealed dislocation in post-spica MRIs and required re-intervention, and none of the hips in the MOR group required re-intervention (P = .107). Medial dye pool width in 3 planes showed no significant difference between MOR and CR.

Conclusion:

There is no difference in the avascular necrosis rate and FCS requirements between CR and MOR under age 1. Post-spica MRI is a favorable tool for evaluating reduction after CR, but its efficacy after MOR is questionable.

Level of Evidence:

Level III, Therapeutic study.

Highlights

• Confirming proper reduction of the hip joint is crucial, especially for posterior subluxations in children younger than 12 months with developmental dysplasia of the hip (DDH). This study aimed to assess and compare the mid-term outcomes of closed versus medial open reduction techniques in this patient demographic.

• The results showed no significant difference between groups regarding avascular necrosis rate and further corrective surgery (FCS) requirement.

• The appropriate reduction was achieved in 38/41 hips. Moreover, medial dye pool width in 3 planes showed no significant difference between medial open and closed reduction techniques.

• The results indicate magnetic resonance imaging (MRI) is effective both in providing information about avascular necrosis and in identifying potential reduction failures after closed reduction. Post-spica MRI is useful to assess the quality of closed reduction. However, the necessity of its use after medial open reduction remains uncertain.

Introduction

Developmental dysplasia of the hip (DDH) is a common disorder in the pediatric age group and requires early treatment to prevent significant morbidity.^1,2 Closed or open reduction followed by spica casting is usually necessary when the initial non-operative methods fail or when the diagnosis is made after the first 6 months.^3-5

Open reduction of the hip may be needed if closed method fails. Both anterior open reduction and medial open reduction (MOR) are generally preferred in children younger than 12 months, with good results.^6,7 Medial open reduction has several advantages, such as allowing direct access to the structures that block reduction and leaving a small and more cosmetically acceptable scar. However, with the narrow exposure through which the surgery is performed, the rate of avascular necrosis is high, and no possibility for capsulorrhaphy is available. Moreover, it is hard to obtain a good visualization of the whole joint. In this case, it may be necessary to confirm proper positioning of the hip with 3-dimensional imaging, just like in closed reduction. Once closed or medial open reduction (CR) was performed, plain radiographs were proven to be unreliable in assessing the quality of reduction,⁸ especially in detecting posterior subluxations. Therefore, 3-dimensional imaging of the affected hip is recommended to confirm proper reduction. For this purpose, computed tomography (CT)⁹ or magnetic resonance imaging (MRI)^10,11 is usually obtained in the first 24 hours following the procedure. Recently, MRI has become the method of choice, with the advantages of avoiding radiation exposure, good visualization of the non-osseous structures,^11-13 excellent assessment of hip reduction,¹⁴ and possible prediction of avascular necrosis.¹⁵

Thus, the present study aims to compare the mid-term results of closed and medial open reduction in patients with DDH under age 1, focusing on radiological and MRI evaluations.

Materials and methods

Patients who underwent closed or medial open reduction for DDH and who were operated on in a single institution between June 2015 and October 2019 were retrospectively reviewed. We excluded patients who did not have a post-reduction MRI and those who had <2 years of follow-up. Thus, the remaining 34 patients with DDH (41 hips) were included in the cohort, with a mean follow-up of 4.2 years (range: 2-6.9 years). Demographics, pre- and post-operative imaging, complications, reinterventions after reduction, and further corrective surgeries (FCSs) were recorded.

All of the dysplastic hips in the cohort were Graf type IV (dislocated hips with an alpha angle of less than 43°). An intraoperative arthrography was performed for all patients to determine the congruency of the reduction. Medial open reduction and CR were applied under same anesthesia based on the acceptable criteria for arthrography.¹⁶ Hips with Tönnis grade 1 arthrography and a larger safe zone with a wide range (>30°) underwent closed reduction, while patients with Tönnis grade ≥2 arthrography underwent medial open reduction. We performed adductor tenotomy in all patients in the MOR group and the hips a narrow safe zone (<30°) in the CR group (7 hips). After proper reduction, bilateral hip spica casting was performed. The MRI (Philips Ingenia, Eindhoven, The Netherlands, 1.5T MRI, UK) images with T1-axial sequences were obtained in the first 24 hours postoperatively to assess the need of reintervention and quality of reduction. Preoperative and latest visit radiographs were evaluated for avascular necrosis (AVN) and acetabular index (AI) as well.¹⁷ For the hips that underwent FCS, the acetabular index was measured on the last radiograph before the FCS. The AVN rates of the 2 groups were compared based on the Kalamchi–McEwen criteria.¹⁸ Reintervention refers to early surgical intervention to achieve concentric reduction, while FCS refers to late surgery for residual acetabular dysplasia or subluxation. All MRIs were performed without any sedation. Our choice for MRI includes only axial sequences, which take less than 10 minutes, with the philosophy of not missing posterior subluxation. The spica cast was remained on the patient for 12 weeks, and after its removal, an abduction brace was used for an additional 12 weeks.

Surgical procedures and MRI evaluations for concentric reduction were performed by a single pediatric orthopedic surgeon. The relationship between the femoral head and triradiate cartilage was investigated. Appropriate anterior and posterior coverage with symmetrical joint space was accepted as concentric reduction. Subluxations/dislocations and reason for reinterventions were also documented. Previously, Gans et al¹⁹ showed a strong correlation of post-reduction medial dye pool (MDP) measurements in 3 planes (neutral, 30° anterior, and 30° posterior) between arthrography and MRI after reduction. They suggested that MDP <16% indicates a favorable reduction in the axial plane. We compared medial dye pool width between the CR and MOR groups in reduced hips to assess the favorable reduction. Using standard PACS software, the femoral head and MDP width were measured in pixels on the mid-axial sections in 3 planes (Figure 1).

Figure 1.

Three-plane (30° anterior, neutral, and 30° posterior) measurement of medial dye pool (MDP) on mid-axial magnetic resonance imaging (MRI) in post-spica MRI. The MDP is calculated by dividing the A, B, and C distances in the femoral head width.

Statistical analyses were performed using the Statistical Package for Social Science Statistics software, version 23.0 (IBM SPSS Corp.; Armonk, NY, USA). Descriptive statistics are presented as mean and SDs for parametric variables and medians and ranges for non-parametric data. Chi-square and Fisher’s exact tests were used for comparing frequencies. Student’s t-test or Mann–Whitney U-test were used to compare means. To confirm a normal distribution, the Shapiro–Wilk’s test, histograms, skewness/kurtosis calculations, and detrended Q–Q plots were used. Threshold for statistical significance was designated as P < .05.

The local ethics committee of Hacettepe University approved the study (Project Number: GO/21/278). Written Informed consent was obtained from all participants.

Results

The study cohort included 41 hips of 34 patients (28 female and 6 male). Closed reduction was applied to 20 hips of 17 patients (6 right sided, 8 left sided, and 3 bilateral), while medial open reduction was applied to 21 hips of 17 patients (4 right sided, 9 left sided, and 4 bilateral). There was no statistically significant difference between the CR and MOR groups in terms of age at the time of reduction (6.6 ± 1.3 vs. 6.7 ± 1.6). Preoperatively, the median Tönnis dysplasia grade of the CR group was 2 (range: 1-4) and that of the MOR group was 3 (range: 2-4, P = .018) (Figure 2).

Figure 2.

Radiographic example for both closed (A, B) and medial open (C, D) reduction. Both patients underwent initial surgery at 6 months. Favorable radiologic outcomes were obtained during the 4-year follow-up.

Preoperative AI values (35.9° ± 2.9 vs. 38.3° ± 3.9) and AI values in the last follow-up (19.2° ± 4.1 vs. 21.9° ± 3.7) between CR and MOR, respectively, were significantly different (P = .03 and P = .02, respectively). Table 1 summarizes the demographics and preoperative radiological assessments of the study groups. At the latest visit, 2/20 (10%) hips in the CR group and 3/21 (14%) hips in the MOR group had group ≥ 1 AVN (P = .454). Further corrective surgery was performed in 3/20 hips in the CR group and 4/21 hips in the MOR group (P = .697). The indications for 7 FCS procedures were residual acetabular dysplasia in 7 hips and subluxation in 1. Mean time from index surgery to FCS was 23.2 months (range: 12-46 months). Five Dega pelvic osteotomies and 2 modified Salter osteotomies were performed for FCS procedures.

Table 1.

Demographics and preoperative radiological assessments of the study groups

	Closed (20 hips)	Medial (21 hips)	P
Age at reduction (months)	6.6 ± 1.3	6.7 ± 1.6	.29
Sex Male Female	2 15	4 13	.41
Time to follow-up (months)	52.2 ± 15.3	49.4 ± 20	.66
Preoperative Tönnis dysplasia grade	2-4 (median 2)	2-4 (median 3)	-
Preoperative AI	35.9 ± 3.0	38.3 ± 3.9	.02*

AI, acetabular index. *Statistically significant difference

The relationship between the femoral head and the acetabular fossa was evaluated with post-spica MRIs, and appropriate reduction was seen in 38/41 hips. Three hips in the CR group had persistent dislocation without concentric reduction of the femoral head within the acetabulum (Figure 3) and undergone anterior open reduction subsequently. None of the hips in the MOR group required re-intervention, and there was no statistically significant difference between groups (P = .107). Excluding 3 dislocated hips from the CR group, MDP width in the neutral, 30° anterior, and 30° posterior planes showed no significant difference between reduced hips in the MOR and CR groups (P = .837, P = .746, P = .871, respectively) (Table 2). We investigated the relationship between MDP widths and AVN or FCS rates and found no statistically significant difference as well.

Figure 3.

Immediate post-spica magnetic resonance image after closed reduction revealed a subluxated left hip. Patient treated with subsequent anterior open reduction.

Table 2.

Outcomes of the patients treated with closed reduction and medial open reduction in age-matched hips with developmental dysplasia of hip

	Closed (20 hips)	Medial (21 hips)	P
AVN (%)  Yes  No	2 (10) 18 (90)	3 (14) 18 (86)	.454
FCS  Yes  No	3 (15) 17 (85)	4 (19) 17 (81)	.697
Re-intervention after Post-spica MRI (%)  Yes  N	3 (15) 17 (85)	0 (0) 21 (100)	.107
Postoperative AI	19.2 ± 4.1	21.9 ± 3.7	.03*
MDP width (%)  Neutral  30° anterior  30° posterior	4.37 ± 4.71 4.57 ± 4.73 3.91 ± 4.54	3.67 ± 3.35 3.79 ± 3.38 3.20 ± 3.37	.837 .746 .871

AI, acetabular index; AVN, avascular necrosis; FCS, further corrective surgery; MDP, medial dye pool; MRI, magnetic resonance imaging.

Discussion

Early diagnosis and adequate treatment of DDH are critical in order to prevent later complications such as degenerative joint disease and pain.²⁰ However, the need for FCS or AVN is one of the main problems that occur as a result of DDH treatment. Several factors have been identified that increase the risk of FCS, such as residual acetabular dysplasia, higher pre-reduction Tönnis grade, older age at reduction, final AVN, or incomplete treatment. Rates of FCS after CR or MOR in the literature range from 15% to as high as 90%.^21-26 In our study, 7/41 hips (17%) required subsequent acetobuloplasty and/or femoral osteotomy, but this rate can be expected to rise with continuing follow-up. Disruption of the medial femoral circumflex artery^27,28 is the major proposed factor for AVN after MOR. Additionally, compression of the capillaries in a non-ossified femoral head causes AVN risk after both CR and MOR as well. Literature contains a high range of AVN rates varying between 10% and 50% after MOR^7,28-32 while showing relatively lower rates in CR, reported between 2% and 36%.^17,33-35 Therefore, CR can be considered a safer method in terms of AVN risk when compared to MOR. However, failure of CR may result in a delayed reduction of hip, thus leading to higher risks of AVN related to delayed open reduction.^36,37. Thus, there are some conditions with a high risk of failure of CR, which leads to a direct indication for MOR can be discussed. Although there is no definite threshold, it has been claimed that as the age at reduction increases, congruency of the hip joint will further deteriorate, leading to CR failure.³⁸ Furthermore, indication for open reduction after failed CR was frequently applied in hips with bilateral DDH and greater preoperative Tönnis grade. In our cohort, we preferred MOR in hips with <30° safe zone or Tönnis grade ≥2 arthrography. In our age-matched comparison, there was no difference in rates of AVN between CR and MOR.

For such reasons mentioned above, the evaluation of the quality of reduction after spica casting in children younger than 12 months has become crucial. It is important to evaluate the reduction as best as possible due to the relatively high rate of malreduction or loss of reduction, especially after closed reduction.^3,39 Imaging methods such as plain radiographs, CT, and ultrasound, when used for reduction quality assessment, have some drawbacks, such as non-ionized radiation exposure, the requirement of a certain level of experience, not allowing soft tissue evaluation, or providing images in a towplane.⁴⁰ Due to the limitations mentioned above, MRI has become a popular alternative for post-reduction evaluation in the last 2 decades.^12,41 Several previous studies have demonstrated that MRI is effective both in providing information about avascular necrosis by evaluating femoral head perfusion and in identifying potential reduction failures after closed reduction.^19,42-45 Dibello et al⁴⁶ evaluated 25 patients (29 hips) who underwent spica casting followed by closed reduction with a post-reduction MRI. Re-intervention was required in only 1 patient who subsequently underwent open reduction. They achieved a concentric reduction in the remaining patients. Jadhav et al⁴⁷ studied 74 patients (107 hips) with a mean age of 16.4 months (range: 4-63 months) who underwent post-reduction MRI. Of the 52 hips that underwent closed reduction, 16 (31%) re-interventions were needed, and in the open reduction group, which includes 55 hips, only 1 (2%) re-intervention was required. However, unlike our study, the patients had pelvic or femoral osteotomy combined with the reductions. Although the studies mentioned above provide valuable information, there is no study in the literature comparing the efficacy of post-reduction MRI in patients who underwent medial open and closed reduction. In contrast to anterior open reduction, the posterior part of the hip joint cannot be evaluated during medial open reduction. We believe that post-spica MRI may be ineffective after the anterior open reduction procedure, which allows better visualization of the hip joint. Thus, in our clinic, we routinely use post-spica MRI only for closed and medial open reductions. In the present study, 3 patients needed a re-intervention in the CR group, but we had no re-interventions in the medial open reduction group.

Unfortunately, there were only 21 CR and 20 MOR hips, and this study may not be sufficient to detect a difference in AVN and re-intervention rates. But considering that this is the first study that compares MRI efficacy regarding re-intervention between CR and MOR in an age-matched cohort, this limitation can be excused. The present study also has a mean follow-up of 4.2 years, which may affect the report of lower AVN and FCS rates. Future studies can demonstrate the necessity to obtain post-spica MRI after MOR.

In conclusion, although MOR has its own difficulties, there is no difference in the avascular necrosis rate and further corrective surgery requirements when compared to closed reduction. Additionally, post-spica MRI is an effective tool for the assessment of proper reduction after closed reduction. Its efficacy in patients who underwent medial open reduction is questionable, but future studies with larger cohorts could be designed.

Funding Statement

The authors declared that this study has received no financial support.