Short-term outcomes of treatment in children presenting with DDH in walking age - An analysis of 84 hips

Abstract

Background

Developmental Dysplasia of Hip (DDH) presenting at walking age is not uncommon, particularly in developing countries. The available treatment modalities in this age group are closed reduction (CR), open reduction (OR), OR with additional femoral and/or pelvic osteotomy. This study was done in patients who presented between 12 and 36 months of age to assess the following: 1) Percentage of hips amenable for successful CR, 2) failure rate after CR and OR in walking age DDH and 3) the need for secondary procedures to address subluxation and residual dysplasia within first 2 years.

Method

ology: After IRB approval, the institutional database was searched for patients admitted with a diagnosis of DDH from January 2009 to January 2019. We identified 142 patients, of which 65 patients with 84 hips formed the study cohort after applying inclusion and exclusion criteria. Demographic details, details of the interventions, brace wear, revision procedures and radiological data were collected from Hospital Information System. We divided the patients in three groups: Group I - CR, Group II –OR, and Group III - OR with an additional bony procedure in the form of femoral and/or pelvic osteotomy.

Results

The mean age at presentation was 20.1 months. We had 10 (11.9%) hips in group I, 39 (46.4%) hips in group II and 35 (41.6%) hips in group III. The mean follow-up was 44.8 months (24–132 months). In Group I, 5 (50%) had re-dislocation and 2 (20%) needed revision intervention for residual dysplasia. In Group II, 4 (10%) had re-dislocation and 4 (10%) needed revision intervention. In Group III, 5 (14.2%) hips needed revision intervention for residual dysplasia. The mean final AI was 24.6°in Group I, 28.2° in Group II and 26.3°in Group III. There was no significant difference in the final AI between the groups (p > 0.05).

Conclusions

An attempted closed reduction has a 50% failure rate, and we recommend a low threshold for open reduction. There is a 10% rate of re-dislocation following open reduction with or without additional bony procedure. About 50% of the dysplastic hips treated without pelvic osteotomy at the time of index procedure fail to remodel and have residual acetabular dysplasia.

1. Introduction

Developmental Dysplasia of the Hip (DDH) is the most common disorder affecting the newborn hip, with an estimated incidence of 1.4 to >35 per 1000 live births.¹ It could be subtle and is often missed during the early years of life. Many times, it is recognized by the abnormality in gait when the child starts walking. Hence, walking age DDH is not uncommon, particularly in countries where hip screening protocols and awareness about healthy swaddling are not in place.

The management of walking age DDH could be challenging to the surgeon in terms of the complexity of the problem and choosing the right treatment option for a given patient. Many treatment options are available for those children presenting in the age group between 1 and 3 ranging from closed reduction (CR), open reduction (OR), OR with additional femoral and/or pelvic osteotomy.^{2, 3, 4, 5, 6, 7, 8, 9} Variable rates of success have been reported in the literature for each of these. Though successful CR for walking age DDH is reported in the literature, up to 30% of hips need secondary reconstructive procedures for recurrent sub-luxation or persistent dysplasia.⁴ Multiple factors could influence the outcomes of surgical intervention in this age group, and controversies still exist regarding the ideal age for pelvic osteotomy and the role of bony procedure during the index procedure.

At the end of clinical and radiological assessment it may not be possible for the surgeon to comment on a definitive management plan for a given child. The usual trend is to discuss the various options with the parents and counsel them for the best option after evaluation under anaesthesia with an arthrogram. However, it would be good to have institutional data regarding the success and failure of various modes of intervention and the need for additional procedures to address residual problems for healthy pre-operative counselling. A retrospective analysis of outcomes of various methods will also provide an opportunity to change the management protocols and practices based on the data.

In this background, we conducted this retrospective analysis in a cohort of patients who underwent the primary intervention for DDH between the age of 12–36 months (1–3 years) to answer the following questions: 1)What percentages of these hips are amenable for successful CR? 2)What is the failure rate after CR and OR in walking age DDH? 3)How many of them need secondary procedures to address subluxation and residual dysplasia? and 4)What is the incidence of complications within first 2 years?

2. Methodology

After approval from the Institutional Review Board, the hospital database was queried for all the patients aged above 1 year and admitted with a diagnosis of DDH between January 2009 to January 2019. The following inclusion criteria were applied: 1) All patients who had their index intervention between 12 and 36 months of age and 2) All patients who had completed at least 2 years of follow-up. The following exclusion criteria were applied: 1) All patients with associated congenital anomalies of the lower limb and those with identified syndromes and 2) All patients who had some form of intervention done elsewhere. We identified 142 hips with the diagnosis of DDH in the above-mentioned period. After applying the inclusion and exclusion criteria, we had 65 patients with 84 hips, which formed the study cohort.

The details of all these 65 patients were collected from their case records. The radiographic data was collected from PACS (Picture Archiving and Communication System). These 84 hips were divided into three groups based on the index intervention. Group I was all the hips who underwent CR. Group II was all those hips that had an OR, and Group III was those hips treated by OR with an additional bony procedure in the form of femoral and/or pelvic osteotomy.

The data collected included age, sex, index procedure, time in spica, rate of re-dislocation and subluxation, number of secondary procedures required, details of all the secondary procedures, occurrence of avascular necrosis (AVN) and other complications, final follow-up time and status of the hip at final follow-up. The IHDI (International Hip Dysplasia Institute) Grade were measured at the following intervals – 1. Pre-operative and Post-operative (at 3 months after spica removal) of index surgery, 2. Pre-operative and Post-operative of each secondary procedure, 3. final follow-up. The Radiographic parameters evaluated included Acetabular Index (AI) and Center-Edge angle (CEA) for children >6 yrs of age. The clinical outcome was measured at the final follow-up using Modified McKay criteria, which includes assessing stability of hip, range of motion, Trendelenburg sign, presence of limp and presence or absence of hip pain. Statistical analysis was done using SPSS v.25. One way ANOVA test was used to compare the means between the Groups and Games-Howell post-hoc test was employed as the numbers in the groups were significantly different based on the Levene test. Chi-square test was used to assess the association between categorical variables. A p-value of <0.05 was considered significant.

2.1. Institutional treatment protocol

All the patients in this age group (1–3years) which were previously untreated were attempted for a trial of closed reduction under anaesthesia. If the hip was reducible, an arthrogram and adductor tenotomy was performed in all of them. The reduction was considered to be stable if it was found to be stable through a 40° range of abduction and had medial dye pooling <6 mm. If the safe zone was less than 40°or the medial dye pooling was >6 mm, open reduction was planned. The open reduction was done through standard bikini incision and Smith-Peterson interval. Iliopsoas tenotomy was routinely done in all the hips. If the femoral ante-version was more than 40°, a femoral osteotomy was performed at the same time. The decision to add a pelvic osteotomy was made based on the cartilaginous acetabular index, anterior coverage and stability of the hip.

Following the index surgery, irrespective of the procedure, all hips were immobilized in spica for 6 weeks. Depending upon the arc of stability, hips were immobilized in 40–60°of abduction and 60–90° of flexion. At 6 weeks, spica was changed under anaesthesia which was retained for 6 more weeks. Following the removal of the spica, patient was mobilized and allowed to walk with an abduction brace, which was continued for the next 6–9 months. During every follow-up visit, the hip was assessed for stability, range of motion, presence of pain, Trendelenburg gait and limb length discrepancy. Radiological evaluation was done with antero-posterior and frog-leg lateral pelvic radiographs to look for containment of the hip and residual dysplasia.

2.2. Secondary procedures

Secondary procedures were performed in two scenarios- 1. Failure of index procedure, 2. Presence of residual dysplasia. Secondary procedures were usually conversion to OR or revision OR with or without bony procedures. Bony procedures involved either a femoral shortening or Varus de-rotation osteotomy with or without a pelvic osteotomy. Pemberton's acetabuloplasty, San Diego acetabuloplasty or Triple innominate osteotomy were the pelvic osteotomies performed.

2.3. Complications

Complications in this study were classified as Major and Minor. Major complications were re-dislocation (failure of closed or open reduction within 6 months of index intervention), avascular necrosis (AVN) and stiff hip. AVN was classified according to Kalamchi & MacEwan classification. Minor complications encountered were plaster sore, soiled spica and broken spica.

3. Results

The mean age at presentation was 20.1 months (12–36 months, SD – 7.08). We had 10 hips in group I, 39 hips in group II and 35 hips in group III. The mean age at index procedure was 19.3 months in group I, 17 months in group II and 23.8 months in group III. There was a significant difference in the mean age between Group II and Group III (p < 0.05). The mean follow-up was 44.8 months (24–132 months). There were 46 girls and 19 boys. Forty-six were unilateral with equal distribution on right and left side, and 19 were bilateral. The distribution of patients in each group and their demographic data are presented in Table 1.

Table 1.

Demographic Data of the patients.

	Group I	Group II	Group III
No. of patients	10	29	26
No. of Hips	10	39	35
No. of Bilateral Hips	0	10	9
Girls	10	18	18
Boys	0	11	8
Mean Age (months)	19.3	17	23
Mean Follow-up (months)	42.6	38.5	53.6

3.1. Group I - closed reduction

A total of 10 hips (11.9%) were treated by CR and spica application. Five of these 10 hips had re-dislocation and needed secondary OR with or without femoral osteotomy, amounting to a failure rate of 50%. Of the five hips with a successful CR, two (40%) needed secondary intervention due to subluxation or residual dysplasia. The overall rate of revision intervention in group I was 70% (Table 2).

Table 2.

Re-dislocation and revision interventions.

	Group I	Group II	Group III
No. of Hips	10	39	35
Re-dislocation (%)	50%	17.9%	22.8%
Subluxation (%)	20%	2.5%	2.8%
Revision Intervention (%)	70%	20.4%	28.5%

3.2. Group II - open reduction

The hips treated by OR and spica application formed the largest group in our series, with 39 hips (46.4%). Of these, 4 hips had re-dislocation, amounting to a failure rate of 10.2%. All of them were treated with revision OR. Four (10.2%) of the 39 hips needed secondary intervention at a mean follow-up of 33 months (range 24–78 months) for residual dysplasia and subluxation. Three of them had pelvic osteotomy and one needed combined femoral and pelvic osteotomy.

3.3. Group III - open reduction with bony procedure

Of the 84 hips, 35 (41.6%) were treated by OR with femoral and/or pelvic osteotomy. Of these, 24 hips had femoral osteotomy alone, and 11 had femoral and pelvic osteotomy. At a mean follow-up of 48 months (range 24–132 months), 5 hips (14.2%) needed secondary intervention in this group for residual dysplasia. All these 5 patients did not have pelvic osteotomy at the time of index procedure.

3.4. Radiographic outcomes

The mean pre-operative AI was 34.8°, 37.8 and 37.9°in Group I, Group II and Group III, respectively. The mean final AI was 24.6°in Group I, 28.2° in Group II and 26.3°in Group III. There was no significant difference in the final AI between the groups (p > 0.05). There were 81 hips (96%) in IHDI grade III and IV pre-operatively, and only 3 were in IHDI grade II. At the final follow-up, 9 hips (10%) persisted in IHDI Grade III and IV (Table 3).

Table 3.

Radiologic parameters.

	Group I	Group II	Group III
No. of Hips	10	39	35
Pre-op IHDI:
I	0	0	0
II	0	2	1
III	3	9	5
IV	7	28	29
Mean Pre-op AI	34.8°	37.8°	37.9°
IHDI at Final Follow-up:
I	8	33	30
II	1	1	2
III	1	2	2
IV	0	3	1
Mean Final AI	24.6°	28.2°	26.3°

3.5. Functional outcomes

Overall, 68 hips (81%) in our series had excellent outcomes and 7 hips (8.3%) had good outcomes irrespective of the type of intervention. Excellent outcomes were found in 80% hips in Group I, 84.6% Hips in Group II and 77.1% in group III (Fig. 1). There was no significant difference in the outcomes between the groups (p = .89). Poor outcomes were seen in 1 hip (10%) in Group I, 4 hips (10.25%) in Group II and 4 hips (11.4%) in Group III, respectively.

Fig. 1.

Bar chart showing the functional outcomes in different Groups.

3.6. Complications

A total of 12 hips (14.3%) had re-dislocation after the index procedure. Of these, 5 were in Group I, 4 were in Group II and 3 were Group III. Of the 84 Hips, 3 (3.5%) hips developed AVN; 1 was in Group I and 2 were in Group II. One patient developed a stiff hip which was managed with Physiotherapy. Minor complications were plaster sore in 3 hips, broken spica in 2 hips and soiled spica in 1 hip.

4. Discussion

Early detection and appropriate commencement of treatment in DDH is essential not only to achieve good outcomes but also to avoid complications associated with late interventions.¹ However, walking age DDH is not an uncommon presentation in developing countries, and it still forms a significant proportion of patients presenting to the orthopaedic units with hip pathologies. The goal of treatment in these children is to achieve a well reduced, stable and concentric hip joint which would in-turn function normally for many years of life. The ideal intervention for a child with DDH who presents primarily between 12 months and 36 months still remains controversial.² The options available are CR, OR, and OR with an additional bony procedure in the form of femoral/pelvic osteotomy.

4.1. Closed reduction

The outcomes of CR in walking age DDH reported in the literature is variable.^{3,4,10, 11, 12, 13, 14, 15} Most of the studies define failure of CR as the need for a subsequent OR following a re-dislocation.¹⁶ Zhang et al., in their series of 138 hips treated by CR, reported a failure rate of 31% and concluded that age more than 18.35 months and IHDI Grade III & IV are more likely to fail with CR.⁴ Schoenecker et al. reported a successful CR in 68% of 38 hips, all of which were >18 months of age.³ However, more than 50% of these hips needed secondary femoral or acetabular procedures for residual dysplasia. They concluded that age <22 months and Tonnis grades I & II are good prognostic factors for a successful CR. Li et al. published the outcomes of 51 hips between 24 and 36 months of age which were treated by CR and spica.¹⁷ They could get a successful CR in 76.5% of their patients with a 5% AVN rate. They also reported that 43.6% of these patients had residual acetabular dysplasia.

In our series, we were able to achieve a CR in 11.9% (10/84) hips. Of these, 5 hips had re-dislocation at the time of spica removal and were converted to OR. As our failure rate with CR was 50%, we recommend a very low threshold for an OR in a child presenting with DDH after 12 months of age. It would be prudent to err on the side of OR in cases with uncertainty regarding the stability of reduction. The need for a secondary procedure in the form of femoral osteotomy with or without acetabuloplasty was high in this group. Two of the 5 hips (40%) who initially had successful CR had subluxation and needed re-intervention. All these patients were not compliant with brace wear, which could be the possible explanation for re-subluxation.

4.2. Open reduction

In our series, 46.4% (39/84) needed OR to achieve a stable concentric reduction. Three (7.7%) of these hips had re-dislocation at the time of spica removal and required revision OR with femoral or pelvic osteotomy. Gholve et al. reported the outcomes of OR in 49 hips with a 5-year follow-up with a failure of OR in 8% of the hips, which needed revision OR.⁶ Swarup et al. also reported that 5.4% of the hips in their series needed a repeat OR due to failure of index OR.¹⁸ Sankar et al. reported a rate of 5.9% (25/421) failure of OR in their series. They concluded that right side DDH, bilaterality, greater pelvic width and decreased abduction of the spica case as risk factors for re-dislocation.¹⁹

During follow-up, four of these 39 hips (10.2%) needed secondary procedures for residual dysplasia. The rate for the need for subsequent ipsilateral surgery after an index OR in a child <3 years with DDH was found to be 14.2% in a study of 260 patients by Swarup et al.¹⁸ Gholve et al. concluded that 49% of patients who underwent an OR in isolation without additional bony procedure needed secondary procedures to address the residual dysplasia.⁶ The 10% rate of secondary procedures after OR in our series may not be a true representation of the need for secondary procedures, as the follow-up is only two years. Miyake et al. reported a long-term follow-up analysis of 85 hips treated with OR with good outcomes in 77.6% of those hips functioning well at skeletal maturity.²⁰

Moris et al. in a recent study comparing residual dysplasia in closed versus open reduction in late presenting DDH concluded that the closed reduction group was associated with increased residual dysplasia and more secondary interventions.²¹ In our series, though some of the hips in closed reduction group remodelled well, the need for secondary intervention was more in this group compared to the open reduction group.

4.3. Open reduction with femoral/pelvic osteotomy

Chen et al. reported the safety of single-stage OR, femoral shortening with de-rotation osteotomy and Salter's osteotomy in 68 hips with good outcomes.⁹ They concluded that the rates of AVN were significantly less when this procedure was performed less than 4 years of age. Erturk et al. did a comparative analysis of 49 hips in 2–5 years age group treated with single-stage OR, femoral osteotomy and Salter's osteotomy and found no significant difference in the rates of AVN between <3 years and >3 years, and concluded that single-stage reconstruction could be safely done.²² The safety of simultaneous bilateral OR with a pelvic osteotomy (Pemberton/Dega) was reported recently in a cohort of 70 patients by Alsiddiky et al.²³ Many studies have shown that combined OR with pelvic osteotomy was better than OR alone in normalization of radiological parameters and AI.^{24, 25, 26}

In our series, 20.1% (24/84) of patients were treated with OR and femoral osteotomy, and 9.2% (11/84) with OR, femoral and pelvic osteotomy. At the time of final follow-up, 34.3% (12/35) of the hips in Group III had residual dysplasia (AI >30°). All these 12 hips had femoral osteotomy without pelvic osteotomy. None of the hips treated by pelvic osteotomy had residual dysplasia (Fig. 2). The remodeling after OR and femoral osteotomy was satisfactory in 50% (12/24) hips. Hence, adding a pelvic osteotomy routinely in all cases in this age group might result in unnecessary surgery in 50% of them. Considering the fact that pelvic osteotomy involves more dissection and blood loss, we prefer to stage the pelvic osteotomy whenever possible. All those hips with stable reduction and a good cartilaginous acetabular roof on the intra-operative arthrogram are allowed to remodel naturally. An add-on pelvic osteotomy is done if the acetabular index does not improve in the next two years. We typically prefer to do the pelvic osteotomy at the time of implant exit in all those hips which failed to remodel. Forlin et al. in their series of 44 hips treated with one-stage OR and pelvic osteotomy with or without femoral osteotomy, reported a re-dislocation rate of 9%.⁸ In our series of 34 hips who underwent OR with an additional bony procedure, there were 3 (8%) re-dislocations; two were in the hips who had femoral osteotomy alone, and one was in a patient who had femoral osteotomy and acetabuloplasty.

Fig. 2.

A- X-ray pelvis with both hips antero-posterior view of a 26-month-old girl showing dislocated L hip with acetabular dysplasia (IHDI grade IV), B- X-ray pelvis with both hips antero-posterior view after open reduction, femoral VDO and Pemberten's osteotomy, C- Follow-up radiograph at 12 weeks showing contained L hip with healing femoral and pelvic osteotomies (AI: R- 14.8°, L-17.7°), D- Follow-up radiograph at 18 months post-surgery showing complete healing with well contained hip without residual dysplasia (AI: R- 12.6, L-13.2°), E− Final follow-up radiograph of the pelvis after implant removal showing good remodeling of the femoral osteotomy and acetabulum (AI: R- 9.7°, L- 5.4°).

The reported rates of AVN following CR varies between 2 and 36%.¹³ Zhang et al. reported an 8.3% incidence of AVN in their series of 72 hips treated with CR.⁴ Novais et al. in their meta-analysis reported an 8% AVN following CR and 19% AVN following OR.²⁷ The incidence of AVN in our series was 3.5%, and two of them was in the OR group, similar to the trend reported by Novais et al. We did not have any AVN in Group III.

5. Limitations of the study

Our study has a number of limitations. Although we had 84 hips in total, the distribution in each group was not uniform and hence comparison between these groups with statistical methods might be questionable. A significant number of hips (42 hips) were lost to follow-up in our series and hence the results may not be the true representation of the outcomes. The follow-up duration is short for this condition and a longer follow-up is needed to give definitive conclusions.

6. Conclusion

An attempted closed reduction in this age group has a high failure rate and we recommend a low threshold for open reduction. A meagre percentage (6% in our series) of hips in this group are amenable for a successful closed reduction and the need for secondary intervention within the next few years needs to be counselled. There is a 10% rate of re-dislocation following open reduction with or without additional bony procedure. About 50% of the hips treated without a pelvic osteotomy at the time of index procedure fail to remodel and have residual acetabular dysplasia.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Research ethics and patient consent

Written consent for publication of patient details were obtained from the parent/guardian.

Declaration of competing interest

All the Authors declare that there is no conflict of interest.