DDH in the Walking Age: Review of Patients with Long-Term Follow-Up

Abstract

Introduction

The best treatment option in children with late detected DDH is still a subject of much controversy and only few studies have investigated the long-term outcome of treatment in such patients. We performed a systematic review to assess long-term outcome of late detected DDH hips treated after walking age.

Methods

Studies met inclusion criteria if they: (1) reported at least 30 hips treated; (2) included children aged between 9 months and 12 years; (3) treatment indication was late detected DDH after walking age; (4) presented a minimum follow-up of 10 years; (5) reported a clinical or radiological outcome. The Kaplan–Meier method was used to evaluate long-term survival according to clinical and radiological outcomes. The rate of total hip replacement (THR) was retrieved.

Results

From a total of 6561 articles, 13 articles with grade IV level of evidence were included in our review. A total of 988 hips in 800 patients with a mean follow-up of 27.9 years (range 10–67) were included. The mean age at surgery was 3.3 years (range, 9 months–12 years). The rate of THR increased according to the length of final follow-up. In particular, all studies reported no case of THR at 23.5 years of follow-up, a rate of 10.2% of THR between 30 and 40 years of follow-up and a rate of 35.6% of THR in patients with follow-up more than 40 years.

Conclusions

In patients with late detected DDH, most THR became necessary more than 30 years after the index procedure and their number increased further after 40 years and more of follow-up. Late detected DDH diagnosed after walking age is a life-long disease.

Introduction

Development dysplasia of hip (DDH) is one of the most frequent hip disorders in children with an incidence of 0.1–6.6 cases per 1000 live births, and it is responsible for up to 30% of primary total hip replacement (THR) in patients up to 60 years of age [1, 2].

Several abduction systems such as Pavlik, Tübingen, von Rosen, Frejka, and others have been successfully used to treat DDH patients younger than 6 months of age. However, in the event of late detected DDH, particularly after walking age, the treatment is more challenging and the long-term functional outcome is variable.

At present, early screening programs have decreased the incidence of late detected DDH, especially in the western hemisphere [3, 4] although, it is still a major concern in developing countries [5]. Thanks to such programs, children with DDH can benefit from early treatment and outcome is generally good. Starting from the twentieth century, open reduction gradually became popular in Europe and North America, and various forms of osteotomy correction gradually emerged, such as Salter Innominate Osteotomy (SIO) [6], Pemberton osteotomy (PO) [7], Dega osteotomy (DO) [8], and triple osteotomy [9] which have been used to treat patients with late detected DDH. However, it is neither clear what the best treatment option in children with late detected DDH after walking age is, nor is the long-term outcome of surgical treatment a good long-lasting certainty.

The current trend is that more and more orthopaedic surgeons perform open procedures to treat late detected DDH, and the age of open reduction also has a decreasing trend. In a survey by the Paediatric Orthopaedic Society of North America (POSNA), 32 and 13% of POSNA members set the minimum age for open reduction at 6 and 12 months, respectively [10]. Similarly, of the 52 cases (71 hips) of DDH treated with open reduction reported by Farsetti et al. [11], the mean age was 16 months (range 3–36). On the other hand, Ning et al. [12] reported 23.8% of patients with late detected DDH treated by open reduction with pelvic and femoral shortening osteotomy were aged between 1.5 and 2.5 years of age. Similar results were also reported by Castaneda et al. [13] and Alexiev et al. [14]. However, the number of publications analysing the long-term outcome of late detected DDH after walking age is relatively scarce and heterogeneous.

The aim of this systematic review is to evaluate the long-term outcome of children with late-onset DDH treated surgically after walking age.

Materials and Methods

Search Strategy

The literature search was completed using Medline, PubMed, Web of Science, EMBASE, Google Scholar, and Cochrane Library databases from the earliest date possible up to May 2021. The review includes retrospective, prospective and longitudinal cohort studies. Narrative and systemic reviews were also included. In addition, editorials and correspondences were screened. All references obtained using the search strategy were imported into Endnote X7 (Thomson Reuters, Philadelphia, PA, USA). The relevant studies were screened for eligibility. The research was not limited by a specific method of treatment or surgical approach. According to the predetermined inclusion and exclusion criteria, two junior researchers (GL and RS) independently screened (titles and abstracts) and extracted the data. Full text of all articles considered relevant after the initial title and abstract screening were retrieved and assessed for eligibility. The references of all selected articles were reviewed as well to find potential articles that were missed. Any remaining non-eligible articles were excluded and duplicate articles were removed. Disagreements were resolved by consensus among junior researchers and senior authors (FC and LY). Our analyses accorded with the Preferred Reporting Items for Systematic Review and Meta-analysis (PRISMA) guidelines for a systematic review of success rate [15].

Search Terms and Delimiting

Search terms included “DDH”, “hip dislocation”, “congenital hip dislocation”, “closed reduction”, “open reduction”, “walking age”, “skeletal traction”, “AVN”, “outcome” and “paediatric” including combinations of the index and free-text terms, as recommended in the Cochrane Handbook for Systematic Reviews of Interventions. The search was restricted to the English language and human participants. Table 1 shows the bibliographic database search strategy (Table 1).

Table 1.

Bibliographic database search strategy

Database	Code	Date	Found
Pubmed	[“Hip dislocation, congenital”(mesh) or (“Developmental dysplasia of the hip”(mesh) or (“DDH”))] and [“closed reduction” or (“open reduction” or (“walking age” or (“skeletal traction” or (“avascular necrosis” or (“outcome” or (“pediatric”))))))]	16 May 2021	2363
Medline	[“Hip dislocation, congenital”(mesh) or (“Developmental dysplasia of the hip”(mesh) or (“DDH”))] and [“closed reduction” or (“open reduction” or (“walking age” or (“skeletal traction” or (“avascular necrosis” or (“outcome” or (“pediatric”))))))]	16 May 2021	1376
Embase	(‘hip dysplasia’/exp or ‘congenital hip dislocation’/exp) and (‘closed reduction procedure’:ab, ti or ‘open reduction procedure’:ab, ti or toddler:ab, ti or ‘skeleton traction’:ab, ti or ‘treatment outcome’:ab, ti or ‘avascular necrosis’:ab, ti or ‘pediatric’:ab, ti)	16 May 2021	993
Web of Science	(developmental dysplasia of the hip or congenital hip dislocation) and (closed reduction or open reduction or outcome or walking age or avascular necrosis or skeletal traction)	16 May 2021	1512
Google Scholar	(developmental dysplasia of the hip or congenital hip dislocation) and (closed reduction or open reduction or outcome or walking age or avascular necrosis or skeletal traction)	16 May 2021	First 200
Cochrane	(developmental dysplasia of the hip or congenital hip dislocation) and (closed reduction or open reduction or outcome or walking age or avascular necrosis or skeletal traction)	16 May 2021	117
Total			6561

Selection Criteria

Studies met the inclusion criteria if they: (1) reported at least 30 patients treated (2) included children from 9 months of age without any underlying neuromotor disorder [16] (3) treatment indication was late detected DDH after walking age; (4) presented a minimum follow-up of 10 years (5) assessed the outcome with a clinical or radiological score.

Exclusion criteria were as follows: (1) pathological dislocation of the hip, (2) studies from which statistical data could not be extracted, (3) studies that did not conform to the inclusion criteria, (4) study with MINORS quality evaluation lower than 8 points [17].

Quality Evaluation

The quality of the included literature was assessed with the MINORS item quality evaluation checklist [18]. The 12 MINORS quality evaluation indicators include (1) clearly stated aim, (2) inclusion of consecutive patients, (3) prospective collection of data, (4) end points appropriate to the aim of the study, (5) unbiased assessment of the study end point, (6) follow-up period appropriate to the aim of the study, (7) loss to follow-up less than 5%, (8) prospective calculation of the study size, (9) an adequate control group, (10) contemporary groups, (11) baseline equivalence of groups, (12) adequate statistical analysis. The first eight items are for both comparative and non-comparative studies, while the last four are for comparative studies only. The items are scored 0 (not reported), 1 (reported but inadequate), or 2 (reported and adequate). The global ideal score is 16 for non-comparative studies and 24 for comparative studies [17].

Literature Screening and Data Extraction

The extracted data included: (1) the characteristics of the included literature, including the first author, year of publication, nationality, number of patients, time of case collection, and type of study; (2) clinical characteristics of patients, including age at diagnosis, gender, number and classification of DDH, treatment duration, treatment success rate, the incidence of complication, and outcome.

Figure 1 displays the flowchart of the total search and study selection (Fig. 1). Of the 73 studies assessed for eligibility, 60 were further excluded after the full text was comprehensively examined. Finally, 13 studies were included in the review. Data on 800 paediatric patients (988 hips) with DDH treated after walking age were collected from the included studies (Table 2).

Fig. 1.

Flowchart of total search and study selection

Table 2.

Population characteristics and follow-up

Author	Number of hips	Number of patients	Male	Female	Mean age at surgery (years)	Min age at surgery (years)	Max age at surgery (years)	Mean follow-up( years)	Min FU (years)	Max FU (years)	Mean age at final follow-up	Traction pre-op
Hess 1996 [19]	33	22	4	17	2.5	1.1	4.3	23, 5	19.7	28.4	25.8	Yes
Böhm 2002 [20]	66	61	10	51	4.1	1.3	8.8	30.9	26.2	35.4	35	No
Thielemann 2003 [21]	100	79	11	68	4, 8	1.1	12.3	14.7	10.4	19.5	19.5	No
Fujii 2004 [22]	68	56	5	51	1.8	0.9	2.7	18	12	24	18	No
Angliss 2005 [23]	191	147	13	134	2.3	1.15	7.9	33	25	48	35.5	Yes
Thomas 2007 [24]	80	60	13	47	2.8	1.5	5	43.3	40	48	45.8	Yes
Rampal 2008 [25]	47	36	5	31	3	1	4.9	14.3	10	30	16.1	Yes
Remmel 2008 [26]	43	43	4	39	1.7	1.15	5.5	32.9	30	37.9	34.6	No
Wu 2010 [27]	49	49	NA	NA	1.7	1.5	3	11.1	10.2	13	13.3	No
Morin 2011 [28]	32	23	2	21	2.6	1.5	5	37	31	44	39	Yes
El-Sayed 2015 [29]	58	48	16	32	4.1	2.1	7.5	16.5	13.1	25.1	20.6	No
Miyake 2018 [30]	85	73	9	64	1.4	0.8	2.7	17.7	13	32	19	No
Scott 2020 [31]	78	58	NA	NA	2.6	1.5	5	41.9	30	48	44.5	Yes
	58	45	NA	NA	2	1.5	4.2	45.6	25	67	47.6	No
Total/average	988	800	92	555	3.3	0.4	4.5	27.9	24	39.3	27.8	46.2%

Cohort Description

Tables 2 and 3 show retrieved articles, population characteristics, and surgical strategy (Tables 2 and 3). We present the studies according to the type of the procedure performed: preoperative traction, open or closed reduction, acetabular and/or femoral osteotomy, soft tissues and mixed procedures. The clinical survival was recorded at the last follow-up visit, based on clinical evaluation or clinical scoring systems when available; an excellent or good outcome was considered as survival. The radiological follow-up was also recorded according to the radiological scoring employed by the specific study. The percentage of hips that underwent THR was retrieved.

Table 3.

Surgical strategy and outcome

Author	Reduction	Index surgery	Femoral osteotomy	Type of femoral osteotomy	Acetabular osteotomy	Redirectional reshaping	Acetabular procedure	Postoperative AVN	Clinical survival	Radiological survival	THR	Type of score
Hess 1996 [19]	Open	Femoral + acetabular osteotomy	Yes	Varus derotation osteotomy	Yes	Yes	Mittelmeier	6%	84%	63%	0%	Tonnis
Böhm 2002 [20]	Open	Femoral + acetabular osteotomy	Yes	Varus derotation osteotomy	Yes	No	Salter	7%	79%	88.9%	6.8%	Severin Harris Merle d’Aubigné
Thielemann 2003 [21]	Open	Femoral + acetabular osteotomy	Yes	Varus derotation osteotomy	Yes	Yes	Pemberton	NA	94%	NA	0%	Tonnis Heine
Fujii 2004 [22]	Open	Soft tissue procedures	No	Na	No	Na	Na	7%	82%	79%	0%	Severinkalamchi
Angliss 2005[23]	Mixed	Femoral osteotomy	Yes	Varus derotation osteotomy	No	Na	Na	13%	60%	26%	14%	Severin
Thomas 2007 [24]	Open	Acetabular osteotomy	No	Na	Yes	No	Salter	5.7%	54%	52%	31%	Severin womac harris
Rampal 2008 [25]	Closed	Acetabular osteotomy	Yes	Na	Yes	No	Salter	2.1%	93.6%	93.6%	0%	Severin kalamchi merle d’aubigné
Remmel 2008 [26]	Open	Femoral osteotomy	Yes	Varus derotation osteotomy	No	Na	Na	4.6%	97%	62.8%	2.3%	Severin merle d’aubigné
Wu 2010 [27]	Open	Acetabular osteotomy	No	Na	Yes	No	Pemberton	Na	71%	86%	0%	Severin mckay kalamchi
Morin 2011 [28]	Closed	Acetabular osteotomy	No	Na	Yes	No	Salter	6.25%	95.2%	84.4%	4%	Severin harris womac
El-sayed 2015 [29]	Open	Femoral + acetabular osteotomy	Yes	Varus derotation osteotomy	Yes	Yes	Dega	Na	75.9%	72,50%	0%	Severin mckayharris merle d’aubigné
Miyake 2018 [30]	Open	Soft tissue procedures	No	Na	No	Na	Na	Na	94.1%	77.6%	0%	Severin kalamchi
Scott 2020 [31]	Open	Acetabular osteotomy	No	Na	Yes	No	Salter	Na	69%	69%	31%	Severin
	Closed	Spica cast	No	Na	No	Na	Na	Na	50%	50%	50%	Severin

Numerical variables are expressed as mean ± standard deviation and count data are expressed in terms of frequency. The data were manually created and imported on MedCalc Statistical Software version 20.009 (MedCalc Software Ltd, Ostend, Belgium) to obtain the Kaplan–Meier survival curves.

Results

According to the above retrieval strategy (Fig. 1), a total of 13 articles met the inclusion criteria [19–31]. These articles included a total of 800 DDH patients treated after walking age (988 hips) with an average follow-up of 27.9 years (range 10–67) and mean age at final follow-up of 27.8 years (range 16.1–47.6), the male: female ratio calculated on the available data was 1: 6. The mean age at surgery was 3.3 years (range 9 months–12 years). Table 2 summarizes data of included articles (Table 2).

Preoperative grading of pathological hip was heterogeneous, with Tönnis classification available only for a minority of studies [19, 20]. Based on the level of evidence assessment, the certainty of the evidence and the strength of the data were of grade IV for all 13 studies [19–31].

Hip Reduction and Traction

Ten studies (660/988 hips, 66.8%) reported on outcome after open reduction [19, 21–27, 30, 31], two studies (137/988, 13.9%) reported on outcome after closed reduction [20, 29, 30], and one study (191/988 hips, 19.3%) reported on both procedures, preoperative traction was performed in six studies [19, 23–25, 28, 30]. In particular, Morin et al. postulated preoperative traction reduced the risk of postoperative avascular necrosis (AVN), especially when associated with closed reduction [28].

Femoral and Acetabular Osteotomies

Acetabular osteotomy was performed in nine studies (five SIO, two PO, one DO, and one Mittelmeier osteotomy) and these were associated with femoral osteotomy in four studies. Femoral varus derotational osteotomy was performed in six studies and was the only surgical procedure in two of these. In two studies, only extensive soft procedures were performed [19–31]. The rate and timing of reoperation after the index procedure were variable in the different studies and could not be analysed in this review.

Avascular Necrosis

The mean rate of postoperative AVN of the femoral head was 6.6% (range, 2.1–13) and it was reported in eight studies [19, 20, 22–26, 28]. Wu et al. compared the hip survival rate between two groups of patients: AVN-present group and AVN-absent group. Outcomes were measured with use of the McKay and the Severin criteria; 96 and 76% of patients had a radiographically satisfactory outcome (Severin class I or II) in the AVN-absent and in the AVN-present group, respectively. According to the McKay criteria, there were significant clinical differences between the two groups of patients. Moreover, excessive inferior displacement of the femoral head following pelvic osteotomy led to a higher risk of AVN [27]. Bohm et al. also reported a slightly increased risk of AVN when open reduction and the SIO were performed simultaneously than when these procedures were staged [20].

Radiological Outcome

Radiological outcome according to the Severin score was reported in 11 studies (855/988 hips, 86.5%) (Fig. 2); in those studies Severin class II and III were considered as a failure. A progressive decrease in radiological hip survival after 15 and 30 years of follow-up was observed, and this finding was similar to the clinical survival analysis.

Fig. 2.

Severin radiological hip survival over time (years)

After multivariate analysis, the preoperative grade of hip dislocation was referred to as a risk factor for future hip failure by four studies [19, 20, 22, 23].

Total Hip Replacement

The rate of THR at last follow-up visit was reported in six studies, (111 THR/988 hips, 11.2%) [20, 23, 24, 26, 28, 31]. The rate of THR increased with the length of follow-up. The Kaplan–Meier survival curve showed a significant increase of THR procedures after 30 years of follow-up, with a steep increase after 40 years of follow-up (Fig. 3). In particular, no cases of THR were reported in studies with a mean follow-up to 23.5 years [19, 21, 22, 25, 27, 29, 30]. In four studies with a mean follow-up ranging between 30 and 40 years, the rate of THR ranged between 2.3 and 14% at the last follow-up visit (34 THR/332 hips; mean: 10.2%) [20, 23, 26, 28]. Two studies with a mean final follow-up of more than 40 years reported a THR rate between 31 and 50% [24, 31] with a significant increase in THR compared to studies with follow-up less than 40 years; in particular at final follow-up, there were 77 THR out of 216 hips (mean 35.6%).

Fig. 3.

Rate of total hip replacement over time (years)

Clinical Outcome

Clinical outcome was heterogeneous, with several different scoring systems reported: Harris, Merle d'Aubigné, McKay, WOMAC [19–31]. Due to such heterogeneity, survival was defined as excellent or good outcome at final follow-up with the specific clinical scoring or clinical evaluation reported. It was not possible to compare survival among different groups, due to the heterogeneity of index surgeries, preoperative traction, type of hip reduction, and outcomes scoring systems. Our analysis, spanning the available literature, included survival analysis with the Kaplan–Meier method and found an overall favourable clinical and radiological outcome for a mean follow-up of 20 years. Between 20 and 30 years of follow-up, the survival was roughly stable, while over 30 years of follow-up we observed a steep decrease in hips survival with a significant increase in the rate of THR. At 40 years or more of follow-up, the rate of THR increased further (Fig. 4). These long-term results are in line with the results of a few series we could not include in this review as they did not meet inclusion criteria [19–31]. This review aimed to determine long-term hip survival and THR rate in DDH patients treated after walking age, regardless of the index treatment and different factors that could contribute to those results at the final follow-up, as the number and type of reoperation between index surgery and last follow-up. Among different studies, different risk factors (age, severity of dislocation, bilateral involvement, postoperative inferior displacement, type of surgery, use of traction, staged procedures) were indicated to be responsible for future hip failure while other factors were found to be protective and correlated with a good or excellent outcome.

Fig. 4.

Overall clinical survival over time (years)

Risk Factors Associated with Increased Risk of Hip Failure

Older age at reduction increased the risk of degenerative arthritis [23] even though Angliss et al. found the classification of preoperative positioning of the femoral head has greater influence on the final outcome than the patient’s age at the time of surgery.

Bilateral involvement was a significant risk factor for failure [24]. In patients with bilateral disease, older age at reduction was associated with an increased hazard of THR after closed reduction than after open reduction and femoral osteotomy [31].

Excessive inferior displacement of the femoral head following pelvic osteotomy led to a higher risk of AVN [27]. Bohm et al. reported a slightly increased risk of AVN when open reduction and SIO were performed simultaneously than when these procedures were staged [20]. For Miyake et al., a cartilaginous acetabular index (CAI) at 5 years old was a predictor of unsatisfactory results, and future acetabular dysplasia was predicted with a CAI equal or above 10 degrees [30].

Factors that Increased Hip Survival

According to Hess et al., femoral and acetabular osteotomy tended to reduce the incidence of AVN [19]. In particular, El Sayed et al. reported femoral shortening facilitates reduction by decreasing the tension on the proximal femoral epiphysis [29].

Gradual reduction minimizes the vascular compromise to the femoral head, as described by Petit and Morel, and later confirmed by Rampal et al. [25]. Also, Morin et al. recommended progressive closed reduction by traction rather than open reduction [28]. For Remmel et al., a two-stage procedure seems to be a successful principle with open reduction followed by intertrochanteric osteotomy at an average of 6.7 weeks [26].

Discussion

The present study indicates the overall rate of good functional and radiological outcomes in patients with late detected DDH diagnosed after walking age is 94% at a mean of 14.5 years of follow-up, with no THR reported at 23.5 years of follow-up [21, 25]. However, data also highlight 10.2 and 35.6% of patients underwent THR at 30–40 years and more than 40 years after index procedure, respectively. At present, it has been widely accepted that open reduction is an option for children with late detected DDH diagnosed after walking age, even though our review found treatment options are heterogeneous. In particular, a significant reason for the great difference in the treatment of late detected DDH among different countries and regions, as well as between different surgeons, is the lack of evidence-based, reliable, feasible, and worldwide accepted treatment guidelines. Although some regional treatment guidelines exist, most guidelines are expert consensus formed by consulting the literature and expert questionnaire survey, which has certain significance in guiding the clinical process. However, it is well known that expert opinions are at the lowest level of evidence-based medicine, and their reliability in obtaining conclusions is far less than that of high-quality clinical randomized controlled studies. For example, some guidelines recommend one-stage open reduction with femoral and pelvic osteotomy correction in children aged 18–24 months of age with DDH. However, closed reduction or open reduction alone is mostly recommended in some paediatric orthopaedic textbooks [18, 32, 33]. In addition, the need for femoral osteotomy remains highly controversial when open reduction is performed [34]. As such, there are many doubts about the practical applicability of these treatment guidelines. Similarly, the level of evidence of the studies included in our review was low, being all studies level IV [19–31].

Results reported by previous works showed radiographic outcome (Severin class I and II) of children with late detected DDH managed by open reduction [11–14, 35–41] was not significantly different compared with those treated with closed reduction [25, 42–49]. Gibson et al. [36] reported 121 cases (147 hips) of DDH treated with open reduction. Age of surgery and age at last follow-up ranged from 12 to 36 months and from 16 to 31 years, respectively. In this series, 59% of the children had satisfactory radiological outcome. Terjesen et al. [50] retrospectively analysed 60 patients (74 hips) with late detected DDH treated with closed reduction, with an average age at initial treatment of 19.6 months (range 8–37) and an average follow-up of 58 years (range 55–60). However, although this work could not be included in our analysis as patients younger than 9 months of age at initial treatment were included, the results were satisfactory in 53% of the hips. According to these long-term studies it appears open reduction in children with late detected DDH does not significantly improve the clinical outcome in all patients. Similarly, the work of Kaneko et al. and of Malvitz et al. could not be included in our analysis as they assessed patients from 7 months and 1 month of age, respectively [48, 51].

Although open reduction in conjunction with pelvic osteotomy may reduce the incidence of residual acetabular dysplasia, many studies have shown it significantly increases the rate of AVN [52–54]. In particular, a slightly increased risk of AVN has been identified when open reduction and SIO are performed simultaneously than when these procedures are staged [20]. Currently, there have been numerous studies showing open reduction is a risk factor for AVN [52, 54]. Preliminary traction before closed or open reduction, can be attempted as to reduce the pressure on the proximal femoral epiphysis and the Petit-Morel technique seems to yield the best results [25]. In particular, in their meta-analysis, Wang et al. [55] included nine comparative studies evaluating the effect of closed and open reduction on the incidence of AVN and found the incidence of AVN was significantly higher in patients with open reduction than in those with closed reduction, which can ultimately affect the efficacy of treatment [52, 53] due to its association with osteoarthrosis [56]. In addition, children with late detected DDH undergoing open reduction can develop complications such as infection and joint stiffness. Today, although some surgeons are moving toward open reduction in patients older than 18 month of age, closed reduction and spica cast immobilization with the hips abducted and flexed remains a frequent treatment option in patients with DDH aged 6–24 months. The rate of dislocation and AVN following CR and cast immobilization varies between 1.7–17.4%, and 0–67%, respectively [1, 11, 27, 53, 54]. Since the pelvis is included in the cast, both hips cannot flex or extend. In case of hip over-abduction and flexion, the cast may induce uneven distribution of pressure between the surfaces of the acetabulum and of the femoral epiphysis, create vascular tension and cause AVN [1, 18, 27, 30].

Residual acetabular dysplasia is relatively common after closed reduction with an incidence between 15 and 32% [57]. However, residual acetabular dysplasia does not necessarily mean failure of treatment. Many studies have demonstrated extracapsular pelvic osteotomy (e.g. SIO, PO, DO) can correct the dysplasia to a normal configuration, and the chance is greater for the patient to have an excellent or good long-term outcome. In particular, the survival curve showed that most additional procedure became necessary more than 20 years after the index surgery [58, 59]. However, when the dysplastic acetabulum can be well corrected without the occurrence of AVN, good long-term results, lasting for up to 30 years, can be expected. Furthermore, our analysis highlighted hips treated after walking age start to deteriorate about 30 years after the initial treatment. However, given the limited number of studies with follow-up more than 45 years, it is not known how such hips would evolve four to five decades after open or closed reduction. In the general population, the incidence of THR is about 1% at an average age of 60 years [2, 60], while in our cohort of patients with late detected DDH the rate of THR is about 50% before the age of 50 years, and this percentage may be even higher in patient aged 60 years or more. Unfortunately, this information is not available as no study has sufficient long-term follow-up [19–31].

Age is another important factor affecting the outcome of children with late detected DDH. In particular, the older the children, the worse is the efficacy of treatment. Woźniak et al. assessed 67 hips treated when patients were younger than 5 years and 71 hips of patients older than 5 years at initial treatment. Interestingly, they found the risk of failure is related to the presence of bilateral DDH, the severity of AVN, and the age at which the treatment is initially carried out [61]. It appears the worse outcome in patients with late detected DDH should be attributed to the pathological changes themselves rather than to the treatment modality. Scott et al. compared the outcome between closed reduction and open reduction with SIO in children with late detected DDH older than 18 months at initial treatment that were followed up for at least 40 years. They found both treatment options were able to improve THR-free survival time, though neither one was capable to prevent osteoarthritis as well as the need for THR; Scott et al. hypothesized outcome may be related to the natural history of the disease rather than the type of treatment applied [31, 62]. As a general rule, Petit and Morel suggested 5 years of age as the cutoff beyond which dislocated hips should not be reduced [28]; our review could not find any strong evidence for patients older than 5 years of age. The best surgical indication should probably be adjusted to each patient and different parameters such as age at diagnosis, severity of dislocation and uni or bilateral involvement (bilateral cases have poorer outcome compared to unilateral cases) [11, 14, 28, 31, 46, 48, 50].

We encountered some limitations in the analysis of our results. The main limitation of our review is the heterogeneity of the data we reviewed. Each study presented numerous variables such as the age of the patients, initial conditions of the pathological hip, treatment carried out, duration of follow-up and outcome measurements detected. Additionally, all studies were retrospective and were influenced by operator-dependent assessment. To obviate such limitations, we have used stringent inclusion criteria such as age at initial treatment more than 9 months of age [16] and follow-up more than 10 years [19–31].

Despite these limitations, it appears pathological hips start to show signs of osteoarthritis about three decades after the initial procedure; 30–40 years after surgery there is a further deterioration and, although data are limited, the hip continues to worsen 40–50 years after the initial treatment. Although older age and higher preoperative Tönnis grade are risk factors for poor outcome, the restoration of normal anatomy protects hips from deteriorating, regardless of the modality of treatment.

Currently, there is no evidence that one treatment modality (open vs closed treatment, traction vs no traction, staged procedure or not, type of pelvic osteotomy, femur osteotomy vs no femur osteotomy) has been identified as optimal, as the outcome may be related to the natural history of the disease rather than the type of treatment applied [31]. Patients with late detected DDH treated after walking age have a life-long disease, with a permanent risk of developing osteoarthritis regardless of the treatment modality. These patients should be followed their whole life since the hip could continue to deteriorate even after the long- term follow-up data we have today.

Declarations

Conflict of interest

On behalf of all authors, the corresponding author states that there is no conflict of interest.

Ethical approval

This article does not contain any studies with human or animal subjects performed by the any of the authors.

Informed consent

For this type of study informed consent is not required.