In the current study, we obtained satisfactory functional and radiological outcomes in most of the patients (regardless of their age) by managing DDH with one stage procedure consisting of OR and DPO, which was performed in all patients; however, in patients above 30 months of age, the need for a concomitant FO was significantly higher. Furthermore, the older patients group needed longer operative time, and their incidence of complications was higher.

Certain anatomical changes of the acetabulum and the femoral head would occur if the femoral head remained dislocated outside the acetabulum; however, these changes could be reversible if the hip was reduced as soon as possible, but the exact upper age at which hip reduction will result in normal acetabular development is uncertain [7, 9]. Many authors suggested that a hip reduced by the age of 4 years could achieve satisfactory acetabular development and growth, and some even extended it up to 8 years of age [14, 25].

Determining the cutoff age to consider a DDH as a late presentation or not is controversial [7, 26]. The debate regarding the optimum age at management could be attributed to the potency of the acetabulum to grow and remodel; the lower limit for acetabular remodeling is 18 months, whereas the upper limit is up to 11 years of age [27, 28]. In the current study, for a single-stage procedure, we included patients below the age of 8 years; they were further divided into two groups, above and below 30 months of age, as was performed in the study by Ning et al. [12].

Correcting the acetabular anatomical characteristics is considered the primary goal of management options, either stimulating the normal acetabulum development by closed reduction of the femoral head or by surgical pelvic osteotomies aiming to improve the acetabulum coverage [29, 30].

Closed reduction in considerably older children with or without adductor tenotomy was reported to offer acceptable results, as long as there is a close follow-up [8]; however, some studies showed that the majority of patients who had successful closed reduction after the age of 18 months required an additional open procedure [31, 32]. This is why most surgeons consider OR for children older than 12–18 months or who failed to achieve a concentric hip reduction by closed maneuvers [9].

Open reduction alone or combined with other procedures such as PO or FO varied in previous studies; in a study by Castañeda et al. [6], including 712 hips with DDH in patients with a mean age of 2.1 years (1–6.5 years), the authors performed OR only in 91 (12.8%) hips—621 (87.2%) hips needed a concomitant PO and a shortening FO was performed in 221 (31%) hips (where all patients were above four years of age). In a study by Charki et al. [33], including 414 hips with DDH, patients had a mean age of 34.6 months (14–96 months), and the type of interventions were as follows: OR only in 18%, OR + FO in 32%, OR + PO in 8%, and the majority had OR + PO + FO (42%); they recommended the addition of a FO in cases with high dislocation (Tönnis 3 or 4) and in cases over 18 months old, while a PO (regardless of sits type) should be added for children older than 36 months and if the AI was > 25, even if the age was as less as 18 months.

On the other hand, if the open reduction was determined, some surgeons advised a whole job single stage surgical procedure, consisting of OR (and capsulorrhaphy), PO, and a FO (if needed) [5, 9, 12]. A single-stage procedure avoids repeated staged surgical intervention with an acceptable outcome. In the current study, all patients underwent OR + DPO, but FO was needed only for patients in group II.

Various pelvic osteotomies were described to obtain previously mentioned goals, the selection of which osteotomy depends on reduction concentricity, patient age, and the status of triradiate cartilage [7, 11]. In the current series, we used DPO exclusively in all patients, which was the same osteotomy performed in 52 hips in the study by Czubak et al. in patients having a mean age of 3.9 years (1.2 to 12.8) [5]. The DPO osteotomy combines the advantages of Salter osteotomy (reorientation of the acetabulum) and Pemberton osteotomy (reshaping of the acetabulum) [34]; furthermore, either anterior or lateral coverage can be obtained by adjusting the osteotomy inner cortical cut [35, 36]. No graft fixation at the osteotomy site was needed owing to the inherent osteotomy site recoil produced by the intact sciatic notch and the posteromedial cortical hinge, and no complications related to fixation devices or the need to be removed [5, 37].

Femoral osteotomy (shortening, derotation, or both) is usually needed if excessive force is required to reduce the hip joint [32]. In the current study, we needed a concomitant FO in 69.4% of patients in group II; however, this was not performed in any patients in group I. Czubak et al. reported performing FO in all hips. However, in a study by Ning et al. [12], who performed a single-stage surgical intervention (using various types of PO) on 864 patients with DDH > 18 months of age(mean age of 5.8 years), they reported performing shortening FO in all hips; however, a derotation was added in selected patients according to the hip anteversion degree measured in preoperative computed tomography (CT) scan. To evaluate factors predicting the need for FO, Sankar et al. [38] evaluated 72 hips with DDH in patients having a mean age of 35.6; they concluded that the patients over the age of 36 months and patients with vertical displacement greater than 30% of the width of the pelvis were more likely to require shortening FO.

Regarding the functional outcomes after single-stage procedures, Ning et al. [12] divided their patients into three groups according to age at surgery: group I: less than 2.5 years; group ii: 2.5–8 years; and group III: > 8 years. After a mean follow-up of 6.2 years (3.2 to 8.9), they reported good or excellent (satisfactory) functional outcomes in a total of 79.4% of patients, with poorer results in group III, and no difference was noted between patients in groups I and II. The same previous observation was concluded from our study, where satisfactory functional outcomes were reported in 94.3% and 80.6% in group I and group II, respectively (all patients were under eight years), with no difference between both groups. Furthermore, after a mean follow up of 4 years (3–9 years), Czubak et al. reported Severin grade I or II (satisfactory) functional outcome in 78.8% of their patients, and no difference was found in patients aged below (group A) or above (group B) 3 years [5]. A systematic review by Wozniak et al. [10] included 23 studies on the outcomes after DPO in DDH. The functional outcomes were reported on 512 hips and were graded as good or very good in 84.8% of the hips by the last follow-up.

Ning et al. reported good or excellent radiological outcomes per Severin classification in 84.7% of their patients [12]. In the current series, most patients reported satisfactory radiographic outcomes according to Severin radiological grades (94.3% in group I versus 83.3% in group II), with no difference between groups. Furthermore, we obtained improvement in the AI measured at the final follow-up in all patients included in the current study compared with the preoperative values (from a mean of 37° to 27.2°, p < 0.05) with no difference between both groups. This was consistent with results from the study by Czubak et al., where the AI was improved in all patients compared with the preoperative values (from a mean of 38.8° to 19.5° for group A and from 39.6° to 21.3° for group B), and there was no difference between the AI measured at the last follow-up between both groups. The improvement of AI was reported in 19 studies (636 hips) in the systemic review by Wozniak et al. [10]; it was reduced to ≤ 20° as reported by 16 studies, and the difference between the mean pre- and postoperative AI was 22.5°. Furthermore, the radiographic evaluation according to Severin classification was grade I and II in 81.7% of 410 hips. At a mean follow-up of 9.3 years (6–4 years), Castañeda et al. [6] reported good radiographic outcomes classified as grade I or II per Severin criteria in 80% of the patients; however, they reported that better radiographic outcomes were obtained in patients who underwent open reduction and pelvic osteotomies.

The effect of performing FO on the outcomes was evaluated in a study by Castañeda et al., where the authors evaluated 645 patients with late presenting DDH; 328 hips received a FO compared with 317 hips that did not. The authors reported slightly better functional outcomes in hips which did not have a FO (measured according to Iowa Hip Score); however, they reported no difference in Severin radiographic scores or the incidence of AVN; furthermore, they concluded that FO is not necessarily related to better outcomes [39]. In the current study, we reported that same finding, as there were no significant differences in the functional, radiological, and complications incidence outcomes between patients within group II who had FO and those who did not.

Avascular necrosis (AVN) of the femoral head is one of the devastating complications that could occur after DDH surgical procedures, which varied among studies with an incidence reaching up to 48% [6, 12, 40]. In the current study, we reported an AVN incidence of 5.6%, which occurred exclusively in group II. A similar incidence (5.8%) was reported in the study by Czubak et al. [5], while Ning et al. reported an incidence of 27.4% and reported them as poor outcomes according to the Kalamchi and MacEwen classification [12]. Wozniak et al. [10] reported an AVN incidence of 18.9% from 19 studies (856 hips). The role of PO or FO on the incidence of AVN was controversial; in a systematic review by Kothari et al. [40], evaluating studies on DDH management after walking age, they concluded that the best results (functional, radiological, and lowest AVN incidence) were demonstrated when OR was combined with a PO; they found no evidence to support improvement in outcomes after adding a FO. Charki et al. [33] reported that the radiological results were better in patients where a PO was added; however, contrary to the previous systematic review, they reported that adding a FO contributes to obtaining better functional outcomes and decreasing the incidence of AVN. Furthermore, some technical considerations were suggested to avoid AVN development, including adequate adductor tenotomy, iliopsoas recession, extensive medial capsular release, and division of the transverse acetabular ligament [41, 42].

The current study has some limitations. First, this was a nonrandomized cohort study with relatively few patients. Second, the wide age range of patients included in group II could affect the outcomes. Third, we did not conduct a correlation analysis to determine factors affecting the incidence of complications, especially AVN. Last, a longer follow-up is needed to demonstrate the effectiveness of the surgical technique utilized in the current study.