The successful rates of CR in the treatment of DDH were inconsistent in the literature which ranged from 43% to 92% . In practical, if concentric, stable reduction of the hip cannot be achieved, OR procedure is an alternative for DDH. For efficacy evaluation of CR in the treatment of DDH, most studies only defined OR following CR as an endpoint of failure. However, we believed that hips with unsatisfactory radiographic outcomes such as residual acetabular dysplasia or subluxation in the long-term follow-ups who did not receive further intervention should be also take into consideration when determining the failure of CR. Because most of these cases should have been received further interventions in case of progressing into a high incidence of degenerative hip disease. But the fact is that not all these patients got further treatment because of their parents refused owing to the asymptomatic state till the latest follow-up . For instance, Yamada et al.  reported a satisfactory successful rate of 92% CR in treating DDH, but 48% of patients who were over six-years old at last follow-up showed residual subluxation. Consequently, the diversity failure rates of CR in the treatment DDH might be partially dependent on the different evaluating criteria for failure. The Severin classification is a widely acceptable assessing system for long-term results after DDH treatment, and the unsatisfactory Severin grades (grading III to VI) deemed to poor containment of acetabulum on femoral head which may be strongly indicated a poor prognosis that may leads to early degenerative joint disease even though these hips had not yet received surgical intervention. It is reported that 46% of the Severin III/IV hips had severe degenerative changes compared with only 3% of the Severin I/II hips . Altogether, the failure should be comprised of the OR cases following CR at early stage and cases with unsatisfactory Severin grades during follow-ups. In present study, we defined the failure of CR combined the cases which had undergone secondary OR (27 patients with 32 hips) and cases that showed unsatisfactory radiographic outcomes with Severin III or above grades (11patients with 11 hips) at last follow-up. In conclusion, the overall failure rate of DDH after CR treatment in present study was 31.16% (43/138). Furthermore, if the prognosis and the related risk factors of failure can be predicted at the time of initial CR, the parents can be informed regarding the outcome and future managements of the child. Many factors have been reported to as the risk factors for the failed CR including an older age, high dislocation grade or large AI and so on. It has been documented that age is an important prognostic factor in the treatment of DDH with CR . Terjesen et al.  reported that the most important independent risk factors for a poor long-term outcome was an age of eighteen months or older at the time of reduction. On the contrary, Barakat et al.  concluded that CR in children between 19 months of age to 36 months should be deliberately conducted by an experienced pediatric orthopaedic surgeon with ability to shift timely to open reduction once the complication occurred in the close follow-up. Unfortunately, a typical study from Bolland’s research reported that 83% of patients over the age of 18 months who underwent CR initially required further open procedures . Herein, we observed the similar outcomes that patients older than 18.35 months at the age of CR may have poor outcomes when compared with younger patients. Altogether, we concluded that using CR as a treatment regime for DDH among patients whose age over than 18 months might not be a reasonable choice.
The failure rate of CR in treating DDH was increased with the severe grading of the dislocation of hips . The higher dislocation grading correlating with increased risk of open reduction . In present study, our results also showed that the more severe dislocation of the DDH before treatment is significantly associated to the inferior outcomes after CR. The successful rate in IHDI grade II were significantly higher than grade III (P = 0.048) or grade IV (P = 0.002) respectively. Although there is no difference in successful rate between grade III and grade IV (P = 0.209), whereas the failed incidence in grade III was 35% (14/40) was also lower than that in grade IV 48.79% (20/41). We inferred that more included cases in future research might be demonstrate more predictable outcomes. Theoretically there are more soft tissues between femoral head and acetabulum in the affected hips with higher IHDI grade, and the pressure between the femoral head and the acetabulum was greater after the closed reduction that would be acted as the obstructs in the “docking” process subsequently result in a failed outcomes including the redislocation, sustained subluxation and/or insufficient acetabular remodeling . Actually, in present study, we employed the medial interval (MI) value attempting to determine the soft tissue obstruction between the acetabulum and femoral head after initial reduction. Our results showed that the satisfactory group demonstrated a less MI than unsatisfactory group, and we also constructed that MI more than 35.35 mm after CR immediately might be strongly indicated a poor outcomes. In present study, we included patients with with treatment history of Pavlik harness or abduction orthosis. Our results showed the orthosis treatment did not affected the CR results. However, this points should be further discussed in further studies, because the failure of orthosis treatment for DDH in infants may involve many variables, especially the compliance to the standard treatment regimen, and these patient-related variables lead to the bias outcomes .
Avascular necrosis (AVN) of femoral head is one of the most concerning complications following CR, which might be result in hip pain, limb-length discrepancy, abnormal gait and premature hip degenerative disease that affect hip function and need further interventions in adulthood . Previous studies reported a discrepant rates of AVN were ranged from 0–67% . Earlier studies have reported that various possible factors related to the AVN, including the age at the onset of treatment, genders, the utilization of prereduction traction , the severity of hip dislocation at treatment , laterality (unilateral/bilateral DDH) , absence of proximal femoral ossific nucleus , failed Pavlik harness treatment , or adductor tenotomy . However, either of these underlying factors was disputed, especially in recent evidence-based studies . Whilst these variations may be a consequence of natural variation due to the relatively small case numbers, different cases selection or the diversities in therapeutic regimes. As there is currently no consensus on whether the prereduction traction is beneficial and efficacy to help reduction, thus we abandoned this procedure before CR in all patients. In present study, our results showed that the AVN occurs in 6/72 (8.33%) of patients with satisfactory outcomes after CR. Furthermore, the occurrence of AVN was unaffected by gender, laterality, the age at CR, presence of the ossific nucleus, adductor tenotomy, seniority of orthopedists, prereduction AI or severity of dislocation. These results are similar to the results from a recent prospective, multicenter research . As AVN after CR was a multifactorial event, the high quality, prospective studies with large samples is still need to elucidate the precise risk factors associated to the AVN after DDH treatment.