The reconstruction of the acetabulum in patients with Crowe II-III DDH is a demanding procedure for orthopedic surgeons. Most surgeons found it technically difficult to achieve acceptable cup coverage at the anatomical acetabulum on account of superolateral bone deficiency[14]. Therefore, femoral head structural autograft was usually utilized at the superolateral rim to provide additional support[15]. However, other authors have proposed the instability of cemented acetabular component with bulk bone grafts[16]. Though some excellent results were reported in cementless THA with autograft[17, 18], this procedure still could be correlated with longer duration of surgery and increased blood loss.
Because the posterosuperior bone above the native acetabulum is almost intact, the acetabular cup can be placed at high hip center to optimize host bone-implant contact[19]. In this study, we sought to show the utility of HHC technique used in patients with Crowe II-III DDH and evaluated the clinical and radiographic results between different heights of hip center.
Early results have shown superior placement and especially lateralization of the cemented acetabular cup resulted in high rate of loosening[7]. In addition, in the cementless THA, aseptic loosening also occurred in long-term follow-up. Watts et al.[20] reviewed 88 primary cementless THA at a mean follow-up of 10 years and found a higher incidence of aseptic loosening and cup revision with superolateral placement of the cup, which was described as more than 10 mm superior and 10 mm lateral to the approximate femoral head center. To avoid this situation, the acetabular component was placed medially adjacent to medial wall during operation in our study. Medialization not only prevented the increase of gravity level arm and joint reaction force, but biomechanically relieved the burden of abductor muscle which was mostly malfunction due to chronically shortened condition and subsequent atrophy. In our study, the mean horizontal distance of the center of rotation which was 30.0 mm in group A and 31.4 mm in group B was comparable to the results described by Flecher et al.[21] (horizontal distance was 30.4 mm when vertical distance was 23.4 mm), Fukui et al.[10] (horizontal distance was 28.9 mm when vertical distance was 28 mm) and Galea et al.[11] (horizontal distance was 31.6 mm when vertical distance was 30.9 mm). However, referring to the anatomical center, only 73 (85.9%) acetabular cups attained the objective of medialization or lateralization less than 10 mm. Lateral cup placement more than 10 mm in group B significantly exceeded that of group A. Obviously, it’s more difficult for medialization when the rotation of the hip was elevated increasingly higher. A possible explanation may be the funnel-shaped geometry of the bony pelvis. Nevertheless, it should be stated that no complications such as loosening and liner wear occurred in our hips with excessive lateralization. Different from previous studies utilizing polyethylene liner in most of cases, COC interface was applied in 91.8% of our cases. Therefore, we considered that the incidence of wear could be reduced by choosing COC interface of which the favorable wear features may efficaciously counteract the excessive joint reaction forces related to lateralization.
Some authors indicated that there is a negative correlation of abductor strength with a high rotation center of the hip. Through a radiological and biomechanical study, Abolghasemian et al.[22] suggested that elevated hip center resulted in a decrease in the muscle length and a corresponding decrease in the preload, leading to the weakness of abductor strength. But in a recent study, Traina et al.[23] demonstrated that restoration of optimal femoral offset and abductor lever arm produced satisfactory results even for a center of hip rotation of > 30 mm. This paper came up with a similar result as well. Though the height of hip center in group B (33.1 ± 4.8) significantly exceeded that in group A (25.1 ± 1.6), the clinical and radiographic outcomes were all excellent after restoration of leg length, FO and ALA, and no significant difference was shown in two groups. In spite of the slack of gluteus medius due to elevated hip center, a larger size stem and appropriate head/neck lengths could be applied as a compensation and could also contribute to correcting leg length discrepancy, avoiding lower limbs of claudication. Further, preserving the continuity of abductors meant a favorable event regarding the restoration of normal gait. In our series, only 8.2% of all hips presented with a positive Trendelenburg sign and 8.7% of patients presented with a limp. The result of Trendelenburg sign was superior to the cases described by Chen et al.[19] (14.2%) and Fukui et al.[10] (13%). Furthermore, a recent gait analysis study by Karaismailoglu et al.[24] claimed that the bilateral HHC technique could provide similar gait characteristics as anatomical reconstruction. Although there was no detailed research on this issue, the rates of limp in unilateral and bilateral HHC were similar (P = 0.912) and low in our cohort, respectively 7.5% and 6.3%.
In our series, the 8-year survival rates of implant were high, respectively 96.7% (95%CI, 90.5%-100%) in group A and 96.2% (95%CI, 89.0%-100%) in group B. Comparison of our survivorships with other studies showed that the HHC technique was a reliable alternative method for Crowe II-III DDH[8, 9, 25]. Meanwhile, higher hip center won’t significantly reduce the survivorship of implants even if it was above 28 mm.
This study shows some limitations. First, our conclusion is based on a relatively small sample size. In addition, the validation of HHC technique needs a longer follow-up, and the consistency associated with a study setting limited to a single surgeon’s practice in single center. Second, this is a retrospective study. However, our patients were identified from a consecutive series with DDH, which may reduce the possibility of selection bias. Third, there is a lack of comparison between HHC technique and other methods. Fourth, the issue of gait in this study is insufficient compared with other gait analysis study. However, we believe that our satisfying results can provide effective supports to HHC technique and can be considered as an important reference to future research.