In the present study, we presented the mid-term follow-up data focused on the bone union status at the site of the femoral osteotomy in patients who underwent cemented stem THA with subtrochanteric femoral shortening transverse osteotomy for Crowe IV DDH. Our results indicate that this procedure is safe and effective based on clinical and radiological outcomes.
Performing THA in patients with Crowe IV DDH is an effective but difficult procedure. To improve the prosthesis survival rate and reduce limb length discrepancies, proper positioning of the acetabular component within the real acetabulum is necessary [25]. However, this procedure may cause the pulling-down distance of the hip center to exceed 40 mm, which can jeopardize neurovascular structures [5, 6]. Subtrochanteric femoral shortening and derotational osteotomy were introduced to address this limitation, but they were also associated with a higher risk for other complications, particularly nonunion at the osteotomy site. Nonunion of the osteotomy site may lead to varus angulation, pain, loss of rotational stability, and prosthetic loosening [7, 26]. The proper union of the osteotomy site depends on several factors, such as the stability of the osteotomy site [27], size of the contact area between the proximal and distal segments [24], available bone stock, and degree of vascularization. Among these factors, the stability of the osteotomy site is the most important and is contingent on the osteotomy type, stem fixation method, stem morphology, and use of additional fixation materials such as cables or plate and screws [26, 28]. However, the impact on osteotomy site stability associated with the osteotomy type and stem fixation method remains contested in the literature.
As a technique, subtrochanteric osteotomy is very adaptable and can be performed with transverse, oblique, double-chevron, and Z-shaped osteotomy shapes [9–12]. Transverse osteotomy reportedly has lower rotational stability than other techniques [7, 27]. However, Muratli et al. [27] conducted a biomechanical study that compared the four osteotomy shapes and demonstrated that no single inherent feature increased osteotomy site stability. In addition, a comprehensive meta-analysis by Li et al. [18] reported no significant differences in complications and survival rates according to the four cutting shapes. However, transverse osteotomy is the preferred approach because it is a relatively simple technique that allows surgeons to adjust the anteversion angle with minimal damage to the periosteum at the osteotomy site [18, 19].
Many surgeons also prefer performing subtrochanteric osteotomy with an uncemented femoral stem [9–12] because cemented stems are prone to failure from cracking or fatigue owing to a narrow cement mantle [29]. Aseptic loosening has also been reported in young and middle-aged patients who received cemented stems [30]. Moreover, cemented stems perform worse in revision surgeries because of the inadequate residual bone stock [31]. They are also associated with a risk for cement leakage into the osteotomy gap [32], which can lead to nonunion of the osteotomy site. Some studies have reported that cemented THA with subtrochanteric osteotomy is an effective treatment for severely dislocated hips [8, 13–17]. However, the nonunion rate with cemented prostheses is 0–20% [8, 13–17] compared with 0–12.5% with uncemented prostheses [1, 33, 34]. The nonunion rate was similar between cemented and cementless stems. Recent improvements in prosthesis materials, product manufacturing processes, and cementing techniques have led to better results with cemented stems [35]. In the clinical setting, the main concern is that the cement will leak into the osteotomy site, leading to nonunion. However, Kawai et al. [36] reported that cemented THA with subtrochanteric transverse osteotomy provides satisfactory short-term results with no major complications in patients with Crowe IV DDH. Sound union of the osteotomy site was noted in all patient hips in the present study despite some patients having cement in the osteotomy gap. This result is encouraging for surgeons who prefer to use cemented stems with subtrochanteric transverse osteotomy for patients with Crowe IV DDH.
When there is potential fragility in the femur, cemented stems are preferred over uncemented ones because they provide more rotational stability [37]. It is important to note that the surgical procedure itself weakens the femur, especially during the femoral trial step, and cementless stems are associated with a higher risk for intraoperative fractures with an incidence of 0–22% [28, 33, 34]. Femoral fractures decrease the stability of the standard-length stem, as a result a longer stem with or without additional fixation, such as a cable or plate, is needed. In comparison, intraoperative fractures rarely occur with cemented stems [8, 13–17]. Should they occur, the cemented stem itself provides good fixation.
To promote union, one study recommends an optimal osteotomy level that depends on the length of the resected femur [24]. Furthermore, the osteotomy ends should be smooth [38] and parallel [14] to maximize the surface contact between the bones. Periosteal stripping should also be kept to a minimum [38]. To reduce the likelihood of cement leakage into the osteotomy gap, we drilled three to five 3-mm holes near the osteotomy sites of the proximal and distal cortical bones for patients included in the present study. The holes allowed the excess cement to leak out rather than deposit into the osteotomy gap during cementing. Another reason was to increase the stability of the cement mantle by the cement column. In addition, it was important to hold both femoral fragments using bone clamps during the process to avoid collecting cement leakage in the osteotomy site.
There is evidence that the stem bridging the osteotomy site is vital for achieving stability, but the recommended length varies in each study. Ozan et al. [39] proposed that the femoral stem should pass the osteotomy site by at least 4–5 cm, whereas Yang et al. [40] suggested that at least 3 cm was sufficient. Meanwhile, Kawai et al. [15] proposed that surgeons should insert the standard-length stem to a depth of at least 7 cm below the osteotomy site. This allows the stem to function as an intramedullary nail, providing stability. Charity et al. [14] concluded that the stem should bypass the osteotomy site by at least twice the diameter of the diaphysis, especially in patients who require more robust fixation, such as in young men. In this study, the average stem length used to bridge the osteotomy site was 97.6 mm (range, 76.5–106.2 mm), consistent with the data presented in the studies mentioned.
While all the patients in the present study demonstrated significant improvement after surgery, some postoperative complications were also noted. Two patients experienced hip dislocation within 3 months after surgery (Fig. 2). This represents a dislocation rate of 14.3% (2/14), which corresponds with the published dislocation rate of 3.8–15% in patients with Crowe IV DDH who undergo subtrochanteric osteotomy [1, 7, 8, 13, 16, 33]. Both patients underwent closed reduction, were required to wear a brace, and were put on bed rest for 4 weeks. No dislocations recurred throughout the study. Moreover, while all the osteotomy sites demonstrated good union, one patient showed delayed healing at 18 months postoperatively (Fig. 1). Analysis of her postoperative radiographs revealed a linear hyperdense shadow in the osteotomy gap, which persisted throughout follow-up. Since no bone growth was observed within the shadow, we suspected that the shadow was cement within the osteotomy gap. The piece of cement prevented bone healing and promoted bone resorption, which was represented by the presence of a low-density shadow within the cortical bone around the osteotomy site. In the serial radiographs of this patient, new bone formation only occurred at the cortical bone periphery, which prolonged bone union time. No other complications occurred during the study period.
We also analyzed the correlation between osteotomy union time and several other factors, such as the length of the femoral resection, length of limb lengthening, distance of the osteotomy site from the lesser trochanter, and length of the stem to bridge the osteotomy site. The results suggest that there is no statistically significant correlation among these factors. Our data are similar to those presented by Akiyama et al. [16]. In contrast, Kawai et al. [36] demonstrated a significant correlation between the length of bone resection and the incidence of delayed union. However, Kawai et al. [36] divided the resected femoral fragments into two or three longitudinal segments and fixed these to the osteotomy site as autologous bone plates. This difference in technique may be the reason behind the differing results in the two studies.
This study had some limitations. First, there was no control group owing to the retrospective study design. Second, the number of enrolled patients is relatively small, with data obtained from only 14 individuals. However, Crowe IV DDH is relatively uncommon, so obtaining a larger sample from a single institution may be difficult [13]. Third, our follow-up period was limited to a maximum of 70 months. Continued follow-up past 70 months may be beneficial to establish more long-term results associated with this procedure.