Limited effects on patient outcomes of conjoint tendon release in anterolateral muscle-sparing total hip arthroplasty

Total hip arthroplasty (THA) provides pain relief and restoration of functionality in osteoarthritic hips and leads to the improvement of the patients’ quality of life [1]. Recent studies have demonstrated that THA with the minimally invasive techniques is effective in reducing soft tissue damage, peri-operative pain, blood loss, and achieves faster recovery times and shorter hospitalization periods [2‐5]. The anterior-based muscle-sparing (ABMS) THA is less popular than the direct anterior approach (DAA) applied to the hip, clinically effective, safe, very low-dislocation rate and, the excellent control of the leg length discrepancy [6]. Minimally invasive total hip replacement via the anterolateral approach in the supine position has yielded excellent clinical results [6, 7].

The conjoined external rotators tendon (CERT; conjoined tendons of the gemellus superior, obturator internus and, gemellus inferior) have been proposed as the key active stabilizers of the hip, along with the internally rotating gluteus minimus, are often described as the “rotator cuff” of the hip [8]. The CERT can help the external rotation of the hip and the stabilization of the hip joint [9] and is important to prevent posterior dislocation of the hip after THA [10, 11]. Although the anterior or anterolateral minimally invasive technique is reported as muscle-sparing THA, the surgeons sometimes need additional soft tissue release for canal preparation of the femur, such as the capsular ligament or the insertion of muscles around the hip including the CERT. These releases enable good canal preparation, insertion of broaches, and trials.

Currently, the clinical effects of CERT release in DAA THA are being investigated, only a few studies have described the effects of CERT release in DAA THA [12]. Consequently, the effects of CERT release in ABMS THA remain unresolved. The purpose of this study was to evaluate the clinical and radiographic results of CERT release in the anterolateral muscle-sparing THA approach.

This was a single-center, retrospective data collection study, approved by our hospital’s institutional review board. The study considered for inclusion 92 hips from 92 patients who underwent unilateral primary THA with ABMS in the supine position at our institution between July 2017 and August 2018. The indication of THA with ABMS in this study was 1) no history of previous surgery on the affected hip, 2) primary osteoarthritis of the hip and osteonecrosis, and 3) secondary osteoarthritis of the hip with Crowe classification from 1 to 3. All hips were examined clinically and radiographically at 6 months and 1 year after surgery. Of these, three hips with opposite THA, two hips with total knee arthroplasty, and one hip with the tibial fracture within 12 months, and one hip with the onset of cerebral infarction on the perioperative period were excluded from the analysis. Finally, in total, 85 hips from 85 patients were included in this study. The baseline demographic data are listed in Table 1, including the age, gender, body mass index (BMI), follow-up periods, and diagnosis.

Among all the included hips, 37 hips (43.5%) retained a CERT (Non-released group) and 48 hips (56.5%) included a CERT release (Released group). The mean ages and BMI were not significantly different (Table 2). The duration of surgery was significantly longer and the amount of blood loss was more extensive in the released group (Table 2). The four hips needed an additional release of the external obturator muscle attachment.

The JOA hip score, all JHEQ scores, and PCS score of the SF-36 were significantly improved without the MCS score (Table 3). There were no significant differences in the JOA hip score, JHEQ, FJF-12, and SF-36 between the non-released group and released groups (Tables 4, 5).

There were significant differences in the stem flexion angle between the groups. The pre-operative and post-operative leg length discrepancies, extension lengths, cup inclinations, anteversions, and stem varus angles were not significant (Table 6).

The clinical results of this study had excellent post-operative outcomes for THA, given that the hip clinical scores were significantly improved (with the exception of the mental section of SF-36), regardless of the CERT release. Furthermore, the leg length discrepancy was improved. The ABMS for THA with preserved CERT was less invasive, the duration of surgery was significantly longer, and the amount of blood loss was more extensive in the released group.

Although the deep external rotators have been proposed as key active stabilizers of the hip described as the ‘rotator cuff’ of the hip [8], the CERT release during the DAA in THA has no impact on gait or patient-reported outcomes within the 12-week period post-surgery [23]. Yao et al. investigated the outcome of DAA for THA and showed that the CERT release was not predictive of the length of stay, one year inpatient pain medication requirements, or outcome scores [12]. These results were consistent with the results of the present study, wherein 56.5% of the hips included a conjoint tendon release irrespective of whether the CERT released or not in anterolateral muscle-sparing THA, and had limited effects on patient outcomes at six months and one-year based on the use of CROMs and PROMs. Specifically, FJS was considered as the most sensitive score and the ultimate goal in joint arthroplasty [17] and tended to show greater dispersion and more marked differences between patients [18].

The importance of preservation of CERT for post-operative hip stability has been described in the case of the posterior approach [24], to prevent posterior dislocation of the hip after THA [10, 11]. The data of the present study showed no dislocation, irrespective of whether the attachment of CERT (released or not) affected the post-operative stability in ABMS for THA. Conversely, if the CERT was released in the non-released group cases, joint instability may be increased.

A cadaveric study has shown considerable variation in the attachments of CERT [13]. The CERT insertion is mostly attached to the anterior part of the femur among the short external rotation muscles [25]. The damage of the obturator internal muscle depended upon the injury at CERT insertions [26] and induced a decrease of the muscle volume [9]. In the surgical steps of the present study, the CERT was released at the attachment site from the anterior to the posterior directions in the case of inadequate canal preparations. The CERT was related to the adjustment of the range of motion and the joint stability, the present study showed the limited effect on patient-reported outcomes, including pain and daily-life activities.

The radiographic evaluation showed that stem sagittal alignment was more flexion in the non-released group, while no significant differences existed in the stem valgus/varus alignment. The stem tended to be inserted from the anterior part of the osteotomy site of the femur if the canal preparation was inadequate. Yoshitani et al. reported that there were no significant differences in clinical or radiographic outcomes between flexion and neutral alignment of the tapered-wedge stem at an average follow-up period of 4.7 years [27]. The difference in the sagittal alignment of our data had a limited effect on the clinical course.

Our study has certain limitations. The first limitation is that this study was a nonrandomized control study. The CERT release depended on the cases, and hips with contracture may have tended to release the capsular band and CERT. The second limitation was stem selection. The choice of the stem was according to the shape of the femur, anteversion, and offset; the differences in the shape of the stem shoulder or stem length affected the capsular band and CERT release, and femur canal preparation. The third limitation pertains to the fact that our study did not include shorter-term outcomes. We may detect short-term clinical differences if shorter-term outcomes were examined.

The CERT release in anterolateral muscle-sparing total hip arthroplasty has a limited effect on postoperative clinical outcomes. Although the CERT release improves femur exposure, the CERT has limited effects on patient outcomes at 6 months and one year post-surgery. The CERT release was more invasive than the preserved CERT, and the thresholds levels may be low. Surgeons should try to leave the CERT for patients with a wide range of motion as much as possible, and release the CERT for inadequate femur canal preparation.