Characteristics of adult gluteal fibrosis
GF appeared to be caused by multiple factors. Intramuscular injection of antibiotics, vitamins, antipyretic, benzyl alcohol, quinine or analgesics is the common etiology that leads to gluteal muscle contracture, which can cause direct muscle damage and subsequently the development of muscle fibrosis in the affected area[15,16]. Except for the injection history, GF is also related to other factors such as scar constitution, poliomyelitis, history of abscesses in buttocks, idiopathic GMC, cytokines (e.g., TGFβ-1, β-3, Smad4, and sphingosine-1-phosphate), immune function abnormalities, physical relationship, as well as children's susceptibility factors, character factors, gender factors and genetic factors [1,16-18]. 100% of our patients involved in this study had an injection history of penicillin dissolved with benzyl alcohol. However, the specific frequency and dosage of injection were unknown. This is probably because most of our patients received the injections in the 1980s and 1990s when they were teenagers and were not able to recall the history.
Children and teenagers were the most common patients of gluteal fibrosis or gluteal muscle contracture in the previous literature. Liu et al. [19] reported the use of radiofrequency vaporization under arthroscopy to treat injection gluteus contracture for 18 patients in 2003. The average age of their patients was 14 years old. Yuan et al. [20] compared the clinical efficiency of the treatment of gluteal muscle contracture with arthroscopic release and conventional open surgery based on 18 patients in 2006. The average age of their patients was 9.45 years old. The initial symptoms of gluteal muscle contracture mainly include changes in gait, such as splay foot walking and body shaking during running. Most of the patients cannot cross their legs actively, and/or have snapping hip. With the growth of age and the progress of disease, most of the patients with gluteal fibrosis would experience a certain degree of bone development abnormalities, such as pelvis tilt, pelvis dense belt, change of iliac bone walking direction, narrowing and lengthening of pelvis, and increase of femoral neck stem angle. At present, most of the patients with GMC were adults. In our clinical observation, we found that a part of adult patients also exhibited some other changes, such as pseudo leg length discrepancy and spinal compensatory scoliosis, resulting in abnormal gait. It was common for adult patients to worry about the effect of surgical treatment before operation. In our study, all patients were adults, with a minimum age of 18 years old and a maximum age of 42 years old (mean, 28.6 years). The follow-up results showed that all the patients achieved good results. The functions of internal rotation, adduction and internal rotation of hip joint were basically restored to normal after operation in 3 patients over 40 years old. Therefore, we supposed that arthroscopic release of gluteal fibrosis in adult patients could also achieve satisfactory results.
There are two main reasons why the diagnosis and treatment were made at such a late age. First of all, these symptoms did not affect the patients’ basic living condition when they were young. They could walk normally, and most of the patients could also run and even engage in vigorous sports such as basketball and football. Thus, neither the patients themselves nor their family realized that this might be a disease and they might need medical treatment. Second, most of these patients resided in remote villages in China when they were young, where the medical and healthcare conditions were relatively backward. Some of the patients did not receive correct diagnosis and proper treatment after seeing a doctor. Some of the patients were though diagnosed with gluteal muscle contracture long ago, but open surgery was generally executed in the past. Some other patients had concerns over the complications of operation and the large wounds left on the buttocks, and therefore were not willing to undergo surgery. Nowadays, it is possible to perform minimally invasive release surgery under arthroscopy, which has less trauma, quick recovery, small wounds, and fewer complications. Many of the patients who were afraid of open surgery before are now willing to accept minimally invasive surgery.
Classification of gluteal fibrosis
The location, range and depth of gluteal muscle contracture varied in different patients. In our clinical work, we found that the severity of contracture obviously affected the effect of arthroscopic release of gluteal fibrosis. Many scholars have studied and discussed the classification and graduation of gluteus contracture, but no widely accepted and applied typing method in clinical practice has been agreed yet. In 2009, Zhao et al. [13] summarized 172 cases of gluteal muscle contracture undergoing open surgery. The patients were classified into three levels (Level I=mild, Level II=moderate, Level III=severe) (typical type and special type) and three types, including Type MA (Gluteus maximus contraction type), Type MEI (Gluteus medius and minimus contraction type) and Type AGM (Gluteus maximus, medius and minimus contraction type). In 2012, Ye et al. [21] introduced a new minimally invasive method for surgical release in 1059 consecutive patients with gluteal muscle contracture. Their patients were assigned to 4 categories: type A, contracture occurred mainly in the iliotibial tract; type B, contracture occurred in the Iliotibial tract and gluteus maximus; type C1, movement of the contraction band was palpable accompanied by an audible snapping sound during squatting; and type C2, movement of the contraction band was not palpable or almost absent accompanied by an audible snapping sound during squatting. In 2013, Liu et al. [22] reported the treatment for a total of 358 patients with gluteus contracture. According to the clinical characteristics and intraoperative situation, their patients were classified into four groups: cable strip, fan-shaped, mixed, and tensor fasciae latae contracture. The aforementioned classification systems did not seem to differ significantly from each other and all these classification systems were practically reliable in understanding the disease pathology and useful in choosing the correct treatment options. The classification system proposed by Liu et al. [13] was based on all types of contractures at different levels with a focus on the functional and pathologic changes, and thus it was adopted to assess the degree of gluteal fibrosis in our study. We found that most of the moderate-severe GMC patients were tensor fascia lata combined with gluteal maximus muscle contracture or mixed type. For such cases, the extent and depth of the release were relatively high, and the difficulty of operation increased significantly. In the previous reports of arthroscopic GMC release, most of the contracture cases were not separated from moderate-severe contracture cases. Therefore, the effect of arthroscopic treatment in moderate-severe GMC patients was not clear. In this study, 118 patients with moderate-severe GMC were followed up and the results showed that arthroscopic tight fibrous band release also achieved good results for adult moderate-to-severe gluteal fibrosis.
Treatment of gluteal fibrosis
Non-operative treatments, including massage, physiotherapy, shortwave diathermy, and active and passive stretching exercises, were only suitable for mild cases, or recommended for patients who were not eligible for surgery or were waiting for surgery [23]. Once the contracture was established, non-surgical treatments would be useless [21,23]. Surgical treatment was the gold standard for all the established GMC cases. Some scholars claimed that open surgery was generally suitable for releasing various degrees and types of GMC, particularly for severe (grade III) GMC, while arthroscopic surgery was mainly suitable for mild GMC (grade II) with a relatively limited area [23-29]. In the meantime, some other scholars thought that the new minimally invasive open release could be considered in all cases of GMC [21]. When performing this operation, the surgeon must have comprehensive knowledge and skills of the anatomical signs and procedures, because the complete segmentation of the contracture belt is the main part of the operation [21]. Although this operation seems simple and easy to perform, the surgeon should keep in mind that it is a blind procedure and has full chances of complications [23].
The success of surgical treatment lies in two points: complete release of fibrous tissue and protection of normal tissue. Incomplete release will lead to residual symptoms and dissatisfaction, while excessive release will damage the stability of the hip joint, which may lead to Trendelenburg gait. Arthroscopic release was not recommended as the primary treatment for level III patients in previous studies. The reason might be related to the operative approach itself. In most literature, the two portals of arthroscopy were longitudinally arranged on the outside of the trochanter and buttock and were located at the proximal and distal side of the buttock. The line connecting the two portals was roughly parallel to the femoral shaft. A large area of subcutaneous tissue was dissociated in operation so as to provide enough operation space. An artificial working space (6 cm × 8 cm) was created in the interval between the subcutaneous fascia and the contracture bands [11]. For level III patients with severe contracture, the artificial working space needs to be larger enough to obtain better exposure of the contracture band. The wide range of subcutaneous separation during the operation can easily cause wound bleeding, which will affect the operative field and increase the operation time and the incidence of complications, such as postoperative wound hydrops and hematocele.
Our arthroscopic portals were different from those reported in previous literature. The two portals (AP and PP) were located in front of and behind the top of the greater trochanter respectively, and the position of PP was adjusted based on the position of the contractile bands determined before surgery and the expected release area. For instance, if the contractile bands were relatively posterior, the PP would be adjusted posteriorly along the horizontal line of the trochanter major. The sciatic nerve was at least 2 cm distal to the PP. The distance between the two approaches is about 10 cm. The line connecting the two portals was roughly perpendicular to the femoral shaft. An artificial working space (2 cm × 10 cm) was created. The tight fibrous band (fascia lata, iliotibial tract and gluteus maximus) around the trochanter major was sectioned from anterior to posterior by a radiofrequency device through the AP. Then, any residual deformities were evaluated carefully. The complete release of contracture was confirmed by flexion, adduction, internal rotation, palpable click, the Ober’s sign and the cross leg sign. In most cases, releasing tight fibrous band of the fascia lata and gluteus maximus were sufficient. If passive adduction was still limited after the release of the fascia lata and gluteus maximus, the contracture bands of gluteus medius would be explored and then released selectively. If passive adduction was still limited after the release of gluteus medius, the contracture bands of gluteus minimus and hip joint capsule would be explored and then released selectively. If passive adduction was still limited after the release of gluteus minimus and hip joint capsule, open treatment was needed. In case of uncontrolled bleeding or deep contracture that was not reachable with an arthroscopic instrument, a small incision should be used. In our study, 2 cases of type III gluteal fibrosis were excluded because the arthroscopic release was unsatisfactory and converted to an open surgery. After the selective release of the deep layer of gluteus medius, the contracture band of gluteus minimus and the partial contracture of the hip joint capsule, satisfactory results were obtained. In Zhao et al.’s study, operative treatment was performed in I to III level patients. All patients in the operative treatment group achieved excellent or fair results, but a number of complications were found in this group (only in level II and III patients), including scar swelling, hematomas, infections, and wound dehiscence[13]. In Amrit S’ study, the patients were classified into type I to IV according to the classification for the location of contraction of external snapping hip, and the excellence rate of surgery reached 100% in type I and type II and reached 92.7% in type III. No long-term postoperative complications were found and no infections, major swelling, hematomas, or wound dehiscence occurred in all of the four types of cases. All patients achieved good results after arthroscopic surgery [30]. For the 118 patients in our study, the GD scale was improved from 55.5±10.6 before operation to 90.1±5.2 at the last follow-up (p<0.05), and the satisfaction rate was up to 95.8%. Comparing our results with Zhao et al.’s and Amrit Set al.’s results, it can be found that patients with GMC can obtain positive outcome after both open and arthroscopic surgery, and can achieve not only complete contracture release but also a lower incidence of complications alongside a higher cosmetic satisfaction rate under the arthroscopic treatment.