Subjects
From March 2018 to March 2019, patients who suffered from femoral neck fracture were randomly divided into 2 groups according to the time of admission. Patients with singular admission were involved in group A, and patients with even days were involved in group B. A supervisor nurse who did not know the surgery situation assigned patients to groups. The same group of surgeons, who had implanted more than 300 prostheses using this approach, performed all operations. Inclusion criteria: femoral neck fractures (within a week), uncemented prosthesis, THA via DAA. Exclusion criteria: advanced age (older than 75 years), hepatic or renal failure, coagulation abnormalities, thrombosis, myocardial infarction, cerebrovascular accidents (within the previous 6 months), long-term use of anticoagulants, long operation time (more than 90 min), unexpected intraoperative condition (intraoperative fracture etc.), unable to undergo an MRI (claustrophobia etc.) and could not be evaluated effectively by MRI. In the end, 56 of 76 patients were included in the study and 20 were excluded. (Fig.1) The surgical team, consisting of three surgeons and two instrument nurses, was responsible for the surgical treatment of the two groups of patients. In Group A, the TFLM was covered by the anterior capsular pad which was rotated to cover the surface of the TFLM, while in group B, the surgeons were reminded to try their best to remember to minimize tension on, traction of the TFLM during the operation. The TFLM had no substantial protection. The approval for this study was obtained from the Institutional Review Boards of the first authors’ affiliated institutions.
Surgical technique
All patients underwent operation under general anesthesia within 5 days of femoral neck fracture. During anesthesia, the same proportion of muscle relaxants were given according to body weight. Intravenous tranexamic acid (1 g) was administered to patients in each group. All patients also received autologous blood transfusion whenever possible during the process. All operations were performed by the same group of surgeons.
The patients were placed in a supine position using a regular table. A typical incision was made over the medial margin of the TFLM. After the location of the TFLM was confirmed, the overlying fascia was incised. After identifying the space between the TFLM and rectus femoris muscle, the TFLM was retracted laterally. After transecting the ascending branch of the lateral circumflex femoral artery, the anterior capsule was exposed satisfactorily.
On the medial side of the capsule, the iliocapsularis muscle lies posterior to the rectus femoris muscle and is attached to the anterior capsule. The iliocapsularis is an elongated muscle that is attached to the full length of the anterior capsule. Though rarely reported, some surgeons have suggested that the iliocapsularis muscle stabilizes the femoral head by contracting within the dysplastic acetabulum. [19, 27, 28] The iliocapsularis muscle serves as an anatomic marker separating the joint capsule of the hip from the hazardous area posteriorly, which contains the arteries and nerves that can be injured by operating on the superomedial aspect of the iliocapsularis muscle.
Keeping the iliocapsularis muscle as the medial boundary of the joint capsule, we dissected along the lateral side of the iliocapsularis, being careful to detach the muscle from the joint capsule to expose the maximum possible area of the anterior articular capsule. After incising the articular capsule, the now separated anterior capsule was folded outward as a layer of soft tissue acting as a “pad” covering the outer musculature and sutured to the skin at the edge of the incision with silk suture material. With this maneuver, the retractor and the handle of the reamer were separated by the “capsular pad” from direct contact with the TFLM, while still allowing the exposure of the hip joint for the THA. The specific steps are described in (Fig. 2 and Fig. 3).
In group B, the anterior capsule was incised in the middle and retracted to the sides to expose the hip joint. The TFLM was protected only by the surgeon’s conscious effort to avoid undue trauma to the muscle during the exposure of the joint to allow adequate visualization for the THA, without the use of special instruments or devices.
In both groups, acetabular and femoral prostheses were implanted in an identical way. Each patient received a non-cemented femoral stem, a press-fit acetabular cup, and a 32-mm ceramic head and ceramic lining. The prosthesis came from a single manufacturer (Waldemar Link Gmbh & Co, Hamburg, Germany). The capsule was then re-sutured closed after the arthroplasty was completed. No other structures were injured during the operation, and all 56 patients had their THA completed within 90 min. Drainage tubes were routinely placed after the operation to allow wound drainage for 48 h. Antibiotics were administered perioperatively for 48 h, and nadroparin calcium as prophylaxis against deep vein thrombosis was administered within 6 h postoperatively. After the patients were fully wake, they were encouraged to walk immediately with the aid of a walker.
At 8, 24, and 48 h after surgery, blood hemoglobin (Hb), serum levels of creatine kinase (CK), lactate dehydrogenase (LDH), and myoglobin (MYO), as well as postoperative drainage from the wound drains [13], were recorded and later compared between the two groups. In addition, we used a VAS scale to score pain at 8, 24, and 48 h after surgery.
No patient developed infection, facture, or other complications, and none of the patients in ether group required any allogeneic blood transfusions during the perioperative period. All patients were discharged from the hospital within one week, with the shortest discharge 3 days after surgery and the longest 7 days after the THA. Patients were required to return to the hospital once a month in the first six months after surgery and Harris scores were calculated in each visit. Harris scores at 1 and 6 months after surgery were recorded and compared between the two groups. A single observer blinded to the patient group collected the relevant clinical data.
An MRI was obtained 4 weeks after surgery and compared with the preoperative MRI. Imaging was performed on a 3.0-T MR tomograph (Achieva 3.0 T TX, Philips, Netherlands) according to a standard protocol. The T2-weighted, cross-sectional images were obtained at the level of the lesser trochanter. The location of the MRI cross-section is shown in (Fig. 4). By fully identifying the surrounding muscle tissue, the TFLM was located and its cross-sectional area and degree of fatty atrophy were measured. The TFLM was identified in the images by its elliptical shape and its position anterolateral to the femur and lateral to the rectus femoris muscle. As a measure of pathologic change, fatty atrophy was first identified and classified by Daniel in 1994. The classification system quantitated fatty changes in the rotator cuff into 5 grades: Grade 0, no fat; Grade 1, scant amounts of fat; Grade 2, fat visible but less than muscle; Grade 3, muscle and fat in equal amounts; and Grade 4, fatty streaks present more than muscle. Florian Engelken modified the classification to give the percentage of fatty tissue increments of 25% in 2014. [29, 30] We adopted the classification method of Daniel et al. and divided the fatty atrophy into 5 grades. The ellipse of the TFLM was measured along its long axis and short axis to obtain the muscle cross-sectional area from the formula.
Analysis of data
In order to determine the effects of the two types of operative attempts to protect the TFLM, we measured the mean level of blood Hb and serum muscle-specific enzymes, as well as the mean changes in fatty atrophy and cross-sectional areas, within groups before and after the THA. For wound drainage, because the data were not normally distributed, we calculated the median drainage (with interquartile ranges as well as actual ranges). All images were analyzed by an observer who was blinded to both groups.
Statistical analysis
We used a Student’s paired t test to compare the differences in the clinical and imaging data with or without use of the capsular pad and a Kruskal Wallace test for wound drainage. P values <0.05 were considered statistically significant. The data were analyzed by GraphPad Prism 5 software.