Application of the Chinese Aircraft-shaped Sleeve system in the treatment of tibial shaft fractures using a suprapatellar approach for tibial intramedullary nailing: a randomised controlled trial

The study was approved by our institutional review board and ethics committee and all participants provided informed consent.

Magnetic resonance images (MRI) of the knee joints of 212 healthy adults with normal lower limb function were obtained from May 2011 to May 2015 from a level I Chinese regional trauma centre. The exclusion criteria were as follows: previous history of knee surgery, trauma, space-occupying lesions in the knee joint and surrounding areas, patellar softening, sulcus dysplasia, synovial hyperplasia, and severe fluid accumulation in the joints. Of the 212 patients, 101 were men and 111 were women, and the age range was 18–73 years (mean 41 years). In total, there were 106 left knees and 106 right knees. The MRIs were obtained using a 1.5-T system (Philips 1.5-T Achieva, Philips, Netherlands) with patients in the supine position with full knee extension. Axial T2-weighted (TR/TE: shortest/100, matrix 256 × 192, FOV 160 × 160, slice thickness 3 mm) sequences were obtained. MRIs were analysed on a workstation (Centricity RIS CE V3.0, General electric medical system (China) Co., Ltd., USA) by one radiologist, and measurements were made using electronic callipers.

The patellar angle (PA) was measured on the axial image of the knee where the patella was the widest. The sulcus angle (SA), sulcus width (SW), and sulcus depth (SD) were measured on axial images of the knee where the transepicondylar axis was fully displayed on the first level from the proximal to the distal ends (Fig. 2).

Sixty inpatients who underwent treatment for a tibial shaft fracture at our centre, between June 2015 and June 2018, were enrolled. The selection criteria were as follows: (1) tibial shaft fractures (OTA type 42), (2) achievement of mature bone, and (3) the use of the Expert Tibial Nail system (ETN; DePuy Synthes, USA), with a suprapatellar approach. The exclusion criteria were as follows: (1) knee injury or stiffness, (2) patella baja, (3) patellar ligament calcification, (4) existing vascular nerve injury, (5) soft tissue injury around the incision area, and (6) metabolic or pathological bone diseases. After patients were enrolled in the study, each received a sealed envelope containing a randomised number, generated using SAS 9.2 (SAS Institute, Inc., Cary, NC, USA) based on the order of admission, for allocation to either the CASS group or the conventional sleeve group, in a 1:1 ratio.

The general characteristics of the patient group were as follows: 33 men and 27 women, with an age range of 22 to 69 years. There were no significant differences in baseline conditions between the CASS and conventional sleeve group. All procedures were performed by one senior surgeon. Due to the specific nature of the intervention methods, double-blinding could not be achieved between the surgeon and patients. Data processing, statistical analysis, and evaluations were performed by independent professionals who had no knowledge of group assignment.

Based on preliminary tests and a previous study [3], we initially defined the primary outcome as the rate of cartilage damage, with an estimated rate of 0.03 in the treatment group (CASS group) and 0.28 in the control group (conventional sleeve group). The ratio between the sample sizes of the two groups was 1:1. Using an α value of 0.05 and β value of 0.2 (power of 0.80), the minimum requirement, based on the SAS sample size calculator procedure [4], was 30 patients per group.

Patients in the CASS group underwent IMN using the original suprapatellar ETN system but with the CASS, which was designed based on the anatomical parameters of the patellofemoral joints measured from the MRI of the sample of 212 healthy individuals, and 3D printed (Fig. 3). The CASS was fabricated using thermoplastic polyurethane (TU), which has a stiffness of 42D–64D. The surgical procedure includes the following (Figs. 4 and 5): preoperative positioning of the patient in the supine position, with a knee pillow under the popliteal fossa of the affected limb to position the knee joint in a semi-extended position (flexion of 10°-20°). A longitudinal skin incision, of approximately 2–4 cm, was made, 1.5 cm above the superior patellar pole. The subcutaneous soft tissue was also incised. Deep separation was performed, with the quadriceps tendon being cut and the joint capsule above the patella opened further. An index finger was used to reach the joint capsule for preliminary separation of soft tissue. The patella was revealed anteriorly. The protective sleeve device was then placed, before placement of the guide wire. To address the shape mismatch in the CASS group, the positioning sleeved (aircraft cabin) was custom-designed based on the average knee parameters for Chinese individuals, using a translucent material. After opening the joint capsule, the positioning sleeve was gently inserted into the femoral sulcus, up to the tibial guide wire insertion point, and a 3.2-mm guide wire was drilled into it. An intraoperative radiograph was obtained to ensure optimal positioning of the insertion and position of the guide wire. The use of the compact, translucent positioning sleeve facilitated manoeuvring of the guide wire within the patellofemoral joint for accurate positioning. The positioning sleeve was removed, and the sleeve system was inserted (including the handle, inner protective sleeve, and centring sleeve) along the guide wire. The inner layer of the sleeve body has a hollow cylindrical shape for inserting a conventional metal circular inner sleeve. The outer layer is O-shaped in the axial position and extends on both sides to become V-shaped, thus resembling aircraft wings, which further stabilise the sleeve, such that even if high pressure is applied to the patellofemoral joint, the sleeve will not slide relative to the patella. Similar to the conventional infrapatellar approach, the entry hole of the nail was made in the direction of the guide wire after removal of the centring sleeve. Closed or open reduction was then performed. A guide wire was inserted for reaming followed by the appropriate intramedullary nail. The distal and proximal locking screws were fixed, and the end cap was inserted. After pulling out the metal guiding device, the surgical space was irrigated thoroughly. We also designed a connector (aircraft tail) for the end of the sleeve to improve flushing. The head has a cross-lock design that can be tightly joined or quickly separated by a rotation. The cross section of the tail is a hollow circle, similar in size to the pulsed irrigation nozzle (WJ-06, Wanjie Medical Device Co., Ltd., China). This section allows passage of the irrigation nozzle, with the nozzle reaching the area surrounding the opening of the tibia through the hollow tube, allowing for thorough irrigation and aspiration of the debris from the joint cavity. In this study, we used 4000 ml of sterile saline for irrigation. The incision was then closed in layers.

The conventional sleeve group underwent surgery using traditional suprapatellar procedures for the ETN system (DePuy Synthes, Raynham, MA, USA). The incision and joint cavity were irrigated repeatedly with a solution volume of 4000 ml.

The following outcomes were obtained for analysis: surgical time, calculated up to the point of remaining; arthroscopic assessment of the condition of the patellofemoral joint cartilage at two time points (once after deep separation, before sleeve insertion, and once after insertion of the end cap), using a Smith & Nephew arthroscope system (Smith & Nephew, Andover, MA, USA); the Outerbridge scale [5] which provides a grading of cartilage damage (grade 0, normal; grade I, cartilage with swelling and softening; grade II, fragmentation and fissuring in an area ≤ 0.5 in. in diameter; grade III, fragmentation and fissuring in an area > 0.5 in. diameter; and grade IV, erosion of the cartilage down to the bone); and the irrigation time. The number of cases with (positive) or without (negative) cartilage damage and residual debris was recorded for both groups. Arthroscopic evaluations were performed by a sports medicine expert, who did not have knowledge of group allocation.

Statistical analysis was performed using SAS 9.2 (SAS Institute, Inc., Cary, NC, USA). The data collected for PA, SA, SW, and SD were divided into two groups by sex. A t test was used to compare the sample means, adopting a completely randomised design. The Shapiro-Wilk test was used to test data normality, and the F test was used to evaluate the homogeneity of variance. The chi-square test, with correction for continuity in a 2 × 2 table, was used to compare the cartilage damage rate and the residual rate of debris between the groups. P < 0.05 indicated statistical significance.