The operation time, length of incision, surgical blood loss, perioperative complications and incidence of bone healing complications were not statistically different between DCLS group and MCCS group, suggesting that the surgical incision in both groups was small with small trauma and less blood loss during the surgery, so both surgeries could be minimally invasive with less complications. The DCLS group had fewer healing complications than did the MCCS group, but there was no statistical difference. The Harris score, fracture healing time, femoral neck shortening, partial weight-bearing time and complete weight-bearing time at the last follow-up were significantly better in the DCLS group compared with the MCCS group. The fracture healing rate and mobility were better in the DCLS group than in the MCCS group, which suggests that DCLS, compared with MCCS, could increase fracture healing rate, improve patients’ mobility and hip function, accelerate earlier bone healing and prevent excessive shortening of femoral neck. So patients in the DCLS group can carry out weight-bearing activities earlier, and have the better quality of life.
Femoral neck fractures are one of the problems in clinical treatment. The prognosis is uncertain. Nonunion and femoral head necrosis are recognized as serious complications after internal fixation. The type of femoral neck fracture and improper treatment are considered to be the main factors of nonunion and femoral head necrosis. Garden classification is the mainstream classification system for femoral neck fractures and guides clinical treatment. Decisions on the treatment of displaced (unstable) fractures (Garden III and IV) in young patients are still controversial. Surgical methods include closed/open reduction internal fixation, hemi-hip replacement, and total hip replacement . Stable fractures(Garden I and II) tend to be treated with internal fixation.
Femoral neck fractures, no matter what treatment method is selected, have a significant impact on the living quality of patients and bring a large economic burden to society . Although internal fixation has a higher incidence of postoperative revision rate, complications, nonunion, delayed bone healing, and poor function in the treatment of undisplaced femoral neck fractures for super-aged patients , internal fixation still is currently preferred for Garden I and II femoral neck fractures . However, there is still no consensus on which internal fixation method can better maintain the stability of fractured ends, promote fracture healing, and avoid and reduce complications such as femoral head necrosis, nonunion, and internal fixation failure.
Three cannulated screws can exert pressure on the fracture end and promote fracture healing. In addition, they occupy a relatively small area of femoral neck, and have less interference with blood flow for femoral head and neck. The triangular distribution can form a three-dimensional frame with bone tissue, which can improve stress against the rotation of femoral head. It can enhance compressive stress intraoperatively and postoperatively between fracture ends, which could promote closer contact between fracture ends and be conducive to fracture healing. However, because the three cannulated screws are not related to each other; the position of the screws are easily affected by subjective and objective factors of the operator. So its resistance to vertical shear and torsion is relatively poor, which can lead to loosening and re-displacement of fracture end, femoral head necrosis, nonunion, and femoral neck shortening [24, 25]. And in the process of healing, lack of sustained and effective solid support will affect the rehabilitation training of the affected limb and growth of the fractures .
Although dynamic hip screw can provide better angular stability and sliding compression, its anti-rotational stability is poor, especially when the hip screw is screwed, which can easily cause poor alignment of the femoral head and neck. Furthermore，dynamic hip screw require large soft tissue exposure, and hip screw insertion damages the cancellous bone of femoral head and neck and destroy its blood supply, which affects the healing of femoral neck fractures.
DCLS consists of three conventional hollow lag screws, one femur lateral plate with locking screw holes and three locking tail caps. During the operation, three conventional hollow lag screws were inserted into femoral neck through the lateral plate to perform static compression and fixation on the fracture end, and then the three locking tail caps were screwed into the locking thread hole on the lateral plate to achieve screw locking. In the process of fracture healing, because the lateral plate is not fixed with the femur, there is good dynamic pressure between the fracture ends under external force loading. Therefore, this design can simultaneously achieve static and dynamic pressure action, to meet the necessary conditions for fracture healing. At the same time, the tail-cap locking design of the system makes the three screws become an integrated rigid frame structure, and the screws can support each other. Therefore, DCLS combines the advantages of MCCS and dynamic hip screw, which can not only improve strong, uniform and accurate compression of the fracture section intraoperatively, but also have stable frame structure to stabilize the broken end of the fracture and controlled dynamic compression to prevent excessive shortening of femoral neck. So it can obtain good initial and continuous stability to prevent displacement of the fracture ends and help fracture healing. Early biomechanical experiment of human corpses showed that DCLS, compared with MCCS, has better biomechanical stability, stronger compressive and torsional resistance .
The tail-cap locking design of DCLS makes the three screws become an integrated rigid frame structure, and the screws can support each other. The load can be evenly distributed among three screws and applied to the lateral plate. On the one hand, The structure can be more even effectively combat stress, bending stress, tensile stress and rotation. On the other hand, although the system does not have the traditional angular stability design, the vertical shear stress can be transferred to the lateral plate by screws, and then the counterforce between the lateral plate and the femoral lateral bone cortex can be used to resist the vertical shear stress. This is completely different from traditional three cannulated lag screws, whose resistance to shear force can only be realized by relying entirely on the limited contact between the screw end cap and the lateral bone cortex. The contact area of this contact is much smaller than that of the lateral plate and the bone cortex in DCLS. So the angular stability of the three cannulated lag screws was worse than that of DCLS. In biomechanical experiments, when the lateral loading was 400N in the horizontal compressive loading test, the compressive stiffness (4 324±l 234) N/mm of DCLS was significantly larger than that of the three cannulated lag screws (3 020±855) N/mm (P=0.0050) . The torsional stiffness of DCLS (11.45 ± 4.95) N·m/º was significantly larger than that of the three cannulated lag screws (6.53±4.83) N·m/º when the torsional load was 2.5 N m/º (P=0.0423) .
At present, three cannulated screws are commonly used for internal fixation of stable femoral neck fractures, but many studies have significant differences in the position distribution and clinical overcomes . The three cannulated screws have large differences in torsion resistance and fracture end stability . The instability of the fracture end is not conducive to fracture healing. Weil et al  used three cannulated screws in an inverted triangle to treat 41 cases of femoral neck fractures, 71% of them had a significant femoral neck horizontal shortening greater than 5mm, and 25% of them had severe shortening greater than 10mm. Significant shortening occurred in 43% of patients in the vertical direction, and severe shortening occurred in 17% of the patients. Screw pullout greater than 5 mm occurred in 41% of patients. 7 cases required late hip replacement. Gupta et al. studied hollow cancellous screws for femoral neck fractures for up to 4 years. The imaging healing time was 7.1 months, the healing rate was 82.22%, the osteonecrosis rate was 6.67%, and the Harris Hip Score was 88.65. Manohara et al.  studied cancellous screw fixation for undisplaced femoral neck fractures in elderly patients, Of the 96 patients followed up for a mean of 39 months, 8/96 (8.3%) underwent revision surgery for femoral head avascular necrosis (5/96, 5.2%) or non-union/implant failure (3/96, 3.1%). Overall, 30/96 (31.3%) patients had a decrease in their mobility status. Chen et al studied patients with femoral neck fractures treated with the dynamic hip system blade or MCCS for an average follow-up of 27 months. No statistically significant differences in the rates of nonunion (4.5% vs. 0) and femoral head avascular necrosis (9.1% vs. 7.1%) were observed. 15.9% of patients reported a femoral neck shortening greater than 10 mm. Other study has found that for femoral neck fractures treated with three hollow screws, nonunion and osteonecrosis were 42% and 17% in the displaced fracture group and 6% and 4% in the non-displaced fracture group .
However, this study had following limitations. The number of cases was relatively small. It was a single-center prospective study and has not been completely randomized and double-blind. The results may be biased. Therefore, this study needs to be verified by further multicenter, randomized, controlled, double-blind clinical trials.