In light of the widespread adoption of the minimally invasive approach within the orthopedic domain, closed reduction complemented by minimally invasive percutaneous instrumentation has progressively emerged as the preferred strategy for managing pelvic fractures. In comparison to conventional open surgery, this approach presents distinct advantages in terms of trauma, incision size, bleeding, pain, complication rates, duration of hospital stay, and the speed of rehabilitation[40, 41]. Nevertheless, the intricate nature of the pelvic structure poses challenges during the reduction process, thereby imposing limitations on the broader clinical application of minimally invasive treatments for pelvic fractures[42]. To address this challenge, scholars continuously strive for innovation. Matta, Yerasimides, and Lefaivre, for instance, have pioneered the development of pelvic fracture reduction frames utilizing X-ray fluoroscopy[27, 43]. Chen et al. accomplished a systematic reduction by refining the Starr pelvic fracture reduction framework[44] .
However, the current reduction frame is bulky, impacting surgical procedures and limited to single-directional traction fixation, making it challenging to achieve multi-dimensional precise reduction operations. Consequently, in practical surgical scenarios, surgeons are still required to manually execute reductions with intraoperative fluoroscopy assistance, heavily relying on the doctor's experience for optimal reduction outcomes. Addressing this issue, scholars are actively engaged in developing robot-assisted reduction systems; however, current research predominantly focuses on diaphyseal fractures[45–51]. At present, there is no intelligent serial RAFR system based on three-dimensional(3D) image guidance that can be truly used for minimally invasive reduction of complex pelvic fractures in clinical practice[8, 52].In this study, the only intelligent serial RAFR system based on 3D image guidance that can be used for minimally invasive reduction of complex pelvic fractures is adopted as the treatment method, which truly realized the intelligent and minimally invasive surgery of pelvic fracture reduction.
Analysis of the Efficacy and Safety of the RAFR System
In previous study conducted by our surgical team[53], the treatment results of 112 patients with unstable pelvic fractures treated with freehand closed reduction exhibited the following metrics: an average intraoperative blood loss of 243ml, an average operation time of 197.91min, an excellent and good reduction rate of 85.71%.The incidence of postoperative complications was 11.61%.
In this study conducted by our surgical team, mean intraoperative blood loss, mean operation time, excellent and good reduction rate, and incidence of postoperative complications were 133.25ml, 204.75min, 85% and 0%, respectively.
Compared with the previous surgical method of our surgical team, the new surgical method has significantly reduced the average intraoperative blood loss, the average operation time is slightly longer, the excellent and good reduction rate is similar, and the incidence of postoperative complications is significantly reduced. In this study, a 100% reduction success rate was achieved using the RAFR system, which was not possible with freehand closed reduction during our previous surgery.In the early stages, the relatively longer reduction time, fluoroscopy time and operation time were attributed to our surgical team's unfamiliarity with the RAFR system. As a new surgical method, we consider it acceptable.
The RAFR system, based on 3D image guidance, treats the pelvis as a 3D entity for reduction planning and operation. After achieving the target reduction positions in both anterior and posterior pelvic rings, fixation is performed to avoid mutual interference. Real-time observation of fracture reduction reduces judgment errors.The intelligent design of the RAFR system reduces technical and personnel requirements, saving on learning costs and human resources. Thanks to the repeatability of the robot, the uniformity of surgical quality has significantly improved.
Safety represents the paramount concern for medical devices. The RAFR system can achieve zero radiation exposure for surgeons during surgery. The system developed a verification procedure for image registration. Only when the registration accuracy is at least greater than 80%, can the next reduction procedure be carried out, which improves the success rate and safety. Real-time dynamic three-dimensional visualization navigation can accurately determine and control the direction, angle, and depth of the holding pin throughout the process, safely establish a screw channel, and reduce the risk of holding-pin loosening and even iatrogenic bone fractures.The reduction robotic arm boasts a maximum load capacity of 160 N, which is less than the force typically exerted by a surgeon's freehand manipulation in conventional surgery. Moreover, it offers synergistic force-position control, ensuring the safety of the reduction operation.The passive holding arm is equipped with a lock to prevent accidental unlocking while in a locked state.The elastic traction device is equipped with a one-dimensional force sensor, which ensures that the traction force does not exceed the safety value. When software or hardware malfunction occurs, the reduction robot will automatically stop, and the surgeon can also manually perform an emergency stop to ensure the safety of the surgery.
Key technologies used by the RAFR system
The overall working principle of this system is designed based on the principle of mirror symmetry of pelvic structure[33, 54–58], and allow the operator to manually adjust the planning error. Due to the limited possibility of obtaining morphological data of the affected hemipelvis prior to fracture in clinical practice, using pelvic symmetry reduction method is more closely aligned with clinical reality. In addition, the RAFR system supports the surgeon to set up and save multiple reduction path points on the basis of computer automatic planning to ensure a smooth reduction process.
Accurate registration of preoperative 3D CT images and intraoperative real-time images is the prerequisite of precise reduction. The RAFR system adopts the new registration mode of preoperative CT and intraoperative CBCT, and realizes dynamic three-dimensional real-time navigation through cross-modal image registration and fusion. The system uses near-infrared (IR) light to locate the position of the tracer placed on both sides of the pelvis. The collected measurement data is of sub-millimeter accuracy and repeatable, which can conduct dynamic three-dimensional real-time navigation, and realize the full visualization of the closed reduction path[8]. The next reduction procedure occurs only if the registration accuracy is at least greater than 80%. Compared to the conventional methods[59–61], The RAFR system does not require pre-surgical implant markers, thereby reducing the risk of pain and infection.
In the case of closed reduction, the stability of the pelvis is critical. The passive holding arms of the RAFR system can easily and rigidly connect to the orthopedic operating bed and the holding pins. The unique design of the U-shaped base keeps it away from the surgical area to facilitate disinfection and paving, while avoiding interference with reduction and fixation operations. In the locked state, the robotic arm can load up to 24kg, providing powerful fixation of the pelvis on the unaffected side and assisting in the accurate reduction of the fracture fragments on the affected side. On the basis of the original two holding pins, a horizontal holding pin was added to the top of the acetabulum to fix the pelvis on the affected side more stably. Under the guidance of 3D visualization, the holding pins are placed in the area with rich bone mass to improve the stability, which is conducive to the robot to stabilize the pelvis and efficiently exert the reduction force. This reduces the risk of needle loosening and iatrogenic fractures.
The surgeon usually needs a force greater than 200N in pelvic reduction surgery, while the maximum force of commonly used medical robotic arms is only 160N, which cannot reach the force required for reduction.The RAFR system effectively reduces the requirement for reduction force through elastic traction, solving the issue of insufficient reduction force associated with existing robotic arms and improving the reduction success rate.The lower limb elastic traction device of the RAFR system balances the resistance generated by the muscles and ligaments to improve the flexibility of reduction operation[62]. We found in clinical practice that the mean reduction operating force was decreased by approximately 35% and 58% under 5kg and 10kg elastic traction compared to rigid traction, respectively.
Shortcomings of the RAFR system
In clinical applications, the RAFR system has some limitations. The reduction path planning is based on pelvic symmetry, making it challenging to provide precise and effective reduction path planning for bilateral pelvic fractures with abnormal morphology or severe comminution on the unaffected side. If both ilium are too comminuted to firmly place the holding pins, surgery would be difficult to proceed as well. If the iliac bones on both sides are excessively comminuted, thereby precluding the stable placement of holding pins, the surgical procedure also becomes impracticable.The RAFR system lacks the ability to automatically adjust the reduction path, and it is unable to automatically unlock when encountering bony obstruction during the reduction process.Although optical positioning systems possess high-precision locational capabilities, they are susceptible to interference from objects within the surgical field, thereby impacting the accuracy and timeliness of tracking.