Power-Tool Use in Orthopaedic Surgery

Background: Power tools are an integral part of orthopaedic surgery but have the capacity to cause iatrogenic injury. With this systematic review, we aimed to investigate the prevalence of iatrogenic injury due to the use of power tools in orthopaedic surgery and to discuss the current methods that can be used to reduce injury. Methods: We performed a systematic review of English-language studies related to power tools and iatrogenic injuries using a keyword search in MEDLINE, Embase, PubMed, and Scopus databases. Exclusion criteria included injuries related to cast-saw use, temperature-induced damage, and complications not clearly related to power-tool use. Results: A total of 3,694 abstracts were retrieved, and 88 studies were included in the final analysis. Few studies and individual case reports looked directly at the prevalence of injury due to power tools. These included 2 studies looking at the frequency of vascular injury during femoral fracture fixation (0.49% and 0.2%), 2 studies investigating the frequency of vertebral artery injury during spinal surgery (0.5% and 0.08%), and 4 studies investigating vascular injury during total joint arthroplasty (1 study involving 138 vascular injuries in 124 patients, 2 studies noting 0.13% and 0.1% incidence, and 1 questionnaire sent electronically to surgeons). There are multiple methods for preventing damage during power-tool use. These include the use of robotics and simulation, specific drill settings, and real-time feedback techniques such as spectroscopy and electromyography. Conclusions: Power tools have the potential to cause iatrogenic injury to surrounding structures during orthopaedic surgery. Fortunately, the published literature suggests that the frequency of iatrogenic injury using orthopaedic power tools is low. There are multiple technologies available to reduce damage using power tools. In high-risk operations, the use of advanced technologies to reduce the chance of iatrogenic injury should be considered. Level of Evidence: Therapeutic Level IV. See Instructions for Authors for a complete description of levels of evidence.

patients undergoing orthopaedic surgery and to discuss the current methods to reduce injury and thus improve patient safety.

Materials and Methods
Search and Information Sources T he methodology of this review is reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines 16 . A systematic review of published literature relating to power tools and iatrogenic injury was undertaken via MEDLINE, Embase, PubMed, and Scopus databases up to April 1, 2020. We used a combination of the search terms "overdrill," "oversaw," "tool," "injury," "danger," "safe," "damage," "overshot," "risk," "drill," "saw," "iatrogenic," "hospital acquired condition," "medical errors," "medical mistake," and "orthopaedic". The electronic database search was further supplemented by manual review of the references within key relevant studies.

Eligibility Criteria
Studies were included if they reported on an orthopaedic procedure that featured the use of a power tool. All study types were eligible, including case reports, animal research and simulation studies, cohort studies, and literature reviews. Included studies were either written in, or translated to, the English language.
Studies were excluded if they involved injuries related to cast-saw, arthroscopic trocar, diathermy, needle, scissors, and blade use. Studies relating to complications outside the scope of orthopaedic surgery, injuries not clearly related to intraoperative power-tool use, or injuries related to temperature-induced damage or infection were also excluded.
The 2 outcomes measured were the prevalence of iatrogenic injury related to power-tool use and the methods or safety mechanisms present to reduce iatrogenic injury.
The first 2 authors (M.C.A.A. and S.Z.) independently conducted the search, screened abstracts, and selected studies for review. Any discrepancies regarding article inclusion were resolved via discussion as recommended by the Cochrane Collaboration guidelines 17 .

Data Collection Process
Relevant data items were extracted from each included article and are presented in Table I. Items included were type of bone, type of iatrogenic injury, power tool used, type of operative procedure, and outcome/recommendation by the authors. The papers were then categorized according to prevalence of iatrogenic injuries, methods to reduce damage using power tools, methods to detect damage when using power tools, and recommendations for power-tool settings.

Statistical Analysis
Statistical analysis was not possible because of the heterogeneity of the study types and clinical data mostly coming from case series. Therefore, descriptive statistics were used where possible.

Source of Funding
This study was supported by a grant from the Wellcome Trust (208858/Z/17/Z). General laboratory funding was provided by the U.K. National Institute for Health Research, The Royal College of Surgeons of England, the Dunhill Medical Trust, and the Michael Uren Foundation.

Results
A total of 3,689 abstracts were retrieved via the electronic databases using the search criteria. Five abstracts were included from the manual reference search of relevant papers. Following duplicate removal and abstract screening, a total of 460 eligible full-text studies remained. After review of these full manuscripts, 88 papers were deemed to fit the inclusion criteria and were subsequently included in the systematic review. Figure 1 shows the study selection flow, according to the PRISMA guidelines 16 . Table I lists data collected from all selected studies, with relevant findings and recommendations.

Prevalence of Iatrogenic Injury Using Power Tools
There were few studies exploring the prevalence of iatrogenic injuries due to power tools. The majority looked at vascular injury, including 2 studies involving femoral fracture fixation 6,18 , 2 involving spinal surgery 19,20 , and 4 involving total joint arthroplasty 12,[21][22][23] . Two of these were systematic reviews that investigated proximal femoral fractures 6 and total hip arthroplasty 12 , with the majority being retrospective studies and 1 questionnaire sent to vascular surgeons 22 . In addition, there were multiple case reports of arterial, nervous, and ureteric injury due to screw placement and drill bit use [24][25][26][27][28][29] .
Novel Methods to Reduce Damage When Using Orthopaedic Power Tools Nine papers involved robotic systems for use during knee arthroplasty 13 , femoral neck fractures 30,31 , and spinal surgery 32,33 or in conjunction with sawing 34 and in laboratory studies 35,36,67 .
Six papers studied simulation [37][38][39][40][41][42] , with the different types summarized effectively by Vanikipuram et al. 42 . These studies showed that simulation can successfully be used to reduce plunging depth in trainee surgeons 38,39 and that computer-based simulation was found to provide effective and transferable skills for inexperienced surgeons 40,41 .
Six papers investigated piezoelectric/vibrational drilling 43-48 , 2 studies looked at dual motor drilling 14,49 , and 4 papers were on recommendations for drill bits 5,[50][51][52] . Two studies by the same authors investigated the Taguchi method and suggested a lower rotational speed of 1,000 rpm and feed rate of 50 mm/min with a twist drill to reduce surface roughness and improve drill-hole quality 53,54 . Six papers looked at imaging techniques in orthopaedic surgery. These include fluoroscopy 55 , magnetic resonance imaging (MRI) 56,57 , fluoro-free navigation techniques 58 , computer-assisted navigation, and 3D image-guided placement 59,60 .         64 , bioimpedance drills 65 , and acoustic emission-signal analysis 66 .

Recommended Settings for Power-Tool Use
Aziz et al. developed an algorithm that detects excessive force and breakthrough of the drill bit during bone drilling, whereby the drill will halt and return to a safe position once the algorithm is triggered 67 . Pandey and Panda calculated the point at which a drill had broken through bone. They found that the best combination of bone-drilling parameters for minimum thrust force is 30 mm/min of feed rate and 1,805 rpm of spindle speed 68 .

Discussion
T his systematic review aimed to determine the prevalence of iatrogenic orthopaedic injuries related to intraoperative power-tool use in the literature and current methods available for reducing the occurrence of these injuries. A total of 88 studies were retrieved and analyzed to help answer these questions, although a wide range of orthopaedic procedures were included. Where possible, we have given recommen-dations to reduce injury on the basis of the studies. However, our intention was to gain appreciation of the breadth of reported iatrogenic injuries in the literature; providing specific recommendations for each procedure is outside the scope of this review.

Prevalence of Iatrogenic Injury
Overall, there were few papers that specifically explored the prevalence of iatrogenic power-tool injuries in orthopaedic surgery. Where reported, types of iatrogenic injuries included vascular, nervous, and ureteral injury.

Power-Tool Use in Orthopaedic Surgery
One systematic review from 2015 looked at vascular injuries that occurred during internal fixation of proximal femoral fractures 6 . The authors estimated the incidence of these injuries to be 0.49%. They showed that, of 182 cases of injury identified, 175 were reported as iatrogenic injuries, mostly in the extra-pelvic vessels and specifically the profunda femoris artery. Interestingly, from their analysis, at least 28 of these cases had a confirmed mechanism of injury involving a drill bit. The authors make several recommendations related to powertool use during internal fixation of proximal femoral fractures. These include encouraging the use of powered instruments under image-intensifier guidance, maintaining the leg in neutral with reduced traction, and keeping the drill bit sharp 6 .
Another systematic review from 2015 investigated the incidence of vascular injury during total hip arthroplasty 12 . The authors described 138 vascular injuries in 124 patients, mostly affecting the common femoral artery (23%) and with the most prevalent mechanism being laceration. However, there was no association between the type of blood vessel injured and surgical approach. The main contributing factors appeared to be aggressive medial retractor placement and injury from screw fixation of the acetabular component. Although not explored in depth during this review, it is important to recognize that the surgeon's (and assistant's) knowledge of anatomy and correct retractor placement is vital to reducing the chance of iatrogenic injury. Other retrospective studies looking at arterial injury in total hip/knee arthroplasty found an incidence of 0.13% 21 and 0.1% 23 , both noting direct laceration as a cause. In addition, a survey sent to vascular surgeons in the U.S. demonstrated 19 instances of popliteal artery injury during total knee arthroplasty (12 cases of which were due to direct injury). However, the response rate was low, with only 13 replies from 190 survey recipients, so underreporting is extremely likely 22 .
Smith et al. conducted a retrospective review of 10 cervical decompression procedures performed by 9 spinal surgeons 20 . They found that the incidence of iatrogenic injury to the vertebral artery was 0.5%, with all cases related to intraoperative motorized power-tool instrumentation. Four of these patients also suffered postoperative neurological deficit, which occurred as a direct result of the arterial injury. The authors give recommendations for avoiding injury, such as dissecting the bone/disc material as close to the midline as possible or using imaging to determine vertebral artery position and artery proximity to the lesion 20 .
A retrospective multicenter study looking at iatrogenic injury to the vertebral artery demonstrated an overall incidence of 0.08%, with C1-2 posterior fixation having the highest incidence (1.35%). This study involved 15,582 surgeries in 21 centers, and 77% of the cases showed no permanent neurological deficit 19 .

Novel Methods for Reducing Damage When Using Orthopaedic Power Tools
The range of robotic systems in surgery is increasing, with numerous systems developed in the last decade to overcome the inaccuracy of manually navigating orthopaedic tools 35,69 .
The benefits of robotic systems include increased safety and a reduced rate of iatrogenic injuries 13,70,71 .
Oscillating saws have the potential to cause soft-tissue damage during total knee arthroplasty 72 , and Cartiaux et al. showed that using robotic navigation in conjunction with these tools has the potential to significantly decrease iatrogenic injury compared with freehand techniques 34 .
Another study looked at robotic-assisted cervical transpedicular screw placement, finding that it achieved 98.8% accuracy in Kirschner wire placement and improved functional outcomes compared with non-robotic-guided placement 33 . Another study of robotic-assisted pedicle screw placement also found increased accuracy in spinal surgery when compared with fluoroscopy-guided techniques 32 .
Shim et al. tested a compact robotic drill prototype using an automated "rolling friction mechanism," which allowed safe removal of the drill tip in an emergency while not compromising the speed and accuracy of the drill 36 .
Piezoelectric surgery uses high-frequency ultrasonic vibrations to cut bone tissue 73 . When compared with conventional drilling, vibrational drilling aims to reduce force, torque, and thermal damage to bone. This is thought to be possible because of the increased precision and reduced bleeding due to a "microcoagulation" effect 44 . For instance, it has been demonstrated that an elliptical vibration-assisted oscillating saw can minimize required cutting force 48 and also reduce risk of soft-tissue injury 74 . This technique can be applied safely in a low-field MRI environment and is a valuable method to facilitate transcortical bone biopsy 45 , but there is minimal comparison of this and traditional methods in the literature. In contrast, another study evaluated the use of ultrasonic bone curette compared with a high-speed drill in spinal surgery. Both groups experienced dural tears, and this study concluded that one method was not significantly better than the other 43 . The suggested optimal settings for vibrational drilling were noted in 1 study to be a drill speed of 1,000 rpm with a frequency of 20 kHz 75 .

Methods of Detecting Damage When Using Orthopaedic Power Tools
To minimize iatrogenic injury, it is important to easily and rapidly identify when injury occurs intraoperatively. One novel example includes the use of a spectroscopy device integrated into a power drill to detect the bone-tissue boundary when drilling holes for intramedullary nailing 15 . This helps to reduce breaching of the periosteum and unintentional soft-tissue injury.
The use of methods providing real-time feedback is increasing. Bolger et al. used an electrical conductivity device to detect iatrogenic spinal pedicle perforation 61 . In 1 multicenter study, it demonstrated a sensitivity of >98% in the detection of breaches, 52% more when compared with the surgeon alone 62 . Similarly, a systematic review of intraoperative somatosensoryevoked potential and transcranial motor-evoked potential methods in cervical spine surgery showed a high sensitivity/ specificity for both (22% to 100%/100%, and 78% to 100%/ 100%, respectively) 64 . Another study used stimulus-evoked Power-Tool Use in Orthopaedic Surgery electromyography to detect proximity to neural structures during iliosacral screw placement. Four of 51 screws were redirected as a consequence of the technique, and all patients had normal neurological status postoperatively 63 . Other novel methods include the use of a bioimpedance-sensing drill to successfully differentiate between cortical and cancellous bone 65 and acoustic-emission signal analysis, which is based on the principle that different bone types will produce varying sound signals when being drilled 66 .

Recommendations for Power-Tool Settings
With such a wide range of equipment, imaging modalities, and device settings available, there can be much heterogeneity in tools and settings used during orthopaedic procedures. Several papers have specific recommendations for power-tool equipment settings to help reduce the risk of iatrogenic injury.
With regard to drill bits, several papers agree that blunt drill bits cause higher damage to bone than sharp bits 51,52 , with 1 study demonstrating significant differences in plunging depth when sharp or blunt drill bits are used 5 . Smooth pins have also been shown to reduce the risk of overdrilling compared with threaded pins and can reduce iatrogenic injury in the form of tissue entanglement 50 .
Two studies looked at dual motor drilling. Unlike use of a conventional drill, this involves a second motor that retracts an attached sleeve at a set rate, accurately advancing the drill bit, and measures the drill bit's energy expenditure and the distance drilled, which is continuously communicated to the surgeon. Gilmer and Lang 14 looked at a dual motor drill for real-time measurement of torque, depth, and bone density. They found that this tool could accurately determine these parameters and, thus, give indications of screw pull-out force and cortex boundaries to prevent screw stripping and overpenetration. The second study showed a mean plunging distance of <1 mm using a dual motor drill and found that there was no difference between novice and experienced surgeons using this technique 49 . It should be noted, however, that both were preliminary studies involving artificial bone specimens and the clinical applicability of the tool would need to be validated in vivo. Other methods in the literature on detecting real-time feedback of bone conditions included the use of laser displacement sensors in a laboratory study 76 .

Conclusions
Although iatrogenic injury in orthopaedic surgery has been described in the literature, it likely is vastly underreported. Despite this, it is important not to overlook the role of power tools in contributing to patient harm and techniques for reducing injury. Such methods should be considered in terms of equipment factors (e.g., drill speed, intraoperative imaging, use of robotic guidance), patient factors (e.g., anatomical variance, safe zones), and surgical factors (e.g., tools to increase haptic feedback, simulation training, and knowledge of critical anatomy). n