The Role of Intraoperative Navigation in Orthopaedic Surgery

Alexa J. Karkenny, MD; Joseph R. Mendelis, MD; David S. Geller, MD; Jaime A. Gomez, MD


J Am Acad Orthop Surg. 2019;27(19):e849-e858. 

In This Article

Intraoperative Navigation in the Orthopaedic Subspecialties

Spinal Deformity and Degenerative Disease

The surgical treatment of spinal deformity and degenerative disease has evolved over the past 20 years with the widespread use of pedicle screws. A common goal for emerging intraoperative navigation platforms has been to maximize the safety of pedicle screw placement.[5] Pedicle screw placement can be technically demanding, as a surgeon must identify the correct entry point, trajectory, and length of the unexposed pedicle or risk injury to the nearby spinal cord, nerve roots, and great vessels. Rotational deformities and narrow thoracic pedicles make screw placement increasingly challenging. Freehand methods for pedicle screw placement are widely accepted and have had reported screw misplacement rates anywhere from 1.7% to 15.7%.[6,7] There is an abundance of literature on the accuracy and safety of pedicle screw placement, some of which is highlighted in Table 2. One multicenter registry reported the accuracy of pedicle screw placement using intraoperative CT-based navigation as 97.5%.[7] Other authors have shown that intraoperative (versus preoperative) CT-based navigation techniques for scoliosis may increase the accuracy and efficacy of placement of a single pedicle screw.[8] Robot-assisted navigation techniques have demonstrated decreased risk of pedicle screw breach compared with freehand fluoroscopy techniques.[9] Navigated pedicle screw placement safety has also been studied among cohorts with a complex anatomy and/or narrow pedicles. Cervical pedicle screws are biomechanically advantageous to other posterior cervical fixation techniques but technically challenging due to the risk of neurovascular injury; one study has reported a 99% accuracy rate of navigated cervical pedicle screw placement.[10] Authors have found real-time visual feedback from navigation useful for preventing attempted instrumentation of pedicles that were absent or otherwise appeared to be impassible and for improving implant density in the setting of syndromic scoliosis.[11,12] Intraoperative navigation has been shown to reduce the odds of unsafe screw placement in narrow thoracic pedicle screws by nearly four times and the odds of a medial breach by nearly eight times compared with a non-navigated cohort.[6]

Radiation exposure during spine surgery is of interest to the public because historically, patients with scoliosis were found to have increased rates of breast and thyroid cancer.[14] Radiation exposure does vary significantly for intraoperative navigation modalities as some rely on CT or fluoroscopy whereas others are imageless. The most recent evidence on patient radiation exposure using CT-based navigation versus fluoroscopy has been inconclusive. One study demonstrated higher effective radiation doses using CT compared to fluoroscopy for posterior spinal fusion, especially for obese children.[13] However, progress has been made to reduce the radiation dose. A pediatric "O-arm" protocol reduced the mean radiation dose (1.17 mSV) to 10 times less than the device's default protocol, which generated satisfactory images in all but one patient in one study.[14] Any comparison is confounded by the surgeon technique, as fluoroscopy time is highly variable, but authors have calculated that the effective dose (0.65 mSV) of one pediatric protocol (80 kV, 20 mA, 80 mA s) for instance, approximates 85 seconds of fluoroscopy time.[15] Surgeons can thus use this information to decipher which imaging modality is optimal for their patients based on their anticipated fluoroscopy time. The radiation dose received by the surgical team is also of interest. The reported yearly radiation dose for a pediatric spine surgeon using fluoroscopic guidance is 3.33 mSv, wearing a lead apron and standing adjacent to a C-arm.[16] Similarly, the highest yearly radiation dose for an unprotected person standing 2 m from the center of an O-arm is 3.32 mSv.[18] The latter suggests that navigation reduces occupational exposure because protective lead, standing behind a lead shield, or exiting the room during the scan can each lower this dose even further.

The operative time and cost associated with intraoperative navigation have also been studied in spinal surgery. The setup time and total operative time using navigation have been shown to be 10 to 20 minutes longer than those without navigation, but navigation has also been shown to cut the time for placement of a single pedicle screw in half.[17,19] Another study found a trend toward longer setup times in the navigated group but a significant decrease in operative length over time in navigated but not freehand cases, indicative of navigation's possible "learning curve."[20] Also, an economic analysis on adult spinal surgery using intraoperative navigation reported fewer reoperations for misplaced screws than a matched cohort for which conventional fluoroscopy was used.[21]


Over the past decade, intraoperative navigation in arthroscopy was largely focused on anterior cruciate ligament (ACL) tunnel positioning in ACL reconstructions to optimize graft kinematics and isometry. Most failed ACL reconstructions are attributable to tunnel malposition, so navigation has been applied in an attempt to increase the accuracy of tunnel placement. A prospective, randomized controlled study comparing tunnel placement for primary ACL reconstruction using a manual technique and a CT-free navigation system did not find any significant differences in tunnel positioning, graft impingement, or function at 2 years postoperatively.[22] More recently, use of a novel optical tracking marker and landmark acquisition method for imageless navigation of ACL tunnel placement did show an improved accuracy of posterior wall margin with navigation.[23] Although navigation has had limited uses in knee arthroscopy, recent research on navigated hip arthroscopy has demonstrated potential to improve this operation. Osteochondroplasty for the treatment of femoroacetabular impingement poses a challenge to the surgeon who must rely on preoperative 2D imaging and intraoperative arthroscopic images to evaluate and then correct a 3D pathology. Navigation-assisted surgery has been proposed to increase its accuracy and success. Kobayashi et al[24] reported their experience with preoperative planning software and intraoperative navigation for the treatment of cam morphology femoroacetabular impingement. The authors performed impingement simulation to calculate an ideal resection area and depth for a virtual osteochondroplasty, and then visualized improvement in range of motion on the simulation model after the virtual surgery. Intraoperative fluoroscopic guidance was used to register anatomic landmarks, the preoperative plan was executed, and a tracked pointer was used to assess the volume of resection. The authors concluded that this technique would help surgeons to first preoperatively identify an impingement point and the extent of bone necessary to resect for effective osteochondroplasty, and then intraoperatively execute a precise plan that can be objectively assessed.

Orthopaedic Oncology

Intraoperative navigation was first introduced in tumor surgery as a means of improving surgical accuracy and margins, but its applications have evolved to improve both oncologic and reconstructive outcomes. The technology has been used in the treatment of pelvic and periarticular resections to help achieve adequate margins while also sparing important nearby structures or articular surfaces. Pelvic tumors pose a surgical challenge, considering the surrounding organs and neurovascular structures as well as the difficult 3D anatomy that surgeons conceptualize to plan resections. Sawbones and cadaver studies have illustrated that navigation improves the accuracy of surgical osteotomies and margins, even in the absence of any soft-tissue consideration.[25] The fusion of CT and MR imaging has enhanced the visualization of soft-tissue structures and soft-tissue tumor extension, further improving surgical guidance and safety.[26] Improved accuracy allows surgeons to pursue more ambitious joint-sparing and/or multiplanar resections, with the overarching aim of realizing better functional outcomes, reconstructive longevity, and quality of life. Figure 3 shows an intraoperative navigation software interface during a wide resection of a chondrosarcoma of the posterior acetabulum with sparing of the articular surface.

Figure 3.

The intraoperative navigation software interface during the wide resection of an ischial spine chondrosarcoma. The ischial spine and posterior acetabulum were curetted down to the subchondral bone, and intraoperative navigation facilitated sparing of the articular surface. The interface displays axial (top left), sagittal (top right), and coronal (bottom left) CT cuts, as well as an AP pelvis radiograph (bottom right) with the surgeon's coupled pointer superimposed.

Navigation-assisted tumor resection can also help a surgeon to carry out a complex preoperative plan involving a custom allograft or megaprosthesis with precision, thereby improving allograft–host bone congruency (and, in turn, bone union) or prosthesis fit, respectively. Outcomes for extremity resection and allograft reconstruction have yielded a nonunion rate of as low as 6%, while encountering few technical limitations and a modest registration time requirement.[27] It remains difficult to compare oncologic outcomes in small reported series because of varying histologies, inherent biological properties, chemotherapy response, tumor sizes, and limited follow-up. However, improvements in intraoperative visualization, design and execution of a planned resection, and surgical accuracy all serve in support of navigation, even in the absence of more robust outcome data.


Intraoperative navigation has also been a valuable tool in the arthroplasty subspecialty. In TKA, the technology has been used to aid in component positioning, gap balancing, and mechanical alignment. Recent literature on navigated TKA is summarized in Table 3. One study demonstrated that navigation-assisted TKA has been shown to significantly decrease the incidence of postoperative component malalignment.[28] Navigated unicompartmental knee arthroplasty (UKA) may also benefit from increased accuracy, as one study reported a statistically significant increase in the proportion of implants within 2° of the target positions in all parameters when compared with a conventional group.[31] Another study echoed these benefits of navigated TKA but questioned whether navigation demonstrated a clinically relevant advantage, reporting no significant differences in functional outcome, quality of life measures, or satisfaction rates at 2 years postoperatively.[29] However, a study using the Australian Orthopaedic Association National Joint Replacement Registry found that patients under 65 years of age who underwent navigated TKA had a significant reduction in revision rates overall and revisions specifically for loosening when compared with conventional TKA.[30] In addition, a recent study on the U.S. Nationwide Inpatient Sample database demonstrated that navigated TKA was associated with lower transfusion rates and perioperative complications but with no significant difference in length of stay or hospital charges compared with conventional TKA.[32]

Intraoperative navigation systems have also been used to assist in successful component positioning for total hip arthroplasty and reverse total shoulder arthroplasty. In total hip arthroplasty, CT-based navigation has been shown to improve accuracy of cup positioning, including both cup inclination and anteversion, in a patient cohort with primary osteoarthritis and hip dysplasia.[33] Additionally, a study on passive navigation for reverse total shoulder arthroplasty using patient-specific glenoid baseplate drill guides has shown the potential to improve the accuracy of glenoid baseplate positioning when compared with a preoperative surgical plan.[34] Although navigated arthroplasty does have its benefits, it is important to note that the temporary reference pins placed outside of the surgical site can pose pin-site complications. A review on 3,136 pin sites in 839 patients reported five pin-site complications, including three infections, one neurapraxia, and one suture abscess, but no fractures.[35]

Orthopaedic Trauma

Surgical navigation has also been studied in various orthopaedic trauma settings. Fractures that are amenable to percutaneous screw fixation lend themselves particularly well to intraoperative navigation. For percutaneous pinning of femoral neck fractures, fluoroscopy-based navigation has been shown to improve parallelism and spread as well as minimize overall complications and revision surgery rates compared with conventional C-arm fluoroscopy.[36] Along the same lines, applications in percutaneous screw fixation for pelvic and acetabular fractures have demonstrated high accuracy of screw placement compared with a CT-based preoperative plan.[37] A cadaveric study on percutaneous iliosacral screw placement by orthopedic trainees using intraoperative fluoroscopic navigation coupled with a CT-based preoperative plan showed that a navigation use significantly increased accuracy, decreased Kirshner wire insertions, and decreased radiation time.[38] A cadaveric study on scaphoid fracture percutaneous fixation showed similar benefits: although navigation required an average of five additional minutes of setup, it significantly reduced guidewire placement time and fluoroscopy time, and had the same accuracy as conventional fluoroscopy.[39]