- Email: [email protected]
Precision of Robotic Guided Instrumentation for Acetabular Component Positioning
The Journal of Arthroplasty 30 (2015) 392–397
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Precision of Robotic Guided Instrumentation for Acetabular Component Positioning Vaibhav Kanawade, MD a, Lawrence D. Dorr, MD b, Scott A. Banks, PhD c, Zenan Zhang, MS c, Zhinian Wan, MD b a b c
Kanawade Patil House, Sangamner, India Orthopedic Department, Keck Medical Center of USC, Los Angeles, California Department of Mechanical & Aerospace Engineering, University of Florida, Gainesville, Florida
a r t i c l e
i n f o
Article history: Received 24 August 2014 Accepted 11 October 2014 Keywords: computer robot THR cup position MAKO
a b s t r a c t Robotic computerized instrumentation that guides bone preparation and cup implantation in total hip arthroplasty was studied. In 38 patients (43 hips) intraoperative cup inclination and anteversion were validated by postoperative CT scans. Planned inclination was 39.9° ± 0.8° and with robotic instrumentation was 38. 0° ± 1.6° with no outliers of 5°; on the postoperative CT scan there were 5 outliers (12%). Planned anteversion was 21.2° ± 2.4° and intraoperatively was 20.7° ± 2.4° with no outlier of 5°; on the CT there were 7 outliers (16%). The center of rotation (COR) was superior by a mean 0.9 ± 4.2 mm and medial by 2.7 ± 2.9 mm. This robotic instrumentation achieved precision of inclination in 88%, anteversion in 84% and COR in 81.5%. © 2015 Elsevier Inc. All rights reserved.
The most common indications for revision of total hip arthroplasty are dislocation and aseptic loosening [1,2]. Multiple studies have correlated these complications to poor implant position [3–8]. Likewise, poorly positioned acetabular cups have been correlated with additional complications of impingement, edge loading, increased wear and pelvic osteolysis [6–13]. An analysis of large cohorts at both a tertiary and a community hospital found at least 50% of cups to be outside the safe zone of Lewinnek et al [12] for both inclination and anteversion [13]. There are similar data for cups placed to a target number for anteversion with outliers beyond 10° of 10–50% [14,15]. Instrumentation using computer navigation has improved the accuracy of component positioning by providing the surgeon quantitative knowledge to give greater precision for intraoperative decisions [14,6–20]. However, computerized instruments are still manually controlled, and manual reaming, even with computerized instruments, can have a difference between the planned and reamed center of rotation for the cup of 6.39 ± 2.44 mm [21]. Robotic computer guided instrumentation was designed to prevent the errors of manual reaming and cup implantation by providing a physical constraint to surgical tools by stereotactic boundaries, i.e., virtual walls, and thereby enabling accurate performance. During reaming, robotic instrumentation allows the surgeon to work within a virtual haptic tunnel, and a fail-safe mechanism stops the reamer if it exceeds the planned bone preparation in any plane by more than 2 mm. Cup impaction likewise is done through this haptic constrained tunnel so accuracy of inclination, anteversion and center of rotation can be achieved as planned. The Conflict of Interest statement associated with this article can be found at http://dx. doi.org/10.1016/j.arth.2014.10.021. Reprint requests: Lawrence D. Dorr, M.D., Keck Medical Center of USC, Orthopedic Department, 1520 San Pablo Street, Suite 2000, Los Angeles, CA 90033. http://dx.doi.org/10.1016/j.arth.2014.10.021 0883-5403/© 2015 Elsevier Inc. All rights reserved.
New technology introduced into clinical practice necessitates validation of its theoretical improvement. Our study was conducted to confirm that the software of this robotic system (MAKO-Stryker, Ft. Lauderdale, FL) performed as accurately and precisely as expected. Two questions were asked: (1) How accurate and precise was acetabular cup inclination and anteversion intraoperatively as compared to the preoperatively planned positions, and as validated by postoperative CT scans? (2) How often was the cup center of rotation (hip center of rotation) within 3 mm superior and 5 mm medial as measured on postoperative anteroposterior (AP) pelvic radiographs? Material and Methods This was an imaging study to validate a surgical technique. It was a prospective study of acetabular component position in primary total hip arthroplasty with the cup implanted using robotic guided instrumentation (MAKO Rio Robot, Ft. Lauderdale, FL). Forty-four patients (48 hips) agreed to participate in this prospective study approved by the institutional review board (IRB) and all patients gave informed consent. One hundred forty-six patients (162 hips) were operated by the surgeon (LDD) during the ten months of this study but only 75 of these patients were operated with the robot. Forty-four of these 75 patients agreed to the study with those who declined doing so because of the necessity of a postoperative CT scan. In three of the 44 patients (5 hips) the robotic arm could not function because it impinged on the tissues within the wound. Therefore, the study population is 40 patients with 43 hips (Table 1). In the 5 cups implanted manually we could still obtain quantitative measurement of their inclination and anteversion by a method named the Fitplane in which the pointer guide touched the metal edge of the cup at five different points, and
V. Kanawade et al. / The Journal of Arthroplasty 30 (2015) 392–397
393
Fig. 1. Computer screen during reaming. Green represents the area to be reamed; white is the area reamed to correct depth; red is the area reamed beyond planned depth. Irregularity of reaming by 1–2 mm is the reason the fail-safe mechanism does not activate until any area is overreamed by 2 mm. Vertical boxes show the remaining depth to be reamed to reconstruct the COR, and when the box turns green the depth is reached. The remaining posterior reaming matches the green area on the screen. The inclination and anteversion numbers can be within 10° of plan during reaming.
the robotic software could then calculate the inclination and anteversion which was displayed on the computer screen. With manual implantation the cup COR could not be determined. All surgeries were performed using the posterior mini-incision [22] by a single surgeon (LDD) with operative time from incision through closure of mean 85 ± 17 minutes. The patient was in the lateral position, and the pelvic array with reflective markers was attached to the pelvic rim with a baseplate secured with 1/8″ threaded pins. Inside the wound, a 4.5 mm screw was inserted in the posterior–superior pelvic bone 1 cm proximal to the acetabular rim, and it was touched with the array guide to confirm the authenticity of the robotic numbers. Surface registration of 32 points of the acetabulum and its rim registered the bony acetabulum into the software which matched it to the virtual 3D pelvis constructed from the CT scan. A registration error of the bony acetabulum of less than 0.5 mm
was accepted. Precise reaming is controlled by a stereotactic interface which restricts the reamers to a predefined volume of resection so that a line-to-line reaming was done for the planned cup size. A fail-safe mechanism stops the reamer if the COR in any direction is exceeded by 2 mm (Fig. 1). The cup was then impacted through the haptic tunnel until it was seated within 0–1 mm of the acetabular surface. The system error for cup inclination and anteversion was 5° and that is the reason that outliers were considered to be beyond 5°. A preoperative CT scan was used to plan the size and coverage of the cup in the acetabular bone as well as the COR in the cephalocaudad, medial–lateral, and anterior–posterior directions (Fig. 2). The same imaging scanner in the radiology department was used for both the preoperative and postoperative CT scans of the pelvis (64 channel multidetector CT, Brilliance 64, Phillips Medical Systems, Best, The
Fig. 2. Preoperative plan on CT scan. The inclination and anteversion, size and center of rotation (COR) of the planned acetabular cup are determined with the relationship of the cup COR (green) to the arthritic hip COR (magenta) displayed. The horizontal numbers show the planned COR in all planes with the horizontal COR 3 mm medial to the hip COR.
394
V. Kanawade et al. / The Journal of Arthroplasty 30 (2015) 392–397
Netherlands). The supine position of the patient was standardized with the longitudinal axis of the body parallel to the imaging table, and both knees and ankles aligned. The pelvic scan was performed at 1 mm intervals and 1 mm thickness with a field of view of 400 mm × 400 mm and pitch of 0.95 mm. The preoperative scan was sent to the MAKO facility for incorporation into the robotic software for that patient's operation (see appendix). The initial target numbers were usually 40° for acetabular inclination and 20° for anteversion in the radiographic plane of Murray [23], but either number could be changed intraoperatively if indicated. Because we used the combined anteversion method [24] it was more commonly intraoperative anteversion that was changed. Preoperative and postoperative anterior–posterior pelvic radiographs were obtained with the patient in the supine position and the hips in 10 to 15° of internal rotation with the x-ray beam centered over the symphysis pubis at a source to film distance of 52 inches. From the preoperative film the following measurements were made: Diagnosis of the causative disease; center of rotation of the operated acetabulum, and of the contralateral hip when normal (27 hips), (Fig. 3a). Two research physicians (VK and ZW) independently performed each measurement. Radiographic cup inclination was measured on the postoperative pelvic film using the method of Callaghan et al [25], and anteversion by the modified method of Ackland et al [26]. Postoperatively, films from the 6 week and 6 month followup visits of
the patient were reviewed and the cup COR measured on the film with the more correct rotation (Fig. 3b). The measurements were adjusted for magnification by use of a magnification marker placed on each film. The criteria of displacement of 3 mm superior and 5 mm medial was used as the safe zone of the COR [27]. A postoperative CT scan was used for validation of the robotic numbers for inclination and anteversion, and the measurements were done by two methods: (1) one method used a matching method for the acetabulae of the preoperative and postoperative scans, and this was done at the University of Florida Biomechanics Laboratory (see appendix). The second method was performed by our research physician (ZW) using the same method used to validate cup inclination and anteversion with computer navigation [14], (see appendix). Statistics For analysis of all measurements, the means and standard deviation were calculated. Statistical analysis was performed with SPSS software version 12, (SPSS Inc., Chicago, IL) and a P value of less than or equal to 0.05 was considered significant with a 95% confidence interval. For measurements of accuracy of the inclination and anteversion, the means with standard deviations are not as important as the precision and bias, which were calculated according to The American Society for Testing and Materials and definitions [28]. Precision is defined as a closeness of agreement between independent measurements (degrees in this study) and represents the reliability and reproducibility of the tests. Bias (systematic error) is the consistent difference between a set of measurements and an accepted reference or true value (CT scans) in this study. A low bias number means that the average of the test number and true value are very close. The repeatability between inclination and anteversion of the intraoperative numbers and the postoperative CT scan values was calculated using the intraclass correlation coefficient. The outliers were calculated as individual variations between intraoperative inclination or anteversion and the postoperative CT scan values beyond 5°. Five degrees is the system error of the MAKO robot. For analysis of hip COR measurements, the differences among preoperative, postoperative and contralateral hip measurements were tested with a paired Student t test (Table 4). Results The planned inclination for 43 hips (40 patients) was mean 39.9° ± 0.8°, (range, 35–40°) and that achieved by intraoperative robotic instrumentation was mean 38.8° ± 1.6°, (range, 35–43°). Postoperatively the CT inclination by the matched method was mean 39.1° ± 3.8 °, (range, 33–48°) and by the Wan method was mean 39.4° ± 3.4°, (range, 31–48°). The outliers between the planned and the robotic achieved implantation for both inclination and anteversion were zero; the postoperative CT validation of the intraoperative inclination was 5 hips (12%) between 5 and 10° with 0 outliers beyond 10° by Table 1 Demographics. Variable
Fig. 3. (a) AP pelvis radiograph showing the method of measurement of the center of rotation (COR) of the preoperative (A) and postoperative (B) hips. Line A is the transteardrop line; line B is perpendicular to line A drawn through the teardrop. Line X is parallel to line B drawn through the lateral edge of the acetabulum. Line Y is parallel to the transteardrop line drawn through the superior edge of the acetabulum (if this is destroyed, use the Ranawat triangle33 to find its original height). 2/3 of the horizontal distance from line B to line X, and 1/3 of the vertical distance from line A to line Y defines the hip COR. The horizontal length of the hip COR is the distance (mm) from the acetabular COR to line B; the vertical height of the hip COR is the distance (mm) from the acetabular COR to line A. (b) Postoperatively, the hip COR was measured by the best fit concentric circle.
Gender Male Female Age (years) Height (cm) Weight (kg) Body mass index Side Left Right Diagnosis Osteoarthritis Developmental dysplasia of hip
Value 18 (19 hips) 22 (24 hips) 63 (48-79) 172 (157–198) 81.3 (50–140) 27.3 (19.5–43.6) 17 26 40 3
V. Kanawade et al. / The Journal of Arthroplasty 30 (2015) 392–397 Table 2 Accuracy of Acetabular Inclination.
Inclination (°) Mean ± SD (Range) Precision Bias Intraclass coefficient a b
Robot 38.8 ± 1.6 (35.6–42.7)
Table 4 Cup Center of Rotation. CT Scan (UF Match)
CT Scan (Manual Match)
39.1 ± 3.8 (33.2–47.6) 7.4a 0.3a 0.331a
39.4 ± 3.4 (31.3–48.2) 6.5b 0.7b 0.342b
Robot vs. CT scan (UF match). Robot vs. CT scan (manual match).
both measuring methods we used. Accuracy (precision and bias) is shown in Table 2. The preoperatively planned anteversion was mean 21.2° ± 2.4°, (range 15–25°) and the intraoperative robotic implantation was mean 20.7° ± 2.4°, (range, 16–26°). The postoperative CT validation by the matched method was mean 18.9° ± 4.1°, (range, 7–26°) and by the Wan method was mean 19.1° ± 4.2°, (range, 5–27°). There were no outliers for the planned versus the intraoperatively achieved positions; but intraoperative anteversion versus that on postoperative CT scans had 7 hips (16%) with outliers between 5 and 10° and 0 outliers beyond 10°. Accuracy by precision and bias is shown in Table 3. The hip center of rotation was contained within a mean 3 mm in the vertical (superior) direction and mean 5 mm horizontally (medially) from the preoperative COR of the arthritic acetabulum (Table 4). In 27 patients with the contralateral hip being normal, there were 5 outliers (18.5%) of the reconstructed hip COR as compared to the true hip COR for both superior and horizontal measurements. Three of these 5 outliers were in dysplastic hips and 2 had a preoperative COR between 10 and 14 mm superior to the true center because of the acetabular geometric changes from arthritis. No complications were observed from the robotic technique.
Discussion For surgeons to use new technology for total hip arthroplasty with confidence requires that it be validated, and be shown to have the potential to improve their operative outcomes. Robotic instrumentation for acetabular bone preparation and cup implantation is such a new computerized tool. If it is to benefit the surgeon's performance it must be simple to use, and provide better precision and more accurate results than manual instrumentation. Two questions were asked in this study of the MAKO instrumentation: (1) Is cup inclination and anteversion accurate and precise? (2) Can the center of rotation of the hip be maintained within 3 mm superior and 5 mm medially? With computer navigation [14] we had precision of 4.4° for inclination and anteversion compared to precision of 7.4° for inclination and 7.3° for anteversion (UF match) with the robotic software in this study, so there was not improved precision with the robot. Compared to
Table 3 Accuracy of Acetabular Anteversion.
Anteversion (°) Mean ± SD (Range) Precision Bias Intraclass coefficient a b
Robot 20.7 ± 2.4 (15.5–25.7)
Robot vs. CT scan (UF match). Robot vs. CT scan (manual match).
395
CT Scan (UF Match)
CT Scan (Manual Match)
18.9 ± 4.1 (16.6–26.4) 7.3a 1.6a 0.423a
19.1 ± 4.2 (5–26.7) 7.3b 1.4b 0.429b
Horizontal (medial) COR Vertical (superior) COR
Pre-op (mm)
Post-op (mm)
Difference (mm)
32 ± 4.3 (24.1 to 45.2) 16.5 ± 5 (6.6 to 29)
29.3 ± 4.1 (21.6 to 40.8) 17 ± 3.5 (12.8 to 227.9)
2.7 ± 2.9 (−2.8 to 7.7) −0.9 ± 4.2 (−9 to 13.4)
manual implantation the robotic instrumentation did have fewer outliers: The inclination target number was missed by 6–10° in 25% of hips, and beyond 10° in 6% of hips [14] whereas with robotic instrumentation there were 5 (12%) outliers between 6 and 10° and zero beyond 10°. For anteversion with manual instrumentation the outliers (of the same surgeon LDD) were 29% of hips by 6–10° and 10% of hips beyond 10° [14], whereas with robotic instrumentation there were 7 hips (16%) with outliers between 6 and 10° and zero beyond 10°. The robotic instrumentation also repeatedly controlled the hip COR within 3 mm superiorly and 5 mm medially. There were limitations and the first was the necessity for a postoperative CT scan which deterred some patients from enrolling in the study. A second limitation is the lack of same study controls, but since the same surgeon had previously been tested for results with manual instrumentation to achieve a target number [14], we did not consider a repeat of this necessary for scientific comparison. The third limitation is that this is an imaging study to validate the performance of the robotic instrumentation and therefore does not answer the important question as to whether clinical results are as good or better than with manual implantation. The fourth limitation is that 5 hips did not have the robot employed because of mechanical impingement of the robot arm with tissues of the hip. This design deficiency has been rectified by retooling the robotic arm to have variable positions available to avoid tissue impingement. The importance of this study is the validation of instrumentation that provides the surgeon an option for accurate and precise implantation of the acetabular component. Several studies have tested cup positions by manual instrumentation and in 40 to 78% of hips the anteversion and/or inclination could not be maintained within the Lewinnek et al [12] safe zone of 20° [13,15,17,29]. This failure to achieve precision with uncemented cups, combined with the unpredictability of uncemented stem anteversion [30], does suggest a reason for the most prevalent complication of hip arthroplasty being dislocation [1,2]. Improved implant position has consistently given more predictable outcomes in published reports [3–8]. One study has shown that use of this same robotic instrumentation maintained the cup positions of inclination and anteversion within the safe zone of Lewinnek et al [12] in 50 of 50 hips (100%) but in only 40 of 50 hips (80%) with conventional acetabular cup placement [31]. One advantage of this robotic instrumentation over computer navigation was control of the hip COR. The fail-safe mechanism to prevent the planned COR being exceeded by 2 mm prevents overreaming, and is a distinct advantage to manual reaming of computer navigation that has been shown to have error that is a mean 6.4 mm [21]. Maintenance of the COR within the planned constraints is favorable for the correct biomechanical reconstruction [27] and the durability of the arthroplasty [32,33]. Unfortunately it is not usually possible to reconstruct the COR within its normal safe zone for hips with geometric deformity of dysplasia or post-trauma, and those with migration of the femoral head of 10 mm or more. In this study of a limited number of hips we were outside this safe zone in 18.5% of them. For robotic instrumentation to become an accepted and standard method of implantation of hip components will require clinical data that support improved clinical outcomes, and that data need to be forthcoming. This study simply provides the precision and accuracy of this machine in clinical use which provides a surgeon the necessary information to consider use of this technology.
396
V. Kanawade et al. / The Journal of Arthroplasty 30 (2015) 392–397
Acknowledgments The authors thank Patricia J. Paul for providing transcription support. Appendix A. MAKO transfer of acetabular anatomical data into software from CT scan The CT scan in DICOM (Digital Imaging Communication in Medicine) format is loaded on the MAKO hip application. With the MIMICS (software), independent segmentation, and a specific 3 dimensional bone model, is created using an anatomic coordinate system which has the preoperative CT table as the longitudinal axis. From this model is obtained the acetabular measurements (Fig. A-1). The origin of coordinates is placed at the acetabular cup center. The x-axis is the medial/lateral direction, y axis is the anterior/posterior direction and z-axis is the proximal/distal direction. This anatomic coordinate system is determined from the alignment of the pelvic anatomy. The medial/lateral axis (x-axis) is parallel to a line drawn through the right and left anterior superior iliac spines, which allows the pelvic tilt to be incorporated so the inclination and anteversion are on the radiographical plane of Murray [23]. Specifically, radiographic inclination is defined as the angle between the acetabular axis and the proximal/distal axis when projected onto the coronal plane (x-z plane); and radiographic anteversion as the angle between the acetabular axis and the coronal plane (x-z plane).
Fig. A-2. The postoperative pelvic model is registered to the preoperative pelvic model (orange), preserving the preoperative coronal plane. All measurements are recorded relative to the anatomic coordinates. In the standard measurement, full preoperative and postoperative pelvises without acetabulum are used to measure the cup position. But to distinguish the preoperative and postoperative pelvises in this figure, hemi-preoperative and full postoperative pelvises were used.
University of Florida Match of the Intraoperative Cup Inclination and Anteversion to the Postoperative CT Scan Acetabulum The University of Florida (UF) Biomechanics Laboratory constructed the pelvic bone models by segmenting the preoperative and postoperative CT scan images with an image intensity threshold filtering ITK-SNAP (www.itksnap.org) [35]. Computerized Assisted Design (CAD) models of cups in all sizes were provided by the manufacturer. The longitudinal axis and anatomic coordinate system is the same as described in the MAKO software technique. After the anatomic coordinate system was defined, the postoperative pelvis and cup were imported onto the preoperative pelvis [34]. Global registration, a function from Geomagic Studio (Geomagic Inc., Morrisville, NC) was used to move and match the full postoperative pelvis without the acetabulum to the preoperative pelvis without the acetabulum based on the 3D to 3D iterative closest point algorithm. The software quantified and recorded the displacements using the 4 × 4 homogenous transformation matrices (Fig. A-2). The center of the imported Computerized Assisted Design (CAD) cup model was moved to the origin of the anatomic coordinate system preserving the preoperative coronal plane (z-axis [proximal/distal] axis of acetabulum). The acetabular axis
was defined as perpendicular to the rim of the cup, passing through the center of the cup. Global registration was used to move the CAD cup model relative to the previously recorded 4 × 4 homogenous transformation matrices to match the postoperative cup model and the coordinates of the acetabular axis were obtained in this step. The radiographic inclination and anteversion were calculated [23]. Manual Match of the Intraoperative Cup Inclination and Anteversion to the Postoperative CT Scan The preoperative and postoperative 3D CT scan pelvic bone model was constructed using Amira (Amira 5.4.3 VSG, Burlington MA, US) (www.vsg3d.com). The same coordinate system used in the UF matching method was established for the preoperative and postoperative bone models, and various manual measurements done: The pelvic tilt (angle between anterior posterior plane (APP) and the coronal X2 plane), and the transverse tilt of the pelvis (angle between the transverse XY plane) and the line joining the bilateral ischial tuberosities). The postoperative 3D pelvic bone model was then transferred to global registration software (Geomagic Studio, Geomagic, Inc., Morrisville, NC), and manually adjusted to match the preoperative 3D pelvic bone model according to the pelvic tilt and the anteroposterior and mediolateral acetabular position preserving the anatomic coordinate system [14]. This model differed by using the Anterior Posterior Pelvic plane and the pelvic tilt. The postoperative cup plane was obtained by the best fit technique, and radiographic inclination and anteversion were calculated [23]. References
Fig. A-1. Anatomic coordinates were derived from the preoperative CT orientation and aligned to the medial- lateral axis formed by ASIS points. The x-axis (red) is the mediallateral direction and the z-axis (blue) is the proximal-distal direction. The y-axis (covered by the center) is the anterior–posterior direction, and is perpendicular to the image shown.
1. Ong KL, Mowat FS, Chan N, et al. Economic burden of revision hip and knee arthroplasty in Medicare enrollees. Clin Orthop Relat Res 2006;446:22. 2. Kurtz S, Mowat F, Ong K, et al. Prevalence of primary and revision total hip and knee arthroplasty in the United States from 1990 through 2002. J Bone Joint Surg Am 2005; 87(7):1487. 3. NJR National Joint Registry. Archived annual reports. Available at: http://www. njrcentre.org.uk/njrcentre/Reports,PublicationsandMinutes/Annualreports/ Archivedannualreports/tabid/87/Default.aspx. [Accessed July 24, 2013]. 4. Swedish Hip Arthroplasty Register. Publications. Avaliable at: http://www.shpr.se/en/ Publications.aspx. [Accessed July 24, 2013]. 5. Puolakka TJ, Pajamaki KJ, Halonen PJ, et al. The Finnish Arthroplasty Register: report of the hip register. Acta Orthop Scand 2001;72(5):433.
V. Kanawade et al. / The Journal of Arthroplasty 30 (2015) 392–397 6. Ulrich SD, Seyler TM, Bennett D, et al. Total hip arthroplasties: what are the reasons for revision? Int Orthop 2008;32(5):597. 7. Schmalzried TP, Guttmann D, Grecula M, et al. The relationship between the design, position, and articular wear of acetabular components inserted without cement and the development of pelvic osteolysis. J Bone Joint Surg Am 1994;76(5):677. 8. Kennedy JG, Rogers WB, Soffe KE, et al. Effect of acetabular component orientation on recurrent dislocation, pelvic osteolysis, polyethylene wear, and component migration. J Arthroplasty 1998;13(5):530. 9. Del Schutte Jr H, Lipman AJ, Bannar SM, et al. Effects of acetabular abduction on cup wear rates in total hip arthroplasty. J Arthroplasty 1998;13(6):621. 10. Patil S, Bergula A, Chen PC, et al. Polyethylene wear and acetabular component orientation. J Bone Joint Surg Am 2003;85-A(Suppl 4):56. 11. De Haan R, Pattyn C, Gill HS, et al. Correlation between inclination of the acetabular component and metal ion levels in metal-on-metal hip resurfacing replacement. J Bone Joint Surg (Br) 2008;90(10):1291. 12. Lewinnek GE, Lewis JL, Tarr R, et al. Dislocations after total hip-replacement arthroplasties. J Bone Joint Surg Am 1978;60(2):217. 13. Callanan MC, Jarrett B, Bragdon, et al. The John Charnley Award: risk factors for cup malpositioning: quality improvement through a joint registry at a tertiary hospital. Clin Orthop Relat Res 2011;469(2):319. 14. Dorr LD, Malik A, Wan Z, et al. Precision and bias of imageless computer navigation and surgeon estimates for acetabular component position. Clin Orthop Relat Res 2007;465:92. 15. Wines AP, McNicol D. Computed tomography measurement of the accuracy of component version in total hip arthroplasty. J Arthroplasty 2006;21(5):696. 16. Kalteis T, Handel M, Bathis H, et al. Imageless navigation for insertion of the acetabular component in total hip arthroplasty: is it as accurate as CT-based navigation? J Bone Joint Surg (Br) 2006;88(2):163. 17. Parratte S, Argenson JN. Validation and usefulness of a computer-assisted cuppositioning system in total hip arthroplasty. A prospective, randomized, controlled study. J Bone Joint Surg Am 2007;89(3):494. 18. Ybinger T, Kumpan W, Hoffart HE, et al. Accuracy of navigation-assisted acetabular component positioning studied by computed tomography measurements: methods and results. J Arthroplasty 2007;22(6):812. 19. Ryan JA, Jamali AA, Bargar WL. Accuracy of computer navigation for acetabular component placement in THA. Clin Orthop Relat Res 2010;468(1):169.
397
20. DiGioia III AM, Jaramaz B, Colgan BD. Computer assisted orthopaedic surgery. Image guided and robotic assistive technologies. Clin Orthop Relat Res 1998;354:8. 21. Honl M, Schwieger K, Salineros M, et al. Orientation of the acetabular component. A comparison of five navigation systems with conventional surgical technique. J Bone Joint Surg (Br) 2006;88(10):1401. 22. Inaba Y, Dorr LD, Wan Z, et al. Operative and patient care techniques for posterior mini-incision total hip arthroplasty. Clin Orthop Relat Res 2005;441:104. 23. Murray DW. The definition and measurement of acetabular orientation. J Bone Joint Surg (Br) 1993;75(2):228. 24. Dorr LD, Malik A, Dastane M, et al. Combined anteversion technique for total hip arthroplasty. Clin Orthop Relat Res 2009;467(1):119. 25. Callaghan JJ, Salvati EA, Pellicci PM, et al. Results of revision for mechanical failure after cemented total hip replacement, 1979 to 1982. A two to five-year follow-up. J Bone Joint Surg Am 1985;67(7):1074. 26. Ackland MK, Bourne WB, Uhthoff HK. Anteversion of the acetabular cup. Measurement of angle after total hip replacement. J Bone Joint Surg (Br) 1986;68(3):409. 27. Dastane M, Dorr LD, Tarwala R, et al. Hip offset in total hip arthroplasty: quantitative measurement with navigation. Clin Orthop Relat Res 2011;469(2):429. 28. ASTM. Annual Book of American Society for Testing and Materials Standard. West Conshohocken, PA: ASTM International; 2002. 29. DiGioia III AM, Jaramaz B, Plakseychuk AY, et al. Comparison of a mechanical acetabular alignment guide with computer placement of the socket. J Arthroplasty 2002;17(3):359. 30. Dorr LD, Wan Z, Malik A, et al. A comparison of surgeon estimation and computed tomographic measurement of femoral component anteversion in cementless total hip arthroplasty. J Bone Joint Surg Am 2009;91(11):2598. 31. Domb BG, El Bitar YR, Sadik AY, et al. Comparison of robotic-assisted and conventional acetabular cup placement in THA: a matched-pair controlled study. Clin Orthop Relat Res 2014;473:329. 32. Karachalios T, Hartofilakidis G, Zacharakis N, et al. A 12- to 18-year radiographic follow-up study of Charnley low-friction arthroplasty. The role of the center of rotation. Clin Orthop Relat Res 1993;296:140. 33. Ranawat CS, Dorr LD, Inglis AE. Total hip arthroplasty in protrusio acetabuli of rheumatoid arthritis. J Bone Joint Surg Am 1980;62(7):1059. 34. Yushkevich PA, Piven J, Hazlett HC, et al. User-guided 3D active contour segmentation of anatomical structures: significantly improved efficiency and reliability. Neuroimage 2006;31(3):1116.
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Precision of Robotic Guided Instrumentation for Acetabular Component Positioning Vaibhav Kanawade, MD a, Lawrence D. Dorr, MD b, Scott A. Banks, PhD c, Zenan Zhang, MS c, Zhinian Wan, MD b a b c
Kanawade Patil House, Sangamner, India Orthopedic Department, Keck Medical Center of USC, Los Angeles, California Department of Mechanical & Aerospace Engineering, University of Florida, Gainesville, Florida
a r t i c l e
i n f o
Article history: Received 24 August 2014 Accepted 11 October 2014 Keywords: computer robot THR cup position MAKO
a b s t r a c t Robotic computerized instrumentation that guides bone preparation and cup implantation in total hip arthroplasty was studied. In 38 patients (43 hips) intraoperative cup inclination and anteversion were validated by postoperative CT scans. Planned inclination was 39.9° ± 0.8° and with robotic instrumentation was 38. 0° ± 1.6° with no outliers of 5°; on the postoperative CT scan there were 5 outliers (12%). Planned anteversion was 21.2° ± 2.4° and intraoperatively was 20.7° ± 2.4° with no outlier of 5°; on the CT there were 7 outliers (16%). The center of rotation (COR) was superior by a mean 0.9 ± 4.2 mm and medial by 2.7 ± 2.9 mm. This robotic instrumentation achieved precision of inclination in 88%, anteversion in 84% and COR in 81.5%. © 2015 Elsevier Inc. All rights reserved.
The most common indications for revision of total hip arthroplasty are dislocation and aseptic loosening [1,2]. Multiple studies have correlated these complications to poor implant position [3–8]. Likewise, poorly positioned acetabular cups have been correlated with additional complications of impingement, edge loading, increased wear and pelvic osteolysis [6–13]. An analysis of large cohorts at both a tertiary and a community hospital found at least 50% of cups to be outside the safe zone of Lewinnek et al [12] for both inclination and anteversion [13]. There are similar data for cups placed to a target number for anteversion with outliers beyond 10° of 10–50% [14,15]. Instrumentation using computer navigation has improved the accuracy of component positioning by providing the surgeon quantitative knowledge to give greater precision for intraoperative decisions [14,6–20]. However, computerized instruments are still manually controlled, and manual reaming, even with computerized instruments, can have a difference between the planned and reamed center of rotation for the cup of 6.39 ± 2.44 mm [21]. Robotic computer guided instrumentation was designed to prevent the errors of manual reaming and cup implantation by providing a physical constraint to surgical tools by stereotactic boundaries, i.e., virtual walls, and thereby enabling accurate performance. During reaming, robotic instrumentation allows the surgeon to work within a virtual haptic tunnel, and a fail-safe mechanism stops the reamer if it exceeds the planned bone preparation in any plane by more than 2 mm. Cup impaction likewise is done through this haptic constrained tunnel so accuracy of inclination, anteversion and center of rotation can be achieved as planned. The Conflict of Interest statement associated with this article can be found at http://dx. doi.org/10.1016/j.arth.2014.10.021. Reprint requests: Lawrence D. Dorr, M.D., Keck Medical Center of USC, Orthopedic Department, 1520 San Pablo Street, Suite 2000, Los Angeles, CA 90033. http://dx.doi.org/10.1016/j.arth.2014.10.021 0883-5403/© 2015 Elsevier Inc. All rights reserved.
New technology introduced into clinical practice necessitates validation of its theoretical improvement. Our study was conducted to confirm that the software of this robotic system (MAKO-Stryker, Ft. Lauderdale, FL) performed as accurately and precisely as expected. Two questions were asked: (1) How accurate and precise was acetabular cup inclination and anteversion intraoperatively as compared to the preoperatively planned positions, and as validated by postoperative CT scans? (2) How often was the cup center of rotation (hip center of rotation) within 3 mm superior and 5 mm medial as measured on postoperative anteroposterior (AP) pelvic radiographs? Material and Methods This was an imaging study to validate a surgical technique. It was a prospective study of acetabular component position in primary total hip arthroplasty with the cup implanted using robotic guided instrumentation (MAKO Rio Robot, Ft. Lauderdale, FL). Forty-four patients (48 hips) agreed to participate in this prospective study approved by the institutional review board (IRB) and all patients gave informed consent. One hundred forty-six patients (162 hips) were operated by the surgeon (LDD) during the ten months of this study but only 75 of these patients were operated with the robot. Forty-four of these 75 patients agreed to the study with those who declined doing so because of the necessity of a postoperative CT scan. In three of the 44 patients (5 hips) the robotic arm could not function because it impinged on the tissues within the wound. Therefore, the study population is 40 patients with 43 hips (Table 1). In the 5 cups implanted manually we could still obtain quantitative measurement of their inclination and anteversion by a method named the Fitplane in which the pointer guide touched the metal edge of the cup at five different points, and
V. Kanawade et al. / The Journal of Arthroplasty 30 (2015) 392–397
393
Fig. 1. Computer screen during reaming. Green represents the area to be reamed; white is the area reamed to correct depth; red is the area reamed beyond planned depth. Irregularity of reaming by 1–2 mm is the reason the fail-safe mechanism does not activate until any area is overreamed by 2 mm. Vertical boxes show the remaining depth to be reamed to reconstruct the COR, and when the box turns green the depth is reached. The remaining posterior reaming matches the green area on the screen. The inclination and anteversion numbers can be within 10° of plan during reaming.
the robotic software could then calculate the inclination and anteversion which was displayed on the computer screen. With manual implantation the cup COR could not be determined. All surgeries were performed using the posterior mini-incision [22] by a single surgeon (LDD) with operative time from incision through closure of mean 85 ± 17 minutes. The patient was in the lateral position, and the pelvic array with reflective markers was attached to the pelvic rim with a baseplate secured with 1/8″ threaded pins. Inside the wound, a 4.5 mm screw was inserted in the posterior–superior pelvic bone 1 cm proximal to the acetabular rim, and it was touched with the array guide to confirm the authenticity of the robotic numbers. Surface registration of 32 points of the acetabulum and its rim registered the bony acetabulum into the software which matched it to the virtual 3D pelvis constructed from the CT scan. A registration error of the bony acetabulum of less than 0.5 mm
was accepted. Precise reaming is controlled by a stereotactic interface which restricts the reamers to a predefined volume of resection so that a line-to-line reaming was done for the planned cup size. A fail-safe mechanism stops the reamer if the COR in any direction is exceeded by 2 mm (Fig. 1). The cup was then impacted through the haptic tunnel until it was seated within 0–1 mm of the acetabular surface. The system error for cup inclination and anteversion was 5° and that is the reason that outliers were considered to be beyond 5°. A preoperative CT scan was used to plan the size and coverage of the cup in the acetabular bone as well as the COR in the cephalocaudad, medial–lateral, and anterior–posterior directions (Fig. 2). The same imaging scanner in the radiology department was used for both the preoperative and postoperative CT scans of the pelvis (64 channel multidetector CT, Brilliance 64, Phillips Medical Systems, Best, The
Fig. 2. Preoperative plan on CT scan. The inclination and anteversion, size and center of rotation (COR) of the planned acetabular cup are determined with the relationship of the cup COR (green) to the arthritic hip COR (magenta) displayed. The horizontal numbers show the planned COR in all planes with the horizontal COR 3 mm medial to the hip COR.
394
V. Kanawade et al. / The Journal of Arthroplasty 30 (2015) 392–397
Netherlands). The supine position of the patient was standardized with the longitudinal axis of the body parallel to the imaging table, and both knees and ankles aligned. The pelvic scan was performed at 1 mm intervals and 1 mm thickness with a field of view of 400 mm × 400 mm and pitch of 0.95 mm. The preoperative scan was sent to the MAKO facility for incorporation into the robotic software for that patient's operation (see appendix). The initial target numbers were usually 40° for acetabular inclination and 20° for anteversion in the radiographic plane of Murray [23], but either number could be changed intraoperatively if indicated. Because we used the combined anteversion method [24] it was more commonly intraoperative anteversion that was changed. Preoperative and postoperative anterior–posterior pelvic radiographs were obtained with the patient in the supine position and the hips in 10 to 15° of internal rotation with the x-ray beam centered over the symphysis pubis at a source to film distance of 52 inches. From the preoperative film the following measurements were made: Diagnosis of the causative disease; center of rotation of the operated acetabulum, and of the contralateral hip when normal (27 hips), (Fig. 3a). Two research physicians (VK and ZW) independently performed each measurement. Radiographic cup inclination was measured on the postoperative pelvic film using the method of Callaghan et al [25], and anteversion by the modified method of Ackland et al [26]. Postoperatively, films from the 6 week and 6 month followup visits of
the patient were reviewed and the cup COR measured on the film with the more correct rotation (Fig. 3b). The measurements were adjusted for magnification by use of a magnification marker placed on each film. The criteria of displacement of 3 mm superior and 5 mm medial was used as the safe zone of the COR [27]. A postoperative CT scan was used for validation of the robotic numbers for inclination and anteversion, and the measurements were done by two methods: (1) one method used a matching method for the acetabulae of the preoperative and postoperative scans, and this was done at the University of Florida Biomechanics Laboratory (see appendix). The second method was performed by our research physician (ZW) using the same method used to validate cup inclination and anteversion with computer navigation [14], (see appendix). Statistics For analysis of all measurements, the means and standard deviation were calculated. Statistical analysis was performed with SPSS software version 12, (SPSS Inc., Chicago, IL) and a P value of less than or equal to 0.05 was considered significant with a 95% confidence interval. For measurements of accuracy of the inclination and anteversion, the means with standard deviations are not as important as the precision and bias, which were calculated according to The American Society for Testing and Materials and definitions [28]. Precision is defined as a closeness of agreement between independent measurements (degrees in this study) and represents the reliability and reproducibility of the tests. Bias (systematic error) is the consistent difference between a set of measurements and an accepted reference or true value (CT scans) in this study. A low bias number means that the average of the test number and true value are very close. The repeatability between inclination and anteversion of the intraoperative numbers and the postoperative CT scan values was calculated using the intraclass correlation coefficient. The outliers were calculated as individual variations between intraoperative inclination or anteversion and the postoperative CT scan values beyond 5°. Five degrees is the system error of the MAKO robot. For analysis of hip COR measurements, the differences among preoperative, postoperative and contralateral hip measurements were tested with a paired Student t test (Table 4). Results The planned inclination for 43 hips (40 patients) was mean 39.9° ± 0.8°, (range, 35–40°) and that achieved by intraoperative robotic instrumentation was mean 38.8° ± 1.6°, (range, 35–43°). Postoperatively the CT inclination by the matched method was mean 39.1° ± 3.8 °, (range, 33–48°) and by the Wan method was mean 39.4° ± 3.4°, (range, 31–48°). The outliers between the planned and the robotic achieved implantation for both inclination and anteversion were zero; the postoperative CT validation of the intraoperative inclination was 5 hips (12%) between 5 and 10° with 0 outliers beyond 10° by Table 1 Demographics. Variable
Fig. 3. (a) AP pelvis radiograph showing the method of measurement of the center of rotation (COR) of the preoperative (A) and postoperative (B) hips. Line A is the transteardrop line; line B is perpendicular to line A drawn through the teardrop. Line X is parallel to line B drawn through the lateral edge of the acetabulum. Line Y is parallel to the transteardrop line drawn through the superior edge of the acetabulum (if this is destroyed, use the Ranawat triangle33 to find its original height). 2/3 of the horizontal distance from line B to line X, and 1/3 of the vertical distance from line A to line Y defines the hip COR. The horizontal length of the hip COR is the distance (mm) from the acetabular COR to line B; the vertical height of the hip COR is the distance (mm) from the acetabular COR to line A. (b) Postoperatively, the hip COR was measured by the best fit concentric circle.
Gender Male Female Age (years) Height (cm) Weight (kg) Body mass index Side Left Right Diagnosis Osteoarthritis Developmental dysplasia of hip
Value 18 (19 hips) 22 (24 hips) 63 (48-79) 172 (157–198) 81.3 (50–140) 27.3 (19.5–43.6) 17 26 40 3
V. Kanawade et al. / The Journal of Arthroplasty 30 (2015) 392–397 Table 2 Accuracy of Acetabular Inclination.
Inclination (°) Mean ± SD (Range) Precision Bias Intraclass coefficient a b
Robot 38.8 ± 1.6 (35.6–42.7)
Table 4 Cup Center of Rotation. CT Scan (UF Match)
CT Scan (Manual Match)
39.1 ± 3.8 (33.2–47.6) 7.4a 0.3a 0.331a
39.4 ± 3.4 (31.3–48.2) 6.5b 0.7b 0.342b
Robot vs. CT scan (UF match). Robot vs. CT scan (manual match).
both measuring methods we used. Accuracy (precision and bias) is shown in Table 2. The preoperatively planned anteversion was mean 21.2° ± 2.4°, (range 15–25°) and the intraoperative robotic implantation was mean 20.7° ± 2.4°, (range, 16–26°). The postoperative CT validation by the matched method was mean 18.9° ± 4.1°, (range, 7–26°) and by the Wan method was mean 19.1° ± 4.2°, (range, 5–27°). There were no outliers for the planned versus the intraoperatively achieved positions; but intraoperative anteversion versus that on postoperative CT scans had 7 hips (16%) with outliers between 5 and 10° and 0 outliers beyond 10°. Accuracy by precision and bias is shown in Table 3. The hip center of rotation was contained within a mean 3 mm in the vertical (superior) direction and mean 5 mm horizontally (medially) from the preoperative COR of the arthritic acetabulum (Table 4). In 27 patients with the contralateral hip being normal, there were 5 outliers (18.5%) of the reconstructed hip COR as compared to the true hip COR for both superior and horizontal measurements. Three of these 5 outliers were in dysplastic hips and 2 had a preoperative COR between 10 and 14 mm superior to the true center because of the acetabular geometric changes from arthritis. No complications were observed from the robotic technique.
Discussion For surgeons to use new technology for total hip arthroplasty with confidence requires that it be validated, and be shown to have the potential to improve their operative outcomes. Robotic instrumentation for acetabular bone preparation and cup implantation is such a new computerized tool. If it is to benefit the surgeon's performance it must be simple to use, and provide better precision and more accurate results than manual instrumentation. Two questions were asked in this study of the MAKO instrumentation: (1) Is cup inclination and anteversion accurate and precise? (2) Can the center of rotation of the hip be maintained within 3 mm superior and 5 mm medially? With computer navigation [14] we had precision of 4.4° for inclination and anteversion compared to precision of 7.4° for inclination and 7.3° for anteversion (UF match) with the robotic software in this study, so there was not improved precision with the robot. Compared to
Table 3 Accuracy of Acetabular Anteversion.
Anteversion (°) Mean ± SD (Range) Precision Bias Intraclass coefficient a b
Robot 20.7 ± 2.4 (15.5–25.7)
Robot vs. CT scan (UF match). Robot vs. CT scan (manual match).
395
CT Scan (UF Match)
CT Scan (Manual Match)
18.9 ± 4.1 (16.6–26.4) 7.3a 1.6a 0.423a
19.1 ± 4.2 (5–26.7) 7.3b 1.4b 0.429b
Horizontal (medial) COR Vertical (superior) COR
Pre-op (mm)
Post-op (mm)
Difference (mm)
32 ± 4.3 (24.1 to 45.2) 16.5 ± 5 (6.6 to 29)
29.3 ± 4.1 (21.6 to 40.8) 17 ± 3.5 (12.8 to 227.9)
2.7 ± 2.9 (−2.8 to 7.7) −0.9 ± 4.2 (−9 to 13.4)
manual implantation the robotic instrumentation did have fewer outliers: The inclination target number was missed by 6–10° in 25% of hips, and beyond 10° in 6% of hips [14] whereas with robotic instrumentation there were 5 (12%) outliers between 6 and 10° and zero beyond 10°. For anteversion with manual instrumentation the outliers (of the same surgeon LDD) were 29% of hips by 6–10° and 10% of hips beyond 10° [14], whereas with robotic instrumentation there were 7 hips (16%) with outliers between 6 and 10° and zero beyond 10°. The robotic instrumentation also repeatedly controlled the hip COR within 3 mm superiorly and 5 mm medially. There were limitations and the first was the necessity for a postoperative CT scan which deterred some patients from enrolling in the study. A second limitation is the lack of same study controls, but since the same surgeon had previously been tested for results with manual instrumentation to achieve a target number [14], we did not consider a repeat of this necessary for scientific comparison. The third limitation is that this is an imaging study to validate the performance of the robotic instrumentation and therefore does not answer the important question as to whether clinical results are as good or better than with manual implantation. The fourth limitation is that 5 hips did not have the robot employed because of mechanical impingement of the robot arm with tissues of the hip. This design deficiency has been rectified by retooling the robotic arm to have variable positions available to avoid tissue impingement. The importance of this study is the validation of instrumentation that provides the surgeon an option for accurate and precise implantation of the acetabular component. Several studies have tested cup positions by manual instrumentation and in 40 to 78% of hips the anteversion and/or inclination could not be maintained within the Lewinnek et al [12] safe zone of 20° [13,15,17,29]. This failure to achieve precision with uncemented cups, combined with the unpredictability of uncemented stem anteversion [30], does suggest a reason for the most prevalent complication of hip arthroplasty being dislocation [1,2]. Improved implant position has consistently given more predictable outcomes in published reports [3–8]. One study has shown that use of this same robotic instrumentation maintained the cup positions of inclination and anteversion within the safe zone of Lewinnek et al [12] in 50 of 50 hips (100%) but in only 40 of 50 hips (80%) with conventional acetabular cup placement [31]. One advantage of this robotic instrumentation over computer navigation was control of the hip COR. The fail-safe mechanism to prevent the planned COR being exceeded by 2 mm prevents overreaming, and is a distinct advantage to manual reaming of computer navigation that has been shown to have error that is a mean 6.4 mm [21]. Maintenance of the COR within the planned constraints is favorable for the correct biomechanical reconstruction [27] and the durability of the arthroplasty [32,33]. Unfortunately it is not usually possible to reconstruct the COR within its normal safe zone for hips with geometric deformity of dysplasia or post-trauma, and those with migration of the femoral head of 10 mm or more. In this study of a limited number of hips we were outside this safe zone in 18.5% of them. For robotic instrumentation to become an accepted and standard method of implantation of hip components will require clinical data that support improved clinical outcomes, and that data need to be forthcoming. This study simply provides the precision and accuracy of this machine in clinical use which provides a surgeon the necessary information to consider use of this technology.
396
V. Kanawade et al. / The Journal of Arthroplasty 30 (2015) 392–397
Acknowledgments The authors thank Patricia J. Paul for providing transcription support. Appendix A. MAKO transfer of acetabular anatomical data into software from CT scan The CT scan in DICOM (Digital Imaging Communication in Medicine) format is loaded on the MAKO hip application. With the MIMICS (software), independent segmentation, and a specific 3 dimensional bone model, is created using an anatomic coordinate system which has the preoperative CT table as the longitudinal axis. From this model is obtained the acetabular measurements (Fig. A-1). The origin of coordinates is placed at the acetabular cup center. The x-axis is the medial/lateral direction, y axis is the anterior/posterior direction and z-axis is the proximal/distal direction. This anatomic coordinate system is determined from the alignment of the pelvic anatomy. The medial/lateral axis (x-axis) is parallel to a line drawn through the right and left anterior superior iliac spines, which allows the pelvic tilt to be incorporated so the inclination and anteversion are on the radiographical plane of Murray [23]. Specifically, radiographic inclination is defined as the angle between the acetabular axis and the proximal/distal axis when projected onto the coronal plane (x-z plane); and radiographic anteversion as the angle between the acetabular axis and the coronal plane (x-z plane).
Fig. A-2. The postoperative pelvic model is registered to the preoperative pelvic model (orange), preserving the preoperative coronal plane. All measurements are recorded relative to the anatomic coordinates. In the standard measurement, full preoperative and postoperative pelvises without acetabulum are used to measure the cup position. But to distinguish the preoperative and postoperative pelvises in this figure, hemi-preoperative and full postoperative pelvises were used.
University of Florida Match of the Intraoperative Cup Inclination and Anteversion to the Postoperative CT Scan Acetabulum The University of Florida (UF) Biomechanics Laboratory constructed the pelvic bone models by segmenting the preoperative and postoperative CT scan images with an image intensity threshold filtering ITK-SNAP (www.itksnap.org) [35]. Computerized Assisted Design (CAD) models of cups in all sizes were provided by the manufacturer. The longitudinal axis and anatomic coordinate system is the same as described in the MAKO software technique. After the anatomic coordinate system was defined, the postoperative pelvis and cup were imported onto the preoperative pelvis [34]. Global registration, a function from Geomagic Studio (Geomagic Inc., Morrisville, NC) was used to move and match the full postoperative pelvis without the acetabulum to the preoperative pelvis without the acetabulum based on the 3D to 3D iterative closest point algorithm. The software quantified and recorded the displacements using the 4 × 4 homogenous transformation matrices (Fig. A-2). The center of the imported Computerized Assisted Design (CAD) cup model was moved to the origin of the anatomic coordinate system preserving the preoperative coronal plane (z-axis [proximal/distal] axis of acetabulum). The acetabular axis
was defined as perpendicular to the rim of the cup, passing through the center of the cup. Global registration was used to move the CAD cup model relative to the previously recorded 4 × 4 homogenous transformation matrices to match the postoperative cup model and the coordinates of the acetabular axis were obtained in this step. The radiographic inclination and anteversion were calculated [23]. Manual Match of the Intraoperative Cup Inclination and Anteversion to the Postoperative CT Scan The preoperative and postoperative 3D CT scan pelvic bone model was constructed using Amira (Amira 5.4.3 VSG, Burlington MA, US) (www.vsg3d.com). The same coordinate system used in the UF matching method was established for the preoperative and postoperative bone models, and various manual measurements done: The pelvic tilt (angle between anterior posterior plane (APP) and the coronal X2 plane), and the transverse tilt of the pelvis (angle between the transverse XY plane) and the line joining the bilateral ischial tuberosities). The postoperative 3D pelvic bone model was then transferred to global registration software (Geomagic Studio, Geomagic, Inc., Morrisville, NC), and manually adjusted to match the preoperative 3D pelvic bone model according to the pelvic tilt and the anteroposterior and mediolateral acetabular position preserving the anatomic coordinate system [14]. This model differed by using the Anterior Posterior Pelvic plane and the pelvic tilt. The postoperative cup plane was obtained by the best fit technique, and radiographic inclination and anteversion were calculated [23]. References
Fig. A-1. Anatomic coordinates were derived from the preoperative CT orientation and aligned to the medial- lateral axis formed by ASIS points. The x-axis (red) is the mediallateral direction and the z-axis (blue) is the proximal-distal direction. The y-axis (covered by the center) is the anterior–posterior direction, and is perpendicular to the image shown.
1. Ong KL, Mowat FS, Chan N, et al. Economic burden of revision hip and knee arthroplasty in Medicare enrollees. Clin Orthop Relat Res 2006;446:22. 2. Kurtz S, Mowat F, Ong K, et al. Prevalence of primary and revision total hip and knee arthroplasty in the United States from 1990 through 2002. J Bone Joint Surg Am 2005; 87(7):1487. 3. NJR National Joint Registry. Archived annual reports. Available at: http://www. njrcentre.org.uk/njrcentre/Reports,PublicationsandMinutes/Annualreports/ Archivedannualreports/tabid/87/Default.aspx. [Accessed July 24, 2013]. 4. Swedish Hip Arthroplasty Register. Publications. Avaliable at: http://www.shpr.se/en/ Publications.aspx. [Accessed July 24, 2013]. 5. Puolakka TJ, Pajamaki KJ, Halonen PJ, et al. The Finnish Arthroplasty Register: report of the hip register. Acta Orthop Scand 2001;72(5):433.
V. Kanawade et al. / The Journal of Arthroplasty 30 (2015) 392–397 6. Ulrich SD, Seyler TM, Bennett D, et al. Total hip arthroplasties: what are the reasons for revision? Int Orthop 2008;32(5):597. 7. Schmalzried TP, Guttmann D, Grecula M, et al. The relationship between the design, position, and articular wear of acetabular components inserted without cement and the development of pelvic osteolysis. J Bone Joint Surg Am 1994;76(5):677. 8. Kennedy JG, Rogers WB, Soffe KE, et al. Effect of acetabular component orientation on recurrent dislocation, pelvic osteolysis, polyethylene wear, and component migration. J Arthroplasty 1998;13(5):530. 9. Del Schutte Jr H, Lipman AJ, Bannar SM, et al. Effects of acetabular abduction on cup wear rates in total hip arthroplasty. J Arthroplasty 1998;13(6):621. 10. Patil S, Bergula A, Chen PC, et al. Polyethylene wear and acetabular component orientation. J Bone Joint Surg Am 2003;85-A(Suppl 4):56. 11. De Haan R, Pattyn C, Gill HS, et al. Correlation between inclination of the acetabular component and metal ion levels in metal-on-metal hip resurfacing replacement. J Bone Joint Surg (Br) 2008;90(10):1291. 12. Lewinnek GE, Lewis JL, Tarr R, et al. Dislocations after total hip-replacement arthroplasties. J Bone Joint Surg Am 1978;60(2):217. 13. Callanan MC, Jarrett B, Bragdon, et al. The John Charnley Award: risk factors for cup malpositioning: quality improvement through a joint registry at a tertiary hospital. Clin Orthop Relat Res 2011;469(2):319. 14. Dorr LD, Malik A, Wan Z, et al. Precision and bias of imageless computer navigation and surgeon estimates for acetabular component position. Clin Orthop Relat Res 2007;465:92. 15. Wines AP, McNicol D. Computed tomography measurement of the accuracy of component version in total hip arthroplasty. J Arthroplasty 2006;21(5):696. 16. Kalteis T, Handel M, Bathis H, et al. Imageless navigation for insertion of the acetabular component in total hip arthroplasty: is it as accurate as CT-based navigation? J Bone Joint Surg (Br) 2006;88(2):163. 17. Parratte S, Argenson JN. Validation and usefulness of a computer-assisted cuppositioning system in total hip arthroplasty. A prospective, randomized, controlled study. J Bone Joint Surg Am 2007;89(3):494. 18. Ybinger T, Kumpan W, Hoffart HE, et al. Accuracy of navigation-assisted acetabular component positioning studied by computed tomography measurements: methods and results. J Arthroplasty 2007;22(6):812. 19. Ryan JA, Jamali AA, Bargar WL. Accuracy of computer navigation for acetabular component placement in THA. Clin Orthop Relat Res 2010;468(1):169.
397
20. DiGioia III AM, Jaramaz B, Colgan BD. Computer assisted orthopaedic surgery. Image guided and robotic assistive technologies. Clin Orthop Relat Res 1998;354:8. 21. Honl M, Schwieger K, Salineros M, et al. Orientation of the acetabular component. A comparison of five navigation systems with conventional surgical technique. J Bone Joint Surg (Br) 2006;88(10):1401. 22. Inaba Y, Dorr LD, Wan Z, et al. Operative and patient care techniques for posterior mini-incision total hip arthroplasty. Clin Orthop Relat Res 2005;441:104. 23. Murray DW. The definition and measurement of acetabular orientation. J Bone Joint Surg (Br) 1993;75(2):228. 24. Dorr LD, Malik A, Dastane M, et al. Combined anteversion technique for total hip arthroplasty. Clin Orthop Relat Res 2009;467(1):119. 25. Callaghan JJ, Salvati EA, Pellicci PM, et al. Results of revision for mechanical failure after cemented total hip replacement, 1979 to 1982. A two to five-year follow-up. J Bone Joint Surg Am 1985;67(7):1074. 26. Ackland MK, Bourne WB, Uhthoff HK. Anteversion of the acetabular cup. Measurement of angle after total hip replacement. J Bone Joint Surg (Br) 1986;68(3):409. 27. Dastane M, Dorr LD, Tarwala R, et al. Hip offset in total hip arthroplasty: quantitative measurement with navigation. Clin Orthop Relat Res 2011;469(2):429. 28. ASTM. Annual Book of American Society for Testing and Materials Standard. West Conshohocken, PA: ASTM International; 2002. 29. DiGioia III AM, Jaramaz B, Plakseychuk AY, et al. Comparison of a mechanical acetabular alignment guide with computer placement of the socket. J Arthroplasty 2002;17(3):359. 30. Dorr LD, Wan Z, Malik A, et al. A comparison of surgeon estimation and computed tomographic measurement of femoral component anteversion in cementless total hip arthroplasty. J Bone Joint Surg Am 2009;91(11):2598. 31. Domb BG, El Bitar YR, Sadik AY, et al. Comparison of robotic-assisted and conventional acetabular cup placement in THA: a matched-pair controlled study. Clin Orthop Relat Res 2014;473:329. 32. Karachalios T, Hartofilakidis G, Zacharakis N, et al. A 12- to 18-year radiographic follow-up study of Charnley low-friction arthroplasty. The role of the center of rotation. Clin Orthop Relat Res 1993;296:140. 33. Ranawat CS, Dorr LD, Inglis AE. Total hip arthroplasty in protrusio acetabuli of rheumatoid arthritis. J Bone Joint Surg Am 1980;62(7):1059. 34. Yushkevich PA, Piven J, Hazlett HC, et al. User-guided 3D active contour segmentation of anatomical structures: significantly improved efficiency and reliability. Neuroimage 2006;31(3):1116.