- Email: [email protected]
Anterior cruciate ligament tunnel placement: Comparison of insertion site anatomy with the guidelines of a computer-assisted surgical system
Anterior Cruciate Ligament Tunnel Placement: Comparison of Insertion Site Anatomy With the Guidelines of a Computer-Assisted Surgical System Volker Musahl, M.D., Andreas Burkart, M.D., Richard E. Debski, Ph.D., Andrew Van Scyoc, B.S., Freddie H. Fu, M.D., and Savio L-Y. Woo, Ph.D.
Purpose: With the development of computer-assisted surgery (CAS) systems, the surgeon’s ability to operate a CAS planning station will become essential. For example, default parameters in computed tomographic (CT) data are being used to place tunnels in anterior cruciate ligament (ACL) reconstruction. The goal of this study was to compare the location of the insertion sites in ACL reconstruction anatomically, via roentgenographic images and via CT scan data and to validate these tunnel placement parameters. Type of Study: Cadaveric analysis. Methods: Eight human cadaveric knees were marked with 6 copper wires 1 mm in diameter around the circumference of the insertions of the ACL. Using lateral roentgenograms and CT scans that were subsequently transferred to the CAS planning station, the tunnel locations were determined. These were based on a distance from the back of the condyle (location A) and from the roof of the notch (location B) on the femur and on a distance posterior from the tuberosity to the posterior margin along the tibial plateau, which is set as the CAS planning station’s default. Locations according to roentgenograms and CT scans were then compared and the accuracy of the CAS planning station was assessed. Results: Comparison of roentgenograms and CT revealed a femoral insertion at 27.5% ⫾ 3.2% and 26.9% ⫾ 3.5% (roentgenograms) and 26.6% ⫾ 1.9% and 26.3% ⫾ 2.4% (CT), respectively. The CAS planning station provided a tunnel location that was 1.3 ⫾ 1.0 mm (0.3 to 2.5 mm) away from the actual femoral ACL insertion. The tibial tunnel was placed according to the copper wire markers and was found to be at 46.2% ⫾ 2.8% (roentgenograms) and 45.4% ⫾ 2.1% (CT). No statistical differences between position in CT and roentgenograms could be detected (P ⬎ .05). Conclusions: The compared methodologies showed similar locations of the ACL insertions, assuring accurate preoperative planning with the CAS system. However, the CAS system requires adjustment to each individual knee anatomy. Key Words: Computer assisted orthopaedic surgery—ACL—Planning station—Tunnel placement—Insertion site.
S
ignificant interest has been shown in the development of computer-assisted surgery (CAS) as having the potential for improving surgical management
From the Department of Orthopaedic Surgery, Musculoskeletal Research Center, Pittsburgh, Pennsylvania, U.S.A.; and the Department of Orthopaedic Sports Medicine, Technical University of Munich, Germany (A.B.). Address correspondence and reprint requests to Freddie H. Fu, M.D., Department of Orthopaedic Surgery, Musculoskeletal Research Center, E 1641 BST, 210 Lothrop St, P.O. Box 71199, Pittsburgh, PA 15213, U.S.A. E-mail: [email protected] © 2003 by the Arthroscopy Association of North America 0749-8063/03/1902-3185$30.00/0 doi:10.1053/jars.2003.50001
154
of anterior cruciate ligament (ACL) deficiency.1-3 Specifically, accurate tunnel placement is expected to help resolve inconsistencies in reconstruction techniques. Therefore, the location of the insertions of the ACL on the tibia and femur is a critical factor in successful surgical reconstruction. Despite extensive reports on ACL anatomy,4-8 approximately 10% to 40% of tunnels for ACL reconstructions have been misplaced.9,10. During arthroscopic surgery, guides11 and fluoroscopically assisted techniques12 have been used to increase the accuracy of graft placement. Furthermore, active and passive CAS systems have been intro-
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 19, No 2 (February), 2003: pp 154-160
ACL TUNNEL PLACEMENT: ANATOMY AND CAS GUIDELINES duced. Passive systems generally perform no action and assist intraoperatively.13-17 Active CAS systems, conversely, perform parts of the surgical procedure autonomously.18 However, surgeons need to program (using image-based methods) these active robots before surgery. The geometry of the ACL insertions has been extensively described. For the nearly round femoral insertion located at the medial aspect of the lateral femoral condyle, the clock position method is the most widely used. The roof of the notch thereby serves as the 12:00 o’clock reference position. This is called Blumensaat’s line as viewed in a lateral roentgenogram. A 10:30 to 11:00 o’clock position for a right knee (and a corresponding 1:00- to 1:30 o’clock position for a left knee) is cited as the tunnel entrance position.19,20 In the other dimension, 7-mm offset guides are usually used to place a tunnel anterior to the posterior margin of the femoral condyle (the over the top position).21,22 Bernard et al.5 proposed a quadrant method that is based on roentgenograms. Assessment of tunnel placement can be achieved intraoperatively using K-wires and fluoroscopy. Although the tibial tunnel placement is generally easier to achieve repeatedly,23 the exact location for a smaller diameter graft (such as 10-mm diameter) within the insertion site is difficult to determine. However, the surgical description of the tibial insertion of the ACL is 7 mm anterior to the posterior cruciate ligament (PCL) and on a line between the anterior border of the posterior horn of the lateral meniscus and the posterior border of the anterior horn of the medial meniscus, intersecting the tibial spine.24 Radiographically, the location of the tibial tunnel location was defined using the anteroposterior diameter of the medial tibial plateau as 43% from the anterior edge.25 Consequently, an unsatisfactory ACL reconstruction outcome can result from errors in surgical technique.26 Because one possible solution to this problem is the integration of CAS into orthopedic surgery, the feasibility and accuracy of such a system needs to be validated. Characteristically, CAS systems require understanding the relationships between CT and bone surface topography. A surgeon who uses arthroscopy to view the notch now needs to interpret the 2-dimensional CT image of this region. Therefore, the goal of this study was to compare the location of the insertion sites of the ACL in anatomy, roentgenographic images, and CT scan data and to validate the guidelines of a CAS planning station for its use of tunnel placement parameters for ACL reconstruction with respect
155
FIGURE 1. Lateral roentgenogram of the femur showing the geometric center of the ACL (x), with respect to its distance from Blumensaat’s line (A1) and from the posterior margin of the femoral condyle (B1), calculated with the quadrant method.5 Width (A2) and depth (B2) of the femoral condyle are shown.
to Blumensaat’s line, the femoral condyles, and the tibial plateau (default). METHODS Eight fresh-frozen human cadaveric knees (aged 50 to 79 years) were tested. The specimens were stored at ⫺20°C and thawed at room temperature for 24 hours before testing. The knees were then dissected free of musculature and capsular tissue, and the infrapatellar fat pat (Hoffa fat pat) and synovial tissue surrounding the ACL were dissected leaving the ACL, PCL, and the medial and lateral structures intact. Six 1-mm diameter copper wires, 6 mm in length, were inserted around the circumference of the femoral and tibial ACL insertion. For the purpose of detecting the greatest dimension of the ACL, the most posterior, as well as the most anterior aspect of the insertion was marked. A lateral roentgenogram was obtained from every knee, each with a precise overlapping of the femoral condyles. To evaluate the anatomic landmarks of the insertion sites of the ACL, 2 methods described in the literature were applied: the radiographic quadrant method on the femur5 (Fig 1) and the radiographic anteroposterior (AP) diameter of the tibial plateau on the tibia.25 These 2 methods were also selected by the
156
V. MUSAHL ET AL.
manufacturer of a surgical robot for definition of anatomic landmarks in their CAS planning station. Therefore, comparison between roentgenograms and CT scans could be made. The radiographic quadrant method is based on anatomical studies by Bernard et al.5 The location of the center of the femoral insertion site of the ACL was evaluated using the greatest diameter of the femoral condyles as a reference. A rectangle was drawn underneath the roof of the notch or Blumensaat’s line. The depth of the notch (A2) and the width of the femoral condyles (B2) exceeding the length of Blumensaat’s line are the margins of that rectangle. Two parameters, A and B, describe the location of the center of the ACL. These parameters, expressed in percent values, are distances from the center of the ACL insertion to Blumensaat’s line (A1),27 calculated as A ⫽ A1/A2 and to the posterior margin of the femoral condyle (B1), calculated as B ⫽ B1/B2. In the literature, A and B were determined as 25%/25% for the center of the femoral insertion site of the ACL (Fig 1).5 Tibial tunnel placement using the CAS planning station, however, is more complex. The manufacturer created a step in the software that superimposes the injured and uninjured knee to use the projected Blumensaat’s line of the uninjured knee as a guide for providing an impingement-free tibial tunnel placement. The surgeon is also free to use predescribed parameters to achieve tunnel placement, such as 43% as described in the literature.25 In this study, the tibial ACL insertion was planned with the CAS planning station software, at the posterior aspect of the marked insertion site. For comparison to the lateral roentgenograms, a line was drawn underneath the cortex of the medial tibial plateau marking the anteroposterior diameter of the tibia (C2) (Fig 2).25 The distance (C1) from the anterior edge of the tibial tuberosity to a point 5 mm anterior to the most posterior marker (p) was calculated on that line, corresponding to the radius of a 10-mm tunnel reamer that was chosen in this study. The point of desired tunnel placement (C) was then calculated as C⫽ C1/C2 (Fig 2). Subsequently, CT scans were taken of all knees. The CT spiral included approximately 250 CT slices (140 kV, 120mA, 3-mm slice collimation, 1.5 pitch, 2-mm reconstruction interval, 0.8 seconds rotation time, helical⫹ Z-interpolation, standard filter) between approximately 10 cm above and below the joint line. The CT data was then transferred to the CAS planning station of the surgical robot CASPAR (computer-assisted surgical planning and robotics; U.R.S.-
FIGURE 2. Lateral roentgenogram of the tibia showing the distance from anterior (C1) of the center of the tibial tunnel (x) with respect to the total tibial anteroposterior diameter (C2). Arrows indicate anterior (a) and posterior (p) markers.
Orho, Rastatt, Germany). This CAS system requires the insertion of reference screws in both the distal femur and the proximal tibia.18 These reference screws contain fiducials that serve as the reference coordinate system in the CAS system as well as the site of registration of the subject (cadaveric specimen or patient’s leg). Furthermore, linked to the CAS system is a surgical planning station that uses the 2-dimensional CT data and a software package (developed by the manufacturer) to place tunnels for ACL reconstruction. This software package allows for a 2-dimensional view of the bony anatomy of the knee in frontal, sagittal, and transverse cuts. Any position on the bone is referred to as an x-, y- and z-dataset with respect to the fiducials in the reference screw and the frontal, sagittal, and transverse cuts are connected to each other via a reference line. Tunnels for ACL reconstruction were then planned on the CAS planning station. The greatest dimension of the tibia and femur needed to be quantified. The quadrant on the femur was placed under Blumensaat’s line. For the tibia, a line was placed directly underneath the cortex of the medial tibial plateau (Fig 2). The software then creates the planned femoral tunnel at (A) 25% and (B) 25% by default. The tibial tunnel was created manually by relying on the marked insertion sites of the ACL as described previously. Because the copper wires made the ACL insertion evident, the planned femoral tunnel (default) could be adjusted until a complete overlapping of tunnel and
ACL TUNNEL PLACEMENT: ANATOMY AND CAS GUIDELINES
157
TABLE 1. Results From Lateral Radiographs and Computed Tomography for the Femur Femur
FIGURE 3. Mean and standard deviation for the geometrical center of the femoral ACL insertion, lateral radiographs versus CT data (P ⬎ .05).
copper wires was achieved (Fig 3). For statistical analysis, the difference between the planned femoral tunnel (default) and the adjusted planned tunnel was noted. The tibial tunnel was created using a parameter (C1) described previously and was noted for comparison with lateral roentgenograms. After the femoral tunnel for ACL reconstruction was created on the CAS planning station, the tunnel positions in sagittal and transverse cuts were compared to evaluate the clock position to the radiographic quadrant method. Therefore, a clock was drawn on the CT prints and the clock position was noted. For validation of the CAS planning station, the robot was then asked to carry out the created plan for the femoral and tibial tunnels. The specimen was rigidly fixed to the robot base with custom-made clamps, and bone motion was recorded with an optical bone motion sensor (U.R.S.-Ortho, Rastatt, Germany). After registration of the specimens through the reference screws, a core reamer attached to the robot’s end effector (diameter, 10 mm) was used to drill the tunnels. The bone plugs were then removed from the core reamers and the 6 previously inserted copper wires were located. The number of copper wires inside the reamer and the number remaining in the insertion site were noted for both the femur and the tibia. For comparison of data points for the center of the ACL insertion in roentgenograms and CT scans, statistical analysis with an unpaired Student t test was performed; significance was set at P ⬍ .05. RESULTS On the lateral roentgenograms, the femoral insertion sites of the ACL were found at 27.5% ⫾ 3.2%
Computed Tomography
Lateral Radiographs
Specimen
A (%)
B (%)
A (%)
B (%)
1 2 3 4 5 6 7 8 Mean⫾SD
33 27 31 25 28 27 26 23 28 ⫾ 3
30 27 29 29 27 30 22 21 27 ⫾ 4
29 30 26 25 27 25 25 26 27 ⫾ 2
28 30 29 25 25 24 25 24 26 ⫾ 2
NOTE. Results are calculated with the quadrant method. Values are for the geometric center of the anterior cruciate ligament and the distance from Blumensaat’s line (A) and distance from the posterior margin of the femoral condyle (B).
(range, 23.4% to 33.3%) and 26.9% ⫾ 3.5% (range, 21.4% to 30.4%) for parameters A and B, respectively (Table 1; Fig 3). The tibial tunnel insertion of the ACL (parameter C) in the AP direction was found at 46.2% ⫾ 2.8% (range, 40.0% to 49.1%) from the total anteroposterior diameter of the tibial plateau, calculated on the medial tibial plateau line (Table 2; Fig 4). After sizing the bony dimensions (as described in “METHODS”), the planning station software created the tunnel at 25%/25% by default. The copper wires, however, marked a circumference that was located at 26.6% ⫾ 1.9% (range, 25% to 30%) and 26.3% ⫾ 2.4% (range, 24% to 30%) for parameters A and B, respectively (Table 1; Fig 5). The distance from that default setting to the actual femoral insertion site of the ACL, as indicated by the copper wire markers, was 1.3 mm (range, 0.2 to 2.5 mm). For the tibial insertion TABLE 2. Results From Lateral Radiographs and Computed Tomography for the Tibia Tibia
Lateral Radiographs
CT
Specimen
C (%)
C (%)
1 2 3 4 5 6 7 8 Mean⫾SD
48 48 40 47 46 49 45 48 46 ⫾ 3
42 43 44 47 46 46 47 48 45 ⫾ 2
NOTE. Values given (C) are from the center of the tibial tunnel.
158
V. MUSAHL ET AL.
FIGURE 4. Mean and standard deviation for center of the tibial tunnel insertion, lateral radiographs versus CT data (P ⬎ .05).
site of the ACL, parameter C was obtained, on average, as 45.4% ⫾ 2.1% (range 42% to 48%) (Table 2). No statistically significant difference between CT and lateral roentgenograms could be detected either for the femur or for the tibia (P ⬎ .05). The tunnel drilling was performed by the robot, and the bone plugs were released from the core reamer. Gross inspection of the insertion site of the ACL revealed a tunnel drill hole in the center of the ACL in all of the femoral tunnels and in the posterior third in all of the tibial tunnels. The diameter of the reamer was 10 mm in all of the tunnel drillings. Therefore, on average, only 3 of 6 (range, 2 to 4) copper wire markers were found in the core reamer together with the bone plug, but 3 copper wires remained at the circumference of the larger insertion. Comparison of the radiographic quadrant method and clock position method revealed a correspondence among the geometric center of the ACL (copper wire markers), the radiographic parameters 26.6%/26.3% (A/B), and a clock position of 10:29 (range, 10:20 to 10:40) (Fig 6).
provide the surgeon with 2-dimensional CT images of the knee. Furthermore, the accuracy for the guidelines to place the femoral tunnel is less than 2 mm. A key to accurate tunnel planning is based on the proper quantification of the greatest diameter of the condyles and the tibial plateau. Unfortunately, this study could not compare fluoroscopy directly to passive and active CAS systems, because it only describes the guidelines of an active CAS system. To achieve proper placement of the tibial tunnel, a step was created in the software that superimposes the injured and uninjured knee. However, this approach does not account for possible pathology in the uninjured knee. Essentially, the surgeon can rely on the predescribed parameter (43%).25 Nevertheless, in this study, the tibial ACL insertion was located (and the tibial tunnel was planned) at the most posterior aspect of the marked insertion site. This was located on average at 45% of the anteroposterior tibial diameter. Carrying out the drilling procedure confirmed this location, because the posterior copper wire markers were present in the core reamer together with the tunnel plug. For the femoral side, parameter A (distance from the back of the condyle) and parameter B (distance from the roof of the notch) correlate well with the previously described parameter by Bernard et al.,5 who described the quadrant method and found 24.8% and 28.5% for parameters A and B, respec-
DISCUSSION Misplacement of tunnels during ACL reconstruction has led to poor outcomes.4,28 The introduction of CAS designed to improve tunnel placement could help in the surgical planning, because the system pays special attention to the patient’s anatomy to achieve more consistent tunnel placement. The objective of this study was to validate the CT-based software of a CAS planning station by comparing anatomy, roentgenographic images, and CT scan data. Overall, the planning station software was found to accurately
FIGURE 5. CT planning station preoperative plan showing the planned tunnel, tunnel start (circle), tunnel direction (arrow), and planned tunnel (oval).
ACL TUNNEL PLACEMENT: ANATOMY AND CAS GUIDELINES
159
FIGURE 6. CT planning station showing comparison of the clock method and quadrant method. (A) Transverse cut, copper wires (arrows), clock superimposed. (B) Sagittal cut (dashed line indicates corresponding cuts).
tively. In another study that described insertion geometry, similar values for these locations were revealed: 25% for parameter A and 28% for parameter B.7 To assist with tunnel placement, 2-dimensional scans or radiographs are widely used preoperatively. However, no direct links to the intraoperative situation exist. Furthermore, fluoroscopically assisted ACL reconstruction is used to increase the precision of ACL reconstruction procedures.12 As indicated in current literature, using active23 or passive29 CAS systems, precision can be further increased. Two differences exist between active and passive CAS systems. Passive systems generally perform no action and provide the surgeon with additional information before and during the surgical procedure; active systems perform parts of the surgery autonomously. A significant decrease in variance of tunnel placement was reported using fluoroscopic CAS.15 Non-image– based approaches for computer-assisted ACL reconstruction have also been developed.1,13 Magnetic resonance imaging (MRI)-based or CT-based CAS systems are able to create 3-dimensional images of the knee and can be used for intraoperative navigation. Accuracy was reported to be less than 1 mm.30 The surgeon that operates a CAS planning station needs to be aware of surface anatomy and its projection in image data. The use of CAS systems is time consuming, costly, and requires a learning curve.31
The benefits of CAS systems, however, are not only the precision in surgery but also its availability as a research tool and the availability of preoperative surgeon training, especially in residency programs. However, the operation of a CAS system requires adjustment to the anatomy of each individual knee. In the future, in vivo trials are planned to further assess the accuracy of ACL reconstruction procedures.
REFERENCES 1. Dessene VS, Lavallee R, Julliard R, et al. Computer assisted knee anterior cruciate ligament reconstruction: First clinical tests. J Image Guid Surg 1995;1:59-64. 2. DiGioia AM III, Jaramaz B, Colgan BD. Computer assisted orthopaedic surgery: Image guided and robotic assistive technologies. Clin Orthop Rel Res 1998;354:8-16. 3. Klos TV, Habets RJ, Banks AZ, et al. Computer assistance in arthroscopic anterior cruciate ligament reconstruction. Clin Orthop Rel Res 1998;354:65-69. 4. Aglietti P, Buzzi AZ, Giron F, et al. Arthroscopic-assisted anterior cruciate ligament reconstruction with the central third patellar tendon: A 5-8 year follow-up. Knee Surg Sports Traumatol Arthrosc 1997;5:138-144. 5. Bernard M, Hertel P, Hornung H, Cierpinski T. Femoral insertion of the ACL: Radiographic quadrant method. Am J Knee Surg 1997;10:14-21; discussion 21-2. 6. Harner CD, Marks PH, Fu FH, et al. Anterior cruciate ligament reconstruction: Endoscopic versus two-incision technique. Arthroscopy 1994;10:502-512. 7. Luites J, Wymenga A, Blankevoort L, Kooloos J. Insertion geometry of the two bundles of the anterior cruciate ligament.
160
8.
9.
10.
11.
12. 13. 14. 15.
16.
17.
18. 19.
V. MUSAHL ET AL.
Presented at the 9th ESSKA Congress, London, September 2000. Staubli HU, Rauschning W. Tibial attachment area of the anterior cruciate ligament in the extended knee position: Anatomy and cryosections in vitro complemented by magnetic resonance arthrography in vivo. Knee Surg Sports Traumatol Arthrosc 1994;2:138-146. Sati M, Bourquin Y, Staeubli H, Nolte LP. Considering anatomic and functional factors in ACL reconstruction: New technology. Presented at the 4th CAOS Meeting, Pittsburgh, PA, June 2000:121-123. Musahl V, Cierpinski T, Hornung H, Hertel P. Sekundaere vordere Kreuzbandplastik nach primaerer Naht und Kreuzbandplastik. Presented at the 63rd Jahrestagung der DGU, Berlin, Germany, November 1999. McGuire DA, Wolchok JC. Consistent and accurate graft passage and interference screw guide wire placement during single incision anterior cruciate ligament reconstruction. Arthroscopy 1997;13:526-529. Goble EM, Downey DJ, Wilcox TR. Positioning of the tibial tunnel for anterior cruciate ligament reconstruction. Arthroscopy 1995;11:688-695. Sati M, Staeubli H, Bourquin Y, et al. Clinical integration of new computer assisted technology for arthroscopic ACL replacement. Oper Tech Orthop 2000;10:40-49. DiGioia AM III. What is computer assisted orthopaedic surgery? Clin Orthop 1998;354:2-4. Klos T, Habets RJ, Banks A, et al. Computer assisted fluoroscopic arthroscopic anterior cruciate ligament reconstruction. Presented at the Computer Assisted Orthopaedic Surgery CAOS/USA Meeting, Pittsburgh, PA, June 1999. Kowal J, Amstutz CA, Nolte LP. On B-mode ultrasound based registration for computer assisted orthopaedic surgery. Presented at the Computer Assisted Orthopaedic Surgery CAOS/ USA Meeting. Pittsburgh, PA, July 2001. Picard F, Moody FE, DiGioia AM, et al. Unreliability of traditional x-ray evaluation in ACL reconstruction. Presented at the Computer Assisted Orthopaedic Surgery CAOS/USA Meeting. Pittsburgh, PA, July 2001. Petermann J, Kober R, Heinze P. Computer-assisted planning and robot-assisted surgery in anterior cruciate ligament reconstruction. Oper Tech Orthop 2000;10:50. Fu FH, Bennett CH, Ma CB, et al. Current trends in anterior
20.
21. 22.
23.
24. 25. 26. 27. 28.
29.
30. 31.
cruciate ligament reconstruction: Part II. Operative procedures and clinical correlations. Am J Sports Med 2000;28:124-130. Harner CD, Fu F, Irrgang JJ, Vogrin TM. Anterior and posterior cruciate ligament reconstruction in the new millennium: A global perspective. Knee Surg Sports Traumatol Arthrosc 2001;9:330-336. Harner CD, Livesay GD, Kashiwaguchi S, et al. Comparative study of the size and shape of human anterior and posterior cruciate ligaments. J Orthop Res 1995;13:429-434. Staeubli HU, Adam O, Becker W, Burgkart R. Anterior cruciate ligament and intercondylar notch in the coronal oblique plane: Anatomy complemented by magnetic resonance imaging in cruciate ligament-intact knees. Arthroscopy 1999;15: 349-359. Burkart A, Debski R, Rudy T, et al. A comparison of precision for ACL tunnel placement using traditional and robotic technique. Presented at the 4th CAOS USA Meeting, Pittsburgh, USA, June 15-17, 2000. Morgan CD, Kalman VR, Grawl DM. Definitive landmarks for reproducible tibial tunnel placement in anterior cruciate ligament reconstruction. Arthroscopy 1995;11:275-288. Amis AA, Jakob RP. Anterior cruciate ligament graft positioning, tensioning and twisting. Knee Surg Sports Traumatol Arthrosc 1998;6:2-12 (suppl 1). Eriksson E. How good are the results of ACL reconstruction? (Editorial, comment.) Knee Surg Sports Traumatol Arthrosc 1997;5:137. Jacobsen K, Bertheussen K, Gjerloff CC. Characteristics of the line of Blumensaat: An experimental analysis. Acta Orthop Scand 1974;45:764-771. Sommer C, Friedrich N, Mueller W. Improperly placed anterior cruciate ligament grafts: Correlation between radioligical parameters and clinical results. Knee Surg Sports Traumatol Arthrosc 2000;8:207-213. Staeubli HU. Computer-assisted knee surgery and navigation. Presented at the 2001 International Society of Arthroscopy, Knee Surgery and Orthopaeic Sports Medicine Congress. Montreux, Switzerland, May 2001. Amiot LP. Computer assisted ACL reconstruction: A feasibility study. Presented at the Second ACL Symposium: State of the Art 2000. Heidelberg, Germany, April 1999. Bernard M. Robot-assisted tunnel placement in ACL surgery. Presented at the 9th ESSKA Meeting, London, September 2000.
Purpose: With the development of computer-assisted surgery (CAS) systems, the surgeon’s ability to operate a CAS planning station will become essential. For example, default parameters in computed tomographic (CT) data are being used to place tunnels in anterior cruciate ligament (ACL) reconstruction. The goal of this study was to compare the location of the insertion sites in ACL reconstruction anatomically, via roentgenographic images and via CT scan data and to validate these tunnel placement parameters. Type of Study: Cadaveric analysis. Methods: Eight human cadaveric knees were marked with 6 copper wires 1 mm in diameter around the circumference of the insertions of the ACL. Using lateral roentgenograms and CT scans that were subsequently transferred to the CAS planning station, the tunnel locations were determined. These were based on a distance from the back of the condyle (location A) and from the roof of the notch (location B) on the femur and on a distance posterior from the tuberosity to the posterior margin along the tibial plateau, which is set as the CAS planning station’s default. Locations according to roentgenograms and CT scans were then compared and the accuracy of the CAS planning station was assessed. Results: Comparison of roentgenograms and CT revealed a femoral insertion at 27.5% ⫾ 3.2% and 26.9% ⫾ 3.5% (roentgenograms) and 26.6% ⫾ 1.9% and 26.3% ⫾ 2.4% (CT), respectively. The CAS planning station provided a tunnel location that was 1.3 ⫾ 1.0 mm (0.3 to 2.5 mm) away from the actual femoral ACL insertion. The tibial tunnel was placed according to the copper wire markers and was found to be at 46.2% ⫾ 2.8% (roentgenograms) and 45.4% ⫾ 2.1% (CT). No statistical differences between position in CT and roentgenograms could be detected (P ⬎ .05). Conclusions: The compared methodologies showed similar locations of the ACL insertions, assuring accurate preoperative planning with the CAS system. However, the CAS system requires adjustment to each individual knee anatomy. Key Words: Computer assisted orthopaedic surgery—ACL—Planning station—Tunnel placement—Insertion site.
S
ignificant interest has been shown in the development of computer-assisted surgery (CAS) as having the potential for improving surgical management
From the Department of Orthopaedic Surgery, Musculoskeletal Research Center, Pittsburgh, Pennsylvania, U.S.A.; and the Department of Orthopaedic Sports Medicine, Technical University of Munich, Germany (A.B.). Address correspondence and reprint requests to Freddie H. Fu, M.D., Department of Orthopaedic Surgery, Musculoskeletal Research Center, E 1641 BST, 210 Lothrop St, P.O. Box 71199, Pittsburgh, PA 15213, U.S.A. E-mail: [email protected] © 2003 by the Arthroscopy Association of North America 0749-8063/03/1902-3185$30.00/0 doi:10.1053/jars.2003.50001
154
of anterior cruciate ligament (ACL) deficiency.1-3 Specifically, accurate tunnel placement is expected to help resolve inconsistencies in reconstruction techniques. Therefore, the location of the insertions of the ACL on the tibia and femur is a critical factor in successful surgical reconstruction. Despite extensive reports on ACL anatomy,4-8 approximately 10% to 40% of tunnels for ACL reconstructions have been misplaced.9,10. During arthroscopic surgery, guides11 and fluoroscopically assisted techniques12 have been used to increase the accuracy of graft placement. Furthermore, active and passive CAS systems have been intro-
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 19, No 2 (February), 2003: pp 154-160
ACL TUNNEL PLACEMENT: ANATOMY AND CAS GUIDELINES duced. Passive systems generally perform no action and assist intraoperatively.13-17 Active CAS systems, conversely, perform parts of the surgical procedure autonomously.18 However, surgeons need to program (using image-based methods) these active robots before surgery. The geometry of the ACL insertions has been extensively described. For the nearly round femoral insertion located at the medial aspect of the lateral femoral condyle, the clock position method is the most widely used. The roof of the notch thereby serves as the 12:00 o’clock reference position. This is called Blumensaat’s line as viewed in a lateral roentgenogram. A 10:30 to 11:00 o’clock position for a right knee (and a corresponding 1:00- to 1:30 o’clock position for a left knee) is cited as the tunnel entrance position.19,20 In the other dimension, 7-mm offset guides are usually used to place a tunnel anterior to the posterior margin of the femoral condyle (the over the top position).21,22 Bernard et al.5 proposed a quadrant method that is based on roentgenograms. Assessment of tunnel placement can be achieved intraoperatively using K-wires and fluoroscopy. Although the tibial tunnel placement is generally easier to achieve repeatedly,23 the exact location for a smaller diameter graft (such as 10-mm diameter) within the insertion site is difficult to determine. However, the surgical description of the tibial insertion of the ACL is 7 mm anterior to the posterior cruciate ligament (PCL) and on a line between the anterior border of the posterior horn of the lateral meniscus and the posterior border of the anterior horn of the medial meniscus, intersecting the tibial spine.24 Radiographically, the location of the tibial tunnel location was defined using the anteroposterior diameter of the medial tibial plateau as 43% from the anterior edge.25 Consequently, an unsatisfactory ACL reconstruction outcome can result from errors in surgical technique.26 Because one possible solution to this problem is the integration of CAS into orthopedic surgery, the feasibility and accuracy of such a system needs to be validated. Characteristically, CAS systems require understanding the relationships between CT and bone surface topography. A surgeon who uses arthroscopy to view the notch now needs to interpret the 2-dimensional CT image of this region. Therefore, the goal of this study was to compare the location of the insertion sites of the ACL in anatomy, roentgenographic images, and CT scan data and to validate the guidelines of a CAS planning station for its use of tunnel placement parameters for ACL reconstruction with respect
155
FIGURE 1. Lateral roentgenogram of the femur showing the geometric center of the ACL (x), with respect to its distance from Blumensaat’s line (A1) and from the posterior margin of the femoral condyle (B1), calculated with the quadrant method.5 Width (A2) and depth (B2) of the femoral condyle are shown.
to Blumensaat’s line, the femoral condyles, and the tibial plateau (default). METHODS Eight fresh-frozen human cadaveric knees (aged 50 to 79 years) were tested. The specimens were stored at ⫺20°C and thawed at room temperature for 24 hours before testing. The knees were then dissected free of musculature and capsular tissue, and the infrapatellar fat pat (Hoffa fat pat) and synovial tissue surrounding the ACL were dissected leaving the ACL, PCL, and the medial and lateral structures intact. Six 1-mm diameter copper wires, 6 mm in length, were inserted around the circumference of the femoral and tibial ACL insertion. For the purpose of detecting the greatest dimension of the ACL, the most posterior, as well as the most anterior aspect of the insertion was marked. A lateral roentgenogram was obtained from every knee, each with a precise overlapping of the femoral condyles. To evaluate the anatomic landmarks of the insertion sites of the ACL, 2 methods described in the literature were applied: the radiographic quadrant method on the femur5 (Fig 1) and the radiographic anteroposterior (AP) diameter of the tibial plateau on the tibia.25 These 2 methods were also selected by the
156
V. MUSAHL ET AL.
manufacturer of a surgical robot for definition of anatomic landmarks in their CAS planning station. Therefore, comparison between roentgenograms and CT scans could be made. The radiographic quadrant method is based on anatomical studies by Bernard et al.5 The location of the center of the femoral insertion site of the ACL was evaluated using the greatest diameter of the femoral condyles as a reference. A rectangle was drawn underneath the roof of the notch or Blumensaat’s line. The depth of the notch (A2) and the width of the femoral condyles (B2) exceeding the length of Blumensaat’s line are the margins of that rectangle. Two parameters, A and B, describe the location of the center of the ACL. These parameters, expressed in percent values, are distances from the center of the ACL insertion to Blumensaat’s line (A1),27 calculated as A ⫽ A1/A2 and to the posterior margin of the femoral condyle (B1), calculated as B ⫽ B1/B2. In the literature, A and B were determined as 25%/25% for the center of the femoral insertion site of the ACL (Fig 1).5 Tibial tunnel placement using the CAS planning station, however, is more complex. The manufacturer created a step in the software that superimposes the injured and uninjured knee to use the projected Blumensaat’s line of the uninjured knee as a guide for providing an impingement-free tibial tunnel placement. The surgeon is also free to use predescribed parameters to achieve tunnel placement, such as 43% as described in the literature.25 In this study, the tibial ACL insertion was planned with the CAS planning station software, at the posterior aspect of the marked insertion site. For comparison to the lateral roentgenograms, a line was drawn underneath the cortex of the medial tibial plateau marking the anteroposterior diameter of the tibia (C2) (Fig 2).25 The distance (C1) from the anterior edge of the tibial tuberosity to a point 5 mm anterior to the most posterior marker (p) was calculated on that line, corresponding to the radius of a 10-mm tunnel reamer that was chosen in this study. The point of desired tunnel placement (C) was then calculated as C⫽ C1/C2 (Fig 2). Subsequently, CT scans were taken of all knees. The CT spiral included approximately 250 CT slices (140 kV, 120mA, 3-mm slice collimation, 1.5 pitch, 2-mm reconstruction interval, 0.8 seconds rotation time, helical⫹ Z-interpolation, standard filter) between approximately 10 cm above and below the joint line. The CT data was then transferred to the CAS planning station of the surgical robot CASPAR (computer-assisted surgical planning and robotics; U.R.S.-
FIGURE 2. Lateral roentgenogram of the tibia showing the distance from anterior (C1) of the center of the tibial tunnel (x) with respect to the total tibial anteroposterior diameter (C2). Arrows indicate anterior (a) and posterior (p) markers.
Orho, Rastatt, Germany). This CAS system requires the insertion of reference screws in both the distal femur and the proximal tibia.18 These reference screws contain fiducials that serve as the reference coordinate system in the CAS system as well as the site of registration of the subject (cadaveric specimen or patient’s leg). Furthermore, linked to the CAS system is a surgical planning station that uses the 2-dimensional CT data and a software package (developed by the manufacturer) to place tunnels for ACL reconstruction. This software package allows for a 2-dimensional view of the bony anatomy of the knee in frontal, sagittal, and transverse cuts. Any position on the bone is referred to as an x-, y- and z-dataset with respect to the fiducials in the reference screw and the frontal, sagittal, and transverse cuts are connected to each other via a reference line. Tunnels for ACL reconstruction were then planned on the CAS planning station. The greatest dimension of the tibia and femur needed to be quantified. The quadrant on the femur was placed under Blumensaat’s line. For the tibia, a line was placed directly underneath the cortex of the medial tibial plateau (Fig 2). The software then creates the planned femoral tunnel at (A) 25% and (B) 25% by default. The tibial tunnel was created manually by relying on the marked insertion sites of the ACL as described previously. Because the copper wires made the ACL insertion evident, the planned femoral tunnel (default) could be adjusted until a complete overlapping of tunnel and
ACL TUNNEL PLACEMENT: ANATOMY AND CAS GUIDELINES
157
TABLE 1. Results From Lateral Radiographs and Computed Tomography for the Femur Femur
FIGURE 3. Mean and standard deviation for the geometrical center of the femoral ACL insertion, lateral radiographs versus CT data (P ⬎ .05).
copper wires was achieved (Fig 3). For statistical analysis, the difference between the planned femoral tunnel (default) and the adjusted planned tunnel was noted. The tibial tunnel was created using a parameter (C1) described previously and was noted for comparison with lateral roentgenograms. After the femoral tunnel for ACL reconstruction was created on the CAS planning station, the tunnel positions in sagittal and transverse cuts were compared to evaluate the clock position to the radiographic quadrant method. Therefore, a clock was drawn on the CT prints and the clock position was noted. For validation of the CAS planning station, the robot was then asked to carry out the created plan for the femoral and tibial tunnels. The specimen was rigidly fixed to the robot base with custom-made clamps, and bone motion was recorded with an optical bone motion sensor (U.R.S.-Ortho, Rastatt, Germany). After registration of the specimens through the reference screws, a core reamer attached to the robot’s end effector (diameter, 10 mm) was used to drill the tunnels. The bone plugs were then removed from the core reamers and the 6 previously inserted copper wires were located. The number of copper wires inside the reamer and the number remaining in the insertion site were noted for both the femur and the tibia. For comparison of data points for the center of the ACL insertion in roentgenograms and CT scans, statistical analysis with an unpaired Student t test was performed; significance was set at P ⬍ .05. RESULTS On the lateral roentgenograms, the femoral insertion sites of the ACL were found at 27.5% ⫾ 3.2%
Computed Tomography
Lateral Radiographs
Specimen
A (%)
B (%)
A (%)
B (%)
1 2 3 4 5 6 7 8 Mean⫾SD
33 27 31 25 28 27 26 23 28 ⫾ 3
30 27 29 29 27 30 22 21 27 ⫾ 4
29 30 26 25 27 25 25 26 27 ⫾ 2
28 30 29 25 25 24 25 24 26 ⫾ 2
NOTE. Results are calculated with the quadrant method. Values are for the geometric center of the anterior cruciate ligament and the distance from Blumensaat’s line (A) and distance from the posterior margin of the femoral condyle (B).
(range, 23.4% to 33.3%) and 26.9% ⫾ 3.5% (range, 21.4% to 30.4%) for parameters A and B, respectively (Table 1; Fig 3). The tibial tunnel insertion of the ACL (parameter C) in the AP direction was found at 46.2% ⫾ 2.8% (range, 40.0% to 49.1%) from the total anteroposterior diameter of the tibial plateau, calculated on the medial tibial plateau line (Table 2; Fig 4). After sizing the bony dimensions (as described in “METHODS”), the planning station software created the tunnel at 25%/25% by default. The copper wires, however, marked a circumference that was located at 26.6% ⫾ 1.9% (range, 25% to 30%) and 26.3% ⫾ 2.4% (range, 24% to 30%) for parameters A and B, respectively (Table 1; Fig 5). The distance from that default setting to the actual femoral insertion site of the ACL, as indicated by the copper wire markers, was 1.3 mm (range, 0.2 to 2.5 mm). For the tibial insertion TABLE 2. Results From Lateral Radiographs and Computed Tomography for the Tibia Tibia
Lateral Radiographs
CT
Specimen
C (%)
C (%)
1 2 3 4 5 6 7 8 Mean⫾SD
48 48 40 47 46 49 45 48 46 ⫾ 3
42 43 44 47 46 46 47 48 45 ⫾ 2
NOTE. Values given (C) are from the center of the tibial tunnel.
158
V. MUSAHL ET AL.
FIGURE 4. Mean and standard deviation for center of the tibial tunnel insertion, lateral radiographs versus CT data (P ⬎ .05).
site of the ACL, parameter C was obtained, on average, as 45.4% ⫾ 2.1% (range 42% to 48%) (Table 2). No statistically significant difference between CT and lateral roentgenograms could be detected either for the femur or for the tibia (P ⬎ .05). The tunnel drilling was performed by the robot, and the bone plugs were released from the core reamer. Gross inspection of the insertion site of the ACL revealed a tunnel drill hole in the center of the ACL in all of the femoral tunnels and in the posterior third in all of the tibial tunnels. The diameter of the reamer was 10 mm in all of the tunnel drillings. Therefore, on average, only 3 of 6 (range, 2 to 4) copper wire markers were found in the core reamer together with the bone plug, but 3 copper wires remained at the circumference of the larger insertion. Comparison of the radiographic quadrant method and clock position method revealed a correspondence among the geometric center of the ACL (copper wire markers), the radiographic parameters 26.6%/26.3% (A/B), and a clock position of 10:29 (range, 10:20 to 10:40) (Fig 6).
provide the surgeon with 2-dimensional CT images of the knee. Furthermore, the accuracy for the guidelines to place the femoral tunnel is less than 2 mm. A key to accurate tunnel planning is based on the proper quantification of the greatest diameter of the condyles and the tibial plateau. Unfortunately, this study could not compare fluoroscopy directly to passive and active CAS systems, because it only describes the guidelines of an active CAS system. To achieve proper placement of the tibial tunnel, a step was created in the software that superimposes the injured and uninjured knee. However, this approach does not account for possible pathology in the uninjured knee. Essentially, the surgeon can rely on the predescribed parameter (43%).25 Nevertheless, in this study, the tibial ACL insertion was located (and the tibial tunnel was planned) at the most posterior aspect of the marked insertion site. This was located on average at 45% of the anteroposterior tibial diameter. Carrying out the drilling procedure confirmed this location, because the posterior copper wire markers were present in the core reamer together with the tunnel plug. For the femoral side, parameter A (distance from the back of the condyle) and parameter B (distance from the roof of the notch) correlate well with the previously described parameter by Bernard et al.,5 who described the quadrant method and found 24.8% and 28.5% for parameters A and B, respec-
DISCUSSION Misplacement of tunnels during ACL reconstruction has led to poor outcomes.4,28 The introduction of CAS designed to improve tunnel placement could help in the surgical planning, because the system pays special attention to the patient’s anatomy to achieve more consistent tunnel placement. The objective of this study was to validate the CT-based software of a CAS planning station by comparing anatomy, roentgenographic images, and CT scan data. Overall, the planning station software was found to accurately
FIGURE 5. CT planning station preoperative plan showing the planned tunnel, tunnel start (circle), tunnel direction (arrow), and planned tunnel (oval).
ACL TUNNEL PLACEMENT: ANATOMY AND CAS GUIDELINES
159
FIGURE 6. CT planning station showing comparison of the clock method and quadrant method. (A) Transverse cut, copper wires (arrows), clock superimposed. (B) Sagittal cut (dashed line indicates corresponding cuts).
tively. In another study that described insertion geometry, similar values for these locations were revealed: 25% for parameter A and 28% for parameter B.7 To assist with tunnel placement, 2-dimensional scans or radiographs are widely used preoperatively. However, no direct links to the intraoperative situation exist. Furthermore, fluoroscopically assisted ACL reconstruction is used to increase the precision of ACL reconstruction procedures.12 As indicated in current literature, using active23 or passive29 CAS systems, precision can be further increased. Two differences exist between active and passive CAS systems. Passive systems generally perform no action and provide the surgeon with additional information before and during the surgical procedure; active systems perform parts of the surgery autonomously. A significant decrease in variance of tunnel placement was reported using fluoroscopic CAS.15 Non-image– based approaches for computer-assisted ACL reconstruction have also been developed.1,13 Magnetic resonance imaging (MRI)-based or CT-based CAS systems are able to create 3-dimensional images of the knee and can be used for intraoperative navigation. Accuracy was reported to be less than 1 mm.30 The surgeon that operates a CAS planning station needs to be aware of surface anatomy and its projection in image data. The use of CAS systems is time consuming, costly, and requires a learning curve.31
The benefits of CAS systems, however, are not only the precision in surgery but also its availability as a research tool and the availability of preoperative surgeon training, especially in residency programs. However, the operation of a CAS system requires adjustment to the anatomy of each individual knee. In the future, in vivo trials are planned to further assess the accuracy of ACL reconstruction procedures.
REFERENCES 1. Dessene VS, Lavallee R, Julliard R, et al. Computer assisted knee anterior cruciate ligament reconstruction: First clinical tests. J Image Guid Surg 1995;1:59-64. 2. DiGioia AM III, Jaramaz B, Colgan BD. Computer assisted orthopaedic surgery: Image guided and robotic assistive technologies. Clin Orthop Rel Res 1998;354:8-16. 3. Klos TV, Habets RJ, Banks AZ, et al. Computer assistance in arthroscopic anterior cruciate ligament reconstruction. Clin Orthop Rel Res 1998;354:65-69. 4. Aglietti P, Buzzi AZ, Giron F, et al. Arthroscopic-assisted anterior cruciate ligament reconstruction with the central third patellar tendon: A 5-8 year follow-up. Knee Surg Sports Traumatol Arthrosc 1997;5:138-144. 5. Bernard M, Hertel P, Hornung H, Cierpinski T. Femoral insertion of the ACL: Radiographic quadrant method. Am J Knee Surg 1997;10:14-21; discussion 21-2. 6. Harner CD, Marks PH, Fu FH, et al. Anterior cruciate ligament reconstruction: Endoscopic versus two-incision technique. Arthroscopy 1994;10:502-512. 7. Luites J, Wymenga A, Blankevoort L, Kooloos J. Insertion geometry of the two bundles of the anterior cruciate ligament.
160
8.
9.
10.
11.
12. 13. 14. 15.
16.
17.
18. 19.
V. MUSAHL ET AL.
Presented at the 9th ESSKA Congress, London, September 2000. Staubli HU, Rauschning W. Tibial attachment area of the anterior cruciate ligament in the extended knee position: Anatomy and cryosections in vitro complemented by magnetic resonance arthrography in vivo. Knee Surg Sports Traumatol Arthrosc 1994;2:138-146. Sati M, Bourquin Y, Staeubli H, Nolte LP. Considering anatomic and functional factors in ACL reconstruction: New technology. Presented at the 4th CAOS Meeting, Pittsburgh, PA, June 2000:121-123. Musahl V, Cierpinski T, Hornung H, Hertel P. Sekundaere vordere Kreuzbandplastik nach primaerer Naht und Kreuzbandplastik. Presented at the 63rd Jahrestagung der DGU, Berlin, Germany, November 1999. McGuire DA, Wolchok JC. Consistent and accurate graft passage and interference screw guide wire placement during single incision anterior cruciate ligament reconstruction. Arthroscopy 1997;13:526-529. Goble EM, Downey DJ, Wilcox TR. Positioning of the tibial tunnel for anterior cruciate ligament reconstruction. Arthroscopy 1995;11:688-695. Sati M, Staeubli H, Bourquin Y, et al. Clinical integration of new computer assisted technology for arthroscopic ACL replacement. Oper Tech Orthop 2000;10:40-49. DiGioia AM III. What is computer assisted orthopaedic surgery? Clin Orthop 1998;354:2-4. Klos T, Habets RJ, Banks A, et al. Computer assisted fluoroscopic arthroscopic anterior cruciate ligament reconstruction. Presented at the Computer Assisted Orthopaedic Surgery CAOS/USA Meeting, Pittsburgh, PA, June 1999. Kowal J, Amstutz CA, Nolte LP. On B-mode ultrasound based registration for computer assisted orthopaedic surgery. Presented at the Computer Assisted Orthopaedic Surgery CAOS/ USA Meeting. Pittsburgh, PA, July 2001. Picard F, Moody FE, DiGioia AM, et al. Unreliability of traditional x-ray evaluation in ACL reconstruction. Presented at the Computer Assisted Orthopaedic Surgery CAOS/USA Meeting. Pittsburgh, PA, July 2001. Petermann J, Kober R, Heinze P. Computer-assisted planning and robot-assisted surgery in anterior cruciate ligament reconstruction. Oper Tech Orthop 2000;10:50. Fu FH, Bennett CH, Ma CB, et al. Current trends in anterior
20.
21. 22.
23.
24. 25. 26. 27. 28.
29.
30. 31.
cruciate ligament reconstruction: Part II. Operative procedures and clinical correlations. Am J Sports Med 2000;28:124-130. Harner CD, Fu F, Irrgang JJ, Vogrin TM. Anterior and posterior cruciate ligament reconstruction in the new millennium: A global perspective. Knee Surg Sports Traumatol Arthrosc 2001;9:330-336. Harner CD, Livesay GD, Kashiwaguchi S, et al. Comparative study of the size and shape of human anterior and posterior cruciate ligaments. J Orthop Res 1995;13:429-434. Staeubli HU, Adam O, Becker W, Burgkart R. Anterior cruciate ligament and intercondylar notch in the coronal oblique plane: Anatomy complemented by magnetic resonance imaging in cruciate ligament-intact knees. Arthroscopy 1999;15: 349-359. Burkart A, Debski R, Rudy T, et al. A comparison of precision for ACL tunnel placement using traditional and robotic technique. Presented at the 4th CAOS USA Meeting, Pittsburgh, USA, June 15-17, 2000. Morgan CD, Kalman VR, Grawl DM. Definitive landmarks for reproducible tibial tunnel placement in anterior cruciate ligament reconstruction. Arthroscopy 1995;11:275-288. Amis AA, Jakob RP. Anterior cruciate ligament graft positioning, tensioning and twisting. Knee Surg Sports Traumatol Arthrosc 1998;6:2-12 (suppl 1). Eriksson E. How good are the results of ACL reconstruction? (Editorial, comment.) Knee Surg Sports Traumatol Arthrosc 1997;5:137. Jacobsen K, Bertheussen K, Gjerloff CC. Characteristics of the line of Blumensaat: An experimental analysis. Acta Orthop Scand 1974;45:764-771. Sommer C, Friedrich N, Mueller W. Improperly placed anterior cruciate ligament grafts: Correlation between radioligical parameters and clinical results. Knee Surg Sports Traumatol Arthrosc 2000;8:207-213. Staeubli HU. Computer-assisted knee surgery and navigation. Presented at the 2001 International Society of Arthroscopy, Knee Surgery and Orthopaeic Sports Medicine Congress. Montreux, Switzerland, May 2001. Amiot LP. Computer assisted ACL reconstruction: A feasibility study. Presented at the Second ACL Symposium: State of the Art 2000. Heidelberg, Germany, April 1999. Bernard M. Robot-assisted tunnel placement in ACL surgery. Presented at the 9th ESSKA Meeting, London, September 2000.