- Email: [email protected]
The accuracy of computed tomography for determining femoral and tibial total knee arthroplasty component rotation
The Journal of Arthroplasty Vol. 15 No. 6 2000
The Accuracy of Computed Tomography for Determining Femoral and Tibial Total Knee Arthroplasty Component Rotation Laith M. Jazrawi, MD, Luther Birdzell, Frederick J. Kummer, PhD, and Paul E. Di Cesare, MD
Abstract: Patellofemoral complications, instability, and tibial polyethylene wear after total knee arthroplasty (TKA) resulting from malrotation of the tibial or femoral components (or both) may be difficult to diagnose based on physical examination and standard knee radiographs. The preoperative assessment of implant rotational alignment is critical in planning treatment because the femoral or tibial component (or both) may need to be revised if malpositioned. The purpose of this study was to ascertain the accuracy of computed tomography (CT) scan for determining rotational alignment of femoral and tibial components in TKA. TKA components were inserted in human cadaver specimens at neutral and 5° of external or internal rotation. For each position, the amount of rotation, determined from digital photographs, was compared with CT scan. The correlation coefficient between these two values averaged 0.87, which was significant at P ⬍ .05. The CT scan protocol described in this study can be applied clinically to patients with patellofemoral complaints to confirm or rule out the presence of component malrotation. Key words: computed tomography, knee, rotation, arthroplasty.
Patellofemoral complications (eg, patella subluxation or dislocation, patellar clunk, wear or loosening of the patellar component, and patella fracture) are the most common complications after total knee arthroplasty (TKA), occurring in 30% of cases [1,2]. Poor patella tracking or dislocation can be the result of malrotation of the tibial or femoral components (or both), an excessively tight lateral retinaculum, improper patellar component positioning, patellar
component loosening, improper axial orientation of the tibial and femoral components, or an overstuffed joint (ie, increased thickness of the patella bone– component construct) [2]. Malrotation also causes rotational incongruity between femoral and tibial components, resulting in increased contact stresses along the tibia during flexion [3]. Although most of these causes, including improper axial alignment, can be determined from plain radiographs or physical examination, implant rotational malalignment cannot. This information is critical in diagnosing or planning treatment of patellofemoral problems because the femoral or tibial component (or both) may need to be revised if malpositioned. Although a previous clinical study was performed investigating the use of computed tomography (CT) scan in determining component rotation in TKA, no one to our knowledge has investigated the accuracy of CT scan for determining rotational alignment. The pur-
From the Musculoskeletal Research Center, Department of Orthopaedic Surgery, New York University–Hospital for Joint Diseases, New York, New York. Submitted June 28, 1999; accepted February 14, 2000. No benefits or funds were received in support of this study. Reprint requests: Paul E. Di Cesare, MD, Musculoskeletal Research Center, Hospital for Joint Diseases, 301 East 17th Street, New York, NY 10003. Copyright r 2000 by Churchill Livingstonet 0883-5403/00/1506-0011$10.00/0 doi:10.1054/arth.2000.8193
761
762 The Journal of Arthroplasty Vol. 15 No. 6 September 2000
Fig. 1. Digital image of externally rotated femoral component. Black arrows mark the epicondylar axis; white arrows mark femoral pegs.
pose of this study was to ascertain the accuracy of CT scan for determining rotational alignment of femoral and tibial components in TKA.
Materials and Methods Six embalmed, paired human femora and tibiae were selected based on size to fit (within 2 mm) a large Genesis I (Smith & Nephew, Memphis, TN) femoral and tibial component. The femoral component was manufactured from cobalt-chromium and the tibial component from titanium. Tibial and femoral specimens were secured in a bench vice and prepared for cemented TKA by 1 senior resident (L.M.J.) with guidance from the senior author (P.E.D.), who is an experienced total joint arthroplasty surgeon, using standard protocol for the Genesis I system. For each knee, components were inserted in neutral (0°), 5° external rotation (⫹), and 5° internal rotation (⫺) in relation to the epicondylar axis of the femur and the posterior condyles of the tibia using a goniometer for initial
Fig. 2. Digital image of internally rotated tibial component. Black arrows mark the posterior tibial axis; white arrows mark the posterior aspect of the tibial component.
alignment. At each position of rotation, digital images were recorded using a digital camera (Sony Mavica MVC-FD7, Sony, Tokyo, Japan) mounted on an adjustable tripod. Femoral component rotation was measured as the angle formed between a 0.062-mm Kirschner wire (K-wire) placed in the epicondylar axis of the femur, as described by Berger et al [4] and a second K-wire secured to the top of the inner pegs of the femoral component with polymethyl methacrylate to mark its geometric center (Fig. 1). Tibial component rotation was determined from the angle formed between a K-wire fixed with polymethyl methacrylate to the posterior aspect of the tibial component and a second K-wire (secured with stainless steel tacks) 2 cm below the joint line along the posterior condyles of the tibial specimens (Fig. 2). Images were converted from JPEG format to TIFF format using Adobe PhotoShop 4.0 (Adobe Systems Incorporated, San Jose, CA) and imported into NIH Image Analysis (National Institutes of Health, Washington, DC), in which the angle of tibial and femoral
Accuracy of CT in TKA
●
Jazrawi et al.
763
Fig. 3. NIH Image Analysis software (National Institutes of Health, Washington, DC) used to determine external rotation of femoral component in Fig. 1. Images were converted from JPEG format to TIFF format using Adobe PhotoShop 4.0 (Adobe Systems Incorporated, San Jose, CA) and subsequently imported into NIH Image Analysis to determine degree of rotation (NIH Image Analysis cannot recognize JPEG files). The angle formed measured 3.65° of external rotation for the femoral component.
component rotation was determined (Fig. 3). The NIH Image Analysis program measures angles in digital images. The program places black lines as seen in Fig. 3 over the K-wires and measures the angle formed at their vertex. Adobe PhotoShop 4.0 was used to convert the image from JPEG format to TIFF format so that it could be recognized by the NIH Image Analysis program. Measurements were performed 3 times and the mean determined.
Fig. 4. Computed tomography image of femoral component shows transposed orientation of the anatomic epicondylar axis, marked by the black arrows, and the orientation of the femoral prosthesis axis (determined by the femoral component pegs), marked by the white arrows. Images were subsequently imported into the NIH Image Analysis program (National Institutes of Health, Washington, DC) to determine the degree of component rotation. The angle formed measured 4.05° of external rotation.
CT scan (General Electric High Speed Advantage scanner, model number CT01, Milwaukee, WI) was performed on each specimen after digital photography. The software package for angle determination comes standard with the scanner. CT axial images (5 mm in thickness) were modified with a soft algorithm, which minimizes artifact from hardware. Femoral imaging began at the level of the epicondylar axis and extended distally to include the K-wire–
764 The Journal of Arthroplasty Vol. 15 No. 6 September 2000 labeled pegs of the femoral component. The angle function off the main menu was entered. The orientation of the femoral component then was determined by superimposing the image of the K-wire–labeled epicondylar axis on the CT image containing the orientation of the femoral pegs (Fig. 4). Tibial scans started from the K-wire–labeled tibial component distally through the posterior condyles of the tibia. The orientation of the tibial component was determined by superimposing the K-wire–labeled tibial component CT image (Fig. 5A) on the CT image containing the K-wire–labeled posterior tibial axis (Fig. 5B). These CT images were imported into NIH Image Analysis, in which the angle of tibial and femoral component rotation was determined. Correlation between component rotations measured from digitized images of the specimens and CT scans was determined by the Spearman rank correlation coefficient with statistical significance set at P ⬍ .05.
Results Correlation between actual component rotation and CT scan–determined rotation was statistically significant (P ⬍ .05) (Table 1). Actual mean femoral component rotation for neutral, external, and internal rotation was 0.11 (SD 0.03), ⫹5.92 (SD 0.63), and ⫺5.46 (SD 0.71), and CT-measured rotation was 0.17 (SD 0.02), ⫹6.04 (SD 0.67), and ⫺5.53 (SD 0.87). Correlation coefficients were 0.93, 0.85, and 0.85. Actual mean tibial component rotation for neutral, external, and internal rotation was 0.21 (SD 0.07), ⫹4.14 (SD 0.83), and ⫺5.65 (SD 1.38), and CT-measured rotation was 0.29 (SD 0.05), ⫹4.36 (SD 0.54), and ⫺6.24 (SD 1.05). Correlation coefficients were 0.90, 0.85, and 0.83.
Discussion Alignment has been implicated as a primary or contributing cause of many complications in TKA. Nagamine et al [5] showed that malrotation of components in TKA is associated with poor clinical outcomes. Rotational malalignment between the femoral and tibial components can be a cause of polyethylene wear, particularly in conforming knee designs [6], causing rotational incongruity between femoral and tibial components, resulting in increased contact stresses along the tibia during flexion or extension depending on component malposition [3]. Figgie et al [7,8] outlined criteria for proper axial alignment in TKA and concluded that component rotation was an important factor. Although
Fig. 5. (A) Transverse computed tomography (CT) image of tibial component. CT scan determines orientation of tibial component by superimposing tibial component orientation, as seen in this photo of a CT image containing the Kirschner wire–labeled posterior tibial axis, as seen in B. (B) Transverse CT image contains superimposed orientation of tibial component (white arrow) and orientation of posterior tibial axis (black arrow). This image was then imported to NIH Image Analysis (National Institutes of Health, Washington, DC), and the degree of rotation was determined.
axial and rotational alignment are recognized as critical factors influencing TKA outcome, only axial alignment has been studied extensively because, in contrast to rotational alignment, axial alignment can be measured readily from standing long-leg radiographs [2,7–22]. A simple analogous test for determining component rotation has not been available. Most techniques available are simply a subjec-
Accuracy of CT in TKA ●
Jazrawi et al.
765
Table 1. Mean Rotational Values for Computed Tomography and Digital Photography
Digital CT Correlation P value
Tibia (0)
Femur (0)
Tibia (⫹5)
Femur (⫹5)
Tibia (⫺5)
Femur (⫺5)
0.21 ⫾ 0.07 0.29 ⫾ 0.05 0.9 .01
0.11 ⫾ 0.03 0.17 ⫾ 0.02 0.93 .009
4.14 ⫾ 0.83 4.36 ⫾ 0.54 0.85 .01
5.92 ⫾ 0.63 6.04 ⫾ 0.67 0.85 .01
5.65 ⫾ 1.38 6.24 ⫾ 1.05 0.83 .04
5.46 ⫾ 0.71 5.53 ⫾ 0.87 0.85 .03
NOTE. Values are mean ⫾ SD. (0) ⫽ neutral rotation; (⫺5) ⫽ degrees of internal rotation; (⫹5) ⫽ degrees of external rotation. CT, computed tomography.
tive measurement of rotation performed at the time of revision surgery [23]. Eckhoff et al [12] described a technique to measure rotational alignment from plain radiographs. Strict criteria for radiographic technique, lack of clinical correlation, and the necessity of a femoral component with pegs have limited its use. Berger et al [1,24] have investigated clinically the use of CT scan in determining component rotation and showed that excessive internal rotation of femoral and tibial components was associated with an increased incidence of patellofemoral complaints in patients. Their CT protocol is based on the surgical epicondylar axis and tibial tubercle as landmarks. The present study differs because we evaluated the ability of CT scan to determine accurately implant rotation compared with actual measurements made from direct visualization (as could be determined from a revision TKA). It provides the basis for further clinical studies investigating the use of CT scan for determining component rotation. It also supports and extends the clinical studies of Berger et al [1,24], confirming the ability of CT scan to determine component malposition accurately. For femoral component analysis, the surgical epicondylar axis was used as the anatomic reference because it is used by most contemporary alignment techniques to properly rotate the femoral component [25]. The technique used in this study employed the femoral component pegs to mark the component’s geometric axis, which was then compared with the anatomic epicondylar axis to determine the precise femoral component rotation. Because some knee systems do not contain pegs, a different component reference point, such as posterior condylar angle [1,24], can be employed. It also is possible that difficulty may arise in identifying the epicondylar axis on CT scans without K-wire references. Previous clinical studies have not encountered this problem [1,24], however. For tibial component analysis, the posterior tibial axis was used as the anatomic reference point. The implant reference point was a tangential line to the posterior aspect of the tibial component. The angle formed between
the intersection of these lines was recorded as the degree of tibial component rotation. In cases in which the posterior tibial axis cannot be defined, as in severely arthritic knees, the tibial tubercle can be used as the anatomic reference point or more reliably, the transtibial axis [1,24,26,27]. CT is a noninvasive technique that can be used to determine accurately the degree of femoral and tibial total knee component rotation. The CT scanner protocol described in this study can be applied easily to patients with clinical patellofemoral complaints, maltracking, or dislocation to confirm or rule out the presence of component malrotation. This preoperative information can guide orthopaedic surgeons to determine the cause of patellofemoral complaints and to aid in the preoperative planning of patients in whom surgery is deemed necessary.
References 1. Berger RA, Crossett LS, Jacobs JJ, et al: Patellofemoral complication following TKA: the consequences of malrotation with normal axial alignment. J Arthroplasty 13:234, 1998 2. Rhoades DD, Noble PC, Reuben JD, et al: The effect of femoral component position on patellar tracking after total knee arthroplasty. Clin Orthop 260:43, 1990 3. Lewis P, Rorabeck CH, Bourne RB, et al: Posteromedial tibial polyethylene failure in total knee replacements. Clin Orthop 299:11, 1994 4. Berger RA, Rubash HE, Seel MJ, et al: Determining the rotational alignment of the femoral component in total knee arthroplasty using the epicondylar axis. Clin Orthop 286:40, 1993 5. Nagamine R, Whitesides LA, White SE, et al: Patellar tracking after total knee arthroplasty: the effect of tibial tray malrotation and articular surface configuration. Clin Orthop 304:263, 1994 6. Northmore-Ball MD, Freeman MAR, Barnes KA: Subclinical rotational malposition: a potential cause of catastrophic wear in conforming condylar knee arthroplasty. Knee 2:177, 1995 7. Figgie M, Goldberg V, Figgie H: Salvage of symptomatic patellofemoral joint following cruciate substitut-
766 The Journal of Arthroplasty Vol. 15 No. 6 September 2000
8.
9.
10. 11.
12.
13.
14.
15.
16. 17.
18.
ing total knee arthroplasty. Am J Knee Surg 1:48, 1988 Figgie M, Golberg V, Figgie H: The effects of alignment of the implant on fracture of the patella after total condylar knee arthroplasty. J Bone Joint Surg Am 71:1031, 1989 Aglietti P, Gandenzi A: Patellofemoral functional results and complications with the posterior stabilized total knee prosthesis. J Arthroplasty 3:17, 1988 Huo M, Sculco T: Complications in primary total knee arthroplasty. Orthop Rev 19:781, 1990 Lynch A, Rorabeck C, Bourne R: Extensor mechanism complications following total knee arthroplasty. J Arthroplasty 2:135, 1987 Eckhoff DG, Piatt BE, Gnadinger CA, et al: Assessing rotational alignment in total knee arthroplasty. Clin Orthop 318:176, 1995 Ranawat C: The patellofemoral joint in total condylar arthroplasty: Pros and cons based on five to ten year follow-up observations. Clin Orthop 205:93, 1986 Vince K, Insall J, Kelly M: The total knee arthroplasty: ten to twelve year results of a cemented knee replacement. J Bone Joint Surg Br 71:793, 1989 Merkow R, Soudry M, Insall J: Patella dislocation following total knee replacement. J Bone Joint Surg Am 67:1321, 1985 Bryan R, Rand J: Revision total knee arthroplasty. Clin Orthop 170:116, 1982 Buechel FF: Treatment of the patella in revision total knee surgery using a rotating bearing patella replacement. Orthop Rev 76S:34, 1990 Clayton M, Thiripathi R: Patellar complications after total condylar arthroplasty. Clin Orthop 170:131, 1982
19. Nagamine R, Whitesides LA, Otani T, et al: Effect of medial displacement of the tibial tubercle on patellar position after rotational malposition of the femoral component in total knee arthroplasty. J Arthroplasty 11:104, 1996 20. Stiehl JB, Cherveny PM: Femoral rotational alignment using the tibial shaft axis in total knee arthroplasty. Clin Orthop 331:47, 1996 21. Arima J, Whiteside LA, McCarthy DS, et al: Femoral rotational alignment, based on the anteroposterior axis, in total knee arthroplasty in a valgus knee. J Bone Joint Surg Am 77:1331, 1995 22. Anouchi YS, Whitesides LA, Kaiser AD, et al: The effects of axial rotational alignment of the femoral component on knee stability and patella tracking in total knee arthroplasty demonstrated on autopsy specimens. Clin Orthop 287:170, 1993 23. Insall J, Binazzi R, Soudry M, et al: Total knee arthroplasty. Clin Orthop 192:13, 1985 24. Berger RA, Crossett LS, Jacobs JJ, et al: Malrotation causing patellofemoral complications after total knee arthroplasty. Clin Orthop 356:144, 1998 25. Eckhoff DG, Metzger RG, Vandewalle MV: Malrotation associated with implant alignment technique in total knee arthroplasty. Clin Orthop 321:28, 1995 26. Poilvache PL, Insall JN, Scuderi GR, et al: Rotational landmarks and sizing of the distal femur in total knee arthroplasty. Clin Orthop 331:35, 1996 27. Mantas J, Bloebaum RD, Skedros JG, et al: Implications of reference axes used for rotational alignment of the femoral component in primary and revision knee arthroplasty. J Arthroplasty 7:531, 1992
The Accuracy of Computed Tomography for Determining Femoral and Tibial Total Knee Arthroplasty Component Rotation Laith M. Jazrawi, MD, Luther Birdzell, Frederick J. Kummer, PhD, and Paul E. Di Cesare, MD
Abstract: Patellofemoral complications, instability, and tibial polyethylene wear after total knee arthroplasty (TKA) resulting from malrotation of the tibial or femoral components (or both) may be difficult to diagnose based on physical examination and standard knee radiographs. The preoperative assessment of implant rotational alignment is critical in planning treatment because the femoral or tibial component (or both) may need to be revised if malpositioned. The purpose of this study was to ascertain the accuracy of computed tomography (CT) scan for determining rotational alignment of femoral and tibial components in TKA. TKA components were inserted in human cadaver specimens at neutral and 5° of external or internal rotation. For each position, the amount of rotation, determined from digital photographs, was compared with CT scan. The correlation coefficient between these two values averaged 0.87, which was significant at P ⬍ .05. The CT scan protocol described in this study can be applied clinically to patients with patellofemoral complaints to confirm or rule out the presence of component malrotation. Key words: computed tomography, knee, rotation, arthroplasty.
Patellofemoral complications (eg, patella subluxation or dislocation, patellar clunk, wear or loosening of the patellar component, and patella fracture) are the most common complications after total knee arthroplasty (TKA), occurring in 30% of cases [1,2]. Poor patella tracking or dislocation can be the result of malrotation of the tibial or femoral components (or both), an excessively tight lateral retinaculum, improper patellar component positioning, patellar
component loosening, improper axial orientation of the tibial and femoral components, or an overstuffed joint (ie, increased thickness of the patella bone– component construct) [2]. Malrotation also causes rotational incongruity between femoral and tibial components, resulting in increased contact stresses along the tibia during flexion [3]. Although most of these causes, including improper axial alignment, can be determined from plain radiographs or physical examination, implant rotational malalignment cannot. This information is critical in diagnosing or planning treatment of patellofemoral problems because the femoral or tibial component (or both) may need to be revised if malpositioned. Although a previous clinical study was performed investigating the use of computed tomography (CT) scan in determining component rotation in TKA, no one to our knowledge has investigated the accuracy of CT scan for determining rotational alignment. The pur-
From the Musculoskeletal Research Center, Department of Orthopaedic Surgery, New York University–Hospital for Joint Diseases, New York, New York. Submitted June 28, 1999; accepted February 14, 2000. No benefits or funds were received in support of this study. Reprint requests: Paul E. Di Cesare, MD, Musculoskeletal Research Center, Hospital for Joint Diseases, 301 East 17th Street, New York, NY 10003. Copyright r 2000 by Churchill Livingstonet 0883-5403/00/1506-0011$10.00/0 doi:10.1054/arth.2000.8193
761
762 The Journal of Arthroplasty Vol. 15 No. 6 September 2000
Fig. 1. Digital image of externally rotated femoral component. Black arrows mark the epicondylar axis; white arrows mark femoral pegs.
pose of this study was to ascertain the accuracy of CT scan for determining rotational alignment of femoral and tibial components in TKA.
Materials and Methods Six embalmed, paired human femora and tibiae were selected based on size to fit (within 2 mm) a large Genesis I (Smith & Nephew, Memphis, TN) femoral and tibial component. The femoral component was manufactured from cobalt-chromium and the tibial component from titanium. Tibial and femoral specimens were secured in a bench vice and prepared for cemented TKA by 1 senior resident (L.M.J.) with guidance from the senior author (P.E.D.), who is an experienced total joint arthroplasty surgeon, using standard protocol for the Genesis I system. For each knee, components were inserted in neutral (0°), 5° external rotation (⫹), and 5° internal rotation (⫺) in relation to the epicondylar axis of the femur and the posterior condyles of the tibia using a goniometer for initial
Fig. 2. Digital image of internally rotated tibial component. Black arrows mark the posterior tibial axis; white arrows mark the posterior aspect of the tibial component.
alignment. At each position of rotation, digital images were recorded using a digital camera (Sony Mavica MVC-FD7, Sony, Tokyo, Japan) mounted on an adjustable tripod. Femoral component rotation was measured as the angle formed between a 0.062-mm Kirschner wire (K-wire) placed in the epicondylar axis of the femur, as described by Berger et al [4] and a second K-wire secured to the top of the inner pegs of the femoral component with polymethyl methacrylate to mark its geometric center (Fig. 1). Tibial component rotation was determined from the angle formed between a K-wire fixed with polymethyl methacrylate to the posterior aspect of the tibial component and a second K-wire (secured with stainless steel tacks) 2 cm below the joint line along the posterior condyles of the tibial specimens (Fig. 2). Images were converted from JPEG format to TIFF format using Adobe PhotoShop 4.0 (Adobe Systems Incorporated, San Jose, CA) and imported into NIH Image Analysis (National Institutes of Health, Washington, DC), in which the angle of tibial and femoral
Accuracy of CT in TKA
●
Jazrawi et al.
763
Fig. 3. NIH Image Analysis software (National Institutes of Health, Washington, DC) used to determine external rotation of femoral component in Fig. 1. Images were converted from JPEG format to TIFF format using Adobe PhotoShop 4.0 (Adobe Systems Incorporated, San Jose, CA) and subsequently imported into NIH Image Analysis to determine degree of rotation (NIH Image Analysis cannot recognize JPEG files). The angle formed measured 3.65° of external rotation for the femoral component.
component rotation was determined (Fig. 3). The NIH Image Analysis program measures angles in digital images. The program places black lines as seen in Fig. 3 over the K-wires and measures the angle formed at their vertex. Adobe PhotoShop 4.0 was used to convert the image from JPEG format to TIFF format so that it could be recognized by the NIH Image Analysis program. Measurements were performed 3 times and the mean determined.
Fig. 4. Computed tomography image of femoral component shows transposed orientation of the anatomic epicondylar axis, marked by the black arrows, and the orientation of the femoral prosthesis axis (determined by the femoral component pegs), marked by the white arrows. Images were subsequently imported into the NIH Image Analysis program (National Institutes of Health, Washington, DC) to determine the degree of component rotation. The angle formed measured 4.05° of external rotation.
CT scan (General Electric High Speed Advantage scanner, model number CT01, Milwaukee, WI) was performed on each specimen after digital photography. The software package for angle determination comes standard with the scanner. CT axial images (5 mm in thickness) were modified with a soft algorithm, which minimizes artifact from hardware. Femoral imaging began at the level of the epicondylar axis and extended distally to include the K-wire–
764 The Journal of Arthroplasty Vol. 15 No. 6 September 2000 labeled pegs of the femoral component. The angle function off the main menu was entered. The orientation of the femoral component then was determined by superimposing the image of the K-wire–labeled epicondylar axis on the CT image containing the orientation of the femoral pegs (Fig. 4). Tibial scans started from the K-wire–labeled tibial component distally through the posterior condyles of the tibia. The orientation of the tibial component was determined by superimposing the K-wire–labeled tibial component CT image (Fig. 5A) on the CT image containing the K-wire–labeled posterior tibial axis (Fig. 5B). These CT images were imported into NIH Image Analysis, in which the angle of tibial and femoral component rotation was determined. Correlation between component rotations measured from digitized images of the specimens and CT scans was determined by the Spearman rank correlation coefficient with statistical significance set at P ⬍ .05.
Results Correlation between actual component rotation and CT scan–determined rotation was statistically significant (P ⬍ .05) (Table 1). Actual mean femoral component rotation for neutral, external, and internal rotation was 0.11 (SD 0.03), ⫹5.92 (SD 0.63), and ⫺5.46 (SD 0.71), and CT-measured rotation was 0.17 (SD 0.02), ⫹6.04 (SD 0.67), and ⫺5.53 (SD 0.87). Correlation coefficients were 0.93, 0.85, and 0.85. Actual mean tibial component rotation for neutral, external, and internal rotation was 0.21 (SD 0.07), ⫹4.14 (SD 0.83), and ⫺5.65 (SD 1.38), and CT-measured rotation was 0.29 (SD 0.05), ⫹4.36 (SD 0.54), and ⫺6.24 (SD 1.05). Correlation coefficients were 0.90, 0.85, and 0.83.
Discussion Alignment has been implicated as a primary or contributing cause of many complications in TKA. Nagamine et al [5] showed that malrotation of components in TKA is associated with poor clinical outcomes. Rotational malalignment between the femoral and tibial components can be a cause of polyethylene wear, particularly in conforming knee designs [6], causing rotational incongruity between femoral and tibial components, resulting in increased contact stresses along the tibia during flexion or extension depending on component malposition [3]. Figgie et al [7,8] outlined criteria for proper axial alignment in TKA and concluded that component rotation was an important factor. Although
Fig. 5. (A) Transverse computed tomography (CT) image of tibial component. CT scan determines orientation of tibial component by superimposing tibial component orientation, as seen in this photo of a CT image containing the Kirschner wire–labeled posterior tibial axis, as seen in B. (B) Transverse CT image contains superimposed orientation of tibial component (white arrow) and orientation of posterior tibial axis (black arrow). This image was then imported to NIH Image Analysis (National Institutes of Health, Washington, DC), and the degree of rotation was determined.
axial and rotational alignment are recognized as critical factors influencing TKA outcome, only axial alignment has been studied extensively because, in contrast to rotational alignment, axial alignment can be measured readily from standing long-leg radiographs [2,7–22]. A simple analogous test for determining component rotation has not been available. Most techniques available are simply a subjec-
Accuracy of CT in TKA ●
Jazrawi et al.
765
Table 1. Mean Rotational Values for Computed Tomography and Digital Photography
Digital CT Correlation P value
Tibia (0)
Femur (0)
Tibia (⫹5)
Femur (⫹5)
Tibia (⫺5)
Femur (⫺5)
0.21 ⫾ 0.07 0.29 ⫾ 0.05 0.9 .01
0.11 ⫾ 0.03 0.17 ⫾ 0.02 0.93 .009
4.14 ⫾ 0.83 4.36 ⫾ 0.54 0.85 .01
5.92 ⫾ 0.63 6.04 ⫾ 0.67 0.85 .01
5.65 ⫾ 1.38 6.24 ⫾ 1.05 0.83 .04
5.46 ⫾ 0.71 5.53 ⫾ 0.87 0.85 .03
NOTE. Values are mean ⫾ SD. (0) ⫽ neutral rotation; (⫺5) ⫽ degrees of internal rotation; (⫹5) ⫽ degrees of external rotation. CT, computed tomography.
tive measurement of rotation performed at the time of revision surgery [23]. Eckhoff et al [12] described a technique to measure rotational alignment from plain radiographs. Strict criteria for radiographic technique, lack of clinical correlation, and the necessity of a femoral component with pegs have limited its use. Berger et al [1,24] have investigated clinically the use of CT scan in determining component rotation and showed that excessive internal rotation of femoral and tibial components was associated with an increased incidence of patellofemoral complaints in patients. Their CT protocol is based on the surgical epicondylar axis and tibial tubercle as landmarks. The present study differs because we evaluated the ability of CT scan to determine accurately implant rotation compared with actual measurements made from direct visualization (as could be determined from a revision TKA). It provides the basis for further clinical studies investigating the use of CT scan for determining component rotation. It also supports and extends the clinical studies of Berger et al [1,24], confirming the ability of CT scan to determine component malposition accurately. For femoral component analysis, the surgical epicondylar axis was used as the anatomic reference because it is used by most contemporary alignment techniques to properly rotate the femoral component [25]. The technique used in this study employed the femoral component pegs to mark the component’s geometric axis, which was then compared with the anatomic epicondylar axis to determine the precise femoral component rotation. Because some knee systems do not contain pegs, a different component reference point, such as posterior condylar angle [1,24], can be employed. It also is possible that difficulty may arise in identifying the epicondylar axis on CT scans without K-wire references. Previous clinical studies have not encountered this problem [1,24], however. For tibial component analysis, the posterior tibial axis was used as the anatomic reference point. The implant reference point was a tangential line to the posterior aspect of the tibial component. The angle formed between
the intersection of these lines was recorded as the degree of tibial component rotation. In cases in which the posterior tibial axis cannot be defined, as in severely arthritic knees, the tibial tubercle can be used as the anatomic reference point or more reliably, the transtibial axis [1,24,26,27]. CT is a noninvasive technique that can be used to determine accurately the degree of femoral and tibial total knee component rotation. The CT scanner protocol described in this study can be applied easily to patients with clinical patellofemoral complaints, maltracking, or dislocation to confirm or rule out the presence of component malrotation. This preoperative information can guide orthopaedic surgeons to determine the cause of patellofemoral complaints and to aid in the preoperative planning of patients in whom surgery is deemed necessary.
References 1. Berger RA, Crossett LS, Jacobs JJ, et al: Patellofemoral complication following TKA: the consequences of malrotation with normal axial alignment. J Arthroplasty 13:234, 1998 2. Rhoades DD, Noble PC, Reuben JD, et al: The effect of femoral component position on patellar tracking after total knee arthroplasty. Clin Orthop 260:43, 1990 3. Lewis P, Rorabeck CH, Bourne RB, et al: Posteromedial tibial polyethylene failure in total knee replacements. Clin Orthop 299:11, 1994 4. Berger RA, Rubash HE, Seel MJ, et al: Determining the rotational alignment of the femoral component in total knee arthroplasty using the epicondylar axis. Clin Orthop 286:40, 1993 5. Nagamine R, Whitesides LA, White SE, et al: Patellar tracking after total knee arthroplasty: the effect of tibial tray malrotation and articular surface configuration. Clin Orthop 304:263, 1994 6. Northmore-Ball MD, Freeman MAR, Barnes KA: Subclinical rotational malposition: a potential cause of catastrophic wear in conforming condylar knee arthroplasty. Knee 2:177, 1995 7. Figgie M, Goldberg V, Figgie H: Salvage of symptomatic patellofemoral joint following cruciate substitut-
766 The Journal of Arthroplasty Vol. 15 No. 6 September 2000
8.
9.
10. 11.
12.
13.
14.
15.
16. 17.
18.
ing total knee arthroplasty. Am J Knee Surg 1:48, 1988 Figgie M, Golberg V, Figgie H: The effects of alignment of the implant on fracture of the patella after total condylar knee arthroplasty. J Bone Joint Surg Am 71:1031, 1989 Aglietti P, Gandenzi A: Patellofemoral functional results and complications with the posterior stabilized total knee prosthesis. J Arthroplasty 3:17, 1988 Huo M, Sculco T: Complications in primary total knee arthroplasty. Orthop Rev 19:781, 1990 Lynch A, Rorabeck C, Bourne R: Extensor mechanism complications following total knee arthroplasty. J Arthroplasty 2:135, 1987 Eckhoff DG, Piatt BE, Gnadinger CA, et al: Assessing rotational alignment in total knee arthroplasty. Clin Orthop 318:176, 1995 Ranawat C: The patellofemoral joint in total condylar arthroplasty: Pros and cons based on five to ten year follow-up observations. Clin Orthop 205:93, 1986 Vince K, Insall J, Kelly M: The total knee arthroplasty: ten to twelve year results of a cemented knee replacement. J Bone Joint Surg Br 71:793, 1989 Merkow R, Soudry M, Insall J: Patella dislocation following total knee replacement. J Bone Joint Surg Am 67:1321, 1985 Bryan R, Rand J: Revision total knee arthroplasty. Clin Orthop 170:116, 1982 Buechel FF: Treatment of the patella in revision total knee surgery using a rotating bearing patella replacement. Orthop Rev 76S:34, 1990 Clayton M, Thiripathi R: Patellar complications after total condylar arthroplasty. Clin Orthop 170:131, 1982
19. Nagamine R, Whitesides LA, Otani T, et al: Effect of medial displacement of the tibial tubercle on patellar position after rotational malposition of the femoral component in total knee arthroplasty. J Arthroplasty 11:104, 1996 20. Stiehl JB, Cherveny PM: Femoral rotational alignment using the tibial shaft axis in total knee arthroplasty. Clin Orthop 331:47, 1996 21. Arima J, Whiteside LA, McCarthy DS, et al: Femoral rotational alignment, based on the anteroposterior axis, in total knee arthroplasty in a valgus knee. J Bone Joint Surg Am 77:1331, 1995 22. Anouchi YS, Whitesides LA, Kaiser AD, et al: The effects of axial rotational alignment of the femoral component on knee stability and patella tracking in total knee arthroplasty demonstrated on autopsy specimens. Clin Orthop 287:170, 1993 23. Insall J, Binazzi R, Soudry M, et al: Total knee arthroplasty. Clin Orthop 192:13, 1985 24. Berger RA, Crossett LS, Jacobs JJ, et al: Malrotation causing patellofemoral complications after total knee arthroplasty. Clin Orthop 356:144, 1998 25. Eckhoff DG, Metzger RG, Vandewalle MV: Malrotation associated with implant alignment technique in total knee arthroplasty. Clin Orthop 321:28, 1995 26. Poilvache PL, Insall JN, Scuderi GR, et al: Rotational landmarks and sizing of the distal femur in total knee arthroplasty. Clin Orthop 331:35, 1996 27. Mantas J, Bloebaum RD, Skedros JG, et al: Implications of reference axes used for rotational alignment of the femoral component in primary and revision knee arthroplasty. J Arthroplasty 7:531, 1992