Extramedullary versus intramedullary tibial cutting guides in megaprosthetic total knee replacement
© Karade et al.; licensee BioMed Central Ltd. 2012
Received: 15 June 2012
Accepted: 26 September 2012
Published: 2 October 2012
In a standard total knee replacement, tibial component alignment is a key factor for the long term success of the surgery. The purpose of this study is to compare the accuracy of extramedullary and intramedullary tibial cutting guides used in indigenous and imported implants respectively, in positioning of the tibial components in megaprosthetic knee replacements.
A comparative study of the accuracy of extramedullary and intramedullary tibial cutting guides was carried out in 92 megaprosthetic knee replacements for distal femoral tumors. For the proximal tibia cut for tibial component placement, an extramedullary guide was used in 65 patients and an intramedullary guide was used in 27 patients. Tibial component alignment angles were measured in postoperative X-rays with the help of CAD software.
There was more varus placement in coronal plane with extramedullary cutting guide (−1.18 +/− 2.4 degrees) than the intramedullary guide (−0.34 +/− 2.31 degrees) but this did not reach statistical significance. The goal of 90 +/− 2 degrees alignment of tibial component was achieved in 54% of patients in the extramedullary group versus 67% in the intramedullary group. In terms of sagittal plane alignment, extramedullary guide showed less accurate results (2.09 +/− 2.4 degrees) than intramedullary guide (0.50 +/− 3.80 degrees) for tibial component alignment, though 78% of patients were aligned within the goal of 0–5 degrees of tibial slope angle in extramedullary group versus 63% in intramedullary group. The mean error in the measurements due to rotation of the knee during taking the X-rays was less than 0.1 degrees and distribution of the X-rays with the rotation of knee was similar in both the groups.
Overall, in megaprosthetic knee replacement intramedullary guides gave more accurate results in sagittal plane and exhibited similar variability as of extramedullary guides in coronal plane.
KeywordsMegaprosthetic knee replacement Cutting guide Tibial component Tibial cut Statistical analysis
In a standard total knee replacement (TKR), the proximal tibia has to be cut for placement of the tibial component base plate. The tibial component alignment is a key factor for the long term success of the surgery[1–3]. This has not been well documented for megaprosthtic knee replacement surgery. Though the implant is constrained and a stemmed tibial component is used, accurate placement may have a role in long term survival of the implant.
In megaprosthetic knee replacement, the tibial component stem is required to be parallel to mechanical axis of tibia in both coronal and sagittal plane. The mechanical axis is defined as the line connecting the center of femoral head and center of talus bone. The mechanical axis of the tibia is nearly parallel to the anatomical axis which is the line connecting midpoints of outer cortical diameter at around 5 and 15 cm distal to the knee joint[4, 5]. The anatomical axis runs along the tibial canal as well. In indigenous implant system (ResTOR knee system), the tibial component stem is 10cm in length and 10mm in diameter. The stem is not long enough to engage the narrowest portion of the tibial intramedullary canal for self-alignment along the anatomical axis. As the tibial component base plate aligns itself along the cut plane, an accurate alignment of the cut plane with respect to the anatomical axis of the bone becomes very important. The cut should be perpendicular to the anatomical axis of the tibia. Extramedullary and intramedullary cutting guides are used predominantly to obtain accurate tibial cuts. Indigenous implant system (ResTOR knee system Sushrut, India) comes with an extramedullary cutting guide whereas the imported implant system (GMRS knee system, Stryker, USA) comes with an intramedullary cutting guide. Both guides have an alignment rod which has to be first placed parallel to the anatomical axis. Then a saw blade is guided to make a cut perpendicular to the rod. Both types of conventional guides use anatomical landmarks to position their alignment rod parallel to the anatomical axis of tibia. Besides bone landmarks, use of palpable tendons as reference has also been reported[7, 8]. Both cutting guides have specific advantages and limitations. Extramedullary guides maybe easier to use due to the long familiarity with standard knee replacements. Intramedullary guides have direct access to tibial canal and hence the anatomical axis, and are believed to be more accurate. There is however, a lack of consensus on which of the cutting guides gives better accuracy of cuts. A few studies have shown that intramedullary guides give more accurate cutting plane alignment than extramedullary guides in total knee arthroplasty[2, 3, 9]. No study appears to have been reported to compare the tibial cutting guides in megaprosthetic knee replacement for distal femoral tumor. The research objective of this study is to compare the accuracy of intramedullary and extramedullary guides used in the two implant systems as mentioned earlier for tibial resection.
Patients and surgical technique
Baseline patient characteristics
n (number of patients)
Mean age in years (S.D.)
Number of Left/Right legs
In the intramedullary group, an entry hole was created in the articular surface near the base of the anterior tibial spine using a six millimeter drill. This position is usually located in anterior one-third of the tibial articular surface. The entry hole was reamed progressively till a tight canal fit was obtained (10–21 mm). The medullary contents were decompressed by suction. Reamers were used to size the canal and intramedullary guide assembly was inserted using the appropriate diameter stem. Rotation alignment was referenced to tibial tubercle. The cutting block with a posterior slope of zero degree was assembled to the intramedullary guide assembly. The design of the guide ensures that the cutting block lies over the anterior portion of tibia, and the slit opening of the cutting block is perpendicular to the intramedullary rod. A stylus was used to determine the thickness of bone resection. The cutting block was secured to the tibia with pins. Then a cutting saw blade was inserted through the slit opening of the cutting block and the tibial bone slice was cut.
Measurement technique and X-ray validation
Unique features observed in benchmark X-rays of knee taken at different angles
Unique feature observed
Fibula can be seen distinctly and patella is at the center covering the knee joint
Fibula is partially covered by tibia and patella is not at the center
Fibula is totally covered by anterior part of tibia and patella is not at center
Fibula is totally covered by posterior part of tibia and patella is not at center
Fibula and patella can be seen distinctly
A power analysis showed that with the present sample size, a mean difference of 1.5° can be detected with 86% probability (alpha = 0.05, one sided comparison of two independent means). Outlier data detection was performed using Grubbs' test assuming that the errors are normally distributed (central limit theorem). Two null hypotheses were stated: (i) the intramedullary group shows less mean error than extramedullary group in coronal view and (ii) the intramedullary group shows less mean error than extramedullary group in sagittal view. Statistical significance of the difference in mean errors was calculated using independent t-test for two independent samples. Statistical significance of the difference in the percentage of correct tibial alignment was calculated using Chi-square test for two independent samples. Statistical significance of the difference in terms of surgeon, gender of the patient, number of left/right leg and number of X-rays with knee rotation was also determined using Chi-square test for two independent samples. The mathematical calculations and the statistical analysis were performed using Microsoft Excel 2007 and Minitab statistical software respectively. A p value greater than 0.05 was taken as statistical significance.
Results for postoperative measurements in both groups
Postoperative measurements in coronal view
Mean TCA1 +/− S.D. in degrees
−1.18 +/− 2.40
−0.34 +/− 2.31
p = 0.06
Number of X-rays within optimal range (percentage)
p = 0.25
Number of X-rays within valgus alignment (percentage)
p = 0.07
Mean error Δ in degrees
p = 0.65
Number of X-rays tends to 200 leg rotation (percentage)
p = 0.70
Postoperative measurements in sagittal view
Mean TCA2 +/− S.D. in degrees
2.09 +/− 2.40
0.50 +/− 3.28
p = 0.01
Number of X-rays within optimal range (percentage)
p = 0.16
Number of X-rays within posterior slope (percentage)
p = 0.03
Mean error Δ in degrees
p = 0.46
Number of X-rays tends to 200 leg rotation (percentage)
p = 0.38
In this study, the extramedullary group exhibited more mean error in coronal plane alignment of tibial component (−1.18 +/− 2.4°) compared to intramedullary group (−0.34 +/− 2.31°). However, this difference was not statistically significant (p = 0.06). The percentage of the optimal cases in both groups was also not significantly different. In the extramedullary group, the percentage of varus placements (75%) was three times the percentage of valgus placements (25%). The reason for this may lie in the distal alignment of the extramedullary rod. The proximal part of the extramedullary rod was aligned over the medial third of tibial tubercle and the distal part was aligned over the center of ankle. However, the mechanical axis (hence the anatomical axis parallel to it) runs through the mid-point of talus bone which usually lies on the medial side of ankle center. If this fact is not considered then the alignment rod will not be parallel to mechanical axis, but will be rotated toward the lateral side resulting in most of the final tibial component placement to be in varus. However, when the percentage of the varus cuts in the extramedullary group were compared to that in the intramedullary group, then no statistically significant difference was found. A few researchers have reported better accuracy using extramedullary guides when the distal alignment of extramedullary rod is taken as three millimeters medial to the mid-point of ankle. For intramedullary guides, one of the reasons mentioned by surgeons for inaccuracy is the instability of intramedullary rod (IM rod). The intramedullary guide relies on the parallelism of IM rod and tibial canal (and hence the anatomical axis). Usually the IM rod is thinner than tibial canal, which may cause a tilt of the rod inside the canal and hence a tilt in the cutting block alignment. The entry point position is also a key factor. The ideal entry point position is on the tibial articular surface corresponding to the proximal continuation of tibial canal. For example, in tibia with obvious varus deformity, the entry point needs to be more lateral than usual. Hence, an ideal entry point position should be preoperatively determined with the help of X-rays.
The sagittal alignment (TCA2) of a replaced knee is kinematically important because most of the knee motion occurs in this plane. In this context, intramedullary guide showed better accuracy than with an extramedullary guide (p < 0.05). In this study there was no significant difference in the number of cases falling in the optimal range of 0–5°. The reason for both these observations lies in the tendency of extramedullary guide to give a posterior slope in the tibial cut. This tendency makes the mean tibial component angle in sagittal view (TCA2) to deviate away from the desired 0° angle. It also increases the number of cases having a posterior slope. Indeed, there were 88% cases with posterior slope in extramedullary group compared to 67% in intramedullary group (p < 0.05). To avoid an undesirable anterior slope, surgeons try to give a posterior slope by moving the distal part of alignment rod a little away from the anterior tibial surface. This explains the tendency of the extramedullary guide to give a posterior slope in the tibial cut. In the intramedullary guide, once the IM rod is inserted inside the tibial canal, there is no possibility of adjustment of the tibial slope by moving the rod. The surgeon can only use different cutting blocks (giving 3-5° slope). These results for sagittal plane are in contrast to the results presented in a study performed by Cashman JP et al., where extramedullary guides showed better accuracy, for total knee arthroplasty (TKA) surgery. A prospective study performed by Maestro et al. on TKA surgery cases showed no statistical difference in sagittal plane positioning of the tibial component.
Extramedullary guides can be used in both deformed and non-deformed bones but are sensitive to surgical performance, while intramedullary guides rely on non-deformed tibial canals. Extramedullary guides are easy to use and can be employed for patients with low fat accumulation over their leg. Fat embolism and intramedullary fracture are considered as drawbacks for using intramedullary guides. The principle of using both the cutting guides is to make their alignment rod parallel to anatomical axis by using landmarks and then cut perpendicular to the alignment rod. The principle is perfect but the design and the procedure need some modification and improvements. To get a proper guidance for the center of the ankle, the design of extramedullary guide may need to be modified and used with proper preoperative planning. A marker can be placed externally over the ankle center which can be determined intraoperative by X-rays. This external marker can be used to align the distal part of alignment rod of the extramedullary guide. The design of the intramedullary guide needs to be improved in such a way that while making IM rod parallel to the tibial canal there should be no error. The IM rod should not wobble inside the canal. To ensure a correct position of the entry point, a preoperative plan must be made with the help of X-rays. Finally, while using either guide, the lower limb must be rigidly fixed to avoid disturbances during cutting block alignment, cutting block fixation and cutting. A combination cutting guide can be developed where the alignment of IM rod of the intramedullary guide can be cross-checked with an external alignment rod of extramedullary guide.
Rotation of knee while taking an X-ray also incorporates an error which was however, found to be very small (< 0.10). In this study, a few postoperative X-rays had a rotation of knee between 00 and 200, when compared to the benchmark X-rays. Equation 1 shows that larger the measurement angle θ more is the error included by a knee rotation of the same angle Ф. It is shown that the error Δ is very small for smaller knee rotation angles Ф and rapidly increases as Ф rises to 900 (Figure5). The distribution of X-rays with knee rotation was similar in both extramedullary and intramedullary group. A standard method should be devised to take a correct coronal view and sagittal view X-rays. Patellar position can be used as an external reference for taking a true coronal and sagittal view X-ray.
There are a few limitations in this study. The first one is that the measurements were made on 2D X-rays where the rotation of tibial component cannot be measured. Measurements on 3D CT, when compared to X-ray based measurements show better precision in such cases. Measurements made in 3D CT can also be used to validate the X-ray based measurements. The second limitation is that the errors which were measured may have some component involved due to inaccurate cutting by saw blade. Even if the alignment of the cutting plane is correctly determined, error can occur during the cutting process. In a few studies, cutting errors of around 1° have been reported mostly in sagittal view (compared to coronal view) because of deflection of saw blade in anterior-posterior direction. Third, the study was not based on randomized control trials. However, the cutting guides were linked with the implants which were chosen on the basis of only affordability by the patient. The study was based on evaluation through postoperative X-rays, and hence the preoperative conditions of the patient were not reported. More number of cases (especially in intramedullary group) can improve the statistical power of the study for more reliable results.
In conclusion, intramedullary guides used in conjunction with imported implants gave more accurate results in the coronal plane alignment of tibial component compared to extramedullary guides used with indigenous implants for megaprosthetic total knee replacement. No significant difference was found in the percentage of cases falling in the optimal range of the alignment in both coronal and saggital planes. Given the need to use tumor megaprostheses with stems in all patients, we suggest the use of intramedullary guides in all cases. Since the imported implant system is unaffordable by most of the patients, there is a need to develop an intramedullary guide for the indigenous implant. This work is currently underway.
Total knee replacement
Tibial component angle
Total knee arthroplasty
actual angle between anatomical axis and tibial implant axis
angular measurements of tibial component angle (TCA1 or TCA2)
angular error in measurements of θ
rotation of the knee.
The project was carried out in the OrthoCAD Network Research Cell at IIT Bombay and was partially supported by the NKN X-ray project funded by National Informatics Center, New Delhi. The authors would also like to thank Dr. Vivek Shetty (Hinduja hospital, Mumbai), Dr. Vijay Shetty (Hiranandani hospital, Mumbai) and Dr. Aashish (Tata Memorial Hospital, Mumbai) for their valuable feedback.
- Macdonald W, Styf J, Carlsson LV, Jacobsson CM: Improved tibial cutting accuracy in knee arthroplasty. Med Eng Phys. 2000, 26 (9): 807-812.View ArticleGoogle Scholar
- Maestro A, Harwin SF, Sandoval MG, Vaquero DH, Murcia A: Influence of intramedullary versus extramedullary alignment guides on final total knee arthroplasty component position: a radiographic analysis. J Arthroplasty. 1998, 13 (5): 552-558. 10.1016/S0883-5403(98)90055-9.View ArticlePubMedGoogle Scholar
- Reed MR, Bliss W, Sher JL, Emmerson KP, Jones SM, Partington PF: Extramedullary or intramedullary tibial alignment guides: a randomized, prospective trial of radiological alignment. J Bone Joint Surg. 2002, 84 (6): 858-860. 10.1302/0301-620X.84B6.12702.View ArticleGoogle Scholar
- Chiu KY, Yau WP, Ng TP, Tang WM: The accuracy of extramedullary guides for tibial component placement in total knee arthroplasty. Int Orthop. 2008, 32 (4): 467-471. 10.1007/s00264-007-0354-5.PubMed CentralView ArticlePubMedGoogle Scholar
- Yoo JH, Chang CB, Shin KS, Seong SC, Kim TK: Anatomical references to assess the posterior tibial slope in total knee arthroplasty: a comparison of 5 anatomical axes. J Arthroplasty. 2008, 23 (4): 586-592. 10.1016/j.arth.2007.05.006.View ArticlePubMedGoogle Scholar
- Shakespeare D: Conventional instruments in total knee replacement: what should we do with them?. Knee. 2006, 13 (1): 1-6. 10.1016/j.knee.2005.06.002.View ArticlePubMedGoogle Scholar
- Rajadhyaksha AD, Mehta H, Zelicof SB: Use of tibialis anteriortendon as distal landmarkfor extramedullary tibial alignmentin total knee arthroplasty. An anatomical study. Am J Orthop. 2009, 38 (3): E68-70.PubMedGoogle Scholar
- Schneider M, Heisel C, Aldinger PR, Breusch SJ: Use of palpable tendons for extramedullary tibial alignment in total knee arthroplasty. J Arthroplasty. 2007, 22 (2): 219-226. 10.1016/j.arth.2006.04.023.View ArticlePubMedGoogle Scholar
- Lozano LM, Segur JM, Macule F, Nunez M, Torner P, Castillo F, Suso S: Intramedullary versus extramedullary tibial cutting guide in severely obese patients undergoing total knee replacement: a randomized study of 70 patients with body mass index > 35 kg/m2. Obes Surg. 2008, 18 (12): 1599-1604. 10.1007/s11695-008-9564-1.View ArticlePubMedGoogle Scholar
- Durkin M, Aram L, Amirouche F, Gonzalez M: Sensitivity and error analysis in conjunction with tibial intramedullary guides in total knee arthroplasty. J Prosthet Orthot. 2001, 13 (2): 54-59. 10.1097/00008526-200106000-00015.View ArticleGoogle Scholar
- Cashman JP, Carty FL, Synnott K, Kenny PJ: Intramedullary versus extramedullary alignment of the tibial component in the Triathlon knee. J Orthop Surg Res. 2011, 6: 44-10.1186/1749-799X-6-44.PubMed CentralView ArticlePubMedGoogle Scholar
- Hewitt B, Shakespeare D: A straightforward method of assessing the accuracy of implantation of knee prostheses. Knee. 2001, 8 (2): 139-144. 10.1016/S0968-0160(00)00078-8.View ArticlePubMedGoogle Scholar
- Holme TJ, Henckel J, Cobb J, Hart AJ: Quantification of the difference between 3D CT and plain radiograph for measurement of the position of medial unicompartmental knee replacements. Knee. 2010, 18 (5): 300-305.View ArticlePubMedGoogle Scholar
- Jeffcote B, Shakespeare D: Varus/valgus alignment of the tibial component in total knee arthroplasty. Knee. 2003, 10: 243-247. 10.1016/S0968-0160(02)00101-1.View ArticlePubMedGoogle Scholar
- Yau WP, Chiu KY: Cutting errors in total knee replacement: assessment by computer assisted surgery. Knee Surg Sport Traumatol Arthrosc. 2008, 16 (7): 670-673. 10.1007/s00167-008-0550-x.View ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.