Kenneth A. Krackow, MD


Published 3/1/2003

SpotCheck: The role of CAS in TKA

Does computer-assisted surgery (CAS) make you a better surgeon? Is it really necessary to improve outcomes for total knee arthroplasty (TKA)? What role should CAS play in TKA?

Kenneth A. Krackow, MD
The State University of New York at Buffalo

Total knee arthroplasty is perhaps the orthopaedic procedure that requires the most particular, complicated, and accurate instrumentation. Current prostheses require a minimum of seven different cuts and several fixation holes or channels, every one of which has to be positioned with regard to the others to an extremely high degree of accuracy—less than 1 mm of translation, and ideally within 1 degree of arc. Errors of greater magnitude, because of the number of such positions, can lead to highly unsatisfactory results. Furthermore, the cuts themselves must be assessed both to an absolute bone surface reference and in relation to ligament tension and overall spacing. These basic factors suggest that computerization, along with sophisticated tracking and measurement equipment, can be of significant benefit.

Several simple examples can demonstrate the benefits and additional information derived from computer navigation during TKA. At the beginning of the procedure, navigation is the only way to accurately measure the point of maximal extension. In other words, exactly how much flexion contracture is present? In other situations, the question may be how much recurvatum is present? As the surgeon goes through the steps of deformity management and places trial components, the same questions recur.

When varus or valgus deformity exists, the amount of balancing to address this deformity is determined principally from what can be called a tension-stress examination—distracting the tibia distally in relationship to the femur and holding the medial and lateral soft tissues at similar tension. The deformity (the amount of release or tightening necessary) is not determined by measuring a weightbearing radiograph or negated by simply rotating the tibia into a neutral position below the femur. In both cases, the surgeon has a measurable angle between the femoral and tibial axes, but does not have any measurable information about the lengths and positions of the ligaments or other soft-tissue elements.

As the surgeon places trial components and performs a range of motion (ROM) and ligament examination, essentially all complete computer navigation systems enable a quantitative assessment of complete ROM, tibiofemoral alignment, and ligament stability. In the absence of the computer, the surgeon cannot accurately assess these factors. With computer navigation, the surgeon can precisely place the components by navigating position during final cementation.

To bring all of this into clinical relevance, consider the words of Lawrence D. Dorr, MD, who spoke at the 25th Current Concepts in Joint Replacement Meeting this past December. Dr. Dorr very eloquently summarized how the major current problems and shortfalls in both total knee and total hip arthroplasty were technical errors. He similarly discussed and concluded that most of these problems could be better managed by computer navigation. When we consider the issues a bit further, it seems obvious that the incorporation of robotic bone preparation to provide the safest and most accurate bone cuts will ultimately follow.

Mary I. O’Connor, MD
Mayo Clinic, Jacksonville, Fla.

The role of CAS in TKA continues to evolve. This technology was developed as a means of improving surgical precision with the expectation that this will translate into better clinical outcomes. Several well-designed clinical trials have shown that use of computer navigation improves the likelihood of achieving limb alignment within 3 degrees of the mechanical axis. In our prospective, randomized study of 200 patients, a statistically significant reduction in deviation from a neutral mechanical axis was achieved with navigation

(P = 0.004) compared to patients who had TKA without navigation (manuscript in preparation).

Kenneth A. Krackow, MD
Mary I. O’Connor, MD

The influence of mechanical axis alignment on implant survival is not, however, precisely known. One study of 400 patients operated on by a single surgeon using three modern implant designs showed that a mechanical axis within 3 degrees of neutral did not improve survival at 15-year follow-up. Another study also showed that a wide range of anatomic alignment did not affect modern implant durability at intermediate follow-up of 6 to 15 years. Although varus malalignment, all polyethylene implants, undersizing of components, and obesity have been found to negatively affect tibial component survival, body mass index and sizing may override alignment as factors influencing failure, according to a study of tibial components.

The reality is that we still see failure of primary TKA at relatively early follow-up. A prospective, multicenter study of 318 patients undergoing knee revision found a mean time to revision of only 7.9 years, with factors related to failure of instability (29 per­cent), wear (24.5 percent) and malalignment (9.4 percent), with most patients having several factors. Although malalignment may be well tolerated by some patients, in others it will contribute to failure.

Use of computer-assisted navigation clearly can improve accuracy of limb alignment and decrease failures related to malalignment, but how navigation will evolve to improve the surgeon’s ability to achieve appropriate ligament balancing and rotation of the components is not established. Current systems allow for assessment of ligament balancing but data is lacking on effectiveness of this tool for ensuring appropriate postoperative soft-tissue balances. In general, experience with current image-free navigation systems does not show improved precision in rotational placement of components.

Navigation systems will incorporate ongoing advances in technology. Most current systems are image-free, requiring no input of preoperative patient-specific images into the computer. A model of the patient’s knee is generated based on the surgeon’s identification of bone landmarks and the computer’s ‘morphing’ this information into a database of images. This process, however, creates the potential for error. Use of intraoperative computed tomography-based images, currently being applied by our neurosurgical colleagues, may improve accuracy for orthopaedic procedures.

An additional evolution of computer-assisted navigation is the application of this technology even before the patient enters the operating room. Based on preoperative magnetic resonance images, computer-designed, patient-specific cutting blocks are manufactured that the surgeon applies in the operating room without use of any navigation tool. Whether this approach will result in improved outcomes is unclear.

In conclusion, the role of the computer in assisting orthopaedic surgeons with surgical planning and execution of TKA will continue to evolve and be refined. The focus needs to expand from achieving improvement in limb alignment to better ligament balancing and prevention of wear. The truth is that even the most gifted orthopaedic surgeons do not consistently execute surgical skill and judgment at the highest level. Although improvements in implant design and bearing surfaces may create a more forgiving environment for lack of surgical precision, tools that decrease the inherent human variability in performance will benefit patients. The added expense of such technology remains a real challenge in the current healthcare environment.

Kenneth A. Krackow, MD, discloses the following: Stryker (royalties, paid consultant, speaker’s bureau, stock options, institutional/research support) and Smith & Nephew (royalties). Mary I. O’Connor, MD, discloses the following: DePuy, a Johnson & Johnson Company (royalties, institutional/research support) and Zimmer (royalties).


  1. Parratte S, Pagnano MW, Trousdale RT, Berry DJ: The mechanical axis may be the wrong target in computer assisted TKA. Paper presented at the 75th Annual Meeting of the American Academy of Orthopaedic Surgeons,; March 5-7, 2008; San Francisco, CA.
  2. Callaghan JJ, Liu SS, Warth LC: Computer-assisted surgery: A wine before its time. In the affirmative. J Arthroplasty 2006; 21(4 suppl 1):27-28.
  3. Berend ME, Ritter MA, Hyldahl HC, Meding JB, Redelman R: Implant migration and failure in total knee arthroplasty is related to body mass index and tibial component size. J Arthroplasty 2008;23(6 Suppl 1):104-109.
  4. Mulhall KJ, Ghomrawi HM, Scully S, Callaghan JJ, Saleh KJ: Current etiologies and modes of failure in total knee arthroplasty revision. Clin Orthop Relat Res 2006; 446:45-50.