Navigation as a Predictor of Soft Tissue Release During 90 Cases of Computer-assisted Total Knee Arthroplasty

Abstract

Soft tissue balance during total knee arthroplasty (TKA) procedures may be challenging to achieve in knees with severe valgus or varus deformities. The purpose of this study was to evaluate the role of navigation in predicting soft tissue release during 90 computer-assisted TKA procedures performed within a 2-year period. Fourteen soft tissue releases were performed, 13 of which were for genu varum, including 9 deep medial collateral ligaments and 4 deep and superficial medial collateral ligaments, and 1 for genu valgum deformity. Forty-one cases were intraoperatively overcorrectable (45.5%), and 20 cases were reducible to a neutral axis. The results proved that navigation can predict the need to perform a soft tissue release during computer-assisted TKA.

Soft tissue balance during total knee arthroplasty (TKA) may be challenging for surgeons to achieve in knees with severe valgus or varus deformities. Traditionally, orthopedic surgeons have assessed ligament balance in 3 stages during surgery.^1-3 Stage 1 assessment occurs when a patient is under anesthesia, before the leg is draped, which allows surgeons to check the reducibility of the deformity without referring to the lower leg mechanical axis. The evaluation at stage 1 assessment provides a general indication of the contracture rather than an exact measurement. Stage 2 assessment is made after the distal femur is resected. At this time, surgeons are able to use distracters, as well as alignment guides, to give an accurate indication of how much contracture remains and what soft tissue releases must be balanced. Stage 3 assessment is performed after the proximal tibia is resected, when all peripheral rim osteophytes have been removed. Subsequently, the knee is re-extended, and the capsular sleeve is tensed with two laminar spreaders to manage residual imbalance.

This traditional method of balancing ligaments varies slightly because it is based on the surgeon’s intuition. Surgeons consider a well-balanced knee and a well-aligned lower leg impossible to obtain if no accurate reference to the mechanical axis is created during the operation. Computer-assisted TKA gives surgeons permanent access to a mechanical axis reference.⁴

The aim of this study was to evaluate the role of navigation using the OrthoPilot (B. Braun Aesculap, Tuttlingen, Germany) to predict soft tissue release during 90 computer assisted-total knee TKA procedures performed between December 2001 and December 2003.


Figure 1: Manual stress exerted in valgus or varus knees at 10° flexion before prostheses implantation.	Figure 2: Manual stress exerted in valgus after medial soft tissue release with data recorded after load line was acquired.

Materials and Methods

The study group included 57 women and 33 men with a mean age of 73±6.13 years (range: 60-90 years). The series comprised 75 knees with genu varum deformities, 13 knees with genu valgum deformities, and 2 knees without any axial deformity (femoropatellar arthrosis). The mean hip-to-knee angle was 171.24°±3.81° (range: 163°-178°) for genu varum knees, and 186.15°±5.19° (181°-200°) for genu valgum knees.

The reducibility of the deformity was evaluated by the load line after exerting manual stress in valgus or in varus at 10° of flexion before implanting the prosthesis (Figure 1). Data were recorded after the load line was acquired, before removing the osteophytes, and were compared with the data required to perform a soft tissue release (Figure 2). The authors considered a knee sufficiently balanced and aligned when a side-to-side difference inferior to 4°, with a mechanical axis at 180°, was achieved (Figure 3). Fourteen soft tissue releases were performed during the operation: 13 soft tissue releases for genu varum, 9 for deep medial collateral ligament (12%), 4 for deep and superficial medial collateral ligament (5.3%), and 1 for genu valgum deformity (7.7%).

The results were classified into three groups: overcorrection, normalcorrection, and undercorrection. In the undercorrection group, four subgroups were identified: subgroup A, or undercorrection of 3° (177°-179°), subgroup B, or undercorrection of 5° (175°-176°), subgroup C, or undercorrection of 8° (172°-174°) and subgroup D, or undercorrection of 8° (171°).

Figure 3: Computer-navigated image of balanced and aligned knee.

Results

Forty-one cases, including 11 genu valgum, were overcorrectable (45.5%). Twenty cases, including 1 genu valgum, were neutral reducible (22.2%). No soft tissue releases were performed in both sets of cases.

Twenty-nine cases presented a nonreducible deformity: 15 in group A (16.6% of all 90 patients), 7 in group B (7.8%), 6 in group C (6.6%), and 1 in group D. In group A, deep medial collateral ligament releases were performed in 3 of 15 cases (20%). In group B, deep medial collateral ligament releases were performed in 4 of 7 cases (57%). In group C, medial collateral ligament soft tissue releases were performed in all 6 cases. In 3 cases, the authors performed a deep and superficial medial collateral ligament release (50%); in 2 cases, the authors performed a deep medial collateral ligament release only; and in 1 case, the authors performed a posterolateral corner release (genu valgum deformity). In group D, in 1 case, the authors performed a deep and superficial mediocollateral ligament release.

Discussion

Navigation has proven to be useful in predicting soft tissue release during TKA. When the reducibility of the deformity was compared with the lower leg mechanical axis obtained by computer, the authors found no need to perform a release in 67.7% of the cases because the lower leg was well aligned. The results were not subjective but based on objective data obtained from the computer. For underreducibility, soft tissue release was performed selectively after implanting the prosthesis. In case of no reducible deformity inferior to 3°, the need to perform a deep medial collateral ligament release reached 20%. In case of no reducible deformity inferior to 5°, the need to perform a deep medial collateral ligament release reached 57%. In both groups, a deep and superficial medial collateral ligament release was not necessary. In the case of no reducible deformity inferior to 8°, the need to do a medial collateral ligament release reached 100%; 66.6% of cases had a large release (deep and superficial medial collateral ligament release for the medial side and posterolateral corner for the lateral side, and 33.4% had a small release (deep medial collateral ligament release). In one case of undercorrection over 8°, a large release was necessary.

Engh⁵reported performing a release in 50% of cases for genu varum deformity. In this series, the incidence of medial collateral ligament release was 17.3% (13 of 75 cases) with a large release in only 5.3% of the cases (4 of 75 cases).⁵ The difference is extremely significant and could arguably support the fact that computer navigation decreases the need to perform a release. Although the series included only 13 cases of genu valgum, the need to perform a release was low (7.7%), and this excellent result must be supported by larger study samples.

Finally, this study confirms that systematic release in severe genu varum or genu valgum must be avoided if information on the reducibility of the deformity is not available.

Conclusion

Navigation can predict the need to perform a release during computer-assisted TKA. The study investigators deem it important to check the reducibility of the deformity before performing any release. The less the deformity is reducible, the greater the need to perform a release. Navigation reduces the incidence of large soft tissue release, as observed in this series.

References

Laskin RS. Total Knee Replacement. Berlin and Heidelberg, Germany; New York, NY: Springer Verlag; 1991.
Insall JN. Surgical approaches to the knee. In: Insall JN, ed. Surgery of the Knee. New York, NY: Churchill Livingstone; 1984:b;41-54.
Freeman MAR. Arthritis of the Knee. Berlin and Heidelberg, Germany; New York, NY: Springer Verlag; 1980.
Saragaglia D, Picard F. Computer assisted implantation of total knee endoprosthesis with no preoperative imaging: the kinematic model. In: Stiehl JB, Konerman WH, Haaker RG, eds. Navigation and Robotics in Total Joint and Spine Surgery. Berlin and Heidelberg, Germany; New York, NY: Springer Verlag; 2003:226-233.
Engh GA. The difficult knee: severe valgus and varus. Clin Orthop Relat Res. 2003; 416:58-63.

Authors

Drs Saragaglia, Chaussard, and Rubens-Duval are from the Department of Orthopedic Surgery and Sport Traumatology, Greonoble University South Teaching Hospital, Echirolles, France.