Original Article
Computer Navigation of Soft Tissues in Total Knee Replacement Yogeesh D. Kamat, MD (Res), MS, DNB, FRCS (Orth) 1 Ajeya R. Adhikari, MS, MCh, FRCS (Orth) 3 1 Department of Orthopaedics, Elective Orthopaedic Centre, Epsom,
United Kingdom 2 Department of Orthopaedics, Elective Orthopaedic Centre, Epsom, United Kingdom 3 Department of Orthopaedics, Elective Orthopaedic Centre, London, United Kingdom
Kamran M. Aurakzai, MRCS 2
Address for correspondence Yogeesh D. Kamat, MD (Res), MS, DNB, FRCS (Orth), Department of Orthopaedics, Elective Orthopaedic Centre, Epsom, Surrey, KT18 7EG, United Kingdom (e-mail:
[email protected]).
J Knee Surg 2013;26:145–150.
Abstract
Keywords
► computer navigation ► TKR ► tissue balancing
Following the success of computer navigation in producing consistently accurate alignment, the focus has shifted to use of these techniques for soft tissue assessment during total knee replacement (TKR). We undertook a prospectively randomized clinical study to compare two methods of tissue balancing in TKR. One method, called bone referencing (BR) employed independent cutting of the femur and tibia followed by subjective assessment with trial prostheses and soft tissue release as deemed necessary. The other method, termed ligament balancing (LB), involved cutting the tibia first and titration of tissue balance and alignment parameters to guide femoral cuts. Our total sample comprised 77 subjects with 80% statistical power. To assess tissue balance we used (a) coronal laxity testing and (b) computer navigation generated passive knee range of movement graphs. The graphical assessment was validated with coronal laxity testing. There was no difference between the resultant tissue balances achieved. However, correlation with preoperative status revealed the LB technique to show better results in a smaller subgroup of knees with greater preoperative tissue imbalance. We advocate variation of tissue balancing technique to suit the individual knee, based on preoperative assessment, to achieve an optimal result in all TKR.
Soft tissue balancing and bony alignment are intimately related and both are critical to the success of a total knee replacement (TKR). Use of computer navigation has been demonstrated to result in increased consistency of accurate prosthetic placement and other more immediate advantages including decreased blood loss1–3 in TKR surgery. The importance of assessment and balancing of soft tissues is well recognized in the literature. Measurement of balance during TKR has been mostly restricted to subjective coronal plane laxity assessment by the surgeon. Besides, prosthetic alignment has a confounding effect on tissue balance assessment. Use of computer navigation has brought forth a dual advantage: (a) alignment being consistently accurate, variables on
soft tissue balancing can be now reliably assessed and (b) recently, software has been developed to enable generation of kinematic graphs depicting intraoperative alignment and balance throughout the range of movement.4–6 Coronal plane laxity (varus and valgus) can also be measured accurately at different degrees of knee flexion.7 Two methods for soft tissue balancing in common use are as follows:
received March 4, 2012 accepted after revision May 30, 2012 published online July 30, 2012
Copyright © 2013 by Thieme Medical Publishers, Inc., 333 Seventh Avenue, New York, NY 10001, USA. Tel: +1(212) 584-4662.
(A) Bony cuts are made; femur then tibia independently based on alignment. Soft tissue balance is then assessed with spacer blocks or trial prostheses and tissue release performed if necessary. This technique was named as “Bone Referencing(BR).”
DOI http://dx.doi.org/ 10.1055/s-0032-1322600. ISSN 1538-8506.
145
146
Computer Navigation of Soft Tissues
Kamat et al.
(B) The tibia is first cut. Femoral cuts are then made using assessment of flexion and extension gaps as well as keeping alignment along mechanical axis in mind. Tissue releases, if required are performed simultaneously with femoral cuts. This was named as the “Ligament Balancing (LB)” technique. Our aims in undertaking this study were as follows: (i) to use intraoperative quantitative assessment of knee soft tissue balance with current computer navigation software and (ii) compare the two methods of soft tissue balancing used in TKR.
Materials and Methods We undertook a prospective randomized study of patients undergoing TKR for osteoarthritis by a single surgeon at our elective orthopaedic center. Sample size calculations were performed following a pilot study and suggested a requirement of 38 in each group to yield 80% statistical power. Approval was obtained from the Regional Ethics Committee and our Institutional Review Board. The computer navigation system used was Galileo (Smith & Nephew Orthopaedics). This system incorporates a minirobot that helps place the femoral cutting block. All bony cuts were navigated to an accuracy of 1 degree and 1 mm. Standard midline skin incision and medial para-patellar approach was used in all TKR. The fixed reference arrays were applied to the tibia and femur and all registration points acquired in the standard fashion. A passive knee range of movement graph was obtained looking at varus/valgus alignment throughout the range of motion and rotation of the femur relative to the tibia. Varus and valgus stresses were applied manually to assess coronal stability of the knee in 0 and 30 degrees of flexion. All this was done before any bony cuts or ligament release. Measurements in both parameters described above were referred to as “baseline.” In the case of the LB technique, the tibial cut was made first. A force feedback ligament tensioner that assessed the gap in millimeters and force in Newtons was used. Assessment of flexion and extension gaps was performed using the software before any soft tissue release and bony cuts were made. The flexion gap was then made rectangular by undertaking appropriate bony cuts (dorsal femoral condyles) and an additional soft tissue release if required. The extension gap was then matched accordingly. The advantage of using computer navigation with the tensioner was precise measurement of the following: (i) rotation of the femoral component so as to keep alignment within acceptable range, (ii) amount of bone resected from the distal and posterior aspects of the femur, and (iii) tissue tensions on the medial and lateral sides of the knee. Soft tissue release could thus be fine-tuned. In case of the BR technique, the femoral and then tibial cuts were made independent using navigation. Femoral external rotation was constantly kept at 3 degrees (measured by posterior condylar referencing and verifying using the trans-epicondylar axis). The Journal of Knee Surgery
Vol. 26
No. 3/2013
Soft tissues were then subjectively assessed with trial prostheses, and releases were performed if required. In both techniques, passive range of movement graphs and varus/ valgus stress testing were repeated at this stage using trial prostheses. Measurements from there were referred to as “postbone resection.” The same operating surgeon tested coronal laxity in all cases, so as to minimize variation in the force applied. Readings were taken off the computer screen as stress was applied. Passive range of movement graphs were all plotted in a standard fashion by the navigation software. These graphs denoted: (a) alignment of the limb in relation to the mechanical axis and (b) relative rotation of femur to tibia as the knee was passively flexed (►Fig. 1). An independent observer blinded to the technique of operation made measurements from the graph plots. These included the following: (i) (from the “baseline” graph) standard deviation of the values on the plot at 0, 30, and 90 degrees knee flexion so as to give the degree of skewing, indicating the preoperative soft tissue status of the arthritic knee; and (ii) (from the “postbone resection” graph) total area of deviation of the plot from a straight line indicating residual bounciness or imbalance in the soft tissues. The intraoperative parameters gathered with the assistance of computer navigation were all recorded in an electronic database (Microsoft access). Parametric t-tests were used for comparison of results with p value < 0.05 taken as level of
Figure 1 (a) is ’baseline’ and (b) is a ’postbone resection’ graph of passive knee movement plotted with the computer navigation software. The X axis denotes range of movement and Y axis denotes (i) mechanical axis alignment in the coronal plane (valgus/varus- the light gray) and (ii) tibia rotation relative to femur (dark gray).
Computer Navigation of Soft Tissues
Kamat et al.
Table 1 Preoperative Status of Both Groups Parameter
Bone Referencing
Ligament Balancing
Av. age at operation
69.3 years (40–86, SD 10.3)
72.1 years (49–85, SD 9.6)
Sex (No. of Males/Females)
12 males/27 females
17 males/21 females
Av. BMI
28.9 (21–34, SD 6.13)
29.4 (22–35, SD 4.51)
Preop Oxford Knee Score
40.1 (30–55, SD 7.8)
39.2 (30–52, SD 6.9)
Preop Knee Society Score (Objective)
35.9 (14–78, SD 14.5)
33.7 (2–68, SD 16.1)
Preop Knee Society Score (Functional)
48.3 (15–80, SD 12.3)
52.1 (15–80, SD 11.9)
Av., average; BMI, body mass index; SD, standard deviation; Preop, preoperative.
significance. A regression analysis was undertaken to correlate standard deviation of baseline graph plots with the degree of residual imbalance in the postbone resection ones.
Results Total 77 cases were performed: 39 in the BR group and 38 in the LB group. ►Table 1 shows a list of preoperative parameter averages in both the groups. Statistical comparison was not performed on preoperative data, as this was a randomized trial. However there was no obvious difference between the two groups.
Coronal Laxity Measurements ►Table 2 shows average total coronal laxity measurements at full extension and 30 degrees flexion at the postbone resection stage in both groups. There was no significant difference. An excellent correlation was found between the following:
(a) mean coronal laxity calculated from values obtained by stress testing, and; (b) points on the baseline and postbone resection passive kinematics graph plots at corresponding points of knee flexion. This served as a validation of our measurements from the graphs. The correlation coefficients were as follows: (A) baseline: (1) full extension–0.91 and (2) 30 degrees knee flexion–0.94; (B) postbone resection: (1) full extension–0.95 and (2) 30 degrees knee flexion–0.94.
Results from Kinematics Graphs ►Table 3 shows the average measurements made from the passive kinematic graphs. The t-test did not reveal any significant difference between the residual tissue imbalances using both techniques, that is, both techniques achieved a similar overall soft tissue balance. A regression analysis was undertaken to correlate standard deviation values on the baseline kinematics graph plot and the value of residual tissue imbalance in the postbone resection graph (►Fig. 2a, b). In the case of BR a moderate correlation was found (correlation coefficient r ¼ 0.44). On the other hand, the LB technique demonstrated a negative slope (correlation coefficient r ¼ -0.03) that is, increasing baseline standard deviation made no difference to the magnitude of postbone resection residual imbalance in these knees. The baseline passive kinematics graphs were all qualitatively analyzed, as different patterns of graph plots were observed. When comparisons between the individual subgroups of complex kinematic patterns were made, the LB technique was found to result in better tissue balance in these smaller numbers.
Operative Data and Postoperative Recovery The LB technique resulted in greater variation in external rotation of the femoral component (►Fig. 3a, b). We did not have any case of patellar maltracking (assessed clinically intraoperatively). The mean tourniquet times for the BR and LB groups were 77 minutes (range: 65 to 110 minutes) and 86 minutes (range: 65 to 120 minutes), respectively. This was a significant difference (p ¼ 0.001). The mean number of days till the
Table 2 Comparison of Total Coronal Laxity Bone Referencing
Ligament Balancing
p Value
Total coronal laxity full extension
2.07 (0–6, SD 1.19)
2.07 (0–4, SD 1.09)
0.93
Total coronal laxity 30 degrees flexion
2.65 (1–7, SD 1.33)
2.75 (1–6, SD 1.35)
0.98
The Journal of Knee Surgery
Vol. 26
No. 3/2013
147
148
Computer Navigation of Soft Tissues
Kamat et al.
Table 3 Comparison of Residual Tissue Imbalance from Passive Kinematics Graphs
Postop mean residual tissue imbalance
Bone Referencing
Ligament Balancing
p Value (t-Test)
3.62 (SD ¼ 3.97)
2.82 (SD ¼ 3.25)
0.33
SD, standard deviation; Postop, postoperative.
The development of computer navigation assistance in TKR can be viewed as having taken place in three phases. Phase I was the initial success in producing consistently accurate prosthetic alignment. Clinical outcome studies (phase II) however failed to provide better performance for computer navigated TKR8,9 even at mid-term follow- up.10 Dedicated navigators have nevertheless explored new avenues and embarked upon exciting prospects (phase III). One is objective measurement of soft tissue balance. Computer navigation techniques have been able to quantify soft tissue releases7,11 whereas in previous times, assessment of soft tissues was surgeon-dependent and description very subjective. The other exciting prospect is exploration of passive knee kinematics with intraoperative dynamic feedback. Even though we are unable to study the effect of muscle forces that come
into play with weight-bearing, the assessment of passive knee motion provides clues at a time when the knee is amenable to adjustments. Comparison between the two techniques of soft tissue balancing has been undertaken in the laboratory setting in small numbers of knees.12 To the authors’ best knowledge, this is the first clinical comparison of the two techniques with an adequate sample size. Our study has been possible due to developments in computer navigation software that have: (a) enabled use of both techniques for the same design of prosthesis and (b) provided means of measuring outcome quantitatively so as to enable statistical testing. Navigation assisted assessment besides increasing accuracy, also made measurements easily reproducible for every individual knee in the study. A potential criticism might be the use of a manual force for coronal laxity testing. However Ritschl et al13 demonstrated that when a minimum force of 100N is achieved, then further increase in manual force does not alter the coronal deviation much. Their cadaveric study revealed that between forces of 80 and 200N, change in the lengths of the collateral soft tissue structures (and consequently joint opening) of the knee was
Figure 2 (a): Correlation between SD (preop) status and postbone resection residual imbalance in the bone referencing group. (b): Correlation between SD (preop) status and postbone resection residual imbalance in the ligament balancing group. Preop, preoperative.
Figure 3 External rotation given to the femoral component (posterior condylar referencing) with the use of both techniques of balancing: (a) bone referencing and (b) ligament balancing.
operative wounds were dry were 4.3 and 4.7 in the BR and LB groups, respectively. The difference was not statistically significant (p ¼ 0.59).
Discussion
The Journal of Knee Surgery
Vol. 26
No. 3/2013
Computer Navigation of Soft Tissues minimal. We found it difficult to restrain hip rotation that is induced when coronal stress is applied with the knee in flexion and have not tested coronal laxity beyond 30 degrees knee flexion. Nevertheless, excellent correlation of coronal laxity readings, with corresponding points on the passive kinematics graph plots, validated use of the kinematic graphs in accurately measuring the total laxity/imbalance of the knee in the whole range of motion under consideration (0 to 90 degrees flexion). The ability to track the relationship between femur and tibia in the coronal, sagittal, and axial planes has allowed navigation software to generate kinematic graphs throughout the range of movements. In our study, these graph plots enabled detection of subtle differences in laxity that would not have been possible with coronal laxity readings alone. Retrospective evaluation of patterns of kinematic graphs also provides clues to kinematic types of knees. Siston et al, having evaluated passive knee kinematics in a similar fashion, describe different patterns of knee movement.6 Improvements in techniques to evaluate knee kinematics14 could potentially make computer navigation assisted kinematics assessment comparable to the gold standard fluoroscopic evaluation.15 There are advantages and disadvantages of both soft tissue balancing techniques. Individual surgeons prefer a particular technique depending on their training and experience. The BR technique is relatively simple, and took significantly less time in our hands. However it involves ligament assessment quite late in the procedure, when bony cuts have already been performed. Making further alterations might be difficult at that stage. Fehring showed that using fixed bony landmarks can result in a trapezoidal joint gap in "40% of the cases16 with possible consequences of polyethylene wear or limited range of motion. The LB technique attempts to achieve equilibrium between alignment and soft tissue tension before making bony cuts. LB took longer but may be suitable in a certain subset of knees that present difficulties for soft tissue balancing. An important finding of our study was the close relationship between bony cuts and soft tissue release. Conventionally, 3 degrees external rotation of the femoral component has been shown to produce a rectangular flexion gap.17,18 Our results with the LB technique show that we reproduced 3 degrees external rotation in only 15 out of 38 TKR, that is, "40%. The patellofemoral articulation and tracking19,20 are not affected by controlled variation of femoral external rotation. This implies that adjusting external rotation of the femur helps balance the flexion gap with much less soft tissue release. The LB technique has been discussed at length in relation to the aim of this paper, which is a comparative analysis of the two techniques. It is possible using a tensioner, without the assistance of computer navigation. We have noted the variation in external rotation but for the first time have been able to measure it, which is what navigation brings to the operating theater of the future. The focus for current developments in the field of computer navigation is the planning of soft tissue management.21 Complex play of soft tissues might produce bad results in seemingly well-aligned and nondeformed knees. Computer navigation can provide much more than accurate prosthetic
Kamat et al.
alignment: it can be used to assess the arthritic knee before embarking on bone cuts. It would help define subsets of knees that might require special operative attention to improve outcome as opposed to the ones that will be managed with routine techniques of TKR surgery. The operative technique might accordingly be catered to suit a particular knee to produce a good end result in all.
Conclusion Measurement of soft tissue balance using passive range of movement kinematics graphs correlates very well with coronal laxity testing. There is no significant difference in soft tissue balance achieved using the two standard techniques. However, the measured tension technique provides better balancing in knees with greater baseline tissue imbalance. Intraoperative passive kinematic graphs and stress data analysis can help in identifying this subset of knees and are continuing to provide means of better understanding the soft tissue envelope.
References 1 Bolognesi M, Hofmann A. Computer navigation versus standard
2
3
4 5
6
7
8
9
10
11
12
instrumentation for TKA: a single-surgeon experience. Clin Orthop Relat Res 2005;440:162–169 Chauhan SK, Clark GW, Lloyd S, Scott RG, Breidahl W, Sikorski JM. Computer-assisted total knee replacement. A controlled cadaver study using a multi-parameter quantitative CT assessment of alignment (the Perth CT Protocol). J Bone Joint Surg Br 2004; 86(6):818–823 Picard F, Deakin AH, Clarke JV, Dillon JM, Gregori A. Using navigation intraoperative measurements narrows range of outcomes in TKA. Clin Orthop Relat Res 2007;463:50–57 Chen E, Ellis RE, Bryant JT, Rudan JF. A computational model of postoperative knee kinematics. Med Image Anal 2001;5(4):317–330 Martelli S, Ellis RE, Marcacci M, Zaffagnini S. Total knee arthroplasty kinematics. Computer simulation and intraoperative evaluation. J Arthroplasty 1998;13(2):145–155 Siston RA, Giori NJ, Goodman SB, Delp SL. Intraoperative passive kinematics of osteoarthritic knees before and after total knee arthroplasty. J Orthop Res 2006;24(8):1607–1614 Picard F, Deakin AH, Clarke IV, Dillon JM, Kinninmonth AW. A quantitative method of effective soft tissue management for varus knees in total knee replacement surgery using navigational techniques. Proc Inst Mech Eng H 2007;221(7):763–772 Martin A, von Strempel A. Two-year outcomes of computed tomography-based and computed tomography free navigation for total knee arthroplasties. Clin Orthop Relat Res 2006;449:275–282 Spencer JM, Chauhan SK, Sloan K, Taylor A, Beaver RJ. Computer navigation versus conventional total knee replacement: no difference in functional results at two years. J Bone Joint Surg Br 2007; 89(4):477–480 Kamat YD, Aurakzai KM, Adhikari AR, Matthews D, Kalairajah Y, Field RE. Does computer navigation in total knee arthroplasty improve patient outcome at midterm follow-up? Int Orthop 2009;33(6):1567–1570 Lüring C, Hüfner T, Perlick L, Bäthis H, Krettek C, Grifka J. [Soft tissue management in knees with varus deformity. Computerassisted sequential medial ligament release]. Orthopade 2005; 34(11):1118–1124, 1120–1122, 1124 Mihalko WM, Whiteside LA, Krackow KA. Comparison of ligament-balancing techniques during total knee arthroplasty. J Bone Joint Surg Am 2003;85-A(Suppl 4):132–135 The Journal of Knee Surgery
Vol. 26
No. 3/2013
149
150
Computer Navigation of Soft Tissues
Kamat et al.
13 Ritschl P, Machacek F, Gruber F, Fuiko R, Wulser M. Mechanical
Properties of Soft Tissues relevant for Ligament Balanced TKR measured in vitro. Poster presented at: CAOS International 2006, Montreal 14 Godest AC, de Cloke CS, Taylor M, et al. A computational model for the prediction of total knee replacement kinematics in the sagittal plane. J Biomech 2000;33(4):435–442 15 Banks SA, Hodge WA. Accurate measurement of three-dimensional knee replacement kinematics using single-plane fluoroscopy. IEEE Trans Biomed Eng 1996;43(6):638–649 16 Fehring TK. Rotational malalignment of the femoral component in total knee arthroplasty. Clin Orthop Relat Res 2000;380(380): 72–79
The Journal of Knee Surgery
Vol. 26
No. 3/2013
17 Whiteside LA. Soft tissue balancing: the knee. J Arthroplasty
2002;17(4, Suppl 1):23–27 18 Whiteside LA. Positioning the femoral component: the effect of
proper ligament balance. Am J Knee Surg 2000;13(3):173–180 19 Bottros J, Gad B, Krebs V, Barsoum WK. Gap balancing in total knee
arthroplasty. J Arthroplasty 2006; 21(4, Suppl 1):11–15 20 Wyss TF, Schuster AJ, Münger P, Pfluger D, Wehrli U. Does total
knee joint replacement with the soft tissue balancing surgical technique maintain the natural joint line? Arch Orthop Trauma Surg 2006;126(7):480–486 21 Clemens U, Miehlke RK. Advanced navigation planning including soft tissue management. Orthopedics 2005;28(Suppl 10): s1259–s1262
