Healio
Healio
October 01, 2008
12 min read
Save

Unicompartmental Knee Replacement: A Comparison of Four Techniques Combining Less Invasive Approach and Navigation

Abstract

We developed a nonimage-guided navigation system for unicompartmental knee replacement, suitable for both conventional and minimally invasive approaches. We performed a radiologic analysis of the accuracy of implantation with conventional nonnavigated instrumentation, conventional, open navigated instrumentation, minimally invasive navigated experimental instrumentation derived from conventional instruments, and minimally invasive navigation-dedicated instrumentation. Navigated technique allowed improving the accuracy of the radiologic implantation. Minimally invasive implantation was effective, but the accuracy may not reach that of the conventional navigated technique. Minimal invasive techniques have to be validated, because a loss of accuracy will negatively influence long-term outcomes.

The accuracy of implantation is an accepted prognostic factor for the long-term survival of unicompartmental knee replacement (UKR).1 However, most UKR systems offer limited and potentially inaccurate instrumentation that relies on substantial surgeon judgment for prosthesis placement. Rates of inaccurate implantation as high as 30% have been reported with conventional, free-hand instrumentation.2 An intramedullary femoral guiding device can improve these results,3 but does not allow reproducible optimal implantation.

Computer-assisted systems have been developed for total knee replacement (TKR) and have proved to allow a higher precision of implantation for such implants compared with conventional instruments.4 The OrthoPilot system (B. Braun Aesculap, Tuttlingen, Germany) has also been validated in clinical use by a prospective, randomized study.5 This system is considered nonimage based, because it relies only on an intraoperative kinematic analysis of the lower limb.

We developed an adaptation of this technique for unicompartmental knee prosthesis (UKP) implantation, without any extramedullary or intramedullary guiding device, suitable for both conventional and minimally invasive approaches. We hypothesized that the navigation system will allow for placement of the prosthesis in a better position than that accomplished with the conventional technique, and that the minimally invasive navigated approach will not decrease the accuracy of the procedure. This study reports the radiologic results of four groups of patients who underwent UKP implantation with conventional nonnavigated instrumentation: conventional open navigated instrumentation; minimally invasive navigated experimental instrumentation derived from conventional instruments; and minimally invasive navigation-dedicated instrumentation.

Operative Techniques

Conventional Nonnavigated Instrumentation

The conventional technique has been described more extensively elsewhere.3 After a medial parapatellar approach, typically 18 cm in length, the tibial resection guide was fixed on an extramedullary rod, after visual alignment with the tibial axis on both coronal and sagittal planes. The guide was pinned on the tibia, and proximal tibial resection was performed with an oscillating saw, preserving the tibial attachment of both cruciates. The femoral canal was entered at the most proximal point of the intercondylar notch, and an intramedullary rod was fixed in the femoral canal, representing the femoral coronal and sagittal anatomical axes. A distal femoral resection guide was fixed on this rod with a coronal orientation defined on preoperative long leg radiographs according to the angle between the mechanical axis and the anatomic axis of the femur, and distal femoral resections were performed with an oscillating saw. A second femoral guide was applied on this distal resection to perform the dorsal femoral resection and the chamfer resection.

Conventional Open Navigated Technique

The navigation system used is an intraoperative nonimage-based one (OrthoPilot; B. Braun Aesculap, Tuttlingen, Germany).6 After a medial parapatellar approach, typically 18 cm in length, two infrared localizers were placed on screws in the distal femur and in the proximal tibia and one strapped on the dorsal part of the foot. The relative motion of two adjacent localizers was tracked by an infrared camera (Polaris; Northern Digital, Toronto, Canada). The dedicated software calculated the center of rotation of this movement and so defined the respective joint center of the hip, knee, and ankle joints. These centers were use to calculated the mechanical axes of both the femur and tibia in both the coronal and sagittal planes. A localizer was then fixed on tibial or femoral resection blocks, and the software displayed in real time the orientation of these blocks compared with the mechanical leg axes. The surgeon can fix the block with the desired orientation before performing the bony resection with a classical motorized saw blade. The trial implants were tested, and the definitive prosthesis was cemented if the test was satisfactory.

Minimally Invasive Experimental Navigated Technique

The same nonimage-based navigation system was used, but the instruments were modified to allow their placement through an 8-cm skin incision. However, the software had to be modified because the minimally invasive approach did not allow the direct palpation of the lateral femorotibial joint. The position of the lateral articular points was calculated by the software with help of the radiographic preoperative planning.

Minimally Invasive Navigation-dedicated Technique

This technique has been described more extensively elsewhere.7 The software is basically the same as for the minimally invasive experimental navigated technique. The procedure begins with a quadriceps-sparing medial arthrotomy, typically 6 cm in length. Kinematic registration is performed as usual. Anatomic registration is limited to the medial femorotibial joint. The tibial resection guide is oriented using a hands-free technique. A navigated bow is fixed with two bicortical screws on the distal femur and oriented along the knee flexion-extension axis. On this bow, the distal and posterior resection guides are fixed and oriented according to the navigation system but not fixed directly within the joint. Resections are performed with a saw blade for the posterior resection and a burr for the distal resections.

Materials and Methods

Five hundred seventy-four patients have undergone a medial osteoarthritis at the investigators’ institution from January 1996 to December 2006 with implantation of a UKR for medial osteoarthritis comprising 256 cases with the conventional manual technique (group A), 90 cases with the conventional navigated technique (group B), 108 cases with the experimental minimally invasive technique (group C), and 120 cases with the minimally invasive navigation-dedicated technique (group D), successively. Two different protheses were used, the Search UKR (B. Braun Aesculap) in groups A, B, and C, and the Univation UKR (B. Braun Aesculap) in group D. Both prostheses were designed to be implanted as follows: coronal femorotibial mechanical angle of 0º to 5º of remaining varus deformation, coronal orientation of the femoral component of 90º ± 2º compared with the coronal femoral mechanical axis, sagittal orientation of the femoral component of 90º ± 2º (Search) or 80º ± 2º (Univation) compared with the distal anterior femoral cortex, coronal orientation of the tibial component of 90º ± 2º compared with the coronal tibial mechanical axis, and sagittal orientation of the tibial component of 88º ± 2º compared with the proximal posterior tibial cortex. All patients underwent a complete radiologic examination in the first 3 months after the index procedure, with anteroposterior (AP) and lateral plain knee radiographs and AP and lateral long leg radiographs.

Thirty UKRs in each group were randomly selected and compared. The following angles were measured on long leg Radiographs by a single observer (J.Y.J.): mechanical femorotibial angle (normal = 0°, varus deformation was described with a positive angle); coronal orientation of the femoral component compared with the mechanical femoral axis (normal = 90°, varus deformation was described with an angle < 90°); sagittal orientation of the femoral component compared with the distal anterior femoral cortex (normal = 90°, flexion deformation was described with an angle < 90°); coronal orientation of the tibial component compared with the mechanical tibial axis (normal = 90°, varus deformation was described with an angle < 90°); and sagittal orientation of the tibial component compared with the proximal posterior tibial cortex (normal = 90°, flexion deformation was described with angle < 90°).

Table 1: Radiographic Results

Individual analysis was performed as follows: one point was given for each fulfilled item, giving a maximal accuracy note of 5 points. The accuracy note was compared among all groups with an ANOVA test with post-hoc Bonferrini-Dunn correction. Prosthesis implantation was considered satisfactory when the accuracy note was 5 (all fulfilled items); the rate of satisfactory implanted prostheses was compared in all groups with a chi-square test. Mean angular values in all groups were compared for each criterion with an ANOVA test with post-hoc Bonferrini-Dunn correction; the sagittal orientation of the femoral component of the group D was corrected to compensate for the different goal. The rate of prostheses implanted within the desired range for each criterion was also compared in all groups with a chi-square test. All statistical tests were performed with a .05 limit of significance.

Results

A total of 120 patients were selected (45 men); mean age was 67 years (SD, 6). Mean body mass index was 29.6 (SD, 4.5). Preoperative pain Knee Society Score (KSS) was 56 points (SD, 12), and preoperative functional KSS was 61 points (SD, 12). Mean preoperative coronal femorotibial mechanical angle was 7.8º (SD, 5.1). There were 54 grade 2, 59 grade 3, and 7 grade 4 degenerative changes according to Ahlback.8 There were no significant differences in any preoperative parameter among all groups.

Radiographic results at the early follow-up are reported in Tables 1 and 2.

Mean global accuracy note was 1.5 (SD, 1.2) in group A, 4.5 (SD, 0.6) in group B, 3.3 (SD, 1.2) in group C, and 4.2 (SD, 1.1) in group D (P < .001). The rate of perfect implantation was 6/30 in group A, 18/30 in group B, 13/30 in group C, and 18/30 in group D (P < .001). Mean femorotibial angle was 0.9º (SD, 4.0) in group A, 1.5º (SD, 2.2) in group B, 1.3º (SD, 2.1) in group C, and 2.6º (SD, 2.7) in group D (NS). The rate of fulfilled item was 20/30 in group A, 25/30 in group B, 22/30 in group C, and 26/30 in group D (NS).

Mean coronal orientation of the femoral component was 88.0º (SD, 2.9) in group A, 89.1º (SD, 1.4) in group B, 91.1º (SD, 5.0) in group C, and 88.0º (SD, 3.0) in group D (NS). The rate of fulfilled item was 21/30 in group A, 26/30 in group B, 23/30 in group C, and 27/30 in group D (NS).

table 2: Radiogrpahic Results

Mean sagittal orientation of the femoral component was 89.3º (SD, 2.8) in group A, 89.6º (SD, 1.6) in group B, 87.6º (SD, 3.5) in group C, and 81.6º (SD, 4.2) in group D (NS). The rate of fulfilled item was 21/30 in group A, 27/30 in group B, 21/30 in group C, and 26/30 in group D (NS).

Mean coronal orientation of the tibial component was 88.2º (SD, 2.6) in group A, 89.1º (SD, 1.4) in group B, 91.1º (SD, 5.0) in group C, and 87.8º (SD, 2.7) in group D (NS). The rate of fulfilled item was 22/30 in group A, 28/30 in group B, 24/30 in group C, and 26/30 in group D (NS).

Mean sagittal orientation of the tibial component was 86.4º (SD, 3.2) in group A, 89.6º (SD, 1.3) in group B, 87.6º (SD, 3.5) in group C, and 87.8º (SD, 2.9) in group D (NS). The rate of fulfilled item was 21/30 in group A, 28/30 in group B, 24/30 in group C, and 26/30 in group D (NS).

Subgroup analysis showed a significant difference between global accuracy note in group A vs group B (P < .001), group C (P = .05), and group D (P < .001) and between rate of perfect implantation in group A vs group B (P < .001), group C (P = .05), and group D (P < .001).

All other subgroup differences were not significant.

Discussion

The restoration of the physiologic alignment of the lower limb is an accepted prognostic factor for long-term survival of a TKR.9 Navigation systems have proven to improve the accuracy of implantation of a TKR.4-6,10 The precision of the system used was experimentally calculated to be of 1° for angle measurement and of 1 mm for distance measurement.11 However, outliers still can occur. There are several possible additional reasons for the observed errors including lack of precision of the radiologic measurement technique, lack of rigid fixation of the resection block on the bone, and lack of precise guiding of the saw blade by the resection blocks bending of the saw blade.12 These causes of errors are inherent in all systems; however, a modification of the reference software should not introduce other causes of errors.

UKR is a valuable alternative to high tibial osteotomy13 or TKR for the treatment of isolated medial osteoarthritis.14 However, the exact indications are still controversial, because some investigators have reported a low survival rate of such implants.15 Inaccurate implantation is a known factor for early failure.1 There is no general agreement on the ideal positioning of a UKR, and the positioning we wanted to achieve can only be seen as a personal opinion. However, the goal of an instrumentation is to allow surgeons to place the prosthesis in the position they choose. It is then valuable to compare the positioning of the three groups of UKR implantation techniques, which were expected to implant the prosthesis in the same position, whatever this position should be. Most instrumentation offers imprecise guiding systems depends primarily on the surgeon’s skill.16 Even intramedullary guiding systems do not offer reproducible optimal implantation technique.3

The conventional navigated instrumentation used in this study is similar to that used for TKR implantation. It has been shown to allow achieving a significantly more accurate implantation measured on postoperative radiographs compared with the manual technique. The accuracy of implantation was similar to that obtained with the reference TKR software.

The conventional manual and navigation techniques involve conventional skin incision and approach, with splitting of the vastus medialis and lateral subluxation of the patella. We developed a navigated minimally invasive technique, which allows performing the entire procedure through a shorter skin incision. Our first experience is interesting. All procedures succeeded with a 5- to 10-cm skin incision. We observed a trend toward decreased accuracy of implantation of the prosthesis with the experimental minimally invasive navigated technique (group C). The calculation of the location of the anatomic point was less precise than the direct palpation, and this point has been addressed in the development of the software. The results of the last version of the software (group D) seem to be as satisfactory as for the reference group with open navigated technique (group B).

We did not yet study the influence of the minimally invasive approach on rehabilitation time. This point has already been investigated, and minimally invasive procedures might allow an earlier discharge and faster rehabilitation.17

Follow-up for the navigated prostheses is currently too short to know whether clinical outcome or survival rates will be improved. Longer follow-up is required to determine the respective advantages and disadvantages of this techniques and the potential benefit of a minimally invasive implantation.

Conclusion

Navigated implantation of a UKR with the nonimage-based system used allowed improvement of the accuracy of the radiologic implantation without significant inconvenience and with little change in the conventional operative technique. Minimally invasive implantation was effective, but the accuracy may not reach that of the conventional navigated technique. Minimally invasive techniques have to be validated, because a loss of accuracy will negatively influence long-term outcomes.

References

  1. Cartier P, Sanouillier JL, Grelsamer RP. Unicompartmental knee arthroplasty surgery. 10-year minimum follow-up period. J Arthroplasty. 1996;11 (7):782-788.
  2. Tabor OB Jr, Tabor OB. Unicompartmental arthroplasty: a long-term follow-up study. J Arthroplasty. 1998;13 (4):373-379.
  3. Jenny JY, Boeri C. Accuracy of implantation of a unicompartmental total knee arthroplasty with 2 different instrumentations: a case-controlled comparative study. J Arthroplasty. 2002;17(8):1016-1620.
  4. Victor J, Hoste D. Image-based computer-assisted total knee arthroplasty leads to lower variability in coronal alignment. Clin Orthop Relat Res. 2004;(428):131-139.
  5. Saragaglia D, Picard F, Chaussard C, Montbarbon E, Leitner F, Cinquin P. Mise en place des prothèses totales du genou assistée par ordinateur: comparaison avec la technique conventionnelle.
  6. Rev Chir Orthop. 2001;87(1):18-28.
  7. Jenny JY, Boeri C. Unicompartmental knee prosthesis implantation with a non-image-based navigation system: rationale, technique, case-control comparative study with a conventional instrumented implantation. Knee Surg Sports Traumatol Arthrosc. 2003;11(1):40-45.
  8. Jenny JY, Ciobanu E, Boeri C. The rationale for navigated minimally invasive unicompartmental knee replacement. Clin Orthop Relat Res. 2007;463:58-62.
  9. Ahlback S. Osteoarthrosis of the knee. A radiographic investigation. Acta Radiol Diagn (Stockh). 1968; suppl.277; 1-72.
  10. Jeffery RS, Morris RW, Denham RA. Coronal alignment after total knee replacement. J Bone Joint Surg Br. 1991;73(5):709-714.
  11. Jenny JY, Boeri C. Implantation d’une prothèse totale de genou assistée par ordinateur. Etude comparative cas-témoin avec une instrumentation traditionnelle. Rev Chir Orthop . 2001;87(7):645-652.
  12. Delp SL, Stulberg SD, Davies B, Picard F, Leitner F. Computer assisted knee replacement. Clin Orthop Relat Res.1998; (354):49-56.
  13. Plaskos C, Hodgson AJ, Inkpen K, McGRaw RW. Bone cutting errors in total knee arthroplasty. J Arthroplasty. 2002;17(6):698-705.
  14. Schai PA, Suh JT, Thornhill TS, Scott RD. Unicompartmental knee arthroplasty in middle-aged patients: a 2- to 6-year follow-up evaluation. J Arthroplasty. 1998;13 (4):365-372.
  15. Newman JH, Ackroyd CE, Shah NA. Unicompartmental or total knee replacement? Five-year results of a prospective, randomized trial of 102 osteoarthritic knees with unicompartmental arthritis. J Bone Joint Surg Br. 1998;80(5):862-865.
  16. Bert JM. 10-year survivorship of metal-backed, unicompartmental arthroplasty. J Arthroplasty. 1998;13 (8):901-905.
  17. Argenson JN, Chevrol-Benkeddache Y, Aubaniac JM. Modern unicompartmental knee arthroplasty with cement: a three to ten-year follow-up study. J Bone Joint Surg Am. 2002;84(12):2235-2239.
  18. Price AJ, Webb J, Topf H, Dodd CA, Goodfellow JW, Murray DW. Rapid recovery after Oxford unicompartmental arthroplasty through a short incision. J Arthroplasty. 2001;16(8):970-976.

Authors

Dr Jenny is from Hôpitaux Universitaires de Strasbourg, Centre de Chirurgie Orthopédique et de la Main, Strasbourg, France.

Dr Jenny receives royalties from and is a consultant for B. Braun Aesculap.

Correspondence should be addressed to: Jean-Yves Jenny, MD, Centre de Chirurgie Orthopédique et de la Main, 10 avenue Baumann, F – 67400 Illkirch-Graffenstaden, France.