Total knee arthroplasty (TKA) performed through a less invasive approach is associated with a quicker postoperative recovery and shorter hospitalization compared to the standard approach.1,2,4,9,10,14,22,24 Prospective trials have shown these advantages can be obtained safely and without compromising correct implant positioning.2-4,10,14,22,24 Less-invasive approaches, evolved from traditional exposures, have been developed to preserve as much of the extensor mechanism as possible.2,4,10,14,22,24
Avoiding incision of the quadriceps tendon above the proximal pole of the patella, Tria and Coon24 developed the so-called “quadriceps-sparing” approach. The limited parapatellar approach has been described by Scuderi et al22 and Tenholder et al23 as a medial parapatellar approach with a quadriceps tendon incision above the proximal patella limited to 2 to 4 cm.
Simultaneously, other authors have modified the original midvastus and subvastus approaches to preserve the vastus medialis obliquus (VMO) insertion on the patella.2,6,11,14,21 The less-invasive mini-midvastus and mini-subvastus variants avoid patella eversion and, in the case of the mini-midvastus, limit the amount of muscle split to a few centimeters.2,10,14 A recent magnetic resonance and cadaver study by Pagnano17 showed the VMO insertion extends down to the midpole of the patella in most cases; therefore, a “quadriceps-sparing” approach invariably violates this insertion, even if incision is made only to the proximal patellar pole.
We asked whether there were differences in terms of operative time and complications, perioperative blood loss, postoperative recovery, and early clinical and radio- graphic results between two modern, less-invasive approaches for TKA, namely the mini-subvastus and the “quadriceps-sparing” approaches.
MATERIALS AND METHODS
Between October 2004 and June 2005 we enrolled 60 patients in a prospective, randomized, blinded (patient and evaluator) study. A power analysis with 95% CI was performed and suggested 60 patients were needed to show a difference in achievement of rehabilitation goals, with a value (1-ß) of 0.8. One VAS score point and one day for rehabilitation milestones were considered clinically important differences in the primary outcome. Inclusion criteria consisted of primary total knee arthroplasty for a diagnosis of osteoarthritis. All patients had unilateral TKA performed by a single surgeon (PA). Thirty patients were treated using a mini-subvastus approach, whereas the other group of 30 patients was treated using a “quadriceps-sparing” approach. Exclusion criteria included inflammatory arthritis, restricted motion (flexion less < 90°), patella infera (Insall-Salvati ratio < 0.6), morbid obesity, risk of ischemic skin complications, and very severe deformities (> 20° on the mechanical axis). Major demographic variables were similar in the two groups (Table 1).
The followup for this study was 3 months because only early differences between the groups were of interest. No patients were lost to followup. The study was approved by the authors' Institutional Review Board and informed consent for participation in the study, and to take part to the randomization process for surgery, was obtained.
Patients were randomized to two groups using computer generated random numbers and a closed-envelope design. Patients, physical therapists, the pain management team, and the chart analyst were not informed of the patients' group assignment. The skin incision was the same.
All surgery was performed using Zimmer minimally invasive surgery (MIS) instrumentation (Zimmer, Warsaw, IN) in both groups. All patients in each group received a NexGen® Legacy Posterior Stabilized (LPS) Flex fixed-bearing TKA (Zimmer) with a Mini Keel® tibial component (Zimmer). During surgery, the patient was placed in a supine position, leaving both legs freely mobile on the table. Preparation and draping of the involved leg was followed by Esmarch exsanguination and tour- niquet inflation. An anteromedial parapatellar skin incision was made. For the mini-subvastus group, the deep fascia was exposed and an inverted L-shaped arthrotomy was made along the medial border of the patella tendon extending at an angle along the inferior VMO border (Fig 1). To gain adequate exposure with sufficient lateral patellar mobility we divided the medial patellofemoral ligament, performed a blunt dissection of the VMO off the vastoadductor membrane, removed most of the fat pad, and in some cases divided the VMO muscle fascia. The patella then could be displaced laterally, but was not everted.2 In the “quad- riceps-sparing” procedures, a modified straight medial parapatellar incision according to Insall12 was made, extending from 2 to 3 cm proximal to the superior patella pole to 2.5 cm below the joint line (Fig 2). This is not exactly as described by Tria and Coon24 but our interpretation of this approach. This is why we speak about “quadriceps-sparing” in inverted brackets. In both groups the patella was laterally dislocated without eversion. All the procedures followed the same sequence of surgical steps. The proximal tibia was first resected in situ without tibiofemoral subluxation. The tibial cut had posterior slope of 5° and was guided by standard extramedullary alignment. Distal femoral resection was performed through a medially mounted MIS cutting block 6° to 7° valgus to the intramedullary reference in varus knees and 4° to 5° in valgus knees. Femoral rotation was referenced from the posterior condyles with 3° external rotation for varus knees and 5° for valgus knees. We ascertained correct flexion and extension space symmetry and leg alignment with spacer blocks. Lastly, with femoral and tibial trial components in place, the patella was rotated to 90° and was resected freehand for resurfacing. We encountered no access problems by performing patellar resection last. Patellofemoral osteophytes were removed at the beginning of the procedure. With all trial components in place, we confirmed patellofemoral tracking and kinematics. Ligament and soft tissue releases were performed as needed. Applying standard cementing techniques, we implanted the components sequentially, starting with the tibial tray. Lastly, we proceeded with layered closure over a single closed suction drain, which was removed 24 hours after surgery.
All patients received epidural anesthesia and followed an identical postoperative pain protocol. This included a lumbar plexus nerve block and patient-controlled epidural analgesia administered through an epidural catheter. Thromboprophylaxis consisted of warfarin sodium (International Normalized Ratio target point = 2) in combination with a calf pump for the first 48 hours, and compression stockings worn for 4 weeks after surgery. All patients received Doppler ultrasound examination on the fifth postoperative day to screen for deep venous thrombosis (DVT).
Rehabilitation started in the recovery area with continuous passive knee movement set at 60°. Mobilization began on the first postoperative day assisted by a physical therapist with full weightbearing on the operated leg. All patients in this study remained in the hospital and had rehabilitation until safe, single- crutch mobilization was possible. Patients then were discharged home without first staying at an intermediate rehabilitation facility. To ascertain the level of difficulty encountered, we studied tourniquet time, overall operation time, rate of complications, and incidence of DVT. Total blood loss was taken as the sum of recorded intraoperative loss and loss from drains during the first 24 hours.
Our physical therapists (FV, TC), who were blinded to the approach used during surgery, assessed all patients twice a day postoperatively until the first unassisted straight-leg raise was performed and 90° of knee flexion were obtained. Knee flexion was again assessed by the therapist at 30, 60, and 90 days. The visual analog pain score was used to assess the level of pain at 24 hours after surgery.
Anteroposterior (AP), lateral, and long-leg weightbearing views were obtained at the first visit 2 weeks postoperatively when all knees were devoid of skin staples, allowing for an unbiased radiographic assessment. All radiographs were assessed independently by two observers (AB, LS) with regard to implant position according to the Knee Society TKA roentgeno- graphic evaluation form.7 Component and overall alignment of neutral ± 2° was rated as correct.
In addition to descriptive statistics, data were analyzed with Microsoft® Excel software (Microsoft®, Redmond, WA) using a two-tailed, unpaired t-test analysis. These tests were applied to the interval measures of the study. Each measure approximated a normal distribution. A p value less than 0.05 was considered significant.
We found similar tourniquet time, total operative time, lateral retinacular release rate, and total blood loss in the two groups (Table 2). We encountered no major intraoperative complications in either group. One patient in the mini-subvastus group had partial damage of the patellar tendon which was inadvertently cut. This was recognized and repaired with nonabsorbable sutures without sequelae. We encountered difficulties in exposure and visualization in obese patients, patients with large knees, muscular male patients, and patients with more severe deformities or patella baja. In five knees the quadriceps incision had to be extended proximally 3 to 4 cm to allow completion of the procedure. However, the mini-subvastus was more versatile and allowed treatment of 30 knees without particular struggle.
There were no cases of proximal DVT in either group, but there were four cases of distal DVT in the mini- subvastus group and two cases of distal DVT in the “quad- riceps-sparing” group. One patient in the “quadriceps- sparing” group had a postoperative hemarthrosis. This resolved without drainage and by adjusting the warfarin dosage (Table 3).
The average postoperative pain on Day 1, as recorded on the visual analog scale, was comparable for both groups. Active straight-leg raise was achieved slightly quicker (p < 0.04) in the mini-subvastus group (1.4 ± 1.3 days; range, 1-7 days) than in the “quadriceps-sparing” group (1.9 ± 1.5 days; range, 1-7 days). Ninety degrees of flexion was achieved at approximately the second postoperative day in both groups. Flexion at 30 days did not substantially differ between the two groups (115° for mini- subvastus versus 112° for quadriceps sparing). At 90 days, average flexion was 119°, and 117°, respectively, in the mini-subvastus group and in the “quadriceps-sparing” group.
Radiographic analysis revealed correct (neutral ± 2°) AP-plane component position in all patients in both groups. None of the tibial components were misaligned (neutral ± 2°) in varus or valgus and the tibial component slope was within 1° to 5° in all knees. There were three knees in the mini-midvastus group and four knees in the “quadriceps-sparing” group where the tibial component showed a medial shift of 3 to 5 mm compared to the resected bone surface (Fig 3). This was probably related to component undersizing attributable to limited intraoperative visualization of the posterolateral corner of the tibial plateau. Asymptomatic residual cement fragments around the lateral tibial component rim were detected in two cases in the mini-subvastus group and three of the “quadriceps- sparing” group (Table 4).
Less invasive TKA techniques are rapidly evolving. Minimal knee capsular incisions first were described for unicompartmental knee arthroplasty.19 The quest for quicker rehabilitation and shorter hospitalization after TKA has expanded minimal-incision applications to this procedure. To date, literature on minimally invasive techniques for TKA is limited to anecdotal reports on the experience of one surgeon, and only few prospective, single-cohort or matched-control comparative studies are available.2,3,5,10,14,23,24 In a previous study from our group, we prospectively compared two nonrandomized groups of TKAs performed with the standard medial parapatellar approach or with a minimally invasive approach through a mini-subvastus incision.2 We found the minimally invasive approach had some advantages concerning functional recovery and postoperative pain compared to the standard approach. These advantages were obtained at the risk of a higher rate of complications. This was our initial experience and there is a learning curve. Our results reflected those of other studies using minimally invasive approaches, such as the mini-midvastus or the “quadriceps-sparing” approaches, reported in the literature.5,10,14,15,24
Our study was limited to a followup of 3 months. We are unable to ascertain the longer-term effects of these two approaches. Our major outcome variables were limited to postoperative pain (VAS), perioperative factors or complications, and knee flexion up to 90 days, and the radio- graphic appearance two weeks postoperatively. Although knee motion may change over time, since the two groups were similar in terms of knee flexion, we presume the longer-term flexion will remain similar in both groups. We did not report other variables such as patient satisfaction or hospitalization times.
We adopted the mini-subvastus approach because of our initial experience with minimally invasive TKA. The standard subvastus approach has shown several advantages in terms of functional recovery compared with the medial parapatellar approach.8,11,16,18,20 Shortening the subvastus dissection a few centimeters proximally preserves the descending genicular artery branches (the musculoarticular branch).8,21 This approach reduces the risk of VMO denervation with the midvastus approach.6,13,18 Recently, Pagnano et al,17 in a cadaver and MRI study of 200 specimens, demonstrated the VMO tendon inserts on the patella distally, down to its midpole, in most cases. Therefore, the subvastus approach is the only current approach preserving the quadriceps insertion on the patella. We compared the mini-subvastus approach to the “quadriceps- sparing” approach because the latter is described as the least invasive of the minimally invasive approach variants.22,24
We found no differences in terms of tourniquet time, total blood loss, intraoperative complications, subjective and objective results and radiographic data between the two approaches. The “quadriceps-sparing” approach, however, was technically more demanding. It was difficult to obtain the exposure of both femoral condyles simultaneously. Nevertheless, when required, the “quadriceps- sparing” approach can easily be extended or converted in a limited or standard parapatellar approach. Extending the incision 5 to 6 cm in the quadriceps tendon may represent a good compromise between having a relatively short incision without interfering with the technique. Functional recovery of the five patients who had a few cm extension of the incision did not differ from those of the rest of the “quadriceps-sparing” group. At 3 months, functional results were comparable to those of standard TKA series for both groups.2 From a radiographic standpoint, although a good alignment was obtained in all the cases, there were a few instances in which the tibial component tended to be positioned a few millimeters medially or there was excessive cement, particularly in the posterolateral tibia. Mini- incision can risk component malalignment. There were no differences between the two groups in this respect. Exposure in both groups is more difficult than traditional approaches, particularly in the “quadriceps-sparing” group, and requires attention to the placement of the various re- tractors, protection of the bone (particularly if osteoporotic), and protection of the patella and patella tendon. We have found poor visualization of the posterolateral corner is not an absolute contraindication to the quadriceps sparing approach, but it adds complexity to the procedure and it makes this approach less versatile than the subvastus approach. We do not recommend using the quadriceps sparing approach in larger patients or muscular males.
The advantage in functional recovery of a minimally invasive approach after TKA seems limited to the first weeks or months after the procedure. We observed no difference between patients who were treated using the mini-subvastus compared with those who were treated with the “quadriceps-sparing” approach. No differences were encountered when the “quadriceps-sparing” incision was converted to a limited parapatellar incision. Utilizing a limited parapatellar approach we observed satisfactory functional recovery, comparable to the other mini- approaches, without compromising TKA technique and allowing the exposure of all the landmarks.
The authors thank Fabio Valgiusti and Tommaso Ciappi for assessing the patients postoperatively.
1. Berger RA, Sanders S, Gerlinger T, Della Valle C, Jacobs JJ, Rosenberg AG. Outpatient total knee arthroplasty with a minimally invasive technique. J Arthroplasty
. 2005;20(Suppl 3):33-38.
2. Boerger TO, Aglietti P, Mondanelli N, Sensi L. Mini-subvastus versus medial parapatellar approach in total knee arthroplasty. Clin Orthop Relat Res
3. Bonutti PM, Mont MA, Kester MA. Minimally invasive total knee arthroplasty: a 10-feature evolutionary approach. Orthop Clin North Am
4. Bonutti PM, Mont MA, McMahon M, Ragland PS, Kester M. Minimally invasive total knee arthroplasty. J Bone Joint Surg Am
. 2004;86(Suppl 2):26-32.
5. Dalury DF, Dennis DA. Mini-incision total knee arthroplasty can increase risk of component malalignment. Clin Orthop Relat Res
6. Engh GA, Parks NL. Surgical technique of the midvastus arthrotomy. Clin Orthop Relat Res
7. Ewald FC. The Knee Society total knee arthroplasty roentgeno- graphic evaluation and scoring system. Clin Orthop Relat Res
8. Faure BT, Benjamin JB, Lindsey B, Volz RG, Schutte D. Comparison of the subvastus and paramedian surgical approaches in bilateral knee arthroplasty. J Arthroplasty
9. Goble EM, Justin DF. Minimally invasive total knee replacement: principles and technique. Orthop Clin North Am
10. Haas SB, Cook S, Beksac B. Minimally invasive total knee replacement through a mini midvastus approach: a comparative study. Clin Orthop Relat Res
11. Hoffmann AA, Plaster RL, Murdock LE. Subvastus (Southern) approach for primary total knee arthroplasty. Clin Orthop Relat Res
12. Insall J. A midline approach to the knee. J Bone Joint Surg Am
13. Jojima H, Whiteside LA, Ogata K. Anatomic consideration of nerve supply to the vastus medialis in knee surgery. Clin Orthop Relat Res
14. Laskin RS, Beksac B, Phongjunakorn A, Pittors K, Davis J, Shim JC, Pavlov H, Petersen M. Minimally invasive total knee replacement through a mini-midvastus incision: an outcome study. Clin Orthop Relat Res
15. Laskin RS. Minimally invasive total knee arthroplasty: the results justify its use. Clin Orthop Relat Res
16. Matsueda M, Gustilo RB. Subvastus and medial parapatellar approaches in total knee arthroplasty. Clin Orthop Relat Res
. 2000; 371:161-168.
17. Pagnano MW, Meneghini RM, Trousdale RT. Anatomy of the extensor mechanism in reference to quadriceps-sparing TKA. Clin Orthop Relat Res
18. Parentis MA, Rumi MN, Deol GS, Kothari M, Parrish WM, Pellegrini VD Jr. A comparison of the vastus splitting and median parapatellar approaches in total knee arthroplasty. Clin Orthop Relat Res
19. Repicci JA, Eberle RW. Minimally invasive surgical technique for unicondylar knee arthroplasty. J South Orthop Assoc
. 1999;8: 20-27. discussion 27
20. Roysam GS, Oakley MJ. Subvastus approach for total knee arthroplasty: a prospective, randomized, and observer-blinded trial. J Arthroplasty
21. Scheibel MT, Schmidt W, Thomas M, von Salis-Soglio G. A detailed anatomical description of the subvastus region and clinical relevance for the subvastus approach in total knee arthroplasty. Surg Radiol Anat
22. Scuderi GR, Tenholder M, Capeci C. Surgical approaches in mini- incision total knee arthroplasty. Clin Orthop Relat Res
23. Tenholder M, Clarke HD, Scuderi GR. Minimal-incision total knee arthroplasty: the early clinical experience. Clin Orthop Relat Res
24. Tria AJ Jr, Coon TM. Minimally incision total knee arthroplasty: early experience. Clin Orthop Relat Res