Clinical Experience Using a Minimally Invasive Surgical Approach for Total Knee Arthroplasty: Early Results of a Prospective Randomized Study Compared to a Standard Approach

Methods 

Institutional review board approval was obtained for this study at each of the participating centers. All patients had osteoarthritis of the knee and had a primary knee arthroplasty with the Stryker Scorpio posterior-stabilized TKA (Stryker Orthopaedics, Mahwah, NJ). This study, with 6 investigators at each of the participating centers, had patients prospectively stratified into minimally invasive and standard approaches by a 50:50 envelope randomization process. The study participants had intraoperative parameters (incision length, blood loss, time of surgery) and early postoperative clinical assessments of patient recovery and satisfaction recorded. Patient demographics for the two study groups were similar (Table 1). Intraoperative evaluation included data collected concerning time of surgery (incision to closure), blood loss, length of incision (measured in centimeters at 90° flexion and full extension at end of case), soft tissue releases, and intraoperative range of motion before and after closure.

Table 1. Demographic Parameters for the Two Groups

	Demographic Parameters	Limited Approach	Standard Approach
	Number	40	40
	Age (range), y	67 (48-84)	70 (54-79)
	Sex (male/female)	29:11	24:15
	Body mass index, mean (range)	32 (19-49)	30 (20-40)
	Preoperative Knee Society score
	Objective, mean (range)	49 (13-92)	57 (31-90)
	Function, mean (range)	51 (20-90)	54 (30-90)
	Preoperative quadriceps strength
	Excellent	8	14
	Good	26	19
	Fair	6	7

Postoperative clinical data collected through discharge included range of motion (active and passive), functional assessment of various activities (straight leg raise, walking), visual analog pain scores, and other parameters, as well as pain indicators, total blood loss, length of stay, and early complications.

Postoperative clinical data was collected at 2, 6, and 12 weeks after surgery and included SF-12, functional questions, Knee Society scores [33], Bartlett patella scores [34] (records anterior knee pain, quadriceps muscle strength, chair rise ability, and stair climb), range of motion, and use of analgesics.

Radiographic Evaluation 

Standard anteroposterior and lateral radiographs of all knees were evaluated preoperatively to determine the degree of deformity 34, 35, 36. Patients with more than 10° of varus or valgus deformity were excluded from the study. Postoperative radiographs were used to assess alignment, component position. Radiographic parameters including tibiofemoral angle, distal femoral angle, and proximal tibial angle were evaluated to determine coronal plane deformity [32]. Lateral films were analyzed for the degree of flexion and extension of all components. Zonal analysis of any radiolucencies were also evaluated.

Data Analysis 

The data were compiled and tabulated using the PhDx Internet-based data base (PhDx Systems Inc, Albuquerque, NM). Descriptive statistics were calculated because of the small number of patients in this study to avoid statistical bias. The descriptive statistics were calculated using SigmaStat for Microsoft Windows Version 3.0 (SPSS, Chicago, Ill). For statistical analyses, a Mann-Whitney Rank Sum Test was used for different subgroups to evaluate their effect on outcomes. These analyses included various demographic variables (length of stay, length of incision, type of incision, Knee Society scores, muscle strength, and complications).

Results 

The overall functional and objective results were similar for the standard and MIS approach groups at all time periods (see Table 2). The MIS group had a mean length of incision in extension of 9 cm (range, 6.9-12.5 cm). The standard group had a mean length of incision of 16 cm (range, 13-20 cm). Various intraoperative indices were also similar between the 2 groups. The mean blood loss was 108 mL in the MIS group (range, 20-400 mL) versus 110 mL (range, 20-300 mL) in the standard group. The time of surgery averaged 10 more minutes in the MIS approach group, 69 minutes (range, 32-175 minutes) versus 59 minutes (range, 42-109 minutes) in the standard group.

Table 2. Results of MIS and Standard Approaches to Total Knee Arthroplasty

	Parameters	MIS Approach	Standard Approach	P
	Mean length of stay, d	4.0	4.1	.588
	Tourniquet time (range), min	69 (32-175)	59 (39-103)	< .001
	Length of incision (range), cm	9 (6.9-12.5)	15 (13-20)	< .001
	Intraoperative blood loss (range), mL	108 (25-400)	110 (20-300)	.450
	6-Wk postoperative Knee Society score
	Objective, mean (range)	76 (43-99)	73 (37-100)	.147
	Function, mean (range)	63 (50-100)	60 (30-100)	.130
	12-Wk postoperative Knee Society score
	Objective, mean (range)	81 (61-99)	77 (51-100)	.121
	Function, mean (range)	74 (55-100)	73 (45-100)	.610
	12-Wk postoperative quadriceps strength
	Excellent	17	16	.946
	Good	19	17
	Fair	4	7
	Complications—wound
	Superficial	4	1	.166
	Deep	0	0

The mean Knee Society objective score for the MIS group was 81 points (range, 61-99 points), which was slightly higher but similar to the standard approach group that had a mean of 77 points (range, 51-100 points) at 3 months follow-up. The functional Knee Society scores were also similar between the 2 groups with a mean of 74 points in the MIS group (range, 55-100 points) versus 73 points in the standard group (range, 45-100 points) at 3 months follow-up. The distribution of excellent, good, fair, and poor results was similar in the 2 groups (see Table 2).

Postoperatively, it was found that length of stay, amount of analgesic use, return of knee function, and total blood loss were similar in both groups.

Surgical adverse events were similar in both groups. In the MIS group, 4 patients had drainage, delayed wound healing, and local soreness; one tibial component subsided but did not affect outcome, and one patient had a femoral fracture after a motor vehicle accident and required early revision. In the standard approach group, one tibial component subsided, one patient had supracondylar femur fracture due to a fall, and one patient had delayed wound healing.

Medical complications were similar in both groups. In the MIS group, one patient developed sciatica, which resolved. One patient, who had a myocardial infarction, recovered uneventfully. A patient with postoperative mental changes recovered uneventfully. In the standard approach group, one patient developed congestive heart failure and made a full recovery. Another patient had a deep venous thrombosis that required prolonged anticoagulation.

Radiographic Results 

Preoperatively, all knees in both groups had less than 10° of deformity (range, 3°-9°). There were 30 varus knees (>3° varus deformity) in the MIS group and 32 varus knees in the standard group. There were 5 valgus knees in each group. Postoperatively, all knees were within the range of ±3° of anteroposterior alignment (goal, 96°; range, 93°-99°). Lateral femoral angles were all within ±3°, although they averaged 2° of extension in the MIS group versus 0.5° extension in the standard group. Anteroposterior tibial angle averaged 89° in the MIS group (goal, 90°) and 90° in the standard approach group, with no patients outside the goal of ±3°. The lateral tibial angle averaged 86° in both groups (goal, 87°), with all patients within ±3° (Fig. 1A-D).

View Large Image Download to PowerPoint

Fig. 1. (A) Preoperative anteroposterior view of a knee with tricompartmental osteoarthritis. (B) Preoperative lateral view of a knee with tricompartmental osteoarthritis. (C) Postoperative anteroposterior view of knee in panel (A) following limited approach. (D) Postoperative lateral view of knee in panel (B) following limited approach.

Discussion 

The early results and techniques of MIS TKA have been described by Bonutti et al 21, 22, and the results have been good. The purpose of this study was to see if these excellent results were reproducible by other surgeons at multiple centers. Therefore, a prospective, randomized study was carried out, and the early results are presented here. At short-term, 3 month follow-up, the 2 prospectively studied groups had similar clinical and radiographic outcomes.

There is limited literature on MIS TKA for bi- or tricompartmental knee arthroplasties 5, 21, 22, 25, 26, 27, 28, 29, 30. Bonutti et al 21, 22, 23 have described early experience with encouraging results with up to 5 years follow-up. Tria and Coon [32] have also reported encouraging results with up to 2 years follow-up. Using the MIS technique, TKAs showed return of early motion, less blood loss, and shorter hospital stays compared with historic control of standard knee arthroplasty. In the present study, we were not able to show any greater benefit of MIS versus standard techniques.

It is noteworthy that the results were similar in this multicenter study in the first 3 months, with an increase in local wound problems (4 vs 1). The increased incidence of delayed wound healing in the MIS group without impacting on outcome is significant, suggesting that skin stretching and retracting, which often are necessary with MIS, should be avoided.

Patient selection criteria, the surgeon's experience, and the surgeon's ability to adapt MIS for TKA all need further evaluation. It is possible that some skillful surgeons will continue to explore, thus developing a superior and reproducible technique.