Re-learning curve for conventional total knee arthroplasty following 30 consecutive computer-assisted total knee arthroplasties

Abstract

A learning curve for returning to conventional total knee arthroplasty (TKA) after using computer-assisted (CAS) TKA has not yet been established. In this study, the postoperative mechanical axes of the first 30 consecutive CAS TKAs performed by a single surgeon were compared to his subsequent 120 conventionally performed TKAs. A “re-learning curve” of 30 conventional TKAs was necessary to attain an average postoperative mechanical axis statistically indistinguishable from the average CAS mechanical axis (1.99°). This is a trend of which surgeons should be aware when converting from CAS TKA to conventional TKA. As a secondary goal, the authors identify the first clinical parameter, preoperative deviation from neutral mechanical axis, that may potentially serve as a guide for the selective use of CAS in TKA.

• Alignment
• computer-assisted surgery
• mechanical axis
• total knee arthroplasty

Introduction

The aim of a total knee arthroplasty (TKA) is to reduce or completely alleviate the pain associated with an arthritic or diseased joint while improving mobility. Although the success of a TKA procedure depends on many factors, neutral alignment of the mechanical axis is considered one of the most important parameters for long-term success, with mal-alignment of greater than 3.0° resulting in high failure rates and decreased prosthesis longevity [1–6]. To help achieve optimal postoperative alignment, computer-assisted (CAS) navigation systems have been developed and have been shown to improve prosthesis alignment [1], [7–13], with some authors suggesting that improved knee function and quality of life result from this improvement [10], [11]. Whether these data correlate with an increased survival rate for the prosthesis remains to be determined, however.

Recently, Maniar et al. [14] described the learning curve of a single surgeon converting from conventional TKA to CAS TKA. It was found that even without prior CAS experience a surgeon could expect significant improvement in postoperative alignment almost immediately when using CAS. Success, defined as a mechanical axis of less than ±3.0°, was achieved in 95% of the first 100 CAS TKAs and in 100% of the second 100 cases performed.

In the current study, we measured the pre- and postoperative mechanical axes of 150 TKAs performed by a fellowship-trained adult reconstruction orthopaedic surgeon (E.S.) with no prior CAS experience. The first 30 TKAs were performed using CAS, and the following 120 with conventional technique. The primary goals of our study were to evaluate the success of CAS for a surgeon implementing the technology, and to determine what, if any, learning curve a surgeon can expect after returning to conventional TKA. We also assessed the effect of the preoperative mechanical axis on postoperative alignment in both CAS and conventional TKA cases. This had been previously studied by Hsu et al., who found that large preoperative deformities did not affect the postoperative alignment of CAS TKA, but resulted in worse outcomes in conventional TKA, highlighting a possible benefit of using CAS [12].

Materials and methods

Patients

The 30 CAS TKA patients comprised 16 females and 14 males with a mean age of 62.9 years. The 120 conventional TKA patients comprised 77 females and 43 males with a mean age of 64.5 years.

After obtaining approval from the Institutional Review Board at Tufts Medical Center, we retrospectively evaluated the pre- and postoperative long-standing radiographs of 150 patients who underwent consecutive primary TKA procedures. The surgeries were performed by a single surgeon between September 2007 and September 2009. At the onset of the operations the surgeon was well-trained in conventional TKA; however, he had no prior experience using computer-assisted technology as an attending surgeon. The first 30 patients undergoing TKA were operated with CAS using the BrainLab Imageless Knee Navigation System (Munich, Germany). The subsequent 120 patients underwent conventional TKA per the surgeon's prior conventional technique. All patients included in the study received the same cruciate-retaining, fixed bearing prosthesis, sized intraoperatively in accordance with the standard TKA protocol. Tourniquet time was recorded for each group, as were pre- and postoperative Knee Society Scores (KSS) at 6 weeks, 6 months, 12 months, and 24 months.

Data analysis

Routine standing pre- and postoperative radiographs were used to determine axial alignment. Measurements were determined by two trained observers blinded to the operation type using the CARESTREAM Vue PACS software measurement tool (CARESTREAM Health, Rochester, NY). The mechanical axis was defined as the angle between a line connecting the center of the femoral head with the center of the knee and a line connecting the center of the knee with the center of the talus. A mechanical axis of varus deformity was recorded as a positive value, while a mechanical axis of valgus deformity was recorded as a negative value. The absolute value of the mechanical axis measurement was recorded for each patient to establish the deviation from neutral alignment. The mean deviation from neutral alignment was determined for the 30 CAS patients (CAS group) and for four groups of 30 consecutive conventional patients (conventional groups 1 through 4).

Statistical analysis

The preoperative and postoperative mechanical axes for CAS and conventional TKA were compared using Student's two-tailed t-test. The statistical analysis was performed using GraphPad Prism version 5.02 for Windows (GraphPad Software, San Diego, CA). Statistical significance was set at p ≤ 0.05.

Results

Inter-observer variability

Two blinded researchers independently measured the mechanical axis of each patient's preoperative and postoperative frontal radiograph. The inter-observer reliability was assessed using a standard linear regression analysis, yielding a slope of 0.98 (Figure 1). This was determined to be an insignificant level of variability between the mechanical axis measurements of each observer.

Figure 1. Scatter plot comparing the degrees of mechanical axis deviation as measured by each independent observer. The slope of the best-fit line equals 0.98, indicating insignificant variability between measurements.

Mechanical axis comparison

The average preoperative mechanical axis for all 150 patients was 6.7°. The average preoperative mechanical axis of CAS patients was 7.0°, while that of the conventional patients was 6.6°. The average postoperative mechanical axis for all 150 patients was 2.85°. The average CAS and conventional postoperative mechanical axes were 1.99° and 3.06°, respectively (p < 0.05). When the 120 conventional patients were subdivided into consecutive groups of 30, the average mechanical axis of Groups 1–4 was 3.55°, 3.08°, 2.97° and 2.64°, respectively. Only the mechanical axis of Group 1 was significantly elevated compared to the CAS group (p ≤ 0.05). The surgeon's learning curve upon returning to conventional surgery is shown in Figure 2.

Figure 2. The dotted line indicates the average CAS TKA postoperative mechanical axis (1.99°). The data points indicate the average conventional TKA postoperative mechanical axis for each group of 30 consecutive patients.

Figure 3 illustrates the effect of preoperative mechanical axis on postoperative alignment. The slopes of the best-fit lines determined by linear regression analysis are 0.17 and 0.54 for CAS (Figure 3a) and conventional (Figure 3b) TKA, respectively. Only the slope of the best-fit line for conventional TKA was significantly different from 0 (p ≤ 0.05). When all conventional patients with a preoperative mechanical axis above 10.0° were eliminated from the analysis, a slope similar to that for CAS (0.16) was obtained that was not statistically different from a slope of 0 (Figure 3c).

Figure 3. Effect of preoperative mechanical axis (°) on postoperative mechanical axis (°) in CAS TKA (a), conventional TKA (b), and conventional TKA in patients with preoperative mechanical axes of 10.0° or less (c). The vertical dotted line indicates a postoperative mechanical axis of 3.0°, a deviation from neutral that is widely considered acceptable.

Tourniquet time

Tourniquet time was significantly longer in the CAS group compared to the conventional group, at 106 min versus 75 min (p ≤ 0.05).

Discussion

The primary goal of this study was to assess the success of a conventionally trained orthopaedic surgeon beginning practice as a novice in computer-assisted surgery, and to determine how a return to conventional surgery affects subsequent surgical outcome. As indicated by previous studies [14–16], even without prior experience of using CAS, an inexperienced surgeon can expect to achieve an optimal average postoperative mechanical axis upon conversion to CAS technology. In this study, the senior surgeon obtained an average CAS postoperative mechanical axis deviation of 1.99°, considered to be within the range of optimal knee alignment according to prior studies [1–6]. A subsequent return to conventional surgery resulted in a steadily improving learning curve, as shown in Figure 2, with no significant difference in postoperative mechanical axis compared to CAS after the first 30 conventional operations. This is the first study to show that a surgeon who uses CAS on a consistent basis should expect a “re-learning curve” upon return to conventional procedures.

Such a “re-learning curve” following the use of computer-assisted surgery is of interest to the surgeon practicing in today's climate of cost-efficient surgery. While our findings are consistent with those of numerous other authors that CAS results in a mechanical axis closer to neutral, the clinical benefit of this finding has yet to be clearly demonstrated through the use of accepted clinical evaluation tools [17–20]. Furthermore, using CAS for TKA involves recurring costs and often results in increased operative time. In view of this, the authors expect a certain population of surgeons currently using CAS technology to consider reversion to the conventional technique. Our data provides the first example of a surgeon's results upon reversion to conventional surgery. Surgeons currently practicing with computerized technology, as well as those considering a trial period using such technology, should expect that a subsequent return to conventional practice may result in a period of sub-optimal, though clinically acceptable, results.

Of interest, as discussed by Harvie et al., no individual clinical parameters have yet been accepted in the literature to guide the selective use of CAS [19]. A secondary goal of our study was to assess preoperative alignment as a possible clinical criterion on which to base the selective use of CAS. Our data corroborate those initially reported by Hsu et al., indicating that preoperative deformity correlates positively with postoperative deviation from neutral alignment for conventional TKA, but that the use of CAS results in a reliable postoperative alignment regardless of preoperative deformity [12]. In our study, when patients with preoperative deformities greater than 10° were eliminated from the conventional TKA data set, preoperative deformity failed to correlate with postoperative alignment. This suggests that for those with preoperative deformities of less than 10°, the results of conventional TKA are identical to those of CAS TKA. Specifically, patients with less severe deformities would have an optimal radiological outcome regardless of whether CAS or conventional technique is used. The present authors are the first to conclude that optimal postoperative alignment can be achieved while minimizing procedure cost through the selective use of CAS on patients with preoperative deformity greater than 10°. Furthermore, as shown in this report and in previous studies, the adoption of CAS does not require a steep learning curve; therefore, the routine use of conventional technology with the selective use of CAS should present few, if any, problems for the surgeon switching between the two technologies.

Limitations of this study include the relatively small size of the CAS patient cohort and the retrospective nature of the study. Our results suggest that patients with a preoperative mechanical axis of >10.0° may benefit more from CAS than those with a preoperative deformity of <10.0°. However, this data does not include clinical correlation of outcome, nor does it address the practical considerations of a surgeon switching between two types of surgery. Further, larger studies are required to assess the clinical significance of computer navigated TKA as well as the role of the selective use of computerized technology in TKA.

Declaration of interest

The authors report no conflicts of interest in this study.