A total of 90 consecutive RA primary unilateral TKA patients, including 31 males and 59 females, were followed-up for 12 weeks. No complications, such as periprosthetic fracture, aseptic loosening, periprosthetic infection, or dislocation, were observed.
The learning curve was analyzed using the CUSUM method. From the CUSUM chart (Fig. 1), the CUSUM peak occurred after 20 cases, allowing the learning curve to be divided into learning and proficiency groups. The CUSUM learning curve was fitted to the following cubic curve equation: CUSUM (Duration of surgery) = 22.617 + 25.648x-0.377x2 + 0.001x3, R2 = 0.924.
However, a decline in the CUSUM value does not represent successful surgery. Therefore, the RA-CUSUM method was used to evaluate the rate of failure of surgery. From the RA-CUSUM plot (Fig. 2), the graph reached a peak with the 21st case, representing the greatest rate of failure of the surgery, but also indicating that the rate of failure started to decline from this case onwards, representing proficiency for the RA-TKA system. Combined with the general trend and the results of the CUSUM and RA-CUSUM curves, a learning curve for the RA-TKA system was determined. The curve is divided into two groups, the first phase representing the learning phase, which spanned 20 cases (1–20), and the second phase representing the proficiency phase (cases 21–90) which began immediately after.
The two phases were compared in terms of demography, preoperative clinical results, preoperative radiographic data, duration of surgery, postoperative clinical results, and postoperative radiographic data (Table 1). There was no significant difference in age, surgical side, body mass index (BMI), sex, preoperative HKA angle, deviation in preoperative HKA, preoperative VAS, preoperative ROM, or preoperative KSS between the two groups (P > 0.05). The duration of surgery in the learning group (128.00 ± 7.50 min) was significantly longer than that in the proficiency group (96.37 ± 14.63 min) (P < 0.001). In terms of clinical results, there was no significant difference in the postoperative ROM, KSS, or VAS scores between the two groups. In terms of radiographic results, there was no significant difference in postoperative HKA angle or deviation in HKA angle, or deviation in CTCA, CFCA, STCA, or SFCA between the two groups.
Table 1
Table 1
Comparison of the clinical and radiographic data between the two groups using the RA-TKA system.
| Learning group | Proficiency group | Statistic | P-value |
N | 20 | 70 | | |
Age | 70.50 ± 5.54 | 67.80 ± 7.76 | t = 1.754 | 0.087 |
BMI (kg/m2) | 27.56 ± 2.53 | 26.28 ± 3.69 | t = 1.784 | 0.081 |
Male (%) | 35.0 | 34.3 | χ²=0.004 | 0.953 |
Surgical side | | | | |
Left (%) | 45.0 | 52.9 | | |
Right (%) | 55.0 | 47.1 | χ²=0.384 | 0.535 |
Preoperative HKA angle | 170.22 ± 6.18 | 172.52 ± 5.68 | t=-1.568 | 0.121 |
Preoperative HKA angle deviation | 10.39 ± 5.24 | 8.33 ± 4.32 | t = 1.787 | 0.077 |
Preoperative VAS | 6.03 ± 1.55 | 6.82 ± 1.82 | t=-1.769 | 0.080 |
Preoperative ROM | 109.20 ± 13.99 | 101.16 ± 19.57 | t = 1.714 | 0.090 |
Preoperative KSS | 45.30 ± 17.78 | 47.04 ± 16.83 | t=-0.403 | 0.688 |
Operation time | 128.00 ± 7.50 | 96.37 ± 14.63 | t = 13.056 | 0.000* |
Postoperative HKA angle deviation | 2.24 ± 0.78 | 1.83 ± 0.93 | t = 1.956 | 0.058 |
Postoperative CTCA deviation | 1.31 ± 0.99 | 1.28 ± 0.89 | t = 0.102 | 0.919 |
Postoperative CFCA deviation | 1.24 ± 0.87 | 1.04 ± 0.84 | t = 0.899 | 0.371 |
Postoperative STCA deviation | 2.14 ± 1.78 | 1.69 ± 1.20 | t = 1.329 | 0.187 |
Postoperative SFCA deviation | 2.23 ± 1.65 | 1.79 ± 1.46 | t = 1.165 | 0.247 |
Postoperative ROM | 119.25 ± 7.66 | 116.70 ± 11.75 | t = 1.152 | 0.255 |
Postoperative KSS | 60.80 ± 8.06 | 62.74 ± 8.63 | t=-0.901 | 0.370 |
Postoperative VAS | 2.68 ± 1.79 | 2.65 ± 1.62 | t = 0.053 | 0.958 |
BMI = body mass index; ROM = range of motion; KSS = Knee Society score; VAS = 10-cm visual analogue scale; *P < 0.05 |
The rate of outliers of the above five angles (Table 2) was further analyzed. There was no significant difference between the two groups (P > 0.05).
Table 2
Table 2
Outliers in component positions and lower limb alignment in patients operated using an RA-TKA system.
Group | Percentage of knees with implant aligned outside ± 3° from neutral angle |
STCA | SFCA | CTCA | CFCA | HKA |
Learning group | 10.0% | 10.0% | 5.0% | 5.0% | 5.0% |
Proficiency group | 5.7% | 4.3% | 0.0% | 1.4% | 2.9% |
χ² | 0.459 | 0.968 | 3.539 | 0.913 | 0.222 |
P-value | 0.498 | 0.325 | 0.060 | 0.339 | 0.638 |
In terms of complications, there was one case of poor wound healing and two cases of wound exudation in the learning group. In the proficiency group, there were three cases of poor wound healing and four cases of wound exudation. There was one case of lower limb deep vein thrombosis in the learning group and three cases in the proficiency group, which was resolved using drug thrombolysis. No RAS-related complications, such as needle infection or peri-needle fracture, were experienced in either of the two groups. There was no significant difference in the total incidence of complications between the two groups (P > 0.05).
The demographics, duration of surgery, and the clinical and radiographic results of RA-TKA (cases 21–90) performed by the same surgeon in the proficiency phase were compared with those of 90 consecutive cases of conventional TKA in the same period. A total of 98 patients were matched by PSM and included for analysis: (1) 49 patients in the proficiency group, and (2) 49 patients in the conventional group. Prematched and postmatched data are displayed in Tables 3 and 4. As shown in Table 4, there was no significant difference in demographic, preoperative clinical and radiographic results between the two groups (P > 0.05). The postoperative HKA angle deviation was 1.75 ± 0.97° in the proficiency group and 3.07 ± 2.43° in the conventional group, a statistically significant difference (P < 0.001). The deviation in postoperative CFCA in the proficiency and conventional groups was 0.97 ± 0.83° and 1.81 ± 1.76°, respectively, a statistically significant difference (P = 0.003). There were significant differences between the proficiency group and the conventional group for deviation in postoperative CTCA (1.24 ± 0.86° vs 1.70 ± 1.29°, P = 0.039) and postoperative STCA (1.83 ± 1.22° vs. 2.41 ± 1.43°, P = 0.032), but no significant difference between the two groups in the deviation in postoperative SFCA (1.87 ± 1.67° vs 2.18 ± 2.05°, P > 0.05).
Table 3
Table 3
Comparison of demographics in the unmatched data.
| Proficiency group | Conventional group | Statistic | P-value |
N | 70 | 90 | | |
Age | 67.80 ± 7.76 | 70.30 ± 7.49 | t=-2.073 | 0.040* |
BMI (kg/m2) | 26.28 ± 3.69 | 26.46 ± 3.32 | t=-0.320 | 0.750 |
Male (%) | 34.3 | 23.3 | χ²=2.337 | 0.126 |
Surgical side | | | | |
Left (%) | 52.9 | 51.1 | | |
Right (%) | 47.1 | 48.9 | χ²=0.048 | 0.826 |
Preoperative HKA angle | 172.52 ± 5.68 | 174.87 ± 8.53 | t=-2.088 | 0.038* |
Preoperative HKA angle deviation | 8.33 ± 4.32 | 8.77 ± 4.64 | t=-0.616 | 0.539 |
Preoperative VAS | 6.82 ± 1.82 | 6.51 ± 1.75 | t = 1.097 | 0.274 |
Preoperative ROM | 101.16 ± 19.57 | 106.71 ± 14.41 | t=-2.067 | 0.040* |
Preoperative KSS | 47.04 ± 16.83 | 43.96 ± 12.82 | t = 1.274 | 0.205 |
BMI = body mass index; ROM = range of motion; KSS = Knee Society score; VAS = 10-cm visual analogue scale; *P < 0.05 |
Table 4
Table 4
Comparison of the clinical and radiographic data between the matched groups.
| Proficiency group | Conventional group | Statistic | P-value |
N | 49 | 49 | | |
Age | 69.20 ± 6.82 | 69.43 ± 6.94 | t=-0.161 | 0.872 |
BMI (kg/m2) | 26.84 ± 3.77 | 26.44 ± 3.45 | t = 0.556 | 0.579 |
Male (%) | 36.7 | 30.6 | χ²=0.411 | 0.521 |
Surgical side | | | | |
Left (%) | 51.0 | 46.9 | | |
Right (%) | 49.0 | 53.1 | χ²=0.163 | 0.686 |
Preoperative HKA angle | 172.42 ± 5.81 | 172.26 ± 6.65 | t = 0.128 | 0.899 |
Preoperative HKA angle deviation | 8.49 ± 4.33 | 8.74 ± 5.23 | t=-0.276 | 0.790 |
Preoperative VAS | 6.75 ± 1.94 | 6.53 ± 1.83 | t = 0.588 | 0.588 |
Preoperative ROM | 103.86 ± 15.73 | 104.02 ± 15.38 | t=-0.052 | 0.959 |
Preoperative KSS | 42.86 ± 15.50 | 48.12 ± 13.05 | t=-1.819 | 0.072 |
Duration of surgery | 98.22 ± 14.42 | 90.33 ± 11.15 | t = 3.032 | 0.003* |
Postoperative HKA angle deviation | 1.75 ± 0.97 | 3.07 ± 2.43 | t=-3.524 | 0.000* |
Postoperative CTCA deviation | 1.24 ± 0.86 | 1.70 ± 1.29 | t=-2.095 | 0.039* |
Postoperative CFCA deviation | 0.97 ± 0.83 | 1.81 ± 1.76 | t=-3.034 | 0.003* |
Postoperative STCA deviation | 1.83 ± 1.22 | 2.41 ± 1.43 | t=-2.171 | 0.032* |
Postoperative SFCA deviation | 1.87 ± 1.67 | 2.18 ± 2.05 | t=-0.806 | 0.422 |
Postoperative ROM | 116.73 ± 11.35 | 116.16 ± 12.82 | t = 0.234 | 0.816 |
Postoperative KSS | 62.55 ± 7.72 | 61.57 ± 11.17 | t = 0.505 | 0.615 |
Postoperative VAS | 2.63 ± 1.62 | 3.21 ± 2.10 | t=-1.536 | 0.128 |
BMI = body mass index; ROM = range of motion; KSS = Knee Society score; VAS = 10-cm visual analogue scale;*P < 0.05 |
As displayed in Table 5, there was no significant difference in the rate of outliers of the SFCA between the two groups (P > 0.05). The rate of the postoperative HKA angle outliers in the proficiency and conventional groups were 4.1% and 36.7%, respectively, a difference that was statistically significant (P < 0.001). The rate of the postoperative CFCA outliers in the proficiency and conventional groups was 2.0% and 14.3%, respectively, a difference that was statistically significant (P = 0.027). The rate of postoperative CTCA outliers in the proficiency and conventional groups was 0.0% and 8.2% respectively, a difference that was statistically significant (P = 0.041). The rate of postoperative STCA outliers in the proficiency and conventional groups was 4.1% and 16.3%, respectively, a difference that was also statistically significant (P = 0.045). Additionally, the duration of surgery for the proficiency group was 98.22 ± 14.42 min, which was longer than that of the conventional group (90.33 ± 11.15 min), a difference that was statistically significant (P = 0.003).
Table 5
Table 5
Comparison of outliers in component positions and lower limb alignment between the proficiency and conventional groups
Group | Percentage of knees with implant aligned > ± 3° from neutral angle |
STCA | SFCA | CTCA | CFCA | HKA |
Proficiency group | 4.1% | 4.1% | 0.0% | 2.0% | 4.1% |
Conventional group | 16.3% | 14.3% | 8.2% | 14.3% | 36.7% |
χ² | 4.009 | 3.059 | 4.170 | 4.900 | 16.082 |
P-value | 0.045* | 0.080 | 0.041* | 0.027* | 0.000* |