A total of 1985 patients were included, of which 498 had completed all muscle strength tests and 758 had completed all PROs during the first year. Figure 1 shows the distribution of the registered number of patients over the number of clinics during the study period. There were four high-volume (> 100 patients) and 111 low-volume clinics.
In total, 31 (1.6%) second ACL injuries were sustained within 1 year after ACL reconstruction, with an additional 37 (2.4%) second ACL injuries sustained between 1 and 2 years after ACL reconstruction. Table 1 presents the odds ratios for the incidence of a second ACL injury within 1 or 2 years after ACL reconstruction. No significant differences were observed in this study. Full details of the second ACL injury rates stratified by age are available in additional files 2 and 3.
Table 1
Odds ratio for sustaining a second ACL injury, by high- and low -patient volume clinics
|
n
|
1 year incidence, n (%)
|
Odds Ratio [CI]
|
p-value
|
Total
|
1985
|
31 (1.6%)
|
|
0.89
|
HV
|
1221
|
19 (1.5%)
|
0.95 [0.46–1.97]
|
LV
|
733
|
12 (1.6%)
|
|
|
|
2 years incidence, n (%)
|
|
|
Total
|
1686
|
68 (4.0%)
|
1.13 [0.68–1.88]
|
|
HV
|
1001
|
44 (4.4%)
|
0.64
|
LV
|
617
|
24 (3.8%)
|
|
ACL = Anterior cruciate ligament; n = number of patients; HV = high-volume clinics; LV = low-volume clinics; CI = confidence interval.
Return to pre-injury level of sports
The rate of returning to the pre-injury level of sports in patients with a pre-injury Tegner score of 1–5 was 62% for high-volume (HV) and 71% for low-volume (LV) clinics at the 1-year follow-up. For patients who had a higher pre-injury level of sports (Tegner 6–10), the rate of return to the same activity level was 26% for HV and 31% for LV clinics at the 1-year follow-up. A greater proportion of patients returned to the pre-injury level of sports, at 2 and 4 months in the HV group than in the LV group given a pre-injury Tegner level of 1–5 (Fig. 2). No other differences were observed in the rate of return to the pre-injury level of sports regardless of the pre-injury Tegner level at any follow-up, up to 1 year after ACL reconstruction. Patients in the HV group had a significantly higher Tegner score at 2 months than those in the LV group (P = 0.03). No differences were observed at 4, 8, or 12 months at the Tegner level (Fig. 2 and Table 2).
Table 2
Change in Tegner and rate of returning to pre-injury Tegner, first year after ACL-reconstruction.
Return to preinjury level of activity
|
2 months
|
p- value
|
4 months
|
p-value
|
8 months
|
p-value
|
12 months
|
p-value
|
|
HV
|
LV
|
|
HV
|
LV
|
|
HV
|
LV
|
|
HV
|
LV
|
|
Tegner
1-5, (%)
|
34%
|
11%
|
0.04
|
42%
|
23%
|
0.03
|
58%
|
42%
|
0.08
|
62%
|
71%
|
0.35
|
Tegner
6-10, (%)
|
5%
|
2%
|
0.05
|
4%
|
5%
|
0.69
|
9%
|
11%
|
0.58
|
26%
|
31%
|
0.20
|
Patient-reported outcomes
A total of 758 patients had available data on PROs from all follow-ups during the first year after ACL reconstruction, of which 63% and 37% represented the HV and LV groups, respectively (Table 3). There were no significant differences in the demographics.
Table 3
Demographics for patients, with available PROs-data, in high- and low-volume clinics, first year after ACL-reconstruction.
|
HV
|
LV
|
P-value
|
n
|
477
|
281
|
|
Women, n (%)
|
274 (57.4%)
|
162 (57.5%)
|
1.00
|
Age (years)
|
31 ± 11
|
32 ± 11
|
0.42
|
Height (cm)
|
173 ± 9
|
174 ± 9
|
0.19
|
Weight (kg)
|
72 ± 13
|
73 ± 13
|
0.32
|
BMI (kg/m2)
|
23.8 ± 3.4
|
23.8 ± 3.2
|
0.99
|
Days between injury and reconstruction
|
269 ± 475
|
387 ± 908
|
0.05
|
Type of graft
|
|
|
|
Hamstrings
|
375 (81%)
|
203 (80.8%)
|
0.88
|
Patella
|
75 (16.2%)
|
41 (16.4%)
|
Allograft
|
4 (0.9%)
|
4 (1.6%)
|
Quadriceps
|
3 (0.6%)
|
1 (0.4%)
|
Other
|
6 (1.3%)
|
2 (0.8%)
|
Tegner pre-injury, n (%)
|
|
|
|
1
|
5 (1.0%)
|
1 (0.4%)
|
0.15
|
2
|
10 (2.1%)
|
4 (1.4%)
|
3
|
13 (2.7%)
|
8 (2.8%)
|
4
|
26 (5.5%)
|
22 (7.8%)
|
5
|
22 (4.6%)
|
19 (6.8%)
|
6
|
42 (8.8%)
|
33 (11.7%)
|
7
|
89 (18.7%)
|
31 (11.0%)
|
8
|
87 (18.2%)
|
57 (20.3%)
|
9
|
127 (26.6%)
|
77 (27.4%)
|
10
|
56 (11.7%)
|
29 (10.3%)
|
ACL, anterior cruciate ligament; n, number of patients; cm, centimeters; kg, kilogram; BMI, body mass index; m, meters; HV, high-volume clinic; LV, low-volume clinic; PROs, patient reported outcomes.
For categorical variables, n (%) is presented.
For continuous variables, the mean ± SD is presented.
Patients treated at HV clinics reported a greater Knee injury and Osteoarthritis Outcome Score-Sports (KOOS-Sports) score at 2 months after ACL reconstruction, compared with patients treated at LV clinics, although the effect size was small (31.3±22.1 HV vs. 27.2±19.9 LV, P = 0.014, d = 0.186). The results from all KOOS subscales during the first year are shown in Fig. 3 and Table 4. There were no differences between the groups for any of the other PROs included in this study at any follow-up during the first year after ACL reconstruction. The values for K-SES18 and ACL-RSI are listed in Additional file 4.
Table 4
Subscales of the KOOS for high- and low-volume clinics, first year after ACL-reconstruction
PROs
|
Group
|
2 months
|
4 months
|
8 months
|
12 months
|
KOOS-Pain
|
HV
|
77.1 ± 14.2
|
81.1 ± 13.0
|
85.4 ± 12.6
|
87.3 ± 12.8
|
LV
|
75.1 ± 14.1
|
80.0 ± 13.2
|
84.8 ± 12.0
|
87.9 ± 10.6
|
KOOS-Symptom
|
HV
|
63.8 ± 17.3
|
71.5 ± 16.3
|
76.9 ± 16.3
|
79.5 ± 16.1
|
LV
|
63.3 ± 17.3
|
70.0 ± 17.3
|
75.9 ± 16.1
|
79.3 ± 15.2
|
KOOS-Sports
|
HV
|
31.3 ± 22.1*
|
47.7 ± 22.6
|
64.8 ± 21.9
|
71.7 ± 23.1
|
LV
|
27.2 ± 19.9*
|
46.9 ± 21.8
|
63.9 ± 21.4
|
71.9 ± 21.4
|
KOOS-Quality
|
HV
|
38.4 ± 16.2
|
45.2 ± 17.4
|
54.0 ± 18.9
|
60.7 ± 19.8
|
LV
|
36.7 ± 15.6
|
43.7 ± 15.1
|
53.0 ± 18.3
|
60.4 ± 18.9
|
ACL, anterior cruciate ligament; KOOS, Knee injury and osteoarthritis outcome score; PROs, Patient reported outcomes. * Significant (p < 0.05) difference between high- and low-volume clinics. All results are presented as mean ± SD.
Muscle strength
For muscle strength testing, 497 patients attended every follow-up during the first year after ACL reconstruction, of which 68% and 32% represented the HV and LV groups, respectively (Table 5). No significant differences were observed in the demographic characteristics of this cohort.
Table 5
Demographics for patients attending muscle strength testing at all follow-ups, first year after ACL-reconstruction.
|
HV
|
LV
|
P-value
|
n
|
338
|
159
|
|
Women, n (%)
|
182 (53.8%)
|
90 (56.6%)
|
0.63
|
Age (years)
|
30 ± 11
|
31 ± 12
|
0.31
|
Height (cm)
|
174 ± 9
|
174 ± 8
|
0.36
|
Weight (kg)
|
72 ± 13
|
73 ± 13
|
0.67
|
BMI (kg/m^2)
|
23.8 ± 3.5
|
23.7 ± 3.5
|
0.61
|
Days between injury and reconstruction
|
248 ± 472
|
430 ± 1065
|
0.06
|
Type of graft
|
|
|
|
Hamstrings
|
269 (80.5%)
|
129 (82.7%)
|
0.26
|
Patella
|
56 (16.8%)
|
25 (16.0%)
|
Allograft
|
1 (0.3%)
|
2 (1.3%)
|
Quadriceps
|
2 (0.6%)
|
0 (0%)
|
Other
|
6 (1.8%)
|
0 (0%)
|
ACL. Anterior cruciate ligament; n, number of patients; cm, centimeters; kg, kilograms; m, meters; BMI, body mass index; HV, high-volume clinics; LV, low-volume clinics.
For categorical variables, n (%) is presented.
For continuous variables, the mean ± SD is presented.
Patients treated at HV clinics presented more symmetrical hamstring strength at 2 months follow-up, although a small effect size (83.3%, ± 19.5% HV vs. 78.4%, ± 21.0% LV, P = 0.014, d = 0.248) and greater passing rates (≥ 90%) for quadriceps strength at 4 months follow-up (33.4% HV vs. 24.5% LV, P = 0.048) compared with patients treated at an LV clinic (Fig. 4 and Table 6). There were no other differences in the strength symmetry or passing rates during the first year after ACL reconstruction.
Table 6
Muscle strength in the quadriceps and hamstrings muscle groups, first year after ACL-reconstruction.
|
|
|
2 months
|
4 months
|
8 months
|
12 months
|
Muscle test
|
Group
|
n
|
LSI ± SD%
|
Pass, n(%)
|
p-value LSI/Pass
|
LSI ± SD%
|
Pass, n(%)
|
p-value LSI/Pass
|
LSI ± SD%
|
Pass, n(%)
|
p-value LSI/Pass
|
LSI ± SD%
|
Pass, n(%)
|
p-value LSI/Pass
|
Quadriceps strength
|
HV
|
338
|
70.5± 20.0
|
37 (10.9)
|
0.55/0.42
|
79.8± 15.7
|
113 (33.4)
|
0.95/0.05*
|
91.0± 13.9
|
201 (59.5)
|
0.33/1.00
|
93.8± 10.6
|
232 (68.6)
|
0.48/0.35
|
LV
|
159
|
69.3± 17.5
|
13 (8.2)
|
79.7± 14.7
|
39 (24.5)
|
89.5± 14.7
|
94 (59.1)
|
94.7± 11.1
|
116 (73.0)
|
Hamstrings strength
|
HV
|
338
|
83.3± 19.5
|
118 (34.9)
|
0.01*/0.76
|
92.9± 13.5
|
204 (60.4)
|
0.24/0.92
|
97.4± 12.9
|
253 (74.9)
|
0.38/0.83
|
97.8± 13.5
|
274 (81.1)
|
0.45/0.55
|
LV
|
159
|
78.4± 21.0
|
53 (33.3)
|
91.3± 12.7
|
95 (59.7)
|
96.2± 14.4
|
117 (73.6)
|
98.8± 11.7
|
125 (78.6)
|
LSI = limb symmetry index, SD = standard deviation, HV = high-volume clinic, LV = low-volume clinic, n = number of patients, Pass = LSI ≥90%, *= Significant (p < 0.05) difference between high- and low-volume clinics.
For categorical variables, n (%) is presented.
For continuous variables, the mean ± SD is presented.
Muscle function and patient-reported outcomes
The number of patients who had data from muscle strength tests was 497, while 757 had data from PROs; thus, 328 patients had data from both muscle function and PROs at all follow-up visits. Out of the 328 patients, 70% and 30% were in the HV and LV groups, respectively (Table 7). There were no differences in demographics between HV and LV clinics.
Table 7
Demographics for muscle function and PROs for high- and low-volume clinics, first year after ACL-reconstruction
|
HV
|
LV
|
P-value
|
n
|
231
|
97
|
|
Women, n (%)
|
120 (51.9%)
|
51 (52.6%)
|
1.0
|
Age (years)
|
29 ± 10
|
30 ± 11
|
0.37
|
Height (cm)
|
174 ± 9
|
175 ± 8
|
0.22
|
Weight (kg)
|
72 ± 14
|
73 ± 13
|
0.48
|
BMI (kg/m^2)
|
23.7 ± 3.6
|
23.7 ± 2.9
|
0.98
|
Days between injury and reconstruction
|
240 ± 423
|
407 ± 1080
|
0.14
|
Type of graft
|
|
|
0.62
|
Hamstrings
|
187(81.7%)
|
78 (83.0%)
|
Patella
|
39 (17.0%)
|
15 (16.0%)
|
Allograft
|
0 (0%)
|
1 (1.1%)
|
Quadriceps
|
1 (0.4%)
|
0 (0%)
|
Other
|
2 (0.9%)
|
0 (0%)
|
ACL, anterior cruciate ligament; n, number of patients; cm, centimetres; kg, kilogram; BMI, body mass index; m, meters; HV, high-volume clinic; LV, low-volume clinic.
For categorical variables, n (%) is presented.
For continuous variables, the mean ± SD is presented.
There were no significant differences between patients treated at HV clinics and those treated at LV clinics in LSI or passing rates for muscle function tests at any follow-up. Limb symmetry indices and passing rates for all muscle function tests are presented in Additional file 5 and 6. Table 8 presents the differences in PROs at follow-up between the HV and LV groups for patients with data from both muscle function tests and PROs at all follow-up visits during the first year after ACL reconstruction. The HV group presented significantly greater K-SES18 at 2 months follow-up (4.5±1.8 HV vs. 3.9±1.7 LV, P = 0.01, d = 0.30), KOOS-Sports at 2 months follow-up (34.1±22.0 HV vs. 27.2±20.4 LV, P = 0.01, d = 0.32), and ACL-RSI at 12 months follow-up (77.7±29.2 HV vs. 69.3±32.9 LV, P = 0.02, d = 0.28), all results had a small effect size.
Table 8
Patient-reported outcomes for high- and low-volume clinics, all follow-ups, first year after ACL-reconstruction
PROs
|
Group
|
2 months
|
4 months
|
8 months
|
12 months
|
K-SES18 Present
|
HV
|
4.5 ± 1.8*
|
6.3 ± 1.7
|
8.2 ± 1.2
|
8.9 ± 1.1
|
LV
|
3.9 ± 1.7
|
6.0 ± 1.6
|
8.0 ± 1.1
|
8.8 ± 1.1
|
K-SES18 Future
|
HV
|
7.6 ± 1.6
|
7.6 ± 1.6
|
7.8 ± 1.6
|
7.9 ± 1.6
|
LV
|
7.5 ± 1.6
|
7.7 ± 1.4
|
7.9 ± 1.2
|
7.9 ± 1.4
|
KOOS-Pain
|
HV
|
79.9 ± 12.2
|
84.2 ± 10.1
|
88.9 ± 9.1
|
90.5 ± 10.4
|
LV
|
78.1 ± 12.3
|
83.5 ± 10.6
|
87.7. ± 10.4
|
90.2 ± 9.8
|
KOOS-Symptom
|
HV
|
67.2 ± 15.9
|
74.9 ± 14.5
|
80.7 ± 14.4
|
83.6 ± 13.5
|
LV
|
66.4 ± 16.6
|
75.0 ± 14.3
|
79.1 ± 16.3
|
83.3 ± 15.5
|
KOOS-Sports
|
HV
|
34.1 ± 22.0*
|
52.8 ± 21.9
|
72.7 ± 17.7
|
80.9 ± 17.2
|
LV
|
27.2 ± 20.4
|
51.8 ± 20.6
|
70.9 ± 16.0
|
78.6 ± 17.8
|
KOOS-Quality
|
HV
|
40.4 ± 16.2
|
48.1 ± 16.3
|
58.5 ± 17.3
|
67.2 ± 16.3
|
LV
|
37.7 ± 16.6
|
46.9 ± 14.8
|
57.9 ± 16.3
|
65.3 ± 18.1
|
ACL-RSI
|
HV
|
|
|
66.7 ± 30.0
|
77.7 ± 29.2*
|
LV
|
59.6 ± 31.9
|
69.3 ± 32.9
|
PROs, patient-reported outcomes; K-SES18, Knee Self-efficacy scale; KOOS, Knee injury and osteoarthritis outcome score; ACL-RSI, anterior cruciate ligament return to sports index; HV, high-volume clinic; LV, low-volume clinic; *, significant (p < 0.05) difference between HV and LV clinics. Results are presented as mean ± SD.