Key words
anterior cruciate ligament - primary repair - arthroscopy - reconstruction - meta-analysis - randomized controlled trial
Introduction
Anterior cruciate ligament (ACL) injury is one of the most common types of knee joint
injury. According to statistics, 85 out of every 100,000 people aged 16–39
suffer from ACL injury [1]. Mayo et al. successfully
performed one-stage open ACL repair surgery for the first time since 1895 and
reported the good results of the surgery [2]. By the
1970s, after a long-term follow-up study of ACL repair Feagin and other scholars
found that although the early follow-up results were satisfactory, the long-term
curative effect was not good enough, and the rate of patients receiving reoperation
within 5 years was also relatively high [3]
[4]
[5]
[6]. Therefore, primary ACL repair surgery was no longer
popular.
After that, ACLR gradually replaced ACL repair as the mainstream surgical method for
the treatment of ACL injury. However, many studies have pointed out that ACLR
surgery has some deficiencies such as postoperative ligament proprioception loss,
tendon donor site complications, autologous or allogeneic graft infection, etc., and
the postoperative functional recovery still needs to be improved [7]
[8]. This may be
related to the fact that the reconstructed ligament cannot effectively restore the
normal anatomical structure and physiological function of ACL [9]
[10]. Therefore, more
and more scholars are focusing on preserving the biology of ACL to improve the
surgical results. With the wide application of arthroscopy, the use of new surgical
instruments and implants, and the deepening understanding of the biological
knowledge of ACL, people have generated new interest in ACL repair [11]
[12]
[13]
[14]. At the same
time, compared with ACLR, ACL repair has less damage, no donor site complications
and can restore active function earlier [15]
[16].
In recent years, we have reason to think that ACL repair should be re-evaluated with
the rapid development of some arthroscopic techniques and the adjustment of
postoperative rehabilitation strategies. Considering that some results in historical
literature were not satisfactory [17]
[18]
[19]
[20], we will re-evaluate the safety and effectiveness
of one-stage ACL repair technology by meta-analysis. Although some review studies
have been reported in recent years [21]
[22]
[23], there has been
no high-quality systematic review related to randomized controlled trial (RCT). The
objective of this study is to evaluate all clinical RCT research of primary ACL
repair (open and arthroscopic) in recent decades, and compare the results between
ACL repair and reconstruction, so as to provide more reliable evidence for clinical
treatment.
Materials and Methods
Retrieval strategies
We searched PubMed, EMBASE, Springer, Ovid, the Cochrane Library, and other
medical literature databases for the literature related to the comparison of
clinical outcomes between one-stage ACL repair and ACLR in all adults published
between January 1970 and June 2021. Keywords: anterior cruciate ligament,
injury, repair, reconstruction. The type of studies included was RCT only. Also,
review articles on this topic were reviewed to retrieve relevant studies that
might have been missed.
Inclusion criteria
Inclusion criteria included (1) diagnosis of ACL injury; (2) RCT; (3)
intervention: experimental group with ACL repair techniques; control group with
conventional ACLR. (4) The observation indexes included: prognostic indexes
(Tegner, Lysholm, IKDC scores), physical examination results (Lachman test,
range of motion, tibial anterior displacement), reoperation rate, and functional
outcomes.
Exclusion criteria
Exclusion criteria included the following: (1) non-RCT studies; (2) no relevant
interventions were included in the above types of literature; (2) follow-up less
than 12 months; (3) cadaveric studies, biomechanical studies, and in vitro or
animal studies; and (4) duplicate published studies were excluded, and
abstracts, lectures, and reviews were also excluded.
Data extraction and quality evaluation
We extracted relevant data by retrieving information and summarized them into
tables and forest plots. The quality of the included studies was evaluated using
Revman software. The parameters included sequence generation (selection bias),
allocation hiding (selection bias), blindness (performance bias), incomplete
result data (detection bias), selective result reporting (reporting bias), and
other issues. Each parameter could be classified as low risk, high risk, or
unclear.
Statistical analysis
Statistical analyses were performed using Revman manager 5.3 software (Cochrane
Collaboration, NordicCochrane Centre, Copenhagen, Denmark). Continuous variables
were analyzed using weighted mean differences, and categorical variables were
assessed using relative risk or absolute risk differences. p<0.05 was
considered statistically significant. Heterogeneity analysis was tested by
Q-statistic (P<0.1), and I2-statistic
(I2>50%). When there was no statistically significant
heterogeneity, a fixed-effects model was used; conversely, a random-effects
model was used. In addition, we performed subgroup analyses depending on the
intervention.
Results
Study selection
The literature search identified 86 papers that met the study objectives, and we
selected 7 RCTs that met the inclusion criteria [17]
[18]
[19]
[20]
[24]
[25]
[26], with a total of 745 patients, of which a total of 61 patients
were lost to follow-up, with the rate of 8.2%. The literature search
process is shown in [Fig. 1], and the basic
characteristics of these studies are shown in [Table
1].
Fig. 1 Search strategy flow diagram.
Table 1 Basic characteristics of included
studies
|
Included Studies
|
N
|
Age (y)
|
Repair technique
|
Reconstruction technique
|
Injury to operation
|
F/U
|
|
(M%/F%)
|
Mean±SD (Min–Max)
|
|
|
Interval (d)
|
Duration (y)
|
n (% Lost F/U)
|
|
Engebretsen et al. (1990) [24]
|
150 (54%/46%)
|
28.7 (16–50)
|
Primary repair with or without LAD
|
BPTB
|
10
|
2
|
3 (2%)
|
|
Grontvedt et al. (1996) [25]
|
150
|
29 (16–50)
|
Primary repair with or without LAD
|
BPTB
|
10
|
5
|
9 (6%)
|
|
Sporsheim et al. (2019) [26]
|
150
|
29 (16–54)
|
Primary repair with or without LAD
|
BPTB
|
NA
|
30
|
37 (24.7%)
|
|
Schliemann et al. (2018) [17]
|
62 (62%/38%)
|
28.7±11.4
|
DIS
|
ACLR (semitendinosus autograft)
|
21
|
1
|
2 (3.2%)
|
|
Hoogeslag et al. (2019) [20]
|
48 (77%/23%)
|
21.5±2.7
|
DIS
|
ACLR (semitendinosus autograft)
|
21
|
2
|
4 (8.3%)
|
|
Murray et al. (2020) [18]
|
100 (44%/56%)
|
17±1.5
|
BEAR
|
ACLR (semitendinosus autograft)
|
45
|
2
|
4 (4%)
|
|
Sters et al. (2020) [19]
|
85 (66%/34%)
|
28.2±11 (18–46)
|
DIS
|
ACLR (semitendinosus autograft)
|
NA
|
1
|
2 (2.3%)
|
LAD, ligament- augmentation device; BPTB, bone-patella tendon-bone; FU,
follow-up; DIS, dynamic intraligamentary stabilization; BEAR,
bridge-enhanced anterior cruciate ligament repair; ACLR, anterior
cruciate ligament reconstruction; NA, not applicable.
Surgical techniques
ACL repair technique
There were 3 papers on open ACL repair [24]
[25]
[26], including 109 patients. The surgical techniques consisted of
2 main categories: primary repair without augmentation or with ligament
augmentation device (LAD). The surgical procedures were described in detail
in previous literature [24]
[25]
[27]. primary
repair technique of the 3 studies was performed according to the method
reported by Palmer [28].
The arthroscopic ACL repair technique had been reported in 4 papers [17]
[18]
[19]
[20],
including 160 patients. These patients were treated with the DIS technique
and BEAR technique, respectively. The DIS procedure was performed according
to the technique described by Kösters [29] and Eggli [30]. A total of 96
patients were included. The BEAR procedure was performed according to the
technique described by Murray [31] and
included a total of 64 patients.
ACLR technique
ACLR interventions were used in all seven publications, including a total of
415 patients. The ACLR grafts used included: 1. Bone-patellar tendon-bone
graft; 2. autologous semitendinosus-gracilis tendon graft.
Quality assessment
We performed a quality assessment of the seven included RCTs using the
Cochrane Risk of Bias Assessment Tool. The entire assessment was performed
by two reviewers separately, and any disagreements were resolved by a third
reviewer. As shown in [Fig. 2], the quality
of the included studies was high. The funnel plot shows no visual evidence
of publication bias.
Fig. 2
a Risk of bias graph exhibiting the review of the
authors’ judgments about each risk of bias item presented as
percentages across all included studies. b Risk of bias
summary revealing the review of the authors’ judgments about
each risk of bias item for included RCTs. Minus sign represents the
risk of bias present, plus sign indicates the risk of bias absent,
and question mark equals the risk of bias uncertain. c The
funnel plots of the included studies. RR, relative risks; SE,
standard error.
Meta-analysis results
The seven included studies used different knee function scoring systems. We
divided the results of the studies into two groups, the experimental ACL
repair group and the control ACLR group, for comparison. It needs to be
mentioned that we combined the data from the ACL repair with or without LAD
group at the same time for the meta-analysis, and did the independent
subgroup analysis respectively, in order to evaluate the results of the
meta-analysis in a comprehensive manner.
Knee clinical scores
Tegner score and subgroup analysis
We included five studies comparing the results of postoperative Tegner scores in
the two groups. The Tegner scores in the two groups were 3–6.8 and
4–7.1, respectively. The difference between the two groups was
statistically significant, and overall, the postoperative Tegner score was
higher in the ACLR group than in the ACL repair group
(SMD=−0.55, 95%CI −0.88 to −0.21,
p=0.001, I2=0%) ([Fig. 3a]).
Fig. 3
a Difference in the Tegner score and the subgroup analysis;
b Difference in the Lysholm score and the subgroup analysis.
CI, confidence interval; IV, inverse variance; SD, standard deviation.
The solid squares indicate the mean difference and are proportional to
the weights used in the meta-analysis. The solid vertical line indicates
no effect. The horizontal lines represent the 95% CI. The
diamond indicates the weighted mean difference, and the lateral tips of
the diamond indicate the associated 95% CI.
We also performed subgroup analysis by intervention and showed that there was no
statistically significant difference in Tegner scores between arthroscopic ACL
repair and ACLR (SMD=−0.22,95% CI −0.82 to 0.39,
P=0.49, I2=0%). In contrast, the difference
between open ACL repair and ACLR was statistically significant
(SMD=−0.69,95% CI −1.09 to −0.29,
P=0.0007, I2=0%). Overall, the postoperative
Tegner score was higher in the ACLR group than in the open ACL repair group
([Fig. 3a]).
Lysholm score and subgroup analysis
There were five included studies comparing the results of postoperative Lysholm
scores between the two groups. The differences between the two groups were
statistically significant, with higher postoperative Lysholm score in the ACLR
group than in the ACL repair group overall
(SMD=−3.26,95%CI −5.98 to −0.54,
p=0.02, I2=67%) ([Fig.
3b]).
We also performed subgroup analysis by intervention and showed that there was no
statistically significant difference in Lysholm scores between arthroscopic ACL
repair and ACLR (SMD=2.35,95%CI −1.97 to 6.66,
P=0.29, I2=0%). In contrast, the difference between open
ACL repair and ACLR was statistically significant, with higher postoperative
Lysholm score in the ACLR group than in the open ACL repair group overall
(SMD=−4.80,95% CI −6.24 to −3.36,
P<0.05, I2=0%). ([Fig.
3b]).
IKDC scores
There were four included studies of arthroscopic ACL repair comparing the results
of postoperative IKDC scores between the two groups. The IKDC scores in the two
groups were 85.7–95.4 and 84.8–94.3, respectively. The
difference between the two groups was statistically significant, and the
postoperative IKDC scores were higher in the arthroscopic ACL repair group than
in the ACLR group overall (SMD=2.12,95%CI 0.14 to 4.10,
p=0.04. I2=0%) ([Fig.
4a]).
Fig. 4
a Difference in the IKDC score; b Difference in the
incidence of KT-1000 (≥3 mm). CI, confidence interval;
IV, inverse variance; M-H, Mantel-Haenszel. The solid squares indicate
the mean difference and are proportional to the weights used in the
meta-analysis. The solid vertical line indicates no effect. The
horizontal lines represent the 95% CI. The diamond indicates the
weighted mean difference, and the lateral tips of the diamond indicate
the associated 95% CI.
Physical examination results
Physical examination results
Lachman test
There were five included studies comparing the postoperative Lachman test results
(2+/3+) between the two groups. There was no statistically
significant difference in the postoperative Lachman test results between the two
groups, with Lachman 2+/3+rates of 22.3 and 7.8%,
respectively (SMD=0.09, 95% CI −0.06 to 0.24,
P=0.24, I2=90%) ([Fig.
5]).
Fig. 5 Difference in the incidence of Lachman test
(2+/3+) and the subgroup analysis; CI, confidence
interval; IV, inverse variance; M-H, Mantel-Haenszel. The solid squares
indicate the mean difference and are proportional to the weights used in
the meta-analysis. The solid vertical line indicates no effect. The
horizontal lines represent the 95% CI. The diamond indicates the
weighted mean difference, and the lateral tips of the diamond indicate
the associated 95% CI.
We also performed subgroup analysis by intervention and showed that there was no
statistically significant difference in Lachman test results between
arthroscopic ACL repair and ACLR (SMD=−0.02, 95% CI
−0.08 to 0.04, P=0.48, I2=0%). In contrast, the
difference between open ACL repair and ACLR was statistically significant
(SMD=0.18, 95% CI 0.01 to 0.34, P=0.04,
I2=77%). ([Fig. 3a]). Overall,
the rate of postoperative Lachman test 2+/3+was lower in
the ACLR group than in the open ACL repair group, which were 9.7 and
30.7%, respectively.
Anterior-posterior knee stability test (KT-1000)
All three included open ACL repair studies compared the results of the
postoperative KT-1000 test (>=3 mm) between the two
groups. The results showed no statistically significant difference between the
two groups (SMD=1.58, 95% CI 0.54 to 4.62, P=0.40,
I2=90%) ([Fig. 4b]).
Knee flexion mobility
There were four included studies comparing the results of postoperative knee
flexion mobility changes between the two groups. The results showed no
statistically significant difference between the two groups, with 8.2 and
6.9% of knee flexion limitations greater than 10°,
respectively. (SMD=1.22, 95% CI 0.62 to 2.42,
P=0.56, I2=0%) ([Fig.
6a]).
Fig. 6
a Difference in the incidence of flexion limitation
(≥10°); b Difference in the incidence of
extension limitation (≥5°). CI, confidence interval;
IV, inverse variance; M-H, Mantel-Haenszel. The solid squares
indicate the mean difference and are proportional to the weights
used in the meta-analysis. The solid vertical line indicates no
effect. The horizontal lines represent the 95% CI. The
diamond indicates the weighted mean difference, and the lateral tips
of the diamond indicate the associated 95% CI.
Knee extension mobility
There were four included studies comparing the results of postoperative knee
extension mobility changes between the two groups. The results showed no
statistically significant difference between the two groups, with 9.6 and
13.2% of knee extension limitations greater than 5°,
respectively. (SMD=0.76, 95% CI 0.45 to 1.30,
P=0.32, I2=3%) ([Fig.
6b]).
knee functional outcomes
There were 2 studies assessing the strength changes of muscles surrounding the
knee joint in patients after surgery, such as the hamstrings, quadriceps and hip
abductor muscle groups. The results were as follows: There was no statistically
significant difference in the comparison of knee muscle functional outcomes
between the two groups. (SMD=0.27, 95% CI −2.69 to 3.23,
P=0.86, I2=92%) ([Fig.
7]).
Fig. 7 Difference in the functional outcomes of muscle strength
and the subgroup analysis. CI, confidence interval; IV, inverse
variance; SD, standard deviation. The solid squares indicate the mean
difference and are proportional to the weights used in the
meta-analysis. The solid vertical line indicates no effect. The
horizontal lines represent the 95% CI. The diamond indicates the
weighted mean difference, and the lateral tips of the diamond indicate
the associated 95% CI.
Reoperation rate
There were five included studies comparing the reoperation rates during postoperative
follow-up between the two groups. The results showed that the difference between the
two groups in postoperative reoperation rates was not statistically significant,
with rates of 15.5 and 9.8%, respectively (SMD=1.61, 95% CI
0.99 to 2.61, P=0.06, I2=31%) ([Fig. 8]).
Fig. 8 Difference in the incidence of reoperation and the subgroup
analysis. CI, confidence interval; M-H, Mantel-Haenszel. The solid squares
indicate the mean difference and are proportional to the weights used in the
meta-analysis. The solid vertical line indicates no effect. The horizontal
lines represent the 95% CI. The diamond indicates the weighted mean
difference, and the lateral tips of the diamond indicate the associated
95% CI.
Our subgroup analysis by intervention showed that there was no statistically
significant difference in the reoperation rates between arthroscopic ACL repair and
ACLR (SMD=1.02, 95% CI 0.48 to 2.18, P=0.95,
I2=0%). In contrast, the difference between open ACL repair and ACLR
was statistically significant, and the rate of postoperative reoperation was lower
in the ACLR group than in the open ACL repair group overall, which were 7.4 and
15.4%, respectively (SMD=2.05, 95%CI 1.08 to 3.88,
P=0.03, I2=48%) ([Fig.
8]).
Subgroup analysis of LAD
Finally, we performed subgroup analysis on whether to use LAD for ACL repair or not.
And the statistical analysis was performed separately according to the type of data,
and the results were summarized in ([Fig. 9]
[ 10]).
Fig. 9 Difference in the continuous variable results for ACL repair
and the subgroup analysis. CI, confidence interval; IV, inverse variance;
SD, standard deviation; The solid squares indicate the mean difference and
are proportional to the weights used in the meta-analysis. The solid
vertical line indicates no effect. The horizontal lines represent the
95% CI. The diamond indicates the weighted mean difference, and the
lateral tips of the diamond indicate the associated 95% CI.
Fig. 10 Difference in the categorical variable results for ACL repair
and the subgroup analysis. CI, confidence interval; M-H, Mantel-Haenszel.
The solid squares indicate the mean difference and are proportional to the
weights used in the meta-analysis. The solid vertical line indicates no
effect. The horizontal lines represent the 95% CI. The diamond
indicates the weighted mean difference, and the lateral tips of the diamond
indicate the associated 95% CI.
The results showed that for ACL repair with or without LAD assistance, there was no
statistically significant difference between the two groups for comparison of either
subjective knee scores or objective examination findings.
(SMD=−0.18, 95%CI −0.67 to 0.31, P=0.48,
I2=30%; SMD=0.06, 95%CI −0.00 to 0.12,
P=0.06, I2=63%).
Moreover, we performed subgroup analyses of each scoring system and found that there
were no statistically significant differences in the results of each test except for
the Lysholm score and KT-1000 test (P=0.89, 0.32, 0.50, 0.95, 0.68), which
showed better results in the ACL+LAD group than in the ACL repair alone
(P=0.01 and 0.003).
Postoperative rehabilitation protocols
Postoperative rehabilitation protocols
We summarized the protocols reported in the 7 included RCT studies regarding
postoperative rehabilitation, as summarized in the table below. the 3 open ACL
repair studies basically used the same rehabilitation protocol: long leg cast
immobilization for 2 weeks, brace immobilization for 6 weeks, weight bearing after 8
weeks, muscle rehabilitation exercises after 12 weeks, and return to sports after 12
months. Compared to open ACL repair, the rehabilitation protocols of the 4
arthroscopic ACL repair studies would be relatively more aggressive. There was no
cast fixation, with adjustable brace use ranging from 4 days to 12 weeks, full
weight bearing after 4 weeks, and a lower requirement for knee ROM limitation, with
return to sports after 5–6 months ([Table
2]).
Table 2 Postoperative Rehabilitation Protocols
|
Included studies
|
Cast immobilization
|
brace immobilization
|
Weight bearing
|
Knee ROM
|
Guided physiotherapy
|
|
|
|
Partial
|
full
|
|
Closed Chain
|
Proprioceptive exercises
|
running
|
return to sports
|
|
Engebretsen et al. (1990) [24]
|
2 wk
|
30° of flexion for 6 wk
|
After 8 wk
|
|
30–60° at 3–4wk
|
After 12 wk
|
|
NA
|
12mo
|
|
|
|
|
|
30–90° at 5–8wk
|
|
|
|
|
|
Grontvedt et al. (1996) [25]
|
2 wk
|
30° of flexion for 6 wk
|
After 8 wk
|
|
30–60° at 3–4wk
|
After 12 wk
|
|
NA
|
12mo
|
|
|
|
|
|
30–90° at 5–8wk
|
|
|
|
|
|
Sporsheim et al. (2019) [26]
|
2 wk
|
30° of flexion for 6 wk
|
After 8 wk
|
|
NA
|
NA
|
|
NA
|
12mo
|
|
Schliemann et al. (2018) [17]
|
4 d
|
NA
|
After 2 wk
|
|
Unrestricted after 2 wk
|
2wk-3wk
|
3wk-6wk
|
After 6wk
|
5 mo
|
|
Hoogeslag et al. (2019) [20]
|
DIS: A long-leg cast for 5d; ACLR: immediate Unrestricted; then
received a near-identical, structured, criteria-based
rehabilitation protocol
|
|
|
|
|
|
|
|
|
|
Murray et al. (2020) [18]
|
NA
|
locking for 6 wk*
|
4 wk
|
After 4 wk
|
0–50° at 2wk
|
physical therapy protocol from MOON
|
|
|
NA
|
|
|
movable at 6–12wk*
|
|
|
0–90° at 2–6 wk
|
|
|
|
|
|
Sters et al. (2020) [19]
|
NA
|
5d
|
20 kg at 0–2wk
|
After 2 wk
|
Unrestricted after 5d
|
5d to 4 wk
|
After 4 wk
|
NA
|
6 mo
|
The rehabilitation protocols depicted in the table are for the patients who
underwent primary ACL repair or reconstruction.; ROM, range of motion; NA,
not applicable; d, day; wk, week; mo, month; ACLR, anterior cruciate
ligament reconstruction; DIS, dynamic intraligamentary stabilization; MOON,
Multicenter Orthopaedics Outcomes Network; Kg, kilograms.;
*Use of locked hinge knee brace for 6 weeks, then use of
functional ACL brace for 6 to 12 weeks.
Discussion
This meta-analysis evaluates the difference in surgical efficacy between ACL repair
and ACLR. The main finding is that there was no significant difference in clinical
results between ACL repair and ACLR group, including IKDC, range of motion (ROM),
Lachman test, laxity difference, reoperation rate and muscle strength. No matter
whether LAD was used or not, there was no obvious difference in the postoperative
curative effect of ACL repair. Except for Tegner and Lysholm scores, which showed
that the ACLR group had better recovery of postoperative motor function, the above
results were promising. At the same time, subgroup analysis showed that the
short-term follow-up results of arthroscopic ACL repair group were indeed better
than those of open ACL repair group and generally comparable to those of the ACLR
group.
Open primary repair of ACL injury was gradually abandoned many years ago, Part of the
reason is that the follow-up results for patients were not satisfactory, which
showed that pain, joint swelling, instability, persistent symptoms were not
relieved, and the incidence of reoperation rate was high. The research reported by
Feagin showed that although the early curative effect was satisfactory, in the
following 5 years of follow-up, only 5 of the initial 64 patients had symptoms
relieved. 91% of the patients had unstable conditions, and 15 patients need
reoperation [4]. Although other literatures had
reported that the success rate could reach 75% in 6 years, there were still
a large number of literatures confirming the above unsatisfactory results. This
study included three research on open ACL repair. After integrating the data of open
ACL repair with or without LAD, we found that in the short-term follow-up
(2–5year), the results showed that the failure rate was higher in the open
ACL repair group, the instability gradually increased with time and stabilized after
5 years, with lower activity and function levels. The BPTB group was indeed much
better than the open ACL repair group. This may be one of the reasons why
meta-analysis results showed that Tegner and Lysholm scores of open ACL repair group
were lower than those of ACLR group. This was also consistent with the results of
the above historical literature. Although the 5-year follow-up results showed that
there was no significant difference in knee ROM, and the reoperation rate of open
ACL repair was basically the same as that of BPTB group, LARS and other scholars
still suggested that non-enhanced ACL repair should not be performed again. On the
other hand, the results of 30-year follow-up of ACL-repair patients by Anne showed
that with the increase of follow-up time, the stability of knee joint in each group
was increasing, and few patients still had substantial relaxation. The unstable
patients who appeared during the follow-up period of 5–16 years also
obtained more stable year by year. Meta-analysis also showed that there was no
significant difference in knee stability and ROM among the groups. The average age
of 30-year follow-up patients reported by ANNE was 60 years old, which was another
reason for the lower Tegner and Lysholm scores of the two groups, as the activity
level of the elderly is usually lower. In addition, the development of
osteoarthritis (OA) was also a factor to be considered for the recovery of knee
stability.
Another point to note is that the reoperation rate of ACL repair group after 30-year
follow-up reported by ANNE was higher than that of ACLR group. This was not appeared
during short-term follow-up. The main defect of ACL repair is in the difficulty of
healing. The repair of ACL is relatively difficult in the synovial environment.
After the fibrin plug in the synovial space is destroyed by fibrinolytic enzyme,
fibroblasts will cover the surface of ACL, and the early healing will be prevented
[32]. This undoubtedly makes ACL repair more
difficult. Cabaud er al. had confirmed that early ACL repair would always fail
completely, and the use of LAD has better results [33]. LAD was first invented by Kennedy to enhance the repair or
reconstruction of ACL injuries [34]. Schabus first
applied LAD and reported good clinical results [35].
Engebretsen et al. had confirmed in cadaver research that LAD can provide about
75% of the extension and flexion activity load of ACL tissue in the early
stage, which protects the early repair of ACL from being damaged until the tissue
fully grows and the repair is completed [36]. This is
consistent with our meta-analysis results. In short-term follow-up (2–5
years), the clinical outcomes, stability and activity function in LAD group were
indeed better than those without LAD. But the protective function of LAD can only
last for about 1 year. So, with the increase of follow-up time, this advantage
gradually disappeared. As mentioned before, the reoperation rate of ACL repair was
significantly higher than that of BPTB group, which was consistent with the similar
results reported by some scholars: supporting the disadvantage of LAD compared with
autologous BPTB or hamstring tendon transplantation [37]
[38]. This is not only related to the
degeneration and aging of tissues after ACL repair, but also related to the wear of
LAD and the mechanical deterioration of synthetic material fragments [39].
Therefore, through analysis, it was not difficult to figure out why open ACL repair
was abandoned by surgeons at that time: the invasive and rough technology of open
arthrotomy, long time fixation and high revision rate were the important reasons.
Although some scholars put forward the view that the disappointing effect of open
ACL repair was largely due to the wrong choice of patients. After reviewing
historical literature, Van et al. found that the location of ACL rupture seems to
play an important role in the prognosis of open ACL repair [40]. Some studies had confirmed that selective open ACL repair for
proximal ACL rupture had a very good prognosis and would not deteriorate with time,
while the prognosis for middle ACL rupture was disappointing [41]. Unfortunately, there was no more high-quality
research to explore this issue. The RCTs we included also did not statistically
analyze the location of ACL rupture. Therefore, we could not make more
judgments.
With the rapid development of arthroscopy and arthroscopic surgery in recent years,
there is a renewed interest in the primary repair of ACL injury. Especially, the
research on the proprioceptive effect of ACL tissue is deepening. Some studies had
shown that preserving the injured stump of ACL can improve the mechanical stability
of knee joint after operation and allow earlier and more active rehabilitation
exercise, which is very important for athletes with ACL injury [1]. Therefore, the main concern of modern joint
surgeons is how to keep this function in patients with ACL injury, especially
athletes. The intervention measures of arthroscopic ACL repair included in this
study were DIS and BEAR. The above two approaches can effectively preserve their own
ACL stumps while enhancing the biomechanical properties of repaired ACL through
augmentation devices [42]
[43]
[44]
[45].
Biery et al. reported the 2-year follow-up results showed that DIS patients returned
to work and exercise earlier than the traditional ACLR, while there was no
difference in treatment cost, revision rate and clinical outcomes [46]. Eggli et al. followed up DIS patients for 5 years
and found that the postoperative success rate could reach 80% [42]. The results of the above historical literature
were consistent with this study. Through systematic analysis, we found that patients
in arthroscopic ACL repair group, including DIS and BEAR group, not only had no
significant difference in clinical score, physical examination, reoperation rate and
postoperative muscle strength recovery from ACLR group, but also achieved better
results in IKDC score during short-term follow-up. By subgroup analysis, it was not
difficult to find that compared with open ACL repair, arthroscopic ACL repair did
show greater advantages: higher postoperative activity function level and less
reoperation rate.
Another great advantage of arthroscopic ACL repair is that it can resume activity
earlier. We noticed that the rehabilitation program in the early decades included
long-leg cast fixation for at least 5–6 weeks, and then partial weight
bearing was allowed after 8 weeks. As we all know, long-time knee joint fixation is
one of the main causes of knee joint pain, decreased mobility and functional loss
[47]. However, the concept of early rapid
rehabilitation gradually emerged around 1990 [48].
Therefore, the lagging postoperative rehabilitation program was also one of the
important factors for the poor activity function outcome in patients after open ACL
repair. Genelin et al. had confirmed that the use of continuous passive movement
machines in the early postoperative period of patients with primary ACL repair
combined with braces providing limited knee joint movement could further improve the
postoperative effect [49]. The rehabilitation programs
of the arthroscopic ACL repair studies we included basically did not use the
continuous fixation after operation, which was used by open ACL repair, instead of
the early partial or completely unlimited joint movement and the advance of
weight-bearing time. The time for returning to sports reported by several included
studies was basically about 5–6 months. This undoubtedly played an important
role in the recovery of activity function after operation. Unfortunately, we had not
found a study to evaluate the difference in activity recovery speed between ACL
repair and ACLR group, which needs further study.
Limitations
There are some limitations in this study. No RCT study had discussed and analyzed the
location of ACL injury. Considering that the ACL repair effect of patients with
proximal ACL rupture is better, an in-depth study is necessary for patient selection
and refinement of surgical indications. Secondly, because only RCT research is
included, the number of related literatures, the total number of patients, and the
included analysis indicators were few, which might have some impacts on the combined
results. Besides, Except for a 30-year follow-up literature, the follow-up time of
other literatures was short, which might not better judge the long-term effect of
ACL repair patients. In addition, all included studies did not report the follow-up
results of postoperative complications and clinical symptoms of patients. And there
was no clear comparative evaluation on the recovery speed of early activity.
Conclusion
This systematic review using meta-analysis found that at short-term follow-up, the
postoperative clinical efficacy of arthroscopic ACL repair was comparable to ACLR,
but the prognosis of open ACL repair was relatively unsatisfactory. Therefore, we
can make the conclusion that the arthroscopic ACL repair technique is an optional
and promising surgical method to treat ACL injury. Of course, we still need more
prospective controlled studies with long follow-up time to confirm our
conclusions.
