Knee Surgery, Sports Traumatology, Arthroscopy

https://doi.org/10.1007/s00167-020-06334-5

KNEE

Autograft type affects muscle strength and hop performance after ACL
reconstruction. A randomised controlled trial comparing patellar
tendon and hamstring tendon autografts with standard or accelerated
rehabilitation
Riccardo Cristiani1,2 · Christina Mikkelsen1,2 · Peter Wange1,3 · Daniel Olsson4 · Anders Stålman1,2 ·
Björn Engström1,2

Received: 24 June 2020 / Accepted: 13 October 2020

© The Author(s) 2020

Abstract
Purpose To evaluate and compare changes in quadriceps and hamstring strength and single-leg-hop (SLH) test performance
over the first 24 postoperative months in patients who underwent anterior cruciate ligament reconstruction (ACLR) with
bone-patellar tendon-bone (BPTB) or hamstring tendon (HT) autografts and followed either a standard or an accelerated
rehabilitation protocol.
Methods A total of 160 patients undergoing ACLR were randomised in four groups depending on the graft that was used
and the rehabilitation protocol (40 BPTB/standard rehab, 40 BPTB/accelerated rehab, 40 HT/standard rehab, 40 HT/acceler-
ated rehab). Isokinetic concentric quadriceps and hamstring strength at 90°/s and the SLH test performance were assessed
preoperatively and 4,6,8,12 and 24 months postoperatively. The results were reported as the limb symmetry index (LSI) at
the same time point. Linear mixed models were used to compare the groups at the different time points.
Results An average quadriceps strength LSI of 78.4% was found preoperatively. After ACLR, the LSI first decreased at
4 months and then increased from 6 to 24 months, reaching an overall value of 92.7% at the latest follow-up. The BPTB
group showed a significantly decreased LSI at 4, 6, 8 and 12 months compared with the HT group. No significant differences
between the graft groups were found at 24 months.
An average hamstring strength LSI of 84.6% was found preoperatively. After ACLR, the LSI increased from 4 to 24 months
in the BTPB group. In the HT group, the LSI first decreased at 4 months and then increased from 6 to 24 months. An LSI
of 97.1% and 89.1% was found at the latest follow-up for the BPTB and the HT group, respectively. The HT group showed
a significantly decreased LSI at all follow-ups compared with the BPTB group.
An average SLH test LSI of 81% was found preoperatively. After ACLR, the LSI increased from 4 to 24 months, reaching
97.6% overall at the latest follow-up. The BPTB group showed a significantly decreased LSI only at 4 months postopera-
tively compared with the HT group. No significant differences in any of the three tests were found between the standard and
accelerated rehabilitation groups for either of the graft groups at any time point.
Conclusion Muscle strength and SLH test performance recovered progressively after ACLR overall, but they did not all fully
recover, as the injured leg performed on average less than 100% compared with the uninjured leg even 24 months postop-
eratively. After ACLR, inferior quadriceps strength and a poorer SLH test performance were found at 4, 6, 8 and 12 months
and at 4 months, respectively, for the BTPB group compared with the HT group. Persistent, inferior hamstring strength was
found at all postoperative follow-ups in the HT group. Rehabilitation, standard or accelerated, had no significant impact on
the recovery of muscle strength and SLH test performance after ACLR in any of the graft groups.
Level of Evidence Level I.

IRB Ethical approval for this study was obtained from the
regional ethics committee, Karolinska Institutet, Diarienumber
2001-044.

Extended author information available on the last page of the article

13
Vol.:(0123456789)
 Knee Surgery, Sports Traumatology, Arthroscopy

Keywords Anterior cruciate ligament · ACL · Hamstring · Patellar tendon · Limb symmetry index · Quadriceps strength ·
Hop test · Hamstring strength · Muscle strength · Graft

Introduction It was hypothesised that muscle strength and SLH test

performance would recover progressively after ACLR, but
The recovery of symmetrical quadriceps and hamstring that symmetry in comparison with the uninjured leg would
strength is regarded as a key factor after anterior cruci- not be achieved, even two years after surgery. Another
ate ligament (ACL) reconstruction (ACLR) [11, 29, 31]. hypothesis was that the recovery of quadriceps strength and
Muscular asymmetries affect sports performance and are SLH test performance would be delayed in patients who
associated with a higher risk of ACL graft rupture and underwent ACLR with a BPTB autograft and, on the other
knee re-injuries after return to sport [11, 15]. The single- hand, the recovery of hamstring strength would be delayed
leg-hop (SLH) test is commonly used to assess the com- in patients who underwent ACLR with an HT autograft.
bination of muscle strength, neuromuscular control and Finally, it was also hypothesised that an accelerated reha-
confidence in the limb after ACLR [17]. bilitation protocol would result in the earlier recovery of
The most commonly used autografts for ACLR are the muscle strength and SLH test performance in comparison
hamstring tendons (HT) and the bone-patellar tendon-bone with a standard rehabilitation protocol in both graft groups.
(BPTB). However, questions remain about how patients
with either an HT or a BPTB autograft recover knee mus-
cle strength postoperatively. Contrasting results have
been reported in randomised studies comparing the two Material and methods
autografts at postoperative follow-ups ranging from 3 to
24 months after ACLR [2, 5, 18]. Aune et al. [5] reported Ethical approval for this study was obtained from the
that patients with an HT autograft have better isokinetic regional ethics committee (Karolinska Institutet, Diarie-
quadriceps strength than patients with a BPTB autograft number 2001–044) and all the included patients gave their
after 6 months but not after 12 and 24 months. At the same informed consent to participate.
time, the authors reported significantly inferior isokinetic Between December 2002 and June 2005, a total of 160
hamstring strength in the HT group in comparison with patients were included in a single-centre, randomised, con-
the BPTB group at 12 months. Maletis et al. [18] evalu- trolled trial. The randomisation was performed with num-
ated 99 patients who underwent ACLR with either an HT bered, opaque, sealed envelopes. The patients were ran-
or a BPTB autograft at 3,6, 12 and 24 months after sur- domised in 4 groups depending on the graft used and the
gery. They reported better isokinetic quadriceps strength rehabilitation followed: BPTB graft/standard rehabilitation
but poorer isokinetic hamstring strength for the HT group (n = 40), BPTB graft/accelerated rehabilitation (n = 40),
at 6, 12- and 24-month follow-ups. On the other hand, HT graft/standard rehabilitation (n = 40), HT graft/accel-
Aglietti et al. [2] evaluated 120 patients who underwent erated rehabilitation (n = 40). The inclusion criteria were
ACLR with an HT or a BPTB autograft at 4,12- and a unilateral ACL tear with trauma occurring between 1
24-month follow-ups and found no significant differences and 12 months before reconstruction, age between 18 and
between the grafts with respect to isokinetic quadriceps 45 years, no previous ligament surgery on the involved knee,
or hamstring strength at each follow-up. These inconsist- no concomitant ligament injuries and no contralateral knee
ent results do not make it possible to draw any conclusion ligament injuries. Patients were excluded if either an osteo-
about differences in strength recovery after ACLR with an arthritis grade higher than Ahlbäck 1 [3] was found on the
HT or a BPTB autograft. Moreover, to our knowledge, no preoperative weight-bearing X-ray or a cartilage injury of
previous studies have compared the effects on the recovery the severity Outerbridge grade 4 [21, 22] was found at the
of knee muscle strength and SLH test performance of an time of ACLR. Moreover, patients were also excluded in the
accelerated or a standard rehabilitation protocol for both event of medial and/or lateral meniscus repair, as this would
autografts at several follow-ups after ACLR. have changed their rehabilitation.
The purpose of this randomised, controlled trial was Patients underwent an isokinetic quadriceps and ham-
to evaluate and compare changes in quadriceps and ham- string strength and SLH test performance assessment preop-
string strength and SLH test performance, over the first eratively and at 4,6,8,12 and 24 months postoperatively. Par-
two postoperative years, between patients who underwent ticipant randomisation and the number of patients assessed
ACLR with an HT or a BPTB autograft and followed a at each follow-up are reported in Fig. 1. Patient character-
standard or an accelerated rehabilitation protocol. istics for each of the 4 randomised groups are summarised
in Table 1.

13
Knee Surgery, Sports Traumatology, Arthroscopy

Fig. 1  Participant randomisation. ACL anterior cruciate ligament, BPTB bone-patellar tendon-bone, HT hamstring tendon

Table 1  Patient characteristics BPTB standard BPTB accelerated HT standard HT accelerated

Number of patients 40 40 40 40
Age at surgery, mean ± SD 29.3 ± 6.4 28.5 ± 5.5 28.0 ± 6.3 28.8 ± 6.3
Male gender 25 (62.5) 34 (85) 29 (72.5) 27 (67.5)
Injured knee, right 18 (45) 23 (57.5) 21 (52.5) 20 (50)
Time from injury to surgery, 6.6 ± 2.9 6.1 ± 4.2 3.8 ± 2.8 4.6 ± 3.5
mo, mean ± SD
Medial meniscus resection 4 (10) 3 (7.5) 5 (12.5) 7 (17.5)
Lateral meniscus resection 9 (22.5) 9 (22.5) 10 (25) 12 (30)
Cartilage injury 13 (32.5) 9 (22.5) 15 (37.5) 10 (25)

NOTE Data are reported as n (%), unless otherwise indicated

BPTB bone-patellar tendon-bone, HT hamstring tendon, SD standard deviation

Lost to follow‑up Surgical technique

Over the 2 postoperative years, 6 patients (3.7%) sustained All the procedures were performed by four surgeons who
an ACL graft rupture (BPTB/standard rehab = 1; BPTB/ were all familiar with both techniques and performed
accelerated rehab = 3; HT/accelerated rehab = 2) and 4 ACLRs on a daily basis. A routine diagnostic arthroscopy
patients (2.5%) sustained a contralateral ACL rupture (1 and eventual meniscal surgery were performed first, fol-
in each group). A total of 5 patients underwent a surgi- lowed by the ACLR. For the patients randomised to the
cal operation on the contralateral knee or lower limb, 3 BPTB group, the central third of the patellar tendon with two
patients underwent arthroscopy on the ipsi-lateral knee, bone blocks was harvested through a longitudinal incision.
2 patients became pregnant and 1 patient sustained a foot The BPTB autograft was between 9 and 10 mm in diameter.
fracture. These patients were not included in the analysis The bone defects of the patella and the proximal tibia were
after the occurrence of the event. In addition, a variable not bone grafted and the defect in the patellar tendon was
number of patients at each follow-up were unwilling to left open, but the paratendon was sutured with a No. 2–0
participate. The number of patients available at each fol- Vicryl absorbable suture. For the patients randomised to the
low-up in each group is reported in Fig. 1. HT group, the graft was harvested through a longitudinal

13
 Knee Surgery, Sports Traumatology, Arthroscopy

Table 2  Milestones in Standard Accelerated

rehabilitation
Perform isokinetic strength test and single-leg-hop test before ACLR
Postoperatively

1-3 weeks

Perform passive and active full ROM and full weight-

bearing, closed kinetic chain exercises with limited (0-50
degrees) ROM, straight-leg raises (extension, flexion,
abduction, adduction), heel rises, balance/proprioceptive
training, stationary bicycle when 110 degrees of knee
flexion achieved

1-6 weeks

Perform passive and active full ROM and full weight-

bearing, closed kinetic chain exercises with limited (0-50
degrees) ROM, straight-leg raises (extension, flexion,
abduction, adduction), balance/proprioceptive training,
stationary bicycle when 110 degrees of knee flexion
achieved
4-6 weeks

Closed and open kinetic chain exercises with ROM

between 0 and 90 degrees, training of the upper body and
trunk

7-8 weeks

Increase closed and open kinetic chain exercises,

increase balance/proprioceptive training, frontal and
lateral step-ups, lunges with weight, slide board, stair-
master, cycling

7-12 weeks

Increase closed kinetic chain exercises, increase

balance/proprioceptive training, frontal and lateral step-
ups, lunges with weight, slide board, stair-master, heel
rises

9-12 weeks

Start running and plyometric exercises, sub-maximum

closed and open kinetic chain exercises

3-4 months 3-4 months

Increase closed kinetic chain exercises and start open Increase running programme and plyometric exercises,
kinetic chain exercises with full ROM, increase maximum closed and open kinetic chain exercises,
balance/proprioceptive training, start running and cuttings and perturbation training, sports-specific drills
plyometric exercises, cycling, start perturbation training
and sports-specific drills

5-6 months

Increase running and plyometric programme, cuttings,

perturbation training and sports-specific drills

Overall goals of rehabilitation

Full ROM
No pain or effusion during physical activity
Regain balance, proprioception, co-ordination and neuromuscular control
LSI ≥ 90% for isokinetic quadriceps and hamstring strength tests and single-leg-hop test
________________________________________________________________________________________________

ACLR anterior cruciate ligament reconstruction, LSI limb symmetry index, ROM range of
motion
incision over the pes anserinus. The sartorius fascia was and the gracilis tendons were harvested using a semicircu-
incised parallel to its fibres and both the semitendinosus lar tendon stripper. The tendons were cleaned and looped
over a No. 2 Vicryl absorbable suture to create a double

13
Knee Surgery, Sports Traumatology, Arthroscopy

graft, which was pretensioned up to 20 lb. The HT autograft with a tape measure. The patients were initially given a ver-
was between 7 and 9 mm in diameter. The femoral tunnel bal description of the test and they were allowed to perform
was drilled using a transtibial technique in all cases. One practical trials until they felt confident about the test. Three
7 × 20 mm and one 9 × 20 mm metal interference screws trials were performed for each leg, always starting with the
were used to fix the BPTB autograft on the femoral and contralateral uninjured leg. The best trial for each leg was
tibial sides respectively. The HT autograft was fixed using registered.
a Rigidfix Cross Pin device (DePuy Mitek, Raynham, MA) The limb symmetry indexes (LSIs) of the peak quadriceps
on the femoral side and an Intrafix device (DePuy Mitek, and hamstring torque and SLH test were calculated as involved
Raynham, MA) on the tibial side. Both grafts were fixed at limb/uninvolved limb × 100 [9, 29].
a knee flexion angle of approximately 20 degrees.
Statistical analysis
Rehabilitation
A linear mixed effect model was estimated, fitting a separate
The patients followed a standardised or accelerated reha- model for each of the three studied outcomes using the lmer
bilitation programme, depending on the randomisation. All function in the lme4 R-package. The model included factors
the patients underwent supervised rehabilitation 2–3 days a for graft, for rehabilitation and for time and all interactions.
week. The patients randomised to the standardised rehabili- The models were additionally adjusted for gender, meniscus
tation programme were supervised at Sportskadekliniken, resection (yes/no) and cartilage lesion (yes/no). The results
Stockholm, Sweden, whereas the patients randomised to are presented graphically as estimated marginal means
the accelerated rehabilitation programme were supervised with the Kenward Roger method for degrees of freedom.
at Capio Artro Clinic, Stockholm, Sweden. The rehabilita- Contrasts are also presented, first comparing the two graft
tion programme was carefully followed with frequent checks groups and then, within each of the graft groups, comparing
on the milestones (Table 2). the two rehabilitation groups. A difference in increase of 30
kilopond-metre was considered to be clinically significant.
Isokinetic strength and single‑leg‑hop test To achieve a statistical power of 85% and an alpha of 5%, a
performance assessment sample size of 19 patients in each study group was required.
The results are presented with 95% confidence intervals
The patients underwent an isokinetic strength and SLH test and p-values. The significance level in all analyses was 5%
performance assessment using a standardised protocol pre- (two-tailed).
operatively and at 4,6,8,12 and 24 months postoperatively.
All the tests were performed at our outpatient clinic and all
the patients were assessed by a single physiotherapist (C.M.). Results
Isokinetic concentric quadriceps and hamstring strength were
measured bilaterally at 90°/s using the Biodex System 3 (Bio- Quadriceps strength
dex Medical Systems, Shirley, New York, USA). The test was
performed in a range of motion between 90° and 10° of knee An average quadriceps strength LSI of 78.4% was found pre-
flexion, always starting with the contralateral uninjured knee. operatively. After ACLR, the LSI first decreased at 4 months
Prior to the test, the patients warmed up using a stationary and then increased from 6 to 24 months, reaching 92.7%
cycling ergometer at low resistance for 10 min. The patients overall at the latest follow-up. The BPTB group showed a
were given a verbal description of the test and 2–3 practical significantly decreased LSI at 4,6,8 and 12 months compared
trials were allowed before testing. Each patient performed 5 with the HT group. No significant differences between the
maximum quadriceps and hamstring contractions with each graft groups were found at 24 months. No significant dif-
leg. The patients were encouraged verbally during the test. The ferences were found between the standard and accelerated
peak quadriceps and hamstring torque values (highest achieved rehabilitation groups for either of the graft groups at any of
values) were registered. the time points (Fig. 2a–c).
The SLH test [24] was performed with the patient standing
on one leg and being instructed to jump straight ahead as far
Hamstring strength
as possible and land on the same leg. The test was considered
successful if the landing was stable. If the patient landed with
An average hamstring strength LSI of 84.6% was found
an early touchdown of the contralateral limb, experienced loss
preoperatively. After ACLR, the LSI increased from 4 to
of balance or took additional hops after landing, the hop was
24 months in the BPTB group. In the HT group, the LSI
repeated. The hop distance was measured from the starting
first decreased at 4 months and then increased from 6 to
line (toe touching the line), to the heel on landing and recorded

13
 Knee Surgery, Sports Traumatology, Arthroscopy

a b
110 110
LSI quadriceps strength 90°/sec

LSI quadriceps strength 90°/sec

100 100
<0.001 0.02
90 <0.001 90
<0.001
80 80
70 70
60 60
50 50
40 40
30 30
Pre- 12 24 Pre 4 6 8 12 24
4 mo 6 mo 8 mo
op mo mo -op mo mo mo mo mo
BPTB 78.9 61.6 71.7 77.3 85.6 92.7 BPTB stand 77.9 60 71.4 78.3 85.8 92.6
HT 77.9 76.2 82 87.2 91 92.7 BPTB accel 80.3 65.3 72.2 77.1 85.1 92.9

c
110
LSI quadriceps strength 90°/sec

100
90
80
70
60
50
40
30
Pre- 12 24
4 mo 6 mo 8 mo
op mo mo
HT stand 78.9 75.8 81.6 86.1 90.5 91.2
HT accel 77.5 77.4 82.9 90.3 92.8 93.5

Fig. 2  a–c LSI (mean and 95% CI) for quadriceps strength at 90°/ tion and HT/accelerated rehabilitation (c) groups at each time point.
second from preoperative to 24 months postoperative and compari- BPTB bone-patellar tendon-bone, CI confidence intervals, HT ham-
son between BPTB and HT groups (a) BPTB/standard rehabilitation string tendon, LSI limb symmetry index. *Only significant P values
and BPTB/accelerated rehabilitation (b) and HT/standard rehabilita- are reported

24 months. An LSI of 97.1% and 89.1% was found at the at all follow-ups (4, 6, 8, 12 and 24 months) compared
latest follow-up for the BPTB and the HT group respec- with the BTPB group. No significant differences were
tively. The HT group showed a significantly decreased LSI found between the standard and accelerated rehabilitation

13
Knee Surgery, Sports Traumatology, Arthroscopy

a b
110 110
<0.001 <0.001 <0.001 0.002
LSI hamstring strength 90°/sec

LSI hamstring strength 90°/sec

100 100
<0.001
90 90
80 80
70 70
60 60
50 50
40 40
30 30
Pre- 4 6 8 12 24 Pre- 4 6 8 12 24
op mo mo mo mo mo op mo mo mo mo mo
BPTB 84.8 89.1 96.2 98.6 99.2 97.1 BPTB stand 84.7 87.5 95.5 97.9 97.7 96.3
HT 84.5 79.5 85.7 86.5 87.4 89.1 BPTB accel 85 91.8 98.2 99.5 100.8 97.8

c
110
LSI hamstring strength 90°/sec

100
90
80
70
60
50
40
30
Pre- 12 24
4 mo 6 mo 8 mo
op mo mo
HT stand 82.9 79.3 85.4 84.2 86 87.1
HT accel 86.4 79.6 88.1 86.9 89.5 91.6

Fig. 3  a–c LSI (mean and 95% CI) for hamstring strength at 90°/ tion and HT/accelerated rehabilitation (c) groups at each time point.
second from preoperative to 24 months postoperative and compari- BPTB bone-patellar tendon-bone, CI confidence intervals, HT ham-
son between BPTB and HT groups (a), BPTB/standard rehabilitation string tendon, LSI limb symmetry index. *Only significant P values
and BPTB/accelerated rehabilitation (b) and HT/standard rehabilita- are reported

groups for either of the graft groups at any of the time Single‑leg‑hop test performance
points (Fig. 3a–c).
An average SLH test LSI of 81% was found preopera-
tively. After ACLR, the LSI increased from 4 to 24 months,
reaching 97.6% overall at the latest follow-up. The BPTB
group showed a significantly decreased LSI at 4 months

13
 Knee Surgery, Sports Traumatology, Arthroscopy

a b
110 110
100 100
0.005
LSI single leg hop test

LSI single leg hop test

90 90
80 80
70 70
60 60
50 50
40 40
30 30
Pre- 4 6 8 12 24 Pre- 4 6 8 12 24
op mo mo mo mo mo op mo mo mo mo mo
BPTB 79.1 79.4 88.5 91.3 94.9 97.9 BPTB stand 77.8 78.8 91.7 93.4 94.5 98.4
HT 83 85.5 91.1 95.1 96.6 97.5 BPTB accel 80.4 80.1 87.6 90.6 95.3 97.6

c
110
100
90
LSI single leg hop test

80
70
60
50
40
30
Pre- 12 24
4 mo 6 mo 8 mo
op mo mo
HT stand 82.1 84.1 91.3 94.6 95 95.9
HT accel 83.9 86.8 90.9 95.7 98.1 99

Fig. 4  a–c LSI (mean and 95% CI) for single-leg-hop test from pre- ated rehabilitation (c) groups at each time point. BPTB bone-patellar
operative to 24 months postoperative and comparison between BPTB tendon-bone, CI confidence intervals, HT hamstring tendon, LSI limb
and HT groups (a), BPTB/standard rehabilitation and BPTB/acceler- symmetry index. *Only significant P values are reported
ated rehabilitation (b) and HT/standard rehabilitation and HT/acceler-

postoperatively compared with the HT group. No significant Discussion

differences were found at the other time points. No signifi-
cant differences were found between the standard and accel- The most important finding in the present study was that
erated rehabilitation groups for either of the graft groups at muscle strength and SLH test performance recovered pro-
any of the time points (Fig. 4a–c). gressively after ACLR overall, but they did not all fully
recover, as the injured leg performed on average less than
100% compared with the uninjured leg even 24 months

13
Knee Surgery, Sports Traumatology, Arthroscopy

postoperatively. After ACLR, the BPTB group reported control, confidence in the limb and the ability to tolerate
inferior quadriceps strength and a poorer SLH test perfor- loads related to sport-specific activities [17]. They are also
mance at 4,6,8 and 12 months and at 4 months respectively recommended to evaluate the patient’s readiness to return
compared with the HT group. On the other hand, the HT to sport after ACLR [29]. The SLH test for distance is con-
group reported persistent, inferior hamstring strength at all sistently reported in the literature as a means of quantifying
postoperative follow-ups compared with the BPTB group. knee performance after ACLR [9, 17, 24, 29, 30]. In the
Rehabilitation, standard or accelerated, had no significant present study, we found an average deficit of approximately
impact on the recovery of muscle strength and SLH test 20% in this test in the injured leg in comparison with the
performance after ACLR in any of the graft groups. uninjured leg preoperatively. However, after ACLR, the SLH
Muscular strength recovery is one of the primary goals test performance improved progressively, reaching almost
after ACLR. Muscular asymmetries are associated with a symmetry (LSI of 97.6%) overall at the latest follow-up. The
higher risk of ACL graft rupture and knee re-injuries after BPTB group showed an inferior SLH test performance only
return to sport [11, 15]. In particular, quadriceps weakness at 4 months postoperatively compared with the HT group.
is a major concern after ACL injury and reconstruction. No significant differences were observed between the graft
Quadriceps muscle strength is associated with patient- groups at the later follow-ups. One interesting finding was
reported knee outcome and satisfaction [8], as well as with that, overall, for both graft groups, the SLH test failed to
osteoarthritis development after ACLR [20]. In the present show the same level of deficit than quadriceps strength. Pre-
study, we found an overall significant decrease in quadriceps vious studies have suggested that patients are able to com-
strength LSI before ACLR. Previous studies have reported pensate quadriceps muscle weakness with hip and trunk
a reduction in the maximum quadriceps moment during muscles for hopping [23, 33].
walking [6] and quadriceps atrophy [14] after ACL injury. Interestingly, we found that the type of rehabilitation,
After ACLR, quadriceps strength recovered progressively standard or accelerated, had no significant impact on the
in all patients. However, the pattern of recovery was greatly recovery of muscle strength and SLH test performance
affected by the graft type, with the BPTB group show- after ACLR in any of the graft groups. Some investigators
ing significantly inferior quadriceps strength at 4,6,8 and recommended the use of early open kinetic chain exercises
12 months compared with the HT group. No significant dif- after ACLR, owing to the potential for early postoperative
ferences were found between the graft groups at 24 months increases, especially in quadriceps muscle strength [7, 10].
postoperatively. However, quadriceps strength did not fully However, despite the earlier introduction of open kinetic
recover compared with the uninjured leg overall, even at the chain exercises in the accelerated rehabilitation groups
latest follow-up. A recent study [16] showed that individuals (4 weeks vs. 3 months postoperatively for accelerated and
with ACLR demonstrate differences in neural excitability standard rehabilitation protocols respectively), we have not
and smaller quadriceps muscle volume compared with con- found any significant differences with regard to strength and
trols, even at an average of 6 years after surgery. hop symmetry for either of the graft groups.
Hamstring weakness is another concern after ACLR. Contrasting results have been reported in terms of the
Hamstring muscle strength deficits might be a risk factor for recovery of knee muscle strength after ACLR with BPTB
ACL re-tears, since they act as agonists to the ACL by resist- and HT autografts at postoperative follow-ups ranging from
ing anterior tibial translation [13, 28]. It has been shown that 3 to 24 months after ACLR [2, 5, 18]. Comparisons with
hamstring activation reduces the loads on the ACL [25, 32]. previous studies are difficult to make, because of differences
In our study, an average hamstring strength LSI of 84.6% in time intervals from injury to surgery, surgical procedures
was found preoperatively. After ACLR, the LSI increased and preoperative and postoperative rehabilitation protocols.
progressively from 4 to 24 months in the BPTB group, while Moreover, a serious lack of standardisation in testing pro-
it first decreased at 4 months and then increased from 6 to cedures (different dynamometers, angular velocities, mode
24 months in the HT group. However, the HT group showed of muscle contractions, number of repetitions and range of
a significantly reduced LSI at all follow-ups (4, 6, 8, 12 and motion) is present in previous studies. In their Cochrane
24 months) compared with the BPTB group. In the case review, Mohtadi et al. [19] reported an overall trend toward a
of HT graft ACLR, hamstring weakness has always been loss of quadriceps strength and hamstring strength for BPTB
a matter of concern. Ageberg et al. [1] showed persistent and HT autografts respectively after ACLR. However, the
hamstring strength deficits even 3 to 5 years after ACLR authors highlighted the fact that many trials comparing the
performed with an HT graft. two grafts ran a high risk of bias and had poor methodo-
Single-leg-hop tests are well established and have shown logical quality, stating that there was insufficient evidence to
high reliability after ACL injury and ACLR [12, 29, 30]. draw conclusions about differences between the two grafts
These tests are performance-based measurements used to in terms of functional outcome.
assess the combination of muscle strength, neuromuscular

13
 Knee Surgery, Sports Traumatology, Arthroscopy

The present randomised, controlled trial showed that systematic review, Sharma et al. [27] reported that the addi-
asymmetries in muscle strength and hop performance are tion of gracilis harvest reduced hamstring strength by only
persistent even 24 months after ACLR performed with either 3.85% relative to an isolated semitendinosus harvest. Finally,
of the 2 grafts. For this reason, rehabilitations protocols we did not assess the relationship between changes in mus-
should be implemented and more time needs to be spent on cle strength/hop performance and patient-reported outcome
muscle strength rehabilitation. The choice between BPTB measurements, but this was not an aim of the current study.
and HT grafts strongly affects the pattern of recovery of mus-
cle strength. The use of the BPTB graft was associated with
an inferior quadriceps strength LSI at 4,6,8 and 12 months Conclusion
postoperatively compared with the HT graft. On the other
hand, the use of the HT graft was associated with an inferior Muscle strength and SLH test performance recovered pro-
hamstring strength LSI at all postoperative follow-ups in gressively after ACLR overall, but they did not all fully
comparison with the BPTB graft. The implications relat- recover, as the injured leg performed on average less than
ing to the impact of graft choice on knee muscle strength 100% compared with the uninjured leg even 24 months
are important to consider. The graft for ACLR should also postoperatively. After ACLR, inferior quadriceps strength
be chosen based on potential strength deficits that it would and poorer SLH test performance were found at 4,6,8 and
be good to avoid within the sport/activity practised by the 12 months and at 4 months respectively for the BTPB group
patient. Moreover, rehabilitation should be customised, tak- compared with the HT group. Persistent, inferior hamstring
ing the type of graft used into account, as each type of graft strength was found at all postoperative follow-ups in the HT
generates intrinsic muscular deficits. In particular, patients group. Rehabilitation, standard or accelerated, had no sig-
undergoing ACLR with the HT graft require more attention nificant impact on the recovery of muscle strength and SLH
to the recovery of hamstring strength, as indicated by the test performance after ACLR in either of the graft groups.
persistent strength asymmetries at all postoperative follow-
ups. Deficits in hamstring strength might contribute to the
higher ACL re-rupture rate observed with the HT graft com- Funding Open access funding provided by Karolinska Institute.
pared with the BPTB graft [26].
The principal strength of this study was the randomised Compliance with ethical standards
design. The surgical procedure, rehabilitation and the
Conflict of interest Each author declares that no possible conflict of
assessment of muscle strength and hop test performance interest (financial or not financial) exists in connection with this study.
were standardised. Finally, the impact of 2 different types
of rehabilitation (standard or accelerated) on the recovery Ethical approval This study was approved from the Regional Ethics
of muscle strength and hop performance after ACLR with Committee, Karolinska Institutet, Diarienumber 2001–044.
either a BPTB or an HT graft was assessed.
Several limitations are present. The use of LSIs to evalu- Open Access This article is licensed under a Creative Commons Attri-
ate muscle strength and hop performance after ACL injury bution 4.0 International License, which permits use, sharing, adapta-
tion, distribution and reproduction in any medium or format, as long
and reconstruction is often discussed as a possible limita- as you give appropriate credit to the original author(s) and the source,
tion. The use of the uninjured leg as a “control” may not provide a link to the Creative Commons licence, and indicate if changes
be wholly appropriate. Patients may reduce their physical were made. The images or other third party material in this article are
activity after ACL injury and during the first months after included in the article’s Creative Commons licence, unless indicated
otherwise in a credit line to the material. If material is not included in
ACLR and this can lead to a loss of muscle strength in their the article’s Creative Commons licence and your intended use is not
uninjured leg as well. On the other hand, it is also possible permitted by statutory regulation or exceeds the permitted use, you will
that patients may gain strength in their uninjured leg during need to obtain permission directly from the copyright holder. To view a
the rehabilitation period, making the operated leg appear copy of this licence, visit http://creat​iveco​mmons​.org/licen​ses/by/4.0/.
relatively weaker. However, LSIs are still the most used and
validated outcome for measuring muscle strength and hop
performance after ACL injury and ACLR [1, 8, 9, 12, 15, References
17, 24, 29, 30]. All the HT ACLRs in this study were per-
1. Ageberg E, Roos HP, Silbernagel KG, Thomeé R, Roos EM
formed by harvesting both the semitendinosus and gracilis (2009) Knee extension and flexion muscle power after anterior
tendons. This may have affected the results. The use of both cruciate ligament reconstruction with patellar tendon graft or
HTs might result in a greater weakness in hamstring strength hamstring tendons graft: a cross-sectional comparison 3 years post
compared with the use of a triple or quadruple semitendino- surgery. Knee Surg Sports Traumatol Arthrosc 17(2):162–169
2. Aglietti P, Giron F, Buzzi R, Biddau F, Sasso F (2004) Ante-
sus graft alone. The gracilis may potentially help compen- rior cruciate ligament reconstruction: bone-patellar tendon-bone
sate for the loss of the semitendinosus [4]. However, in their

13
Knee Surgery, Sports Traumatology, Arthroscopy

compared with double semitendinosus and gracilis tendon grafts. cruciate ligament reconstruction: the Delaware-Oslo ACL cohort
A prospective, randomized clinical trial. J Bone Joint Surg Am study. Am J Sports Med 40(10):2348–2356
86(10):2143–2155 18. Maletis GB, Cameron SL, Tengan JJ, Burchette RJ (2007) A pro-
3. Ahlbäck S (1968) Osteoarthrosis of the knee. A radiographic spective randomized study of anterior cruciate ligament recon-
investigation. Acta Radiol Diagn (Stockh) 277:7–72 struction. Am J Sports Med 35(3):384–394
4. Ardern CL, Webster KE (2009) Knee flexor strength recovery 19. Mohtadi NG, Chan DS, Dainty KN, Whelan DB (2011) Patellar
following hamstring tendon harvest for anterior cruciate ligament tendon versus hamstring tendon autograft for anterior cruciate
reconstruction: A systematic review. Orthop Rev (Pavia) 1(2):e12 ligament rupture in adults. Cochrane Database Syst Rev. https​://
5. Aune AK, Holm I, Risberg MA, Jensen HK, Steen H (2001) doi.org/10.1002/14651​858.CD005​960.pub2
Four-strand hamstring tendon autograft compared with patellar 20. Øiestad BE, Juhl CB, Eitzen I, Thorlund JB (2015) Knee extensor
tendon-bone autograft for anterior cruciate ligament reconstruc- muscle weakness is a risk factor for development of knee osteoar-
tion. Am J Sports Med 29(6):722–728 thritis. A systematic review and meta-analysis. Osteoarthr Cartil
6. Berchuck M, Andriacchi TP, Bach BR, Reider B (1990) Gait 23(2):171–177
adaptations by patients who have a deficient anterior cruciate 21. Outerbridge RE (1961) The etiology of chondromalacia patellae.
ligament. J Bone Joint Surg Am 72(6):871–877 J Bone Joint Surg Br 43-B:752–757
7. Beynnon BD, Johnson RJ, Naud S, Fleming BC, Abate JA, 22. Outerbridge RE, Dunlop JA (1975) The problem of chondroma-
Brattbakk B, Nichols CE (2011) Accelerated versus nonac- lacia patellae. Clin Orthop Relat Res 110:177–196
celerated rehabilitation after anterior cruciate ligament recon- 23. Powers CM (2010) The influence of abnormal hip mechanics on
struction: a prospective, randomized, double-blind investigation knee injury: a biomechanical perspective. J Orthop Sports Phys
evaluating knee joint laxity using roentgen stereophotogram- Ther 40(2):42–51
metric analysis. Am J Sports Med 39(12):2536–2548 24. Reid A, Birmingham TB, Stratford PW, Alcock GK, Giffin JR
8. Cristiani R, Mikkelsen C, Edman G, Forssblad M, Engström (2007) Hop testing provides a reliable and valid outcome measure
B, Stålman A (2020) Age, gender, quadriceps strength and during rehabilitation after anterior cruciate ligament reconstruc-
hop test performance are the most important factors affecting tion. Phys Ther 87(3):337–349
the achievement of a patient-acceptable symptom state after 25. Renström P, Arms SW, Stanwyck TS, Johnson RJ, Pope MH
ACL reconstruction. Knee Surg Sports Traumatol Arthrosc (1986) Strain within the anterior cruciate ligament during ham-
28(2):369–380 string and quadriceps activity. Am J Sports Med 14(1):83–87
9. Cristiani R, Mikkelsen C, Forssblad M, Engström B, Stål- 26. Samuelsen BT, Webster KE, Johnson NR, Hewett TE, Krych AJ
man A (2019) Only one patient out of five achieves symmetri- (2017) Hamstring autograft versus patellar tendon autograft for
cal knee function 6 months after primary anterior cruciate ACL reconstruction: is there a difference in graft failure rate?
ligament reconstruction. Knee Surg Sports Traumatol Arthrosc A meta-analysis of 47,613 patients. Clin Orthop Relat Res
27(11):3461–3470 475(10):2459–2468
10. Fukuda TY, Fingerhut D, Moreira VC, Camarini PM, Scodel- 27. Sharma A, Flanigan DC, Randall K, Magnussen RA (2016) Does
ler NF, Duarte A Jr, Martinelli M, Bryk FF (2013) Open kinetic gracilis preservation matter in anterior cruciate ligament recon-
chain exercises in a restricted range of motion after anterior cruci- struction? A systematic review. Arthroscopy 32(6):1165–1173
ate ligament reconstruction: a randomized controlled trial. Am J 28. Thomas AC, Villwock M, Wojtys EM, Palmieri-Smith RM (2013)
Sports Med 41(4):788–794 Lower extremity muscle strength after anterior cruciate ligament
11. Grindem H, Snyder-Mackler L, Moksnes H, Engebretsen L, Ris- injury and reconstruction. J Athl Train 48(5):610–620
berg MA (2016) Simple decision rules can reduce reinjury risk 29. Thomeé R, Kaplan Y, Kvist J et al (2011) Muscle strength and hop
by 84% after ACL reconstruction: the Delaware-Oslo ACL cohort performance criteria prior to return to sport after ACL reconstruc-
study. Br J Sports Med 50(13):804–808 tion. Knee Surg Sports Traumatol Arthrosc 19(11):1798–1805
12. Gustavsson A, Neeter C, Thomeé P, Silbernagel KG, Augustsson 30. Thomeé R, Neeter C, Gustavsson A, Thomeé P, Augustsson J,
J, Thomeé R, Karlsson J (2006) A test battery for evaluating hop Eriksson B, Karlsson J (2012) Variability in leg muscle power and
performance in patients with an ACL injury and patients who hop performance after anterior cruciate ligament reconstruction.
have undergone ACL reconstruction. Knee Surg Sports Traumatol Knee Surg Sports Traumatol Arthrosc 20(6):1143–1151
Arthrosc 14(8):778–788 31. Undheim MB, Cosgrave C, King E, Strike S, Marshall B, Falvey
13. Hiemstra LA, Webber S, MacDonald PB, Kriellaars, (2007) Con- E, Franklyn-Miller A (2015) Isokinetic muscle strength and
tralateral limb strength deficits after anterior cruciate ligament readiness to return to sport following anterior cruciate ligament
reconstruction using a hamstring tendon graft. Clin Biomech reconstruction: is there an association? A systematic review and
(Bristol Avon) 22(5):543–550 a protocol recommendation. Br J Sports Med 49(20):1305–1310
14. Kannus P (1988) Ratio of hamstring to quadriceps femoris mus- 32. Withrow TJ, Huston LJ, Wojtys EM, Ashton-Miller JA (2008)
cles’ strength in the anterior cruciate ligament insufficient knee Effect of varying hamstring tension on anterior cruciate ligament
Relationship to long-term recovery. Phys Ther 68(6):961–965 strain during in vitro impulsive knee flexion and compression
15. Kyritsis P, Bahr R, Landreau P, Miladi R, Witvrouw E (2016) loading. J Bone Joint Surg Am 90(4):815–823
Likelihood of ACL graft rupture: not meeting six clinical dis- 33. Zebis MK, Andersen LL, Brandt M et al (2016) Effects of evi-
charge criteria before return to sport is associated with a four dence-based prevention training on neuromuscular and biome-
times greater risk of rupture. Br J Sports Med 50(15):946–951 chanical risk factors for ACL injury in adolescent female athletes:
16. Lepley AS, Grooms DR, Burland JP, Davi SM, Kinsella-Shaw a randomised controlled trial. Br J Sports Med 50(9):552–557
JM, Lepley LK (2019) Quadriceps muscle function following
anterior cruciate ligament reconstruction: systemic differences Publisher’s Note Springer Nature remains neutral with regard to
in neural and morphological characteristics. Exp Brain Res jurisdictional claims in published maps and institutional affiliations.
237(5):1267–1278
17. Logerstedt D, Grindem H, Lynch A, Eitzen I, Engebretsen L, Ris-
berg MA, Axe MJ, Snyder-Mackler L (2012) Single-legged hop
tests as predictors of self-reported knee function after anterior

13
 Knee Surgery, Sports Traumatology, Arthroscopy

Affiliations

Riccardo Cristiani1,2 · Christina Mikkelsen1,2 · Peter Wange1,3 · Daniel Olsson4 · Anders Stålman1,2 ·
Björn Engström1,2

3
* Riccardo Cristiani Aleris Sports Medicine and Orthopedics Sabbatsberg,
[email protected] Crafoords väg 6, 11382 Stockholm, Sweden
4
1 Unit of Medical Statistics, Department of Learning,
Department of Molecular Medicine and Surgery, Stockholm
Informatics, Management and Ethics (LIME), Karolinska
Sports Trauma Research Center, Karolinska Institutet,
Institutet, Stockholm, Sweden
Stockholm, Sweden
2
Capio Artro Clinic, FIFA Medical Centre of Excellence,
Sophiahemmet Hospital, Valhallavägen 91,
11486 Stockholm, Sweden

13