ORIGINAL ARTICLE

Prospective clinical comparisons of semitendinosusversus semitendinosus and gracilis tendon autografts for anatomicdouble-bundle anterior cruciate ligament reconstruction

Yusuke Inagaki • Eiji Kondo • Nobuto Kitamura •

Jun Onodera • Tomonori Yagi • Yasuhito Tanaka •

Kazunori Yasuda

Received: 21 February 2013 / Accepted: 1 June 2013

� The Japanese Orthopaedic Association 2013

Abstract

Background The data available from the previously

reported clinical studies remains insufficient concerning

the hamstring graft preparation in double-bundle anterior

cruciate ligament (ACL) reconstruction.

Objective To test the hypothesis that there are no signifi-

cant differences between the semitendinosus tendon alone

and the semitendinosus and gracilis tendon graft fashioning

techniques concerning knee stability and clinical outcome

after anatomic double-bundle ACL reconstruction.

Methods A prospective study was performed on 120

patients who underwent anatomic double-bundle ACL

reconstruction according to the graft fashioning technique.

The authors developed the protocol to use hamstring ten-

don autografts. When the harvested doubled semitendino-

sus tendon is thicker than 6 mm, each half of the

semitendinosus tendon is doubled and used for the anter-

omedial (AM) and posterolateral (PL) bundle grafts (Group

I). On the other hand, when the harvested semitendinosus

tendon is under 6 mm in thickness, the gracilis tendon is

harvested additionally. The distal half of the semitendino-

sus and gracilis tendons are doubled and used for the AM

bundle graft, and the remaining proximal half of the sem-

itendinosus tendon is doubled and used for the PL bundle

grafts (Group II). Sixty-one patients were included in

Group I, and 59 patients in Group II. The two groups were

compared concerning knee stability and clinical outcome

2 years after surgery.

Results The postoperative side-to-side anterior laxity

averaged 1.3 mm in both groups, showing no statistical

difference. There were also no significant differences

between the two groups concerning the peak isokinetic

torque of the quadriceps and the hamstrings, the Lysholm

knee score, and the International Knee Documentation

Committee evaluation.

Conclusion There were no significant differences

between the two graft fashioning techniques after anatomic

double-bundle ACL reconstruction concerning knee sta-

bility and postoperative outcome. The present study pro-

vided orthopedic surgeons with important information on

double-bundle ACL reconstruction with hamstring tendons.

Level of evidence Level II; prospective comparative

study.

Introduction

The normal anterior cruciate ligament (ACL) consists of the

anteromedial (AM) and posterolateral (PL) bundles, which

have different functions [1, 2]. The idea of reconstructing

both bundles was proposed in the 1980s [3]. In the early

2000s, since Yasuda et al. [4] reported a new concept of

anatomic reconstruction of the AM and PL bundles of the

ACL with 2-year clinical results superior to conventional

This paper was presented at the 8th Biennial International Society

of Arthroscopy, Knee Surgery and Orthopaedic Sports Medicine

(ISAKOS) Congress, 15–19 May 2011, Rio de Janeiro, Brazil.

Y. Inagaki � E. Kondo (&) � N. Kitamura � J. Onodera �K. Yasuda

Department of Sports Medicine and Joint Surgery, Hokkaido

University Graduate School of Medicine, Kita-15, Nishi-7,

Kita-ku, Sapporo, Hokkaido 060-8638, Japan

e-mail: eijik@med.hokudai.ac.jp

Y. Inagaki � Y. Tanaka

Department of Orthopaedic Surgery, Nara Medical University,

Kashihara, Nara, Japan

T. Yagi

Department of Orthopedic Surgery, Yamanote-dori Yagi

Hospital, Sapporo, Hokkaido, Japan

123

J Orthop Sci

DOI 10.1007/s00776-013-0427-9

single-bundle ACL reconstruction, a number of anatomic,

biomechanical, and clinical studies on the anatomic double-

bundle reconstruction procedures have been conducted in

the field of ACL reconstruction, and several clinical trials

have found that postoperative knee stability is superior in

the anatomic double-bundle reconstruction compared with

conventional single-bundle reconstruction [5–13].

The essence of ACL reconstruction is grafting tendon

bundles across the knee joint. Therefore, graft selection,

preparation, and fixation are critical factors to ensure ACL

reconstruction leads to clinical success [14]. However, it

has been well established that the weak points of the

hamstring tendon graft fixed with sutures to bone are (1)

low stiffness of the graft-suture-bone complex, (2) rapid

relaxation of the graft tension after surgery, and (3) diffi-

culty in tension control during graft fixation [15, 16].

Therefore, the authors have developed hamstring tendon

‘hybrid’ autografts which consist of hamstring tendon

connected in series with commercially available polyester

tape (Leeds-Keio Artificial Ligament, Neo Ligament,

Leeds, England, United Kingdom) [10, 15, 16]. The hybrid

graft was used to address the above described weak points

based on biomechanical studies [17–19]. In anatomic

double-bundle ACL reconstruction, the authors developed

the protocol described below to fully utilize the limited

amount of hamstring tendon. Concerning the graft prepa-

ration in double-bundle ACL reconstruction, a few clinical

comparisons between different fashioning techniques have

been reported [20, 21]. On the basis of clinical results [8, 9,

13], the authors hypothesized that there are no significant

differences between the semitendinosus tendon alone and

semitendinosus and gracilis tendon graft fashioning meth-

ods concerning knee stability and clinical outcome after

anatomic double-bundle ACL reconstruction. The purpose

of this study was to test this hypothesis.

Materials and methods

Study design

A prospective, comparative, cohort study was conducted in

143 patients who underwent anatomic double-bundle ACL

reconstruction using hamstring tendon autografts in one of

the author’s affiliate hospitals between 2004 and 2006.

Based on the graft fashioning protocol described below, the

authors performed anatomic double-bundle reconstruction

using semitendinosus tendon autograft or semitendinosus

and gracilis tendon autografts for all patients. Each patient

showed an ACL deficiency in the unilateral knee at the time

of surgery. The diagnosis of injured ligaments was made

based on a detailed history of the knee injury, physical

examinations on pathologic status and abnormal laxity,

routinely performed computed digital radiographs and MRI

scans, and the findings at surgery. Patients with a combined

ligament injury in the posterior cruciate ligament, the lateral

collateral ligament, the PL corner structures of the knee, and

medial collateral ligament (grade 3) were excluded from

this study. In addition, patients with any previous operations

for ligament injuries, a concurrent fracture, or severe

osteoarthritis were excluded. The time from onset of injury

to surgery was 1 month or more. This clinical study design

had been accepted by the institutional review board clear-

ance in this hospital before commencement, based on the

described study design and informed consent.

In 2004, 2005 and 2006, 55, 42, and 46 patients, respec-

tively, were enrolled in this study. Patients were informed

that they were going to be in a study, and that they could

choose another graft type if they did not wish to participate in

this study. Other surgical options in this hospital included

single-bundle reconstruction with hamstring tendon auto-

grafts, or a bone-patella tendon-bone autograft. The patients

who did not wish to take part in this study were not enrolled.

Two years after surgery, each patient was examined

with the standard clinical evaluation methods. One hundred

and twenty patients (83.9 %) underwent the same follow-

up examinations, while 23 patients were lost (Fig. 1); 16

patients were excluded from the evaluation because there

were no muscle torque data taken at the final follow-up.

Two patients were excluded for contralateral ACL injury

and ipsilateral revision ACL surgery, respectively. Finally,

three patients did not attend the regular follow-up after

ACL reconstruction up to 2 years postoperatively. There

were 68 men and 52 women with an average age of

27 years at the time of surgery. In Group I, 61 patients

underwent anatomic double-bundle ACL reconstruction

using the semitendinosus tendon alone. In Group II,

59 patients underwent anatomic double-bundle ACL

143 patients 2004: 55 patients2005: 42 patients2006: 46 patients

Group I Group II

Loss of f/u

61 patients 59 patients

75 patients 68 patients

23 patients

> 2-yr f/u > 2-yr f/u

Fig. 1 Flowchart demonstrating patient movement through the study

Y. Inagaki et al.

123

reconstruction using the semitendinosus and gracilis ten-

dons. There were no significant differences between the

two groups concerning age, gender, height, weight, and the

time to surgery (Table 1).

One senior orthopedic surgeon (K.Y.) performed all

operations using the same procedure for each group. At the

time of reconstruction, the medial or lateral meniscus was

partially resected in 21 patients, and repaired in nine

patients (Table 1). No treatment was administered for

softening or fissuring of the articular cartilage. In each

group, an approximately 3-cm-long incision was made in

the antero-medial portion of the proximal tibia, and the

hamstring tendon was harvested using a tendon stripper.

Concerning the graft selection for each patient, first, a

surgeon harvested the semitendinosus tendon. When the

harvested doubled distal portion of the semitendinosus

tendon was thicker than 6 mm, the semitendinosus tendon

was used for the AM and PL bundle graft. On the other

hand, when the harvested doubled distal portion of the

semitendinosus tendon was under 6 mm in thickness, the

gracilis tendon was harvested additionally. The length of

the semitendinosus tendon or/and gracilis tendon was

measured using a linear scale. The distance from the tibial

insertion of the tendon to its tendinous termination into

muscle was defined as the length of the tendon [22]. The

cross-sectional area of the tendon portion was measured

with a cylindrical gauge (Sizing system, Acufex, Smith &

Nephew Endoscopy, Andover, Massachusetts). After the

harvested tendon was passed through each stainless tube,

the greatest diameter of the gauge was defined as the

diameter of the tendon. Each reconstruction procedure was

performed using the arthroscopically assisted 1-incision

(transtibial tunnel) technique. Each graft was secured with

EndoButtons-CL (Smith & Nephew Endoscopy) on the

femur and with two staples (Meira, Nagoya, Japan) on the

tibia. All patients underwent postoperative management

with the same rehabilitation protocol [9, 23].

Graft fashioning technique

In Group I, when the harvested doubled semitendinosus

tendon was thicker than 6 mm, the semitendinosus tendon

was cut into two parts. The distal half of the semitendi-

nosus tendon was doubled (two strands) with side-by-side

sutures and used for the AM bundle graft, and the

remaining proximal half of the semitendinosus tendon was

also doubled (two strands) with side-by-side sutures and

used for the PL bundle graft (Fig. 2). The lengths of the

autografts were 60–70 and 50–60 mm for the AM and PL

bundle grafts, respectively. The autografts were connected

in series with 10-mm-width polyester tape (Leeds-Keio

Artificial Ligament, Neo Ligament) at the tibial side and

attached to the Endobutton-CL (Smith & Nephew Endos-

copy) (Fig. 2). The size of the Endobutton-CL was adjusted

so that an autogenous tendon portion of 15–20 mm was

located in the femoral and tibial tunnels. In Group II, when

the harvested semitendinosus tendon was under 6 mm in

thickness, the gracilis tendon was harvested additionally.

The distal half of the semitendinosus and the gracilis ten-

dons were doubled (four strands) with side-by-side sutures

and used for the AM bundle graft, and the remaining

proximal half of the semitendinosus tendon was doubled

(two strands) with side-by-side sutures and used for the PL

bundle grafts. However, when the doubled proximal half of

the semitendinosus tendon was under 5 mm in thickness,

the proximal half of the gracilis tendon was added addi-

tionally. Then, three or four strand semitendinosus and

gracilis tendons were used for the PL bundle grafts. After

that, the two hamstring hybrid autografts were fashioned in

the same manner as in Group I.

Anatomic double-bundle ACL reconstruction procedure

The details of this procedure have been previously

described in the literature [4, 13]. Briefly, a tibial tunnel for

the PL bundle was created. To insert a guidewire, a hole-in-

Table 1 Comparison of background factors of patients between

Groups I and II

Group I

(N = 61)

Group II

(N = 59)

P value

Age (years) 28.2 ± 11.9 26.2 ± 10.3 0.320

Male/female (patients) 35/26 33/26 0.873

Height (cm) 166.4 ± 7.9 166.3 ± 8.4 0.954

Weight (kg) 62.5 ± 9.3 65.6 ± 10.7 0.0954

Interval between injury and

operation (months)

23.6 ± 50.4 16.1 ± 46.3 0.394

Meniscal injury (patients)a 9:4 12:5 0.343

Values are expressed as mean ± SDa The number of patients with partial resection versus that with

meniscus repair

Fig. 2 The hamstring tendon hybrid autografts for anatomic double-

bundle anterior cruciate ligament reconstruction. The lengths of

autografts were 60–70 and 50–60 mm for AM and PL bundles,

respectively. AM anteromedial, PL posterolateral

Hamstring graft in DB ACL reconstruction

123

one guide (Wire-navigator, Smith & Nephew Endoscopy,

Tokyo, Japan) was used. The tibial indicator of the Navi-tip

was placed at the center of the PL bundle footprint on the

tibia. Then, keeping the tibial indicator at this point, the

femoral indicator was aimed at the center of the PL bundle

attachment on the femur. A guidewire was drilled through

the sleeve in the tibia. Then, a guidewire for AM bundle

reconstruction was inserted in the same manner. Using a

wire-navigator, the femoral indicator was aimed at the

center of the AM bundle attachment on the femur. The two

tibial tunnels were made with a cannulated drill corre-

sponding to the measured diameter of the prepared

substitute.

To create two femoral tunnels for the AM and PL

bundles in the lateral condyle, first, a guidewire was drilled

at the center of the femoral attachment of the AM bundle

through the AM tibial tunnel by use of an offset guide

(transtibial femoral ACL Drill Guide, Arthrex, Naples, FL,

USA). Using the inserted guidewire, a tunnel was made

with a 4.5-mm cannulated drill. The length of the tunnel

was measured with a scaled probe. Then, the portal for an

arthroscope was changed to the medial infrapatellar portal.

A guidewire was inserted at the center of the PL bundle

attachment on the femur through the PL tibial tunnel. A

4.5-mm-diameter tunnel was drilled, and its length was

measured in the same manner. Finally, two sockets were

created for the AM and PL bundles, respectively, with

cannulated drills, the diameter of which was matched to the

two grafts prepared with the technique described above.

Finally, the graft for the PL bundle was introduced

through the tibial tunnel to the femoral tunnel by use of a

passing pin. The EndoButton was flipped on the femoral

cortical surface. Then, the graft for the AM bundle was

placed in the same manner. For graft fixation, an assistant

surgeon simultaneously applied tension of 30 N to each

graft using two tensiometers (Meira, Nagoya, Japan) at 10�of knee flexion for 2 minutes. Then, a surgeon simulta-

neously secured the two tape portions onto the tibia using

two spiked staples (Meira) in the turn-buckle fashion.

Clinical evaluations

Each patient underwent clinical examination 2 years after

surgery. The side-to-side anterior laxity was measured with

a KT-2000 arthrometer (MEDmetric, San Diego, CA,

USA) at 30� of knee flexion under an anterior drawer force

of 133 N. A well-trained physical therapist who was blin-

ded to the procedure collected the KT-2000 arthrometer

results postoperatively. An experienced orthopedic surgeon

(E.K.), who was also blinded to the procedure, performed

the pivot-shift test, the results of which were subjectively

evaluated by the examiner. In evaluation of the pivot-shift

test [4, 9], the indication of ?? was defined when the

examiner felt a sudden rotational slip movement between

the tibia and femur, a so-called jog, during the test for the

injured knee. The ?? pivot-shift test result showed an

obvious failure of the ACL function. The indication of ?

was defined when the examiner felt some difference in the

rotational movement during the test between the injured

and uninjured knees but did not obviously feel the sudden

rotational slip movement. This condition showed some

insufficiency of the ACL function but did not show a

complete failure of the ACL. As to overall evaluation, the

Lysholm knee score (maximum score 100 points) and the

International Knee Documentation Committee (IKDC)

form were used. Peak isokinetic torque of the quadriceps

and the hamstrings was measured at 60 �/s of angular

velocity using KIN-COM (Chattecx Corp, Chattanooga,

TN, USA) in both knees after surgery. Muscle torque as

measured postoperatively in the uninvolved knee was

represented as a ratio (percentage) to the uninvolved value.

Statistical analysis

All data were shown as means with SD. For each param-

eter, unpaired Student’s t test and Chi-square test were

performed between the two groups. When a significant

result was obtained, a post hoc test with a Fisher protected

least significant difference test was made for multiple

comparisons. Correlations between the side-to-side anterior

laxity and the tunnel diameter were calculated by use of the

Pearson correlation coefficient. A commercially available

software program (StatView, SAS Institute, Cary, NC,

USA) was used for statistical calculation. The significance

level was set at P = 0.05.

Results

In Group I, the lengths and the diameters of the semiten-

dinosus tendon averaged 254.4 and 4.7 mm (Table 2),

respectively. In Group II, the lengths and the diameters of

the semitendinosus tendon averaged 235.2 and 4.4 mm

(Table 2), respectively, while those of the gracilis tendon

averaged 206.0 and 3.4 mm. The length of the semitendi-

nosus tendons in Group I was significantly longer than

those of the semitendinosus tendons (P \ 0.0001) in Group

II. The diameter of the semitendinosus tendons in Group I

was also significantly thicker than those of the semitendi-

nosus tendons (P \ 0.0001) in Group II. Concerning the

diameter of the AM bundle graft, Group II was signifi-

cantly greater (P \ 0.0001) than Group I (Table 2).

Regarding the diameter of the PL bundle graft, there were

no significant differences between the two groups. The

distribution of the AM and PL bundle graft diameter in

Groups I and II are shown in Fig. 3.

Y. Inagaki et al.

123

The postoperative side-to-side anterior laxity measured at

30� of knee flexion with the KT-2000 averaged 1.3 mm in

both groups, showing no statistical difference. In each group,

there was no significant relationship between the femoral and

tibial tunnel diameter of the AM and PL grafts and the

postoperative side-to-side anterior laxity (correlation coef-

ficients range; r = -0.171 to 0.17, P = 0.1189–0.9613).

Regarding the pivot-shift test (Table 3), the Chi-square test

showed no significant difference between the two groups.

Concerning the range of knee motion, there was no patient

with loss of terminal knee extension in Group I. On the other

hand, three patients had loss of knee extension over 5� in

Group II. In both groups, there was no patient with loss of

terminal knee flexion over 15�. Regarding the ratio of the

involved limb to the contralateral limb about the peak is-

okinetic torque of the quadriceps and the hamstrings, there

were no significant differences between the two groups

(Table 3). There were also no significant differences

between the two groups concerning the Lysholm knee score,

and the IKDC evaluation (Table 3).

Discussion

This study demonstrated that there are no significant dif-

ferences between the semitendinosus tendon alone and the

semitendinosus and gracilis tendon graft fashioning tech-

niques concerning the postoperative side-to-side anterior

laxity, the peak muscle torque, the range of knee motion,

the Lysholm knee score, and the IKDC evaluation 2 years

after anatomic double-bundle ACL reconstruction.

The postoperative side-to-side anterior laxity measured

with KT-2000 averaged 1.3 mm in both groups. The Lys-

holm score and the IKDC evaluation achieved good results

in both groups compared with the previously reported

results after anatomic double-bundle reconstruction [4–6,

Table 2 The lengths and diameters of the semitendinosus and grac-

ilis tendons, and the diameters of the anteromedial and posterolateral

tunnels of the femur and tibia in Groups I and II

Group I

(N = 61)

Group II

(N = 59)

P value

Semitendinosus tendon

Length (mm) 254.4 ± 15.9 235.2 ± 22.9 \0.0001

Diameter (mm) 4.7 ± 0.5 4.4 ± 0.5 \0.0001

Gracilis tendon

Length (mm) Not applicable 206.0 ± 26.0

Diameter (mm) 3.4 ± 0.5

Anteromedial bundle graft

Diameter (mm) 6.2 ± 0.5 6.9 ± 0.5 \0.0001

Posterolateral bundle graft

Diameter (mm) 5.9 ± 0.4 5.9 ± 0.3a 0.639

Values are expressed as mean ± SDa In 19 patients, the semitendinosus and gracilis tendons were used

for the PL bundle graft

0

10

20

30

40

50

5 5.5 6 6.5 7 7.5 8

Group I (PL bundle graft)

Pat

ient

s

Diameter (mm)

0

10

20

30

40

50

5 5.5 6 6.5 7 7.5 8

Group II (PL bundle graft)

Pat

ient

s

Diameter (mm)

0

10

20

30

40

50

5 5.5 6 6.5 7 7.5 8

Group II (AM bundle graft)

Pat

ient

s

Diameter (mm)

0

10

20

30

40

50

5 5.5 6 6.5 7 7.5 8

Group I (AM bundle graft)

Pat

ient

s

Diameter (mm)

Fig. 3 The distribution of the AM and PL bundle graft diameters in Groups I and II. AM anteromedial, PL posterolateral

Hamstring graft in DB ACL reconstruction

123

8–11, 13, 23]. The results indicated that this graft fash-

ioning procedure may be an effective method in restoring

knee stability in anatomic double-bundle ACL recon-

struction. The reason for the similar results in the two

groups is that the difference in tunnel diameter between the

two groups was\1 mm, so it may not affect the maturation

and the function of the grafts.

Recently, Zhao et al. [21] compared the clinical results of

a double-bundle ACL reconstruction with four strands ver-

sus eight strands of hamstring tendon graft. They reported

that a double-bundle ACL reconstruction with eight strands

yields significantly better results than a double-bundle ACL

reconstruction with four strands, concerning the side-to-side

difference in anterior knee laxity, the IKDC subjective result,

and the Lysholm score. However, Zhao et al. [21] described

that their double-bundle reconstruction was not an anatomic

reconstruction. On the tibial side, the inner openings of both

bundles were 7-mm anterior to the tip of the tibial spine and

located on a medial–lateral line, side by side. In addition, Li

[24] pointed out that the diameter of an eight-strand ham-

string tendon (mean diameter 8 mm, range 6–11 mm) was

significantly greater than that of a native ACL. Therefore,

notchplasty was performed in more patients in the eight-

strand hamstring tendon group than in the four-strand group.

Because of the obviously decreased volume of the knee

cavity following reconstruction, impingement between the

eight-strand hamstring graft and the posterior cruciate liga-

ment or the femoral notch might occur, which could impair

the strength and reduce the longevity of the graft, and even

the posterior cruciate ligament, because of intermittent shear

force following knee motion. In this study, the mean AM

bundle diameters were 6.2 and 6.9 mm in Groups I and II,

respectively. Therefore, in the author’s double-bundle

reconstruction, most patients did not show graft impinge-

ment, because the placed AM and PL grafts were relatively

thin and anatomically twisted in the intercondylar notch.

Therefore, the authors did not perform the notchplasty except

in chronic cases. However, the authors found some tendency

of a difference in postoperative loss of knee extension

between the two procedures. Namely, three patients showed

loss of knee extension by 5�–10� in Group II, while there

were no patients with loss of knee extension in Group I. From

the clinical viewpoint, the loss of knee extension is one of the

pathological conditions that should be absolutely avoided

after ACL reconstruction [25]. Even though this difference

was not statistically significant, the authors believe it was

clinically important for avoiding graft impingement.

Therefore, surgeons should consider the graft thickness for

graft preparation using the semitendinosus and gracilis ten-

don graft fashioning techniques. Recently, Niki et al. [20]

reported the clinical results of anatomic double-bundle ACL

reconstruction by use of bone-patellar tendon-bone and

gracilis tendon grafts and compared them with the results of

double-bundle ACL reconstruction by use of semitendinosus

tendon or semitendinosus and gracilis tendon grafts. At

2 years follow-up, there were no significant differences in

terms of the Lysholm score, Tegner activity level, and IKDC

evaluation among the three groups. In their hamstring dou-

ble-bundle ACL reconstruction, two double-looped semi-

tendinosus tendons were prepared for both the AM and PL

bundle grafts when the semitendinosus tendon was 24 cm or

longer. Gracilis tendon was harvested for the PL bundle graft

and double looped only when the semitendinosus tendon was

\24 cm long. This protocol was different from the authors’

graft preparation protocol. In our protocol, the authors con-

sider that a 6-mm diameter may be the necessary thickness of

the AM hamstring graft in ACL reconstruction, based on

previous cadaveric studies [26–28]. Although, in general,

long hamstring tendons have thick diameters, while short

tendons have thin diameters, surgeons should consider not

only the graft length but also the graft thickness for graft

preparation.

There are many reports on the effect of additional

gracilis tendon harvesting in single bundle ACL recon-

struction [29–34]. Tashiro et al. [34] reported that patients

with reconstructed single bundle ACLs with semitendino-

sus and gracilis tendons showed lower isokinetic and iso-

metric muscle peak torque in the deep knee flexion range

Table 3 Comparisons in the clinical outcome between Groups I and

II

Group I

(N = 61)

Group II

(N = 59)

P value

Anterior laxitya (mm) 1.3 ± 1.4 1.3 ± 1.5 0.825

Pivot-shift test 0.751

(-) 48 patients 45 patients

(?) 13 patients 14 patients

(??) 0 patient 0 patient

Loss of knee motion 0.0745

Loss of extension ([5�) 0 patient 3 patients

Loss of flexion ([15�) 0 patient 0 patient

Peak isokinetic torque of the

quadricepsb (%)

86.4 ± 15.9 87.6 ± 14.6 0.661

Peak isokinetic torque of the

hamstringsb (%)

93.3 ± 17.8 92.6 ± 15.1 0.842

Lysholm knee score (points) 97.0 ± 4.6 96.7 ± 5.4 0.775

IKDC score 0.593

A (normal) 46 patients 44 patients

B (nearly normal) 15 patients 14 patients

C (nearly abnormal) 0 patient 1 patient

D (abnormal) 0 patient 0 patient

Values are expressed as mean ± SDa Difference of anterior knee laxity between treated knee and unin-

jured knee (mm)b Ratio of treated knee to uninjured knee (%)

Y. Inagaki et al.

123

than those with semitendinosus only. Segawa et al. [33]

and Gobbi et al. [31] also reported the weakness of internal

rotation muscle strength in those with semitendinosus and

gracilis tendons than those with semitendinosus tendon

only. On the other hand, Ardern et al. [30] showed no

statistical difference concerning the muscle strength

between the groups which were grouped in the same

manner as this study. Nakamura et al. [32] reported a

decrease in maximum standing knee flexion angle of the

involved limb compared to the uninvolved limb in patients

who underwent ACL reconstruction with semitendinosus

and gracilis tendons, compared with those with semiten-

dinosus tendon only. On the contrary, the above mentioned

Ardern et al. [30] reported that there was no correlation

between the maximum standing knee flexion angle and the

isometric muscle torque at 105� of knee flexion, so it

should not be used as the clinical parameter of postopera-

tive muscle strength. Therefore, in addition to long-term

results, more detailed evaluation on the muscle strength

such as the value at deep knee flexion, and the internal

muscle torque of knee is needed.

There were several limitations to this study. The first

limitation is that the patients were not truly randomized

because the authors used the originally developed graft

fashioning protocol in anatomic double-bundle ACL

reconstruction. Although age, gender, height, weight, and

the time from injury to surgery were not completely the

same between the two groups, there were no statistical

differences. The second limitation is that the authors only

evaluated the peak isokinetic torque of the quadriceps and

the hamstrings at 60 �/s of angular velocity after ACL

reconstructions with hamstring tendon graft. The third

limitation is that the follow-up period was only 2 years.

Therefore, at the present time, the authors cannot speculate

as to whether there will be differences between the two

different graft fashioning techniques in terms of long-term

outcome of knee function and return to sports. The fourth

limitation is that the authors did not precisely evaluate the

ability of sports performance because, in the short-term

results, these parameters are commonly favorable, inde-

pendent of reconstruction procedures. In the future, the

authors should conduct a long-term follow-up study to

compare the subjective evaluation and the ability of sports

performance between the two groups. However, beyond

these limitations, the present study provided orthopedic

surgeons with important information on double-bundle

ACL reconstruction with hamstring tendons.

Conclusion

In this study, the authors did not find any significant dif-

ferences between the two graft types of hamstring tendon

hybrid autograft (semitendinosus only, or semitendinosus

and gracilis tendons) after anatomic double-bundle ACL

reconstruction, concerning postoperative side-to-side

anterior laxity, peak muscle torque, range of knee motion,

Lysholm knee score, or IKDC evaluation 2 years after the

operation.

Acknowledgments This research was supported in part by Grants-

in-Aid for scientific research (21500400) from the Ministry of Edu-

cation, Science and Culture, Japan. The authors report no conflict of

interest.

References

1. Amis AA, Dawkins GPC. Functional anatomy of the anterior

cruciate ligament: fiber bundle actions related to ligament

replacement and injuries. J Bone Joint Surg Br. 1991;73:260–7.

2. Sakane M, Fox RJ, Woo SL, Livesay GA, Li G, Fu FH. In situ

forces in the anterior cruciate ligament and its bundles in

response to anterior tibial loads. J Orthop Res. 1997;15:285–93.

3. Mott HW. Semitendinosus anatomic reconstruction for cruciate

ligament insufficiency. Clin Orthop Relat Res. 1983;172:90–2.

4. Yasuda K, Kondo E, Ichiyama H, Kitamura N, Tanabe Y, To-

hyama H, Minami A. Anatomic reconstruction of the anterome-

dial and posterolateral bundles of the anterior cruciate ligament

using hamstring tendon grafts. Arthroscopy. 2004;20:1015–25.

5. Aglietti P, Giron F, Cuomo P, Losco M, Mondanelli N. Single-

and double-incision double-bundle ACL reconstruction. Clin

Orthop Relat Res. 2007;454:108–13.

6. Jarvela T. Double-bundle versus single-bundle anterior cruciate

ligament reconstruction: a prospective, randomize clinical study.

Knee Surg Sports Traumatol Arthrosc. 2007;15:500–7.

7. Kondo E, Merican AM, Yasuda K, Amis AA. Biomechanical

comparisons of knee stability after anterior cruciate ligament

reconstruction between two clinically available trans-tibial pro-

cedures: anatomic double-bundle versus single-bundle. Am J

Sports Med. 2010;38:1349–58.

8. Kondo E, Yasuda K. Second-look arthroscopic evaluations of

anatomic double-bundle anterior cruciate ligament reconstruc-

tion: relation with prospective knee stability. Arthroscopy.

2007;23:1198–209.

9. Kondo E, Yasuda K, Azuma H, Tanabe Y, Yagi T. Prospective

clinical comparisons of anatomic double-bundle versus single-

bundle anterior cruciate ligament reconstruction procedures in

328 consecutive patients. Am J Sports Med. 2008;36:1675–87.

10. Kondo E, Yasuda K, Miyatake S, Kitamura N, Tohyama H, Yagi

T. Clinical comparison of two suspensory fixation devices for

anatomic double-bundle anterior cruciate ligament reconstruc-

tion. Knee Surg Sports Traumatol Arthrosc. 2012;20:1261–7.

11. Yagi M, Kuroda R, Nagamune K, Yoshiya S, Kurosaka M.

Double-bundle ACL reconstruction can improve rotational sta-

bility. Clin Orthop Relat Res. 2007;454:100–7.

12. Yagi M, Wong EK, Kanamori A, Debski RE, Fu FH, Woo SL.

Biomechanical analysis of an anatomic anterior cruciate ligament

reconstruction. Am J Sports Med. 2002;30:660–6.

13. Yasuda K, Kondo E, Ichiyama H, Tanabe Y, Tohyama H. Clin-

ical evaluation of anatomic double-bundle anterior cruciate lig-

ament reconstruction procedure using hamstring tendon grafts:

comparisons among 3 different procedures. Arthroscopy.

2006;22:240–51.

14. Hamner DL, Brown CH Jr, Steiner ME, Hecker AT, Hayes WC.

Hamstring tendon grafts for reconstruction of the anterior cruciate

Hamstring graft in DB ACL reconstruction

123

ligament: biomedical evaluation of the use of multiple strands

and tensioning techniques. J Bone Joint Surg Am. 1999;81:

549–57.

15. Yasuda K, Tsujino J, Ohkoshi Y, Tanabe Y, Kaneda K. Graft site

morbidity with autogenous semitendinosus and gracilis tendons.

Am J Sports Med. 1995;23:706–14.

16. Yasuda K, Tsujino J, Tanabe Y, Kaneda K. Effects of initial graft

tension on clinical outcome after anterior cruciate ligament

reconstruction. Autogenous doubled hamstring tendons con-

nected in series with polyester tapes. Am J Sports Med. 1997;25:

99–106.

17. Miyata K, Yasuda K, Kondo E, Nakano H, Kimura S, Hara N.

Biomechanical comparison of anterior cruciate ligament: recon-

struction procedures with flexor tendon graft. J Orthop Sci.

2000;5:585–92.

18. Numazaki H, Tohyama H, Nakano H, Kikuchi S, Yasuda K. The

effect of initial graft tension in anterior cruciate ligament

reconstruction on the mechanical behaviors of the femur-graft-

tibia complex during cyclic loading. Am J Sports Med.

2002;30:800–5.

19. Yamanaka M, Yasuda K, Tohyama H, Nakano H, Wada T. The

effect of cyclic displacement on the biomechanical characteristics

of anterior cruciate ligament reconstructions. Am J Sports Med.

1999;27:772–7.

20. Niki Y, Matsumoto H, Hakozaki A, Kanagawa H, Toyama Y,

Suda Y. Anatomic double-bundle anterior cruciate ligament

reconstruction using bone-patellar tendon-bone and gracilis ten-

don graft: a comparative study with 2-year follow-up results of

semitendinosus tendon grafts alone or semitendinosus-gracilis

tendon grafts. Arthroscopy. 2011;27:1242–51.

21. Zhao J, He Y, Wang J. Double-bundle anterior cruciate ligament

reconstruction: four versus eight strands of hamstring tendon

graft. Arthroscopy. 2007;23:766–70.

22. Ferrari JD, Ferrari DA. The semitendinosus: anatomic consider-

ations in tendon harvesting. Orthop Rev. 1991;20:1085–8.

23. Tohyama H, Kondo E, Hayashi R, Kitamura N, Yasuda K.

Gender-based differences in outcome after anatomic double-

bundle anterior cruciate ligament reconstruction with hamstring

tendon autografts. Am J Sports Med. 2011;39:1849–57.

24. Li B. Concerns about double-bundle reconstruction with 8-strand

hamstring tendon graft. Arthroscopy. 2008;24:969.

25. Shelbourne KD, Urch SE, Gray T, Freeman H. Loss of normal

knee motion after anterior cruciate ligament reconstruction is

associated with radiographic arthritic changes after surgery. Am J

Sports Med. 2012;40:108–13.

26. Takahashi M, Doi M, Abe M, Suzuki D, Nagano A. Anatomical

study of the femoral and tibial insertions of the anteromedial and

posterolateral bundles of human anterior cruciate ligament. Am J

Sports Med. 2006;34:787–92.

27. Mochizuki T, Muneta T, Nagase T, Shirasawa S, Akita K, Sekiya

I. Cadaveric knee observation study for describing anatomic

femoral placement for two-bundle anterior cruciate ligament

reconstruction. Arthroscopy. 2006;22:356–61.

28. Edwards A, Bull AMJ, Amis AA. The attachments of the

anteromedial and posterolateral fibre bundles of the anterior

cruciate ligament part 2: femoral attachment. Knee Surg Sports

Traumatol Arthrosc. 2008;16:29–36.

29. Adachi N, Ochi M, Uchio Y, Sakai Y, Kuriwaka M, Fujihara A.

Harvesting hamstring tendons for ACL reconstruction influences

postoperative hamstring muscle performance. Arch Orthop

Trauma Surg. 2003;123:460–5.

30. Ardern CL, Webster KE, Taylor NF, Feller JA. Hamstring

strength recovery after hamstring tendon harvest for anterior

cruciate ligament reconstruction: a comparison between graft

types. Arthroscopy. 2010;26:462–9.

31. Gobbi A, Domzalski M, Pascual J, Zanazzo M. Hamstring

anterior cruciate ligament reconstruction: is it necessary to sac-

rifice the gracilis? Arthroscopy. 2005;21:275–80.

32. Nakamura N, Horibe S, Sasaki S, Kitaguchi T, Tagami M,

Mitsuoka T, Toritsuka Y, Hamada M, Shino K. Evaluation of

active knee flexion and hamstring strength after anterior cruciate

ligament reconstruction using hamstring tendons. Arthroscopy.

2002;18:598–602.

33. Segawa H, Omori G, Koga Y, Kameo T, Iida S, Tanaka M.

Rotational muscle strength of the limb after anterior cruciate

ligament reconstruction using semitendinosus and gracilis tendon.

Arthroscopy. 2002;18:177–82.

34. Tashiro T, Kurosawa H, Kawakami A, Hikita A, Fukui N.

Influence of medial hamstring tendon harvest on knee flexor

strength after anterior cruciate ligament reconstruction, a detailed

evaluation with comparison of single- and double-tendon harvest.

Am J Sports Med. 2003;31:522–9.

Y. Inagaki et al.

123