Systematic Review

“Ligamentization” in Hamstring Tendon Grafts After Anterior

Cruciate Ligament Reconstruction: A Systematic Review of the
Literature and a Glimpse Into the Future
Leo Pauzenberger, M.D., Stefanie Syré, M.D., and Mark Schurz, M.D.

Purpose: The aims of this study were to provide a systematic review of the literature on “ligamentization” in hamstring
tendon (HT) grafts after anterior cruciate ligament (ACL) reconstruction and analyze the postoperative remodeling process
in human patients and animal models. Methods: We performed a search in the Medline, PubMed, Embase and The
Cochrane Library databases, followed by a manual search of reference lists to identify relevant articles. Only studies that
investigated the ligamentization of hamstring grafts by histologic examination and comprehensively reported on the
remodeling process were deemed eligible for review. Results: A total of 4 studies were determined appropriate for
systematic review: 2 of the included studies investigated human ACL grafts and 2 were performed in animal models. The
studies included the examination of 79 human hamstring graft biopsy specimens and grafts of 27 skeletally mature sheep.
To verify the remodeling process, authors reported on various aspects of cellularity, vascularity, and collagen organization.
None of the included studies reached a level of evidence higher than 3. Conclusions: A postoperative ligamentization
process can be found in HT grafts after ACL reconstruction and shows a typical progression through 3 distinguishable
remodeling phases in humans and in animal models, whereas the progression and intensity of remodeling is distinctly
increased in animals. Clinical Relevance: Because postoperative remodeling influences biomechanical properties of
hamstring grafts, a better understanding of this process and its timing could lead to substantial improvements in post-
operative care strategies and indirectly to the optimization of surgical techniques.

R upture of the anterior cruciate ligament (ACL) is

the most common ligament injury of the human
knee, with a reported incidence of 75,000 to 100,000 in
report that 72% to 100% of patients returned to even
the most strenuous athletic activity after ACL recon-
struction. Although strict criteria for return to sports
the United States alone, making ACL reconstruction were applied in these studies, failure rates were still
a very frequently performed orthopaedic procedure.1 reported at 5% to 16%.9-15 Despite great advances in
The goal of ACL reconstruction is restoration of knee diagnosis, treatment, and rehabilitation of knee liga-
stability to allow a safe return to athletic activities, ment injuries, potential for improvement still exists in
reduce the risk of secondary meniscal lesions, and early surgical techniques and postoperative treatment
onset of osteoarthritis.2-5 strategies.
Despite adequate surgical technique, satisfactory early In the 1980s, researchers first reported that patellar
outcome, and rehabilitation, graft failures are not tendon (PT) grafts used for the reconstruction of ACL-
uncommon; mid- to long-term reinjury rates were re- deficient knees undergo a functional adaption and
ported to be between 3% and 19%.6-8 Various authors a continuous postoperative remodeling processdfrom
the tendinous graft toward a tissue closely resembling
the structural properties of the native ACL.16,17 This
From the Department of Trauma Surgery, Medical University of Vienna,
regularly occurring remodeling process became popu-
Vienna, Austria.
The authors report that they have no conflicts of interest in the authorship larly known as ligamentization.17
and publication of this article. Although magnetic resonance imaging (MRI) pres-
Received March 18, 2013; accepted May 6, 2013. ents a viable alternative, the standard method for
Address correspondence to Leo Pauzenberger, M.D., Department of Trauma illustration of the remodeling process is histologic
Surgery, Medical University of Vienna, General Hospital, Level 6C, Währinger
examination of graft tissue. In ACL graft healing,
Gürtel 18-20, 1090 Vienna, Austria. E-mail: [email protected]
Ó 2013 by the Arthroscopy Association of North America several distinct phases can be observed upon histologic
0749-8063/13187/$36.00 examination: an early phase with tissue properties still
http://dx.doi.org/10.1016/j.arthro.2013.05.009 similar to the initial tendon graft or hypocellularity,

1712 Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 29, No 10 (October), 2013: pp 1712-1721
 LIGAMENTIZATION IN HAMSTRING TENDON GRAFTS 1713

followed by a proliferative phase with extensive Data Extraction

remodeling and revascularization, and a final phase of Results of the data extraction were recorded in an
maturation with characteristic restructuring of the graft adapted form previously proposed in a review by Claes
toward the properties of the native ACL.18-22 et al.,19 which was modified specifically for our purposes.
Of special clinical interest are findings suggesting that Every study was evaluated for model, study type, surgical
biomechanical properties of ACL grafts are greatly technique, graft origin, graft type, graft fixation, number
influenced by the remodeling process, whereas the of surgeons, clinical outcome, number of grafts exam-
mechanical characteristics decrease to a minimum ined, number and type of controls examined, study
during the extensive remodeling phase and recover interval, points of examination, examiner independence,
toward, but never reach, levels of the native ACL in the method of examination, method to verify the process of
late phase of graft maturation.18,23-26 ligamentization, and reported stages of ligamentization.
Given regular progress and clinical testing, it is
assumed that the healing process is sufficiently Selection of Studies
advanced 6 months after surgery to recommend The search was limited to articles found in the data-
a return to sports, but because clinical examination bases Medline, PubMed, Embase, and The Cochrane
cannot predict maximal loading capacities or the indi- Library between their earliest entry points and the day of
vidual progression of graft remodeling in reconstructed our search. Only controlled original studies analyzing the
ACLs, a better understanding of the ligamentization intra-articular remodeling process in HT or STG grafts
process could potentially improve postoperative care after ACL reconstruction by histologic examination with
strategies in the future. a minimum follow-up of 1 year were deemed eligible for
Although hamstring tendon (HT) autografts or allo- systematic review. To provide a comprehensive review
grafts replaced the PT as the primary graft of choice for of currently available research, we also decided to
ACL reconstructions, studies that comprehensively include controlled studies in animal models in our search
illustrate the ligamentization process of semitendinosus- but discuss those results separately. Articles focusing on
gracilis (STG) grafts are relatively rare. Recently, a the ligamentization process in PT or other grafts or graft-
systematic review by Claes et al.19 illustrated the post- tunnel integration were not included.
operative remodeling of human tendon grafts but did not
take possible differences between hamstring and PT Study Quality Assessment
grafts into consideration. Thus, the purpose of this review The methodological quality of selected studies was
was to systematically analyze and discuss available assessed with regard to study design, including model,
findings on the ligamentization of hamstring tendon sample size, presence of controls, surgical technique,
grafts in the current literature. use of independent or blinded examiners, or both, and
time of follow-up.
Methods
Results
Search Strategy
We performed a systematic review of the literature Search Results
in Medline, PubMed, Embase, and The Cochrane Our search revealed 906 potentially relevant articles.
Library databases on July 7, 2012 to identify all pub- After the removal of duplicates, titles and abstracts of
lished studies reporting on the postoperative remod- the remaining 383 articles were screened for relevance.
eling processes of HT grafts after ACL reconstruction. This process identified 17 articles that were obtained for
The search included all entries from the respective detailed evaluation.
database’s starting date to the day of our search, Additionally, the reference lists of these articles were
including articles published online as “Epub ahead of manually screened for relevant publications to ensure
print.” For the computerized search, a combination of comprehensiveness of the literature search. This com-
search terms related to the concepts of ligamentization, plementary search revealed one additional article,
remodeling, ACL reconstruction, hamstring, STG, and which was retrieved for further evaluation.28
semitendinosus graft was used. Additionally, refer- Over the course of detailed analysis, one article was
ences of selected articles were screened manually to excluded because it did not separate the results of HT
ensure a comprehensive review of the current litera- and PT grafts.21 One study was excluded because
ture. To document the search process in detail, the primary suture was used as therapy for ACL rupture,29
“Preferred Reporting Items for Systematic Reviews and and another study was excluded because biopsy
Meta-Analyses” (PRISMA)27 flow diagram, illustrating samples were taken from the site of ACL rupture.28 One
the number of identified, included, and excluded study was excluded because it focused on genetic
studies, as well as the reasons for exclusion was used expression of various factors and gene products over
(Fig 1). the course of remodeling.30 One study did not provide
1714 L. PAUZENBERGER ET AL.

Fig 1. The PRISMA flow diagram

illustrating the search process.27 (HT,
hamstring tendon; PT, patellar tendon.)

native HT or ACL controls.31 Nine studies did not report time points for examination; the animals were killed at
a number, method, or interval of examinations ad- this time, and the ACL grafts were examined. In these
equate to comprehensively illustrate the process of studies, 27 skeletally mature sheep weighing between
ligamentization.32-40 At the end of the selection process, 60 and 90 kg were used for histologic examination; an
2 studies in human patients41,42 and 2 studies in animal additional 20 sheep were examined for biomechanical
models43,44 were selected for in-depth review. testing.43,44
The assessment of study quality revealed that none of
Characteristics of Included Studies the included studies reached a level of evidence higher
All included studies focused on the process of liga- than 3. The highest quality of included studies was
mentization in hamstring autografts and presented reached by the reports of Sánchez et al.42 and Janssen
adequate study protocols for verification as well as et al.,41 which reached a level of evidence of 3. There
a comprehensible documentation of the remodeling were no randomized controlled trials found in the
process. Overall, included publications reported the literature that focused on the remodeling process in
examination of 79 human grafts and 47 sheep grafts.41-44 human or animal HT grafts after ACL reconstruction.
A summary of all included studies is given in Table 1. Only 3 articles explicitly reported the use of one or
more blinded independent examiners to conduct
Study Quality Assessment histologic analyses.42-44
Both human studies were designed as case-control
studies using biopsy samples taken during second-look ACL Reconstruction
arthroscopies that had become necessary because of In all included studies, the transplants used for ACL
reasons unrelated to the ACL reconstruction in one reconstruction were autologous STG or HT grafts, and
study and to cyclops lesions, meniscal tears, loose the type of graft preparation and use varied among the
bodies, and/or chondral lesions, removal of staples, or studies. One study reported the use of a quadruple
plica syndrome in the other. The studies included the hamstring graft in an arthroscopically assisted trans-
examination of 79 human patients between 18 and 54 tibial surgical technique with femoral Bone Mulch
years of age.41,42 Screw fixation and WasherLoc Tibial Fixation Device
Both studies performed in animal models were (Arthrotek/Biomet, Warsaw, IN), whereas graft tension
designed as controlled prospective studies with fixed at fixation was 90 to 100N with the knee fully
 LIGAMENTIZATION IN HAMSTRING TENDON GRAFTS 1715

Table 1. Study Characteristics

Study Quality Janssen (2011) Sanchez (2010) Kondo (2011) Goradia (2000)
Model Human Human Sheep Sheep
Study type Case-control study Case-control study Controlled animal study Controlled animal study
Level of evidence 3 3 Not mentioned Not mentioned
Independent examiner Not mentioned 3 independent examiners, 2 examiners, blinded Independent examiner
blinded
ACL reconstruction
Surgical technique Arthroscopic technique Arthroscopic technique ACL excision, open Open technique
PRGF addition (n ¼ 22) technique
Graft origin Autograft Autograft Autograft Autograft
Graft type Quadruple hamstring Hamstring tendons Doubled semitendinosus Double semitendinosus
Femoral fixation Bone Mulch Screw Transcondylar screws EndoButton EndoButton
Tibial fixation Washer Loc screw Bone plug, 2 metal staples Double spike plate Sutures, screw
Surgeons 1 surgeon (H.S.) Not mentioned Not mentioned 1 surgeon (V.G.)
Clinical outcome No signs of abnormal All knees were stable, no Significantly increased Stable at postoperative
laxity, negative pivot shift clinical symptoms before anterior-posterior examination
test result in included biopsy procedure translation
patients
Graft examination
No. of ACL grafts n ¼ 42 n ¼ 15, n ¼ 22 (PRGF) n ¼ 16 (histologic), n ¼ 20 n ¼ 11
(biomechanical)
No. of control ACLs n ¼ 17 (native HT) n ¼ 2 (native HT) n ¼ 16 (histologic), n ¼ 4 (native
n ¼ 8 (ACL) n ¼ 20 (biomechanical) semitendinosus)
Length of follow-up 117 mo 24 mo 52 wk 52 wk
Points of examination 6-12, 13-24, >24 mo 6-12, 13-18, 19-24 mo 0, 2, 6, 12, 24, 52 wk 4, 8, 12, 24, 52 wk
Method of examination Histologic analysis, Histologic analysis, light Histologic analysis, Histologic analysis,
immunostaining, microscopy polarized/conventional polarized/conventional
polarized/conventional light microscopy light microscopy
light microscopy
Ligamentization
Verification of Cellular, vascular, and Vascularity, cellularity, Cellularity, shape of the Cellularity, vascularity,
ligamentization myofibroblast density, extracellular matrix nuclei, collagen striations collagen organization,
collagen fibril alignment characteristics inflammation, crimp
pattern
Stages of ligamentization 3 3 3 3
Early healing phase 6-12 mo 6-12 mo 2-6 wk 4-8 wk
Remodeling phase 13-24 mo 13-18 mo 6-24 wk 8-24 wk
Maturation phase >24 mo 19-24 mo 24-52 wk 24-52 wk
ACL, anterior cruciate ligament; PRGF, platelet-rich growth factor.

extended. Postoperative rehabilitation started on the The clinical outcome after ACL reconstruction or at
first day, allowing full weight bearing when tolerated the time of the biopsy procedure, or both, was
and closed-chain exercises, supported by passive and mentioned in all studies. One study included in their
active flexion and extension exercises. Running, heavy report only patients with no signs of abnormal laxity
work activities, and return to competitive contact sports and negative pivot shift test results41; all knees were
were allowed from 2, 3, and 4 months, respectively.41 reported as stable and without any clinical symptoms
Sánchez et al.42 reported the use of “standard proce- before the time of biopsy procedures in the other
dure” and HT grafts without further explanation, tibial human study.42
fixed with a transcondylar screw, and femoral fixation Regarding clinical outcome in animal models, one
with a platelet-rich growth factor-treated bone plug and study reported a stable knee situation after ACL
2 metal staples. reconstruction,43 whereas the other found significantly
In all animals, an arthrotomy was performed and the increased anterior-posterior translation in all sheep
native ACL was excised, followed by ACL reconstruc- with reconstructed ACLs, but this study also showed
tion with a doubled semitendinosus autograft. The that the anterior-posterior translation decreased con-
grafts were fixed with femoral EndoButtons (Smith & tinuously until 52 weeks and was significantly lower
Nephew Endoscopy, Andover, MA) and a double spike than in knees with excised ACL without recon-
plate or bicortical cancellous screws in the tibia. After structions.44
surgery, animals were allowed to put full weight on the The surgeons who performed the ACL reconstruc-
operated knees without restriction of motion.43,44 tions were explicitly mentioned in only 2 studies.41,43
1716 L. PAUZENBERGER ET AL.

Graft Examination hypocellularity, especially toward the center of the

Native HTs were reported as controls by all authors, graft; acellular necrosis was not found in all grafts.
and one study also included native ACLs as histologic During this early phase, hypervascularity relative to
reference.41 The 2 studies of human grafts depended on a native ACL was already evident in the periphery of
examination of grafts during necessary second-look the graft. No cells with rod-like nuclei as seen in the
arthroscopies and thus ranged from first examinations native HT were present during this phase, but the size of
at 6 months postoperatively to last follow-ups of up to spindle-shaped fibroblast-like cell nuclei was increased
117 months after initial ACL surgery.41,42 Animals were compared with that in native HT and ACL. The collagen
killed at fixed time points from week 0 to week 52. bundles of the graft tissue showed random orientation
Both included animal studies had a minimum of 5 without identifiable uniform sinusoidal crimp pattern.
examination points scheduled, which allowed for an After this initial healing phase, a phase of intense
adequate evaluation of the ligamentization process and remodeling started between 6 and 8 weeks after
its phases.43,44 surgery. Although the granulation tissue presented as
The method of examination in both human studies a thick envelope at the beginning of this phase, it
included staining of samples with H&E followed by decreased in the peripheral areas of the graft with
evaluation under conventional light microscopy41,42; progression of the remodeling phase and was continu-
additionally, evaluation with polarized light microscopy ously replaced by longitudinally oriented collagen. In
was reported in one study.41 Biopsy specimens were contrast to the general hypocellularity of the early
obtained with a 3 mm basket42 or a 2.75-mm straight healing phase, the graft showed hypercellularity
forceps,41 resulting in specimens of 3  5 to 10 mm42 compared with the native ACL. The graft showed an
and 2 to 3 mm, respectively.41 The location of the inhomogeneous picture with areas of hypercellularity
biopsy procedure was reported as directly under the and hypocellularity. Fibroblast-like cell nuclei in the
femoral condyle42 or superficially from the mid- central portion of the graft predominantly had a
substance of one graft bundle.41 spindle-shaped form but then shifted to rod-shaped and
Authors of both animal studies reported the use of oval nuclei. During this phase, the graft remained
H&E staining and examination of the specimens with hypervascular, whereas the hypervascularity started to
conventional and polarized light microscopy, and tissue subside toward the end of the remodeling phase.
samples were taken from the middle intra-articular Collagen was still randomly oriented in the center
portion of the graft44 or the intra-articular portion was of the graft, whereas collagen in peripheral areas
excised and prepared for histologic examination.43 showed a more longitudinal orientation but without
a sinusoidal crimp pattern. With progression of the
Verification of the Remodeling Process remodeling phase, collagen in the central portion of the
To document the ligamentization process, its phases, graft showed a longitudinal orientation with a mod-
and the changes occurring in the remodeling graft, all erate number of mononuclear cells, and a uniform
authors investigated different aspects. crimp pattern could be identified.
All studies reported on cellularity aspects of the The first signs of a final ligamentization phase were re-
remodeling grafts, including cellularity, nuclear shape, ported from week 24 and were still present at 52 weeks.
nuclear orientation, and necrosis.41-44 In addition, both Granulation tissue was completely replaced at this time by
studies of human grafts reported on vascular aspects, ingrown collagenous tissue and was no longer distin-
including vascularity, vascular density or degree of guishable. The size and number of fibroblast-like cell
revascularization, and vascular maturity.41,42 Addi- nuclei were similar to earlier specimens but were
tionally, aspects of the extracellular matrix (crimp predominantly spindle-shaped or oval. The collagen
pattern, presence of glycosaminoglycans, collagen fibril fibers were well organized and longitudinally oriented
alignment)42 and myofibroblast density41 were re- throughout the graft. A uniform crimp pattern, appearing
ported. Besides cellular aspects, authors of animal similar to that of the normal ACL, could be seen in
studies also documented vascularity, collagen organi- increasing percentage throughout the graft. Cellularity
zation, inflammation, crimp pattern, shape of nuclei, continued to decrease toward levels of the native ACL,
and collagen striations.43,44 with no more signs of necrotic areas.43,44

Ligamentization of HT Grafts in Animal Models Ligamentization of Human HT Grafts

In included animals, the beginning of an early healing A phase of early healing could be identified in the first
phase could be seen from 2 to 4 weeks. During this months after ACL reconstruction. During this initial
phase, hamstring grafts showed granulation tissue in phase, synovial healing was found on the outside of the
the peripheral areas; this enveloping granulation tissue hamstring graft. The graft remained a tendinous struc-
became thicker toward the end of the initial healing ture with collagen crimp pattern similar to that of
phase. The early healing phase was characterized by a native HT, but irregular collagen orientation could
 LIGAMENTIZATION IN HAMSTRING TENDON GRAFTS 1717

Fig 2. Ligamentization in human ham-

string (*) and patellar (y) tendon
grafts.19-22,41,42 Adapted from Figure 2
in Claes et al.19 (ACL, anterior cruciate
ligament.)

also be seen. Vessel density was reported to be lowest studies reported the identification of 3 consecutive distin-
during this phase, without signs of revascularization. guishable phases of postoperative remodeling.
Localized reduction of cellularity could be observed, Controlled human studies illustrating ligamentization
but the graft’s overall cell density always remained of PT grafts found that an initial healing phase was
increased compared with the native HT. All cells were finished 6 months after surgery, followed by intense
viable and evenly distributed throughout the whole remodeling, which was shown to recede at 12 months,
graft. An increase of cell density at the end of this initial to a final maturation phase, which was shown to be
healing phase suggested early remodeling transitioning ongoing at 48 months (Fig 2).20-22 Compared to these
toward a phase of intense proliferation. findings, the included studies suggest a delayed pro-
Vascularity continuously increased, and the periphery gression of remodeling in human hamstring grafts,
of the graft was highly vascularized and cellular. Myo- whereas the ligamentization process occurs in a manner
fibroblast density, as well as overall cell density, was similar in principle to that reported for PT grafts and
maximally increased during this phase, indicating shows typical stages of remodeling.41,42
intense remodeling processes. Collagen bundles app- A recent review by Scheffler et al.,18 primarily based
eared more regularly oriented and aligned than in the on the data of animal studies focusing on PT graft
early postoperative period. remodeling, analyzed the ligamentization process in
This intense remodeling was followed by a final great detail and found that after an initial healing
maturation phase. In a continuous evolution, the phase, a phase of intense remodeling occurs up until 12
enveloping tissue and the initial graft became hardly weeks. Although comparison is difficult, time lines of
distinguishable. Throughout, cells seemed well oriented the included studies in sheep might suggest a slower
within a mature collagen matrix, which showed ligamentization process of HT grafts in animals as well
a regular crimp pattern similar to that of the native (Fig 3).
ACL. Nonetheless, areas of irregular cell and collagen The data from included studies could indicate a slower
orientation might remain, especially in the periphery of progression of remodeling in hamstring autografts
the graft. Overall cell density and myofibroblast density compared with PT grafts, but because of a lack of strictly
decreased but never reached the levels of a normal defined criteria for the ligamentization process, inter-
ACL. Vessel density increased further but did not reach pretation of results remains difficult. Whether the
the vessel density found in a native ACL. Although the seeming differences in the time line of postoperative
grafts were adapting over time toward a structure remodeling between HT and PT grafts are inherent or
similar to the ACL, distinct differences remained, and can be attributed to study design, surgical technique, or
the properties of an intact ACL could not be fully other factors is a topic of interest for future research.
restored.41,43 A possible reason for the heterogeneity of time lines
seen in human studies is the dependence of histologic
Discussion examinations on the necessity for second-look arthros-
According to the progression of cellularity, vessel density, copies that cannot be planned in advance. Patients in
and other histologic signs of graft maturity, all included whom second-look arthroscopy is necessary do not have
1718 L. PAUZENBERGER ET AL.

Fig 3. Timeline of the ligamentiza-

tion process of HT grafts in animal
models.18,43,44 (ACL, anterior cruciate
ligament.)

an ideal postoperative knee situation, which could human grafts, which could flaw results and may be one
influence the remodeling process, even when the reason reason for the differences found between animal and
for arthroscopy is unrelated to the primary reconstruc- human graft remodeling. Studies included in our review
tion.42,42 The authors of the included human studies described special vigilance for reproducible biopsy tech-
used routine arthroscopically assisted surgical tech- niques and separation of harvested tissue.41,42
niques, whereas in animals authors chose open surgical The present review shows that, in principle, the
approaches.41-44 Surgical technique, and especially an remodeling process is similar in human and animal HT
anatomically incorrect reconstruction of the ACL, could autografts, with major differences in the rate of
influence postoperative remodeling through a possible progression and intensity.41-44 As previously reported
increase in laxity or nonanatomical tension properties. in other grafts,18 HT autografts in animals showed
Goradia et al.43 reported knee stability after surgery, a more intense remodeling process, with more severe
whereas Kondo et al.44 found increased anterior-poste- deterioration of cellular properties during the early
rior laxity during the whole study, but no apparent healing phase43,44; this led to partial acellular necrosis
significant differences in timeline or reports of ligamen- of the graft in one study, which was not seen in human
tization could be identified. Janssen et al.41 reported that grafts.41,42 Xu et al.31 investigated the intrastrand
their surgical technique was not anatomically correct, maturation of hamstring grafts in a rabbit model and
but all patients’ knees were stable and without abnormal also found a chronological progression through the
laxity, and the time line of remodeling presented in their typical remodeling stages, with signs of necrosis during
study complies with an expected overall prolonged the initial healing phase. The initial decrease in cellu-
process in humans.19 Sánchez et al.42 also reported larity seems typical for animal grafts, whereas human
regular knee stability in all patients. Further research HT grafts presented with viable cells and increased
needs to be conducted to investigate the connection cellular density at all times compared with HT and ACL.
between surgical techniques, fixation systems, other Neither human study included examinations of the
possible influencing factors (e.g., rehabilitation program earliest healing phase, which might have prevented the
used, underlying injury) and the rate of ligamentization. identification of necrotic areas.41,42
Nonetheless, the goal has to be an anatomically correct A viable alternative to histologic examination could
reconstruction to provide optimal conditions for regular be the monitoring of postoperative remodeling with
progression of remodeling and early restoration of magnetic resonance (MR) scans.26,45-47 In 2001, Weiler
biomechanical properties. et al.26 published their findings from a study in sheep,
Regarding the points of examination and intervals which investigated the value of MRI for the prediction
used in the included studies, different approaches arise of ACL graft vascularity, biomechanical properties, and
from the chosen study design. The animal studies43,44 noninvasive observation of the ligamentization process.
both used a prospective design with fixed time points for The authors found that the changes in MR signal
examination, whereas the human studies41,42 were intensity of a tendon graft over time are consistent with
dependent on the necessity of second-look arthros- histologically determined phases of remodeling: In an
copies. With the availability of total extraction and initial phase representing the beginning of remodeling,
therefore relatively unlimited possibilities for examina- MR signal intensity continuously increased, and cellu-
tion of the graft, animal studies can potentially provide larity and vascularity reached a maximum intensity
more detailed data, especially with regard to explicit simultaneously. Thereafter signal intensity slowly
description of location and extensiveness of changes decreased, marking maturation of the graft and resto-
within the graft. Mentioned in the literature19,28 is the ration of its structural properties, closely resembling the
arguably superficial nature of biopsy procedures in native ACL. Furthermore, a significant negative linear
 LIGAMENTIZATION IN HAMSTRING TENDON GRAFTS 1719

correlation was found between biomechanical param- References

eters and MR signal intensity, which also illustrates the 1. Griffin LY, Agel J, Albohm MJ, et al. Noncontact anterior
progression of mechanical properties during ligamen- cruciate ligament injuries: Risk factors and prevention
tization previously shown in various studies.18,26 strategies. J Am Acad Orthop Surg 2000;8:141-150.
Bachmann et al.45 investigated the postoperative 2. Lohmander LS, Englund PM, Dahl LL, Roos EM. The
progression of signal intensity and appearance on MR long-term consequence of anterior cruciate ligament and
scans of semitendinosus grafts during the first 2 years meniscus injuries: Osteoarthritis. Am J Sports Med 2007;35:
1756-1769.
after ACL reconstruction in 74 patients. The authors
3. Daniel DM, Stone ML, Dobson BE, et al. Fate of the ACL-
reported that all expected changes in the remodeling
injured patient. A prospective outcome study. Am J Sports
graft could be identified by MRI. STG grafts presented Med 1994;22:632-644.
with low signal intensity compared with a normal PT up 4. Levy AS, Meier SW. Approach to cartilage injury in the
to 3 months, and signal intensity increased thereafter, anterior cruciate ligament-deficient knee. Orthop Clin
starting from the periphery of the graft progressing North Am 2003;34:149-167.
toward the center. From 1 to 2 years, signal intensity 5. Hawkins RJ, Misamore GW, Merritt TR. Followup of the
slowly decreased toward the initial low levels, begin- acute nonoperated isolated anterior cruciate ligament
ning from the center of the graft. These results suggest tear. Am J Sports Med 1986;14:205-210.
that the ligamentization process and the associated 6. Hui C, Salmon LJ, Kok A, et al. Fifteen-year outcome of
changes of biomechanical characteristics can be ad- endoscopic anterior cruciate ligament reconstruction with
patellar tendon autograft for “isolated” anterior cruciate
equately monitored by MRI. Further research using the
ligament tear. Am J Sports Med 2011;39:89-98.
advantages of verifying the ligamentization process
7. Deehan DJ, Salmon LJ, Webb VJ, Davies A, Pinczewski LA.
with a noninvasive tool such as MRI could help find Endoscopic reconstruction of the anterior cruciate ligament
definitive connections between postoperative remod- with an ipsilateral patellar tendon autograft. A prospective
eling and graft failure in a noninvasive way. Because longitudinal five-year study. J Bone Joint Surg Br 2000;82:
the remodeling process influences the biomechanical 984-991.
properties of the graft, we think that monitoring of 8. Pinczewski LA, Lyman J, Salmon LJ, et al. A 10-year
ligamentization could lead to improved postoperative comparison of anterior cruciate ligament reconstructions
rehabilitation strategies and indirectly to the optimiza- with hamstring tendon and patellar tendon autograft: A
tion of surgical techniques. controlled, prospective trial. Am J Sports Med 2007;35:
564-574.
Limitations 9. Beynnon BD, Johnson RJ, Fleming BC, et al. Anterior
Because the validity of a systematic review is ulti- cruciate ligament replacement: Comparison of bone-
patellar tendon-bone grafts with two-strand hamstring
mately determined by the quality of included studies,
grafts. A prospective, randomized study. J Bone Joint Surg
we have to report limitations of our study regarding the
Am 2002;84-A:1503-1513.
overall power of conclusions, resulting from the quality 10. Fu FH, Shen W, Starman JS, Okeke N, Irrgang JJ. Primary
of articles found in the literature. Additionally, as anatomic double-bundle anterior cruciate ligament
a result of the absence of clear criteria for the verifica- reconstruction: A preliminary 2-year prospective study.
tion of the ligamentization process, we found consid- Am J Sports Med 2008;36:1263-1274.
erable heterogeneity in the investigated parameters, 11. Colombet P, Robinson J, Jambou S, et al. Two-bundle,
detail of description, and study design, which made it four-tunnel anterior cruciate ligament reconstruction.
difficult to derive homogeneous data. In combination, Knee Surg Sports Traumatol Arthrosc 2006;14:629-636.
these limitations resulted in an inability to perform 12. Mascarenhas R, Tranovich M, Karpie JC, et al. Patellar
a quantitative analysis and allowed for only a descrip- tendon anterior cruciate ligament reconstruction in the
high-demand patient: Evaluation of autograft versus allo-
tive analysis of data found. These limitations show the
graft reconstruction. Arthroscopy 2010;26(9 suppl):S58-S66.
need to establish a more practicable tool for future
13. Sajovic M, Strahovnik A, Komadina R, Dernovsek MZ.
research into the ligamentization process. The effect of graft choice on functional outcome in ante-
rior cruciate ligament reconstruction. Int Orthop 2008;32:
Conclusions 473-478.
In principle, ligamentization occurs in human and 14. Zaffagnini S, Marcacci M, Lo Presti M, et al. Prospective
and randomized evaluation of ACL reconstruction with
animal hamstring grafts in a way similar to that previ-
three techniques: A clinical and radiographic evaluation at
ously reported in PT grafts. Furthermore, the post- 5 years follow-up. Knee Surg Sports Traumatol Arthrosc
operative healing process of HT grafts shows a typical 2006;14:1060-1069.
progression through 3 chronological, distinguishable 15. Zaffagnini S, Bruni D, Russo A, et al. ST/G ACL recon-
remodeling phases, whereas the progression and struction: Double strand plus extra-articular sling vs
intensity of remodeling is distinctly increased in double bundle, randomized study at 3-year follow-up.
animals. Scand J Med Sci Sports 2008;18:573-581.
1720 L. PAUZENBERGER ET AL.

16. Arnoczky SP, Tarvin GB, Marshall JL. Anterior cruciate 31. Xu Y, Ao Y-F. Histological and biomechanical studies of
ligament replacement using patellar tendon. An evalua- inter-strand healing in four-strand autograft anterior
tion of graft revascularization in the dog. J Bone Joint Surg cruciate ligament reconstruction in a rabbit model. Knee
Am 1982;64:217-224. Surg Sports Traumatol Arthrosc 2009;17:770-777.
17. Amiel D, Kleiner JB, Roux RD, Harwood FL, Akeson WH. 32. Scranton PE, Lanzer WL, Ferguson MS, Kirkman TR,
The phenomenon of “ligamentization”: anterior cruciate Pflaster DS. Mechanisms of anterior cruciate ligament
ligament reconstruction with autogenous patellar tendon. neovascularization and ligamentization. Arthroscopy 1998;
J Orthop Res 1986;4:162-172. 14:702-716.
18. Scheffler SU, Unterhauser FN, Weiler A. Graft remodeling 33. Papachristou G, Nikolaou V, Efstathopoulos N, et al. ACL
and ligamentization after cruciate ligament reconstruc- reconstruction with semitendinosus tendon autograft
tion. Knee Surg Sports Traumatol Arthrosc 2008;16:834-842. without detachment of its tibial insertion: A histologic
19. Claes S, Verdonk P, Forsyth R, Bellemans J. The “liga- study in a rabbit model. Knee Surg Sports Traumatol Arthrosc
mentization” process in anterior cruciate ligament recon- 2007;15:1175-1180.
struction: what happens to the human graft? A systematic 34. Zaffagnini S, De Pasquale V, Marchesini Reggiani L, et al.
review of the literature. Am J Sports Med 2011;39:2476- Electron microscopy of the remodelling process in
2483. hamstring tendon used as ACL graft. Knee Surg Sports
20. Abe S, Kurosaka M, Iguchi T, Yoshiya S, Hirohata K. Light Traumatol Arthrosc 2010;18:1052-1058.
and electron microscopic study of remodeling and matu- 35. Marumo K, Saito M, Yamagishi T, Fujii K. The “liga-
ration process in autogenous graft for anterior cruciate mentization” process in human anterior cruciate ligament
ligament reconstruction. Arthroscopy 1993;9:394-405. reconstruction with autogenous patellar and hamstring
21. Falconiero RP, DiStefano VJ, Cook TM. Revascularization tendons: A biochemical study. Am J Sports Med 2005;33:
and ligamentization of autogenous anterior cruciate liga- 1166-1173.
ment grafts in humans. Arthroscopy 1998;14:197-205. 36. Lane JG, McFadden P, Bowden K, Amiel D. The liga-
22. Rougraff B, Shelbourne KD, Gerth PK, Warner J. mentization process: A 4 year case study following ACL
Arthroscopic and histologic analysis of human patellar reconstruction with a semitendinosis graft. Arthroscopy
tendon autografts used for anterior cruciate ligament 1993;9:149-153.
reconstruction. Am J Sports Med 1993;21:277-284. 37. Weiss M, Unterhauser FN, Weiler A. Crimp frequency is
23. Yoshikawa T, Tohyama H, Katsura T, et al. Effects of local strongly correlated to myofibroblast density in the human
administration of vascular endothelial growth factor on anterior cruciate ligament and its autologous tendon
mechanical characteristics of the semitendinosus tendon grafts. Knee Surg Sports Traumatol Arthrosc 2012;20:889-
graft after anterior cruciate ligament reconstruction in 895.
sheep. Am J Sports Med 2006;34:1918-1925. 38. Bollen S, Pease F, Ehrenraich A, Church S, Skinner J,
24. Tohyama H, Yasuda K. Extrinsic cell infiltration and Williams A. Changes in the four-strand hamstring graft in
revascularization accelerate mechanical deterioration of anterior cruciate ligament reconstruction in the skeletally-
the patellar tendon after fibroblast necrosis. J Biomech Eng immature knee. J Bone Joint Surg Br 2008;90:455-459.
2000;122:594-599. 39. Kondo E, Yasuda K, Katsura T, et al. Local administration
25. Papageorgiou CD, Ma CB, Abramowitch SD, Clineff TD, of autologous synovium-derived cells improve the
Woo SL. A multidisciplinary study of the healing of an structural properties of anterior cruciate ligament auto-
intraarticular anterior cruciate ligament graft in a goat graft reconstruction in sheep. Am J Sports Med 2011;39:
model. Am J Sports Med 2001;29:620-626. 999-1007.
26. Weiler A, Peters G, Mäurer J, Unterhauser FN, Südkamp NP. 40. Suzuki D, Takahashi M, Abe M, Nagano A. Biochemical
Biomechanical properties and vascularity of an anterior study of collagen and its crosslinks in the anterior cruciate
cruciate ligament graft can be predicted by contrast- ligament and the tissues used as a graft for reconstruction
enhanced magnetic resonance imaging. A two-year study in of the anterior cruciate ligament. Connect Tissue Res
sheep. Am J Sports Med 2001;29:751-761. 2008;49:42-47.
27. Moher D, Liberati A, Tetzlaff J, Altman DG, PRISMA 41. Janssen RPA, van der Wijk J, Fiedler A, et al. Remodelling
Group. Preferred reporting items for systematic reviews of human hamstring autografts after anterior cruciate
and meta-analyses: The PRISMA statement. J Clin Epi- ligament reconstruction. Knee Surg Sports Traumatol
demiol 2009;62:1006-1012. Arthrosc 2011;19:1299-1306.
28. Johnson LL. The outcome of a free autogenous semite- 42. Sánchez M, Anitua E, Azofra J, et al. Ligamentization of
ndinosus tendon graft in human anterior cruciate recon- tendon grafts treated with an endogenous preparation
structive surgery: a histological study. Arthroscopy 1993;9: rich in growth factors: Gross morphology and histology.
131-142. Arthroscopy 2010;26:470-480.
29. Seitz H, Menth-Chiari WA, Lang S, Nau T. Histological 43. Goradia VK, Rochat MC, Kida M, Grana WA. Natural
evaluation of the healing potential of the anterior cruciate history of a hamstring tendon autograft used for anterior
ligament by means of augmented and non-augmented cruciate ligament reconstruction in a sheep model. Am J
repair: An in vivo animal study. Knee Surg Sports Traumatol Sports Med 2000;28:40-46.
Arthrosc 2008;16:1087-1093. 44. Kondo E, Yasuda K, Katsura T, et al. Biomechanical and
30. Roseti L, Buda R, Cavallo C, et al. Ligament repair: A histological evaluations of the doubled semitendinosus
molecular and immunohistological characterization. tendon autograft after anterior cruciate ligament recon-
J Biomed Mater Res A 2008;84:117-127. struction in sheep. Am J Sports Med 2012;40:315-324.
 LIGAMENTIZATION IN HAMSTRING TENDON GRAFTS 1721

45. Bachmann G, Cassens J, Heinrichs C, et al. MRT of the platelet-derived growth factors. Arthroscopy 2010;26:
knee in the follow-up of the plastic repair of the anterior 50-57.
cruciate ligament with autologous semitendinosus tendon 47. Figueroa D, Melean P, Calvo R, et al. Magnetic resonance
[article in German]. Rofo 1994;161:446-452. imaging evaluation of the integration and maturation of
46. Radice F, Yánez R, Gutiérrez V, et al. Comparison of semitendinosus-gracilis graft in anterior cruciate ligament
magnetic resonance imaging findings in anterior reconstruction using autologous platelet concentrate.
cruciate ligament grafts with and without autologous Arthroscopy 2010;26:1318-1325.