Original Article

Comparison of the Radiologic, Arthroscopic, and Clinical

Outcomes between Repaired versus Unrepaired Medial
Meniscus Posterior Horn Root Tear During Open Wedge
High Tibial Osteotomy
O-Sung Lee, MD1 Seung Hoon Lee, MD2 Yong Seuk Lee, MD, PhD1

1 Department of Orthopaedic Surgery, Seoul National University Address for correspondence Yong Seuk Lee, MD, Department of
College of Medicine, Bundang Hospital, Seongnam, Seoul, Korea Orthopaedic Surgery, Seoul National University College of Medicine,
2 Department of Orthopaedic Surgery, Incheon Metropolitan City Bundang Hospital, 166 Gumi-ro, Bundang-gu, Seongnam-si,
Medical Center, Seongnam, Korea Gyeonggi-do 463-707, South Korea
(e-mail: smcos1@daum.net; smcos1@snu.ac.kr).
J Knee Surg

Abstract The efficacy and outcomes for the concurrent repair of medial meniscus posterior horn

Downloaded by: University of Michigan. Copyrighted material.

root tear (MMPHRT) during open wedge high tibial osteotomy (OWHTO) are unclear.
This study compared the radiologic, arthroscopic, and clinical outcomes between
repaired and unrepaired MMPHRT during OWHTO. Fifty-seven patients were prospec-
tively enrolled from 2014 to 2016. The radiologic, arthroscopic, and clinical outcomes
were compared between 25 patients who underwent OWHTO with all-inside repair of
MMPRT using FasT-Fix (repaired group) and 32 patients who underwent OWHTO
without repair of MMPRT (unrepaired group) with a mean 2-year follow up in both
groups. The meniscal healing status was classified as complete, partial, or no healing,
according to second-look arthroscopic findings. The medial meniscal extrusion (MME)
was evaluated using magnetic resonance imaging. The width of medial joint space,
joint line convergence angle (JLCA), posterior tibial slope (PTS), Kellgren–Lawrence (KL)
grade, hip-knee-ankle angle, and weight-bearing line ratio was evaluated on simple
standing. The clinical outcomes were evaluated using the Knee Society score and the
Western Ontario and McMaster University score. Healing rates (partial and complete)
of the MMPHRT showed a statistical difference between the two groups (repaired
group vs. unrepaired group, 19/25 (76%) vs. 13/32 (40.6%), p ¼ 0.008). The post-
Keywords operative MME showed no statistical differences between groups (repaired versus
► knee unrepaired group: 4.5  1.3 mm vs. 4.5  2.1 mm, p ¼ 0.909). The postoperative
► open wedge high width of medial joint space, JLCA, PTS, and KL grade all showed no statistical differences
tibial osteotomy between groups after 2 years of OWHTO. Other radiologic parameters and clinical
► meniscus posterior outcomes showed no statistical differences between groups. Repair of the MMPHRT
horn root tear during OWHTO showed a superior healing rate to the unrepaired MMPHRT. However,
► repair repair of the MMPHRT was not related to the radiologic and clinical outcomes.
► healing Therefore, there is no clear evidence of the need for the MMPHRT repair during
► outcome OWHTO.

received Copyright © by Thieme Medical DOI https://doi.org/

October 28, 2018 Publishers, Inc., 333 Seventh Avenue, 10.1055/s-0039-1692992.
accepted after revision New York, NY 10001, USA. ISSN 1538-8506.
May 19, 2019 Tel: +1(212) 584-4662.
Efficacy and Outcomes of MMPHRT Repair Lee et al.

Open wedge high tibial osteotomy (OWHTO) is a well-estab- OWHTOs were diagnosed as MMPHRT, and arthroscopic all-
lished treatment option for medial compartment arthritis of inside repair using FasT-Fix (Smith & Nephew Endoscopy,
the knee joint. This procedure could provide the medial Andover, MA) was attempted on MMPHRT. However, four cases
compartment with a favorable mechanical environment for of MMPRT among the cases in which repair were initially
better healing of the articular cartilage by reducing the load.1–4 attempted were not suitable for repair with FasT-Fix, and as
During OWHTO, medial meniscus posterior horn root tear a result, only debridement and partial meniscectomy were
(MMPHRT) is commonly observed in the osteoarthritic knee. performed (►Fig. 1). All patients were recommended to
The incidence of MMPHRT has been frequently reported and undergo locking plate removal with second-look arthroscopy
this accounts for 80% of patients with osteoarthritis and 27.8% at approximately 2 years after OWHTO. This study obtained the
of medial meniscal tear injuries.5,6 approval of the institutional review board of our hospital.
MMPHRT is considered to be a radial tear or avulsion at the
insertion of the medial meniscus.7,8 A radial tear of the medial Surgical Technique
meniscus posterior root can result in a total meniscectomized A single-surgeon performed all surgical procedures. The
knee due to the loss of hoop tension by disrupting the critical target mechanical axis was the weight-bearing line passing
circumferential fibers.9,10 This pathology may cause meniscal through 62.5% of the width of the tibial plateau, correspond-
extrusion, loss of articular cartilage, joint space narrowing, and ing to a postoperative mechanical valgus of 2 to 4 degrees. An
eventually, progressive osteoarthritis.7,8,11–14 Recently, non- approximately 5-cm incision was made longitudinally at the
operative treatment and partial meniscectomy for MMPHRT anteromedial aspect of the proximal tibia. After release of the
has shown unsatisfactory results.15,16 Therefore, surgical superior border of the pes anserinus and the anterior border
repair of the MMPHRT has been increasingly performed.17–22 of the medial collateral ligament, horizontal osteotomy was

Downloaded by: University of Michigan. Copyrighted material.

Although the surgical treatment of the MMPHRT is increas- performed along guiding pins. Biplanar anterior osteotomy
ingly being attempted, reports of how it should be treated was then performed. The distraction was performed gradu-
during OWHTO are rare. In one study, MMPHRT showed a high ally at the most posterior gap until the target mechanical axis
rate of healing after OWHTO without attempted repair, and was obtained.27 The osteotomy site was fixed with a long
clinical and radiologic outcomes were not different between locking plate (DWLP, TDM, Seongnam, Korea).
healed and nonhealed groups.8 Additionally, it was recently A concomitant arthroscopy during OWHTO was per-
supposed that alignment correction is more important than formed on all patients included in this study. Diagnosis of
repair itself for the healing of MMPHRT, and it is recommended MMPHRT was confirmed when that could be raised easily on
to correct alignment if patients show varus alignments with probing and showed no true meniscal continuity between
MMPHRT.12,15,23 This could imply that MMPHRT could be the tibial attachment site and posterior horn of the medial
a secondary change of the osteoarthritis, and that there might meniscus. Other types of meniscal tears, rather than root
be questionable clinical relevance for its repair. tears, were excluded from this study, even if they had been
Therefore, this study compared the radiologic, arthro- diagnosed as a MMPHRT on preoperative MRI. In the repaired
scopic, and clinical outcomes between repaired and unre- group, the meniscal tear was reduced anatomically close to
paired MMPHRT during OWHTO to evaluate the necessity of the origin of the meniscal root using the tip of the FasT-Fix
MMPHRT repair during OWHTO. The hypotheses of this study 360 curved needle delivery system (Smith & Nephew Endo-
were that (1) there are no differences in the healing rate scopy, Andover, MA).20 When the meniscus was adequately
between the repaired and unrepaired MMPHRT during reduced, the first and second implants were positioned
OWHTO, and (2) other radiologic and clinical outcomes do sequentially in the most suitable location to maintain the
not differ between the groups. hoop of the meniscus. The knot was then snugged down to
form the suture construct. Finally, we trimmed the free end
of the suture materials. If necessary, for the effective repair,
Materials and Methods
multiple sutures were performed. In the unrepaired group, a
Patients Selection simple debridement was performed to refresh the degen-
Between March 2014 and February 2016, 179 patients who erative site of meniscal tear and promote the healing of the
underwent OWHTO were retrospectively reviewed. All patients MMPHRT. In cases of MMPHRT in which repair was initially
underwent preoperative 3-T magnetic resonance imaging attempted but was found to be unsuitable for FasT-Fix repair,
(MRI) scans (Ingenia, Philips Healthcare, Best, the Netherlands), a simple debridement was performed to promote the spon-
and an MMPHRT was defined as a radial tear at the posterior taneous healing of the MMPHRT. No intra-articular drains
attachment of the medial meniscus by a radial linear defect at were inserted in either group.
the posterior insertion in the axial plane, a vertical linear defect
at the root on the coronal plane, and a ghost sign in the sagittal Postoperative Rehabilitation
plane.24–26 During OWHTO, it was confirmed by concomitant For the repaired group, weight bearing was delayed until
arthroscopy. From March 2014 to February 2015, 32 cases 2-weeks postoperatively, and only then was partial weight
among a total of 96 OWHTOs were diagnosed as MMPHRT, bearing with crutches was permitted. Full weight-bearing was
and a simple debridement on the site of meniscal tear was permitted beginning at 4 weeks postoperatively, only if the
performed to promote the healing of the MMPHRT. From patient could tolerate it. Lifestyle modifications were recom-
March 2015 to February 2016, 30 cases among a total of 83 mended to avoid deep flexion of the knee. For the unrepaired

The Journal of Knee Surgery

 Efficacy and Outcomes of MMPHRT Repair Lee et al.

Downloaded by: University of Michigan. Copyrighted material.

Fig. 1 Flow diagram of the study based on Consolidated Standards of Reporting Trials (CONSORT) guidelines. MMPHRT, medial meniscus
posterior horn root tear; OWHTO, open wedge high tibial osteotomy.

group, tolerable weight bearing with crutches was encouraged 5.0.9.2 (INFINITT, Seoul, South Korea) was used for the all
immediately postoperation, and full weight-bearing was radiographic measurements. Medial meniscal extrusion
permitted, if the patient was able. All patients performed (MME) was defined as the amount of meniscus displacement
isometric exercises and passive range of motion (ROM) exer- from the medial edge of the tibial plateau to the periphery of
cises with the operated knee 2 days after surgery until a the meniscal body, at the level of the medial collateral ligament
maximum flexion angle of 130 degrees or more was achieved. in the coronal view using the 3-T MRI scans (Ingenia, Philips
For both groups, squatting with weight was avoided for Healthcare, Best, the Netherlands).24,25,29 The width of the
3 months, and patients were instructed to exercise caution medial joint space was measured from the center of the medial
when rising from a seated position. femoral condyle to the center of the medial tibial plateau on
standing posteroanterior view with 45-degree knee flexion
Evaluation of MMPHRT Healing (Rosenberg’s view).22,30,31 The joint line convergence angle
The healing status of MMPHRT was classified as complete, (JLCA), posterior tibial slope (PTS), weight-bearing line (WBL)
partial, or no healing, according to second-look arthroscopic ratio, and hip-knee-ankle (HKA) angle were measured pre-
findings.8,19,28 Complete healing was defined as meniscal operatively, and at the time of second-look arthroscopy on full-
continuity with no cleft, no lifting on probing, and normal length standing anteroposterior radiographs. The JLCA was
meniscal tension at the repair site. Partial healing included both measured as the angle between the line connecting the distal
lax healing and scar tissue healing, as defined by Seo et al.28 Lax femur and the proximal tibial articular surfaces. The PTS was
healing was defined as apparent increases in meniscus lifting defined as the angle formed between the proximal tibial
and mobility on probing with good meniscal continuity. Scar plateau and a line perpendicular to the anatomical axis of
tissue healing was defined as a meniscus that could be easily the tibial shaft on a lateral radiograph. The WBL was drawn
raised on probing, and showed no true meniscal continuity from the center of the femoral head to the center of the
except for some connecting scar tissue fibers between the tibial superior articular surface of the talus. To calculate the WBL
attachment site and the posterior horn of the medial meniscus. ratio, the denominator was the width of the tibia, measured
No healing was defined as no meniscal continuity, without any using a ruler, and the numerator was the tibial intersection of
evidence of meniscal healing at the repair site. the WBL in the knee joint (with the medial tibial edge at 0%).
The HKA angle was measured as the angle between the line
Evaluation of Radiologic Outcomes from the center of the femoral head to the center of the knee
Radiographic evaluations were performed preoperatively and joint, and the line from the center of the knee joint to the center
at the time of second-look arthroscopy. The INFINITT version of the ankle joint.

The Journal of Knee Surgery

Efficacy and Outcomes of MMPHRT Repair Lee et al.

The Kellgren–Lawrence (KL) grade was used for the eva- in the unrepaired group (OWHTO without repair of MMPRT),
luation of the radiologic severity of osteoarthritis on full- with a mean follow-up of 2 years. There were no statistical
length standing anteroposterior radiographs (grade 0, no differences between groups for preoperative demographics.
degenerative change; grade 1, questionable osteophytes and Additionally, no statistical differences between groups were
no joint space narrowing; grade 2, definitive osteophytes observed with respect to preoperative ROM, WOMAC scores,
with possible joint space narrowing; grade 3, definitive joint and knee scores (►Table 1).
space narrowing with moderate multiple osteophytes and The healing of the MMPHRT was observed in 76.0% (19/25)
some sclerosis; and grade 4, severe joint space narrowing of the patients in the repaired group, whereas it was observed
with cysts, osteophytes, and sclerosis). Additionally, the in 40.6% (13/32) of the patients in the unrepaired group. There
articular cartilage status of the medial femoral cartilage was a statistically significant difference between the groups for
was evaluated according to the International Cartilage Repair healing rate (p ¼ 0.008) (►Table 2). Among the patients
Society (ICRS) grade by an arthroscopic examination during whose MMPHRT was healed, complete healing was observed
OWHTO at the time of metal removal (ICRS 0, a macrosco- in 40% (10/25) of patients in the repaired group, and in 15.6%
pically normal cartilage without notable defects; ICRS 1, a (5/32) of patients in the unrepaired group (p ¼ 0.038). Partial
cartilage with a fibrillated, slightly softened surface or super- healing was observed in 36% (9/25) of patients in the repaired
ficial fissures; ICRS 2, a defect < 50% of the cartilage thick- group, and in 25% (8/32) of patients in the unrepaired group
ness; ICRS 3, a defect > 50% of the cartilage thickness; and the MMPHRT (p ¼ 0.368; ►Table 2). ►Fig. 2 shows a patient
ICRS 4, a full-thickness osteochondral injury). who achieved complete healing of MMPHRT at 2-year follow-
up after OWHTO with MMPHRTrepair (►Fig. 2). ►Fig. 3 shows
Evaluation of Clinical Outcomes a patient with a good postoperative alignment and clinical

Downloaded by: University of Michigan. Copyrighted material.

Flexion contracture and active maximal flexion were mea- outcomes, although she did not achieve a healing of the
sured in the supine position using a goniometer preopera-
tively and at the time of the second-look arthroscopy. The
clinical status of each patient was assessed using the Western
Table 1 Comparison of preoperative demographics and
Ontario and McMaster University (WOMAC) score and the
measurements
Knee Society (KS) knee and functional scores.
Repair Unrepaired p-Value
Analysis of Possible Variables Affecting Healing Status group group
Repair of MMPHRT, undercorrection and overcorrection Patients number 25 32 –
compared with acceptable correction, preoperative KL grade,
Age (y) 58.1  4.2 59.8  5.3 0.240a
preoperative degree of meniscus extrusion, age, body mass
index (BMI), and sex were subjected to multivariate logistic Sex (male/female) 8/18 10/24 0.865b
analysis to identify independent predictors of healing status Side (left/right) 11/15 20/14 0.239b
for MMPHRT after OWHTO. The criteria for undercorrection Height (cm) 158.8  7.7 155.6  8.4 0.189a
and overcorrection of WBL ratio were 50 and 70%, respec- Weight (kg) 68.6  10.5 64.9  10.3 0.240a
tively.32 Preoperative KL grade was divided into grades 3 and
Body mass index 27.1  3.2 26.8  3.7 0.744a
4. Preoperative extrusion of the medial meniscus was (kg/m2)
divided by 3 mm based on the previous study.33 The cut-
Onset of symptom 25.4  14.0 28.5  22.7 0.780a
off values of age and BMI were 60 and 25, respectively. (mo)
Follow-up (y) 1.9  2.4 2.2  2.1 0.120a
Statistical Analysis
Flexion contracture 4.2  4.1 2.7  2.1 0.097a
All statistical analyses were performed with SPSS version
(degrees)
22.0 statistical package (IBM Corp., Armonk, NY). Mean and
Active flexion 129.5  8.1 127  7 0.077a
standard deviation (SD) are presented for continuous vari-
(degrees)
ables. The differences in continuous variables were analyzed
WOMAC Total 41.4  8.4 40.2  10.5 0.158a
with Student’s t-test. The differences in other categorical
scores
variables were analyzed with Pearson’s Chi-square test or by Pain 8.7  2.4 8.8  3.9 0.314a
linear association. Independent risk factors for nonhealing Stiffness 6.6  2.1 4.1  2.0 0.070a
were examined using a multivariable logistic regression Function 26.2  6.3 27.3  12.7 0.317a
model with a backward stepwise procedure. Results are
Knee Knee 46.6  11.9 53.2  18.2 0.169a
expressed using odds ratios (OR) and their 95% confidence scores scores
intervals (CI). Statistical significance was set at p < 0.05. Function 57.6  7.7 57.6  10.9 0.988a
scores

Results Abbreviation: WOMAC, Western Ontario and McMaster University.

Note: the values are presented as mean  standard deviation.
Among a total of 55 patients included in this study, 25 patients a
Derived with Student’s t-test.
were included in the repaired group (OWHTO with all-inside b
Derived with Pearson’s Chi-square test. The statistical significance was
repair of MMPRTusing FasT-Fix) and 32 patients were included set at p < 0.05.

The Journal of Knee Surgery

 Efficacy and Outcomes of MMPHRT Repair Lee et al.

Table 2 Comparison of the healing of MMPHRT between 60, BMI over 25 kg/m2, and sex were not associated with the
repaired group and unrepaired group healing status of MMPHRT after OWHTO (►Table 5).

Repaired Unrepaired p-Value

group group Discussion
(n ¼ 25) (n ¼ 32)
The most important finding of the present study was that
Total number of 19 (76.0) 13 (40.6) 0.008a repair of MMPHRT using FasT-Fix during OWHTO signifi-
MMPHRT healing (%)
cantly increased the healing rate of MMPHRT compared with
Complete healing (%) 10 (40) 5 (15.6) the unrepaired treatment; however, it was not related to the
Partial healing (%) 9 (36) 8 (25) postoperative radiologic and clinical outcomes. Although
No healing (%) 6 (24) 19 (59.4) complete healing of the MMPHRT was achieved in some of
the patients (15.6%) after OWHTO without repair of the torn
Abbreviation: MMPHRT, medial meniscus posterior horn root tear. meniscus, healing was either incomplete or not obtained in
Note: the values are presented as mean  standard deviation.
a most patients (84.4%). Despite the differences in the healing
The comparison was performed between the healing (n ¼ 19, 76%)
and no healing (n ¼ 13, 40.6%) using the Pearson’s Chi-square test. The rates, patients who underwent OWHTO showed similar
statistical significance was set at p < 0.05. radiologic and clinical results, regardless of repair of the
concurrent MMPHRT. Additionally, no repair, undercorrec-
tion of WBL ratio (< 50%), and preoperative KL grade 4 rather
MMPHRT at 2 years after OWHTO without MMPHRT repair than grade 3, were significant predictors of no healing of
(►Fig. 3). MMPHRT after OWHTO.

Downloaded by: University of Michigan. Copyrighted material.

The preoperative and postoperative MMEs of the repaired MMPHRT can lead to loss of the hoop strain by disrupting
group were 4.6  1.9 mm and 4.5  1.3 mm, respectively. the critical circumferential fibers.5,34 The subsequent per-
The preoperative and postoperative MMEs of the unrepaired ipheral displacement of the meniscus weakens its capacity to
group were 4.3  1.6 mm and 4.5  2.1 mm, respectively. protect the articular cartilage, and, as a result, progresses
There were no statistically significant differences between medial compartment osteoarthritis.10,35 With regards to the
the preoperative and postoperative MMEs between groups fate of MMPHRT without surgery, Krych et al36 reported that
(p ¼ 0.195 and p ¼ 0.909, respectively). The preoperative nonoperative management resulted in poor clinical out-
width of the medial joint space were not statistically differ- comes, worsening arthritis, and a relatively high rate of
ent between groups (repaired vs. unrepaired, 3.4  1.0 mm arthroplasty at 5-year follow-up. The arthroscopic partial
vs. 3.7  1.2 mm, p ¼ 0.095), and the postoperative width of meniscectomy (the traditional surgical option of the
the medial joint space was also not statistically different MMPHRT) has been known to provide satisfactory clinical
between groups (repaired vs. unrepaired, 3.7  1.3 mm vs. outcomes.16,37 However, although it may improve clinical
3.8  1.3 mm, p ¼ 0.943; ►Table 3). symptoms of patients with MMPHRT, there has been concern
There were no statistically significant differences in the about the worsening of osteoarthritis after partial menis-
preoperative and postoperative JLCA, PTS, WBL ratio, and cectomy.16,37–39 Han et al reported that 16 of 46 patients
HKA angle between groups. Moreover, the severity of (35%) showed radiographic progression of osteoarthritis at a
osteoarthritis indicated by the KL grade was not statistically mean follow-up of 77 months, and, in addition, only 56% of
different between groups preoperatively, and at the time of patients had improvements in pain.38 Chung et al16 also
the second-look arthroscopy (p ¼ 0.243 and p ¼ 0.789, reported that refixation of the MMPHRT slowed the progres-
respectively). In addition, the preoperative and postopera- sion of osteoarthritis compared with partial meniscectomy,
tive cartilage status of the medial femoral condyle assessed although it did not completely prevent the progression of
by the ICRS grade was also not statistically different between arthritis.
groups (p ¼ 0.426 and p ¼ 0.497, respectively; ►Table 3). Contrary to nonoperative management and partial menis-
In terms of the clinical outcomes, there were no statistically cectomy of the MMPHRT, repair of MMPHRT improved
significant differences in the preoperative and postoperative clinical scores and restored tibiofemoral contact pressures
values of flexion contracture and active maximal flexion. All to a more favorable level.40,41 Additionally, surgeons can
clinical scores improved at the 2-year follow-up after OWHTO expect good healing after the repair of MMPHRT because the
in both groups. There were no statistically significant differ- anterior and posterior horns of the menisci have been known
ences in the postoperative values of KS knee and functional to have good vascular supply.42 Clinically, many authors have
scores, or WOMAC scores between groups (►Table 4). reported better outcomes for repair of MMPHRT compared
Multivariable logistic regression analysis revealed that the with partial meniscectomy and nonoperative manage-
following three factors were significant independent predic- ment.15,36,37 Kim et al37 reported that arthroscopic pullout
tors of healing status of MMPHRT after OWHTO: no repair repair of a medial MRT achieved better clinical and radiologic
(OR ¼ 36.35, 95% CI: 2.78–475.88, p ¼ 0.006), undercorrec- results, and good healing with restoration of hoop tension of
tion of WBL ratio (< 50%; OR ¼ 24.51, 95% CI: 1.64–367.37, the meniscus on MRI and second-look arthroscopy, com-
p ¼ 0.021), and preoperative KL grade 4 (OR ¼ 43.75, 95% CI: pared with partial meniscectomy at a mean follow-up of
3.78–506.57, p ¼ 0.002). Overcorrection of WBL ratio (> 70%), 48.5 months. In a recent systematic review of 172 patients,
preoperative large extrusion of meniscus (> 3 mm), age over arthroscopic transtibial pullout repair improved functional

The Journal of Knee Surgery

Efficacy and Outcomes of MMPHRT Repair Lee et al.

Downloaded by: University of Michigan. Copyrighted material.

Fig. 2 (A) A 58-year-old woman shows a medial meniscus posterior horn root tear (MMPHRT) on MRI; (B) arthroscopic view of repaired MMPHRT
at the time of open wedge high tibial osteotomy; (C) second-look arthroscopy showing complete healing of MMPHRT at postoperative 2 years;
(D) comparison between preoperative and postoperative limb alignments.

scores and prevented the progression of osteoarthritis in among 20 knees treated with pullout refixation using simple
most patients, at least during a short-term follow-up stitches showed a progression of KL grade, and the articular
(30.2 months).43 However, the long-term results of the cartilage of 10 knees which underwent second-look arthro-
MMPHRT repair are still unclear because healing of scopy healed completely at a minimum of 2-years follow-up.
the torn root after repair would be disturbed by the low- Kim et al37 reported that pullout suture of an MMPHRT
healing potential of degenerative meniscal tissues, and showed better clinical and radiologic outcomes compared
remaining medial extrusion would be a biomechanical factor with partial meniscectomy. Among 14 patients treated with
impeding healing of the repaired MMPHRT.5,23,44,45 Indeed, pullout suture, 64.3% showed normal fixation strength,
in a study with a midterm follow-up (mean, 72 months, 71.4% had normal hoop strain, and only 6.7% had retears of
range, 60–110 months), KL grade progressed in 67% of the the meniscus on second-look examinations at a mean follow-
patients with refixation of MMPHRT.16 up of 48.5 months. Seo et al28 reported that pullout refixa-
Surgeons have introduced several techniques for better tion using double-loop sutures showed significant improve-
healing of MMPHRT. Lee et al46 reported that only one (4.8%) ment in clinical outcomes and KL grade progression occurred

The Journal of Knee Surgery

 Efficacy and Outcomes of MMPHRT Repair Lee et al.

Downloaded by: University of Michigan. Copyrighted material.

Fig. 3 (A) A 55-year-old woman performed open wedge high tibial osteotomy without repair of medial meniscus posterior horn root tear.
(MMPHRT); (B) second-look arthroscopy showing the unhealed MMPHRT at postoperative 2 years; (C) MRI showing the MMPHRT at
postoperative 2 years; (D) comparison between preoperative and postoperative limb alignments.

in only one of 21 patients. The suture anchor technique has ulation of torn meniscus compared with previous techni-
been introduced as another repair option, with the advan- ques, such as pullout refixation and the suture anchor
tages of avoiding tibial tunnel and additional incision. Jung technique. Although direct comparisons with other methods
et al18 reported that the suture anchor technique showed were not possible because it was performed concurrently
significant improvements in clinical outcomes, meniscal with OWHTO in our study, the healing rate with our
extrusion, and a healing rate of 90% at a mean follow-up of technique was comparable to those of previous studies.
30.8 months. Additionally, Kim et al17 reported that both Several prognostic factors leading to poor outcomes after
suture anchor refixation and pullout refixation showed the refixation of MMPHRT have been reported.12,15,23
similar improvements in clinical outcomes and progression Previous studies commonly reported that concomitant
of KL grade. In our study, MMPHRT was repaired with a high grade chondral lesions and varus alignment were
method using a Fast-Fix 360 (Smith & Nephew Endoscopy, poor prognostic factors for clinical outcomes.12,15,23 Similar
Andover, MA) similar to the all-inside method, which to previous literature, multivariable logistic regression ana-
Thompson and Pinczewski20 introduced. The technique lysis in our study revealed that no repair, undercorrection of
was dramatically simpler in terms of reduction and manip- WBL ratio (< 50%), and preoperative KL grade 4 rather than

The Journal of Knee Surgery

Efficacy and Outcomes of MMPHRT Repair Lee et al.

Table 3 Comparison of radiologic and arthroscopic results between repaired group and unrepaired group

Repaired group (n ¼ 25) Unrepaired group (n ¼ 32) p-Value

Medial meniscal extrusion
Preoperative 4.6  1.9 4.3  1.6 0.195a
Postoperative 4.5  1.3 4.5  2.1 0.909a
Width of medial joint space (mm)
Preoperative 3.4  1.0 3.7  1.3 0.095a
Postoperative 3.7  1.2 3.8  1.3 0.943a
Joint line convergence angle (degrees)
Preoperative 3.3  1.2 3.0  1.6 0.072a
Postoperative 2.5  1.2 2.4  1.5 0.331a
Posterior tibial slope (degrees)
Preoperative 8.7  1.4 8.6  1.6 0.451
Postoperative 9.1  1.8 9.3  1.5 0.348
WBL ratio (percent)
Preoperative 21.9  9.7 22.2  12.6 0.928a

Downloaded by: University of Michigan. Copyrighted material.

Postoperative 64.1  10.6 62.0  11.6 0.192
Hip-knee-ankle angle (degrees)
Preoperative Varus 6.3  2.2 Varus 6.4  2.3 0.808a
Postoperative Valgus 1.9  1.2 Valgus 1.8  1.4 0.755a
Kellgren–Lawrence grade
Preoperative (grade 1/2/3/4) 0/0/16/9 0/0/25/7 0.243b
Postoperative (grade 1/2/3/4) 0/5/16/4 0/2/28/2 0.789b
ICRS grade of medial femoral condyle
Preoperative (grade 0/1/2/3/4) 0/0/4/9/12 0/1/8/12/14 0.426b
Postoperative (grade 0/1/2/3/4) 1/3/6/9/6 2/6/9/11/7 0.497b

Abbreviations: ICRS, International Cartilage Repair Society; WBL, weight-bearing line.

Note: the values are presented as mean  standard deviation.
a
Derived with Student’s t-test.
b
Derived by linear by linear association. The statistical significance was set at p < 0.05.

Table 4 Comparison of postoperative clinical results between repaired group and unrepaired group

Repaired group Unrepaired group p-Value

Flexion contracture (degrees) 1.1  2.1 1.0  1.6 0.976
Active flexion (degrees) 137.1  7.5 133.1  8.9 0.063
WOMAC scores Total 6.4  5.5 9.2  5.3 0.158
Pain 1.5  1.9 1.4  1.4 0.958
Stiffness 0.9  1.1 1.1  1.5 0.543
Function 4.0  3.4 6.6  4.0 0.317
KS Scores KS knee scores 46.6  11.9 53.2  18.2 0.278
KS function scores 91.8  7.1 89.0  9.9 0.511

Abbreviations: KS, Knee Society; WOMAC, Western Ontario and McMaster University.
Note: the values are presented as mean  standard deviation derived with Student’s t-test. The statistical significance was set at p < 0.05.

grade 3 were significant predictors of no healing of MMPHRT reported as possible factors leading to arthritic progres-
after OWHTO. Additionally, older age, preoperative large sion.12,15,23,39,47 All patients included in our study had
meniscus extrusion ratio, incomplete reduction of meniscal knee osteoarthritis of KL grade 3 or 4 combined with varus
extrusion, and loose healing after refixation were also alignment. Other prognostic factors, except for the varus

The Journal of Knee Surgery

 Efficacy and Outcomes of MMPHRT Repair Lee et al.

Table 5 Multivariate logistic regression models for the possible References

variables affecting the healing status of MMPHRT after OWHTO 1 Magnussen RA, Lustig S, Demey G, Neyret P, Servien E. The effect
of medial opening and lateral closing high tibial osteotomy on leg
OR (95% CI) p-Value length. Am J Sports Med 2011;39(09):1900–1905
2 Bito H, Takeuchi R, Kumagai K, et al. Opening wedge high tibial
No repair 36.35 (2.78–475.88) 0.006 osteotomy affects both the lateral patellar tilt and patellar height.
WBL ratio 24.51 (1.64–367.37) 0.021 Knee Surg Sports Traumatol Arthrosc 2010;18(07):955–960
Undercorrection (< 50%) 3 Osman WS, Yousef MG, El Gebeily MA, Metwaly RG. Tibial slope
and patellar height changes following high tibial osteotomy (a
Overcorrection (> 70%) 4.72 (0.55–40.16) 0.156 comparative study). Osteoarthritis Cartilage 2015;6(03):247–254
Preoperative Kellgren– 43.75 (3.78–506.57) 0.002 4 Yang JH, Lee SH, Nathawat KS, Jeon SH, Oh KJ. The effect of biplane
Lawrence grade 4 medial opening wedge high tibial osteotomy on patellofemoral
joint indices. Knee 2013;20(02):128–132
Preoperative meniscal 1.74 (0.22–13.75) 0.599
5 Bin SI, Kim JM, Shin SJ. Radial tears of the posterior horn of the
extrusion (> 3 mm)
medial meniscus. Arthroscopy 2004;20(04):373–378
Body mass index 2.05 (0.32–13.22) 0.451 6 Robertson DD, Armfield DR, Towers JD, Irrgang JJ, Maloney WJ,
(kg/m2)  25.0 Harner CD. Meniscal root injury and spontaneous osteonecrosis of
Age  60 (y) 5.46 (0.94–31.91) 0.059 the knee: an observation. J Bone Joint Surg Br 2009;91(02):
190–195
Sex (male) 0.21 (0.03–1.47) 0.116 7 Koenig JH, Ranawat AS, Umans HR, Difelice GS. Meniscal root tears:
diagnosis and treatment. Arthroscopy 2009;25(09):1025–1032
Abbreviations: CI, confidence interval; MMPHRT, medial meniscus
8 Nha KW, Lee YS, Hwang DH, et al. Second-look arthroscopic
posterior horn root tear; OR, odds ratio; OWHTO, open wedge high
findings after open-wedge high tibia osteotomy focusing on the

Downloaded by: University of Michigan. Copyrighted material.

tibial osteotomy; WBL, weight bearing line.
posterior root tears of the medial meniscus. Arthroscopy 2013;29
The statistical significance was set at p< 0.05. CI: confidence intervals.
(02):226–231
9 Kanamiya T, Naito M, Hara M, Yoshimura I. The influences of
alignment, could not be corrected by a surgeon at the time of biomechanical factors on cartilage regeneration after high tibial
surgery. Therefore, proper realignment of lower extremities osteotomy for knees with medial compartment osteoarthritis:
would be a better treatment for the patients with the clinical and arthroscopic observations. Arthroscopy 2002;18(07):
MMPHRT combined with osteoarthritis. 725–729
10 Allaire R, Muriuki M, Gilbertson L, Harner CD. Biomechanical
There are several limitations in this study. First, this study
consequences of a tear of the posterior root of the medial
was a retrospective case-control study in which both groups meniscus. Similar to total meniscectomy. J Bone Joint Surg Am
were divided over time. Therefore, a selection bias could have 2008;90(09):1922–1931
occurred. However, the influence of potential confounders was 11 Choi ES, Park SJ. Clinical evaluation of the root tear of the posterior
minimized because a single surgeon performed all surgeries in horn of the medial meniscus in total knee arthroplasty for
a single institution and his skill regarding the OWHTO and osteoarthritis. Knee Surg Relat Res 2015;27(02):90–94
12 Chung KS, Ha JK, Ra HJ, Kim JG. Prognostic factors in the midterm
arthroscopic procedure was already mature at the beginning of
results of pullout fixation for posterior root tears of the medial
this study. Second, the follow-up period was only 2 years after meniscus. Arthroscopy 2016;32(07):1319–1327
surgery. A long-term study is necessary to determine whether 13 Chung KS, Ha JK, Ra HJ, Kim JG. A meta-analysis of clinical and
healing of the MMPHRT affects the progression of arthritis and radiographic outcomes of posterior horn medial meniscus root
survival after OWHTO. Third, results of the MMPHRT repair repairs. Knee Surg Sports Traumatol Arthrosc 2016;24(05):
1455–1468
method in this study were not yet established when the
14 Kim JG, Lee YS, Bae TS, et al. Tibiofemoral contact mechanics
method was used alone, without OWHTO. Finally, it was not
following posterior root of medial meniscus tear, repair, menis-
possible to include all preoperative and postoperative variables cectomy, and allograft transplantation. Knee Surg Sports Traumatol
in the analysis for the multivariate analysis concerning healing Arthrosc 2013;21(09):2121–2125
status. We had to select and categorize some continuous 15 Ahn JH, Jeong HJ, Lee YS, et al. Comparison between conservative
variables, considering their clinical significance, to overcome treatment and arthroscopic pull-out repair of the medial menis-
cus root tear and analysis of prognostic factors for the determina-
the limitations of the statistical analysis method. We analyzed
tion of repair indication. Arch Orthop Trauma Surg 2015;135(09):
the categorical data with a backward stepwise procedure to 1265–1276
maintain the statistical adequacy of the regression analysis. 16 Chung KS, Ha JK, Yeom CH, et al. Comparison of clinical and
radiologic results between partial meniscectomy and refixation
of medial meniscus posterior root tears: a minimum 5-year
Conclusion follow-up. Arthroscopy 2015;31(10):1941–1950
17 Kim JH, Chung JH, Lee DH, Lee YS, Kim JR, Ryu KJ. Arthroscopic
Repair of the MMPHRT during OWHTO showed a superior
suture anchor repair versus pullout suture repair in posterior root
healing rate to the unrepaired MMPHRT. However, repair of tear of the medial meniscus: a prospective comparison study.
the MMPHRT was not related to the postoperative radiologic Arthroscopy 2011;27(12):1644–1653
and clinical outcomes. Therefore, there is no clear evidence of 18 Jung YH, Choi NH, Oh JS, Victoroff BN. All-inside repair for a root
the need for the MMPHRT repair during OWHTO. tear of the medial meniscus using a suture anchor. Am J Sports
Med 2012;40(06):1406–1411
19 Cho JH, Song JG. Second-look arthroscopic assessment and clinical
Conflict of Interest
results of modified pull-out suture for posterior root tear of the
None declared. medial meniscus. Knee Surg Relat Res 2014;26(02):106–113

The Journal of Knee Surgery

Efficacy and Outcomes of MMPHRT Repair Lee et al.

20 M Thompson S, A Pinczewski L. A reduction method for all-inside 34 Harner CD, Mauro CS, Lesniak BP, Romanowski JR. Biomechanical
posterior horn meniscal repair. Arthrosc Tech 2015;4(05): consequences of a tear of the posterior root of the medial
e423–e424 meniscus. Surgical technique. J Bone Joint Surg Am 2009;91
21 Kodama Y, Furumatsu T, Fujii M, Tanaka T, Miyazawa S, Ozaki T. (Suppl 2):257–270
Pullout repair of a medial meniscus posterior root tear using a 35 Lerer DB, Umans HR, Hu MX, Jones MH. The role of meniscal root
FasT-Fix all-inside suture technique. Orthop Traumatol Surg Res pathology and radial meniscal tear in medial meniscal extrusion.
2016;102(07):951–954 Skeletal Radiol 2004;33(10):569–574
22 Chung KS, Ha JK, Ra HJ, Nam GW, Kim JG. Pullout fixation of 36 Krych AJ, Reardon PJ, Johnson NR, et al. Non-operative manage-
posterior medial meniscus root tears: correlation between ment of medial meniscus posterior horn root tears is associated
meniscus extrusion and midterm clinical results. Am J Sports with worsening arthritis and poor clinical outcome at 5-year
Med 2017;45(01):42–49 follow-up. Knee Surg Sports Traumatol Arthrosc 2017;25(02):
23 Moon HK, Koh YG, Kim YC, Park YS, Jo SB, Kwon SK. Prognostic 383–389
factors of arthroscopic pull-out repair for a posterior root tear of 37 Kim SB, Ha JK, Lee SW, et al. Medial meniscus root tear refixation:
the medial meniscus. Am J Sports Med 2012;40(05):1138–1143 comparison of clinical, radiologic, and arthroscopic findings with
24 Lee YG, Shim JC, Choi YS, Kim JG, Lee GJ, Kim HK. Magnetic medial meniscectomy. Arthroscopy 2011;27(03):346–354
resonance imaging findings of surgically proven medial meniscus 38 Han SB, Shetty GM, Lee DH, et al. Unfavorable results of partial
root tear: tear configuration and associated knee abnormalities. meniscectomy for complete posterior medial meniscus root tear
J Comput Assist Tomogr 2008;32(03):452–457 with early osteoarthritis: a 5- to 8-year follow-up study. Arthro-
25 Choi SH, Bae S, Ji SK, Chang MJ. The MRI findings of meniscal root scopy 2010;26(10):1326–1332
tear of the medial meniscus: emphasis on coronal, sagittal and 39 Kim SJ, Choi CH, Chun YM, et al. Relationship between preopera-
axial images. Knee Surg Sports Traumatol Arthrosc 2012;20(10): tive extrusion of the medial meniscus and surgical outcomes after
2098–2103 partial meniscectomy. Am J Sports Med 2017;45(08):1864–1871
26 Furumatsu T, Fujii M, Kodama Y, Ozaki T. A giraffe neck sign of the 40 Marzo JM, Gurske-DePerio J. Effects of medial meniscus posterior

Downloaded by: University of Michigan. Copyrighted material.

medial meniscus: A characteristic finding of the medial meniscus horn avulsion and repair on tibiofemoral contact area and peak
posterior root tear on magnetic resonance imaging. J Orthop Sci contact pressure with clinical implications. Am J Sports Med
2017;22(04):731–736 2009;37(01):124–129
27 Lee YS, Kang JY, Lee MC, Oh WS, Elazab A, Song MK. Effect of the 41 Ozkoc G, Circi E, Gonc U, Irgit K, Pourbagher A, Tandogan RN.
osteotomy length on the change of the posterior tibial slope with a Radial tears in the root of the posterior horn of the medial
simple distraction of the posterior gap in the uni- and biplanar open- meniscus. Knee Surg Sports Traumatol Arthrosc 2008;16(09):
wedge high tibial osteotomy. Arthroscopy 2016;32(02):263–271 849–854
28 Seo HS, Lee SC, Jung KA. Second-look arthroscopic findings after 42 Arnoczky SP, Warren RF. The microvasculature of the meniscus
repairs of posterior root tears of the medial meniscus. Am J Sports and its response to injury. An experimental study in the dog. Am J
Med 2011;39(01):99–107 Sports Med 1983;11(03):131–141
29 LaPrade RF, Ho CP, James E, Crespo B, LaPrade CM, Matheny LM. 43 Feucht MJ, Kühle J, Bode G, et al. Arthroscopic transtibial pullout
Diagnostic accuracy of 3.0 T magnetic resonance imaging for the repair for posterior medial meniscus root tears: a systematic
detection of meniscus posterior root pathology. Knee Surg Sports review of clinical, radiographic, and second-look arthroscopic
Traumatol Arthrosc 2015;23(01):152–157 results. Arthroscopy 2015;31(09):1808–1816
30 Lee DW, Kim MK, Jang HS, Ha JK, Kim JG. Clinical and radiologic 44 Bin SI, Jeong TW, Kim SJ, Lee DH. A new arthroscopic classification
evaluation of arthroscopic medial meniscus root tear refixation: of degenerative medial meniscus root tear that correlates with
comparison of the modified Mason-Allen stitch and simple meniscus extrusion on magnetic resonance imaging. Knee 2016;
stitches. Arthroscopy 2014;30(11):1439–1446 23(02):246–250
31 Chung KS, Ha JK, Ra HJ, Kim JG. Does release of the superficial 45 Lee DH, Lee BS, Kim JM, et al. Predictors of degenerative medial
medial collateral ligament result in clinically harmful effects after meniscus extrusion: radial component and knee osteoarthritis.
the fixation of medial meniscus posterior root tears? Arthroscopy Knee Surg Sports Traumatol Arthrosc 2011;19(02):222–229
2017;33(01):199–208 46 Lee JH, Lim YJ, Kim KB, Kim KH, Song JH. Arthroscopic pullout
32 El-Azab HM, Morgenstern M, Ahrens P, Schuster T, Imhoff AB, suture repair of posterior root tear of the medial meniscus:
Lorenz SG. Limb alignment after open-wedge high tibial osteot- radiographic and clinical results with a 2-year follow-up. Arthro-
omy and its effect on the clinical outcome. Orthopedics 2011;34 scopy 2009;25(09):951–958
(10):e622–e628 47 Kwak YH, Lee S, Lee MC, Han HS. Large meniscus extrusion ratio is
33 Park HJ, Kim SS, Lee SY, et al. Medial meniscal root tears and a poor prognostic factor of conservative treatment for medial
meniscal extrusion transverse length ratios on MRI. Br J Radiol meniscus posterior root tear. Knee Surg Sports Traumatol
2012;85(1019):e1032–e1037 Arthrosc 2018;26(03):781–786

The Journal of Knee Surgery