Skeletal Radiology

, Volume 40, Issue 2, pp 181–187

Arthroscopic all-inside meniscal repair—Does the meniscus heal?

A clinical and radiological follow-up examination to verify meniscal healing using a 3-T MRI

Authors

    • Department of Traumatology and Sports Injuries
  • Herbert Resch
    • Department of Traumatology and Sports Injuries
  • Rosemarie Forstner
    • Department of RadiologyUniversity Hospital of Salzburg
  • Mayer Michael
    • Department of Traumatology and Sports Injuries
  • Bernd Minnich
    • Department of Organismic BiologyUniversity of Salzburg
  • Mark Tauber
    • Department of Traumatology and Sports Injuries
Scientific Article

DOI: 10.1007/s00256-010-0965-6

Cite this article as:
Hoffelner, T., Resch, H., Forstner, R. et al. Skeletal Radiol (2011) 40: 181. doi:10.1007/s00256-010-0965-6

Abstract

Objective

The purpose of this study was to correlate clinical and radiological results using a 3-T MRI to verify meniscal healing after arthroscopic all-inside meniscus repair.

Materials and methods

We selected 27 patients (14 men and 13 women) with an average age of 31 ± 9 years and retrospective clinical examinations and radiological assessments using a 3-T MRI after all-inside arthroscopic meniscal repair were conducted. Repair of the medial meniscus was performed in 19 patients and of the lateral meniscus in eight. In 17 patients (63%), we performed concomitant anterior cruciate ligament reconstruction. The mean follow-up period was 4.5 ± 1.7 years. The Lysholm score and Tegner activity index were used for clinical evaluation. Four grades were used to classify the radiological signal alterations within the meniscus: central globular (grade 1); linear horizontal or band-like (grade 2); intrameniscal alterations and linear signal alterations communicating with the articular surface (grade 3); and complex tears (grade 4).

Results

At follow-up, the average Lysholm score was 76 ± 15 points, with ten of the patients placed in group 6 based on the Tegner activity index. MRI examinations revealed no signal alteration in three patients, grade 1 in 0, grade 2 in five, grade 3 in 13, and grade 4 in six. The MRI findings correlated positively with the clinical scores in 21 patients (78%).

Conclusions

Correlation of clinical and radiological examination was performed using 3-T MRI. In spite of satisfactory clinical outcomes at follow-up, a radiological signal alteration may still be visible on MRI, which was believed to be scar tissue, but could not be proven definitively. Imaging with a 3-Tesla MRI after meniscal suture surgery provides good but no definitive reliability on meniscus healing and therefore gives no advantage compared to 1.5-T MRI, with good clinical outcome using an all-inside arthroscopic meniscal repair.

Clinical relevance

3T-MRI can not substitute diagnostic arthroscopy in patients with persistent complaints after arthroscopic all-inside meniscal repair.

Level of evidence

Retrospective case series (EBM Level IV)

Keywords

3-T MRIArthroscopyMeniscal sutureAll-inside repair techniqueFasT-Fix Device

Introduction

The biomechanical and biological relevance of the meniscus for the knee joint is well documented. Loss of meniscal tissue induces an adverse process with accelerated development of degenerative osteochondral lesions [16]. However, it has been reported that successful meniscal repair leads to a lower incidence of degenerative changes over the long-term when compared to partial or total meniscectomy and nonoperative treatment of meniscal lesions [7]. For this reason, as much meniscal tissue as possible is now retained during surgery [8, 9].

MRI of the knee has proven to be an accurate and convenient diagnostic tool for identifying meniscal tears. MRI is gaining popularity especially in patients with suspicion of re-tear or re-injury. The accuracy of MRI in detecting meniscal tears has been reported at variable levels of sensitivity and specificity [10]. As described by Magee and Williams [11], there are few reports in the literature focusing on meniscal tears investigated by 3-T MRI, and there are currently no reports on 3-T MRI evaluation following meniscal sutures. When compared to other imaging techniques, 3-T MRI may offer more information on meniscal morphology due to higher spatial resolution and shorter acquisition time [12].

The objectives of this study were to verify whether postoperative intrameniscal changes in terms of healing can be demonstrated using a 3-T MRI imaging protocol and to correlate MRI findings with clinical outcomes according to the Lysholm score and Tegner activity index after arthroscopic all-inside meniscal repair.

Materials and methods

Between 2002 and 2007, 32 patients underwent arthroscopic meniscal repair at the authors' institution. Inclusion criteria for meniscal repair were defined as: (1) acute meniscal tear (surgery within 3 weeks of the trauma), (2) tear close to the meniscal rim, (3) tears with vertical tear pattern, and (4) age <45 years. There were five failed meniscal repairs (15.6%) that required a partial meniscectomy after an average time of 1 ± 0.8 years. These five patients were excluded from the study. To avoid interference from clinical impairments caused by cartilage or possible meniscus symptoms, patients with advanced chondropathy detected by 3-T MRI were excluded. According to a modified classification by Kornaat [13], 22 patients (81%) showed no cartilaginous defects, whereas five (19%) had a grade 1 chondropathy. Thus, all 27 patients were included in the study and received follow-up clinical and radiological examinations using 3-T MRI. The clinical evaluation was performed by an independent examiner who was not the surgeon and consisted of functional knee investigation and a questionnaire according to the scale of Lysholm. The Lysholm score represents a well-validated functional score designed for knee injuries [14]. The sports activity level was assigned to the scale of Tegner with regard to the knee function. This is a numeric scale ranging from 1 to 10, whereas 1 represents the lowest athletic level and 10 represents the highest one [15]. The study included 14 male and 13 female patients with an average age of 31 ± 9 years. The medial meniscus was involved in 19 patients and the lateral meniscus in eight. All patients underwent meniscal repair with the FasT-Fix repair system (Smith & Nephew Inc, Andover, MA). Twenty patients (74%) had concomitant anterior cruciate ligament (ACL) ruptures, of which 17 (63%) underwent secondary arthroscopic ACL reconstruction using hamstring tendons 8 weeks after meniscal repair. The average time period between arthroscopy and follow-up was 4.5 ± 1.7 years (range: 2–7 years).

Surgical technique

A standard arthroscopic approach was performed with a flexed knee joint. An internal arthroscopic suture system (FasT-Fix, Smith & Nephew) was used with an ipsilateral standard portal for tears in the medial horn or by way of a contralateral standard portal for tears in the lateral horn to restore form and function of the meniscus. Rasping of the tear margins and trephination of the meniscal wall in order to induce bleeding was done if possible. The suture loop was passed through the meniscal tear and knotted within the joint in terms of horizontal sutures. The number of sutures depended on the tear size and ranged from 1 to 3 (Fig. 1).
https://static-content.springer.com/image/art%3A10.1007%2Fs00256-010-0965-6/MediaObjects/256_2010_965_Fig1_HTML.jpg
Fig. 1

Intraoperative photograph after FasT-Fix meniscal Repair

Immediately following surgery, patients were allowed to actively move their knees within a non-painful range. Forced knee flexion was avoided, especially if there was any effusion or soft-tissue swelling, and progressive early motion was permitted if swelling decreased. Loaded knee flexion should not exceed 90° for the first 8 weeks. If simultaneous reconstruction of the ACL, using hamstrings, was performed, rehabilitation protocols followed the principles for meniscus reconstruction [16].

Clinical examinations

Clinical evaluation was performed by an independent examiner who was not the surgeon using the Lysholm knee score and the Tegner activity index. Range of motion in the sagittal plane and Lachman, Steinman, and McMurray tests were evaluated.

MRI evaluation

All patients underwent MRI of the knee with an eight-channel phased-array coil on a 3.0-T MRI system (Philips Achieva Release 2.5). Sagittal T1 (TR/TE 633/20), and PD FS (TR/TE 2902/30) were performed in the sagittal, coronal, and transaxial planes with slices 3 mm in thickness, a field of view (FOV) of 160 mm, an acquisition matrix of 320 × 252, and a reconstruction matrix of 512 × 512. MRI examination was performed by an independent radiologist who was blinded to patients' data and clinical findings at follow-up. Meniscal signal alterations were classified as defined by Stoller et al. and Mink et al. [17, 18]. Grade 0 was defined as a normal meniscus with a homogenous hypoechoic structure. Grades 1 and 2 describe meniscal degeneration and were characterized by meniscal signal alterations without reaching the articular surface. Grade 1 was defined as a central globular intrameniscal alteration. A grade 2 lesion was a linear horizontal, triangular, or band-like signal intensity that remained within the meniscus. Grade 3 and 4 alterations were meniscal tears communicating with an articular margin of the meniscal surface. Grade 3 represented a linear alteration continuous with the meniscal surface and in grade 4, multiple complex alterations were visualized. Grade 3 and 4 signal intensities in lesions were compared to the signal from joint fluid on T2-weighted images (WI). In previous studies using 1.5-T MRI, menisci were deemed to be healed if no intrameniscal fluid was demonstrated [19, 20]. Accordingly, signal alterations at the site of meniscal repair reaching the articular meniscal surface and displaying a signal as high as that of fluid within a joint were classified as a tear (Fig. 2). In contrast, low to intermediate signal intensity within the meniscus was regarded as scar tissue formation, as in the case of a healed suture repair (Fig. 3) [17, 18, 21].
https://static-content.springer.com/image/art%3A10.1007%2Fs00256-010-0965-6/MediaObjects/256_2010_965_Fig2_HTML.jpg
Fig. 2

Intrameniscal lesion displaying a high signal alteration corresponding to a clinical symptomatic re-tear

https://static-content.springer.com/image/art%3A10.1007%2Fs00256-010-0965-6/MediaObjects/256_2010_965_Fig3_HTML.jpg
Fig. 3

Intrameniscal lesion showing a low signal alteration corresponding to “scar” tissue formation and healing after meniscal suture repair

Statistical analysis

The MRI findings within our study were correlated with clinical symptoms. A correlation was deemed positive if the grading of the radiologist matched the clinical situation (Lysholm score) of the patient. Data were pooled based on Tegner’s score grading and meniscus rupture. Pooled data were analyzed using either a t test or a Mann–Whitney rank sum test (a KS-test if not normally distributed) at a significance level of 0.05 and a power of 0.80. The Pearson Product moment correlation and linear regression analysis were used to indicate the strength and direction of the linear relationship between the Tegner score, MR grading, and the Lysholm score. All statistical calculations were done using SigmaPlot© 11 (Jandel Scientific Inc, San Rafael, CA, USA).

Results

Clinical results

Of the 27 patients included in this study, 19 (70.4%) had no complaints at follow-up, which was rated as clinical success with no or minimal limitations in normal life. The Lachmann test was found to be positive in seven patients, without subjectively evident instability in normal life and the McMurray examination was positive in five. The average postoperative Lysholm Score at follow-up was 76 ± 15 points. According to the Tegner activity index, the plurality of patients were in group 6 (range from group 2 to group 6) which is assigned to recreational sports like tennis, badminton, jogging, etc. at least five times per week (Table 1). Thirteen patients received repair of the medial meniscus and six received lateral repair. The average Lysholm score in the 17 patients with ACL reconstruction was 81 ± 13 points, whereas the three patients who did not receive an ACL replacement, because of personal demand, had an average score of 50 ± 11 points, a statistically significant difference (p < 0.05). The seven patients with an isolated tear of the meniscus had an average Lysholm score of 85 ± 8 points with a subjectively satisfactory result in six of the patients. The sex of the patients did not influence the outcome. The patient data and the results of the clinical and radiological examinations are shown in Table 1.
Table 1

Patient data at follow up (ACL= anterior cruciate ligament)

Patient no.

Gender

Age at time of surgery (years)

Side

Grade

Radiological interpretation

Lysholm score

Tegner activity

ACL

1

M

41

Lateral

II

Suture without re-rupture

82

2

Surgery

2

M

24

Medial

III

Suture without re-rupture

82

6

Normal

3

W

17

Medial

IV

Suture without re-rupture

90

6

Normal

4

M

19

Lateral

II

Suture without re-rupture

90

6

Surgery

5

M

24

Lateral

0

no signal alteration

100

6

Normal

6

W

45

Lateral

II

Suture without re-rupture

83

6

Normal

7

W

29

Medial

III

Suture without re-rupture

89

4

Surgery

8

M

23

Medial

III

Suture without re-rupture

86

3

Surgery

9

W

44

Medial

0

no signal alteration

79

3

Surgery

10

W

38

Medial

IV

Suture without re-rupture

85

4

Surgery

11

M

34

Medial

II

Suture without re-rupture

60

3

Surgery

12

W

33

Lateral

II

Suture without re-rupture

80

6

Surgery

13

W

45

Medial

III

Suture without re-rupture

81

3

Normal

14

M

24

Lateral

III

Suture without re-rupture

100

6

Surgery

15

M

25

Medial

III

Suture without re-rupture

82

6

Surgery

16

W

22

Medial

III

Suture without re-rupture

83

4

Surgery

17

M

30

Medial

III

Suture without re-rupture

65

2

none

18

M

32

Medial

IV

Suture without re-rupture

37

3

none

19

W

37

Medial

III

Suture without re-rupture

95

4

Surgery

20

W

37

Medial

III

Suture without re-rupture

85

3

Surgery

21

W

21

Medial

0

no signal alteration

49

2

none

22

M

20

Lateral

III

re-rupture

53

5

Surgery

23

W

45

Lateral

III

re-rupture

62

3

Surgery

24

M

34

Medial

IV

re-rupture

90

6

Normal

25

W

25

Medial

IV

re-rupture

72

2

Normal

26

M

24

Medial

III

re-rupture

100

6

Surgery

27

M

38

Medial

IV

re-rupture

70

2

Surgery

Imaging results

The locations of preoperative radiologically detected meniscal tears were as follows: anterior horn of the medial meniscus (n = 1), middle segment of the medial meniscus (n = 2), posterior horn of the medial meniscus (n = 16), anterior horn of the lateral meniscus (n = 1), middle segment of the lateral meniscus (n = 2), and posterior horn of the lateral meniscus (n = 5). Before surgery, all patients had at least a grade-3 lesion. At a mean follow-up of 4.5 ± 1.7 years, three patients showed no signal alteration on a 3-T MRI. Five had a grade 2 signal alteration, 13 a grade 3 and 6 a grade 4. In 13 of the 19 patients with grade 3 or 4 lesions, the signal alteration was of low or intermediate intensity and was interpreted to represent scar tissue. In total, in six patients, a re-rupture was detected, seen by fluid entering the meniscus creating an intrameniscal signal with high intensity, continuous with the base of the meniscus. In three patients, an additional meniscal cyst was identified. In ten patients a slight joint effusion was detected.

Correlation between clinical and imaging results

The MRI findings correlated with the Lysholm score-determined clinical outcomes in 78% of cases. Before determining differences in radiological examinations, all data were grouped according to the Tegner activity index (ranging from 2–6) as presenting with or without rupture after surgery; different groups were then compared according to their grade (ranging from 1–4) as rated by the radiologist. No significant differences were found between groups (Table 2).
Table 2

Radiological examination (grading 1–4)

Group 1

n1

Group 2

n2

p value

Tegner score 2–4, no rupture

14

Tegner score 5–6, no rupture

6

n.s. (0.341)

Tegner score 2–4, re-rupture

5

Tegner score 5–6, re-rupture

2

n.s. (0.857)

Tegner score 2–4, re-rupture

14

Tegner score 2–4, no re-rupture

5

n.s. (0.516)

Tegner score 5–6, re-rupture

2

Tegner score 5–6, no re-rupture

6

n.s. (0.143)

Tegner score 2–6, re-rupture

7

Tegner score 2–6, no re-rupture

20

n.s. ( 0.166)

Comparisons of the outcomes of radiological examinations (grading 1–4) between patients who underwent meniscal suture repair with and without rupture relative to their Tegner score rating; Mann-Whitney U-test; n.s. not significant

To determine differences in imaging examinations, all data were grouped according to the Tegner activity index (ranging from 2–6) as being with or without rupture after surgery, and different groups were then compared based on their Lysholm scoring. No significant differences were found between groups (Table 3).
Table 3

Clinical examination (Lysholm score)

Group 1

n1

Group 2

n2

p value

Tegner score 2–4, no rupture

14

Tegner score 5–6, re-rupture

6

n.s. (0.458)

Tegner score 2–4, re-rupture

5

Tegner score 5–6, re-rupture

2

n.s. (0.960)

Tegner score 2–4, re-rupture

14

Tegner score 2–4 no re-rupture

5

n.s. (0.138)

Tegner score 5–6, re-rupture

2

Tegner score 5–6, no re-rupture

6

n.s. (0.355)

Tegner score 2–6, re-rupture

7

Tegner score 2–6, no re-rupture

20

n.s. ( 0.135)

Comparisons of the outcomes of clinical examinations (Lysholm score) between patients who underwent meniscal suture repair with and without rupture relative to their Tegner score rating; Mann-Whitney U-test; n.s. not significant

No significant correlations were found in the regression and correlation analyses between radiological examination and Lysholm scoring in patients with and without rupture who underwent meniscal suture repair relative to their Tegner score rating. However, a positive correlation between the radiological examination (grading) and the clinical appearance (Lysholm score) of the patients was found (Table 4).
Table 4

Radiological examination vs. Lyholm score

Group

n

R

Tegner 1–6 (all)

27

0.0393

Tegner 1–3 (all)

14

0.113

Tegner 4–6 (all)

13

0.0858

Tegner 2–4, no rupture

14

0.00295

Tegner 5–6, no rupture

6

0.171

Tegner 2–6, no rupture

20

0.0143

Tegner 2–4, re-rupture

5

0.357

Tegner 2–6, re-rupture

7

0.487

Regression and correlation analyses between radiological examination (grading 1–4) and Lysholm scoring in patients who underwent meniscal suture repair with and without rupture relative to their Tegner score rating; R Pearson Product Moment Correlation coefficient

Discussion

Preservation of meniscal tissue, especially in younger active individuals, is of utmost importance when treating a meniscal tear. Meniscal repair procedures are associated with the most favorable outcomes for linear, vertical, or oblique tears in the periphery of the meniscus due to the proximity of the vascular supply. The anterior cruciate ligament (ACL) plays a crucial role in meniscal repair [22, 23]. In most cases concomitant to the meniscus lesion, an ACL replacement was performed. The meniscus is at lower risk if the ACL is original or has been reconstructed [24]. Besides increased stability, another major aspect of successful healing is believed to be due to factors including the acuteness of the injury and the presence of postoperative hemarthrosis. The hemarthrosis associated with ACL reconstruction provides a milieu filled with chemotactic and growth factors essential for meniscal healing [24, 25].

The FasT-Fix System represents a second-generation, suture-based device incorporating all arthroscopically placed implants and attached repair suture. Some authors [25, 26] conclude that this suture system is a satisfactory technique, with assets and drawbacks, but has a final clinical success rate of 83% [26]. Nevertheless, Quinby et al. commented on a cadaveric evaluation of the two most commonly used commercially available devices including potential neurovascular risk from the depth of the needle-stick and entrapment of the popliteus tendon and superficial medial collateral ligament [27, 28]. However, we did not observe such complications in the present study.

A novel aspect of this study is the evaluation of meniscal suture outcomes with a 3-T MRI. Magee and Williams [29] were the first to describe the sensitivity and specificity of 3-T MRI compared to arthroscopy in order to detect meniscal tears. They concluded that 3-T MRI compares favorably with regard to sensitivity and specificity in studies using a 1.5-Tesla field strength or lower. The use of 3-T MRI resulted in more accurate and definitive diagnoses.

With regards to the results of von Engelhardt et al., which demonstrated a specificity of 95% and a positive predictive value of 87% of 3-T MRI, a grade 3 lesion has a high probability of corresponding to a meniscal tear during arthroscopy [30]. In our examinations, we had 19 patients with a grade 3 lesion or higher; in ten cases, it was classified as scar tissue. This led us to conclude that in spite of poor radiological grading, the clinical appearance of these patients may be satisfactory because of “functional” scar tissue. Clinical follow-up cannot differentiate between healed, partially healed, and asymptomatic failures. Even with such a high imaging quality, it is not possible to determine for sure if the tear has completely healed. The reason for false-positive and false-negative diagnoses may be due to MR interpretation error or due to the insensitivity of MR in detecting some tears. 3-T MRI following meniscal repair does not permit interpretive distinction between successful and ineffective meniscal healing; thus, repeat arthroscopic visualization remains the gold standard for definite assessment. Another group of authors, using either second-look arthroscopy or CT-arthrography, found that 66% of these failed repairs were clinically asymptomatic [23]. Some authors [31] continue to recommend arthrography for the detection of pre-op meniscal tears, and radial tears in particular. The cost is lower and the diagnostic procedure is quick and easy to perform. MR-arthrography appeared to be superior compared to conventional MRI [32] as already mentioned in the literature, but we preferred conventional MRI. Our retrospective MRI interpretations revealed that most sutures were completely healed after a mean time of 4.5 years, but we did not carry out a second-look arthroscopy or CT-arthrography. As indicated by clinical findings, all patients with MRI-healed meniscal repairs were also satisfied with the outcome. The clinical findings of these patients showed an average Lysholm score of 75. Our comparisons showed that there is no significant correlation between the clinical outcomes and the radiological grading, which means that patients with good clinical outcomes may have had a high radiological grade (3 and 4) and vice versa. We believe that in patients without symptoms, the meniscus is either histologically healed or acts as an autograft and fulfills the mechanical tasks of an injured, but “scar-tissue-healed”, meniscus. It is very difficult to distinguish between scar tissue and a completely healed meniscus on MRI examinations, as we have shown through our comparisons.

Limitations of this study include the retrospective study design. Furthermore, injuries to (or deficiencies of) the ACL are an important factor in meniscal healing. Most of our patients had a concomitant rupture of the ACL, which may have influenced the healing process. Even though failure rates may be higher in unstable knees, meniscus repair is not contraindicated in ACL-deficient knees [33]. Another limitation is that there is no true gold standard (i.e., second-look arthroscopy) for healing and especially there has been no study comparing the image quality of 1.5 to 3.0-Tesla MRI after meniscal suture repair. Compared to a T-MRI of 1.5 or less, a 3-T MRI has a much higher resolution [34]. This may be helpful in interpreting lesions more precisely, but we cannot definitively state that 1.5-T MRI would be insufficient to examine and interpret a sutured meniscus.

Conclusions

Our study showed that internal arthroscopic meniscal repair leads, in most cases, to the clinical healing of the meniscus based on clinical evaluation and patient satisfaction. However, a 3-T MRI does not allow a definitive statement regarding structural meniscus healing.

Copyright information

© ISS 2010