Article Text

Download PDFPDF

Elevation of systemic matrix metalloproteinases 2 and 7 and tissue inhibitor of metalloproteinase 2 in patients with a history of Achilles tendon rupture: pilot study
  1. B Pasternak,
  2. T Schepull,
  3. P Eliasson,
  4. P Aspenberg
  1. Orthopaedics and Sports Medicine, Department of Clinical and Experimental Medicine, Linköping University, Linköping, Sweden
  1. Correspondence to Björn Pasternak, Orthopaedics and Sports Medicine, Department of Clinical and Experimental Medicine, Linköping University, SE-58185 Linköping, Sweden; bjorn; pasternak{at}


Objectives In this study, serum levels of matrix metalloproteinases (MMPs) and tissue inhibitors of metalloproteinases (TIMPs) between patients with a history of Achilles tendon rupture and blood donor controls were compared, and their relation to mechanical properties of the tendons during healing were studied.

Methods More than 3 years after injury, serum levels of MMP-1, MMP-2, MMP-3, MMP-7, MMP-8, MMP-9 and MMP-13, TIMP-1 and TIMP-2 in eight patients who had Achilles tendon rupture were measured. Twelve blood donors served as controls. During the early phase of healing, the tendon modulus of elasticity was calculated from radiostereometric data and tendon cross-sectional area.

Results Patients with a history of Achilles tendon rupture had increased levels of MMP-2 (median difference 10%, p=0.01), MMP-7 (median difference 15%, p=0.02) and TIMP-2 (median difference 36%, p=0.02), compared with controls. Levels of MMP-7, measured 3 years after injury, correlated inversely to tendon modulus of elasticity (rs=20.83, p=0.02) and positively to tendon elongation (rs=0.74, p=0.05) during the early phase of healing. There was a trend towards positive correlation between MMP-7 and cross-sectional area during the early phase of healing (rs=0.67, p=0.08).

Conclusions Patients with a history of Achilles tendon rupture appear to have elevated levels of MMP-2, MMP-7 and TIMP-2 in serum. In these pilot data, the view that the MMP–TIMP system is involved in tendinopathy is supported and that disturbances in proteolytic control might be generalised are suggested.

View Full Text

Statistics from

Tendon rupture occurs in tendons altered by degeneration, which is characterised by disorganised connective tissue and variation in cell density, ranging from hypercellularity to hypocellularity.1 2 These changes are thought to precede rupture. In addition, degenerated tendons exhibit a different regulation profile of proteolytic enzymes and their inhibitors, and altered levels, quality and proportions of collagens and proteoglycans,1 3,,5 suggesting elevated tissue turnover.

The clinical outcome of tendon healing shows large variation and is difficult to predict.6 7 Rerupture of the Achilles tendon often occurs without obvious predisposing factors.8 The reasons for differences in outcome are largely unknown. Data from research in areas other than tendon suggest that alterations in the balance between the tissue degrading matrix metalloproteinases (MMPs) and tissue inhibitors of metalloproteinases (TIMPs) underlie complications after injury and surgery. Patients who develop non-union after long-bone fracture have aberrant levels of MMP-1, MMP-8 and TIMP-1.9 These changes were detected in serum, possibly suggesting an underlying systemic abnormality in proteolytic control. Moreover, locally elevated perioperative levels of several MMPs in the large bowel wall predict an increased rate of anastomotic wound failure after colonic resection.10 This suggests individual differences regarding proteolytic capacity of the extracellular matrix. Such differences could influence the susceptibility to tendon rupture.

We compared serum levels of MMPs and TIMPs between patients with a history of Achilles tendon rupture and controls. We also investigated whether MMPs and TIMPs might be related to early outcome of tendon healing. The blood samples were collected more than 3 years after injury to acquire presumed baseline MMP/TIMP values. Our primary hypotheses were that baseline MMP-1 to TIMP-1 ratio would differ between patients who had tendon rupture and controls and that baseline MMP-1 to TIMP-1 ratio would correlate inversely to mechanical properties of the tendons during healing.

Patients and methods

Eight patients who had participated in a prospective study concerned with biomechanical evaluation of Achilles tendon healing were recruited.11 There were two women and six men (age range 35–52 years). Inclusion criteria for the original study were total unilateral Achilles tendon rupture within the preceding 5 days and ages 18–65 years. Exclusion criteria were cardiovascular, lung, kidney or liver disease, diabetes mellitus, malignancy and rheumatoid arthritis. The patients were operated using the Kessler suture technique in local anaesthesia. Postoperatively, patients wore a plaster cast in equinus position for 3 weeks and were in a neutral position for another 3 weeks. After 6 weeks, patients were provided with special shoes with a 2-cm elevation of the heel, which they wore for another 6 weeks. Between 6 and 24 weeks postoperatively, the patients performed a standardised training program under the supervision of a physiotherapist.

Patients had their Achilles tendon rupture a median 37 (range 35–40) months before the collection of blood samples. Exclusion criteria for providing blood samples were surgery or bacterial infection within the preceding month. Control serum was obtained from 12 healthy blood donors, two women and ten men (age range 35–50 years). Control subjects' history concerning tendon disease was not known. The study was approved by the regional ethical committee in Linköping, and informed consent was obtained from all participants.

Measurements of tendon mechanical properties had been performed using radiostereometric analysis (RSA), which generated data for Young's modulus of elasticity. Briefly, small tantalum beads were implanted into the ruptured tendon ends at surgery and used as markers during RSA measurements. For RSA measurements, the patient sat with their leg horizontal and the foot placed in a frame. A pedal was applied to the forefoot and loaded with weights. The patient was asked to resist the dorsiflexing force to keep the foot in position. The force applied to the pedal was 25 and 200 N. The strain in the Achilles tendon, while resisting these forces, caused the distance between the implanted tantalum markers to change. This could be measured by RSA, which is a kind of three-dimensional x ray with high precision. Using the data for strain per load and cross-sectional area, as provided by ultrasonography, the modulus of elasticity could be calculated. These measurements were made at 6, 12 and 18 weeks after rupture, with the mean value of the three measurements used for analysis. The change in distance between the tantalum beads from 6 to 18 weeks was used as a measure of elongation of the healing tendon callus. Tendon cross-sectional area was determined by ultrasonography at the same time points; again the mean value of the three measurements was used for analyses. A detailed description of the measurement procedures is presented elsewhere.11

Serum was stored at −70°C until analysis. Levels of the collagenases (MMP-1, MMP-8 and MMP-13), gelatinases (MMP-2 and MMP-9), stromelysin 1 (MMP-3) and matrilysin 1 (MMP-7) were determined using Fluorokine Multi Analyte Profiling (F-MAP) kits (R&D Systems, Minneapolis, Minnesota, USA) in a Luminex 100 Bioanalyzer (Luminex, Austin, Texas, USA). These assays recognise proforms, active forms and TIMPcomplexed forms of the respective MMPs. There is a strong correlation between conventional ELISA and multiplex F-MAP measurements.12 MMP-13 could not be detected in any sample by the F-MAP assay at a detection limit of 0.052 ng ml−1. A conventional ELISA (GE Healthcare, Uppsala, Sweden) with a lower detection limit (0.032 ng ml−1) also failed to detect MMP- 13. TIMP-1 and TIMP2 were determined using sandwich ELISA kits (Quantikine; R&D Systems). The TIMP-1 assay recognises free TIMP-1, TIMP-1 bound to pro-MMP-9 and, to some extent, TIMP-1 bound to active MMPs. The TIMP-2 assay recognises free TIMP-2, TIMP-2 bound to pro-MMP-2 and TIMP-2 bound to active MMPs.

Groups were compared using two-tailed Mann–Whitney U tests. MMP-1 to TIMP-1 ratio in relation to the modulus of elasticity, cross-sectional area and elongation was evaluated calculating Spearman's correlation coefficient. Those MMPs and TIMPs that were significantly different between patients and controls were also correlated (Spearman's) to the three mechanical parameters. Approximate 95% CI for medians and for difference in medians were calculated by a non-parametric method.13


Serum levels of MMP-2, MMP-7 and TIMP-2 were elevated in patients who had Achilles tendon rupture more than 3 years ago, compared with controls (table 1).

Table 1

Serum MMPs and TIMPs in patients with a history of Achilles tendon rupture

There was an inverse correlation between MMP-7 and modulus of elasticity, a trend towards positive correlation between MMP-7 and cross-sectional area and positive correlation between MMP-7 and elongation (table 2). Further correlation analyses between MMP-7 and modulus of elasticity at the three different time points were performed for descriptive purposes (table 3).

Table 2

Correlation between serum MMPs/TIMPs at 3 years and early mechanical parameters of tendon healing

Table 3

Correlation between serum MMP-7 at 3 years and modulus of elasticity at three time points during tendon healing


We tested the hypothesis that an underlying deviation in the MMP–TIMP balance is associated with tendon rupture and with poorer outcome of tendon healing. There were no significant findings regarding the primary effect variable, but an exploratory analysis showed that MMP-2, MMP-7 and TIMP-2 were elevated in serum of the patients who previously had Achilles tendon rupture. These findings support the view that the MMP–TIMP system is involved in the aetiology of tendinopathy.1 Furthermore, they suggest that alterations in the MMP–TIMP system might be generalised. The inverse correlation between MMP-7 and modulus of elasticity and the positive correlation between MMP-7 and tendon elongation point to MMP-7 as a determinant of healing quality.

MMP-2 (gelatinase A) breaks down collagen fragments generated by collagenase activity and is also able to cleave intact fibrillar collagens I and III. MMP-7 (matrilysin 1) can degrade type I collagen, the main tendon collagen subtype. MMP-2 and MMP-7 have several additional properties including proteolysis of extracellular matrix components such as proteoglycans and fibronectin and ability to cleave several pro-MMPs to their active forms. Overproduction of MMP-2 and MMP-7 could thus lead to increased tendon tissue degradation through their own activity and through activation of other MMPs. TIMP-2 is a general endogenous inhibitor of MMPs, able to inhibit soluble as well as membrane-bound MMPs. It has several additional properties involved in cell function regulation. The elevation of TIMP-2 observed in our patients could represent a response to the elevated levels of MMPs or a constitutive overproduction of this inhibitor.

Patients who have Achilles tendon rupture are at excessive risk of a new rupture in the contralateral tendon, and there is increased family risk of rotator cuff injuries.14 15 Studies also show associations between polymorphisms in genes coding for connective tissue components and tendon rupture and tendinopathy.16 A polymorphism of the MMP-1 gene was found to be over-represented in patients with aseptic loosening of hip prostheses.17 Together, this suggests that there is a systemic component involving MMPs in the aetiology of tendon rupture and the variation in healing.

Our findings of elevated MMP-2, MMP-7 and TIMP-2 levels in patients with a history of tendon rupture contrast to a study based on tissue from acute Achilles tendon rupture reporting mRNA data. This showed downregulation of MMP-7 and TIMP-2 in acute rupture and no difference in MMP-2 levels between acute rupture and controls.5 On the other hand, another study showed elevated MMP-2 mRNA levels at the site of rupture, compared with a healthy portion of the Achilles tendon.18 It is, however, uncertain whether our results are comparable with acute rupture because the rupture itself is likely to induce transient changes in MMP mRNA levels.

What is already known on this topic

  • ▶. Tendinopathy is a major problem in recreational and professional athletes.

  • ▶. An altered regulation profile of several of the MMP enzymes and their inhibitors (TIMP) is documented in tendinopathy specimens.

  • ▶. There appear to be interindividual differences in proteolytic capacity.

This is a pilot study with a small sample size and multiple group comparisons. The results need confirmation in larger studies aimed at specific proteins, for example, MMP-7.

Blood samples were collected more than 3 years after tendon rupture with the intention of acquiring baseline levels. It is unlikely that the serum levels in the study patients reflect the healing process per se in view of previous studies, showing that systemic MMP-9 returns to baseline at 1 month after major orthopaedic and gastrointestinal surgery, whereas MMP-2 did not change postoperatively.19 Furthermore, studies of human Achilles tendon lesions created by collecting biopsies suggest that tendons have healed at 1 year after injury, as assessed by MRI.20 Additionally, in the biopsy material obtained at 2 years after harvest of the central third of the patellar tendon, there is no immunohistochemical staining for type III collagen, suggesting that there is no more active healing.21 We cannot rule out, however, that the observed differences might reflect ongoing tendinopathy or permanent changes induced by tendon rupture. To obtain true baseline values, prospective studies based on asymptomatic individuals who subsequently develop tendon rupture are needed.

There is some uncertainty to serum MMP measurement related to MMP release from leukocytes and platelets during serum preparation. Some argue that plasma should always be used for determination of MMP concentrations.22 23 However, although serum yields higher absolute values than plasma, there is a strong correlation between the two in levels of MMP-2, MMP-3, MMP-8 and MMP-9, but not in MMP-1.22

In conclusion, we show elevated levels of MMP-2, MMP-7 and TIMP-2 in serum of patients with a history of Achilles tendon rupture. Based on our findings, we hypothesise that a general disturbance in proteolytic control might render tendons prone to rupture, and we encourage further investigation whether general alterations in proteolytic capacity may contribute to tendon degeneration and tendon rupture.

What this study adds

  • ▶. Serum MMP-2, MMP-7 and TIMP-2 appear to be elevated in patients with a history of Achilles tendon rupture.

  • ▶. MMP-7 may be a determinant of tissue strength during healing of the Achilles tendon.

  • ▶. A general disturbance in proteolytic control might be involved in the pathogenesis of tendinopathy.


We thank Teréz Hanqvist, RN for collecting the blood samples.


View Abstract


  • Funding The strategic research programme Materials in Medicine in Linköping, the Swedish Research Council and Landstinget i Östergötland research funds.

  • Competing interests PA and BP are co-inventors of a patented method for the coating of sutures with pharmacological agents, for example, MMP inhibitors. TS and PE have no conflict of interest.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.