Achilles tendons can suffer tendinosis, a partial tear or complete rupture. The epidemiological literature suggests a steadily increasing incidence of Achilles tendon injuries for the past five decades, possibly related to the increased participation in recreational sports (14,21,25). Neither mechanical weakness after a large acute tear nor the risk of progression to complete Achilles tendon rupture nor the level of activity to be allowed without risking a rerupture after repair are known because the strength to failure of the Achilles tendon cannot be measured clinically (2,6,16,22,23,26,31,33-35). Similarly, reliable guiding information on the risk of rupture and guidelines for activity levels on the basis of clinical imaging are not available.
Ultrasonography (US) and magnetic resonance (MR) imaging are used to investigate Achilles tendinopathies (27,40). MR imaging predicted clinical outcome after surgical repair (12,34). US has shown promise, but MR remains more sensitive in detecting and monitoring patients with Achilles tendinosis (17,26). A third modality, calcaneal bone mineral densitometry (BMD), measured lower bone mineralization whether Achilles tendon rupture was treated surgically (1,38) or conservatively (24).
Arguably, the determination of mechanical properties of Achilles tendons should be a reliable tool to predict their risk of rupture but is not applicable clinically and animal models must be used. The purposes of this study were to 1) characterize the mechanical properties of Achilles tendons at various time points after creating a large central defect and 2) assess the ability of three noninvasive clinical imaging modalities (MR imaging, US, and BMD) to predict the mechanical properties of rabbit Achilles tendons after a large central defect. Our hypotheses were that 1) Achilles tendons will recover mechanically from a large defect within 8 wk and 2) the mechanical status of repair can be predicted by noninvasive ancillary investigation. We hoped to produce experimental data that can guide clinical management after large Achilles tears or operative repair (3-5,9,13,15,20,29,39).
The protocol for these experiments was approved by the University of Ottawa Animal Care Committee. This study adhered to animal care standards of the American College of Sports Medicine (ACSM).
Thirty-eight healthy adult female New Zealand White rabbits (Orytolagus cuniculus; Charles River Laboratories, Saint-Constant, Quebec, Canada) with an average weight of 4.5 kg (range, 4.0-4.8 kg) were studied.
Model of Full-Thickness Central Tendon Defect
After an intramuscular injection of ketamine (25 ng·kg−1), midazolam (2 mg·kg−1), and glycopyrrolate (0.1 mg·kg−1), surgery was performed under general anesthesia induced with isoflurane (2%-3%) and oxygen (1-2 L·min−1) administered using a face mask. Through a longitudinal posterolateral skin incision, the crural fascia and Achilles paratenon were opened. With the use of a custom-made Achilles tendon punch, a full-thickness anteroposterior oval defect 2 mm wide × 7 mm long, corresponding to approximately 50% of the tendon width, was created 20 mm proximal to the insertion on the Achilles tendon, so that the continuity of the tendon was preserved on both sides of the defect (Fig. 1). This site in the rabbit corresponds to the most common site of injury in humans (8). The paratenon and fascia were allowed to spontaneously cover the surgical site. A continuous skin suture was performed.
Postoperatively, the rabbits were housed in groups, allowed unrestricted activity, and provided with free access to food and water. For pain control, a fentanyl patch (Duragesic 25; Janssen-Ortho, Inc., Markham, ON, Canada) was applied the day before surgery; the patch was removed 4 d postoperatively. In addition, buprenorphine (0.02 mg·kg−1) was given subcutaneously until postoperative day 3.
The rabbits were divided into two studies. In the interval study, groupings of 10 animals were killed immediately after surgery and at 4 and 8 wk (n = 30). These durations are commonly used clinically for reassessments and to decide on a return to activities, work, or sport. In the longitudinal study, one rabbit was killed immediately after surgery and at 1, 2, 4, 6, 8, 10, and 16 wk (n = 8). The longitudinal study was conducted to assess any trends in mechanical recovery or imaging that may have been missed at the three time points in the interval study.
All animals were killed with an overdose of pentobarbital (100 mg·kg−1 given intravenously). After careful dissection, the calcaneus-Achilles tendon-gastrocnemius-soleus muscle complex of both legs was resected en bloc. Each specimen was wrapped in gauze moistened with a physiologic saline solution without fixation and stored at −13°C until testing. The specimens were thawed to room temperature, underwent MR, US, and BMD, and refrozen. They were thawed a second time before mechanical testing.
The tendons were aligned and imaged in groups of six in a custom-designed waterproof box (11). MR imaging was performed using a Symphony 1.5-T unit (Siemens Medical Systems, Iselin, NJ) and its extremity coil. The tendon was imaged with transverse T1-weighted spin echo (relaxation time (TR) = 400, excitation time (TE) = 14, number of excitation (NEX) = 2, field of view (FOV) = 130 mm, matrix = 576 × 576, slice thickness = 2 mm), transverse dual-echo T2-weighted turbo spin echo (TSE; TR = 4680, TE = 14/84, NEX = 2, FOV = 130 mm, matrix = 576 × 576, slice thickness = 2 mm), and sagittal proton density (PD) TSE (TR = 2720, TE = 20, NEX = 3, FOV = 130 mm, matrix = 512 × 440, slice thickness = 2 mm) sequences.
The maximal anteroposterior and mediolateral dimensions of the tendon 20 mm proximal to the calcaneal insertion were measured at a picture-archiving communication system workstation (Fig. 2). We then defined an oval region of interest in the transverse plane over the injury and measured the mean optical density (OD) for each of the T1-weighted, PD, and T2-weighted MR sequences. Transverse images 18 mm proximal to the tendon-calcaneus insertion were exported, and cross-sectional areas were calculated (ImageJ Image Processing Program, Freeware Version 1.32j). All measurements were performed by one investigator (G.D.) who was blinded to the identity of the specimen.
US images of the tendon were captured using an HDI 5000 (Philips Medical Systems, Markham, Ontario, Canada) by one experienced sonographer (J.D.) blinded to the identity of the specimen. The maximal anteroposterior diameter of the tendon was measured on transverse images obtained at 5-mm intervals with a compact linear array CL10-5 transducer (Fig. 3). Focal intratendinous echoic abnormalities were also sought using three sagittal images, at the medial, mid, and lateral aspects of the tendon with a linear array L12-5 50-mm transducer.
Calcaneal BMD was performed using a dual-energy x-ray absorptiometer (DPX-alpha; Lunar Corporation, Madison, WI). Each specimen was placed into a container filled with raw rice, simulating the natural environment of the soft tissue. The x-ray beam was directed from the dorsum to the plantar aspect of the calcaneus. The mineral density of the entire calcaneus was measured with software adapted for small bone (SmartScan Version 4.7e; Lunar Corporation). All measurements were performed by one experienced nuclear medicine technician (P.S.) blinded to the identity of the specimen.
Mechanical testing consisted of applying a tensile load exclusively to the tendon until rupture. We used a dual-cryogenic fixation assembly, which held the tendon securely at both ends and allowed rupture (30). First, the flexor digitorum superficialis tendon running within the epitenon was divided (11). Then, the tendon-calcaneus complex was secured into a normal saline-filled cryofixation assembly, and the proximal end of the tendon was secured into a second cryofixation assembly. The tendon was adjusted so that the portion to be tested spanned from 10 to 35 mm proximal to the calcaneal insertion, thereby centering the tendon defect. The saline in both cryofixation assemblies was frozen at −20 to −25°C by feeding liquid nitrogen into the double wall of the containers. Frozen saline solution ensured a secure grip on the tendons and excluded the tendon calcaneal insertion and myotendinous junction from mechanical testing. The cryogenic fixation containers were then mounted on a tensile testing machine (MTS Sintech 1G; MTS Systems Corporation, Eden Prairie, MN). We applied petroleum jelly to the tested portions of the tendons to prevent dehydration and thermal injury. A heater was centered to prevent freezing of the portion to be tested, and two cardboard insulators protected the ice surface from melting. This setup ensured a sharp temperature gradient between the frozen ends of the tendon and the portion to be tested, which was done at rabbit body temperature (37°C) (30). Two thermocouples (one in each cryofixation assembly) closely monitored the degree of freezing. For each sample, we conducted 10 consecutive preconditioning tests to a peak load of 100 N at a loading rate of 18 N·s−1.
Testing to failure was then conducted. Each tendon was loaded by elongating it at a displacement rate of 10 mm·s−1 until a 50% decrease in load was detected. We chose this displacement rate to simulate the high-velocity clinical conditions that lead to tendon rupture. Load and crosshead displacement data were recorded at 100 Hz, and a load-deformation curve was generated for each specimen using TestWorks 4 software (MTS Systems Corporation). Peak load was measured. Dividing the peak load at failure by the cross-sectional area gave the tendon stress value (N·mm−2; defined as the maximal tensile load that is sustained by cross-sectional unit of tendon). Finally, stiffness (N·mm−1) was calculated by fitting a linear regression line to the load-deformation data from 30% to 90% of the maximal peak load on the deformation curve.
We used SPSS 11.5.0 for Windows (SPSS, Inc., Chicago, IL) for constitution of the database and statistical analysis. In the interval study, we compared the data between the various time points and between the experimental and contralateral tendons using one-way ANOVA. We conducted post hoc analyses of statistically significant comparisons with multiple t-tests. In the longitudinal study, the data were described, and no statistical test was run.
In the interval study, we used Pearson's correlation coefficient to determine correlation between the clinical imaging results and mechanical testing results for the experimental and contralateral tendons. The data obtained immediately after surgery in the experimental tendons were excluded from the correlation with imaging because of the absence of a healing reaction. A P value of 0.05 or less was regarded as statistically significant. Results are expressed as mean and SD.
No animal was lost during the study. Two contralateral tendons obtained at 4 wk were excluded from all data analyses owing to technical difficulties at the time of mechanical testing. In the longitudinal study, the contralateral tendon was not collected immediately after surgery.
The mean anteroposterior and mediolateral diameters and the cross-sectional area of the experimental tendons 4 and 8 wk after surgery were greater than a) those of the contralateral tendons at the same time points, b) those of the contralateral tendons immediately after surgery, and c) those of the experimental tendons immediately after surgery (all P < 0.05; Figs. 4A and B).
The mean T1-weighted OD at 4 wk in the experimental group was significantly higher than the mean value for the contralateral tendons at 4 wk and for the experimental tendons immediately after surgery (both P < 0.05; Fig. 4C). At 8 wk, the mean T1-weighted OD was still higher for the experimental than for the contralateral tendons (P < 0.05). Similarly, the mean PD OD at 4 wk of the experimental tendons was higher than the mean value at 4 wk for the contralateral tendons, the experimental tendons immediately after surgery, and the experimental tendons at 8 wk (all P < 0.05; Fig. 4D). Similar differences in mean T2-weighted OD were observed between the experimental and contralateral tendons at both time points (Fig. 4E).
The mean anteroposterior diameter in the experimental tendons at 4 and 8 wk was greater than the corresponding values for the contralateral tendons, the contralateral tendons immediately after surgery, and the experimental tendons immediately after surgery (all P < 0.05; Fig. 4F). At 4 wk, a poorly defined hypoechoic area at the site of tear was identified in 3 (30%) of the 10 experimental tendons (Fig. 3); at 8 wk, the experimental tendons did not demonstrate any focal intratendinous echoic abnormality.
There was no difference in mean BMD value for the entire calcaneus between the experimental and contralateral legs at 4 or 8 wk (Fig. 4G).
Seventy-three tendons were tested; all failed at the tendon midsubstance and none failed at the clamp.
The mean peak load at failure immediately after surgery was lower than a) the mean value for the contralateral tendons, b) that of the experimental tendons at 4 wk, and c) that of the experimental tendons at 8 wk (all P < 0.05; Fig. 4H). There was no significant difference in mean peak load at failure between the experimental and contralateral tendons at 4 or 8 wk.
The mean stress value for the experimental group at 4 and 8 wk was significantly lower than that for the contralateral tendons (both P < 0.05; Fig. 4I). There were no statistically significant differences in mean tendon stiffness between the experimental and the contralateral tendons at any time point (data not shown).
On MR imaging, the anteroposterior diameter, mediolateral diameter, and cross-sectional area were higher 2 wk after surgery and remained elevated at all time points up to 16 wk (Fig. 4A). MR imaging T2-weighted signal abnormality returned to contralateral values by 4 wk, T1-weighted OD by 8 wk, and PD remained elevated to 16 wk (Figs. 4C-E).
On US, a well-defined hypoechoic lesion corresponding to the surgical site was identified immediately and 1 wk after surgery. A hypoechoic area was identified at the defect site at 2 and 4 wk, and from 6 to 16 wk, no focal intratendinous echoic abnormality was detected.
No trend in calcaneal BMD was observed (Fig. 4G).
The peak load tolerated by the tendons was lowest immediately after surgery and reached near contralateral values by 2 wk (Fig. 4H). Stress values were markedly lower for the experimental tendon than the contralateral tendon from 1 to 16 wk (Fig. 4I).
Correlation of Imaging Outcome Measures with Mechanical Properties
In the experimental group, high T1-weighted OD correlated with low peak load at failure (P < 0.05; Table 1). Both high T1-weighted OD and PD OD correlated with lower mechanical stress (P < 0.05). Finally, lower BMD of the calcanei correlated with lower stress (P < 0.05). In the contralateral tendons, none of the imaging outcome measures correlated with any of the mechanical variables.
The creation of a full-thickness large central defect in the rabbit Achilles tendon minimizes vascular injuries and inflammation and mimics more closely a full-thickness tendon tear or surgical repair than models using paratenon injections of tumor necrosis factor α (36), collagenases (10), prostaglandins (37), repetitive motion (3), crush injury (7), or immobilization (24). The acute injury, however, does not reproduce the clinical picture of chronic tendinopathy. The dual-cryogenic fixation assembly allowed testing mechanical properties of the isolated Achilles tendon, excluding interference by muscle, enthesis, or bone (30). This is clinically important given that Achilles tears and ruptures occur in midsubstance and not at the insertion of tendon into bone (32). With these devices, we measured mean contralateral, intact Achilles tendon peak load values 200%-300% higher than those previously reported (39). Using this model and methods, we characterized the physical and mechanical aspects of healing of Achilles tendons after a loss of 50% of their widths.
Achilles tendon healing was characterized by a marked increase in size that persisted throughout the study. Mechanically, Achilles tendons responded quickly to the tensile load demands after a large central defect. Peak load at failure, the best indicator of tendon strength had recovered by 4 wk. The longitudinal study indicated that peak load at failure returned to contralateral values by 2 wk. This is earlier than we had hypothesized. These results suggest that the healing/remodeling Achilles tendon adjusted quickly and fully to one maximal tensile mechanical demand. The novel quantification and reporting of decreased Achilles tendon stress in this model is physiologically interesting. Throughout the study, the tendons remained very weak per unit of cross-sectional area (stress). Four weeks after surgery, the combination of a reduction in stress of 205% and an increase in tendon thickness of 203% maintained the peak load at 99% of the contralateral value. These findings suggest that early after a large defect, the laying down of new extracellular matrix (collagen fibers in majority) of lower mechanical strength was offset by a larger number of them restoring peak load at failure.
In addition to measuring the mechanical properties of the achilles tendon (AT) after a large full-thickness defect, we assessed the ability of three diagnostic imaging tools to predict the mechanical properties. Four weeks after surgery, US and MR imaging variables were all abnormal. On MR imaging, the increased T1-weighted, T2-weighted and PD ODs reflected the healing process. The radiological hyperdensity testified to the abnormal internal structure of the tendons and correlated closely with the lower tendon stress. However, tendon stress is not as clinically important a mechanical outcome as is peak load at failure to predict the tendon strength and risk of rupture. In this study, a decrease in T1-weighted OD on repeat MR imaging indicated an increase in strength of the recovering tendon. No other ancillary test predicted the recovery of peak load to failure compared with the contralateral tendons. These results add to earlier results of Khan et al. (18) who had shown only moderate correlation between US or MR and clinical assessment.
How can these results help with clinical investigation and management? The clinician assessing a patient after an acute Achilles tendon injury wants to know whether a partial or complete tear has occurred, with the latter entailing a surgical option. Initially, US, which is fast, available, and cost-effective, can provide information similar to MR imaging. After complete rupture is ruled out, detecting altered size, abnormal OD on MR and echogenicity on US 4 wk or more after the injury may be indicative of lower tendon stress. Clinically, however, information on tendon strength is more meaningful than data on tendon stress. In this study, 4 wk after injury, larger and abnormal tendons on imaging were not weaker than the contralateral tendons. Considering the limitations of the current study, the managing clinician can interpret these findings as suggesting that, 4 wk after a large Achilles tendon injury or surgery, return to activities ought not be delayed based solely on abnormal imaging results. Also, such abnormal size and imaging 4 wk after an acute injury or surgery would not necessitate restricting physical activity to prevent rupture. These conclusions are different from the findings after hind limb immobilization where weaker tendons have undergone matrix changes undetectable by MR (28). Allowing weight bearing after surgery lesion led to stronger Achilles tendons than preventing weight bearing in the absence of a surgical trauma (28). The combination of immobilization and non-weight-bearing postoperatively remains to be studied with similar mechanical-radiological correlations.
In our study, BMD of the calcaneus, which is modulated by foot impact when weight bearing and by Achilles tendon pull, did not correlate with tendon strength. In this model, rabbits were allowed weight bearing postoperatively. This is different from the immobilized rabbit hind limb, prevented from weight bearing where calcaneus BMD correlated with mechanical weakness (28). Although BMD may be useful in predicting Achilles tendon strength after immobilization, the current study does not support the use of BMD for clinical management after Achilles tendon injury or surgical repair.
This study has some limitations. First, the model used does not reproduce the pathophysiology of chronic tendinopathy. An acute lesion was created in a normal tendon as opposed to a degenerated tendon. This could have led to different mechanical and imaging healing. Second, the methods used to mechanically test these tendons (tensile load to failure) differ from the postoperative situation where the animals were submitted to repetitive lower tensile loads which did not lead to failure. These two factors limit the generalization to the clinical setting. Third, the contralateral legs may have compensated for the injured side, possibly increasing their peak load at failure. Despite this compensation, the experimental group reached contralateral peak load levels by 4 wk. Fourth, the longitudinal study indicated that a more detailed evaluation of this injury model within 4 wk of surgery could further delineate clinically relevant relationships between mechanical strength and US and MR imaging findings. Finally, unlike the management of some Achilles partial tears in humans, we did not restrict the rabbits in weight bearing or in range of motion (19).
In conclusion, a decrease in stress characterized Achilles tendons' mechanical recovery in the first 8 wk after a large partial defect. Abnormal MR and BMD imaging findings were predictors of decreased stress. But, more importantly, the enlarged tendons had rapidly compensated for their decreased stress and thus recovering their tensile strength. Considering the limitations related to this model, these findings suggest that abnormal US or MR imaging findings 4 wk after a large partial defect should not prevent a progressive return to usual activities, sports, and work.
This study was supported by the Research Secretariat of the Workplace Safety and Insurance Board of Ontario and the Canadian Institutes of Health Research.
The authors thank Philippe Poitras for designing the tendon punch, Eric Murray and Alain Berthiaume for acquiring the MR data, Josianne Dodier for the US images, Philippe St-Laurent for the BMD calculations, Julie Courchesne for specimen preparation, Gloria Baker for scientific editing, and Elizabeth Coletta for the graphics.
No author has any financial affiliation that may be perceived to have biased this report. The results of the present study do not constitute endorsement by ACSM.
1. Alfredson H, Nordstrom P, Lorentzon R. Prolonged progressive calcaneal bone loss despite early weightbearing rehabilitation in patients surgically treated for Achilles tendinosis. Calcif Tissue Int
2. Alfredson H, Pietilä T, Jonsson P, Lorentzon R. Heavy-load eccentric calf muscle training for the treatment of chronic Achilles tendinosis. Am J Sports Med
3. Archambault JM, Hart DA, Herzog W. Response of rabbit Achilles' tendon to chronic repetitive loading. Connect Tissue Res
4. Backman C, Boquist L, Friden J, et al. Chronic Achilles' paratenonitis with tendinosis: an experimental model in the rabbit. J Orthop Res
5. Benazzo F, Stennardo G, Mosconi M, et al. Muscle transplant in the rabbit's Achilles' tendon. Med Sci Sports Med
6. Bhandari M, Guyatt GH, Siddiqui F, et al. Treatment of acute Achilles tendon ruptures: a systematic overview and metaanalysis. Clin Orthop
7. Constantinescu MA, Greenwald DP, Amarante MT, et al. Effects of laser versus scalpel tenolysis in the rabbit flexor tendon. Plastic Reconst Surg
8. Crenshaw AH, ed. Campbell's Operative Orthopedics
. 8th ed. St. Louis (MO): Mosby-Year Book; 1992. 1904 p.
9. Crum JA, LaPrade RF, Wentorf FA. The anatomy of the posterolateral aspect of the knee. J Orthop Res
10. Dahlgren LA, van der Meulen MC, Bertram JE, et al. Insulin-like growth factor-I improves cellular and molecular aspects of healing in a collagenase-induced model of flexor tendinitis. J Orthop Res
11. Doherty GP, Koike Y, Uhthoff HK, et al. Comparative anatomy of rabbit and human Achilles tendons with magnetic resonance and ultrasound imaging. Comp Med
12. Haims AH, Schweitzer ME, Patel RS, et al. MR imaging of the Achilles tendon: overlap of findings in symptomatic and asymptomatic individuals. Skeletal Radiol
13. Han S, Gemmell SJ, Helmer KG, et al. Changes in ADC caused by tensile loading of rabbit Achilles' tendon: evidence of water transport. J Magnet Res
14. Houshian S, Tscherning T, Riegels-Nielsen P. The epidemiology of Achilles tendon rupture in a Danish county. Injury
15. Jassem M, Rose AT, Meister K, et al. Biomechanical analysis of the effect of varying suture pitch in tendon graft fixation. Am J Sports Med
16. Karjalainen PT, Aronen HJ, Pihlajamaki HK, et al. Magnetic resonance imaging during healing of surgically repaired Achilles tendon ruptures. Am J Sports Med
17. Karjalainen PT, Soila K, Aronen HJ, et al. MR imaging of overuse injuries of the Achilles tendon. AJR Am J Roentgenol
18. Khan KM, Forster BB, Robinson J, et al. Are ultrasound and magnetic resonance imaging of value in assessment of Achilles tendon disorders? A two year prospective study. Br J Sports Med
19. Koike Y, Uhthoff HK, Ramachandran N, et al. Achilles tendinopathy. Crit Rev Phys Rehabil Med
20. Kuo PL, Li PC, Shun CT, et al. Strain measurement of rabbit Achilles' tendons by ultrasound. Ultra Med Biol
21. Leppilathi J, Puranen J, Orava S. Incidence of Achilles tendon rupture. Acta Orthop Scand
22. Lo IK, Kirkley A, Nonweiler B, et al. Operative versus nonoperative treatment of acute Achilles tendon ruptures: a quantitative review. Clin J Sport Med
23. Mafi N, Lorentzon R, Alfredson H. Superior short-term results with eccentric calf muscle training compared to concentric training in a randomized prospective multicenter study on patients with chronic Achilles tendinosis. Knee Surg Sports Traumatol Arthrosc
24. Matsumoto F, Trudel G, Uhthoff HK, et al. Mechanical effects of immobilization on the Achilles' tendon. Arch Phys Med Rehabil
25. Möller A, Åström M, Westlin NE. Increasing incidence of Achilles tendon rupture. Acta Orthop Scand
26. Nehrer S, Breitenseher M, Brodner W, et al. Clinical and sonographic evaluation of the risk of rupture in the Achilles tendon. Arch Orthop Trauma Surg
27. Neuhold A, Stiskal M, Kainberger F, et al. Degenerative Achilles tendon disease: assessment by magnetic resonance and ultrasonography. Eur J Radiol
28. Ohberg L, Lorentzon R, Alfredson H. Eccentric training in patients with chronic Achilles tendinosis: normalised tendon structure and decreased thickness at follow up. B J Sports Med
29. Pomeranz SJ. Foot and ankle. In: Pomeranz SJ, editor. Gamuts and Pearls in MRI and Orthopedics
. Cincinnati (OH): MRI-EFI Publications; 1997. 251 p.
30. Ramachandran N, Koike Y, Poitras P, et al. Dual cryogenic fixation for mechanical testing of soft musculoskeletal tissues. IEEE Trans Biomed Eng
31. Rees JD, Lichtwark GA, Wolman RL, Wilson AM. The mechanism for efficacy of eccentric loading in Achilles tendon injury; an in vivo
study in humans. Rheumatology
32. Rees JD, Wilson AM, Wolman RL. Current concepts in the management of tendon disorders. Rheumatology
33. Rettig AC, Liotta FJ, Klootwyk TE, et al. Potential risk of re-rupture in primary Achilles tendon repair in athletes younger than 30 years of age. Am J Sports Med
34. Shalabi A, Kristoffersen-Wiberg M, Aspelin P, et al. MR evaluation of chronic Achilles tendinosis. A longitudinal study of 15 patients preoperatively and two years postoperatively. Acta Radiol
35. Silbernagel KG, Thomeé R, Eriksson BI, Karlsson J. Continued sports activity, using a pain-monitoring model, during rehabilitation in patients with Achilles tendinopathy-a randomized controlled study. Am J Sports Med
36. Stone D, Green C, Rao U, et al. Cytokine-induced tendinitis: a preliminary study in rabbits. J Orthop Res
37. Sullo A, Maffulli N, Capasso G, et al. The effects of prolonged peritendinous administration of PGE1 to the rat Achilles tendon: a possible animal model of chronic Achilles tendinopathy. J Orthop Sci
38. Therbo M, Petersen MM, Nielsen PK, et al. Loss of bone mineral of the hip and proximal tibia following rupture of the Achilles tendon. Scand J Med Sci Sports
39. Thermann H, Frerichs O, Holch M, et al. Healing of the Achilles' tendon: an experimental study. Foot Ankle Int
40. Weinstabl R, Stiskal P, Neuhold A, et al. MR and ultrasound study of Achilles tendon injury. Unfallchirurgie