a Departamento de Ortopedia, Santa Casa de São Paulo, São Paulo, SP, Brasil
b Serviço de Ortopedia e Traumatologia, Hospital Israelita Albert Einstein, Universidade Federal de São Paulo (Unifesp), São Paulo, SP,Brasil
c Instituto de Ortopedia, Hospital das Clínicas, Faculdade de Medicina, Universidade de São Paulo (USP), São Paulo, SP, Brasil
d Escola Paulista de Medicina, Universidade Federal de São Paulo (Unifesp), São Paulo, SP, Brasil
Introduction
One of the most frequent causes of shoulder pain is degener-ative traumatic injuries to the rotator cuff, particularly in theavascular area of the tendon,1which may affect individuals inany age group and may become worse with aging and throughwork or recreational activities.
Approximately 54% of adults over the age of 60 yearspresent partial tearing or complete rupture of the rotator cuff,in comparison with only 4% of those aged between 40 and60 years.4Furthermore, around half of these patients do notpresent histories of trauma: this suggests that in these cases,the degeneration of the rotator cuff occurs gradually andresults in incomplete tearing and possibly complete rupture.This event may lead to loss of shoulder function to varyingdegrees.5,6It can also occur at different levels: shoulder pain,weakness in making arm abduction movements and loss ofmobility.
Under these conditions, the treatment options include con-servative measures (rest and avoidance of the causal factor),pharmacological measures (non-steroidal anti-inflammatorydrugs) and rehabilitation (physiotherapy), along with infiltra-tion using corticoids in the subacromial space (ICSS).
ICSS is used in the initial treatment of pathological condi-tions of the rotator cuff and may produce good results withregard to pain relief and gains in range of motion, proba-bly because of the anti-inflammatory and analgesic effectsof these medications,7,8resulting from the anti-inflammatoryeffect of the corticoids. Among the corticoids used, the oneseen most often is methylprednisolone, which is indicated asshort-term adjuvant therapy for this condition, particularlyfor relieving acute crises or avoiding exacerbation, because ofits solubility and short-term action, which causes less tissuedamage.
However, the current clinical recommendations for localuse of these medications for treating pathological conditionsof the rotator cuff are that three infiltrations over the courseof a year should be the maximum number, with a spacing of atleast three months between them.9This precaution is takenbecause of the side effects reported in the literature, such astendon atrophy, alterations to the healing process, structuralalterations to the collagen fibers and metabolic alterationsto collagen synthesis. These could lead to diminution of thebiomechanical properties of the tendon and even cause com-plete tendon rupture.
Few studies have examined the specific clinical effects ofcorticoids on the rotator cuff. The literature is basically com-posed of case reports and experimental studies on animalsusing the tendon of the sural triceps and the patellar ten-don. The lack of consensus and the lack of specific studiesevaluating the effect of ICSS on the tendon of the rotator cuffstrengthen the idea that there is a need to comprehend and/orexplain the mechanisms that are implicated in the effects ofcorticoids. Currently, corticoids are only used empirically totreat this condition.12The rotator cuff of rates is consideredto be a very useful in vivo model for studying the diseases ofthe rotator cuff,13although these studies have not yet eval-uated the degree of tendon resistance after clinical use ofICSS.
In this light, the present study had the primary aim of eval-uating the effect of methylprednisolone on the mechanicalresistance of the rotator cuff in rats, along with the secondaryobjective of evaluating the histological alterations relating to cellularity, vascularization and changes to collagen fiber pat-terns in the tendon of the rotator cuff in rats.
Materials and methods
Seventy-five adult male Wistar rats (Rattus norvegicus) ofaverage weight 300 g were used. They were kept under stan-dardized lighting conditions (dark-light cycles of 12 h each)and temperature conditions (22 ± 2?C), with free access towater and feed. The experiments were conducted in the lab-oratory of the surgical center of our university, between thehours of 08:00 and 17:00, after the animals had become accli-matized. The protocols were approved by our institution'sEthics Committee for Animal Use (no. 12.016.4.01.111).
The total sample size was subdivided as follows. A mini-mum group size of n = 6 was respected, which was consideredto be the smallest number sufficient for the statisticalanalyses proposed, with regard to the species selected, asdescribed in the literature on this field. The calculation forthe number per group was made in accordance with theformula: [(Z?: 2 + Z?)2p*(1 - p*)2]:2i; where p* = (p0 + p1)/2 andi (delta) = (p0 + p1). Four treatment groups were evaluated,which began with 18 animals each. From these groups, sixanimals were removed at the different observation times, i.e.respectively at the times of the first, second and third admin-istrations of this treatment. In addition to these animals, threerats were used initially, in a pilot study to standardize thetechnique for substance administration and to recognize theanimal's anatomy.
The animals were randomly divided between the groups,which received the following treatments: group 1 (sham),which did not undergo any intervention or receive any treat-ment over the course of the experimental period and served asa control group for the resistance analyses; group 2 (vehicle),which received treatment using the diluent of the corticoidtested, in accordance with the manufacturer's specification;and groups 3 and 4 (experimental), which received subacro-mial injections of two different doses of the corticoid tested,respectively.
The corticoid investigated was methylprednisolone at thedoses of 0.6 mg/kg (group 3) or 6 mg/kg (group 4). These doseswere selected based on a reference study14that used 0.6 mg/kgto observe deleterious effects from similar treatment on theperitenon of the sural triceps of rats. In addition, pharma-cokinetic issues were taken into consideration and the doseof 6 mg/kg was chosen with the aim of investigating possi-ble dose-effect correlations for the parameters observed inthe present study, as well as possible adverse effects fromits administration. The drug was administered in the sub-acromial space of the animals' right shoulder, at a volumeof 0.1 ml/100 g of the animal's weight. Three administrationswere performed. The time of application of the first dose wasconsidered to be time zero, followed by injections at the timesof 24 h and 7 days after the first administration.
For the procedure of subacromial injection, the animalsreceived anesthesia induced by means of an intraperitonealinjection of ketamine hydrochloride plus xylazine hydrochlo-ride (40-75 mg/kg of ketamine + 5-10 mg/kg of xylazine). Theintra-articular infiltrations were made via an anterior route, with the animal in dorsal decubitus on a table that was tilted at45?and the upper right limb rotated externally. Following this,topographically, the apex of the acromion and the humeralhead were located. The needle was introduced between thesetwo, along an oblique path going down from the horizontal,until it reached contact with the humeral head. It was thenwithdrawn slightly to enable administration of the volume.
Twenty-four hours after the first application (day 1), 7 dayslater (second application) or 14 days later (third application),counting from the start of the treatment with methylpred-nisolone, different groups of animals (n = 6) were sacrificedand dissected. The tendon of the rotator cuff was carefullyremoved in order to separate the tendon from the humerus.This was done by raising it from its humeral insertion andamputating it at its muscle-tendon junction, so as to form aunit that was then evaluated in accordance with the method-ology described below.
Mechanical resistance test
This test was conducted in accordance with what wasrecorded by other authors,15using a bench with a man-ual clamp. Its upper extremity was connected to a lineardynamometer and its lower extremity was connected to areceptacle. The entire apparatus was suspended at approx-imately one meter from the ground. For this biomechanicaltest, the force (P = m × g) exerted on the tendon was measuredin newtons (N). The tendon unit was dissected by means of anincision above the protuberance of the scapular spine, withaccess superiorly to the supraspinatus muscle, together withits tendon. This was used for the resistance tests and histologi-cal evaluation. The distal portion of the tendon was deinsertedfrom the greater tubercle of the humerus and tenotomy wasperformed close to the junction with the muscle. One extrem-ity of each unit was gripped by the bench clamp and the otherend was connected to the dynamometer. The receptacle atthe lower extremity was filled with distilled water at a con-stant flow rate. In this model, as the water flowed in, the forceexerted on the tendon increased, such that the correspondingforce on the dynamometer produced by mass of the receptaclefilled with water together with gravity could be read directlyfrom the scale, in newtons. The magnitude of the force at themoment when each animal's tendon snapped was recordedindividually.
Histological analysis
For the histological analysis, the separated tendon unit wasembedded in paraffin and sections of thickness five microns(5 ) were cut longitudinally and transversally. The slides wereviewed at magnifications of 100×, 200× and 400× after theyhad been stained using a preparation of hematoxylin-eosin(HE), in order to observe the alterations to the collagen fiberbundles and to the nuclei and cytoplasm of the fibrocytes andfibroblasts. The parameters of cellularity, collagen thickness,occurrences of edema and vascular proliferation were eval-uated and categorized as scores of 0, 1, 2 or 3, as describedin a previous study.15These parameters or histological indi-cators were examined in order to demonstrate the behaviorof the tendon after the interventions made in this study. All
the analyses were performed by a single pathologist, who wasunaware of the respective treatment groups of the animalsfrom which the samples were obtained.
The data are presented as the mean ± SEM (standard errorof the mean), for values obtained from ordinal variables (ten-sion in N, to evaluate the resistance of the tendon) or themedian for categorical variables (degree of inflammation). Theanalysis was performed using GraphPad Instat 4.0®, usingone or two-way ANOVA followed by the Kruskal-Wallis orBonferroni test, respectively. p-Values < 0.05 were taken to besignificant.
Results
The analysis on the resistance tests on the tendons obtainedfrom the animals in the different groups revealed that thetreatments on the animals using methylprednisolone at thetwo doses evaluated in this study did not alter the tension (inN) that was needed to break them when they were evaluatedwithin 24 h after the first administration (Table 1).
However, as can be seen in Table 1, at the times of sevenand 14 days after the first subacromial infiltration of corticoid,significant differences (p < 0.05) were observed in relation tothe control group. On the seventh day, the mean values for theforce applied to the tendons in the group treated with 6 mg/kgof the substance were lower than those obtained from animalsin the control group (difference: -6.4 N; 95% CI: -13.3 to 0.4).A similar effect was observed on day 14 for the two doses ofmethylprednisolone evaluated (0.6 mg/kg: difference: -6.5 N;95% CI: -14.2 to 0.4; and 6 mg/kg: difference: -6.0 N; 95% CI:-12.3 to 0.3).
Regarding the histological analysis comparing differentparameters (as scores) in the treated groups in relation to thecontrol group, the evaluation under the optical microscoperevealed significant alterations (p > 0.05) for the parame-ters evaluated, at the different observation times after thetreatment with methylprednisolone started, as presented inTable 2.
Discussion
This study demonstrated that there is a relationship betweensubacromial use of corticoids and reduction in mechanical resistance and in the histological parameters of the tendonof the rotator cuff. This strengthens the preliminary findingsfrom similar studies.
In relation to the mechanical resistance of the rotator cuff,alterations were observed starting from the third administra-tion (after a period of 14 days), for the dose of 0.6 mg/kg ofmethylprednisolone. For the higher corticoid dose (6 mg/kg),significant alterations were already observed from the secondadministration, after a period of 7 days.
Conservative treatment using subacromial injections ofcorticoids for managing rotator cuff injuries is a commonpractice within the field of medicine, but its side effects meanthat greater care is required when using this. Although severaltheories about the anti-inflammatory effect of this drug exist,the way in which the degeneration of the tendon tissue takesplace has not been clearly expressed.
In the present study, it could be seen that 24 h after sub-acromial application of corticoid, there was no significantalteration (p > 0.05) to the resistance between the groups stud-ied, independent of the dose used. Contrary results wereobserved in another study15that also evaluated the influenceof methylprednisolone on the tendon of the sural triceps ofrats. IN the latter study, it was noted that the force neededto break the tendon decreased within the first 24 h after firstadministration of the drug. This effect was maintained for atleast the next two weeks. This divergence between the twostudies was probably because in the present study, unlike theprevious one, healthy tendons without any injury prior to thetreatment were evaluated, given that fragmentation of the tis-sue might facilitate its rupture.
On the other hand, in the present study, from the timeof the second administration, i.e. 7 days after the start ofthe treatment, a reduction in the resistance of the tendon ofthe rotator cuff in the animals treated with the higher doseof methylprednisolone (6 mg/kg) was noted in relation to thecontrol group. A similar effect was recorded by other authors14in relation to the observation time of 7 days, although witha dose 10 times smaller (i.e. 0.6 mg/kg). Although the corti-coid doses applied in the present study were different to thoseevaluated in the previous study, this dose-effect relationshippresented greater deleterious effects on the tendon, as shownin the histological analysis performed in the present study(Table 2) and also in a previous study that used similar method-ology for analyzing the Achilles tendon in rats.
According to the present findings, the adverse effectsof corticoids seem to follow a dose-dependent relationship,given that although treating the animals with a dose of0.6 mg/kg promoted significant alterations with regard to theparameters of collagen presence and vascular proliferationafter the third administration (day 14) in relation to the shamgroup, histological alterations with the dose of 6 mg/kg hadalready started to be observed from the time of the secondadministration (day 7), regarding the fibrillar appearance andpresence of collagen.
In relation to the possible mechanisms through whichthese corticoid effects might be occurring, the histologicalalterations seen in the tendons of the animals treated withmethylprednisolone demonstrated possible degeneration ofthe tendon tissue caused by this substance. It has been sug-gested that after tissue damage, vascular proliferation maysubsequently occur, with the aim of supplying oxygen and cel-lular nutrition, in an attempt to repair the damaged tissue.16This notion is in line with our histological findings, in whichwe observed increased vascular proliferation on the 14th dayafter the start of the treatment, with the two corticoid dosesevaluated.
On the other hand, production of collagen by the extracel-lular matrix of the tendon is a source of resistance to tendontraction. Simply decreasing the quantity of these fibers mayimply diminished tendon strength.17Likewise, methylpred-nisolone was shown to produce a lower quantity of collagenand lower resistance of the tendon of the rot, subsequentto partial injury induced by means of an incision.18Thepresent study reinforces these findings, given that the pres-ence of collagen was affected by both doses, on days 7 and 14,respectively through treatment with the doses of 6 mg/kg and0.6 mg/kg.
Finally, in relation to cellularity, it is known that the tendonis a structure composed mainly of collagen fibers, with rela-tively low presence of cells, which are mainly fibroblasts.17Although these cells were not investigated in the presentstudy, the findings relating to the presence of collagen indi-rectly suggest to us that their presence at the sites of corticoidadministration increased, starting from the time of the sec-ond administration of the higher dose tested, given that thesecells are responsible for production of these fibers. This ideais corroborated by a study on the tendons of the semitendi-nosus muscle of humans, from which it was observed thatincreased cellularity might suggest that the tendon was moresusceptible to injury.
Conclusion
The analysis on the results obtained from the present studymade it possible to demonstrate a relationship between sub-acromial use of methylprednisolone and both diminishedmechanical resistance and histological alterations in the ten-don of the rotator cuff, at the doses and times evaluated in thepresent study.
Future studies may provide greater support regarding themechanisms implicated in this corticoid effect, which mayinfluence the treatment of musculoskeletal conditions duringdegenerative processes.
Conflicts of interest
The authors declare no conflicts of interest.
s1. Moore KL. Anatomia orientada para a prática clínica. 5th ed.Rio de Janeiro: Guanabara Koogan; 2007.2. Nové-Josserand L, Walch G, Adeleine P, Courpron P. Effect ofage on the natural history of the shoulder: a clinical andradiological study in the elderly. Rev Chir Orthop ReparatriceAppar Mot. 2005;91(6):508-14.3. White RH. Shoulder pain. West J Med. 1982;137(4):340-5.4. Bartolozzi A, Andreychik D, Ahmad S. Determinants ofoutcome in the treatment of rotator cuff disease. Clin OrthopRelat Res. 1994;(308):90-7.5. McLaughlin HL. Rupture of the rotator cuff. J Bone Joint SurgAm. 1962;44:979-83.6. McMahon PJ, Debski RE, Thompson WO, Warner JJ, Fu FH,Woo SL. Shoulder muscle forces and tendon excursionsduring glenohumeral abduction in the scapular plane. JShoulder Elbow Surg. 1995;4(3):199-208.7. Adebajo AO, Nash P, Hazleman BL. A prospective double blinddummy placebo controlled study comparing triamcinolonehexacetonide injection with oral diclofenac 50 mg TDS inpatients with rotator cuff tendinitis. J Rheumatol.1990;17(9):1207-10.8. Petri M, Dobrow R, Neiman R, Whiting-O'Keefe Q, SeamanWE. Randomized, double-blind, placebo-controlled study ofthe treatment of the painful shoulder. Arthritis Rheum.1987;30(9):1040-5.9. Lashgari CJ, Yamaguchi K. Natural history and nonsurgicaltreatment of rotator cuff disorders. In: Norris TR, editor.Orthopaedic knowledge update. Shoulder and elbow. 2nd ed.Rosemont: American Academy of Orthopaedic Surgeons;2002. p. 155-62.10. Hugate R, Pennypacker J, Saunders M, Juliano P. The effects ofintratendinous and retrocalcaneal intrabursal injections ofcorticosteroid on the biomechanical properties of rabbitAchilles tendons. J Bone Joint Surg Am. 2004;86(4):794-801.11. Wiggins ME, Fadale PD, Ehrlich MG, Walsh WR. Effects of localinjection of corticosteroids on the healing of ligaments. Afollow-up report. J Bone Joint Surg Am. 1995;77(11):1682-91.12. Wei AS, Callaci JJ, Juknelis D, Marra G, Tonino P, Freedman KB,et al. The effect of corticosteroid on collagen expression ininjured rotator cuff tendon. J Bone Joint Surg Am.2006;88(6):1331-8.13. Thomopoulos S, Soslowsky LJ, Flanagan CL, Tun S, Keefer CC,Mastaw J, et al. The effect of fibrin clot on healing ratsupraspinatus tendon defects. J Shoulder Elbow Surg.2002;11(3):239-47.14. Tatari H, Kosay C, Baran O, Ozcan O, Ozer E. Deleteriouseffects of local corticosteroid injections on the Achillestendon of rats. Arch Orthop Trauma Surg. 2001;121(6):333-7.15. Lech O, Severo AL, Silva LHP, Marcolan AM, Lütkemeyer E,Kim JH. Efeito do uso de corticoide em tendões previamentetraumatizados: estudo experimental. Rev Bras Ortop.1996;31(3):187-92.16. Chan BP, Chan KM, Maffulli N, Webb S, Lee KK. Effect of basicfibroblast growth factor. An in vitro study of tendon healing.Clin Orthop Relat Res. 1997;(342):239-47.17. Junqueira LCU, Carneiro J. Histologia básica. 11th ed. Rio deJaneiro: Guanabara Koogan; 2008. p. 103-23.18. Mikolyzk DK, Wei AS, Tonino P, Marra G, Williams DA, HimesRD. Effect of corticosteroids on the biomechanical strength ofrat rotator cuff tendon. J Bone Joint Surg. 2009;91(5):1172-80.19. Kartus J, Movin T, Papadogiannakis N, Christensen LR,Lindahl S, Karlsson J. A radiographic and histologic evaluationof the patellar tendon after harvesting its central third. Am JSports Med. 2000;28(2):218-26.
