Influence of a standardized closed soft tissue trauma on resistance to local infection. An experimental study in rats

Transcription

1 ELSEVIER Journal of Orthopaedi Researh 21 (2003) Journal of Orthopaedi Researh Influene of a standardized losed soft tissue trauma on resistane to loal infetion. An experimental study in rats Thomas Kalike a,b,*, Urs Shlegel, Gert Printzen, Erih Shneider, Gert Muhr a, Stephan Arens a Brrufsgenos.sensliuftlihe Klinikrn Bergrnannsheil, Chirnrgishe KIinik und Poliklinik, Uniaersitutsklinik. Burklr-le-la-Ca~np-Plats I, Bohum, Germany A 0-Researh Institute, Clavadelersrrassr, 7270 Davus, Switzerland Chernishrs Zentrallahor ier Univrrsitutsklinikrn. Insrlspitul, 3010 Bern, Switzerland Reeived 10 April 2001; aepted 6 July 2002 Abstrat Purpose: The etiology of loal posttraumati infetion in the loomotor system depends on the amount, virulene and pathogeniity of the inoulated miroorganisms and the loallsystemi host damage due to the type and extent of the aident or iatrogeni trauma. The relative effet of these fators remains unlear. In partiular, it is still unlear today whether-in presene of miroorganisms-soft tissue damage and its pathophysiologial onsequenes lead to infetion after soft tissue trauma, or whether the baterial ontamination is the primarily ause for posttraumati infetion. The aim of the projet was to gain information on the onsequenes of a soft tissue injury in terms of resistane to loal infetion. Sine linial populations are too heterogeneous, the problem was investigated in a standardized, redued (no surgery or implants) experimental in vivo model. Met/zod In female Sprague-Dawley-rats with a standardized losed soft tissue trauma to the tibialis anterior musle (group 1: n = 13) or without (group 11: n = 13), we ompared the inidene of loal infetion after a pairwise loal, perutaneously injeted baterial hallenge with various onentrations of Staphyloous aureus (2 x x lo6 olony forming units, CFU). The standardized losed soft tissue trauma was reated by appliation of a speially designed, omputer ontrolled impat devie. The ontaminated soft tissue and the underlying bone were removed under sterile onditions after five days and quantitatively evaluated for baterial growths. Infetion was defined as positive baterial growth at the soft tissue and/or bone. A stepwise experimental design with an up-and-down dosage tehnique was used to adjust the baterial hallenge in the area of the ID50 (50% infetion dose). Statistial evaluation of the differene between the infetion rates of both groups was performed by two-sided fisher exat test (p < 0.05). Results: The overall infetion rate was 46%. For the group with soft tissue trauma the ID50 was 1.32 x lo5 CFU and 1.05 x lo6 CFU for the group without soft tissue trauma. The infetion rate was 69% (9 of 13 animals) for the group with soft tissue trauma and 23%) (3 of 13 animals) for the group without soft tissue trauma. This differene is statistially signifiant 0, = 0.047). Conlusions: The infetion rate after a standardized losed soft tissue injury was signifiantly higher and the ID50 lower than without soft tissue trauma. Our results demonstrate that in presene of miroorganisms it is not primarily the baterial ontamination but rather the soft tissue damage and its pathophysiologial onsequenes resulting in dereased infetion resistane that seondarily lead to infetion Orthopaedi Researh Soiety. Published by Elsevier Siene Ltd. All rights reserved. Krjwwds; Closed soft tissue trauma; Infetion; Animal model; Rat * Corresponding author. Address: Berufsgenossenshaftlihe Kliniken Bergmannsheil, Chirurgishe Klinik und Poliklinik, Universitiitsklinik, Biirkle-de-la-Camp-Platz 1, Bohum, Germany. Tel.: , address: tkaelike@ompuserve.de (T. Kiilike). Introdution Aording to the urrent pathophysiologial model, the etiology of loal posttraumati infetion in the loomotor system depends on the amount, virulene and pathogeniity of the inoulated miroorganisms and on the loal/systemi host damage due to the type and extent of the aident or iatrogeni trauma [9,13, 19,31,361. Trauma-indued strutural and funtional /03/$ - see front matter Orthopaedi Researh Soiety, Published by Elsevier Siene Ltd. All rights reserved PlI: S (02)

2 3 74 T. Kalike et ul. I Journul of Orfliopaedi Researh 21 (2003) damage to the loal host tissue auses devasularization, malperfusion, disturbane of endothelial permeability, hypoxia, aidosis, hematoma, edema and inreased intraompartmental pressure [10,11,17,39]. In vivo mirosopy studies by Menth-Chiari et al. (1998) and Shaser et al. (1999) demonstrated that the damaging effets of losed soft tissue trauma result in a substantial and sustained impairment of nutritive perfusion, indiated by a signifiant redution in funtional apillary density [24,32]. This may result in an impaired humoral and ellular immune ompetene [15]. On a loal level it may derease resistane to the pathogeni mirobiologial load with subsequent manifestation of infetion in the traumatized tissue [3]. As early as 1976 Gustillo and Anderson demonstrated the effet of the ondition of the soft tissue as a major risk fator in posttraumati infetion in a linial study [12]. Consequently, there has been inreasing linial and sientifi interest in the effet of soft tissue injury on resistane to loal infetion over the last deade. In open soft tissu trauma, it seems obvious that the baterial load invading the injured soft tissues through the open skin barrier is a very important etiologial fator for infetion. However in losed trauma with unperforated skin there is no baterial ontamination of the injured soft tissues through an open wound. Nevertheless numerous linial investigations have shown that operative treatment of losed fratures with severe soft tissue injury is assoiated with a higher risk of infetion than is the ase for losed fratures without severe soft tissue injury [12,17,33,40]. These investigations indiate that--even without a major baterial ontamination-soft tissue damage and its pathophysiologial onsequenes at as a atalyst for infetion sine they redue resistane to infetion. The effet of an isolated soft tissue injury on the development of posttraumati infetion has not yet been evaluated under standardized, reproduible onditions despite the fat that the pathophysiologi hanges mentioned above indiate the fundamental effet on the loal resistane to infetion of soft tissue injuries. In linial populations the orrelation between soft tissue trauma and infetion annot be investigated under suffiiently standardized, reproduible onditions (e.g. major interobserver deviations in quantifiation of trauma [37], heterogeneity of trauma, pre- and omorbidity). Additionally, under linial onditions soft tissue trauma in the loomotor system is often ombined with fratures. Data in literature onerning the inidene of suh infetions mainly refer to fratures with soft tissue trauma. Therefore, we used the experimental model established by Shaser et al. (1999) to indue a standardized losed soft tissue trauma on the lower leg of the rat [32]. Additionally, we loally inoulated Stuphyloous aweus. The primary aim of the projet was to gain information on the effet of soft tissue injury on resis- tane to loal infetion. The seondary aim was to establish an infetion model after losed soft tissue trauma in the lower leg of the rat to evaluate further infetionrelated parameters. Materials and methods Experimentul model and unimuls In female Sprague-Dawley-rats with (group I: n = 13) or without (group 11: n = 13) a standardized losed soft tissue trauma to the tibialis anterior musle we ompared the inidene of loal infetion after a pairwise loal, perutaneously injeted baterial hallenge with various onentrations of S. uureus (2 x x lo6 olony forming units, CFU). The bateria were inoulated into the anterior tibial musle immediately after trauma to the soft tissues. We used a omputer-assisted, ontrolled impat tehnique powered by an eletri motor as used by Shaser et al. (1999) to produe the standardized losed soft tissue trauma [32]. An impulse of 0.28 kgmls to the anterior tibial musle of the lower leg was applied. The diameter of the piston was 10 mm. The right lower leg was shaved and positioned in a plasti mold to avoid passive movement and to ensure optimal energy transmission. Immediately after ontusion and injetion of the bateria we performed radiographi ontrols of the right tibia in two views to exlude fratures and preexisting pathologies in the tibia. The impulse was seleted so that it would not frature the bone beneath the soft tissues and would not ause ompartment syndrome. The proedure did not result in an open wound. The linially visible signs of the soft tissue injury were swelling of the soft tissues and bruising. The study was approved by the National Animal Protetion Authorities (number , Kantonales Veterinaramt Graubunden, Switzerland). Buterial inoulum We obtained a human-pathogeni beta-hemolysing phage-typed S. uureus strain (V ) from an infeted hip prosthesis and prepared a broth inoulum suspension as desribed by Melher et al. [22]. We used baterial onentrations of 2 x x lo6 CFU per 100 PI. Anesthesia The animals were anesthetized before appliation of the standardized losed soft tissue trauma and inoulation of S. uureus. We used isofluran in an indution hamber followed by intraperitoneal appliation of ml/kg midazolam/fentanyl/fluanison. Postoperative analgesia was performed with 0.2 mg/kg buprenorphine in a flavored gelatine base. Mirobiology After five days the animals were killed by COz inhalation in an indution hamber. The anterior tibial musle and the underlying tibia were removed under sterile onditions and quantitatively evaluated for baterial growth. Lysotyping of positive baterial growth was onduted aording to standard international phenotyping proedures (International bateriophage Set). Infetion was defined as positive baterial growth in the soft tissue and/or bone. Experimental proedure and statistis We applied a grouped sequential experimental proedure with an up-and-down dosage tehnique. As desribed in previous publiations this tehnique allows the determination of the level of baterial onentrations at whih differenes in the infetion rates of the ompared groups are most evident [1,2,4,5,14,22,23]. This onentration is lose to the level of the infetion dose of 50%1 (ID50) [30]. In this proedure, the baterial onentration is sequentially adapted to the predited ID50 in eah test phase, whereby a high baterial onentration was used initially in the first test phase in whih both

3 T. Kalike et al. I Journal of Orthopaedi Researh 21 (2003) % d.e 80% r % $ 40% E 20% 0% Phases Total I II 111 IV V VI n=26n=2 n=4 n=4 n=4 n=8 n=4 lnoulum 2x10~ i/i doses 1x106 - ~- 2/2-1/1 (incfu) 2x10~ - i/i 111-2/2-1x10~ /2 1/1 2x10~ - - 1/1 - Number of animals Fig. 1. Infetion rates for the group with soft tissue trauma and the group without soft tissue trauma in total and related to eah of six experimental phases are shown in the upper setion. The number of animals in total and in eah phase is noted at the top. Inoulum doses between 2 x lo4 CFU and 2 x lo6 CFU were used. Below the results from eah phase for either group, the inoulum doses and the number of animals used at these doses are indiated, refleting the sequential up-and-down dosage tehnique. the group with and without soft tissue trauma developed an infetion. In subsequent test phases the baterial onentration in both groups was gradually redued until negative test results were ahieved in both groups. Having established the upper and lower thresholds, approximately ID50 was applied for the remaining animals in the subsequent test phases. The experiment onsisted of six phases (Fig. 1). For a priori alulation of the neessary sample size based on a statistial power of 0.8 and a type-i-error probability ( < 0.05 we had to estimate the expeted effet size of the losed soft tissue trauma on infetion resistane ompared with the absene of suh a trauma. No suffiient data was available from the literature to support this assumption. However, we assumed that the effet of the soft tissue trauma would be more important than the effet of the material for frature fixation implants, for example, on infetion resistane. For this parameter we had previously demonstrated an infetion rate of 75% in rabbits for the titanium DCP versus 35% for the steel DCP by the same experimental proedure using approximately IDS0 [S]. Therefore, we predited an effet size (w) of 0.6, orresponding to infetion rates of 80% (SO% + 30%) for the group with a losed soft tissue trauma and 20% (50% - 30%) for the group without trauma. On this basis we alulated a neessary sample size of at least 12 animals in eah group in order to obtain the above-mentioned power of 0.8 and deided a priori to perform a final statistial evaluation to assess the differenes between the infetion rates of both groups for a total sample size of 26 animals (13 per group). The statistial evaluation was based on the twosided fisher exat test (p < 0.05 as level of signifiane) [8]. Results There were no ompliations related to the proedure or anesthesia. All animals reovered from the anesthesia within a few hours. None of the animals died or suffered from open wound infetion. Lysotyping of positive baterial growth onfirmed the presene of no germ other than the inoulated strain of S. aureus in all speimens. infetion rate related to inoulum doses (CFU) 100% 8.g 80% 0 60% % E - 20% 0% 2.x lo4 1 x lo5 zx lo5 1 x 106 2x 106 lnoulurn (CFU) Fig. 2. Infetion rate related to inoulum doses (CFU) showing the positive results as perentages of eah group. The numbers indiate the number of infeted animals and the total number of animals at eah inoulum dose. The overall infetion rate was 46%. For the group with soft tissue trauma the infetion rate was 69% (9 of 13 animals) and 23"/0 (3 of 13 animals) for the group without soft tissue trauma. This differene is statistially signifiant (p = 0.047) [8]. In eight out of 9 animals in group I with positive baterial growth the miroorganisms were found in the soft tissue and the bone. One animal had an isolated baterial growth in the soft tissue. In 1 of 3 animals in group I1 with positive baterial growth the miroorganisms were found in the soft tissue and the bone. The other 2 animals had an isolated baterial growth in the soft tissue. In both groups mirobiologial evaluation did not show speimens with isolated baterial growth in the bone only. The infetion rates, number of animals, and inoulum doses in the six experimental phases are shown in Fig. 1. Apart from Phase I, a higher rate of infetion assoiated with the soft tissue trauma was apparent in all phases of the experiment. Calulations based on the umulative infetion rates showed that the ID50 for the group with soft tissue trauma was 1.32 x lo5 CFU and for the group without soft tissue trauma 1.05 x lo6 CFU. Fig. 2 shows the obvious differene in the infetion rates for the groups with and without soft tissue trauma in the range of the ID50. Disussion On the one hand, many animal models exist for the experimental indution of a soft tissue trauma [6,7,16, 21,341. These studies demonstrate impairment of ontratile funtion and hanges in funtional reovery and histology following soft tissue trauma and test the effiay of immobilization and anti-inflammatory mediation. Soft tissue trauma has been indued in these models by insertion of foreps through a skin inision (rush injury), employment of a spring hammer and the use of the drop-mass or pneumati impat

4 3 76 T. Kalike et ul. I Journul of Orthopedi Researh 21 (2003) tehnique. In these studies, mild to severe soft tissue traumata were indued, but no progression was made from a trauma to an infetion model. On the other hand, numerous infetion models in animals have been desribed for the investigation of osteomyelitis, implant-related infetion, effiieny of antibioti therapy or other mirobiologial aspets [18,20,26,28,29,40], but these rarely fous on the relation between infetion and isolated soft tissue trauma. Wihmann et al. (1998) demonstrated the influene of a losed frature with only minimal soft tissue injury and an inisional soft tissue trauma (laparatomy) with severe soft tissue injury-both in ombination with hemorrhage-on systemi immune depression (lymphoyte funtion) in mie [38,39]. The losed frature with only minimal soft tissue injury did not derease the immune funtion in ontrast to the severe soft tissue injury [38,39]. Soballe et al. (1998) investigated soft tissue infetions following iatrogeni trauma and found that eletri auter lesions redued infetion resistane after baterial ontamination of laparatomy wounds [35]. In these studies, the exat standardization and reliable reprodution of the soft tissue damage seems questionable (losed frature ombined with an inisional soft tissue traumakauter lesion ombined with surgery). Aording to our own experimental experiene, infetion models reat very sensitively to even minor, unintended hanges of the applied trauma. Impreise standardization of this trauma may therefore lead to unreliable results. Establishing a ombined trauma-infetion model that redues all parameters with an influene on infetion exept the parameter under investigation while simultaneously fulfilling the onditions of a standardized, reproduible investigation is essential. Although in linial pratie severe soft tissue trauma generally ours in ombination with fratures, in our model neither frature of the underlying bone, nor additional surgial trauma to the soft tissue should be indued. Both an additionally inrease suseptibility to infetion and would inevitably make the experimental groups less homogenoeus [33,40]. For this reason we used the model established by Shaser et al. (1999) to indue soft tissue trauma on the lower leg of the rat, thus meeting the need for a standardized, reproduible investigation while simulating the highly relevant linial situation of high-energyheloity trauma to the extremities in an optimal way [32]. Relying on the grouped sequential method of baterial inoulation established in our previous experiments, we were able to establish a ombined trauma-infetion model to evaluate the effet of soft tissue trauma on loal resistane to infetion for the first time [1,2,4,5]. The results of our study show that loal resistane to infetion is signifiantly redued after soft tissue trauma demonstrated by signifiantly inreased infetion rates. The differene in infetion rates between the groups in our study was 46%. This differene was higher than the differene for various implant and iatrogeni fators obtained in our previous studies [l, 2,4,5]. Our results underline the importane of soft tissue trauma as one of the most important risk fators in the etiology of posttraumati infetion in the loomotor system. Our results demonstrate that it is not primarily the baterial ontamination but rather the soft tissue damage and its pathophysiologial onsequenes resulting in dereased infetion resistane that seondarily lead to infetion. Below a ritially high amount of miroorganisms the baterial ontamination is well tolerated without deteriorating to manifested infetion, as long as the soft tissues are healthy and their defene mehanisms undamaged. At first glane, this information does not seem to be of impressive originality-espeially for a linian, who is routinely onfronted with the major relevane of the soft tissue in ompliations involving infetion after loomotor trauma. Nevertheless, linial and experimental researh has foused on the redution of additional operation-related tissue trauma by developing new tissue preserving stabilization tehniques and implants. These tehnial innovations may ontribute to a deline in infetion rates after injuries with substantial soft tissue damage [1,17,25,27]. The aim of further linial and experimental studies should be to fous on understanding the natural pathophysiologial hanges after soft tissue trauma and to find a way to influene those mehanisms to minimize infetion rates. We believe that our experimental model on the lower leg of the rat is suitable for the evaluation of further open questions on the etiology of loal infetion and on the effet of different types and grades of soft tissue trauma, the effet of various bateriologial aspets and the dynamis of (anti-)infetious mehanisms whih beome ative between traumakontamination and the manifestation of infetion. Of high linial relevane might be the experimental investigation of the reovery time of traumatized soft tissue until the risk of infetion is diminished to the point where surgial proedures are no longer at inreased risk. Our infetion model in the rat may failitate the appliation of modern investigative methods (e.g. immunology/moleular biology). The rat is the standard animal for suh methods and, in ontrast to the rabbit, the neessary tools (primers, antibodies) for rats are easily available at relatively low ost. These investigative tools may play an important role in understanding the pathophysiologial hanges that take plae following soft tissue trauma. Based on the insight gained from this study that it is primarily soft tissue damage and its pathophysiologial onsequenes that lead to infetion as a result of redued infetion resistane and not only baterial ontamination, further studies should fous on the effiay of the treatment options after soft tissue injury.

5 T. Kiilike rt al. I Journal of Orthopaedi Reseurh 21 (2003) Aknowledgements This projet was supported by a grant from the researh ommission of the AOIASIF-Foundation, Switzerland. Although none of the authors have reeived benefits for personal or professional use from a ommerial party related diretly or indiretly to the subjet of this artile, benefits have been or will be reeived but are direted solely to a researh fund, foundation, eduational institution, or other non-profit institution with whih one or more of the authors is assoiated. Referenes Arens S, Eijer H, Shlegel U, Printzen G, Perren SM, Hansis M. Influene of the design for fixation implants on loal infetion. An experimental study of DC-Plates vs. Point-Contat-Fixators in rabbits. J Orthop Trauma 1999;13: Arens S, Hansis M, Shlegel U, Eijer H, Printzen G, Ziegler WJ, et al. Infetion after open redution and internal fixation with dynami ompression plates4linial and experimental data. Injury 1996;27(Suppl 3): Arens S. Hansis M, Siebert C. Die Bedeutung des Wirtsshadens bei der Entstehung postoperativer Wundinfektionen. Hyg Med 1994; 19: Arens S, Kraft C, Shlegel U, Printzen G, Perren SM, Hansis M. Suseptibility to loal infetion in biologial internal fixation. Experimental study of open vs. minimally invasive plate osteosynthesis (MIPO) in rabbits. Arh Orthop Trauma Surg 1999;119:82-S. Arens S, Shlegel U, Printzen G, Ziegler WJ, Perren SM, Hansis M. Influene of the materials for frature fixation implants on the development of loal infetion. An experimental study of steel versus titanium DC-plates in rabbits. J Bone Joint Surg 1996; 78B:647-5 I. Criso JJ, Jokl P, Heinen GT, Connell MD, Panjabi MM. A musle ontusion injury model. biomehanis, physiology, and histology. Am J Sports Med 1994;22: Fisher BD, Baraos VE, Reid DC. Effet of systemi inhibition of prostaglandin synthesis on musle protein balane after trauma in the rat. Can J Physiol Pharmaol 1991;69: Freedman KB, Bernstein J. Sample size and statistial power in linial orthopaedi researh. J Bone Joint Surg 1999;81A: Gristina AG. Biomaterial-entered infetion: mirobial adhesion versus tissue integration. Siene 1987:237: Gristina AG, Naylor PT, Myrvik QN. Mehanisms of musuloskeletal sepsis. Orthop Clin North Am 1991;22: Guillou PJ. Biologial variation in the development of sepsis after surgery or trauma. Lanet 1993;342: [12] Gustillo RB, Anderson JT. Prevention of infetion in the treatment of 1025 open fratures of long bones: retrospetive and prospetive analysis. J Bone Joint Surg 1976;58A: [I31 Hansis M. Pathophysiology of infetion--a theoretial approah. Injury 1996;27(3):5-8. [I41 Hauke C, Shlegel U, Melher GA, Printzen G, Perren SM. Einfluss des Implantatmaterials auf die loale Infektresistenz bei der Tibiamarknagelung. Eine experimentelle Vergleihsstudie am Kaninhen mit Marknageln aus rostfreiem Stahl und Reintitan. Swiss Surg 1996;l(Suppl 2):45. Hoh RC, Rodriguez R, Manning T, Bishop M, Mead P, Shoemaker WC, et al. Effets of aidental trauma on ytokine and endotoxin prodution. Crit Care Med 1993;21:839-4s. Jarvinen M, Sorvari T. Healing of a rush injury in rat striated musle. 1. Desription and testing of a new method of induing a standard injury to the alf musle. Ata Pathol Mirobiol Sand A 1975;83: Krettek C. Fraktur und Weihteilshaden. Chirurg 1998;69: Mader JT. Animal models of osteomyelitis. Am J Med 1985; 78: Mader JT, Shirtliff M, Calhoun JH. The host and the skeletal infetion: lassifiation and pathogenesis of aute baterial bone and joint sepsis. Bai!lieres Best Prat Res Clin Rheumatol l999;13( I): Mayberry-Carson KJ, Tober-Mayer B, Lambe DW, Costerton JW. Osteomyelitis experimentally indued with Baterroides tlietaitaomivon and Staphyloous epidermidis. Influene of a foreign-body implant. Clin Orthop 1992;280: MGeahie JK, Grounds MD. Initiation and duration of musle preursor repliation after mild and severe injury to skeletal musle of mie: an autoradiographi study. Cell Tissue Res 1987:248: Melher GA, Claudi B, Perren SM, Shlegel U, Muntzinger J, Printzen G. Influene of type of medullary nail on the development of loal infetion. J Bone Joint Surg 1994;76B: [23] Melher GA, Metzdorf A, Shlegel U, Ziegler WJ, Perren SM, Printzen G. Influene of reaming versus nonreaming in intramedullary nailing on loal infetion rate: experimental investigation in rabbits. J Trauma 1995;39: [24] Menth-Chiari WA, Curl WW, Rosenrane E, Smith TL. Contusion of skeletal musle inreases leuoyte-endothelial ell interations: an intravital-mirosopy study in rats. J Trauma 1998;45: [25] Nast-Kolb D, Betz A, Shweiberer L. Der Wandel in der Unfakhirurgie der letzten 10 Jahre-in Beitrag zur Infektionsprophylaxe. Chirurg 1991;62: [26] Naylor PT, Myrvik QN, Gristina A. Antibioti resistane of biomaterial-adherent oagulase-negative and oagulase-positive staphylooi. Clin Orthop 1990;261:12&33. [27] Ninkovi M, Shoeller T, Wehselberger G, Benedetto KP, Anderl H. Infektprophylaxe bei komplexen Extremitatentrauma durh sofortige definitive Rekonstruktion mittels freiem Gewebetransfer. Chirurg 1997;68: [28] Nordon CW, Gillespie WJ, Nade S. Infetions in bone and joints. Oxford: Blakwell Sientifi Publiations; [29] Petty W, Spanier S, Shuster JJ, Silverthorne C. The influene of skeletal implants on inidene of infetion. Experiments in a anine model. J Bone Joint Surg Am 1985;67: [30] Reed LJ, Munh H. A simple method of estimating fifty perent endpoints. Am J Hyg 1938;27: [31] Robson MC. Infetion in the surgial patient: an imbalane in the normal equilibrium. Clin Plast Surg 1979;6: [32] Shaser K, Vollmar B, Menger M, Shewior L, Kroppenstedt S, Rashke M, et al. In vivo analysis of miroirulation following losed soft-tissue injury. J Orthop Res 1999; 17: [33] Siebert CH, Arens S, Hansis M. The role of surgial trauma in the aetiology of postoperative wound infetion -quantifiation of surgery indued trauma. Hyg Med 1995;20: [34] Smith TL. Curl WW, Smith BP, Holden MB, Wise T, Marr A. et al. New skeletal musle model for the longitudinal study of alterations in miroirulation following ontusion and ryotherapy. Mirosurgery 1993;14:

6 378 T. Kalike et al. I Journal of Orthopaedi Researh 21 (2003) [35] Soballe PW, Nimbkar NV, Hayward I, Nielson TB, Druker WR. Eletri autery lowers the ontamination threshold for infetion of laparatomies. Am J Surg 1998;175: [36] Stone HH. Soft tissue infetions. Am Surg 2000;66(2): [37] Templemann DC, Gulli B, Tsukayama DT, Gustillo RB. Update on the management of open fratures of the tibia1 shaft. Clin Orthop 1998;350: [38] Wihmann MW, Remmers D, Ayala A, Chaudry IH. Der Beitrag von Weihteiltrauma undloder Knohenfraktur zur Immunde- pression nah hamorrhagishem Shok im Tierexperiment. Unfallhirurg 1998;101:3741. [39] Wihmann MW, Zellweger R, Williams C, Ayala A, DeMaso CM, Chaudry IH. Immune funtion is more ompromised following losed bone frature and hemorrhagi shok than hemorrhage alone. Arh Surg 1996;131: [40] Worlok P, Slak R, Harvey L, Mawhinney R. The prevention of infetion in open fratures: an experimental study of the effet of frature stability. Injury 1994;25:31-8.