osteomelitis

Diagnosis and Management ofChronic Infection

Abstract

High-energy penetrating extremity injuries are often associated withsevere open fractures that have varying degrees of soft-tissuecontamination and tenuous soft-tissue coverage. The result is arelatively high prevalence of chronic osteomyelitis compared withthat in civilian trauma patients. Diagnosing chronic osteomyelitisrequires a careful history and thorough physical and radiographicexaminations. Cross-sectional imaging can help delineate theextent of bony involvement, and scintigraphy can be used as adiagnostic tool and to gauge response to treatment. Clinical stagingalso directs surgical management. Adequacy of débridementremains the most important clinical predictor of success; thus,adopting an oncologic approach to complete (ie, wide) excision isimportant. Reconstruction can be safely performed by a variety ofmethods; however, proper staging and patient selection remaincritical to a successful outcome. Although systemic and depotdelivery of antibiotics plays a supporting role in the treatment ofchronic osteomyelitis, the ideal dosing regimens, and the durationof treatment, remain controversial.

Most commonly, the termchronic infection in combat-

related extremity trauma connotesosteomyelitis. The cause is usuallymultifactorial but stems from thehigh-energy nature of the initial in-jury. Severe open fractures, withvarying degrees of gross contamina-tion and tenuous soft-tissue enve-lopes, are commonplace. Through-out the phases of treatment,considerable efforts are directed to-ward limb salvage or preservation ofresidual limb length. Despite the ju-dicious use of internal and externalfixation in these patients, the preva-lence of osteomyelitis, chronic orotherwise, is relatively high com-pared with that in civilian traumapatients.1-3

Chronic infection negatively affectsseveral aspects of recovery in this

predominately young, active patientpopulation. Severe open fractures,already predisposed to delayed unionand nonunion in the absence of com-plication, become extremely difficultto treat, particularly in the presenceof infected internal fixation.2,4

Weight-bearing restrictions affectambulatory status, functional mobil-ity, and independence. In addition,serial débridement procedures, pro-longed hospitalization, and long-term antibiotic therapy are associ-ated with considerable expense,delays in rehabilitation, and loss ofproductivity.5 Chronic osteomyelitisdeserves many of the same clinicalconsiderations as malignant tumors.For example, an accurate tissue (al-beit microbiologic) diagnosis guidestreatment;6,7 staging carries prognos-tic value8 and guides treatment;6 and

Jonathan Agner Forsberg, MD

MAJ Benjamin Kyle Potter, MD

George Cierny III, MD

Lawrence Webb, MD

From the Department ofOrthopaedic Surgery, MemorialSloan-Kettering Cancer Center, NewYork, NY (Dr. Forsberg),Orthopaedic Surgery Service, WalterReed Army Medical Center,Washington, DC (Dr. Potter),REOrthopaedics, San Diego, CA(Dr. Cierny), and the GeorgiaOrthopaedic Trauma Institute,Macon, GA (Dr. Webb).

J Am Acad Orthop Surg 2011;19(suppl 1):S8-S19

Copyright 2011 by the AmericanAcademy of Orthopaedic Surgeons.

S8 Journal of the American Academy of Orthopaedic Surgeons

the treatment often involves com-plete (ie, wide) excision of the in-volved bone,6,9 followed by complexbony and soft-tissue reconstruction.10

Diagnosis

Diagnosing chronic osteomyelitis re-quires a thorough evaluation. Acomprehensive physical examinationis necessary to identify any systemicmanifestations of infection, althoughthese findings are rare. Focusedphysical examination of the extremi-ties should also be performed, withan emphasis on the condition of thesoft tissues overlying the area of in-terest. Neurologic and vascular func-tion should be assessed and any de-formities (including limb lengthdiscrepancies) noted. A thorough so-cial, past medical, and surgical his-tory should be taken to identify anycomorbid conditions for the purposeof “describing the host”8 and assign-ing an accurate physiologic class (Athrough C).11

Imaging remains the cornerstone ofthe evaluation process.9 Radiographsprovide useful information in terms ofbone loss, deformity, and the type andnumber of implants. Cross-sectionalimaging can help identify the presenceof an abscess or sequestrum, as well asdelineate the extent of medullaryedema and, perhaps more importantly,the extent of cortical involvement.Contrast MRI studies should be ob-tained whenever possible, althoughthese studies demonstrate signal degra-dation in the presence of internaland/or external fixation, particularly

when intramedullary implants are pres-ent. For this reason, CT should be con-sidered in this setting. Indium 111–la-beled white blood cell scintigraphy ismore specific than three-phasetechnetium-99m bone scan in identify-ing infection12 and thus is routinelyused, not only to help diagnose andlocalize focal areas of osteomyelitisbut also to evaluate response totreatment and, ultimately, the eradi-cation of infection.

Laboratory evaluation, with theexception of microbiologic testing, isoften nonspecific. Leukocyte countwith differential is usually normal.7

The erythrocyte sedimentation rate,in contrast, is typically elevated butlacks specificity necessary to diag-nose extremity infections. C-reactiveprotein, procalcitonin and other in-flammatory cytokine levels are reli-ably elevated in the setting of acute,posttraumatic extremity infec-tions13,14 but have not been fully eval-uated in the setting of chronic osteo-myelitis. Liver and kidney functionas well as the patient’s human immu-nodeficiency virus status should beassessed. An accurate microbiologicassessment should be performed inall cases and include tissue and fluidcultures for aerobic, anaerobic acid-fast bacilli, and fungal organisms.Gram stain results at the time of ini-tial débridement/biopsy should alsobe recorded.

Classification

Chronic osteomyelitis is a biofilm in-fection caused by complex colonies

of phenotypically diverse microor-ganisms propagating freely within amicrobial-based, polysaccharide ma-trix that provides an immunity tohost defenses and systemic concen-trations of antimicrobial agents.15,16

Once the biofilm bacteria establishmacromolecular attachments to ex-posed, nonviable surfaces within thewound (ie, tissue, implants, foreignbodies), there is no way to eradicatethe disease shy of either killing thehost or physically removing the in-flammatory nidus, including the col-ony and all substrate attachments.17

Thorough surgical débridement, acompetent host response, andpathogen-specific antimicrobial cov-erage are, therefore, the essentialcomponents of treatment protocols:the débridement removes the biofilmburden while antibiotics rid the hostof residual bacterial phenotypes,leaving the host free to heal a live,clean, manageable wound.

Published in 1985, the Cierny-Mader classification of adult osteo-myelitis11 is the first system to articu-late treatment with the naturalhistory of the disease.6,18 In this sys-tem, the biofilm nidus is character-ized by one of four anatomic typeswhose complexity and associatedrisk for treatment failure escalate nu-merically (Figure 1). In type I, med-ullary osteomyelitis (Figure 2), thenidus is endosteal whereas, in type II,superficial osteomyelitis (Figure 3), itis confined to an outer surface ofbone that remains exposed, usuallyunprotected by a refractory, soft-tissue deficit. Localized osteomyeli-

Dr. Cierny or an immediate family member has stock or stock options held in Royer Biomedical and serves as a board member,owner, officer, or committee member of the Limb Lengthening and Reconstruction Society and the Musculoskeletal Infection Society.Dr. Webb or an immediate family member is a member of a speakers’ bureau or has made paid presentations on behalf of theMusculoskeletal Transplant Foundation; serves as a paid consultant to Zimmer; has received nonincome support (such as equipmentor services), commercially derived honoraria, or other non-research–related funding (such as paid travel) from Synthes, Smith &Nephew, Stryker, and Kinetic Concepts; and serves as a board member, owner, officer, or committee member of the OrthopaedicTrauma Association Southeastern Fracture Consortium. Neither of the following authors or any immediate family member hasreceived anything of value from or owns stock in a commercial company or institution related directly or indirectly to the subject ofthis article: Dr. Forsberg and Dr. Potter.

Jonathan Agner Forsberg, MD, et al

2011, Vol 19, Supplement 1 S9

tis, type III, is a well-marginated se-questration of an attached orfloating fragment of bone (Figure 4),often combining the features of bothtypes I and II osteomyelitis. In local-ized osteomyelitis, the entire niduscan be excised with a wide marginwithout causing the osseous segmentto become unstable. In type IV le-sions, diffuse osteomyelitis, the pro-

cess is permeative and involves a seg-ment of bone and/or an entire joint(Figure 5), and it often exhibits char-acteristics of types I, II and III. TypeIV lesions are mechanically unstableeither before and/or after a completeand thorough débridement.

Examples for each anatomic typeinclude the following: type I, he-matogenous osteomyelitis, or an

infected fracture-union followingmedullary stabilization; type II, full-thickness wounds resulting frompressure or venous stasis ulcers, aninfected fracture-union with a soft-tissue deficit, or nonhealing Pap-ineau grafts;19 type III, infectedfracture-union with butterfly-fragment sequestration or previousplate fixation; and type IV, peripros-thetic infections, chronic septic ar-thritis, or infected nonunions.

In this same system, the patient-host is stratified with regard to his orher physiologic capacity to detour in-fection, withstand treatment, and/orbenefit from cure. A hosts arehealthy patients; B hosts are patientswho possess comorbidities known tohave a detrimental effect on woundhealing. The effect can be local (BL),systemic (BS), or combined (BL/S),depending on the circulatory, he-matopoietic, metabolic, immuno-logic, and nutritional status of thepatient. Finally, a C host is a patient,compromised or otherwise healthy,who qualifies only for palliative, notcurative, treatment (eg, patients whoderive no quality-of-life improve-ment with cure, the morbidity oftreatment is excessive, the prognosisis poor, and/or the patient’s coopera-tion is lacking). By combining theanatomic type with the host classifi-cation, a clinical stage is designated.For example, a diabetic patient (BS/Lhost) with an infected nonunion ofthe distal tibia (type IV osteomyeli-tis) is designated as a stage IV BS/Losteomyelitis.

Deriving an accurate clinical stageis important for many reasons. Itaids in planning the course of treat-ment, identifies progressive stages ofthe disease,6 and serves as a prognos-tic indicator.18 In addition, determi-nation of an accurate clinical stageguides the nature of the surgical pro-cedure20 and is reproducible, so itcan be used to compare outcomes inpatient and treatment cohorts.8,21

A graphic depiction of the four anatomic types of osteomyelitis, matched withtheir respective treatment formats and components of their reconstruction.(Adapted from Cierny G III: Chronic osteomyelitis: Results of treatment. InstrCourse Lect 1990;39:495-508.)

Figure 1

Diagnosis and Management of Chronic Infection


Management

Clinical staging begins with a reviewof the patient’s medical and surgicalhistory and ends with assignment ofthe treatment format (Table 1). Tojustify the morbidity and risk of limbsalvage, the expected outcome mustoffer a distinct advantage over am-putation or observation alone. Iftreatment for cure is contraindicatedor deemed to be excessive, the pa-

tient is classified as a C host, and heor she is offered palliation (eg,incision/drainage, oral antibiotics,ambulatory aides, pain medication).Amputation is indicated when limbsalvage and palliation are neithersafe nor feasible.

Amenable comorbidities are re-versed and/or minimized (host opti-mization) before surgical interven-tion. Thereafter, soft tissues areresected to supple, well-perfusedmargins. Bone is tangentially excised

until exposed surfaces bleed in a uni-form, haversian (cortical) or sinusoi-dal (cancellous) pattern (ie, the pa-prika sign).40 All foreign bodies andsurgical implants are removed, thewound is lavaged of debris, and thesurgical field is prepared for closure(double setup).

Multiple tissue samples, notwound swabs, are collected fromdeep wound surfaces (eg, loculatedfluids, reactive granulations, foreignbodies), and specimens are set up for

Type I osteomyelitis of the distal femur. A, Clinicalphotograph demonstrating draining fistula (arrow) at themedial knee in a man with an open fracture. He hadundergone intramedullary fixation of the femur morethan 20 years previously. B, Lateral radiographdemonstrating the healed fracture (white arrow) andpossible sequestra distally (black arrows). C, Axial T1-weighted postcontrast magnetic resonance image of thedistal femur. The solid arrow indicates the medial sinus.The dashed arrow indicates the medullary nidus.

Figure 2

Type II osteomyelitis of the distal femur. A, Clinicalphotograph showing two sinus tracts (black arrows) atthe lateral knee, near the tip of a partially failed free flap(white arrow), 2 years after resection and postoperativeradiation done for liposarcoma. B, AP radiograph withperiosteal new bone formation seen along the lateralfemur (arrow). C, A T1-weighted magnetic resonanceimage depicting no involvement of the medullarycontents (arrow), consistent with superficial (type II)osteomyelitis.

Figure 3



culture and sensitivity testing for aer-obic and anaerobic bacteria. Whenorthopaedic implants, foreign bod-ies, and/or sequestra are retrieved,quantitative cultures following soni-cation sequencing23 and/or quantita-tive polymerase chain reaction pyro-sequencing24 are performed toidentify the microbial populationswithin the biofilm colony. Frozen-

section biopsies confirm the presenceof inflammation, validate the needfor fungal and/or mycobacterial cul-ture setups in the laboratory, andusually rule out the presence ofpathologic states mimicking infec-tion (eg, neoplasms, pseudotumors,autoimmune disorders, dysplasias).

Following débridement, each recon-struction must take into consideration

(1) the advantages and disadvantagesof attempting to supplement the exist-ing soft-tissue envelope, (2) the me-chanical integrity of the remainingbony segments, and (3) how best tomanage residual dead space (Table 1:VI, A, 2; VI, B, 4;VIII, A, 3; and XI).Wound closure by any means is imper-ative when vital structures (eg, vessels,nerves, tendons) are exposed and/or

Type III osteomyelitis of the distal tibia. A, Scarcontracture, soft-tissue loss, and a draining sinus(arrow) in the distal leg of a 39-year-old diabetic whosustained an open tibial fracture 13 years beforepresentation. B, AP radiograph demonstrating healedfractures, a distal cortical defect (white arrow), andmedullary sequestration proximally (black arrow). C, Aninversion recovery magnetic resonance scan depictingthe cloaca (white arrow), a medullary nidus (dashedarrow), and a cortical sequestrum (asterisk).

Figure 4

Type IV osteomyelitis of the distal femur, knee joint, andproximal tibia. A and B, Two different patients, whoeach presented with a soft-tissue deficit (white arrows)at the knee and/or proximal leg associated with anunderlying, diffuse osteomyelitis. C, AP radiographillustrating deformity with nonunion of a tibial plateaufracture, antibiotic beads (black arrow), and clinicianmarkings at the sites of intended resection (whitearrows). D, Lateral radiograph of the second patientshowing air within the joint (white arrow) containing awell-fixed, long-stemmed revision total knee prosthesis.

Figure 5



when the reconstruction of choice (ie,surgical implants, allografts) requires aclean surgical field to succeed. Closureis safeguarded by the systemic admin-istration of pathogen-specific antibiot-ics and the elimination of dead spaceby either the apposition of viable tis-sues (eg, soft-tissue transpositionsand/or transfers; acute limb shortening)or the implantation of an antibiotic de-pot (eg, antibiotic beads, sponges, orspacers). In a closed wound, the highconcentrations of antibiotics created bya space-filling, high-dose antibioticdepot will eliminate all remaining phe-notypes from the biofilm colony. Fur-thermore, antibiotic-impregnated poly-methylmethacrylate (PMMA) beadsand/or spacers will maintain any work-able dead space needed for future use(known as the spacer effect).41 Fol-lowing wound healing and patientresuscitation, the depot is removed,the “space” reclaimed, and recon-struction performed as a clean surgi-cal procedure.21,31 When flaps failand/or circumstances preclude theiruse, bone transport,32-34 combinedmethods of limb shortening/lengthening,25 and negative pressure–assisted closures are valuable tools.42

A live, clean wound will heal bysecondary intention and thereby cir-cumvent the need for restoration ofthe soft-tissue envelope. For thesereasons, open reconstruction tech-niques, such as open cancellous bonegrafting,19 acute limb shortening,25

vascularized bone flaps, and openmethods of bone transport,43 can beused, but they have several potentialdisadvantages. The components usedin the reconstruction must also belive (Table 1: VI, A, 2 and XI); theopen wound has a prolonged expo-sure to contamination and/or super-infection; external devices are thefixation of choice; and the surfacearea for bone grafting is limited.Nevertheless, despite these limita-tions, open techniques can be usefulin carefully selected patients, particu-

larly when a salvage solution isneeded.

Depot and SystemicDelivery of Antibiotics

Local depot delivery of antibiotics toinfected wounds has become a criti-cal component of musculoskeletal in-fection management in the last twodecades. By this method, one canachieve local antibiotic concentra-tions several-fold greater than bothbacterial minimum inhibitory con-centrations and the levels safely at-tainable with systemic administra-tion, with negligible systemictoxicity. The elution rates and ulti-mate local concentrations of antibi-otics are dependent on the deliveryvehicle, surface area of the deliveryvehicle, type and concentration ofantibiotics, fluid presence and fluidturnover rate, time in vivo, and per-manence or bioabsorbable nature ofthe vehicle.44 Antibiotic-impregnatedPMMA is versatile and can be usedto make spacers for prosthetic jointresections or segmental bone defectsand beads to increase surface areaand resultant elution rate of antibiot-ics, as well as to coat intramedullaryimplants when needed to simultane-ously treat osteomyelitis and bonyinstability. These techniques havebeen widely adopted and haveproven utility in the management ofa variety of deep, fracture- andimplant-related infections.44-46 Sus-tained supratherapeutic local antimi-crobial concentrations exceeding 6weeks in duration are routinelyachievable with these delivery meth-ods.47 The so-called membrane tech-nique of PMMA spacer placement,followed by subsequent spacer re-moval and bone grafting, has dem-onstrated both good clinical successand the bioactivity of the biologicand osteoinductive layer that formsaround such spacers.46

More recently, several bioabsorb-able delivery vehicles have becomeavailable with favorable elutioncharacteristics and putatively similarefficacy. These delivery vehicles havethe added benefits of obviating theneed for removal and, in many cases,of being osteoconductive and/or os-teoinductive.26,48,49 Direct and mini-mal carrier application of local anti-biotics is now being investigated viaseveral modalities, although the du-ration of effect persistence will osten-sibly be decreased.50-54 Local adju-vants therefore represent anincreasingly critical component ofthe physician’s armamentarium inthe fight against chronic musculo-skeletal infections. As increasinglybiocompatible and bioabsorbabletechnologies continue to develop andgreater supporting evidence becomesavailable, a shift away from PMMA-based therapy appears both inevita-ble and advisable in the absence ofplanned secondary procedures for re-instrumentation, bone grafting, orsoft-tissue reconstruction. As infec-tion eradication rates improve, addi-tional focus may be warranted in as-sessing and minimizing local tissuetoxicity because of supratherapeuticantibiotic concentrations in search ofthe optimal balance between infec-tion eradication and osseous unionor soft-tissue healing.55,56

There is now evidence that localdepot antibiotic delivery has equiva-lent or better efficacy with regard toinfection prevention or eradicationthan does systemic therapy.57 How-ever, the efficacy of combining theseantibiotic treatment modalities ap-pears to be additive, if not synergis-tic;45 particularly in the setting ofchronic infection, primary treatmentsuccess rates are not yet highenough, and long-term recidivism istoo frequent, to routinely eschewsystemic in favor of local therapy. Ageneral guideline for the duration ofsystemic antimicrobial therapy is 6



Table 1

Treatment of Adult Chronic Osteomyelitisa

I Patient evaluationII Preoperative testing

A. Laboratory testing: full metabolic panel, CBC with differential, coagulation panel, UA, ESR, CRP level, colonizationtesting22

B. Diagnostics: vascular indices, ultrasonography, oxygen tensions (tc PO2)C. Radiography: plain radiography, MRI, CT, nuclear scan, PET, angiographyD. Tissue specimens: cultures, histology sections, PCR pyrosequencing23,24

III Clinical stagingA. Anatomic type: I, medullary; II, superficial; III, localized; IV, diffuseB. Physiologic class: A host, B host, C host

IV Treatment formatA. Limb salvage; amputation; palliation: C hosts; no treatment for cure

V Host optimizationA. Reverse amenable comorbidities

VI First surgeryA. One-stage treatment

1. Débridement, cultures, biopsy, antibiotics (all treatment formats)2. Dead space management (limb salvage, amputation)

a. Wound: secondary intention, primary vs delayed closureb. Bone: vascularized bone flaps, acute shortening25

c. Fixation: orthoses, external fixatorsd. Depots: antibiotic beads26-30

B. First stage of multistage treatments1. Débridement, cultures, biopsy, systemic antibiotics2. Double setup: change instruments; re-prep and redrape; new gowns, gloves21,31

3. Temporary fixation: external fixation, antibiotic-coated devices4. Dead space management (limb salvage, amputation)

a. Wound: secondary intention, primary vs delayed closureb. Bone: bone transport,32-38 vascularized bone flapsc. Fixation: orthoses, external fixators, devices (coated)b 36-38

d. Depots: antibiotic beads, antibiotic spacers38,39

VII Outpatient follow-upA. Wound surveillance; laboratory values: ESR, CRP; physical rehabilitation

VIII Second surgery (second stage)A. Definitive reconstruction

1. Prophylactic antibiotics, device removal, débridement, cultures (frozen biopsy negative—no inflammation)2. Double setup: change instruments; re-prep and redrape; new gowns, gloves3. Reconstruction

a. Wound: primary closureb. Bone: bone grafts, vascularized bone flaps, prosthetic jointsc. Fixation: orthoses, external or internal fixation, prosthetic jointsd. Depots: antibiotic beads, permanent spacers,18,36 devices (coated)b

B. Staged reconstruction no. 21. Prophylactic antibiotics, device removal, débridement, cultures (frozen biopsy positive—acute inflammation)2. VI B treatments vs amputation

IX Outpatient follow-upA. Wound surveillance; laboratory values: ESR, CRP; physical rehabilitation



weeks for most patients, with exten-sion to 3 months for patients withretained infected implants,58 al-though extended suppressive therapyto fracture union, followed by subse-quent implant removal, has been ad-vocated.59

High-volume, low-pressure irri-gation remains a critical compo-nent of the débridement process.Intermediate- and high-pressure la-vage systems are readily availableand widely used, but they decreasewound bioburden and contamina-tion at the expense of host tissuedamage, which may be responsiblefor the reported bacterial reboundphenomenon observed followingtheir use.60 The addition of deter-gents, antiseptics, or antimicrobialsto irrigant solutions has not consis-tently been demonstrated to improveoutcomes.61

Negative-pressure wound therapywith reticulated open-cell foamdressings (NPWT/ROCF) is an im-portant wound adjunct that in-creases patient comfort and care con-venience while improving localcirculation, accelerating granulation

tissue formation, and increasingedema clearance, but it only variablyaffects bacterial bioburden and maybe more effective at preventing thantreating infections.62 Silver nano-particulate-impregnated ROCF andinfusion of antiseptics have recentlybecome commonplace supplementsto NPWT, but to date, clinical evi-dence of efficacy is limited.

Future Directions

Part of the reason for the current dif-ficulty with antibiotic resistance isattributable to our misunderstandingof what an antibiotic is and what itimposes on the bacterial genome.This is exemplified by the declara-tion credited to the surgeon generalWilliam Stewart in 1967: “The timehas come to close the book on infec-tious diseases. We have basicallywiped out infection in the UnitedStates.”63 Antibiotics came to be(and, at the time of this writing, con-tinue to be) used as a feed additive insubtherapeutic doses in the dairy andlivestock industries. In fact, 70% of

all antibiotics used in the UnitedStates are administered in this fash-ion.64 What has been little appreci-ated is the resilience and adaptabilityof the bacterial genome, which were,from an evolutionary viewpoint,minimally thwarted by the wide-spread use of antibiotics. Currently,penicillin is as effective against acutehematogenous osteomyelitis as is aplacebo, whereas, shortly after its in-troduction in World War II, penicil-lin was nearly 100% curative of thisdisease.65 As Nobel laureate Chris-tian de Duve points out, “Given anadequate supply of nutrients, a singlebacterial cell can generate 280,000billion individuals (‘generations’) ina single day.”66 This rapid duplica-tion and exchange of bacterial ge-netic material represents a continu-ous probing of the environment bythe bacterial genome for selective ad-vantage by spontaneous mutations.In 70 years, penicillin has gone fromwonder drug to placebo. The mes-sage is clear: indiscriminate use ofantibiotics facilitates the emergenceof resistance.

To compound matters, the devel-

Table 1 (continued)

Treatment of Adult Chronic Osteomyelitisa

X Third surgery (third stage)A. Definitive reconstruction (frozen biopsy negative—no inflammation)

1. VIII A treatmentsB. Staged reconstruction no. 3 (frozen biopsy positive—acute inflammation)

1. VI B treatments vs amputationXI Fourth surgery (fourth stage)

A. Biologic reconstructionsc

1. No devices, no foreign bodiesXII Outpatient follow-up

A. Wound surveillance, laboratory values: ESR, CRP; physical rehabilitation

CBC = complete blood count, CRP = C-reactive protein, ESR = erythrocyte sedimentation rate, PCR = polymerase chain reaction,PET = positron emission tomography, UA = urinalysisa In the San Diego treatment algorithm for chronic osteomyelitis, patients entering a palliation protocol and/or not requiring reconstruction fol-lowing débridement exit after the initial surgery (VI, A, 2). Osseous reconstructions and the use of surgical implants are staged to follow inter-vals of local and systemic antibiotic therapy.b Devices (coated): a medullary rod, cortical plate or prosthesis coated with antibiotic-impregnated bone cementc Biologic reconstructions—eg, bypass synostosis, acute shortening/fusion, bone transport, resection arthroplasty.Adapted from Cierny G III, DiPasquale D: Adult osteomyelitis, in Cierny G III, McLaren AC, Wongworawat MD, eds: Orthopaedic KnowledgeUpdate: Musculoskeletal Infection. Rosemont, IL, American Academy of Orthopaedic Surgeons, 2009, pp 135-153.



opment of new antibiotics by thepharmaceutical industry has beenslow. No fundamentally new class ofantibiotics has been brought to mar-ket since the 1970s. Thus, cliniciansneed to be resourceful in the use ofthe tools available. Buchholz andEngelbrecht67 as well as Klemm27

were early proponents of local anti-biotic elution from impregnatedPMMA. Theoretically, when the lo-cal antibiotic concentration is kepthigh (ie, levels well above minimalbacterial concentration levels of rele-vant pathogens), bacterial growthcan be eliminated; low systemic con-centrations associated with this strat-egy minimize the likelihood of ad-verse systemic effects.28 The concepthas been successfully employed andincorporated in several treatmentregimens.68 One problem is the per-manence of methylmethacrylate as acarrier. Frequently, this warrants ad-ditional surgery for removal. Hence,the use of a resorbable carrier hasappeal. Recent clinical reports of cal-cium sulfate, polylactic acid, and cal-cium phosphate, as well as other bio-absorbable “carrier” materials,69-72

are encouraging but limited. Onestudy involved the use of tobramycinwith calcium sulfate carrier pellets in25 patients with chronic osteomyeli-tis following débridement, with erad-ication reported in 92%.26 A secondstudy on the use of either vancomy-cin or tobramycin in calcium sulfatecarrier pellets in six patients withchronic osteomyelitis following dé-bridement showed no infection andprogressive bony healing in five ofthe six at a mean follow-up of 28months.73

Another aspect of musculoskeletalinfection that has prompted researchincludes bacterial adhesion. This fea-ture is thought to be pivotal to thechronicity of osteomyelitis as well asthe “foreign body effect.” Elek andConen74 elegantly demonstrated a10,000-fold enhancement of the

“minimal pustule forming dose” ofstaphylococcus in human skin by thepresence of a single silk suture. Gib-bons and Socransky75 elucidated thecomplex affinity of Streptococcusmutans for the enamel of the humantooth by identifying specific enzymesthat break down sucrose and polym-erize the component glucose into in-soluble glucan. The specific affinityof glucan for the enamel of the toothprovides the “glue” that binds thatsurface to bacterial microcolonies (ie,plaque) acting as a syncytium. Bacte-rial metabolism generates the acid re-sponsible for enamel dissolution andformation of dental caries. By under-standing the complex relationshipsbetween glucan, the bacteria, plaqueformation, and acid generation, thepathophysiology of dental caries andgum disease was elucidated, and ap-propriate therapies were designed togreatly lower their incidence and se-verity. Paralleling these insights,Gristina and Costerton76 and others28

have shown the role of the glyco-calyx and biofilm in adhesion todead bone and implants and its piv-otal role in the production of andpersistence of musculoskeletal infec-tion.

One of the major obstacles to basicresearch in this area is the fact thatstandard bacterial culture (eg, Co-lumbia blood agar plate) selects forbacteria devoid of glycocalycealcoats. The free-floating, “naked”bacteria seen in pure laboratory cul-tures are not the same as theirglycocalyceal-coated cousins adher-ing to bone or implant in an area ofosteomyelitic focus. Hence, new bac-terial culture techniques that selectfor and preserve the glycocalyx and,therefore, the bacteria in their adher-ent mode, need to be developed,standardized, and used as a systemfor assessing the effectiveness of ther-apeutic agents.77

New concepts of bacterial syncytiaand quorum sensing may help pro-

vide a way of averting the elabora-tion of virulent factors characteristicof fulminance.78 That is, by prevent-ing the ability of resident bacteria to“sense a quorum,” it may be possibleto keep the resident bacteria in theirpeaceful commensal state andthereby avoid the confrontation withhost white cells and the elaborationof tissue-destructive lysosomal en-zymes and free radicals, which char-acterize a full-blown local tissue in-fection. Investigation of quorum-sensing signal molecules or theirreceptors, and understanding theirinteractions, may provide novelstrategies for treatment and preven-tion.78,79 The development of appro-priate models to enable standard-ization and assessment of theseparameters is in its infancy.80

Other insights into musculoskeletalinfection include those of Hudson etal,81 who demonstrated the intracel-lular presence of staphylococci in os-teoblasts. Later work demonstratedthat the intracellular staphylococciwere potentially viable and “reinfec-tive,” given their “passage” throughtheir host osteoblast.82 Additionally,it was shown that antibiotic “pres-sure” caused an adaptive change inintracellular staphylococci—that is,the elaboration of an extracellularcapsule.83 Pillai et al84 have used nan-otechnologic methodologies to cou-ple the antibiotic nafcillin topoly(lactic-co-glycolic acid) andthereby facilitate its ability to pene-trate the cell membrane of the osteo-blast. Using this Trojan horse strat-egy, they have reported clearance ofintracellular staphylococci.

Other aspects of microbial patho-physiology that provide potentiallyfertile areas for research are bacterialviruses, or bacteriophages. One ofthe largest centers investigating thisapproach is the Eliava Institute ofBacteriophage, Microbiology and Vi-rology in Tbilisi, Georgia. Currently,treatment using bacteriophages is not



approved in countries other thanGeorgia. The FDA obstacles againstthe use of the methodology are sig-nificant. Although employing bacte-riophages is theoretically appealing,the level of evidence for efficacioustreatment of chronic infection usingthis approach has been low. Onecompelling and potentially promis-ing line of inquiry is the multidisci-plinary approach combining phageresearch with nanotechnology.85

Phage is species- and strain-specific;tying that specificity to a drug is ap-pealing and at least theoretically of-fers the potential for a unique classof therapeutic drugs.

Understanding the fundamental as-pects of the pathophysiology of mi-crobial musculoskeletal infection isnecessary for the development of ef-ficacious treatment strategies. De-spite new and exciting lines of in-quiry, it is still very early in thisprocess.

Summary

The treatment of chronic, posttrau-matic osteomyelitis in the extremityis challenging and often requires acommitment by both the patient andthe treating surgeon toward com-plete (ie, wide) resection of the in-volved bone. Reconstruction can besafely performed by a variety ofmethods; however, proper stagingand patient selection remain criticalto a successful outcome. Consensusregarding the use of depot-deliveredantibiotics, as well as the timing andduration of systemic antibiotics, islacking and deserving of furtherstudy.

References

1. Pollak AN, Jones AL, Castillo RC, BosseMJ, MacKenzie EJ; LEAP Study Group:The relationship between time to surgicaldebridement and incidence of infection

after open high-energy lower extremitytrauma. J Bone Joint Surg Am 2010;92(1):7-15.

2. Shawen SB, Keeling JJ, Branstetter J,Kirk KL, Ficke JR: The mangled footand leg: salvage versus amputation. FootAnkle Clin 2010;15(1):63-75.

3. Mody RM, Zapor M, Hartzell JD, et al:Infectious complications of damagecontrol orthopedics in war trauma.J Trauma 2009;67(4):758-761.

4. McGuigan FX, Forsberg JA, AndersenRC: Foot and ankle reconstruction afterblast injuries. Foot Ankle Clin 2006;11(1):165-182, x.

5. Masini BD, Waterman SM, Wenke JC,Owens BD, Hsu JR, Ficke JR: Resourceutilization and disability outcomeassessment of combat casualties fromOperation Iraqi Freedom and OperationEnduring Freedom. J Orthop Trauma2009;23(4):261-266.

6. Cierny G III: Chronic osteomyelitis:results of treatment. Instr Course Lect1990;39:495-508.

7. Trampuz A, Zimmerli W: Diagnosis andtreatment of implant-associated septicarthritis and osteomyelitis. Curr InfectDis Rep 2008;10(5):394-403.

8. Bowen TR, Widmaier JC: Hostclassification predicts infection afteropen fracture. Clin Orthop Relat Res2005;433:205-211.

9. Calhoun JH, Manring MM: Adultosteomyelitis. Infect Dis Clin North Am2005;19(4):765-786.

10. Cierny G III, Zorn KE, Nahai F: Bonyreconstruction in the lower extremity.Clin Plast Surg 1992;19(4):905-916.

11. Cierny G III , Mader JT, Penninck JJ: Aclinical staging system for adultosteomyelitis. Contemp Orthop 1985;10(5):17-37.

12. Sayle BA, Fawcett HD, Wilkey DJ,Cierny G III , Mader JT: Indium-111chloride imaging in chronicosteomyelitis. J Nucl Med 1985;26(3):225-229.

13. Forsberg JA, Elster EA, Andersen RC,et al: Correlation of procalcitonin andcytokine expression with dehiscence ofwartime extremity wounds. J Bone JointSurg Am 2008;90(3):580-588.

14. Neumaier M, Scherer MA: C-reactiveprotein levels for early detection ofpostoperative infection after fracturesurgery in 787 patients. Acta Orthop2008;79(3):428-432.

15. Gristina AG: Biomaterial-centeredinfection: Microbial adhesion versustissue integration. Science 1987;237(4822):1588-1595.

16. Hall-Stoodley L, Costerton JW, Stoodley

P: Bacterial biofilms: from the naturalenvironment to infectious diseases. NatRev Microbiol 2004;2(2):95-108.

17. Ehrlich GD, Hu FZ, Post JC: Role forbiofilms in infectious disease, inGhannoum MA, ed: Microbial Biofilms.Washington DC: ASM Press, 2004:332-358.

18. Cierny G III, DiPasquale D: Treatment ofchronic infection, in Extremity WarInjuries: State of the Art and FutureDirections. J Am Acad Orthop Surg2006;14(10):S105-S110.

19. Papineau LJ: Osteocutaneous resection-reconstruction in diaphysealosteomyelitis. Clin Orthop Relat Res1974;101:306.

20. Cierny G III: The classification and treat-ment of adult osteomyelitis, in EvartsCM, ed: Surgery of the MusculoskeletalSystem, ed 2. Philadelphia, PA: WBSaunders, 1990.

21. Cierny G III, DiPasquale D:Periprosthetic total joint infections:Staging, treatment, and outcomes. ClinOrthop Relat Res 2002;23-28.

22. Rao N, Cannella B, Crossett LS, YatesAJ Jr, McGough R III: A preoperativedecolonization protocol forStaphylococcus aureus preventsorthopaedic infections. Clin OrthopRelat Res 2008;466(6):1343-1348.

23. Wolcott RD, Dowd SE: A rapidmolecular method for characterisingbacterial bioburden in chronic wounds.J Wound Care 2008;17(12):513-516.

24. Edwards RA, Rodriguez-Brito B, WegleyL, et al: Using pyrosequencing to shedlight on deep mine microbial ecology.BMC Genomics 2006;7:57.

25. Sales de Gauzy J, Vidal H, Cahuzac JP:Primary shortening followed by callusdistraction for the treatment of aposttraumatic bone defect: Case report.J Trauma 1993;34(3):461-463.

26. McKee MD, Wild LM, Schemitsch EH,Waddell JP: The use of an antibiotic-impregnated, osteoconductive,bioabsorbable bone substitute in thetreatment of infected long bone defects:Early results of a prospective trial.J Orthop Trauma 2002;16(9):622-627.

27. Klemm KW: Antibiotic bead chains. ClinOrthop Relat Res 1993;295:63-76.

28. Wahlig H, Dingeldein E, Bergmann R,Reuss K: The release of gentamicin frompolymethylmethacrylate beads: Anexperimental and pharmacokinetic study.J Bone Joint Surg Br 1978;60(2):270-275.

29. Adams K, Couch L, Cierny G, CalhounJ, Mader JT: In vitro and in vivoevaluation of antibiotic diffusion fromantibiotic-impregnated



polymethylmethacrylate beads. ClinOrthop Relat Res 1992;278:244-252.

30. Walenkamp GHIM: Gentamicin-PMMABeads: A Clinical, Pharmacokinetic, andToxicological Study. Darmstadt, FRGermany, E Merck, 1983, pp 19-22.

31. Cierny G III: Managing the débridementdefect, in Coombs R, Fitzgerald R, eds:Infection in the Orthopaedic Patient.London, UK: Butterworth Publishers,1988:123-131.

32. Cattaneo R, Catagni M, Johnson EE:The treatment of infected nonunions andsegmental defects of the tibia by themethods of Ilizarov. Clin Orthop RelatRes 1992;280:143-152.

33. Green SA: Skeletal defects: A comparisonof bone grafting and bone transport forsegmental skeletal defects. Clin OrthopRelat Res 1994;301:111-117.

34. Ilizarov GA, Lediaev VI: Replacement ofdefects of long tubular bones by meansof one of their fragments. Vestn Khir ImI I Grek 1969;102(6):77-84.

35. Ilizarov GA, Ledyaev VI: Thereplacement of long tubular bone defectsby lengthening distraction osteotomy ofone of the fragments. 1969. Clin OrthopRelat Res 1992;280:7-10.

36. Cierney G III, DiPasquale D: Adultosteomyelitis, in Cierny G III, McLarenAC, Wongworawat MD, eds:Orthopaedic Knowledge Update:Musculoskeletal Infection. Rosemont IL:American Academy of OrthopaedicSurgeons, 2009:135-155.

37. Madanagopal SG, Seligson D, RobertsCS: The antibiotic cement nail forinfection after tibial nailing. Orthopedics2004;27(7):709-712.

38. Younger AS, Duncan CP, Masri BA:Treatment of infection associated withsegmental bone loss in the proximal partof the femur in two stages with use of anantibiotic-loaded interval prosthesis.J Bone Joint Surg Am 1998;80(1):60-69.

39. Nelson CL: The current status ofmaterial used for depot delivery of drugs.Clin Orthop Relat Res 2004;427(427):72-78.

40. Sachs BL, Shaffer JW: A staged Papineauprotocol for chronic osteomyelitis. ClinOrthop Relat Res 1984;184:256-263.

41. Volz RG, Kloss J, Peltier LF: The use ofmethylmethacrylate as a temporaryspacer following en bloc resection of thedistal femur. Clin Orthop Relat Res1980;147:185-187.

42. Webb LX: New techniques in woundmanagement: vacuum-assisted woundclosure. J Am Acad Orthop Surg 2002;10(5):303-311.

43. Cierney G III, DiPasquale D: Limb

lengthening for bone loss due toinfection, in Handy RC, ed: Managementof Limb-length Discrepancies.Rosemont, IL: American Academy ofOrthopaedic Surgeons, forthcoming.

44. Zalavras CG, Patzakis MJ, Holtom P:Local antibiotic therapy in the treatmentof open fractures and osteomyelitis. ClinOrthop Relat Res 2004;427:86-93.

45. Ostermann PA, Seligson D, Henry SL:Local antibiotic therapy for severe openfractures: A review of 1085 consecutivecases. J Bone Joint Surg Br 1995;77(1):93-97.

46. Pelissier P, Masquelet AC, Bareille R,Pelissier SM, Amedee J: Inducedmembranes secrete growth factorsincluding vascular and osteoinductivefactors and could stimulate boneregeneration. J Orthop Res 2004;22(1):73-79.

47. Bertazzoni Minelli E, Caveiari C, BeniniA: Release of antibiotics frompolymethylmethacrylate cement.J Chemother 2002;14(5):492-500.

48. Beardmore AA, Brooks DE, Wenke JC,Thomas DB: Effectiveness of localantibiotic delivery with an osteoinductiveand osteoconductive bone-graftsubstitute. J Bone Joint Surg Am 2005;87(1):107-112.

49. Shirtliff ME, Calhoun JH, Mader JT:Experimental osteomyelitis treatmentwith antibiotic-impregnatedhydroxyapatite. Clin Orthop Relat Res2002;401:239-247.

50. Cavanaugh DL, Berry J, Yarboro SR,Dahners LE: Better prophylaxis againstsurgical site infection with local as wellas systemic antibiotics: An in vivo study.J Bone Joint Surg Am 2009;91(8):1907-1912.

51. Mader JT, Stevens CM, Stevens JH,Ruble R, Lathrop JT, Calhoun JH:Treatment of experimental osteomyelitiswith a fibrin sealant antibiotic implant.Clin Orthop Relat Res 2002;403:58-72.

52. Mendel V, Simanowski HJ, Scholz HC,Heymann H: Therapy with gentamicin-PMMA beads, gentamicin-collagensponge, and cefazolin for experimentalosteomyelitis due to Staphylococcusaureus in rats. Arch Orthop TraumaSurg 2005;125(6):363-368.

53. Orhan Z, Cevher E, Yildiz A, Ahiskali R,Sensoy D, Mülazimoglu L:Biodegradable microspherical implantscontaining teicoplanin for the treatmentof methicillin-resistant Staphylococcusaureus osteomyelitis. Arch OrthopTrauma Surg 2009;May 12:[Epub aheadof print].

54. Teupe C, Meffert R, Winckler S,Ritzerfeld W, Törmälä P, Brug E:Ciprofloxacin-impregnated poly-l-lactic

acid drug carrier: New aspects of aresorbable drug delivery system in localantimicrobial treatment of boneinfections. Arch Orthop Trauma Surg1992;112(1):33-35.

55. Edin ML, Miclau T, Lester GE, LindseyRW, Dahners LE: Effect of cefazolin andvancomycin on osteoblasts in vitro. ClinOrthop Relat Res 1996;333:245-251.

56. Isefuku S, Joyner CJ, Simpson AH:Gentamicin may have an adverse effecton osteogenesis. J Orthop Trauma 2003;17(3):212-216.

57. Moehring HD, Gravel C, ChapmanMW, Olson SA: Comparison ofantibiotic beads and intravenousantibiotics in open fractures. ClinOrthop Relat Res 2000;372:254-261.

58. Trebse R, Pisot V, Trampuz A:Treatment of infected retained implants.J Bone Joint Surg Br 2005;87(2):249-256.

59. Dunbar RP Jr: Treatment of infectionafter fracture fixation: Opinion. Retainstable implant and suppress infectionuntil union. J Orthop Trauma 2007;21(7):503-505.

60. Owens BD, White DW, Wenke JC:Comparison of irrigation solutions anddevices in a contaminatedmusculoskeletal wound survival model.J Bone Joint Surg Am 2009;91(1):92-98.

61. Anglen JO: Comparison of soap andantibiotic solutions for irrigation oflower-limb open fracture wounds: Aprospective, randomized study. J BoneJoint Surg Am 2005;87(7):1415-1422.

62. Webb LX, Pape HC: Current thoughtregarding the mechanism of action ofnegative pressure wound therapy withreticulated open cell foam. J OrthopTrauma 2008;22(10 suppl):S135-S137.

63. Gorbach SL, Bartlett JG, Blacklow NR:Infectious Diseases, ed 3. Philadelphia,PA, Lippincott Williams & Wilkins,2004.

64. Mellon M, Benbrook C, Benbrook KL:Hogging It: Estimates of AntimicrobialAbuse in Livestock. Cambridge, MA:Union of Concerned Scientists, 2001.

65. Altemeier WA, Wadsworth CL: Anevaluation of penicillin therapy in acutehematogenous osteomyelitis. J BoneJoint Surg Am 1948;30(3):657-673.

66. Cierney G III: A Guided Tour of theLiving Cell. Scientific American LibrarySeries, vol 1. New York, NY: W. H.Freeman, Scientific American Books,1984.

67. Buchholz HW, Engelbrecht H: Depoteffects of various antibiotics mixed withPalacos resins [German]. Chirurg 1970;41(11):511-515.



68. Hsieh PH, Chen LH, Chen CH, Lee MS,Yang WE, Shih CH: Two-stage revisionhip arthroplasty for infection with acustom-made, antibiotic-loaded, cementprosthesis as an interim spacer. J Trauma2004;56(6):1247-1252.

69. Branstetter JG, Jackson SR, HaggardWO, Richelsoph KC, Wenke JC: Locallyadministered antibiotics in wounds in alimb. J Bone Joint Surg Br 2009;91(8):1106-1109.

70. Garvin K, Feschuk C: Polylactide-polyglycolide antibiotic implants. ClinOrthop Relat Res 2005;437:105-110.

71. Nelson CL, McLaren SG, Skinner RA,Smeltzer MS, Thomas JR, Olsen KM:The treatment of experimentalosteomyelitis by surgical débridementand the implantation of calcium sulfatetobramycin pellets. J Orthop Res 2002;20(4):643-647.

72. Noel SP, Courtney HS, Bumgardner JD,Haggard WO: Chitosan sponges tolocally deliver amikacin and vancomycin:A pilot in vitro evaluation. Clin OrthopRelat Res 2010;468(8):2074-2080.

73. Gitelis S, Brebach GT: The treatment ofchronic osteomyelitis with abiodegradable antibiotic-impregnatedimplant. J Orthop Surg (Hong Kong)2002;10(1):53-60.

74. Elek SD, Conen PE: The virulence ofStaphylococcus pyogenes for man: Astudy of the problems of woundinfection. Br J Exp Pathol 1957;38(6):573-586.

75. Gibbons RJ, Socransky SS: Intracellularpolysaccharide storage by organisms indental plaques: Its relation to dentalcaries and microbial ecology of the oralcavity. Arch Oral Biol 1962;7:73-79.

76. Gristina AG, Costerton JW: Bacterialadherence to biomaterials and tissue:The significance of its role in clinicalsepsis. J Bone Joint Surg Am 1985;67(2):264-273.

77. Brown MR, Costerton JW, Gilbert P:Extrapolating to bacterial life outside thetest tube. J Antimicrob Chemother 1991;27(5):565-567.

78. Miller MB, Bassler BL: Quorum sensingin bacteria. Annu Rev Microbiol 2001;55:165-199.

79. Huber B, Eberl L, Feucht W, Polster J:Influence of polyphenols on bacterialbiofilm formation and quorum-sensing.Z Naturforsch C 2003;58(11-12):879-884.

80. Danino T, Mondragón-Palomino O,Tsimring L, Hasty J: A synchronized

quorum of genetic clocks. Nature 2010;463(7279):326-330.

81. Hudson MC, Ramp WK, Nicholson NC,Williams AS, Nousiainen MT:Internalization of Staphylococcus aureusby cultured osteoblasts. Microb Pathog1995;19(6):409-419.

82. Ellington JK, Harris M, Webb L, et al:Intracellular Staphylococcus aureus: Amechanism for the indolence ofosteomyelitis. J Bone Joint Surg Br 2003;85(6):918-921.

83. Ellington JK, Harris M, Hudson MC,Vishin S, Webb LX, Sherertz R:Intracellular Staphylococcus aureus andantibiotic resistance: Implications fortreatment of staphylococcalosteomyelitis. J Orthop Res 2006;24(1):87-93.

84. Pillai RR, Somayaji SN, Rabinovich M,Hudson MC, Gonsalves KE: Nafcillin-loaded PLGA nanoparticles fortreatment of osteomyelitis. BiomedMater 2008;3(3):034114.

85. Petty NK, Evans TJ, Fineran PC,Salmond GP: Biotechnologicalexploitation of bacteriophage research.Trends Biotechnol 2007;25(1):7-15.



osteomelitis

Documents

chronic osteomyelitisrequires

chronic orotherwise

chronic osteomyelitisdeserves

degrees of soft

durationof treatment

phases of treatment

civilian trauma patients

tenuous softtissue coverage