Metastatic Bone Disease: Diagnosis, Evaluation, and Treatment · the diagnosis was osteosarcoma and not metastatic carcinoma. The treatment options now became incredibly com-plex,

The PDF of the article you requested follows this cover page.

This is an enhanced PDF from The Journal of Bone and Joint Surgery

2009;91:1518-1530. J Bone Joint Surg Am.J. Sybil Biermann, Ginger E. Holt, Valerae O. Lewis, Herbert S. Schwartz and Michael J. Yaszemski

Metastatic Bone Disease: Diagnosis, Evaluation, and Treatment

This information is current as of June 4, 2009

Reprints and Permissions

Permissions] link. and click on the [Reprints andjbjs.orgarticle, or locate the article citation on

to use material from thisorder reprints or request permissionClick here to

Publisher Information

www.jbjs.org20 Pickering Street, Needham, MA 02492-3157The Journal of Bone and Joint Surgery

http://www2.ejbjs.org/misc/reprints_perms.dtl

http://www.jbjs.org

http://www.jbjs.org

Selected

Instructional

Course Lectures

The American Academy of Orthopaedic Surgeons

MARY I. O’CONNOR

EDITOR, VOL. 59

COMMITTEEMARY I. O’CONNORCHAIRMAN

FREDERICK M. AZARPAUL J. DUWELIUSKENNETH A. EGOLPAUL TORNETTA III

EX-OFFICIO

DEMPSEY S. SPRINGFIELDDEPUTY EDITOR OF THE JOURNAL OF BONE AND JOINT SURGERY

FOR INSTRUCTIONAL COURSE LECTURES

JAMES D. HECKMANEDITOR-IN-CHIEF,

THE JOURNAL OF BONE AND JOINT SURGERY

Printed with permission of the American Academy ofOrthopaedic Surgeons. This article, as well as other lecturespresented at the Academy’s Annual Meeting, will be availablein March 2010 in Instructional Course Lectures, Volume 59.The complete volume can be ordered online at www.aaos.org,or by calling 800-626-6726 (8 A.M.-5 P.M., Central time).

1517

Metastatic Bone Disease: Diagnosis,Evaluation, and Treatment

By J. Sybil Biermann, MD, Ginger E. Holt, MD, Valerae O. Lewis, MD, Herbert S. Schwartz, MD,and Michael J. Yaszemski, MD, PhD

An Instructional Course Lecture, American Academy of Orthopaedic Surgeons

Metastatic bone disease is a major health-care issue, affecting 4.9 million individ-uals in the United States. The cost of bonemetastasis from cancer was estimated tobe thirteen billion dollars per year in theUnited States in 20051, and the annualincident number of cancer cases in theUnited States is expected to double overthe next fifty years2. With improvedmedical treatment of many cancers,patients are living longer, which placesthem at increased risk for the develop-ment of metastatic disease3,4. The skel-eton is the third most common target ofmetastatic cancer and can be one of theearliest sites affected, especially in indi-viduals with breast or prostate cancer.Ultimately, 60% to 84% of all cases ofmetastatic disease invade bone, andapproximately 70% of patients withmetastatic bone disease experience bonepain5. Patients with metastatic cancerinvolving bone are also at increasedrisk for fractures, spinal cord compres-sion, hypercalcemia, and immobility re-sulting in substantial medical-associatedmorbidities.

Current treatment options forpatients with bone metastases are pri-

marily palliative. These options consistof local therapies, systemic treatment,and analgesics. Unfortunately, osseousmetastases are generally refractory tosystemic therapy. Local irradiation maybe sufficient, but surgical treatment isnecessary for patients with a pathologicfracture and often necessary for pa-tients with an impending pathologicfracture.

DiagnosisOrthopaedic surgeons are often thepractitioners to whom a person withmusculoskeletal pain is initially referred.If plain radiographs identify a bonelesion in the symptomatic area, theorthopaedic surgeon is then faced with adilemma of how to proceed.

Plain radiographs yield more in-formation about a bone tumor than anyother diagnostic modality. Certain basicguidelines for interpretation of plainradiographs alert the clinician to payparticular attention to the anatomic siteof the bone tumor, the zone of transi-tion between the tumor and the hostbone, and the presence of any internalcharacteristics that may determine the

nature of the matrix that the tumorproduces.

Aggressive features that can beidentified on a plain radiograph in-clude a tumor of >5 cm in diameter,interruption of the cortex, periostealreaction, and pathologic fracture. Cor-tical interruption with concurrentsymptoms can be considered evi-dence of a nondisplaced pathologicfracture.

Benign bone tumors are morecommon in young people, whereasmalignant bone tumors, especiallymetastatic carcinomas, are much morecommon in individuals who are morethan forty years old. The patient’s med-ical history should be elicited to identifyany personal or family history of ma-lignant tumors, cancer risk factors, andsystemic symptoms. The physical ex-amination is important to identify theprecise area of tenderness and thepresence or absence of a soft-tissuemass. If a tumor originated in bone,the soft-tissue mass should not bemobile over the bone. Neurovascularcompromise is uncommon, as is distaledema.

Disclosure: The authors did not receive any outside funding or grants in support of their research for or preparation of this work. Neither they nor amember of their immediate families received payments or other benefits or a commitment or agreement to provide such benefits from a commercialentity.

J Bone Joint Surg Am. 2009;91:1518-30

1518

TH E J O U R N A L O F B O N E & JO I N T SU R G E RY d J B J S . O R G

VO LU M E 91-A d NU M B E R 6 d J U N E 2009ME TA S TAT I C B O N E DI S E A S E : DI AG N O S I S , EVA LUAT I O N ,A N D TR E AT M E N T

Laboratory tests offer some cluesthat may facilitate staging. The mostimportant laboratory tests in the eval-uation of an adult with a bone lesion aremeasurements of serum calcium, serumimmunoglobulin, and prostate-specific-antigen levels and the erythrocyte sed-imentation rate. Hypercalcemia is notuncommon in patients with multiplemyeloma or metastatic cancer, and itcan be life-threatening. A serum proteinelectrophoresis with a monoclonal pro-tein spike is indicative of myeloma. Anelevated level of serum prostate-specificantigen is unique to prostate carcinoma.The erythrocyte sedimentation rate isa nonspecific value that is often elevatedin individuals with infection, immuno-logical disorders, or marrow cell neo-plasms such as lymphoma, Ewingsarcoma, histiocytosis, or leukemia. Apregnancy test is warranted for a womanof child-bearing age to safely allowfurther radiographic imaging. Correla-tion among the history, findings onphysical examination, and findings onplain radiographs is the key to thedecision-making process, and the clini-cian bears the ultimate responsibilityfor correlating these important param-eters. If a primary malignant tumor isconsidered a possibility, referral to anorthopaedic oncologist is justified at thisstage. If the correlating clinical andradiographic data indicate that the tu-mor is benign, observation may bechosen. If it is more probable that thetumor is malignant, a sophisticateddiagnostic and staging strategy shouldbe employed. Even the decision toperform a needle biopsy should bemade judiciously by an experiencedphysician.

The orthopaedic surgeon shouldbe aware that disorders other than bonetumors cause bone lesions. These in-clude infection, stress fracture, myositisossificans, metabolic bone disease, os-teonecrosis, and synovial proliferativediseases.

Staging StudiesThe chance that a solitary bone lesion isa metastatic carcinoma in an individualolder than forty years of age is approx-imately 500 times higher than the

chance that the tumor is a primary bonesarcoma.

There are at least six good reasonsto conduct a staging workup prior tobiopsy6:

1. The tumor may be a sarcoma;thus, a staging workup could prevent aninappropriately placed biopsy site orneedle trajectory.

2. There may be a site that is easierto biopsy or one that is associated withless morbidity.

3. Preoperative embolization maybe needed to prevent bleeding.

4. A biopsy can be avoided if thediagnosis can be made on the basis ofthe laboratory analysis, such as withmultiple myeloma.

5. A working diagnosis or preop-erative suspicion of a primary bonetumor on the basis of imaging studiescan help the surgical pathologist tomake the diagnosis more accurately onfrozen-section analysis when surgery iscontemplated at the time of biopsy.

6. Complete imaging combinedwith histopathologic analysis may makeit more likely for the pathologist toaccurately identify the source.

The most common steps in theworkup for suspected bone metastasisof unknown origin consist of a medicalhistory; physical examination; routinelaboratory analysis; plain radiography ofthe involved bone and chest; whole-body bone scintigraphy (bone scan-ning); and computed tomography of thechest, abdomen, and pelvis with oraland intravenous contrast media. Evalu-ation in this fashion will identify theprimary site in 85% of patients with ametastatic bone tumor7. Local imaging,including magnetic resonance imagingand computed tomography, of theinvolved site needs to be performedonly if a diagnosis of primary disease(sarcoma) is under consideration. Be-cause breast carcinoma is common andrarely presents as a metastasis of un-known primary origin to bone, it mayor may not be necessary to include amammogram in the diagnostic strategyor workup.

Positron emission tomography isan emerging technology that has a highsensitivity for identifying malignant

tumors, infections, and other physio-logic processes in the skeleton and softtissues throughout the body. Its speci-ficity, however, is low. In several studies,positron emission tomography com-bined with computed tomographyidentified the primary tumor in ap-proximately 50% of individuals with ametastasis of previously unknown ori-gin8-10. Use of positron emission to-mography alone (without concomitantcomputed tomography) decreases thespecificity to 30%.

Once all of the data have beengathered, biopsy can be performed. Itcan be either percutaneous (needle) oropen (incisional). The pros and cons ofthe two techniques are beyond the scopeof this lecture. Nonetheless, it is im-perative to make a pathologic diagnosisprior to proceeding with any furthermedical, surgical, or radiation treat-ments. Unless the patient has a knownhistory of histologically confirmed me-tastasis, radiographic findings are in-sufficient evidence on which to basetreatment when cancer is suspected.

Pitfalls in Diagnosis and PrematureSurgical InterventionPerhaps the worst medical scenario for apatient who presents with an unknownbone lesion is the ‘‘rodded’’ sarcoma.A typical example is that of a forty-four-year-old man who sustained a patho-logic fracture while swinging a hammer(Figs. 1-A and 1-B). A lytic lesion wasdetected in the humeral shaft. Statisticaldata and the radiologist’s opinion sug-gested a metastatic malignant tumor ofunknown origin. The surgeon assumedthe radiographic diagnosis (metastaticcarcinoma) to be correct and thusscheduled surgery for that evening. Thepostoperative radiograph showed anantegrade locked intramedullary nail inthe humerus. Material obtained fromreaming during the surgery was sent tothe pathology department, and oneweek later the surgeon was notified thatthe diagnosis was osteosarcoma and notmetastatic carcinoma. The treatmentoptions now became incredibly com-plex, not to be outdone by the likelyconsequences with regard to the pa-tient’s survival and the medicolegal

1519



issues. The problem could have beenentirely eliminated, if the diagnosis hadbeen made first. The lesson to belearned from this case is to proceed withthe staging protocol described above.One should first perform a biopsy andfrozen-section analysis during the sur-gery and not proceed with additionalsurgery until the diagnosis is confidentlymade and the best management of theproblem is clear.

Considerations in SurgicalManagementGoals of treatment of metastatic bonedisease include pain relief, preservationof function, and provision of a long-lasting construct that can be usedimmediately. Accomplishing these goals

in an environment of osteolysis andmechanical instability can be challeng-ing. Common errors that lead to treat-ment failures include underestimatingthe life expectancy of the patient,underestimating the abnormal bonebiology in pathologic defects, andundertreating current disease while notplanning for future disease. Currentstrategies for treatment of metastaticdisease have increased overall patientsurvival; therefore, careful consider-ation must be given not only to theimmediate stability of the surgical con-struct but also to its durability. Despitesome increases in survival, the short-ened life span of this population makesa reoperation due to fixation failureespecially undesirable.

Effect of Life Expectancy onFixation OptionsA clear understanding of the life ex-pectancy of patients with metastaticbone disease can help to prevent manycommon fixation errors and failures.The life expectancy of patients withmetastatic disease strongly influencesthe choice of the method of fixationof a pathologic fracture. Patients whoare likely to survive for a longer dura-tion are more likely to undergo somelevel of fracture-healing and also tobenefit from more extensive resectionsand reconstructions11,12. However, thesepatients are more likely to survivelonger than the reconstruction wasintended to last13. Surgical failures oc-cur when local disease progresses orhardware failure occurs as a result offracture nonunion.

Some general guidelines shouldbe followed while it is kept in mind thattreatment must be individualized. Thesix-month survival rates associatedwith tumors that commonly metasta-size to bone have been reported to be98% for prostate cancer, 89% for breastcancer, 50% for lung cancer, and 51%for kidney tumors14. When the ortho-paedic surgeon is not equipped toestimate life expectancy and weigh therisks and benefits of reconstructionagainst the anticipated life span, amedical oncologist or an orthopaediconcologist should be consulted. Pre-dicting the length of an individualpatient’s life is difficult and unreliable;therefore, it is better to assume that thepatient will live longer than oneanticipates.

Role of Patient Activity Leveland Expectations in Choice ofFixation MethodA more active, mobile patient with alonger life expectancy should be con-sidered for a more aggressive procedurethat preserves a higher level of func-tion as compared with what would beappropriate for an ill, immobile patientwith a short life expectancy. A longersurgical procedure with greater riskscan be justified for a patient who willobtain both short and long-term bene-fits. It is important that the surgical

Fig. 1-A

Figs. 1-A and 1-B A ‘‘rodded’’ sarcoma. An intramedullary nail was placed to treat a

pathologic fracture in the humerus. One week later, it was discovered that the correct

diagnosis was osteosarcoma. Fig. 1-A A pathologic fracture in the humerus with a

radiographic diagnosis of metastatic carcinoma.

1520



goals be explained and reiterated so thatthe patients and families have realisticexpectations. In particular, they shouldbe reminded that care is palliative andnot curative.

Need for Immediate StabilityThe provision of immediate structuralstability and the longevity of the con-struct must be considered when choos-ing the fixation technique. A mindsettoward treating normal bone that isfractured must be abandoned andreplaced with an understanding of theimplications of treating perpetuallymechanically unstable bone with large,nonregenerating defects. Customaryuses of polymethylmethacrylate, autog-enous bone grafts, bone-graft substi-tutes, and porous ingrowth implantsmust be reevaluated15. Constructs thatwould never be used to treat standard(non-pathologic) traumatic fractures

are the mainstay of treatment of path-ologic fractures secondary to metastaticdisease.

Polymethylmethacrylate is a nec-essary adjunct that provides immediatestructural stability and increased bio-mechanical rigidity when combinedwith the use of implants16. Its behavior ispredictable, it does not degrade overtime, it conforms to unusual tumorcavity geometry, and it allows evaluationof early construct failure due to tumorrecurrence. It has been proposed that itsexothermic properties also result insome local tumor necrosis16. Polyme-thylmethacrylate should always be con-sidered first for the filling of a largedefect caused by metastatic tumorosteolysis.

In stark contrast, autogenousbone graft and bone-graft substitutesshould rarely be used. Autogenous bonegraft causes donor site morbidity and

pain, which is not justified in thispatient population. Also, bone graftsand bone-graft substitutes require os-teointegration, which is unpredictablein the setting of tumor osteolysis.Equally important is the fact that, whileintegrating into the defect, the constructdoes not allow unrestricted weight-bearing, which negates the importantconcept of immediate use.

Reasons to Not Plan for Bone IngrowthAnother discordant concept is the use ofbone-ingrowth prostheses rather thancemented prostheses in this popula-tion17. Non-weight-bearing is com-monly required for osteointegration ofporous ingrowth implants. This maynot be acceptable for a debilitatedpatient, who may also have upper-extremity disease precluding weight-bearing. In addition, the use of a porousingrowth prosthesis at a site of micro-scopic residual disease is not recom-mended as the residual tumor can causerapid loosening. One must also con-sider the possibility that postoperativeradiation therapy contributes to bonenecrosis and may lead to peri-implantfailure15.

Differences Between Pathologic andConventional Traumatic FracturesIn order to treat metastatic bone diseasethe surgeon must have a solid grasp ofthe biology of metastatic tumors andunderstand how it differs from that ofnormal bone that is fractured. Fracture-healing rates in association with lesionsthat commonly metastasize to bone varywidely, with reported prevalences of37% for breast cancer, 0% for lungcancer, 44% for kidney tumors, and67% for myelomas18. A life expectancyof longer than six months is the mostpositive factor predicting fracture un-ion11. A knowledge of fracture-healingrates in particular settings may affect thechoice of fixation device; e.g., a distalfemoral tumor with a pathologic frac-ture resulting from metastatic lungcancer may be better treated with amegaprosthesis, whereas a lesion froma myeloma may be amenable to openreduction and internal fixation with aplate and polymethylmethacrylate.

Fig. 1-B

An intramedullary nail has been placed.

1521



Understanding how aggressivelya metastatic tumor needs to be treatedhelps in determining the appropriatedegree of resection and reconstruction.The resections used to treat metastaticdisease are most commonly intrale-sional. Intralesional resection leaves alarge defect as well as microscopic (andsometimes macroscopic) disease thatrequires radiation therapy. The resec-tion increases the size of the defectcaused by the destructive tumor, andthis defect requires a large, immediatelystructurally stable void-filler as bonewill not traverse a large defect, whichconsequently will not heal. Expecting alarge defect to heal is a common mistakethat leads to a high failure rate. Treat-ment of massive defects can requireinnovative and unconventional recon-struction solutions—e.g., shortening ofthe bone with intercalary resection to

achieve immediate bone apposition andstability or, more commonly, filling ofthe defect with polymethylmethacrylate.

Fixation that encompasses theentire bone, when possible, is a key ele-ment in treating metastatic disease andthe large defects that it produces. Notonly does it increase the mechanicalstability of a construct that includes alarge defect, it may also provide pro-phylaxis against future disease. Thedisease that presents in the future mayarise from separate hematogenous de-posits, or it may have been spread alongthe medullary canal when a nail wasinserted through a site of macroscopicdisease.

Role of Postoperative RadiationPostoperative external beam radiationis necessary in most cases to obliterateresidual microscopic disease and thus

prevent disease progression and furtherosteolysis19. However, it also affects thebone’s blood supply, which can lead topostradiation necrosis. Postoperativeradiation should include the entireoperative field, which usually encom-passes the entire bone20. Patients shouldbe followed for the remainder of theirlives to identify any postradiation ne-crosis at the tumor site as well as in theadjacent joints. Necrosis in these loca-tions may require further treatmentsuch as resection of necrotic bone andreconstruction with a prosthesis.

Fixation Specific to Tumor LocationJust as each patient requires individual-ized treatment, each osseous locationrequires special consideration with re-gard to the best type of fracture fixation.In general, pathologic fractures resultingfrom metastatic disease are treated by

Fig. 2-A

A sixty-two-year-old man with a history of squamous cell carcinoma of the tongue presented with worsening pain in

the proximal part of the left thigh and the left hip of three months’ duration. The anteroposterior pelvic radiograph

shows a lytic lesion in the left acetabular dome. An open biopsy confirmed metastatic squamous cell carcinoma.

1522



repairing or removing existing bone15.When there is enough remaining bonewith structural integrity, it may be usedto anchor a nail or a plate augmentedwith polymethylmethacrylate. Whenthe host bone is mechanically incom-petent, there is massive bone loss, ora joint surface is destroyed, bone isremoved and replaced with a prosthesis.

Although there is agreement re-garding the indications for most fixationmethods, there is some controversyabout the best treatment for metastaticbone disease. Controversial areas in-clude the humeral diaphysis, the ace-tabulum, and the femoral neck. Sometreatments are chosen solely on the basisof the surgeon’s preference, whereasothers are selected on the basis of expe-rience combined with scientific princi-ples. The principles previously describedin this paper must be adhered to in orderto treat any metastatic lesion optimally.In addition to predicting the patient’s

life expectancy and understanding tumorbiology, one must be aware of thedifferent biomechanical properties ofthe different areas of the same bone toachieve the best outcomes.

The humeral diaphysis may befixed with an intramedullary nail or acombination of a plate and polymeth-ylmethacrylate21. Although intramed-ullary nailing provides whole-bonefixation, it does not allow tumor de-bulking unless the tumor is exposedthrough a separate incision22. It may beimportant to debulk a tumor that is notradiosensitive, such as a renal cell car-cinoma. In cases in which debulking isplanned, the tumor is usually exposedwell enough for plate fixation, anotherreason why open reduction and internalfixation is a logical decision. Openreduction and internal fixation with aplate combined with use of polymeth-ylmethacrylate allows additional biome-chanical fixation as well as reduction and

control of less radiosensitive tumors.Biomechanically, a combination of plateand polymethylmethacrylate fixation issuperior to intramedullary nail fixation23.

Reconstruction of periacetabulardefects requires a stable construct thatdistributes weight from the lower ex-tremity to the remaining pelvis andspine24. This may be accomplished withuse of a standard reconstruction cup, anantiprotrusio cage, polymethylmetha-crylate, or polymethylmethacrylate withscrew augmentation as described byHarrington25. According to Harrington,the size of the construct that should beused to treat a periacetabular lesionincreases as the size of the lytic defectincreases25. Lesions that do not breachthe joint may be treated with screws orpins and polymethylmethacrylate alone(Figs. 2-A and 2-B), whereas a largerdefect is better treated with an antipro-trusio cage that bypasses the lytic zoneand rests on solid host bone.

Fig. 2-B

A postoperative anteroposterior radiograph shows a reconstruction with polymethylmethacrylate and cannulated

screws that was performed through an extended iliofemoral approach.

1523



Common errors include under-estimating the size of the lytic defect andunderutilizing the surrounding ilium andischium to anchor an implant, or usingstandard arthroplasty implants becausethey are familiar to the surgeon26. If ahemispherical cup is chosen it should becemented into the acetabulum. A betterconstruct is a cup-cement combinationanchored to a large portion of the iliumwith the Harrington technique27. How-ever, this technique is demanding and isusually abandoned in favor of an anti-protrusio cage. Standard22 press-fit ar-throplasty cups loosen, especially whenthere is rapid tumor progression, andjumbo revision cups cannot appropri-ately fill large acetabular defects whenfixation onto normal ilium (and ischiumwith discontinuity) is required. Anti-protrusio cages provide this necessaryilium/ischium fixation and, when aug-mented with polymethylmethacrylate,provide a construct that is immediatelystable for full weight-bearing.

Arthroplasty is the treatmentmethod of choice for femoral necklesions22,28. Reconstruction nails are com-monly used in this setting, but they havea high failure rate because of the largebiomechanical stresses on the femoralneck combined with lesions that pro-gress instead of healing 22. Argumentshave been made in favor of intramed-ullary nail fixation of lesions that occurin the femoral neck and femoral shaft,but these lesions can be addressed with along-stem prosthesis. When a long-stemprosthesis is used, the femoral necklesion is removed and the implantbypasses any additional femoral lesions.Polymethylmethacrylate should be usedwith care in this setting. Patients withextensive metastatic disease in thefemur who undergo stabilization withan intramedullary rod or a long-stemprosthesis combined with polymethyl-methacrylate may have acute pulmo-nary compromise and hypotensionintraoperatively. The anesthesiologistshould be aware of this risk.

Once arthroplasty is chosen, thequestion of whether a cemented longstem or short stem should be usedarises. A long-stem prosthesis is tech-nically more difficult to insert, but it

fixes the entire bone, thereby providingprophylaxis against fractures throughlesions that develop in the future. Anargument against the use of a longstem is that it can be associated withintraoperative hypotension and post-operative sequelae secondary to fatemboli29. A secondary consideration isthat a revision will be difficult shouldan infection or periprosthetic fractureoccur22. Although a short stem is tech-nically easier to insert, it does notprotect the entire bone from fractureand thus requires close follow-up be-cause periprosthetic metastases mayoccur. However, if such metastases dodevelop, the bone can be protectedfrom fracture with a periprostheticplate.

Avoiding Common Fixation ErrorsThat Cause Construct FailuresCommon fixation errors that may leadto construct failure can be avoided byunderstanding concepts that are para-mount to the care of patients withmetastatic bone disease.

The local biology that led to thelytic bone lesion and pathologic fracturealso reduces the ability of the bone toheal. Not taking this into account whentreating a pathologic fracture may leadto failure of fixation and additionaldifficulty for the patient. Relying onradiation or chemotherapy to overcomeosteolytic damage resulting from a tu-mor is a recipe for failure. Periprostheticfixation failure results from incompletetumor debulking, inadequate bone fix-ation, failure to fix the entire length ofbone, use of degradable substancesinstead of polymethylmethacrylate tofill the bone void, or use of familiarimplants and techniques instead of onesthat are more appropriate for fixationof pathologic fractures and may requiregreater exposure.

As stated earlier in this paper, thesurgeon must remember that pathologicfractures occur in perpetually mechan-ically unstable bone with large, non-regenerating defects. Fixation constructsthat would never be considered in astandard trauma setting are the main-stay of treatment of pathologic fracturessecondary to metastatic disease.

Minimally Invasive Treatment ofBone MetastasesRadiation therapy, usually externalbeam radiation, remains the standard ofcare for patients with localized bonepain but no impending risk of fracture.Recently, clinicians have begun to ex-plore alternative strategies for the treat-ment of osseous metastases.

Minimally invasive procedures areexcellent options for the treatment ofskeletal metastases in patients who areotherwise poor surgical candidates be-cause of their age, comorbidities, or theextent of their disease. In addition,patients in whom bone metastases arerefractory to radiation therapy are ex-cellent candidates for these minimallyinvasive procedures.

Radiofrequency AblationThe aim of radiofrequency ablation is toablate tumors as widely as possible butnot beyond the outer margin of thetumor. To achieve this, an electrode,which is essentially an uninsulatedlength of wire that acts as a monopolaremitter of energy, is inserted directlyinto the tumor. Alternating electriccurrent, emitted from the tip of theelectrode, heats the tissue and causes celldeath due to coagulative necrosis30.Thermal diffusion progressively raisesthe temperature of the tissue surround-ing the probe until a steady state isreached. The surrounding blood flowcools the tissue and reduces the extentof thermal coagulation.

With the patient under local an-esthesia or conscious sedation, a smallskin incision is made and the probe isadvanced up to the farthest part of theosteolytic lesion. Once the location ofthe probe is confirmed with use ofcomputed tomography or ultrasound,the electrode is advanced through theinsulated needle tip. The temperature ofthe treated tissue as well as the skin ismonitored constantly. The probe can beintroduced into large lesions multipletimes through additional routes, and theablation is then extended. The use ofmultiple probes allows the clinician totreat larger lesions.

Radiofrequency ablation, whichhas been extensively studied for the

1524



treatment of tumors of the liver andkidney 31-33, cardiac arrhythmias34,35, un-resectable pulmonary tumors36-39, andosteoid osteomas40-43, was, to ourknowledge, first reported for the treat-ment of bone metastases by Dupuy et al.in 200044. Dupuy et al. found a signif-icant reduction in pain in a heteroge-neous population of patients. Callstromet al.45 reported on the results of radio-frequency ablation in twelve patientswith a metastatic bone lesion, measur-ing between 1 and 11 cm, for whomradiation therapy or chemotherapy hadfailed to provide symptomatic relief.On the average, the patients benefitedfrom the treatment within one week.Four weeks after treatment, both themean pain score and the mean score forpain interference with activities of dailyliving, as measured with the Brief PainInventory, had decreased significantly.Narcotic use was also significantlyreduced.

Goetz et al.46 performed a multi-center study of percutaneous image-guided radiofrequency ablation ofpainful bone metastases. They reportedon forty-three patients for whom thestandard treatment of osseous metasta-ses had failed or who were poor candi-dates for such treatment. The meanscore for the worst pain before treat-ment, as recorded on the Brief PainInventory-Short Form, was 7.9 (range, 4to 10) of a possible 10 points. Ninety-five percent of the patients experienceda significant initial decrease in pain, andthe worst pain was decreased signifi-cantly at four, twelve, and twenty-fourweeks.

More recently, Callstrom et al.conducted a prospective clinical trialof the use of percutaneous radiofre-quency ablation guided by computedtomography and ultrasound in fourteenpatients who had one or two painfulosseous metastases47. Each patient hada score of ‡4 points (of a possible 10)for worst pain in a twenty-four-hourperiod. The authors found that fourweeks after treatment, the mean painscore decreased significantly, as did themean score for pain interference withactivities of daily living. All patientsreported a reduction in narcotic use,

and no serious complications wereobserved.

The mechanism by which radio-frequency ablation provides pain reliefis likely multifold. It has been theorizedthat the intense heat that is generatedmay destroy local sensory nerves, thuseffectively ‘‘numbing’’ the area46. Also,the decrease in tumor burden (celldeath) may decrease the production ofcytokines and tumor factors involvedboth in the sensitization of the nerveendings and in the stimulus of osteo-clastic activity, or the radiofrequencyablation may prevent tumor progres-sion and thus prevent development ofadditional painful microfractures of thebone46.

Radiofrequency ablation canprovide effective palliation of local-ized and painful bone metastases. Anadditional benefit is that it can beperformed on an outpatient basis.Furthermore, a biopsy, which isdiagnostic in approximately 75% ofcases, can be performed at the timeof the procedure without decreasingor changing the structural integrityof the treated site. However, radiofre-quency ablation is contraindicatedwhen there is no safe needle accessto the lesion, when there are im-portant structures (especially nerves)within millimeters of the lesion, orwhen the lesions are immediatelysubcutaneous.

Percutaneous CryoplastyThe use of freezing temperatures forthe therapeutic destruction of tissuestarted in England in 1845 when JamesArnott described the use of iced saltsolutions to freeze certain canceroustumors. He reported a reduction intumor size and amelioration of pain48,49.Improved freezing techniques werepossible early in the 1990s, when solid-ified carbon dioxide came into use andlater when liquid nitrogen and nitrousoxide became available. Currently, withthe development of an argon-basedsystem and a smaller applicator diam-eter, use of this technique has becomemore feasible at other disease sites. Likeradiofrequency ablation, percutaneouscryoplasty was initially used for non-

osseous lesions such as hepatic andrenal tumors50-54.

As a result of the size of the probeand the lack of proper insulation, theuse of first-generation devices was lim-ited to the intraoperative setting andopen procedures. However, the devel-opment of sealed cryoprobes with smalldiameters (1.7 and 2.4 mm) and withinsulation along the shaft not onlyallowed the use of these devices percu-taneously, but also afforded the userbetter control over the shape and size ofthe ablated area (with the use of mul-tiple probes).

Cryoprobes are inserted percuta-neously into the tissue with the patientunder general anesthesia or conscioussedation. Argon gas is forced through asegmentally insulated probe. The rapidexpansion of the gas results in rapidcooling, with the temperature reaching2100�C within a few seconds, and thegeneration of an ice ball. Active thawingof the ice ball is achieved by the in-stillation of helium gas, instead of argongas, into the cryoprobe. A single cryo-probe provides an ice ball of approxi-mately 3.5 cm in diameter. The use ofmultiple cryoprobes not only allows thegeneration of large ice balls (>8 cm indiameter), and hence the managementof large lesions, but also permits theclinician to contour the shape of the iceball. The ablation zone can be shapedby varying the geometry of the probeplacement. Up to eight cryoprobes canbe used independently at a time, thusdecreasing the procedure time for largelesions. In addition, synchronous abla-tion with several cryoprobes eliminatespossible residual disease at the interfaceof overlapping zones.

Cell death from cryoablation isdue to two mechanisms: intracellular iceformation and cellular dehydration. Therapid freezing immediately adjacent tothe probe results in intracellular iceformation and subsequent cell destruc-tion; at a further distance from theprobe, the gradual cooling causes os-motic differences across the cell mem-brane, resulting in secondary cellulardehydration and cell death.

Cryoablation treatments are moretime-consuming than radiofrequency

1525



ablation. Although complete radiofre-quency ablation may require severaloverlapping procedures, the time foreach procedure is short (five to tenminutes). However, each freeze-thaw-freeze cycle of cryoablation requirestwenty-five to thirty minutes and anadditional ten minutes for warmingprior to probe removal47. An advantageof cryoablation is that the edges of theice ball can be seen with currentlyavailable imaging. The use of cryoabla-tion for the treatment of primary andmetastatic bone lesions was, to ourknowledge, first reported by Sewellet al.55. In their study, sixteen tumors infourteen patients underwent cryoabla-tion under magnetic resonance imagingguidance. They reported a reduction inpain in the postoperative period, andthis reduction was apparently sustained.

Callstrom et al. reported the re-sults of a prospective study of fourteenpatients with osseous metastases thatwere treated with cryoablation47. Thelesions ranged from 1 to 11 cm indiameter. After treatment, the meanscores for the worst pain and paininterference with activities of daily liv-ing both decreased significantly. Allpatients who were taking narcotic painmedication reported a reduction in itsuse. No complications were reported.The authors concluded that percutane-

ous cryoablation is a safe and effectivemethod for palliation of pain due toosseous metastases. However, they didnote that they did not treat patients whowere at risk for pathologic fracture, andthis may have slightly skewed theresults.

CementoplastyPercutaneous injection of polymethyl-methacrylate into metastatic vertebralbody lesions has been used to palliatepain. Cementoplasty is an extension ofthe concept of vertebroplasty. It consistsof the injection of opacified bone cementinto an osseous cavity, and its goals arestabilization and pain relief. Like theprocedures described above, cemento-plasty provides pain relief, but it has theadded potential benefit of restoring themechanical stability of the bone.

To perform the cementoplasty, aneedle is hammered into the osseouslesion percutaneously under three-dimensional imaging (computed tomog-raphy or magnetic resonance imaging).Venography can then be performed toevaluate the filling pattern and identifysites of leakage. Next, the polymethyl-methacrylate is injected into the cavityunder continuous fluoroscopic guid-ance. Although complete filling of theosteolytic defect with polymethylme-thacrylate is preferred, complete filling

may not be necessary to confer stability.Several authors have recommended ra-diation of the site after the procedure toensure local tumor control47.

Serious complications includepulmonary embolus and fracture56,57,and imaging guidance is necessary dur-ing the procedure to prevent and/orassess the leakage of the polymethyl-methacrylate into the soft tissues and, inthe case of acetabular lesions, into thehip joint.

Cotten et al. reported the out-comes of acetabular cementoplasty forthe treatment of twelve periacetabularlesions56. All patients received postop-erative radiation therapy at an average oftwenty-one days after the procedure.Nine of the patients had pain relief,which was sustained in all but two ofthem. Improvement in mobility andwalking was noted within three days.

More recently, Kelekis et al. treatedtwenty-three lesions in fourteen patientswith cementoplasty57. The lesions werelocated in the superior and inferior pubicrami and within the ischial tuberosity. Allpatients had pain that was refractory toradiation and narcotic therapy. Themean duration of follow-up was ninemonths. The authors found that theprocedure produced effective pain reliefin 92% of the patients.

Cementoplasty andRadiofrequency AblationAlthough cementoplasty has been usedto treat osteolytic lesions, several clini-cians prefer to ablate a metastatictumor prior to the cementoplasty47,58,59.This is especially true when a bonemetastasis is bulky and extends outsidethe bone. In such cases, cementoplastyalone may be insufficient and combi-nation therapy can have a synergisticeffect. The coagulation necrosis pro-duced by the radiofrequency heat abla-tion makes homogeneous distributionof the polymethylmethacrylate in thelesion possible. The combination ofthe two modalities provides local tumorcontrol, tumor necrosis, stabilization,and pain relief.

The procedure is a combinationof the two techniques described above.Briefly, after the radiofrequency abla-

Fig. 3

The strategy described by Tomita et al. (Reprinted, with permission, from: Tomita K,

Kawahara N, Kobayashi T, Yoshida A, Murakami H, Akamaru T. Surgical strategy for spinal

metastases. Spine. 2001;26:298-306.)

1526



tion is performed, a needle is intro-duced into the osteolytic lesion andpositioned under computed tomogra-phy guidance. The polymethylmetha-crylate is then injected into the lesionunder fluoroscopic guidance.

Toyota et al. treated twenty-threebone metastases in seventeen adultpatients with radiofrequency ablationfollowed by cementoplasty60. The meantumor size was 5 cm (range, 2 to 12 cm).The technical success rate was 100%.The patients were followed for anaverage of two years. Initial pain reliefwas achieved in 100% of the patients,and the mean duration of pain relief wasseven months. Three patients had re-currence of the pain from two weeks tothree months after the procedure.

Nakatsuka et al. reported theoutcomes of radiofrequency ablationand cementoplasty for the treatment oftwenty-three metastatic bone lesions inseventeen patients61. The mean tumorsize was 4.9 cm (range, 1.2 to 15 cm).The procedure was technically success-ful in 96% of the patients. Local ther-apeutic effects were evaluated withcontrast-enhanced magnetic resonanceimaging, and tumor necrosis, indicatedby a lack of tumor enhancement, wasobserved in 71% of the cases. Pain wasrelieved within one week in 100% of thepatients, but it recurred in five patientsat a mean of 4.9 months. The authorsconcluded that the combination ofcementoplasty and radiofrequency ab-lation is a valuable treatment alternative.

Spinal MetastasesThe spine is the most common site formetastatic disease, and 40% to 80% ofpatients with cancer have spinal metas-tasis at the time of death. There areapproximately 18,000 new cases of spi-nal metastasis per year in the UnitedStates. The rate of skeletal metastasesrelative to that of primary bone tumorsis 40:1, and skeletal metastasis must beconsidered in the differential diagnosisof a patient who presents with a spinallesion.

Surgical treatment may be ap-propriate for spinal metastasis if thereis a neurologic deficit resulting fromcompression by a surgically accessiblelesion or if the patient has intractablepain. Surgical treatment may also be

Fig. 4

The strategy described by Walker et al. (Reproduced, with modification, from: Walker MP, Yaszemski MJ, Kim CW, Talac R, Currier BL. Metastatic

disease of the spine: evaluation and treatment. Clin Orthop Relat Res. 2003;415 Suppl:S165-75.) Reprinted with permission.

1527



appropriate to establish a histologicdiagnosis, obtain long-term local con-trol, address impending or actual in-stability, or prevent or reduce deformity.Each of these indications is related tothe goals of treatment for patients withspinal metastasis, which are to protector improve neurologic function; tointerfere as little as possible with sys-temic treatment; to be certain of thediagnosis of the spinal lesion prior totreating it; and to reduce unremittingpain so that the patient can return to hisor her prior level of daily function,maximize mobility without using abrace if possible, and improve thequality of life in as little time aspossible62-64. The need for a histologicdiagnosis warrants additional discus-sion. Patients with a personal history ofcancer do experience spinal conditionsthat are not metastases, such as infec-tions and benign tumors, just as personswithout cancer do. A special situation isa spinal lesion in a patient with cancerwho has not yet had a metastasis fromthe cancer. It behooves the treatingphysician to verify that the lesion iscancer prior to initiating treatment formetastatic disease, so that a benignlesion or infection, should either bepresent, receives the appropriate treat-ment. Even if a patient has establishedmetastases elsewhere, it is appropriateto perform a biopsy to obtain specimensfor pathologic analysis and microbio-logic culture, prior to initiating treat-ment if the clinical situation, laboratorystudies, and imaging studies lead touncertainty regarding the spinal lesionin question.

The causes of pain from spinalmetastases include tumor invasion ofa vertebral body, pathologic vertebralfracture, spinal instability, and nerveroot or spinal cord compression. Spinalinstability is difficult to quantify, butthere are several systems with which toclassify it65,66. Once the decision has beenmade that a patient is an appropriatesurgical candidate, it must be estab-lished that he or she does not have anycontraindications to surgery, such asimpaired nutritional status, anemia,coagulopathy, hypercalcemia, too shorta life expectancy, or the inability to

obtain skeletal fixation if reconstructionis part of the surgical plan. Life expec-tancy is particularly difficult to predict,but the surgeon and medical oncologistshould try to make as good an estimateas possible. We have thought that anappropriate life expectancy prior toundergoing major spinal surgery is atleast three months, but this is arbitraryand varies depending on the exactclinical setting.

There are several algorithms inthe literature that one can use to guidedecision-making regarding patientswith spinal metastatic disease. The ap-proach described by Tomita et al.67 is arevised version of the Tokuhashi scoreand includes consideration of the tu-mor’s aggressiveness and the extent ofskeletal and visceral metastases. Scoresfor these factors are combined to cal-culate a total score, which then links totreatments ranging from total en bloctumor resection to palliative care. Thesystem by Walker et al.68 involves asequence of questions about neurologicdeficits, stability, pain, and the tumor’sresponsiveness to radiation. The an-swers to these questions lead to adecision regarding operative or nonop-erative care. The strategies describedby Tomita et al. and by Walker et al.are presented in Figures 3 and 4,respectively.

Between 1980 and 2000, therewere several studies in which it wasconcluded that appropriate surgical de-compression yielded useful improve-ment in neurologic function in about80% of patients with spinal metasta-sis69,70. Klimo et al. performed a meta-analysis of twenty-eight articles on thetreatment of spinal metastases that hadbeen published between 1984 and200271. The data were derived fromtwenty-four articles in which a total of999 patients received surgical treatmentand four articles in which a total of 543patients received radiation treatment,and the studies mostly provided Level-III evidence. Eight hundred and forty-three of the 999 patients in the surgicallytreated group were able to walk after thesurgery, whereas only 357 of the 543patients in the radiation-treated groupwere able to walk after the radiation. Of

384 patients who were unable to walkbefore surgical treatment, 228 regainedthe ability to walk after the surgery, butonly seventy-nine of 265 patients whocould not walk before radiation wereable to do so after it. Thus, the surgicallytreated patients were 1.3 times morelikely to be able to walk after the surgerythan the radiation-treated patients werelikely to walk after the radiation, andthey were twice as likely to regainambulatory function after having beenunable to walk at the time that treat-ment began. The overall success rateswith regard to the ability to walkwere 84% and 66% after surgery andradiation, respectively. The authorsnoted that the neurologic status, overallhealth, extent of the disease (spinal orextraspinal), and type of primary lesionall have an impact on the appropriatetreatment selection.

Patchell et al.72 reported the re-sults of a prospective, randomized,multi-institutional nonblinded trial ofpatients in whom a metastatic tumorresulting in spinal cord compressionwas treated either with surgery fol-lowed by radiation (fifty patients) orwith radiation alone (fifty-one pa-tients). The primary end point was theability to walk, which was achieved forsignificantly more patients in the sur-gery group (forty-two of fifty; 84%)than in the radiation group (twenty-nine of fifty-one; 57%). In addition, ofsixteen patients who were unable towalk at the time that they receivedsurgery, ten regained the ability afterthe surgery, whereas only three of thesixteen patients who could not walkbefore the radiation regained the abil-ity after the radiation. This differencewas significant. The patients in theradiation treatment arm of the studywho lost their ability to walk and thenhad surgical decompression did notdo as well as the nonambulatorypatients who had surgery primarily.The patients treated with surgeryretained the ability to walk signifi-cantly longer (122 days) than did thosetreated with radiation therapy alone(thirteen days). The rules for stoppinga clinical trial were applied to thisstudy because of the large difference in

1528



treatment effect between the twogroups, and the conclusion was thatdirect decompressive surgery in addi-tion to postoperative radiation therapyis superior to treatment with radiationtherapy alone for patients with spinalcord compression caused by metastaticcancer.

The concept of performing surgi-cal decompression first and then ad-ministering radiation was also supportedby several other studies73,74. Wise et al.addressed this issue from the perspectiveof complication rates in a retrospectivestudy74. They reported on eighty-eightpatients who had a spinal procedure forthe treatment of metastatic disease; forty-five had preoperative radiation, andforty-three did not. Six of the forty-fivepatients with preoperative radiation hada major complication, and ten of theforty-five had a minor complication. Incomparison, four of the forty-three pa-tients who had surgery first had a majorcomplication and three of the forty-threehad a minor complication. All of thedeep wound infections in the entire

series were in the patients who had hadpreoperative radiation.

In summary, multiple factorsmust be carefully considered to arriveat a treatment plan for patients whohave spinal metastatic disease. Oneshould remember that spinal metasta-ses are common and that primaryspinal tumors are uncommon, and onemust be certain of the diagnosis priorto initiating treatment. When meta-static disease is present, the treatmentmust address the tumor cells in addi-tion to achieving neurologic decom-pression and spinal column stability.The treatment recommendations frommedical and radiation oncology need tobe tailored to each particular patient’sclinical situation.

J. Sybil Biermann, MDUniversity of Michigan Hospitals, 1500 EastMedical Center Drive, CC 7304/5946,Ann Arbor, MI 48109-5946. E-mail address:[email protected]

Ginger E. Holt, MDHerbert S. Schwartz, MDVanderbilt Orthopaedic Institute,Medical Center East, South Tower Suite 4200,1215 21st Avenue South, Nashville,TN 37232-8774.E-mail address for G.E. Holt:[email protected] address for H.S. Schwartz:[email protected]

Valerae O. Lewis, MDMD Anderson Cancer Center,P.O. Box 301402, Unit 444, Houston,TX 77230-1402.E-mail address: [email protected]

Michael J. Yaszemski, MD, PhDMayo Clinic, 200 First Street S.W., Rochester,MN 55905. E-mail address:[email protected]

Printed with permission of the AmericanAcademy of Orthopaedic Surgeons. This article,as well as other lectures presented at theAcademy’s Annual Meeting, will be available inMarch 2010 in Instructional Course Lectures,Volume 59. The complete volume can beordered online at www.aaos.org, or by calling800-626-6726 (8 A.M.-5 P.M., Central time).

References

1. Schulman KL, Kohles J. Economic burden ofmetastatic bone disease in the U.S. Cancer.2007;109:2334-42.

2. American Cancer Society. Cancer facts and fig-ures. 2006. http://www.cancer.org/docroot/STT/stt_0_2006.asp?sitearea=STT&level=1.

3. Bagi CM. Targeting of therapeutic agents to boneto treat metastatic cancer. Adv Drug Deliv Rev.2005;57:995-1010.

4. Oien KA, Evans TR. Raising the profile of cancer ofunknown primary. J Clin Oncol. 2008;26:4373-5.

5. Varadhachary GR, Abbruzzese JL, Lenzi R. Diag-nostic strategies for unknown primary cancer. Can-cer. 2004;100:1776-85.

6. Simon MA, Finn HA. Diagnostic strategy for boneand soft-tissue tumors. J Bone Joint Surg Am.1993;75:622-31.

7. Rougraff BT, Kneisl JS, Simon MA. Skeletalmetastases of unknown origin. A prospective studyof a diagnostic strategy. J Bone Joint Surg Am.1993;75:1276-81.

8. Ambrosini V, Nanni C, Rubello D, Moretti A,Battista G, Castellucci P, Farsad M, Rampin L,Fiorentini G, Franchi R, Canini R, Fanti S. 18F-FDGPET/CT in the assessment of carcinoma of unknownprimary origin. Radiol Med. 2006;111:1146-55.

9. Pelosi E, Pennone M, Deandreis D, Douroukas A,Mancini M, Bisi G. Role of whole body positronemission tomography/computed tomography scanwith 18F-fluorodeoxyglucose in patients with biopsyproven tumor metastases from unknown primary site.Q J Nucl Med Mol Imaging. 2006;50:15-22.

10. Scott CL, Kudaba I, Stewart JM, Hicks RJ,Rischin D. The utility of 2-deoxy-2-[F-18]fluoro-D-glucose positron emission tomography in the inves-tigation of patients with disseminated carcinoma ofunknown primary origin. Mol Imaging Biol. 2005;7:236-43.

11. Nathan SS, Healey JH, Mellano D, Hoang B,Lewis I, Morris CD, Athanasian EA, Boland PJ.Survival in patients operated on for pathologicfracture: implications for end-of-life orthopedic care.J Clin Oncol. 2005;23:6072-82.

12. Bohm P, Huber J. The surgical treatment of bonymetastases of the spine and limbs. J Bone JointSurg Br. 2002;84:521-9.

13. Kelly CM, Wilkins RM, Eckardt JJ, Ward WG.Treatment of metastatic disease of the tibia. ClinOrthop Relat Res. 2003;415 Suppl:S219-29.

14. Damron TA. Treatment principles and predic-tion of the impending pathologic fracture. In:Schwartz HS, editor. Orthopaedic knowledge up-date: musculoskeletal tumors 2. Rosemont, IL:American Academy of Orthopaedic Surgeons; 2007.p 369-74.

15. Ogilvie CM, Fox EJ, Lackman RD. Currentsurgical management of bone metastases in theextremities and pelvis. Semin Oncol. 2008;35:118-28.

16. Sim FH, Daugherty TW, Ivins JC. The adjunctiveuse of methylmethacrylate in fixation of pathologicalfractures. J Bone Joint Surg Am. 1974;56:40-8.

17. Weber KL, Randall RL, Grossman S, Parvizi J.Management of lower-extremity bone metastasis.J Bone Joint Surg Am. 2006;88 Suppl 4:11-9.

18. Gainor BJ, Buchert P. Fracture healing in meta-static bone disease. Clin Orthop Relat Res. 1983;178:297-302.

19. Frassica DA. General principles of external beamradiation therapy for skeletal metastases. ClinOrthop Relat Res. 2003;415 Suppl:S158-64.

20. Townsend PW, Smalley SR, Cozad SC, Rosen-thal HG, Hassanein RE. Role of postoperative radi-ation therapy after stabilization of fractures causedby metastatic disease. Int J Radiat Oncol Biol Phys.1995;31:43-9.

21. Frassica FJ, Frassica DA. Evaluation and treat-ment of metastases to the humerus. Clin OrthopRelat Res. 2003;415 Suppl:S212-8.

22. Damron TA, Rock MG, Choudhury SN, GrabowskiJJ, An KN. Biomechanical analysis of prophylacticfixation for middle third humeral impending pathologicfractures. Clin Orthop Relat Res. 1999;363:240-8.

23. Weber KL, Lewis VO, Randall RL, Lee AK,Springfield D. An approach to the management of thepatient with metastatic bone disease. Instr CourseLect. 2004;53:663-76.

24. Marco RA, Sheth DS, Boland PJ, Wunder JS,Siegel JA, Healey JH. Functional and oncologicaloutcome of acetabular reconstruction for the treat-ment of metastatic disease. J Bone Joint Surg Am.2000;82:642-51.

25. Harrington KD. New trends in the managementof lower extremity metastases. Clin Orthop RelatRes. 1982;169:53-61.

26. Wunder JS, Ferguson PC, Griffin AM, PressmanA, Bell RS. Acetabular metastases: planning for

1529



reconstruction and review of results. Clin OrthopRelat Res. 2003;415 Suppl:S187-97.

27. Harrington KD. The management of acetabularinsufficiency secondary to metastatic malignant dis-ease. J Bone Joint Surg Am. 1981;63:653-64.

28. Camnasio F, Scotti C, Peretti GM, Fontana F,Fraschini G. Prosthetic joint replacement for longbone metastases: analysis of 154 cases. ArchOrthop Trauma Surg. 2008;128:787-93.

29. Herrenbruck T, Erickson EW, Damron TA, HeinerJ. Adverse clinical events during cemented long-stemfemoral arthroplasty. Clin Orthop Relat Res. 2002;395:154-63.

30. Simon CJ, Dupuy DE. Percutaneous minimallyinvasive therapies in the treatment of bone tumors:thermal ablation. Semin Musculoskelet Radiol.2006;10:137-44.

31. de Baere T, Bessoud B, Dromain C, Ducreux M,Boige V, Lassau N, Smayra T, Girish BV, Roche A,Elias D. Percutaneous radiofrequency ablation ofhepatic tumors during temporary venous occlusion.AJR Am J Roentgenol. 2002;178:53-9.

32. de Baere T, Kuoch V, Smayra T, Dromain C,Cabrera T, Court B, Roche A. Radio frequencyablation of renal cell carcinoma: preliminary clinicalexperience. J Urol. 2002;167:1961-4.

33. de Baere T, Risse O, Kuoch V, Dromain C,Sengel C, Smayra T, Gamal El Din M, Letoublon C,Elias D. Adverse events during radiofrequency treat-ment of 582 hepatic tumors. AJR Am J Roentgenol.2003;181:695-700.

34. Belhassen B, Rogowski O, Glick A, Viskin S, IlanM, Rosso R, Eldar M. Radiofrequency ablation ofaccessory pathways: a 14 year experience at the TelAviv Medical Center in 508 patients. Isr Med Assoc J.2007;9:265-70.

35. Topilski I, Rogowski O, Glick A, Viskin S, Eldar M,Belhassen B. Radiofrequency ablation of atrioven-tricular nodal reentry tachycardia: a 14 year experi-ence with 901 patients at the Tel Aviv SouraskyMedical Center. Isr Med Assoc J. 2006;8:455-9.

36. Steinke K. Radiofrequency ablation of pulmonarytumours: current status. Cancer Imaging. 2008;8:27-35.

37. Steinke K, Sewell PE, Dupuy D, Lencioni R,Helmberger T, Kee ST, Jacob AL, Glenn DW, King J,Morris DL. Pulmonary radiofrequency ablation—aninternational study survey. Anticancer Res. 2004;24:339-43.

38. Zhu JC, Yan TD, Morris DL. A systematic reviewof radiofrequency ablation for lung tumors. AnnSurg Oncol. 2008;15:1765-74.

39. Simon CJ, Dupuy DE, DiPetrillo TA, Safran HP,Grieco CA, Ng T, Mayo-Smith WW. Pulmonary radio-frequency ablation: long-term safety and efficacy in153 patients. Radiology. 2007;243:268-75.

40. Rosenthal DI. Radiofrequency treatment. OrthopClin North Am. 2006;37:475-84, viii.

41. Sans N, Galy-Fourcade D, Assoun J, Jarlaud T,Chiavassa H, Bonnevialle P, Railhac N, Giron J,Morera-Maupome H, Railhac JJ. Osteoid osteoma:CT-guided percutaneous resection and follow-up in38 patients. Radiology. 1999;212:687-92.

42. Woertler K, Vestring T, Boettner F, WinkelmannW, Heindel W, Lindner N. Osteoid osteoma: CT-guided percutaneous radiofrequency ablation andfollow-up in 47 patients. J Vasc Interv Radiol.2001;12:717-22.

43. Rosenthal DI, Alexander A, Rosenberg AE,Springfield D. Ablation of osteoid osteomas with apercutaneously placed electrode: a new procedure.Radiology. 1992;183:29-33.

44. Dupuy DE, Hong R, Oliver B, Goldberg SN.Radiofrequency ablation of spinal tumors: tempera-ture distribution in the spinal canal. AJR Am JRoentgenol. 2000;175:1263-6.

45. Callstrom MR, Charboneau JW, Goetz MP, RubinJ, Wong GY, Sloan JA, Novotny PJ, Lewis BD, WelchTJ, Farrell MA, Maus TP, Lee RA, Reading CC,Petersen IA, Pickett DD. Painful metastases involvingbone: feasibility of percutaneous CT- and US-guidedradio-frequency ablation. Radiology. 2002;224:87-97.

46. Goetz MP, Callstrom MR, Charboneau JW, FarrellMA, Maus TP, Welch TJ, Wong GY, Sloan JA, NovotnyPJ, Petersen IA, Beres RA, Regge D, Capanna R,Saker MB, Gronemeyer DH, Gevargez A, Ahrar K,Choti MA, de Baere TJ, Rubin J. Percutaneous image-guided radiofrequency ablation of painful metastasesinvolving bone: a multicenter study. J Clin Oncol.2004;22:300-6.

47. Callstrom MR, Charboneau JW, Goetz MP, RubinJ, Atwell TD, Farrell MA, Welch TJ, Maus TP. Image-guided ablation of painful metastatic bone tumors: anew and effective approach to a difficult problem.Skeletal Radiol. 2006;35:1-15.

48. Gage AA. History of cryosurgery. Semin SurgOncol. 1998;14:99-109.

49. Gage AA, Gonder MJ, Soanes WA, Emmings FG.Cancer cryotherapy. Mil Med. 1967;132:550-6.

50. Sewell PE, Howard JC, Shingleton WB, HarrisonRB. Interventional magnetic resonance image-guidedpercutaneous cryoablation of renal tumors. SouthMed J. 2003;96:708-10.

51. Shingleton WB, Sewell PE Jr. Cryoablation ofrenal tumours in patients with solitary kidneys. BJUInt. 2003;92:237-9.

52. Zhou XD, Tang ZY. Cryotherapy for primary livercancer. Semin Surg Oncol. 1998;14:171-4.

53. Callstrom MR, Charboneau JW. Technologies forablation of hepatocellular carcinoma. Gastroenterol-ogy. 2008;134:1831-5.

54. Desai MM, Gill IS. Current status of cryoablationand radiofrequency ablation in the management ofrenal tumors. Curr Opin Urol. 2002;12:387-93.

55. Sewell PE Jr, Jackson MS, Dhillon GS. Percuta-neous MRI guided cryosurgery of bone tumors[abstract]. Radiology. 2002;225 Suppl:514.

56. Cotten A, Deprez X, Migaud H, Chabanne B,Duquesnoy B, Chastanet P. Malignant acetabularosteolyses: percutaneous injection of acrylic bonecement. Radiology. 1995;197:307-10.

57. Kelekis A, Lovblad KO, Mehdizade A, Somon T,Yilmaz H, Wetzel SG, Seium Y, Dietrich PY, RufenachtDA, Martin JB. Pelvic osteoplasty in osteolyticmetastases: technical approach under fluoroscopicguidance and early clinical results. J Vasc IntervRadiol. 2005;16:81-8.

58. Schaefer O, Lohrmann C, Herling M, UhrmeisterP, Langer M. Combined radiofrequency thermalablation and percutaneous cementoplasty treatmentof a pathologic fracture. J Vasc Interv Radiol.2002;13:1047-50.

59. Marcy PY, Palussiere J, Descamps B, Magne N,Bondiau PY, Ciais C, Bruneton JN. Percutaneouscementoplasty for pelvic bone metastasis. Support

Care Cancer. 2000;8:500-3. Erratum in: SupportCare Cancer. 2000;8:510.

60. Toyota N, Naito A, Kakizawa H, Hieda M,Hirai N, Tachikake T, Kimura T, Fukuda H, Ito K.Radiofrequency ablation therapy combined withcementoplasty for painful bone metastases: initialexperience. Cardiovasc Intervent Radiol. 2005;28:578-83.

61. Nakatsuka A, Yamakado K, Maeda M, YasudaM, Akeboshi M, Takaki H, Hamada A, Takeda K.Radiofrequency ablation combined with bone cementinjection for the treatment of bone malignancies.J Vasc Interv Radiol. 2004;15:707-12.

62. Ecker RD, Endo T, Wetjen NM, Krauss WE.Diagnosis and treatment of vertebral column metas-tases. Mayo Clin Proc. 2005;80:1177-86.

63. Hosono N, Ueda T, Tamura D, Aoki Y, YoshikawaH. Prognostic relevance of clinical symptoms inpatients with spinal metastases. Clin Orthop RelatRes. 2005;436:196-201.

64. North RB, LaRocca VR, Schwartz J, North CA,Zahurak M, Davis RF, McAfee PC. Surgical manage-ment of spinal metastases: analysis of prognosticfactors during a 10-year experience. J NeurosurgSpine. 2005;2:564-73.

65. Asdourian PL, Weidenbaum M, DeWald RL,Hammerberg KW, Ramsey RG. The pattern of verte-bral involvement in metastatic vertebral breast can-cer. Clin Orthop Relat Res. 1990;250:164-70.

66. Taneichi H, Kaneda K, Takeda N, Abumi K,Satoh S. Risk factors and probability of vertebralbody collapse in metastases of the thoracic andlumbar spine. Spine. 1997;22:239-45.

67. Tomita K, Kawahara N, Kobayashi T, Yoshida A,Murakami H, Akamaru T. Surgical strategy for spinalmetastases. Spine. 2001;26:298-306.

68. Walker MP, Yaszemski MJ, Kim CW, Talac R,Currier BL. Metastatic disease of the spine: evalua-tion and treatment. Clin Orthop Relat Res. 2003;415Suppl:S165-75.

69. Sundaresan N, Rothman A, Manhart K, KelliherK. Surgery for solitary metastases of the spine:rationale and results of treatment. Spine. 2002;27:1802-6.

70. Wai EK, Finkelstein JA, Tangente RP, Holden L,Chow E, Ford M, Yee A. Quality of life in surgicaltreatment of metastatic spine disease. Spine.2003;28:508-12.

71. Klimo P Jr, Thompson CJ, Kestle JR, SchmidtMH. A meta-analysis of surgery versus conventionalradiotherapy for the treatment of metastatic spinalepidural disease. Neuro Oncol. 2005;7:64-76.

72. Patchell RA, Tibbs PA, Regine WF, Payne R, SarisS, Kryscio RJ, Mohiuddin M, Young B. Directdecompressive surgical resection in the treatment ofspinal cord compression caused by metastatic can-cer: a randomised trial. Lancet. 2005;366:643-8.

73. Ghogawala Z, Mansfield FL, Borges LF. Spinalradiation before surgical decompression adverselyaffects outcomes of surgery for symptomatic meta-static spinal cord compression. Spine. 2001;26:818-24.

74. Wise JJ, Fischgrund JS, Herkowitz HN, MontgomeryD, Kurz LT. Complication, survival rates, and riskfactors of surgery for metastatic disease of thespine. Spine. 1999;24:1943-51.

1530



Metastatic Bone Disease: Diagnosis, Evaluation, and Treatment · the diagnosis was osteosarcoma and not metastatic carcinoma. The treatment options now became incredibly com-plex,

Documents