cryosurgery in the treatment of gct

8/7/2019 cryosurgery in the treatment of gct

1/13

CLINICAL ORTHOPAEDICS AND RELATED RESEARCHNumber359,pp 176-1880 1999 Lippincott Williams & Wilkins, Inc.

Cryosurgery in the Treatment ofGiant Cell TumorA Long Term Followup Study

Martin M . Malawer, MD*; Jacob B ickels, MD**; Isaac M eller, MD**;Richard G . Buch, MDT; Robert M . Henshaw, MD*; andYehuda Kollender, MD**

Between 1983 and 1993,102 patients with giantcell tumor of bone were treated at three institu-tions. Sixteen patients (15.9%) presented withalready having had local recurrence. All pa-tients were treated with thorough curettage ofthe tumor, burr drilling of the tumor innerwalls, and cryotherapy by direct pour tech-nique using liquid nitrogen. The average fol-lowup was 6.5 years (range, 4-15 years). Therate of local recurrence in the 86 patientstreated primarily with cryosurgery was 2.3%(two patients), and the overall recurrence ratewas 7.9% (eight patients). Six of these patientswere cured by cryosurgery and two underwentresection. Overall, 100 of 102 patients werecured with cryosurgery. Complications associ-ated with cryosurgery included six (5.9%)pathologic fractures, three (2.9%)cases of par-tial skin necrosis, and two (1.9%)significant de-generative changes. Overall function was good

From the *Washington Cancer Institute, WashingtonHospital Center, Washington DC; **The National Unitof Orthopedic Oncology, Tel-Aviv Sourasky MedicalCenter, Sackler Faculty of Medicine, Tel-Aviv Univer-sity, Tel-Aviv, Israel; and ?Saint Paul Medical CancerCenter, Center for Bone and Soft Tissue Sarcom a, Med-ical City Hospital, Dallas, Texas.Reprint requests to Martin M . Malawer, M D, Washing-ton Cancer Institute, Washington Hospital Center, 110Irving Street NW , Washington, DC 20010.Received: March 3, 1998Revised: July 2, 1998Accepted: August 11, 1998

to excellent in 94 patients (92.2%),moderate inseven patients (6.9%),and poor in one patient(0.9%).Cryosurgery has the advantages of jointpreservation, excellent functional outcome, andlow recurrence rate when compared with otherjoint preservation procedures. For these rea-sons, it is recommended asan adjuvant to curet-tage for most giant cell tumors of bone.

Giant cell tumor of bone first was describedin 1818 by Cooper and Travers.10 Its localaggressiveness was described by Nelaton 53and its malignant potential by Virchow.65During the preroentgen era, most giant celltumors were treated by radical amputation.45Development of precise clinical criteria us-ing radiologic studies permitted better tumoridentification and less radical treatment.4.9

The descriptor benign first was applied togiant cell tumor by Bloodgood 4 to differenti-ate these tumors from other bony malignanciesthat required amputation. He stated that a sig-nificant number of patients with giant cell tu -mor could be cured by multiple excisions. Gi-ant cell tumor now is considered a benignaggressive lesion. This terminology is mis-leading, because 3% of giant cell tumors areprimarily malignant'3.'4,'6,52,64 or will undergomalignant transformation and metastasize ei-ther after radiation therapy6.5',5*or after sev-eral local rec~rrences.24,26~31

176


2/13

Number 359Februarv. 1999Giant cell tumor represents approximately5% of all primary bone tumors. Seventy per-

cent of these lesions occur in the third orfourth decades of life.6J3J6J-4The tumor isthought to arise in the metaphyseoepiphysealjunction.l3J6,24,57Large tumors may extendinto the metaphysis and, more rarely, into thediaphysis. The primary areas of involvementare the femoral condyles, tibia1 plateau,proximal humerus, and distal radius.'6,24,3'CRYOTHERAPY IN THETREATMENT OF GIANT CELLTUMORIn 1966, Gage et a1 20 published their initialfindings on the biologic effect of cryotherapyon bone. These authors produced bone necrosisin laboratory animals by circulating liquid ni-trogen around the femurs and observed subse-quent bone regeneration from the periosteumand endosteum. Marcove and Miller 38 firstused cryotherapy in the treatment of metastaticcarcinoma of the proximal humerus in 1969.They used cryosurgery for treatment of variousbenign and metastatic bone tumor~.36,37~39~40,42Marcove et a141-43 described the use ofcryosurgery in the treatment of primary bonesarcomas. During the 1970s, Marcove et a1 42pioneered the development of cryotherapy inthe treatment of giant cell tumor of bone anddescribed the effectiveness of a direct pourmethod in freezing the walls of a curetted cav-ity. This technique used wide incision, thor-ough curettage, and repetitive exposure of thecuretted area to temperatures below -20" C byliquid nitrogen instillation.42 They advocatedthis method as a physical adjuvant in the hopeof decreasing the high rates of local recurrenceafter curettage, thus avoiding the need for ex-tensive resection and reconstruction.42

Extensive data within the field of cryobi-ology show that five mechanisms are in-volved in the cytotoxicity produced by liq-uid nitrogen: (1) thermal shock, (2) electrolytechanges, (3) formation of intracellular icecrystals and membrane disruption, (4)denatu-ration of cellular proteins, and ( 5 )microvas-

Cryosurgery in Giant Cell Tumor 177cular failure.23,29,44,47,49The formation of intra-cellular ice crystals is considered the mainmechanism of cellular necrosis. During cryo-therapy, rapid freeze causes intracellular icecrystals to form; this is followed by a slowthaw that causes intracellular crystallizationand membrane destruction. Malawer et a1 34emphasized the role of microvascular throm-bosis and described a 7 to 12 mm rim of bonenecrosis when liquid nitrogen was used in adog model. A second freeze and thaw cycle ismore effective because of the increased con-ductivity of the cold after the first cycle.44Marcove et a142 stated that three freeze andthaw cycles produce tumor cell death up to 2cm from the cavity margin.

Cryosurgery has been associated with in-jury to the adjacent rim of bone, cartilage,and soft tissues caused by exposure to liquidnitrogen with secondary fractures, skin in-jury with wound healing problems, and tem-porary neurapraxia (Table 1). The reportedrate of local recurrence varies, ranging from7.1% to 57% (Table 2).The purpose of this study was to evaluatethe efficacy of cryosurgery as a physical adju-vant in the treatment of giant cell tumor ofbone. Particular attention was given to the rateof local recurrence and the extent of complica-tions that have given this modality a poor rep-utation. The study was performed at three on-cology centers, using the same technique ofcurettage, cryosurgery, and reconstruction. Itis the largest report published of giant cell tu-mors treated by cryosurgery with long termfollowup. This is a timely subject in the face ofsurgical advances with cryotherapy in thetreatment of other cancers.66,70,71MATERIALS AND M ETHOD SOne hundred two consecutive patients with giantcell tumor of bone were treated between January1983 and June 1993 at three institutions. All par-ticipating surgeons trained together and used thesame technique of curettage, resection, cryother-apy, and reconstruction. There were 52 male and50 female patients. Ages ranged from 15 to 72years (average, 27 years). The average followup


3/13

178 Malawer et al Clinical Orthopaedicsand Related ResearchTABLE 1. Literature Review on Complication Rate After Cryosurgery

Author Joint NerveCases Fracture Infection Degeneration Palsy Other~ ~ ~ ~-Marcove et aI3 42 52 13 8 2 4Marcove et aI4 18 7 - - 4 Joint stiffness (3)Jacobs and Clemencyz5 12 6Malawer and D ~ n h a r n ~ ~ 25 2 __ - - Flap necrosis ( 1)Synovial fistula (1)Aboulafia et al l 9Marcove et aP9 7 - 2 - - Rectal fistula (1)1Marcove et aI4O 51 5 1Schreuder et aI6 26 1 2

Total 200 34 12 2 10Percent 17 6 1 5

- - -~~ -- - - -- - --

was 6.5 years (range, 4-15 years). Sixteen pa-tients (15.9%) presented with local recurrences;these patients had undergone one to three previ-ous surgical procedures. All patients underwentstaging studies that included plain radiography,computed tomography (CT), and chest radi-ograph. Figure 1 shows the anatomic distributionof the tumor. Using Campanaccis staging systemfor giant cell tumor of bone: 15 tumors wereclassified as Stage I, 47 tumors as Stage 11 , and 40tumors as Stage 111.

If the clinical presentation and the imagingstudies were compatible with diagnosis of a clas-sic benign giant cell tumor of bone, the biopsy(frozen section) and surgery were performed dur-

Scapula ( I )

Sacrum ( 1 ) Pelvis ( 5 )Proximal Femur (12) Distal Radius (6)

Carpal, metacarpalTarsal, metatarsal bones (13)Distal Femur (28)Proximal Fibula ( 5 )Proximal Tibra (20)

Distal Tibia (11)

Fig 1. Anatomic site of giant cell tumor in 102patients treated with cryosurgery.

ing the same session. In case of atypical clinicalor radiologic presentation, either CT guided coreneedle or open incisional biopsy were performedand surgery was delayed until histopathologicevaluation had been completed.

Three patients presented with a closed patho-logic fracture of the distal femur after minortrauma. This group of patients was treated with anopen reduction, curettage, burr drilling, and inter-nal fixation. Cryosurgery, as described in the sur-gical technique section, was performed 4 to 6months later when fracture healing was estab-lished clinically and radiologically.2Surgical Technique

When possible, a pneumatic tourniquet wasused during the procedure to decrease localbleeding and prevent blood from acting as a heatsink and being a thermal barrier for the cryother-apy. Because of the metaphyseoepiphyseal loca-tion of giant cell tumors in long bones,cryosurgery, with the exception of the proximalfemur, is an extracapsular procedure. Violation ofthe joint cavity must be avoided because of thepossibility of contamination of the joint cavitywith tumor cells and potential injury to the carti-lage after direct exposure to liquid nitrogen.Pelvic lesions were approached using the utilitar-ian incision, described by Enneking.17 Sacral andscapular lesions were approached using a longitu-dinal posterior incision. After exposure of the in-volved bone and soft tissues, a cortical windowthe size of the longest longitudinal dimension ofthe tumor was made. A large cortical window isessential to expose the entire tumor and avoid in-


4/13

Number 359February, 1999 Crvosuraerv in Giant Cell Tumor 179TABLE 2. Literature Review of Local Recurrence Rate After Cu rettage, Curettageand Burr Drilling, Resection, and Cryosurgery for G iant Cell Tumor of Bone

Curettage andCurettage Burr Drilling Resection Cryosurgery

Author R LR n LR n LR n LRJohnston and DahlinZ7Hutter et alz4Mnaymneh et aI5lJohnson and Rileyz8Dahlin et allzGoldenberg et alzZM ~ G a r t h ~ ~Marcove et a137,4*Larsson et ai3'Persson and Woulerss6Enneking and ShirleyIgSung et a163Jacobs and Clernencyz5McDonald et a146Carnpanacci et ,I6Malawer and D ~ n h a r n ~ ~Gitelis et al*lSanjai et aP0Aboulafia et allMarcove et a139O'Donnell et aP4Kattapurarn et aI3OYip69TotalPercent

71231617

1362030125285

151

-

-----9--197-

648

411384

739

143142941

-

---

--5--83-

26540.8

Present studyPatients with giant cell tumor, treated primarily with cryosurgeryPatients with already recurrent giant cell tumor, treated with cryosurgeryTotal

149

216

667

---510752758206

------15

339

2404

220

-

--2082000

------04412.9

n = number of treated patients; LR = number of patients with recurrent disease.

adequate curettage. It has to be elliptical with itsaxis parallel to the long axis of bone to reduce thestress rising effect (Fig 2). The tumor was ap-proached through the retained thinned or de-stroyed cortex to minimize additional bone loss. Allgross tumor was removed with hand curettes. Thiswas followed by high speed burr drilling with Mi-das Rex@(Midas Rex, Forth Worth, TX) or BlackMax* (Anspach, Lake Park, FL) of the inner reac-tive shell (Fig 3). Before introduction of the liquidnitrogen, bony perforations were identified andsealed, and the surrounding skin, soft tissues, and

neurovascular bundle were protected by mobiliza-tion and shielding with Gelfoam@(Upjohn, Kala-mazoo, MI). Large skin flaps were retracted toprotect them from any possible spillage of the liq-uid nitrogen (Fig 4).

The direct pour (open) technique as de-scribed by Marcove et al 42 was used; liquid ni-trogen (-196" C) was poured through a stainlesssteel funnel into the tumor cavity, and care wastaken to fill the entire cavity. A thermocouple wasused to monitor the freeze within the cavity, cav-ity wall, adjacent soft tissue, and the area 1 to 2


5/13

180 Malawer et al Clinical Orthopaedicsand Related Research

Correct IncorrectFig 2. A large cortical window is essential toexpose the entire tumor and avoid inadequatecurettage.mm from the periphery of the cavity. The sur-rounding soft tissues were irrigated with warmsaline solution to decrease the possibility of ther-mal injury. Two freeze and thaw cycles were ad-ministered. In each cycle, liquid nitrogen was leftin the cavity until it had evaporated completely.Each cycle lasted for 1 to 2 minutes and was pro-portional to the volume of poured liquid nitrogen.Spontaneous thaw was allowed to occur for 3 to 5minutes. The temperature of the cavity was moni-tored with a thermocouple; once it rose above 0"C, the cycle was considered complete. Afterevaporation, the cavity was irrigated with saline.

Reconstruction then was performed. Threetypes of reconstructions were used depending o nthe site and size of the cavity. These were classi-fied as Type 1, no reconstruction, usually for smallcavities of less than 2 cm in nonweightbearing ar-eas; Type 2, polymethylmethacrylate plus or minusbone graft, before the routine use of internal fixa-tion; and Type 3, polymethylmethacrylate plus orminus bone graft plus internal fixation with in-tramedullary hardware (Figs 5-7). Proximal femur

Fig 4. Liquid nitrogen is poured through a stain-less steel funnel. Temperature within the cavity,and in the surrounding bone and soft tissues ismonitored with thermocouples. Tissues are irri-gated continuously with warm saline solution.

lesions were reconstructed with a side plate andcompression screw (Fig 8). The subchondral sur-faces were reconstructed with autologous bonegraft before cementation. There were nine Type 1,20 Type 2, and 73 Type 3 reconstructions.Postoperative ManagementRoutine perioperative prophylactic antibioticswere administered for 3 to 5 days. The woundswere examined on the third day after surgery. Ifthe skin was intact, passive and active motion ofthe adjunct joint was begun. Patients with lesionsof the lower extremities were kept nonweight-bearing for 6 weeks. Radiographs were obtained6 weeks postoperatively to rule out fracture andto establish bone graft incorporation. If healinghad progressed satisfactorily, weightbearing was

Fig 3. To remove all macro-scopic tumor, curettage has tobe followed by meticulous burrdrilling.


6/13

Number 359February, 1999 Cryosurgery in Giant Cell Tumor 181

Fig 5. Type 3 reconstruction with intramedullaryhardware and reinforcement with polymethyl-methacrylate and corticocancellous bone graft.

allowed. For the first 2 years after surgery, pa-tients were observed in the outpatient clinic every3 months. On each visit, physical examinationand radiographs were performed. Patients wereexamined semiannually for an additional 3 yearsand annually thereafter.

Fig 7. Plain radiograph of Type 3 reconstruc-tion of t h e proximal tibia.

DataAnalysisAll clinical records and imaging studies were an-alyzed for each patient by an orthopaedic oncolo-gist and musculoskeletal radiologist. The site andstage of each lesion was observed on radiographs.The rates of local recurrence, fracture, neu-rapraxia, wound complications, and degenerativechanges were determined. Functional evaluationwas done according to the American Muscu-loskeletal Tumor Society system, 18 and was de-termined by the orthopaedic oncologist at eachpatients most recent followup.

RESULTSOn e hundred tw o patients with g iant cell tu-mo r of bon e were treated with curettage, burrdrilling, and cryosurgery w ith either Type 1,Type 2, or Type 3 reconstruction. The aver-age followup was 6.5 years with a minimumof 4 years.Fig 6. Plain radiograph of Type 3 reconstruc-tion of the distal femur.


7/13

182 Malawer et al Clinical Orthopae dicsan d Related Research

Fig 8. Plain radiograph of Type 3 reconstruc-tion of the proximal femur.

Local RecurrenceLocal recurrence developed in eight patients(7.9%), of which seven were located in boneand one in the soft tissues. The rate of localrecurrence among the 86 patients with noprior treatment was 2.3% (two patients),whereas the recurrence rate among the 16patients who were referred with local recur-rence was 37.5% (six patients). Aftercryosurgery, none of the three patients whopresented with a pathologic fracture had a lo-cal recurrence.

Local recurrences appeared 9 to 48 monthsafter surgery (average, 16 months). Six of theeight patients with local recurrences weretreated by recurettage and cryosurgery; the twoother patients underwent resection surgery.One of these patients had an endoprosthetic re-placement and the second underwent resectionarthrodesis (radiocarpal fusion). One hundred

of 102 patients in the present series were curedwith cryosurgery. All of the patients were dis-ease free at their most recent followup.FracturePostoperative fracture occurred in six pa-tients (5.9%), none of whom had undergoneinternal fixation. Therefore, the fracture rateamong patients treated by internal fixation(Type 3 reconstruction) is 0% (0 of 73 pa-tients) and 21% (six of 29 patients) amongpatients who were not treated with internalfixation (Type 1 or Type 2 reconstruction).All fractures occurred during the first 2 yearsafter the operation, all around the knee joint(distal femur, four; proximal tibia, two), andoften after minor trauma to the extremity.Five fractures eventually united after conser-vative treatment by means of closed reduc-tion and external immobilization with cast orbraces for an average of 9 months. The oneremaining patient required surgery for anasymptomatic nonunion of the tibia.Wound, Soft Tissue InjuryThere were no cases of early or late bone orsoft tissue infection, wound dehiscence, orfull thickness skin necrosis. Three patients(2.9%) sustained partial skin necrosis. Thisdamage resulted from contact with leakingliquid nitrogen and was managed satisfacto-rily by nonsurgical treatment. A peronealnerve palsy was observed in one patient andrecovered spontaneously after 6 months. Novenous or arterial thromboses were ob-served. No neurologic deficits were ob-served in the one patient who was treated forgiant cell tumor of the sacrum. In that case,as in any other anatomic location, nerveswere retracted and protected with Gelfoam.@Degenerative ChangesRadiographic and clinical evidence of de-generative changes around the knee joint de-veloped in two patients. One had mild symp-toms that were managed with conservativetreatment and the other required a total kneereplacement.


8/13

Number 359February, 1999FunctionFunction was estimated to be good or excel-lent in 94 patients (92.2%), moderate inseven patients (6.9%), and poor in one pa-tient (0.9%).

DISCUSSIONThe purpose of this study was to determinethe efficacy of cryosurgery in the treatmentof giant cell tumor of bone. One hundred twoconsecutive patients with giant cell tumor ofbone were treated with cryosurgery with along term followup. This is the largest reportto date of giant cell tumors treated bycryosurgery.

Giant cell tumor is a benign aggressive le-sion. For that reason, absence of local recur-rence, rather than patient survival, is the ma-jor criterion used to assess adequacy ofsurgical treatment. Adequacy of the surgicalmargin, rather than the radiologic stage ofthe tumor, is the major determinant of localtumor contro1.6.46Treatment StrategiesDuring the past several decades, surgeons haveused various modalities in the treatment of gi-ant cell tumors of bone: (1) curettage,totoxic agents such as phenol,~2.~4.~5,21.54,63zincchloride,48 alcoho1,15,63and H202,55,56(3) curet-tage and a physical adjuvant (polymethyl-methacrylate3,5656 and cryosurgery25,36,37,39,42),( 5 )radiation therapy,6,27,51fj2and (6) emboliza-tion, which is practiced in unresectable tu-mors.8 In a classic study from the MemorialSloan-Kettering Hospital, Hutter et a124 re-ported that recurrence rates in giant cell tumorstreated by curettage alone were higher thanthose in tumors treated by resection or curet-tage in combination with physical adjuvants.Table 2 summarizes a large combined clinicalexperience of 648 patients with giant cell tu-mor treated by curettage with an average localrecurrence rate of 40.8% (265 patients).

6,12,22,27,28,32,45,48,51,54,60,63,69(2) curettage and cy-

(4) primary resection, 6,12,19,21,24,27,46,48,51,60,63,69

Cryosurgery in Giant Cell Tumor 183After the neoplastic tissue is curetted away

from the inner wall of the lesion, the reactiveshell consistently reveals an irregular contour.This irregularity makes it virtually impossibleto remove all the tissue with a curette.16 Whencurettage is followed by burr drilling, the rateof local recurrence seems to decrease signifi-cantly; however, although burr drilling is abasic step in most nonresection surgeries ofgiant cell tumors, there are only a few seriesof patients treated with curettage and burrdrilling alone (Table 2).

The difficulties with local control ledsome investigators to recommend en bloc re-sections for persistent cases of giant cell tu -mor. An analysis of 14 studies involving 339patients treated with resection surgeryyielded an average recurrence rate of 12.9%(Table 2). Although this group of patients hasone of the lowest recurrence rates, jointfunction was limited because most tumorsare epiphyseometaphyseal and, therefore,necessitate intraarticular re~ection.6.16~21~63Wide excision and replacement with an allo-graft or a prosthesis is considered too exten-sive surgery to obtain local control, andcurettage plus an adjuvant modality is themain technique used in the treatment of mostgiant cell tumors of bone.

Phenol, which coagulates all proteinaceoussubstances, may remove microscopic tumorresidua that remains after curettage.I4J5,63Be-cause the number of reported patients treatedwith only curettage and phenol is quite smalland the recurrence rate is extremely variable(5% to 66%),12J5,21,s4,63the efficacy of phenolas an adjuvant to curettage is questionable.ODonnell et a1 s4 compared two groups of pa-tients treated with burr drilling and either phe-nol or no adjuvant and found exactly the samerecurrence rate (16.6%).

The two most commonly used physical ad-juvants are polymethylmethacrylate andcryosurgery. Originally, polymethylmethacry-late was used when simple filling with autolo-gous bone was insufficient and arthrodesis wasin question.56 Because the cement filled defectis stable mechanically, patients can bear weight


9/13

184 Malawer et al Clinical Orthopaedicsand Related Researchimmediately and rehabilitate q~ickly.3~56It washypothesized that the heat of polymerization ofthe polymethylmethacrylate could induce tu-mor necrosis and advance the excision marginafter curettage. Moreover, the monomer has adirect toxic effect that results in hypoxia.50Ex-perimental data showed that the heat of po ly-merization drops sharply between the centerof the polymethylmethacrylate and the inter-face with the adjacent bone.68Willuns et al,67who reviewed the effect of heat in a dogmodel, reported that bone marrow necrosisoccurs at 60" C, variable and time dependentnecrosis occurs between 50" C and 60" C, andno necrosis occurs below 48" C. They con-cluded that necrosis of tumor cells was ques-tionable under surgical conditions because themaximum temperature at the cancellous boneinterface in their dog model, using a lateralcondyle filled with polymethylmethacrylate,never exceeded 46" C.67 Malawer et a134 usinga skeletally mature mongrel dog in a tumormodel of the distal femur, compared wholemount sections with plain radiographs, hem a-toxylin and eosin sections, and tetracyclinefluorescence. No evidence of adjacent bonynecrosis was seen when the cavity was filledwith polymethylmethacrylate alone. The mainrole of polymethylmethacrylate is to providemechanical stability. Structural reconstruction,using polymethylmethacrylate and internalfixation (Type 3 reconstruction in this study),is essential to provide m echanical support andprevent fractures through the large curetted,frozen bone cavity. In addition, immediate fix-ation allows early rehabilitation of the adja-cent joint. A proven benefit of polymethyl-niethacry late is that recurrences are readilydiscernible at the bon e-cement interface.56The use of polymethylmethacrylate to filldefects has been criticized because of concernthat its stiffness would lead to early degenera-tive changes when used to support a subchon-dral defect. 56 Wilkins et a1 67 disputed this the-ory and suggested that the stiffness of thepolymethylmethacrylate is not a significantcause of secondary osteoarthritis. However, ithas been shown that the incidence of degener-

ative joint changes after the use of poly-methylmethacrylate alone to fill large sub-chondral bone defects is related to the proxim-ity of the cavity to the articular cartilage.7W hen the distance of the tumor from the artic-ular cartilage was less than 1 cm, the incidenceof degenerative changes was 2.5 times greaterthan wh en the distance was greater than 1cm.7The use of subchondral bone graft, as advo-cated by C am panacci et a17 and routinely usedin the present series, may decrease the likeli-hood of degenerative changes by forming athicker bony interface between the poly-methylmethacrylate and the articular cartilage.In the present series there were two patientswith degenerative joint changes aftercryosurgery. The c linical and radiologic find-ings were no different than for any other pa-tient with noninflammatory arthntides, but thefact that these changes occurred in the samecompartment in which the surgery was per-formed suggests that they might be related toit. In the one patient who underwen t total kneereplacement, surgical specimen was not sentfo r pathologic evaluation.Marcove et a1 3 6 , 3 7 , 3 9 ~ ~ ~reported their re-sults with treating giant ce ll tumor by curet-tage, cryosurgery, and bone grafting or pack-ing the cavity with polymethylmethacrylate.They summarized the experience with twopatient grou ps.@ A 36% recurrence rate w asobserved in the first group (25 patients).Th at recurrence rate, although high, is lowerthan the 50% rate after curettage that was thestandard in that time (Table 2). After Mar-cove refined the surgical technique to in-clude a wider exposure and more carefulcurettage, the rate of recurrence dropped to12% in the second group (27 patients).4* Inthe present study, the recurrence rate afterminimum followup of 4 years was 2.3%among the 8 6 patients w ho w ere treated pri-marily by cryosurgery and 7.9% in the entiregrou p of 102 patients that included 16 pa-tients with recurrent tumor. This is amongthe lowest reported rec urrence rates after anysurgical intervention for giant cell tumor ofbone. Moreover, because 84% to 97% of lo-


10/13

Number 359February, 1999 Cryosurgery in Giant CellTumor 185cal recurrences appear within 2 years,22 andall recurrences were manifest within 3 yearsin the series of Campanacci et a1,6 it is un-likely that a longer followup period signifi-cantly would change these results.

Postoperative fracture is the most commonand serious complication associated withcryosurgery.25.42 Fracture is an inherent riskafter reconstruction of any large bone defect,and especially after cryosurgery near aweightbearing joint. After cryosurgery, bonenecrosis and disruption of osteoid extend theperiod through which reossification occursand delay bone healing.34 Vigorous freezingincreases the likelihood of cure at the cost ofhigher rate of pathologic fractures, whereasinadequate freezing of bone surrounding thetumor may predispose to local recurrence.Marcove et a142 made only a minimal attemptto reconstruct these defects and reported a25% fracture rate that is similar to the fracturerate of the current series when internal fixa-tion was not used. The fractures they reportedoccurred before the use of polymethyl-methacrylate combined with internal fixation.In the present series there were six postopera-tive fractures and all occurred in patients whohad not undergone internal fixation (six of 29cases). As a result, the use of internal fixationis recommended in all patients with giant celltumors who are undergoing cryosurgery.

Wide exposure and adequate mobilizationof skin flaps and adjacent neurovascularbundle, along with continuous irrigation oftissues with warm saline solution, reducesthe incidence of skin necrosis. Three patientsin the present series had a superficial skinnecrosis that healed with conservative localcare. That low rate of skin necrosis (< 3%)compared favorably with the 8% rate re-ported by Marcove et al.42 N o patients in thisstudy had a postoperative infection. It proba-bly is the result of the protective measuresused, including perioperative antibiotics,protection of the skin edges during the pro-cedure, and postoperative elevation of theextremity to reduce venous stasis and edemaof the flap.

Joint function, evaluated by the AmericanMusculoskeletal Tumor Society system waswell preserved (good to excellent function) in92% of the patients in the current series. Thisrate is similar to the rate reported by Jacobsand Clemency,25 who reported preservation ofjoint function in 10 of 12 patients treated bycryosurgery. It also compares favorably withresults among the patients treated by resection.As recommended by Cowell and Curtissll thefollowup in the present study is greater than 2years, as that period being the minimum pe-riod of time required in reporting functionaloutcome in patients who have had a recon-structive surgical procedure.

To perform a controlled study to evaluatethe efficacy of cryosurgery in local controlover giant cell tumor of bone, one has to ran-domize patients to two treatment groups. Thefirst group would be treated with curettage,burr drilling, and cryosurgery, and the secondwith curettage and burr drilling alone. Thisstudy was not performed in this fashion.Given the local aggressiveness of this tumorthat could result in loss of the adjacent jointand the increased risk of malignant transfor-mation after local recurrence, the authorsthought that it would be unethical not to use aphysical adjuvant to curettage and burrdrilling.As recommended by Rudicel and Es-daile,59 it is valid statistically and ethicallypreferable to randomize surgical proceduresto different institutions, each skilled and ex-perienced in a specific procedure. This elimi-nates any bias that results from asking onesurgeon to perform two or more different pro-cedures with the same slull for a given diseaseprocess and, therefore, results of that studywere compared with contemporary publishedresults of alternative treatment modalities.

Cryosurgery is recommended as a physi-cal adjuvant to curettage in the treatment ofgiant cell tumor of bone. It extends the mar-gin of a simple curettage or resection curet-tage and makes it biologically equivalent tothat of a wide resection. Compared withother techniques, cryosurgery with compos-ite fixation not only preserves joint function


11/13

186 Malaweret al Clinical Orthopaedicsand Related Researchbut also significantly decreases th e rate of lo-cal tumor recurrence. Th e routine use of in -ternal fixation with polymethylmethacrylateand bone graft is recommended. Careful at-tention to soft tissue protection and surgicalreconstruction significantly decreases thepreviously pub lished reports of high rates offracture and infection. Resection surgery isreserved for malignant giant cell tumor ofbone or for either primary or recurrent giantcell tumor with an extensive bone destruc-tion and soft tissue com pone nt that representless than 5% of the cases in experience. Mo stprimary giant cell tumors of bone can betreated successfully with curettage, burrdrilling, and cryosurgery, thus avoiding theneed for resection and joint reconstruction.References

1. Aboulafia AJ, Rosenbaum DH, Sicard-RosenhaumL,Jelinek JS, Malawer MM: Treatment of large suh-chondral tumors of the knee with cryosurgery andcomposite reconstruction. Clin Orthop 307:189-199,1994.2. Alkalay D, Kollender Y,Mozes M, Meller I: Giantcell tumors with intraarticular fracture. Two-stagelocal excision, cryosurgery and cementation in 5 pa-tients with distal femoral tumor followed for 2 4years. Acta Orthop Scand 67:291-294, 1996.3. Bini SA, Gill K, Johnston JO: Giant cell tumor ofbone. Curettage and cement reconstruction. Clin Or-thop 321:245-250,1995.4. Bloodgood JC: A conservative treatment of giantcell sarcoma with the study of bone transplantation.Ann Surg 56:210-239,1912.5. Campanacci M: Giant cell tumor and chondrosar-coma. Grading, treatment and results. Cancer Res6. Campanacci M , Baldini N, Boriani S, Sudanese A:.Giant cell tumor of bone. J Bone Joint Surg69A:10&114, 1987.7. Campanacci M, Capanna R, Fahbri N, Bettelli G:Curettage of giant cell tumor of bone. Reconstruc-tion with subchondral grafts and cement. Chir Or-gani Mov 75(Suppl):212-213,1990.8. Carrasco CH, Murray JA: Giant cell tumors. OrthopClin North Am 20:395-405, 1989.9. Coley WB: Prognosis in giant cell sarcoma of thelong bones. Ann Surg 79:321-357,561-595, 1924.

10. Cooper A, Travers B: Surgical Essays. Ed 3. Lon-don, Cox and Son 195,1818.11. Cowell HR, Curtiss Jr PH: The randomized clinicaltrial. J Bone Joint Surg 67A:1151-1152, 1985. Edi-torial.12. DaNin DC, Crupps RE, Johnson EW: Giant-cell tu-mor: A study of 195 cases. Cancer 25:1061-1070,1970.

54:257-261, 1976.

13. Dorfman HD, Czerniak B: Giant Cell Lesions. InDorfman HD, Czerniak B (eds). Bone Tumors. StLouis, CV Mosby 559-598,1998.14. Eckardt JJ, Cooper KL, Unni KK, Sim FH : MayoClinic Tumor Rounds. Benign giant cell tumor. Or-thopedics 3:1142-1152, 1980.15. Eckardt JJ, Grogan TJ: Giant-cell tumor of hone.Clin Orthop 204:45-58, 1988.16. Enneking WF: Lesions of Uncertain Origin Origi-nating in Bone. Giant-Cell Tumor. In Enneking WF(ed). Musculoskeletal Tumor Surgery. Vol 2. NewYork, Churchill Livingstone 1435-1476, 1983.17. Enneking WF: The Anatomic Considerations in Tu-mor Surgery: Pelvis. In Enneking WF (ed). Muscu-loskeletal Tumor Surgery. Vol 2. New York,Churchill Livingstone 483-529,1983,8. Enneking WF, Dunham W, Gebhardt MM, MalawerMM, Pritchard DJ: A system for functional evalua-tion of reconstructive procedures after surgical treat-ment of tumors o f the musculoskeletal system. Clin9. Enneking WF, Shirley PD: Resection-arthrodesis formalignant and potentially malignant lesions aboutthe knee using an intramedullary rod and local bonegrafts. J Bone Joint Surg 59A:223-236,1977.20. Gage AA , Greene JCW, Neiders ME, Emmlings FG :Freezing bone without excision. JAMA 197:77@774,1966.21. Gitelis S , Mallin BA, Piasecki P, Turner F: Intrale-sional excision compared with en bloc resection forgiant cell tumor of bone. J Bone Joint Surg75A: 1648- 1655, 1993.22. Goldenberg R, Campbell C, Bonfiglio M: Giant celltumor. An analysis of 218 cases. J Bone Joint Surg52A:619-664,1970.23. Harris L, Griffths J: Relative effects of cooling andwarming rates on mammalian cells during thefreeze-thaw cycle. Cryobiology 14:662469, 1977.24. Hutter RVP, Worcester Jr JN, Francis KC, et al: Be-nign and malignant giant-cell tumor of bone. A clin-icopathological analysis of the natural history of thedisease. Cancer 15:653-690, 1962.25. Jacobs PA, Clemency RE: The closed cryosurgicaltreatment of giant cell tumor. Clin Orthop 192:149-158,1985.26. Jewel1 JH, Bush L F Benign giant cell tumor ofbone with a solitary pulmonary metastasis. J BoneJoint Surg 46A:848-852, 1964.27. Johnson EW, Dahlin DC: Treatment of giant cell tu-mor of bone. J Bone Joint Surg 41A:895-904,1959.28. Johnson KA , Riley Jr LH: Giant-cell tumor of bone.An evaluation of 24 cases treated at the Johns Hop-kins Hospital between 1925 and 1955. Clin Orthop

29. Karow AR, Webb WR: Tissue freezing, a theory forinjury and survival. Cryobiology 299-108, 196530. Kattapuram AS, ODonnell R, Huszar M, et al: Sur-gical management of innominate giant cell tumor.Clin Orthop 329:281-287, 1996.3 1. Kay RM, Eckardt JJ, Seeger LL, Mirra JM, Hak DJ:Pulmonary metastases of benign giant cell tumor ofbone. Six histologically confirmed cases, includingone of spontaneous regression. Clin Orthop 302:

Orthop 286:241-246, 1993.

62: 187-191, 1969.

219-230,1994.


12/13

Number 359February, 1999 Cryosurgery in Giant Cell Tumor 18732.

33.

34.

35.36 .

37.

38.

39.

40.

41.

42.

43.

44.45.46.

47.

48.

49.

50.

Larsson SE, Lorentzon R, Boquist L: Giant cell tu-mor of bone. A demographic, clinical andhistopathological study of all cases recorded in theSwedish Cancer Registry for the years 1958-1968. JBone Joint Surg 57A:167-173, 1975Malawer MM, Dunham W Cryosurgery and acryliccementation as surgical adjuncts in the treatment ofaggressive (benign) bone tumors. Analysis of 25 pa-tients below the age of 21. Clin Orthop 262:42-57,1991.Malawer MM, Marks MR, McChesney D, et al: Theeffect of cryosurgery and polymethylmethacrylate indogs with experimental bone defects comparable totumor defect. Clin Orthop 226:299-310, 1988.Manaster BJ, Doyle AJ: Giant cell tumors of bone.Radio1 Clin North Am 31:299-323, 1993.Marcove RC: A 17-year review of cryosurgery in thetreatment of bone tumors. Clin Orthop 163:Marcove RC, Lyden JP, Huvos AG, Bullough PG:Giant cell tumors treated by cryosurgery. J BoneJoint Surg 55A: 1633-1644, 1973.Marcove RC, Miller TR: Treatment of primary andmetastatic bone tumors by cryosurgery. JAMA207: 1890-1894,1969.Marcove RC, Sheth DS, Brien EW, Huvos AG,Healey JH: Conservative surgery for giant cell tu-mors of the sacrum. The role of cryosurgery as asupplement to curettage and partial excision. CancerMarcove RC, Sheth DS, Takemoto S , Healey JS:The treatment of aneurysmal bone cyst. Clin OrthopMarcove RC, Stove11 PB, Huvos AG, Bullough PG:The use of cryosurgery in the treatment of low andmedium grade chondrosarcoma. Clin Orthop 122:147-156, 1977.Marcove RC, Weis LD, Vaghaiwalla MR, et al:Cryosurgery in the treatment of giant cell tumor ofbone. A report of 52 consecutive cases. CancerMarcove RC, Zahr KA , Huvos AG, Ogihara WCryosurgery in osteogenic sarcoma: Report of threecases. Cancer 1052-60, 1984.Mazur P: Cryobiology: The freezing of biologicalsystems. Science 168:939-949, 1970.McCarthy EF: Giant-cell tumor of bone: An histori-cal perspective. Clin Orthop 153:14-25, 1980.McDonald DJ, Sim FH, McLeod RA, Dahlin DL:Giant cell tumor of bone. J Bone Joint SurgMcGann LE, Kruuv J, Frim J, Frey HE: Factors af-fecting the repair of sublethal freeze-thaw damage inmammalian cells. Suboptimal temperature and hy-poxia. Cryobiology 12530-539, 1975.McGrath PJ: Giant cell tumor of bone. An analysisof fifty-two cases. J Bone Joint Surg 54B:216-229,1972.Miller RH, Mazur P Survival of frozen-thawed hu-man red cells as a function of cooling and warmingvelocities. Cryobiology 13:404-414, 1976.Mjoberg B, Pettersson H, Rosenquist R, RydholmA: Bone cement, thermal injury and radiolucentzone. Acta Orthop Scand 55:597-600, 1984.

231-234,1982.

7411253-1260,1994,

3111157-163,1995.

41 ~957-969, 1978.

68A:235-242,1986.

5 1. Mnaymneh WA, Dudley RH, Mnaymneh GL: Giantcell tumor of bone. An analysis and follow-up studyof the forty-one cases observed at the MassachusettsGeneral Hospital between 1925 and 1960. J BoneJoint Surg 46A:63-75, 1964.52. Nascimento AG, Huvos AG, Marcove RC: Primarymalignant giant cell tumor of bone: A study of eightcases and review of the literature. Cancer 44:1393-1402,1979,53. Nelaton E: Dune Nouvelle Espece de Tumerus Be-nignes des Os, ou Tumerus a Myeloplaxes. Paris,Adrien Delahaye 65-72, 1860.54. ODonnell RJ, Springfield DS, Motwani HK, et al:Recurrenceof giant cell tumors of the long bones af-ter curettage and packing with cement. J Bone JointSurg 76A:1827-1833, 1994.55. Persson BM, Elselund L, Lovdahl R, Gunterberg B:Favorable results of acrylic cementation for giantcell tumors. Acta Orthop Scand 55:209-214, 1984.56. Persson BM, Wouters HW Curettage and acrylic ce-mentation in surgery of giant cell tumors of bone.Clin Orthop 120:125-133, 1976.57. Picci P, Manfrini M, Zucchi V, et al: Giant cell tumorin skeletally immature patients. J Bone Joint Surg65A:468-490, 1983.58. Rock MG, Sim FH, Unni KK, et al: Secondary ma-lignant giant-cell tumor of bone. Clinicopathologi-cal assessment of nineteen patients. J Bone JointSurg 68A: 1073-1079, 1986.59. Rudicel S , Esdaile J: The randomized clinical trial inorthopedics: Obligation or option? J Bone Joint Surg67A:1284-1293,1985.60. Sanjay BKS, Frassica FJ, Frassica DA, et al: Treat-ment of giant-cell tumor of the pelvis. J Bone JointSurg 75A: 1466-1475,1993,61. Schreuder HWB, Veth RPH, Pruszczynski M, et al:Aneurysmal bone cysts treated by curettage,cryotherapy and bone grafting. J Bone Joint Surg62. Seider MJ, Rich TA, Ayala AG, Murray JA: Giantcell tumors of bone: Treatment with radiation ther-apy. Radiology 161537-540, 1986.63. Sung HW , Kuo DP, Shu WP, et al: Giant-cell tumorof bone: Analysis of two hundred and eight cases inChinese cases. J Bone Joint Surg 64A755-761,1982.64. Unni KK: Giant Cell Tumor (Osteoclastoma). InUnni KK (ed). Dahlins Bone Tumors. General As-pects and Data on 11,087 Cases. Philadelphia, JBLippincott Company 263-283, 1996.65. Virchow R: Die Krankhaften Geschwulste. Vol 2.Berlin, Hirschwald 90-97, 1846.66. Weaver ML, Atkinson D, Zemel R: Hepaticcryotherapy in treating colorectal metastases. Can-cer 76:210-214, 1995.67. Wilkins RM, Okada Y, Sim FH, Chao EYS, GorgkiJ: Methylmethacrylate Replacement of SubchondralBone: A Biomechanical, Biochemical, and Morpho-logic Analysis. In Enneking WF (ed). Limb-SparingSurgery in Musculoskeletal Oncology. New York,Churchill Livingstone 479-485, 1987.68. Willert HG : Clinical Results of The TemporaryAcrylic Bone Cement Plug in The Treatment of BoneTumors: A Multicentric Study. In Enneking WF (ed).

79B:20-25,1997.


13/13

188 Malawer et al Clinical Orthopaedicsand Related ResearchLimb -Sparing Surgery in Mu sculoskeletal Oncology.New York, Churchill Livingstone 445458, 1987.69. Yip KMH, Leung PC, Kumta SM : Giant cell tumorof bone. Clin Orthop 323:6&64,1996.

70. Zacarian SA: Cryosurgery for SkinCancers and Cu -taneous Disorders. St Louis, CV Mosby7-20, 1985.71. Zippe C D: Cryosurgical ablation for prostate cancer:A current review. Semin Urol 13: 148-156, 1995.

cryosurgery in the treatment of gct

Documents

Transcript of cryosurgery in the treatment of gct