Abstract
Background: Aneurysmal bone cyst is a benign, locally destructive lesion of bone. The rates of local recurrence after curettage have varied widely. Therefore, we performed a retrospective study of patients who had had an aneurysmal bone cyst in order to identify the rate of local recurrence and the prognostic factors related to local recurrence after use of contemporary methods of curettage with a high-speed burr.Methods: We reviewed the cases of forty patients who had been managed by the same surgeon for an aneurysmal bone cyst, as diagnosed on the basis of the latest pathological review, between January 1, 1976, and December 31, 1993. The patients were evaluated with regard to age, gender, the duration and type of symptoms, the presence or absence of pathological fracture, the status of the growth plate, the bone and part of the bone that were involved, the type of operative procedure, the outcome, the radiographic stage, the findings on magnetic resonance imaging and computerized tomography (when it became available) and on bone scintigraphy, and histological parameters. The median duration of follow-up was eighty-seven months (range, fifteen to 267 months). According to the criteria of Enneking, no patient had a stage-1 lesion (one with a surrounding rim of cortical bone), twenty-four had a stage-2 lesion (one with a clearly defined border but no cortical bone), and sixteen had a stage-3 lesion (one with no clearly defined border).Results: Of the forty patients, thirty-four had curettage with use of a high-speed burr. Of these thirty-four, twenty-two had filling of the defect with a cancellous autogenous graft; four, with a cancellous allograft; and three, with polymethylmethacrylate. In five patients, no material was put into the defect. The remaining six patients had resection through the margin of the lesion.Four (12 percent) of the thirty-four patients who had curettage had a local recurrence. No patient who had an excision through the margin of the lesion had a local recurrence. All local recurrences were in skeletally immature girls who were three, four, ten, and eleven years old. Univariate analysis with use of the chi-square, Fisher exact, and Wilcoxon log-rank tests showed that local recurrence was associated only with a young age (p = 0.0036) and open growth plates (p = 0.039). All local recurrences occurred within two years postoperatively, at two, seven, nine, and twenty-four months, and all were treated successfully with a second operation.Conclusions: Rates of local control of almost 90 percent can be achieved with thorough curettage with use of a mechanical burr and without use of liquid nitrogen, phenol, or other adjuvants in patients who have an aneurysmal bone cyst of an extremity. A young age and open growth plates are associated with an increased risk of local recurrence.
Aneurysmal bone cyst is a benign, locally destructive lesion of bone that was first described as a distinct entity by Jaffe and Lichtenstein9 in 1942. Mirra stated that this lesion accounted for approximately 1 percent of all bone tumors that were biopsied (no numbers were given)14. Most aneurysmal bone cysts are seen in patients who are in the second decade of life; the lesion is considered rare in patients who are less than five years old.
The treatment for these lesions has consisted of curettage with or without subsequent bone-grafting. The rate of local recurrence after this procedure has been reported to range from 10 percent (two of twenty patients11) to as high as five of seven patients1-3,7,11,13,15,18,21,22,24. This is a wide range, and in many instances the rate has been considered unacceptably high. Therefore, some authors have advocated the use of local adjuvants such as polymethylmethacrylate, liquid nitrogen, and phenol to decrease the rate of local recurrence. Schreuder et al. are the only authors, to our knowledge, to have demonstrated a low rate of local recurrence (one of twenty) with use of liquid nitrogen as an adjuvant20. Most authors have not found a meaningful improvement in the rates of recurrence when local adjuvants were applied1,13,15. Aneurysmal bone cysts in expendable bones can be treated with resection, with an extremely low rate of recurrence1,3,11,16. Other authors have used embolization to treat aneurysmal bone cysts in areas that are not amenable to operative intervention5,16.
The purpose of the current study was to review our operative experience with aneurysmal bone cysts of the extremities in order to ascertain the rate of local recurrence after curettage with use of a high-speed mechanical burr and to identify prognostic factors related to local recurrence.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
†University of Colorado Health Sciences Center, 4200 East 9th Avenue, Box B202, Denver, Colorado 80262.
‡Portland Orthopaedic Clinic, P.O. Box 23200, Portland, Oregon 97281.
§University of Chicago Hospitals and Clinics, MC 3079 (T. D. P. and M. A. S.) and MC 6101 (A. G. M.), 5841 South Maryland Avenue, Chicago, Illinois 60637.
#Nizwa Hospital, P.O. Box 1222, Nizwa 611, Sultanate of Oman.
We conducted a retrospective review of the cases of forty consecutive patients who had been managed for a primary aneurysmal bone cyst of an extremity at our musculoskeletal oncology service between January 1, 1976, and December 31, 1993; all patients were managed by the senior one of us (M. A. S.). No patient who had a secondary aneurysmal bone cyst (an aneurysmal bone cyst associated with another, underlying lesion [a primary tumor]) was included in the study group18. The data were obtained from office and hospital charts, operative reports, and our extensive orthopaedic oncology files. These files contain pathology reports, operative photographs and photographs of specimens, photomicrographs, and all relevant imaging studies.
Each record was evaluated for variables that we believed might influence the prognosis, including the patient's age and gender; the duration and type of symptoms; the presence or absence of pathological fracture; the width of the physeal plate; the bone and portion of the bone that were involved; the operative procedure; local recurrence; the radiographic stage; the findings on magnetic resonance imaging, computerized tomography, and bone scintigraphy; and histological parameters (mitotic count, degree of cellularity, and the presence or absence of nuclear atypia).
All patients had plain radiography, thirty had computerized tomography, twenty had magnetic resonance imaging, and thirty-five had bone scintigraphy. All imaging studies were evaluated and graded by three of us (C. P. G., Jr.; T. D. P.; and M. A. S.). When there was a discrepancy between the stage of the lesion determined on the basis of plain radiographs and that determined on the basis of computerized tomography scans of patients who had had both studies, the final determination was based on the computerized tomography scans. We defined skeletal immaturity as the presence of radiolucency at the level of the growth plate as seen on two orthogonal radiographs.
We applied the radiographic portion of Enneking's staging system for benign bone tumors to all of the plain radiographs and computerized tomography scans6. According to this system, a latent lesion (stage 1) is one that is surrounded by a margin consisting of a mature rim of cortical-like reactive bone without deformation or expansion of the encasing bone. An active lesion (stage 2) has a well defined demarcation between it and the cortical bone, but there is no cortical-like reactive shell (Figs. 1-A and 1-B); the margin is often irregular, and expansion or deformation of the overlying cortex is common. An invasive lesion (stage 3) is characterized by an ill defined border, incomplete reactive bone margins, and cortical destruction; some exhibit soft-tissue extension (Figs. 2-A and 2-B). On the basis of these criteria, no patient had a stage-1 lesion, twenty-four (60 percent) had a stage-2 lesion, and sixteen (40 percent) had a stage-3 lesion as seen on computerized tomography scans and plain radiographs. Thirty patients (75 percent) had both computerized tomography scans and plain radiographs, and in twenty-two (73 percent) of them the lesion was assigned the same stage on the basis of both studies. In four patients the lesion was assigned a higher stage on the basis of computerized tomography scans compared with that assigned on the basis of radiographs, and in four patients it was assigned a lower stage.
Both the computerized tomography scans and the magnetic resonance images were evaluated for the presence of fluid-fluid levels because these have been reported to suggest a diagnosis of aneurysmal bone cyst4,23. Five (17 percent) of the thirty patients who had computerized tomography had fluid-fluid levels, compared with fourteen (70 percent) of the twenty who had magnetic resonance imaging. Eight of fifteen patients who had no demonstrable fluid-fluid levels on computerized tomography had this finding on magnetic resonance imaging, but no patient who had no fluid-fluid levels on magnetic resonance imaging had this finding on computerized tomography. All bone-scintigraphy studies showed increased uptake at the site of the lesion. Fourteen (40 percent) of the thirty-five patients who had bone scans had a so-called halo that appeared as an area of increased nucleotide uptake surrounding an area of relatively decreased uptake8.
The original histological slides were re-reviewed by one of us (A. G. M.), a musculoskeletal pathologist. We performed a histological evaluation of representative paraffin-embedded sections from the lesion. The diagnostic criteria for aneurysmal bone cyst—blood-filled cavernous spaces lined by thin septa of variable cellularity composed of plump stromal and osteoclast-like giant cells—were based on those proposed by Lichtenstein as summarized by Mirra14. The mitotic index (the number of mitoses per ten high-power fields) was measured by counting the number of mitotic figures in a minimum of ten nonoverlapping high-power fields (430 times magnification) per slide that were available for review. Between ten and fifty fields were examined for each patient, depending on the quantity of material that was available. Cytological atypia was graded on the basis of nuclear size and chromatin texture, with grade 0 indicating no atypia; grade 1, enlargement and minimum heterogeneity of the chromatin; grade 2, more pronounced heterogeneity of the chromatin; and grade 3, pleomorphism. Only grade-0 and grade-1 atypia were seen in the current study. Cellularity was estimated in the most cellular areas of the septa and the wall of the lesion; areas with secondary changes, such as fracture or periosteal reaction, were avoided. A grade of 1 indicated low cellularity, similar to that seen in desmoplastic fibroma; a grade of 2, intermediate cellularity, similar to that seen in fibrous dysplasia or fibrous histiocytoma; and a grade of 3, marked cellularity, similar to that seen in giant-cell tumor of bone.
Univariate statistical analysis was performed when appropriate with use of the chi-square, Fisher exact, and Wilcoxon log-rank tests19. Data for each patient, recorded at the time of the most recent visit to the clinic, during which a physical examination had been performed and a radiograph had been made, were obtained from the patient's chart.
The study group consisted of thirteen male patients (33 percent) and twenty-seven female patients (68 percent). The median age at the time of the operation was fourteen and one-half years (range, three to fifty-eight years). The median duration of follow-up was eighty-seven months (range, fifteen to 267 months). Only two patients were followed for less than three years (fifteen and twenty-four months). The median duration of the symptoms before the diagnosis was made was three and one-half months (range, birth to 120 months). The most common initial symptom was pain, which occurred in thirty-seven patients (93 percent). Six patients (15 percent) reported a mass, and three (8 percent) described a decreased range of motion. In one patient (3 percent), the aneurysmal bone cyst was an incidental finding on radiographs that had been made for another purpose. Seven patients (18 percent) had a pathological fracture at the time of the diagnosis.
Thirty-six patients (90 percent) had not had a previous operation for treatment of the lesion. Four patients (10 percent) were referred to us after having had a previous operation elsewhere and sustaining a local recurrence.
Ten (25 percent) of the forty aneurysmal bone cysts involved the femur; six (15 percent), the tibia; five (13 percent), the ulna; four (10 percent) each, the radius and pelvis; three (8 percent), the humerus; two (5 percent) each, the scapula and fibula; and one (3 percent) each, the clavicle, metatarsal, patella, and phalanx of the toe. The growth plate was open in eighteen patients (45 percent) and closed in twenty-two (55 percent). Three lesions occurred on the surface of the bone, and thirty-seven were considered to be based in the medullary canal. Twenty-seven lesions (68 percent) arose from the metaphysis of a long bone, and twelve of these lesions extended into the epiphysis. Four of the twelve lesions had epiphyseal extension across an open growth plate. Six lesions (15 percent) were based in the diaphysis of a long bone, and seven (18 percent) occurred in a flat bone.
The most common method of treatment at our institution was a debulking curettage performed with a curet followed by lavage with saline solution. A mechanical curettage with use of a high-speed burr then was performed, followed by a second lavage and mechanical curettage. If it was technically possible to remove the thin shell of bone and the periosteum, we excised them, mainly to gain access to the cavity for curettage and insertion of the mechanical burr. This operative strategy allowed preservation of the maximum amount of structural integrity of the long bone.
Thirty-four (85 percent) of the forty patients had curettage. The resultant defect was filled with a cancellous autogenous graft in twenty-two patients, with a cancellous allograft in four, and with polymethylmethacrylate in three. The defect was not filled in the remaining five patients. Two of the three patients who received polymethylmethacrylate were considered by the surgeon to need immediate structural support. The third patient originally had been diagnosed, on evaluation of frozen sections, as having a giant-cell tumor and had been managed with use of polymethylmethacrylate; however, on the final pathological review of the operative specimen, the diagnosis was changed to that of an aneurysmal bone cyst. Of the five patients in whom the defect was not filled, three had a lesion in a flat bone and did not need structural support; in the other two, the senior one of us chose not to use any filling material. We did not use adjuvant agents such as liquid nitrogen or phenol, and no patient received radiation.
Six patients (15 percent) had an excision through the margin of the lesion as the initial procedure at our institution. Three of these patients had an expansile lesion in an expendable bone (the iliac wing, the fibular diaphysis, and the metatarsal head). The fourth patient had a pathological fracture of the diaphysis of the radius; this was treated with excision followed by insertion of a nonvascularized fibular autogenous graft fixed with a plate and screws. The fifth patient, who was fifty-three years old, was diagnosed as having a giant-cell tumor on the basis of frozen sections and was managed with a resection and a distal femoral replacement early in the series; however, after the entire operative specimen had been assessed, the diagnosis was changed to an aneurysmal bone cyst. The sixth patient had resection of the distal aspect of the ulna after having been managed for multiple local recurrences at another institution.
Overall, only four patients (10 percent) in our study group had a local recurrence (Fig. 3). Four (12 percent) of the thirty-four patients who had had curettage had a recurrence, compared with none of the six who had been managed with marginal resection. All four recurrences were within the first two years postoperatively.
The first patient who had a local recurrence, a three-year-old girl with a central meta-epiphyseal stage-3 lesion in the distal aspect of the humerus, had had two operations elsewhere; the first consisted of curettage and insertion of an autogenous iliac-crest bone graft, and the second, performed four months later, consisted of curettage and insertion of an autogenous nonvascularized fibular autogenous graft. Both procedures were followed by a rapid local recurrence. We performed curettage with external fixation after the second recurrence. A biopsy performed two years later during a release of the elbow revealed latent asymptomatic nonprogressive disease throughout the previously involved area. At the time of this writing, six years later, she had had no progression of the disease.
The second patient who had a local recurrence, a four-year-old girl with a stage-2 lesion, had had a pathological fracture of the diaphysis of the radius (Fig. 4-A). She initially was managed with curettage and insertion of an autogenous iliac-crest bone graft. After a local recurrence nine months later (Fig. 4-B), she had a resection through the margin of the lesion (Fig. 4-C) and insertion of a nonvascularized fibular autogenous graft (Fig. 4-D). Twenty-six months after the resection, she was disease-free (Fig. 4-E).
The third patient who had a local recurrence was a ten-year-old-girl who had had a stage-3 lesion of the metaphysis of the distal aspect of the femur, which initially was treated with curettage and insertion of an autogenous iliac-crest bone graft. Seven months later, she had a recurrence in the juxta-epiphyseal region, which was treated with repeat curettage supplemented with polymethylmethacrylate. Seventy-seven months after the second curettage, the patient was disease-free.
The fourth patient who had a local recurrence was an eleven-year-old girl who had had a stage-3 lesion and a pathological fracture of the proximal aspect of the radius. She was initially managed with curettage and insertion of an autogenous iliac-crest bone graft. Two months later, she had a recurrence in the juxta-epiphyseal region, which was treated with marginal excision of the proximal aspect of the diaphysis and reconstruction with use of a free nonvascularized fibular graft. Seventy-three months after the second treatment, the patient had no evidence of disease.
Although none of the six patients who were managed with resection through the margin of the lesion had a local recurrence, we could not demonstrate a significant benefit for this treatment compared with curettage because of the small number of patients who had had a primary excision. Because of the negligible rate of local recurrence after primary excision in our series, statistical analysis of the remaining variables was limited to the thirty-four patients who had been managed with curettage.
The median age of the patients who had a local recurrence was seven years, whereas the median age of those who did not have a local recurrence was fifteen years; this difference is significant (p = 0.0036). In addition, if the arbitrary age of ten years that was suggested by Freiberg et al.7 is used, then the rate of recurrence of three of six in the group of patients who were ten years of age or younger is significantly higher than the rate of 4 percent (one of twenty-eight) in the group of patients who were older than ten years (p = 0.012). Similarly, the rate of local recurrence was zero of eighteen among patients who had closed growth plates compared with four of sixteen among those who had open growth plates (p = 0.039). Patients who had open growth plates and had a lesion that crossed this relative barrier to spread were considered as a subset12. Four of our patients fit into this category, and two of them sustained a local recurrence. Because of the small numbers of patients and local recurrences in our study, we were unable to verify any relationship between local recurrence and gender, crossing of the growth plate by the lesion, previous local recurrence, pathological fracture, location of the lesion, radiographic stage, or findings on magnetic resonance imaging. Interestingly, all four local recurrences were in female patients.
Forty histological samples from thirty-six patients were reviewed with regard to mitotic activity, the degree of cellularity, and the presence of nuclear atypia; these three variables were considered to reflect biological invasiveness. The criteria that we chose to predict recurrence are those that commonly predict more invasive behavior of tumors. Interestingly, however, the four recurrent lesions were apparently less invasive, as predicted by the histological criteria, than the original lesions in these patients. Of the thirty-six initial lesions, twenty-six (72 percent) demonstrated some mitotic activity, nuclear atypia, and moderate cellularity. In contrast, none of the recurrent lesions had mitotic activity, only one showed atypia, and three of four exhibited low cellularity. None of the recurrent lesions demonstrated an increase in any histological measure of invasiveness compared with the primary lesions.
Aneurysmal bone cysts frequently have been reported to have a relatively high rate of local recurrence1-3,7,11,13,15,18,21,22,24. Because of the apparently high rate after curettage, several authors have recommended the use of various adjuvants, including liquid nitrogen, phenol, and radiation1,13,15,20. Compared with curettage alone, these adjuvants are associated with increased morbidity (for example, pathological fracture may occur after use of liquid nitrogen), and they do not appear to decrease the rate of local recurrence compared with the treatment described in the current study. Excision through the margin of the lesion has been recommended for expendable bones and is the only treatment that has consistently decreased the rate of local recurrence1,3,11,15. However, resection through the margins of lesions in long bones, especially if they are nonexpendable or have open growth plates, is often too extensive for the treatment of benign lesions.
We reviewed our database to determine the rate of local recurrence and the prognostic features related to local recurrence in patients who had an aneurysmal bone cyst of an extremity that was treated with contemporary methods of curettage with a high-speed burr. The use of adjuvants (polymethylmethacrylate) was infrequent (three patients). Marginal resection was performed in only six patients, in whom the involved bone was expendable, the structural integrity would have been lost if curettage were performed, or the planned reconstruction was simple. Our findings show that the rate of local recurrence can be kept to a minimum (four [12 percent] of thirty-four patients in the current study) if curettage is thorough and is performed with a power burr. Furthermore, local recurrence was significantly related only to a young age and open growth plates. Gender, a high radiographic stage, occurrence in a long bone, pathological fracture, and violation of an open growth plate showed some trend toward an increased risk of local recurrence. However, significant relationships between local recurrence and gender, previous local recurrence, pathological fracture, anatomical location, radiographic stage, findings on magnetic resonance imaging or computerized tomography, findings on bone scintigraphy, or histological characteristics may not be detectable with small numbers of patients and local recurrences, as we had in our study.
Several other authors have suggested that a young age is associated with an increased rate of local recurrence. In a large retrospective study at the Mayo Clinic, Vergel De Dios et al. noted that 93 percent (twenty-seven) of twenty-nine patients who had a local recurrence were less than twenty years old24. Similarly, Tillman et al., in an earlier report on this same patient population, noted that local recurrence occurred in 32 percent (nine) of twenty-eight patients who were less than fifteen years old compared with 12 percent (four) of thirty-four who were more than fifteen years old22. Biesecker et al. also noted a trend toward higher rates of recurrence in younger patients1. In contrast, Ruiter et al. could not demonstrate a difference in the rates of recurrence between age-groups18.
There have been three relatively recent articles on aneurysmal bone cysts in young children. Cole noted a high rate of recurrence in children between the ages of five and fourteen years; seven of thirteen children had a recurrence after curettage alone3. Freiberg et al. reported local recurrence after curettage in five of seven patients who were less than ten years old7. Ozaki et al. recently reported a low rate of recurrence in fourteen patients who were less than ten years old17. One of five patients who had been managed with curettage and bone-grafting had a local recurrence, as did one of five who had been managed with curettage supplemented with polymethylmethacrylate. None of their patients had a recurrence after resection.
We found a significantly higher rate of local recurrence (p = 0.0036) in our younger patients. A variable closely related to chronological age is the maturity of the growth plates. We noted a significantly higher rate of recurrence (p = 0.039) in patients with open growth plates, suggesting that a young age and open physes are indicators of a poor prognosis relative to local recurrence. To our knowledge, this relationship has not been evaluated in other studies. We realize that it is difficult to determine exactly when the growth plate is closed on radiographs, and closure is not always found to be associated with age or gender.
The high rate of local recurrence in young and skeletally immature patients may be related to a more active biological process of the lesion in this population. Alternatively, a reluctance on the part of the surgeon to perform a sufficiently extensive procedure in or near an open growth plate, especially in a lower extremity, could result in residual tissue after curettage, thus facilitating local recurrence. If this second hypothesis is correct, recurrence could be expected to take place next to the open growth plate. In the current series, two of the three lesions that originally had been treated near an open growth plate recurred next to the growth plate and one did not. Our study group was not large enough to confirm either hypothesis. Therefore, both biological invasiveness and timidity on the part of the surgeon remain plausible explanations.
Surprisingly, although all of the local recurrences were in young female patients, gender did not emerge as a significant predictor of local recurrence with the numbers available. Few other authors have attempted to demonstrate an association between gender and recurrence. Biesecker et al. noted that all local recurrences in their series were in long bones of female patients, but this finding was not significant with the numbers available1. The lack of significance of this variable in our study may very well have been due to the small number of local recurrences and to the higher percentage of female patients than usually has been reported18,22,24.
Thirty-four (85 percent) of our forty patients were managed with curettage with use of a high-speed power burr, and only four (12 percent) of them had a local recurrence. No patient who was managed with resection had a local recurrence. The reported rates of local recurrence of aneurysmal bone cysts after curettage have varied widely1-3,7,11,13,15,18,21,22,24. The most commonly used adjuvant has been liquid nitrogen1, and the rates of recurrence have been similar to those reported after curettage alone or in association with bone-grafting, except in the series of Schreuder et al., who reported only one local recurrence (4 percent) after the treatment of twenty-seven lesions with curettage and cryotherapy20. However, the rate of fracture after cryotherapy has been reported to be approximately 8 percent (four of fifty-one)1,13. We believe that the 12 percent rate of local recurrence after curettage alone in the current series compares favorably with the previously reported rates and that it does not justify the potential morbidity associated with use of adjuvants, especially because none of the four recurrent lesions recurred again after a second treatment.
In summary, acceptable rates of local control can be achieved with use of high-speed curettage alone in patients who have an aneurysmal bone cyst of an extremity. All four local recurrences in our patients took place within two years after the operation, suggesting that radiographic surveillance for local recurrence should continue for at least two years. The closely related variables of a young age and open growth plates are both significant prognostic indicators of an increased risk of local recurrence after extensive mechanical curettage.
Biesecker, J. L.; Marcove, R. C.; Huvos, A. G.; and Mike, V.: Aneurysmal bone cysts. A clinicopathologic study of 66 cases. Cancer,26: 615-625, 1970.26615
1970
[PubMed]
Campanacci, M.; Capanna, R.; and Picci, P.: Unicameral and aneurysmal bone cysts. Clin. Orthop.,204: 25-36, 1986.20425
1986
[PubMed]
Cole, W. G.: Treatment of aneurysmal bone cysts in childhood. J. Pediat. Orthop.,6: 326-329, 1986.6326
1986
Davies, A. M.; Cassar-Pullicino, V. N.; and Grimer, R. J.: The incidence and significance of fluid-fluid levels on computed tomography of osseous lesions. British J. Radiol.,65: 193-198, 1992.65193
1992
DeRosa, G. P.; Graziano, G. P.; and Scott, J.: Arterial embolization of aneurysmal bone cyst of the lumbar spine. A report of two cases. J. Bone and Joint Surg.,72-A: 777-780, June 1990.72-A777
1990
Enneking, W. F.: A system of staging musculoskeletal neoplasms. Clin. Orthop.,204: 9-24, 1986.2049
1986
[PubMed]
Freiberg, A. A.; Loder, R. T.; Heidelberger, K. P.; and Hensinger, R. N.: Aneurysmal bone cysts in young children. J. Pediat. Orthop.,14: 86-91, 1994.1486
1994
Hudson, T. M.: Scintigraphy of aneurysmal bone cysts. AJR: Am. J. Roentgenol.,142: 761-765, 1984.142761
1984
[PubMed]
Jaffe, H. L., and Lichtenstein, L.: Solitary unicameral bone cyst with emphasis on the roentgen picture, the pathologic appearance and the pathogenesis. Arch. Surg.,44: 1004-1025, 1942.441004
1942
Kaplan, E. L., and Meier, P.: Nonparametric estimation from incomplete observations. J. Am. Statist. Assn.,53: 457-481, 1958.53457
1958
Koskinen, E. V. S.; Visuri, T. I.; Holmström, T.; and Roukkula, M. A.: Aneurysmal bone cyst. Evaluation of resection and curettage in 20 cases. Clin. Orthop.,118: 136-146, 1976.118136
1976
[PubMed]
Kuettner, K. E.; Pauli, B. U.; and Soble, L.: Morphological studies on the resistance of cartilage to invasion by osteosarcoma cells in vitro and in vivo. Cancer Res.,38: 277-287, 1978.38277
1978
[PubMed]
Marcove, R. C.; Sheth, D. S.; Takemoto, S.; and Healey, J. H.: The treatment of aneurysmal bone cyst. Clin. Orthop.,311: 157-163, 1995.311157
1995
[PubMed]
Mirra, J. M.: Aneurysmal bone cyst. In Bone Tumors. Clinical, Radiologic, and Pathologic Correlations, edited by J. M. Mirra, P. Picci, and R. H. Gold. Ed. 2, pp. 1267-1311. Philadelphia, Lea and Febiger, 1989.
Morton, K. S.: Aneurysmal bone cyst: a review of 26 cases. Canadian J. Surg.,29: 110-115, 1986.29110
1986
Murphy, W. A.; Strecker, W. B.; and Schoenecker, P. L.: Transcatheter embolisation therapy of an ischial aneurysmal bone cyst. J. Bone and Joint Surg.,64-B(2): 166-168, 1982.64-B(2)166
1982
Ozaki, T.; Hillmann, A.; Lindner, N.; and Winkelmann, W.: Aneurysmal bone cysts in children. J. Cancer Res. and Clin. Oncol.,122: 767-769, 1996.122767
1996
Ruiter, D. J.; van Rijssel, T. G.; and van der Velde, E. A.: Aneurysmal bone cysts: a clinicopathological study of 105 cases. Cancer,39: 2231-2239, 1977.392231
1977
[PubMed]
SAS Language Guide for Personal Computers, 6.10. Cary, North Carolina, SAS Institute, 1998.
Schreuder, H. W. B.; Veth, R. P. H.; Pruszczynski, M.; Lemmens, J. A. M.; Koops, H. S.; and Molenaar, W. M.: Aneurysmal bone cysts treated by curettage, cryotherapy and bone grafting. J. Bone and Joint Surg.,79-B(1): 20-25, 1997.79-B(1)20
1997
Szendroi, M.; Cser, I.; Konya, A.; and Renyi-Vamos, A.: Aneurysmal bone cyst. A review of 52 primary and 16 secondary cases. Arch. Orthop. and Trauma Surg.,111: 318-322, 1992.111318
1992
Tillman, B. P.; Dahlin, D. C.; Lipscomb, P. R.; and Stewart, J. R.: Aneurysmal bone cyst: an analysis of ninety-five cases. Mayo Clin. Proc.,43: 478-495, 1968.43478
1968
[PubMed]
Tsai, J. C.; Dalinka, M. K.; Fallon, M. D.; Zlatkin, M. B.; and Kressel, H. Y.: Fluid-fluid level: a nonspecific finding in tumors of bone and soft tissue. Radiology,175: 779-782, 1990.175779
1990
[PubMed]
Vergel De Dios, A. M.; Bond, J. R.; Shives, T. C.; McLeod, R. A.; and Unni, K. K.: Aneurysmal bone cyst. A clinical pathologic study of 238 cases. Cancer,69: 2921-2931, 1992.692921
1992
[PubMed]