Consequences of local recurrence of soft tissue sarcoma: 205 patients from the Scandinavian Sarcoma Group Register

From the Scandinavian Sarcoma Group Register, information on 1,224 surgically-treated patients with soft tissue sarcoma (STS) of the extremity or trunk wall, diagnosed between 1987 and 1995, was collected. 205 patients, one third of whom were referred to a center with a local recurrence, had a total of 284 local recurrences. This analysis describes the treatment for these local recurrences, complications and risk of further recurrences. 169 patients were surgically treated for their first local recurrence. An intralesional or marginal margin was achieved in 110 of these patients, 59 of whom were also given radiotherapy. 54 of the 169 patients had a second local recurrence. The second local recurrence rate was 0.50 if the first local recurrence had been treated with only surgery with a marginal margin, compared to 0.28 if treated with surgery with a marginal margin and radiotherapy or with a wide margin (p = 0.0008). In extremity STS, the crude amputation rate for local recurrences was 0.22 (31 of 142)-i.e., higher than for primary tumors 0.09 (96 of 1065) (p < 0.0001). A high local recurrence rate after treatment outside of sarcoma centers has earlier been shown. We conclude that the consequences of local recurrence in terms of morbidity and costs justifies referral of STS patients for multidisciplinary evaluation and multimodality treatment.

Growth factor-induced morphological, physiological and molecular characteristics in cerebral endothelial cells

The capacity of vascular endothelial cells to modulate their phenotype in response to changes in environmental conditions is one of the most important characteristics of this cell type. Since different growth factors may play an important signalling role in this adaptive process we have investigated the effect of endothelial cell growth factor (ECGF) on morphological, physiological and molecular characteristics of cerebral endothelial cells (CECs). CECs grown in the presence of ECGF and its cofactor heparin exhibit an epithelial-like morphology (type I CECs). Upon removal of growth factors, CECs develop an elongated spindle-like shape (type II CECs) which is accompanied by the reorganization of actin filaments and the induction of alpha-actin expression. Since one of the most important functions of CECs is the creation of a selective diffusion barrier between the blood and the central nervous system (CNS), we have studied the expression of junction-related proteins in both cell types. We have found that removal of growth factors from endothelial cultures leads to the downregulation of cadherin and occludin protein levels. The loss of junctional proteins was accompanied by a significant increase in the migratory activity and an altered protease activity profile of the cells. TGF-beta1 suppressed endothelial migration in all experiments. Our data provide evidence to suggest that particular endothelial functions are largely controlled by the presence of growth factors. The differences in adhesiveness and migration may play a role in important physiological and pathological processes of endothelial cells such as vasculogenesis or tumor progression.

Cytological diagnosis of skeletal lesions. Fine-needle aspiration biopsy in 110 tumours

We have previously shown that cytological diagnosis based on fine-needle aspiration biopsy (FNAB) is a safe and efficient method for the discrimination between benign, primary malignant and metastatic bony lesions. We have now studied metastatic lesions to assess the diagnostic accuracy and to ascertain whether FNAB allows identification of the primary lesion. Between 1990 and 1997, 447 patients were referred for diagnosis of skeletal lesions of unknown type. Of these 119 proved to have metastatic disease, either myeloma or lymphoma. Nine were excluded leaving 110 consecutive patients with metastatic carcinoma (80), myeloma (16) or lymphoma (14). FNAB gave a correct diagnosis in 102 of the 110 patients (93%). In eight it was inconclusive. It correctly diagnosed 15 of 16 patients with myeloma, 12 of 14 with lymphoma, and 75 of 80 with metastatic carcinoma. Furthermore, the site and type of malignancy were correctly suggested in two-thirds of patients with metastatic carcinoma. Overall, only seven open biopsies were carried out. We conclude that time-consuming and costly investigations can be reduced by choosing FNAB as the initial diagnostic method for skeletal lesions of unknown origin. The choice of radiological examinations, laboratory tests and surgical biopsies can be determined by using FNAB.

Surgical margins, local recurrence and metastasis in soft tissue sarcomas: 559 surgically-treated patients from the Scandinavian Sarcoma Group Register

The prognostic importance of surgical margins on local recurrence rates and metastasis-free survival (MFS) was studied in 559 patients with soft tissue sarcoma of the extremities and trunk wall. The patients were all surgically treated, but received no adjuvant treatment. The median follow-up for the survivors was 7.4 (range: 0.1 - 12.5) years. Independent prognostic factors for MFS were analysed by Cox models. The overall 5-year MFS was 0.72 (95% confidence intervals (CI) 0.68 - 0.76). High histopathological malignancy grade (relative risk (RR) 3.0; 95% CI 1.5 - 6.3) and an inadequate surgical margin (RR 2.9; 95% CI 1.8 - 4.6) were independent risk factors for local recurrence. High histopathological malignancy grade and large tumour size (> 7 cm) were the most important risk factors for metastasis. Local recurrence was associated with an increased risk of metastasis (RR 4. 4; 95% CI 2.9-6.8), but an inadequate surgical margin was not a risk factor for metastasis (RR 1.1; 95% CI 0.8-1.7). This study confirms that, as regards metastasis, tumour-related risk factors (malignancy grade and tumour size) are more important risk factors than treatment-related factors. Local recurrence was associated with an increased metastasis rate, whereas inadequate surgical margin was a risk factor for local recurrence but not for metastasis. Hence, the proposed causal association between local recurrence and metastasis is doubtful, and if it exists is a weak association.

Neural induction of the blood-brain barrier: still an enigma

1. The study of the blood-brain barrier and its various realms offers a myriad of opportunities for scientific exploration. This review focuses on two of these areas in particular: the induction of the blood-brain barrier and the molecular mechanisms underlying this developmental process. 2. The creation of the blood-brain barrier is considered a specific step in the differentiation of cerebral capillary endothelial cells, resulting in a number of biochemical and functional alterations. Although the specific endothelial properties which maintain the homeostasis in the central nervous system necessary for neuronal function have been well described, the inductive mechanisms which trigger blood-brain barrier establishment in capillary endothelial cells are unknown. 3. The timetable of blood-brain barrier formation is still a matter of debate, caused largely by the use of varying experimental systems and by the general difficulty of quantitatively measuring the degree of blood-brain barrier "tightness." However, there is a general consensus that a gradual formation of the blood-brain barrier starts shortly after intraneural neovascularization and that the neural microenvironment (neurons and/or astrocytes) plays a key role in inducing blood-brain barrier function in capillary endothelial cells. This view stems from numerous in vitro experiments using mostly cocultures of capillary endothelial cells and astrocytes and assays for easily measurable blood-brain barrier markers. In vivo, there are great difficulties in proving the inductive influence of the neuronal environment. Also dealt with in this article are brain tumors, the least understood in vivo systems, and the induction or noninduction of barrier function in the newly established tumor vascularization. 4. Finally, this review tries to elucidate the question concerning the nature of the inductive signal eliciting blood-brain barrier formation in the cerebral microvasculature.

Zinc protects against apoptosis of endothelial cells induced by linoleic acid and tumor necrosis factor alpha

BACKGROUND

Zinc requirements of the vascular endothelium may be increased in inflammatory conditions, ie, atherosclerosis, in which apoptotic cell death is prevalent.

OBJECTIVE

We hypothesized that zinc deficiency may potentiate disruption of endothelial cell integrity mediated by fatty acids and inflammatory cytokines by enhancing pathways that lead to apoptosis and up-regulation of caspase genes.

DESIGN

Endothelial cells were maintained in low-serum medium or grown in culture media containing selected chelators, ie, diethylenetriaminepentaacetate or N,N,N', N'-tetrakis(2-pyridylmethyl)-ethylenediamine (TPEN), with or without zinc supplementation. Subsequently, cells were treated with linoleic acid, tumor necrosis factor alpha (TNF-alpha), or both. We studied the effect of zinc deficiency and supplementation on the induction of apoptosis by measuring caspase-3 activity, cell binding of annexin V, and DNA fragmentation.

RESULTS

Our results indicated that linoleic acid and TNF-alpha independently, but more markedly in concert, up-regulated caspase-3 activity and induced annexin V binding and DNA fragmentation. Zinc deficiency, especially when induced by TPEN, dramatically increased apoptotic cell death induced by cytokines and lipids compared with control cultures. Supplementation of low-serum- or chelator-treated endothelial cells with physiologic amounts of zinc caused a marked attenuation of apoptosis induced by linoleic acid and TNF-alpha. Morphologic changes of cells observed during zinc deficiency were prevented by zinc supplementation. Media supplementation with other divalent cations (eg, calcium and magnesium) did not mimic the protective role of zinc against apoptosis.

CONCLUSIONS

Our data indicate that zinc is vital to vascular endothelial cell integrity, possibly by regulating signaling events to inhibit apoptotic cell death.

Clinical course in synovial sarcoma: a Scandinavian sarcoma group study of 104 patients

We analyzed treatment and outcome in 104 Scandinavian patients with synovial sarcoma in the extremities or trunk wall, diagnosed between 1986 and 1994. Only surgically treated patients without metastases at diagnosis were included. Median follow-up of survivors was 6 (3-11) years. 34 patients developed metastases. The overall 5- and 7-year survival rates were 0.76 (95% CI 0.66-0.83) and 0.69 (0.58-0.78), respectively. Large tumor size and amputation were significantly associated with impaired metastasis-free survival. Patients with local recurrence had a higher risk of metastases following the local event. Local excision with inadequate margin was associated with a higher risk of local recurrence.

110 subfascial lipomatous tumors. MR and CT findings versus histopathological diagnosis and cytogenetic analysis

PURPOSE

To evaluate whether liposarcoma, atypical lipomatous tumors and lipoma can be differentiated radiologically.

MATERIAL AND METHODS

We have retrospectively analyzed CT and/or MR images of 110 subfascial lipomatous lesions. The amount of fat within the tumors was visually graded from the images as: none, 1-75%, 75-95% or 95-100%. The structure of non-fatty tumor components was compared. The images were compared to histopathology and in 37 cases to cytogenetic findings.

RESULTS

Only 4 of 20 liposarcomas contained fat. All 4 lesions, histopathologically diagnosed as atypical lipomatous tumors, contained fat but less than 75% of tumor volume. All lesions with more fat than 75% of tumor volume were histologically diagnosed as lipomas. However, one-third of the karyotyped lipomas had ring chromosomes which are considered typical for atypical lipomatous tumors.

CONCLUSION

When a tumor is composed more or less solely of fat, the diagnosis of a lipoma or atypical lipomatous tumor with a phenotype simulating a lipoma can be assumed. When the fat content is less than 75% of the tumor volume or non-fatty nodules are found, biopsies from different tumor components have to be performed to exclude malignancy. When no fat is found, imaging does not help in differentiating lipoma or liposarcoma from other soft tissue tumors.

Clinical significance of genetic imbalances revealed by comparative genomic hybridization in chondrosarcomas

DNA copy number changes were studied by comparative genomic hybridization (CGH) in 50 chondrosarcoma samples from 45 patients. Mean number of genetic aberrations in primary tumors was 4.8 +/- 1.8. The most frequently gained regions were 20q12-qter (37%), 20q (32%), 8q24.1-qter (27%), 20p (24%), and 14q24-qter (24%). Losses were 5.5 times less frequent than gains and observed mainly at Xcen-q21, 6cen-q22, and 18cen-q11.2 (11% each). Recurrent and metastatic tumors showed a mean of 4.0 +/- 2.2 aberrations per sample. The most frequently gained regions were chromosome 7 (4 cases), 5q14-q32 (4 cases), 6p (3 cases), and 12q (3 cases). Losses of DNA sequences were 3.4 times less frequent than gains. Histological tumor grade was significantly associated with metastasis-free survival (P = .002) and overall survival (P = .003), being the strongest prognostic factor tested. A statistically significant correlation was found between gain at 8q24.1-qter and shorter overall survival (P = .01) but not with local recurrence or metastasis-free survival. Gain at 14q24-qter was associated with a trend to shorter overall survival (P = .05) but neither with an increased risk for local recurrence nor with metastasis-free survival. In a multivariate analysis, only the tumor grade associated with overall survival (P = .02). In a multivariate analysis together with the tumor grade, gain at 8q24.1-qter did not retain its significance (P = .44), indicating that this imbalance is not an independent prognostic factor.

Malignant fibrous histiocytoma of bone: a retrospective EMSOS study of 125 cases. European Musculo-Skeletal Oncology Society

In an effort to learn more about malignant fibrous histiocytoma (MFH) of bone and its prognosis with different treatment approaches, the European Musculo-Skeletal Oncology Society (EMSOS) initiated a retrospective survey among its members. Data requested included patient and treatment variables and outcome. The information on all patients with histologically proven, primary, localized osseous extremity MFH was analyzed if surgical tumor removal was performed and disease status was documented for at least one follow-up date. 125 such patients were evaluable (74 male, 51 female; median age 34 years; tumor site femur 81, tibia 26, humerus 12, other 6). Local treatment was surgery only (110) or surgery plus radiotherapy (15). Chemotherapy was used in 97/125. On last follow-up, 85 patients remained in remission, 33 had developed metastases, 6 a local recurrence, and 1 a combined relapse. With a median follow-up of 3.9 years for patients at risk, actuarial 5-year disease-free survival was 59%. In univariate analyses, younger age and the use of chemotherapy were associated with a more favorable outcome, as was limb-salvage surgery. 23 of 66 tumors with information on response to preoperative chemotherapy responded well (> 90% necrosis). Among these 23, only one relapsed.

Ki-67 is strongly prognostic in synovial sarcoma: analysis based on 86 patients from the Scandinavian Sarcoma group register

In a study based on formalin-fixed paraffin-embedded material from 86 patients with primary synovial sarcoma located in the extremities or on the trunk wall, the prognostic importance of MIB-1 index, p53-expession and tumour size was analysed. Multivariate analysis identified two metastatic risk factors: increasing tumour size and MIB-1 > 9%. The 5-year metastasis-free survival-rate for patients with tumour size < or = 5 cm + MIB-1 < 10% was 0.83 (95% confidence interval (CI) 0.64-0.92) compared to 0.31 (95% CI 0.11-0.53) in cases with tumour size > 5 cm + MIB-1 > or = 10%. Our study shows that metastatic disease in synovial sarcoma is closely related to MIB-1 index. Using our model based on tumour size and MIB-1 index, cases with good and poor prognosis can easily be discriminated. Therefore our model can be used to identify patients who should be considered for adjuvant chemotherapy.

MR imaging of lipoma and liposarcoma

PURPOSE

To evaluate whether lipoma, atypical lipomatous tumors, and liposarcomas can be differentiated by MR images.

MATERIAL AND METHODS

The MR images of 59 lipomatous lesions and liposarcomas were retrospectively reviewed. Apart from size, surgical site, location and margins, the percentage of fat of the tumor volume was assessed as none, 1-75%, 75-95%, or 95-100%.

RESULTS

None of the 18 liposarcomas contained fat that could be recognized by MR imaging. The 3 atypical lipomatous tumors all contained fat but less than 75% of the tumor volume. In 32 of 38 ordinary lipomas, the percentage of fat was 95-100%, and in 4 less than 95% of the tumor volume. Two lipomas did not contain fat that could be recognized by MR imaging.

CONCLUSION

A lesion which predominantly has a fat signal is, in all probability, an ordinary lipoma. Lesions with less fat, but still mostly fatty, may either be lipoma or atypical lipomatous tumor. In this group, the discrimination between these two entities cannot be based upon imaging features. In the absence of a fat signal, liposarcoma or lipoma cannot be differentiated from other soft tissue tumors.

Fine needle aspiration (FNA) cytology in the diagnosis of recurrent soft tissue sarcoma

We have used FNA cytology to diagnose suspected local recurrences of soft tissue sarcoma. Since 1991, a total of 95 FNA cytologies were performed on 86 patients. There were 47 local recurrences, of which 44 were diagnosed correctly by FNA cytology; one biopsy was inconclusive, and two lesions were incorrectly assessed as benign. Thirty-nine patients proved to have benign lesions in the scar area examined cytologically on 50 occasions. None of the specimens was regarded as malignant, but in four cases FNA cytology was inconclusive. Overall, there were 5% inconclusive cytological biopsies, 0% falsely malignant and 5% falsely benign. The inconclusive and false-negative cytological diagnoses had no important clinical consequences. FNA biopsy provides a simple means of diagnosing local recurrence of soft tissue sarcoma.

Cytotoxic and genotoxic effects of 4-hydroxynonenal in cerebral endothelial cells

Oxygen free radicals are produced in the central nervous system (CNS) as a consequence of normal physiological metabolic reactions of neuronal cells, but there is evidence accumulating that they are also implicated in the processes leading to a number of pathological changes in the brain. A general mechanism whereby oxygen free radicals induce tissue damage is lipid peroxidation (LPO), which generates a large variety of water-soluble carbonyl compounds. Due to their high reactivity, we focused our investigations on 4-hydroxyalkenals, in particular on 4-hydroxynonenal (HNE), the major 4-hydroxyalkenal. Two phenotypes of cerebral endothelial cells (cECs) were treated with various concentrations of 4-hydroxynonenal and the cyto- and genotoxic effects studied. The cytogenetic endpoints determined were chromosomal aberrations and the induction of micronuclei. Three hours of incubation with HNE induced significantly elevated levels of chromosomal aberrations at concentrations > or = 1 microM and micronuclei at concentrations > or = 10 microM in both cEC phenotypes, compared to the controls. Cytotoxicity was observed at a concentration of 50 microM HNE and was significantly higher in the elongated and spindle-shaped cEC phenotype (type II) than in the epithelial cEC phenotype (type I). The results indicate that cECs are affected by HNE even at low concentrations with minor differences between the two cEC phenotypes.

Comparison of technetium-99m-MIBI and technetium-99m-tetrofosmin uptake by musculoskeletal sarcomas

Technetium-99m-MIBI was initially developed for heart studies but it can also be used to depict tumors, predict multidrug resistance and evaluate chemotherapy. Recently, 99mTc-tetrofosmin, which exhibits similar physical properties, has been launched for heart studies. Tumor uptake and prediction of multidrug resistance have also been reported regarding the latter tracer. A comparison of these two tracers regarding the detectability of musculoskeletal sarcoma has been made.

METHODS

Twenty patients with musculoskeletal sarcoma of the extremities or pelvis underwent planar examination after the administration of 99mTc-MIBI and 99mTc-tetrofosmin with an interval of 2-7 days. The tumor activity was compared with one ipsilateral and one contralateral background region.

RESULTS

There was a small, but not significant, difference in favor of 99mTc-MIBI with regard to both background regions.

CONCLUSION

Technetium-99m-MIBI and 99mTc-tetrofosmin can both be used to visualize musculoskeletal sarcomas. The choice may depend on which agent is used routinely for myocardial studies in the laboratory.

Use of 99mTc-MIBI scintigraphy in the evaluation of the response of osteosarcoma to chemotherapy

The use of gamma camera scintigraphy with technetium-99m hexakis-2-methoxyisobutylisonitrile (99mTc-MIBI) for assessment of the response of high-grade osteosarcoma to preoperative chemotherapy was evaluated. Twelve patients with osteosarcoma of the extremities underwent planar examination with 99mTc-MIBI before and after preoperative chemotherapy according to the recommendations of the Scandinavian Sarcoma Group. After calculating a quotient for the tumour and the average activity of both extremities and correcting for background activity, the change in uptake between the two examinations was assessed. This was compared with histological examination of the ultimately resected specimen in 11 patients and progressive clinical disease in one. All the 11 tumours undergoing histological examination showed cellular necrosis of between 50% and 100% as well as a reduced uptake of 99mTc-MIBI, while the single progressive tumour showed an increased uptake. There was a correlation between the reduction of radiopharmaceutical uptake and the histological response in the entire series, while the variation was too large to allow conclusions in individual patients. This variation may have biological reasons or may be due to the planar imaging technique, which only allows semiquantitative evaluation. The technique reflects response to therapy but is not yet clinically applicable for the identification of poor responders, which would serve as a basis for alteration of the chemotherapy regimen. In order to evaluate whether such a role could be fulfilled, further studies using single-photon emission tomography with correction for attenuation and scattering of photons are necessary.

Posterior decompression and stabilization for spinal metastases. Analysis of sixty-seven consecutive patients

The neurological function, survival, and rehabilitation of sixty-seven consecutive patients who had been managed operatively for spinal metastases with epidural compression were assessed. The epidural compression was in the thoracic spine in forty-one patients and in the lumbar spine in twenty-six. According to the system of Frankel et al. for the assessment of neurological function, twenty-six patients had a major neurological deficit (grade B or C), thirty-two had a minor deficit (grade D), and nine had no deficit (grade E). None of the patients had an operation to treat a pathological vertebral fracture without epidural compression. The operative treatment included wide decompression through a posterior approach followed by stabilization without bone-grafting. A Cotrel-Dubousset device was used in thirty-two patients; an Olerud posterior fixator, in sixteen; an Isola device, in twelve; and another device, in seven. The most common complication was wound infection (eleven patients). There were no perioperative or immediate postoperative deaths (within fourteen days). The rate of survival was 51 per cent (thirty-four of sixty-seven) at six months and 22 per cent (fifteen of sixty-seven) at twelve months. Over-all, forty-four of the fifty-eight patients who had had a neurological deficit preoperatively had complete or partial neurological recovery within the first two weeks postoperatively. The nine patients who had not had a neurological deficit preoperatively retained normal neurological function postoperatively. Thirty-eight of the forty-four patients who were alive at three months and twenty-nine of the thirty-four who were alive at six months were still able to walk. Thirty-nine of the forty-nine patients who survived more than two months were able to return home for a median of seven months. Fourteen patients had a reoperation on the spine. Six of these patients had recurrent epidural compression at another level of the spine, and five had recurrent compression at the previously treated level. Three patients had a reoperation because of loosening of the implant. The results of this study suggest that neurological function can be maintained or improved by decompression and stabilization through a posterior approach as treatment for spinal metastases.

Incorporation of β-Galactosidase-Expressing Endothelial Cells into the Skeletal Muscle Microvascular Bed of Mice

Cloned murine endothelial cells (cEC) were used as a carrier system for introducing a foreign gene into the microvascular bed of the hind limb of inbred mice. cEC were transfected with a β-galactosidase-neo fusion construct, which enables both selection for DNA uptake in the presence of G 418 and the staining of cells for β-galactosidase activity. Transfected cEC adhered and integrated readily into confluent monolayers of nontransfected cEC (up to 26% of total cell number). Seeding lacZ-transfected cEC on explanted arteries revealed rapid adhesion of the cells (within minutes) to the intact endothelium. After injection of 106transfected EC via the femoral artery into the microvascular bed of the hind limb their presence was documented by β-galactosidase staining after various time periods (1 h to 4 wk). Implanted cEC were detected in numerous elements of the microcirculation both in frozen sections and in squash preparations of the hind limb muscle and in the femoral bone up to 4 wk after the injection. The microvascular bed of skeletal muscle of the mouse as a recipient site for transduced syngeneic endothelial cells is, thus, a suitable experimental model to study various strategies for somatic gene therapy. Copyright © 1997 Elsevier Science Inc.

Expression of a chemotactic cytokine (MCP-1) in cerebral capillary endothelial cells in vitro

A full-length cDNA encoding the porcine monocyte chemoattractant protein-1 (pMCPC-1) was isolated from growth-stimulated porcine cerebral capillary endothelial cells (cEC); the pMCP-1 cDNA showed 89% identity to human MCP-1 and was isolated by use of subtractive hybridization and differential screening of two phenotypically different sub-populations of cloned cEC. pMCP-1 was abundantly expressed in cEC grown in the presence of FCS, ECGF and heparin whereas lower expression was observed in cEC kept in FCS-supplemented medium only. As shown by Northern blot analysis, no pMCP-1 transcripts were present in total RNA derived from freshly isolated brain capillaries, large brain vessels or whole brain homogenate. MCP/JE expression was also demonstrated in ECGF/heparin-treated murine cEC. Astrocytes and smooth muscle cells grown in FCS-supplemented medium did not show MCP-1 expression. Treatment of porcine cEC with TNF-alpha increased pMCP-1 mRNA levels in a dose-dependent manner. These data further support the notion that cerebral capillary endothelial cells actively participate in processes of CNS host defense.

Reconstruction in metastatic destruction of the acetabulum. Support rings and arthroplasty in 12 patients

In 12 patients with acetabular metastases, we reconstructed the hip with a support ring, cement, and Charnley prosthesis with a 22 mm head. There were no immediate postoperative deaths or cardiovascular complications. Dislocations occurred in 5 patients, 1 of whom required open reduction. At follow-up, 3 patients were alive 11, 15, and 18 months after surgery. 9 patients died after 8 (2-13) months. All patients obtained relief from pain, became ambulatory and were discharged to their homes. 1 patient had rapid disease progression with pathologic fracture and cranial dislocation of the support ring. Among the remaining 11 patients, the hip has not again caused pain or limited function. There were no signs of loosening, except in the patient with fracture of the hemipelvis. Remineralization was observed in 2 breast cancer patients after chemotherapy. Hip reconstruction can restore painless function until death in cancer patients with severe destruction of the acetabulum.

Cytological diagnosis of bone tumours

We evaluated the diagnostic accuracy of fine-needle aspiration biopsy in a prospective study of 300 patients with previously undiagnosed bone lesions. Patients with suspected local recurrence of a primary bone tumour or a metastatic lesion of a previously diagnosed malignancy were excluded. Fine-needle aspiration biopsy was performed under radiological control as an outpatient procedure. The series was grouped into three major categories: 1) benign bone lesions including infections; 2) primary malignant bone tumours; and 3) metastases including lymphomas and myelomas. We compared the cytological diagnosis with the final diagnosis as assessed by histological examination and/or the clinical and radiological features. Material considered conclusive for cytological diagnosis was obtained from 251 of the 300 patients. Of the 49 failures, there were 24 aspirates with insufficient cellular yield and 25 in which a diagnosis could not be made although the cytological material was adequate in quantity. Most of the inconclusive aspirates (36/49) were obtained from benign bone lesions. The diagnosis was correct in 239 (95%) of the 251 cases providing adequate cytological material. There were eight (3%) falsely benign diagnoses, one (0.3%) falsely malignant, and three cases in which we were unable to differentiate between sarcoma and a metastasis. Chondrosarcoma (2/12) gave the greatest diagnostic difficulty and Ewing's sarcoma the least (0/9). There were no decisive errors of treatment. All falsely benign or malignant diagnoses were questioned, and led to open biopsy since they did not correlate with the clinical and radiological features. Our study suggests that fine-needle aspiration biopsy is a valid option for the diagnosis of bone tumours. It is a simple outpatient procedure which gives sufficient cytological material for the correct diagnosis in 80% of cases. As with histological analysis of material from open biopsy, the cytological assessment must agree with the clinical and radiological findings.