The aim of this study was to evaluate whether intra-operative ultrasound (= IOUS) is a suitable tool to detect residual tumour tissue after gross total resection in supratentorial gliomas. During a period of 18 months 45 patients with supratentorial gliomas (38 high-grade and 9 low-grade, according to the WHO-grading system [42]) were operated on. A series of 78 biopsies was taken from the resection cavity under continuous sonographic control at the end of surgery. Gross total tumour resection was intended in 34 patients (= 76%). The biopsy specimens were matched with the sonographic features at each biopsy site. The sonographic appearance of the resection margins were classified into 2 groups: (1) Irregular hyperechoic areas extending from the cavity into the iso-echogenic brain tissue and (2) a dense small (< or = 3 mm in diameter) rather regular hyperechoic rim surrounding the resection cavity. 47 out of 53 biopsies taken from hyperechoic areas (group I) (36 high-grade/11 low-grade) revealed solid tumour tissue (= 89%). 34 (= 72%) of these 47 areas were microscopically assessed as inconspicuous by the surgeon. 6 samples (4 high-grade/2 low-grade) contained tumour infiltration zone. 25 biopsies (23 high-grade/2 low-grade) taken from the hyperechoic rim [group 2] were diagnosed as follows: Normal brain tissue in 11, tumour infiltration zone in 8 and solid tumour tissue in 6 cases. Of 34 cases with "gross total removal" according to the surgeon's assessment 25 showed sonographic signs of residual tumour tissue, which was confirmed histologically as solid tumour tissue in 22 of these cases. It is concluded, that IOUS following resection of supratentorial gliomas can detect residual tumour tissue with high specificity and thus improve gross total resection. However, a thin hyperechoic rim surrounding the resection cavity (less than 3 mm in diameter) is a non-specific finding, which can mask thin residual tumour layers and therefore needs further evaluation of its nature.

Transcranial color-coded real-time sonography (TCCS) was performed in 57 patients with primary intracranial brain tumors (n = 49) or arteriovenous malformations (n = 8) to evaluate its diagnostic potential. In 46 patients (81%), lesions could be identified employing this technique. In 7 patients, transcranial ultrasound examination was not feasible because of bone thickness; in the remaining 4 patients, the tumor was indistinguishable from adjacent brain tissue despite sufficient insonation, suggesting that these neoplasms are isoechogenic. The sonographic features of brain tumors were very similar: a hyperechogenic matrix of the lesion was interspersed by hypoechogenic pixels. Larger hypoechogenic areas (0.5-1 cm) gave evidence of tumor necrosis. Differences between the findings of TCCS and computed tomography concerning tumor size were found in 7 patients, in whom TCCS revealed an area of smaller extension within the corresponding hypodense area on the computed tomographic scan. Perifocal brain edema could not be detected by ultrasound examination. In 13 patients, a thin, hypoechogenic peritumoral halo was disclosed that did not correlate with perifocal brain edema identified by computed tomography and that may have been due to compression of adjacent parenchyma. In patients with arteriovenous malformations, TCCS permitted the identification of the main feeders, the nidus, and the draining venous system by color-coded depiction of intravascular blood flow. In conclusion, TCCS is an additional method for initial diagnosis and highly suitable for follow-up in tumor patients and provides valuable information about tissue characteristics and blood flow.

Transcranial colour-coded real-time sonography (TCCS) was carried out in 25 patients with brain tumours to determine whether this noninvasive method provides additional information about the extent of solid tumour, its differentiation from oedema, and its tissue components. All 25 patients had serial computed tomography (CT)-guided stereotactic biopsies. Comparison of ultrasound, CT and histological findings revealed that the vast majority of contrast enhancing areas on CT were hyperechogenic (32/33; 97%) and contained tumour tissue (29/32; 91%). Hyperechogenic areas always represented solid tumour (23/23 patients), even when CT showed low density non-enhancing lesions. In lesions hypoechogenic on TCCS and low density on CT, histology consistently revealed necrotic tumour (7/7). Biopsies obtained from parenchyma with normal echogenicity revealed tumour in only 3 of 16 specimens. Despite the high specificity of TCCS in the differentiation of tumour components, its sensitivity to tumour was inferior to that of CT (24/25; 96%). TCCS thus allows noninvasive preoperative identification of tumour tissue and its extent setting.

No abstract

In this prospective study the role of intra-operative Colour-Duplex-Sonography (= CDS) during surgery of arteriovenous malformations (= AVM) is evaluated. During the last three years 20 consecutive patients with supratentorial AVMs were examined by intra-operative CDS in order to evaluate the potential of CDS to 1) localize the AVM, 2) differentiate between embolized and perfused parts, 3) identify feeding and draining vessels and 4) control the complete excision of the AVM. All AVMs were localized supratentorially, 9 were grade I and II (according to Spetzler and Martin [31]), 8 grade III and 3 grade IV. 11 were partly embolized and 8 associated with an intracerebral bleeding. In all cases the nidus was correctly localized sonographically by its typical bidirectional flow pattern in Colour-mode. CDS guided the surgeon directly to all (11 cases) deep-seated AVMs (2 to 4 cm subcortically). The smallest nidus measured 10 mm. 28 of 34 angiographically defined main feeding and 18 of 23 draining vessels were identified. 14 patients were controlled sonographically at the end of the resection regarding the completeness of excision. In 11 patients CDS was negative and was confirmed by either postoperative angiography or MRI in 10 patients. In one case residual AVM tissue was missed by CDS. Positive CDS findings in 3 cases were all confirmed by microscopic re-inspection, angiography and CCT. Our results suggest that CDS is able to localize AVMs intra-operatively with minimal instrumentation. It allows safe navigation to deep-seated malformations with high accuracy. Feeding and draining vessels can be identified and completeness of resection can be controlled.