[Does the core biopsy of solid liver lesions permit an exact histological classification? Results of a prospective study under routine clinical conditions].

PURPOSE

The aim of this prospective study was to answer the question to what extent percutaneous core-cut biopsy of solid lesions of the liver can permit (sub-)classification under routine conditions.

MATERIAL AND METHODS

Subject of this study were 80 percutaneous core-cut biopsies of solid liver lesions in 75 patients. The biopsies were done consecutively under routine conditions by 10 different radiologists with Tru-Cut-needles and a biopsy gun with sonographic (n = 73) or CT (n = 7) guidance. After receiving the histological analysis, the radiologists then prospectively divided the biopsies into "valid" und "unclear" using clinical and radiologic data of the respective patient. The results were verified by a second biopsy, follow-up or surgery.

RESULTS

80% of all histological analyses were rated as "valid", 20% as "unclear". The accuracy of the "valid" reports in respect of ranking was 95% and 90% for the accurate tumour classification. These values were reduced to 81.3% and 74%, respectively, with regard to the total number of cases. The negative predictive value was reduced from 78% to 51.7%.

CONCLUSION

Core-cut biopsy of solid liver lesions guided by imaging yields good results even under routine conditions as long as all histological reports have been correlated with imaging and clinical informations.