Combined loss of heterozygosity (LOH) on 1p and 19q is reported in 50% to 90% of oligodendroglial tumors and has emerged as a strong and favorable prognostic factor. Fluorescence in situ hybridization (FISH) and polymerase chain reaction (PCR) are the most widely used techniques. The aim of this study was to evaluate the reliability of FISH to predict LOH at 1p and 19q when performed on touch preparations from 40 oligodendroglial tumors, even if the majority of the nuclei showed chromosomal imbalance. PCR was used as the gold standard. The presence of none or one target signal was reported as FISH-LOH, whereas all other losses were defined as FISH-imbalance. The sum of nuclei with FISH-LOH and imbalance was calculated in each case (FISH-sum) and cut-off values were defined as the mean FISH-sum value in controls plus 3 standard deviations; 27.7% for 1p and 33.2% for 19q. These corresponded well with the optimal cut-off values for our data, calculated using the minimum error rate classification procedure (35.6% for 1p and 33.1% for 19q). Concurrent FISH and PCR results were encountered in 95% for 1p and 87.5% for 19q. FISH-sum was the best and simplest discriminating variable for correct classification of LOH status. Under these conditions, even a dominant population of nuclei showing FISH-imbalance represented an LOH status in the tumor cells. FISH on touch preparations is a quick and reliable method for 1p/19q testing, does not require normal DNA and can be easily performed in an immunohistochemistry unit.