If the 135 patients with BNP.30 pg/ml were subjected to MUGA and LVEF,50% again was considered reference value, then only 8 of the 21 heart failure patients would have been identified. Another possibility could be to use BNP as an additive measure together with LVEF-measurement in the surveillance of cardiotoxic cancer treatment. If these two values were combined 18 of the 21 patients with CHF were identified suggesting that BNP possess meaningful and additive information to the current surveillance based on MUGA. However, by combining BNP and MUGA only 1 more patient was identified compared to BNP.30 pg/ml alone and 150 patients had either BNP.30 pg/ml or EF,50, increasing the number of ����false positives���� compared to BNP.30 pg/ml as the only measure. It was also speculated if serial measurements of BNP could have improved the diagnostic performance of chemotherapy-induced CHF, since it is well known that repeated measurements in patients with chronic heart failure provide incrementally unique information. Unfortunately, this was only done in 73 of the patients, of whom 5 patients developed Cyamemazine congestive heart failure, making the material too small for meaningful statistical analyses. Several published DIM-C-pPhtBu studies have investigated the role of BNP in diagnosing and risk stratification of chemotherapy-associated cardiotoxicity, however most studies have simply compared BNP levels with LVEF measured by MUGA or echocardiography at baseline or during/after treatment with antracyclines as we did also in our previous publication. Results have shown either no association or a correlation between impaired ejection fraction and increased levels of BNP. There are multiple reasons for these conflicting results including inadequate sample size, heterogeneity of the studied population with regards to cancer diagnosis, age, chemotherapeutic agent and treatment regime/cumulative dose. Furthermore, the timing of blood samples drawn for BNP measurements, the laboratory methods and cut off values contributes to the inconsistency of the results.