The Utility of Radionuclide Ventriculography (RNV) Prior to Anthracycline Chemotherapy in Patients with Acute Myeloid Leukemia: A Retrospective, Single Institution Study.

Background: Despite cardiotoxicity, anthracyclines remain one of the most important classes of cytotoxic agents in the management of hematological malignancy. For over two decades, monitoring of left ventricular ejection fraction (LVEF) has been recommended to identify patients at risk of developing overt cardiac failure. As anthracycline cardiotoxicity is closely correlated with cumulative dose, and current chemotherapy protocols limit exposure to these agents, we sought to investigate whether routine LVEF estimation by RNV remains necessary in patients undergoing treatment for acute myeloid leukemia (AML). We also sought to determine whether subgroups of patients could be identified who were at higher or lower risk of cardiotoxicity.

Methods: Between 2000 and 2006, 104 patients underwent induction chemotherapy for AML at the Alfred Hospital. Retrospective review of prior cardiac history, cumulative anthracycline dose and results of RNV assessment was undertaken. RNV was performed according to established protocols in the best left anterior oblique projection to obtain maximal septal separation between left and right ventricles allowing a robust and reproducible measure of LVEF.

Results: Median patient age was 59 years (range 17–74); 50% were female. 4 patients had prior exposure to anthracylines and one patient had a history of cardiac disease. Induction chemotherapy consisted of cytosine arabinoside (standard dose 93%; high dose 7%) combined with an anthracycline (idarubicin 94%, daunorubicin 5%, amsacrine 1%). Of the 104 patients who underwent induction chemotherapy, 25 (24%) had one and 47 (45%) had two courses of cytosine arabinoside based consolidation chemotherapy. All patients had LVEF measured by RNV prior to induction. 62% and 32% of patients had repeat assessment prior to their first and second consolidation chemotherapy cycles respectively. For all patients receiving idarubicin, the cumulative dose was < 60 mg/m² in 43% and ≥60 mg/m² in 57%. Five patients had chemotherapy management changed as a result of RNV monitoring: one did not receive an anthracycline due to baseline LVEF impairment and four patients had amsacrine in place of idarubicin after induction due to a reduction in LVEF. Six patients in total had a reduction in LVEF of greater than 20%, however none developed clinically significant cardiac decompensation. Two of these had risk factors for anthracycline induced cardiotoxicity: one prior anthracycline treatment with mediastinal radiotherapy and one a pre-existing reduction in LVEF. Cumulative idarubicin dose, cytosine arabinoside dose, age or gender did not predict for a fall in LVEF.

Conclusion: Despite minimal overall anthracycline cardiotoxicity, clinically significant falls in LVEF occurred in 6% of patients and was largely unpredictable. However, even though RNV enabled this group of patients to be easily identified the optimal management of patients with asymptomatic falls in LVEF remains to be determined.