Fluoropyrimidine-associated cardiotoxicity: revisited

Background: The syndrome of 5-fluorouracil (5-FU)-associated cardiotoxicity remains poorly defined. Patients and methods: We performed a literature review (1969 - 2007) and compiled data derived from 377 evaluable cases out of 448 reported cases. Results: Patient age ranged from 14 to 86 years. Of the patients 65% were 55 years old and the male:female ratio was 1.5:1. The most commonly treated tumors were gastrointestinal (60%), head and neck (22%) and breast (4%). Of the patients 14% had a history of heart disease whereas cardiac risk factors were found in 37%. Mode of administration included: continuous infusion (72%); bolus (22.5%); intermediate infusion (3%); oral (2%); and intraperitoneal (1 patient). The dosages of 5-FU used were < 750 mg/m(2)/day (36%), 751 - 999 (16%), 1,000 (26%), 1,001 - 1,499 (4%) and 1,500 (16%). Of the patients 54% received 5-FU in combination with other chemotherapeutic agents (cisplatin 44%) whereas 51% received 5-FU alone or with leucovorin. Only 4% patients had undergone previous or concomitant radiation therapy to the mediastinum. Of cardiac incidents that happened 69% were seen during or within 72 h of the first cycle of 5-FU. Angina occurred in 45% of patients whereas myocardial infarction was seen in 22%, arrhythmias in 23, acute pulmonary edema in 5, cardiac arrest and pericarditis in 1.4 and heart failure in 2. Electrocardiographic evidence of ischemia or ST-T changes were recorded in 69% of patients, but abnormal cardiac enzymes were found in only 12%. The cardiac symptoms were reproducible in 47%, including in one patient subsequently treated with 5-FU p.o. Symptoms were also elicited when the same patients were treated with lower doses or different schedules. Of the patients 68% responded to conservative anti-anginal therapy, although prophylactic coronary vasodilators had limited efficacy. Overall, 8% of patients showing cardiotoxicity on 5-FU administration died. Furthermore, 13% reexposed to 5-FU died. Conclusions: Our review suggests that 5-FU cardiotoxicity is an infrequent but real phenomenon that is independent of dose and may be related to a continuous infusion schedule. The presence of cardiac risk factors is not predictive. Patients should be observed closely and 5-FU administration discontinued if cardiac symptoms develop. A rechallenge with 5-FU should be reserved only for those patients in whom there is no reasonable alternative therapy and should be performed in the setting of aggressive prophylaxis and close monitoring.