OBJECTIVES: To evaluate the technologies used to reduce anthracycline-induced cardiotoxicity in children. Also to evaluate cardiac markers to quantify cardiotoxicity, and identify cost-effectiveness studies and future research priorities. DATA SOURCES: Eight electronic databases were searched from inception to January 2006. Bibliographies of related papers were assessed for relevant studies and experts contacted to identify additional published references. REVIEW METHODS: A systematic review of the evidence was undertaken using a priori methods. RESULTS: Four randomised controlled trials (RCTs) met the inclusion criteria of the review, each considering a different cardioprotective intervention; all trials included children with acute lymphoblastic leukaemia, and one also included children with non-Hodgkin's lymphoma. However, all had methodological limitations. No cost-effectiveness studies were identified. One RCT and six cohort studies on the use of cardiac markers met the inclusion criteria of the review, but also had methodological limitations. Of the two RCTs that considered continuous infusion versus bolus (rapid) infusion, one found that continuous infusion of doxorubicin did not offer any cardioprotection over bolus; the other suggested that continuous infusion of daunorubicin had less cardiotoxicity than bolus. Two studies considered cardioprotective agents, one concluded that dexrazoxane prevents or reduces cardiac injury without compromising the antileukaemic efficacy of doxorubicin and the other reported a protective effect of coenzyme Q10 on cardiac function during anthracycline therapy. One RCT suggested that cardiac troponin T can be used to assess the effectiveness of the cardioprotective agent dexrazoxane. Two cohort studies considering atrial natriuretic peptide and two considering brain (B-type) natriuretic peptide suggested that these chemicals are elevated in some subgroups of children treated with anthracyclines for cancer. N-terminal B-type natriuretic peptide levels were significantly elevated in children treated with anthracyclines who had cardiac dysfunction. One cohort study found that serum lipid peroxide was higher in younger children treated with doxorubicin than correspondingly aged children not receiving doxorubicin. No differences in carnitine levels were found in children treated with doxorubicin and a group of healthy children in one cohort study. CONCLUSIONS: It is difficult to draw conclusions about the effectiveness of technologies for reducing or preventing cardiotoxicity and about the use of cardiac markers in children as the evidence is limited in quantity and quality. The lack of standardisation for monitoring and reporting cardiac performance is problematic. Not all studies report effectiveness in terms of cardiac outcomes and event-free survival with supporting statistical analyses. Studies are mostly small and of short duration, making generalisation difficult. Increasing numbers of survivors of childhood cancer treated with anthracyclines will experience cardiac damage and require long-term surveillance and management. This will have an impact on cardiac services and costs. Diverse medical problems and other late sequelae that affect cardiac outcome will have an impact on other specialist services. Mechanisms to reduce or prevent cardiotoxicity from anthracycline therapy and cardiac markers to improve monitoring could alter the extent of this impact on service provision. RCTs of the different methods for reducing or preventing cardiotoxicity in children treated with anthracyclines for cancer with long-term follow-up are needed to determine whether the technologies influence the development of cardiac damage. Cost-effectiveness research is also required.