Randomized comparison of cooked and noncooked diets in patients undergoing remission induction therapy for acute myeloid leukemia.

Changes clinical practice - Patients with acute leukemia undergoing induction chemotherapy may eat a diet with fresh fruits and vegetables.

The article indicates that rates of fever, infection, and survival are comparable between patients receiving cooked or uncooked diets. Thus, the study questions a long held belief that patients with acute myeloid leukemia (AML) need to be maintained on a special neutropenic diet. In my experience, compliance is often difficult with a cooked diet and patients miss the fruits and vegetables, another "loss" at a difficult time. Thus, the ability to liberalize diet would make a major improvement in quality of life.

The authors report the experience of 153 patients with AML receiving induction therapy. Patients were randomized to a typical neutropenic cooked diet with no fresh fruits and vegetables or to a more liberalized diet with fresh fruit and vegetables. There were no difference in rates of infection, fever, or survival between the two groups.

The study shows that clinical practice guidelines are often not based on sound scientific data and are best studied in a randomized setting. The study indicates that patient outcome is not affected by a more liberal diet, which is likely to be more pleasing to patients. The study is limited by the following features: (1) compliance to the diet on either arm was often difficult to document, given the nature of the study; (2) only 153 patients of the 404 potentially eligible patients were randomized; (3) survival results were poor overall.

For patients with acute myeloid leukemia and high-risk myelodysplastic syndrome receiving remission induction therapy, a neutropenic diet consisting of no fresh fruits and vegetables did not reduce the risk of a major infection in this study.

The neutropenic diet, which consists of restrictions of fresh fruits and vegetables, has been recommended for patients receiving highly myelosuppressive chemotherapy. However, the efficacy of such an intervention was not definitive. This trial was a randomized controlled study in which patients with acute myeloid leukemia (AML) or high-risk myelodysplastic syndrome (MDS) who were admitted to a protective environment for remission induction therapy were eligible. Randomization was stratified according to an early risk of mortality (ERM) score. All subjects received antibacterial and antifungal prophylaxis. The primary outcome was major infection defined as pneumonia, bacteremia or fungemia. A total of 153 patients were randomized, 78 to the cooked food arm and 75 to the raw food arm. The groups were similar with respect to age, ERM score and chemotherapy. 29% in the cooked food group and 35% in the raw food group developed a major infection (p=0.6). Fever of unknown origin was documented in 51% in the cooked food group and 36% in the raw food group (p=0.07). The neutropenic diet does not reduce major infections and should not be recommended in this setting. Whether or not this finding is generalizable to patients not in a protective environment or to those not receiving antibacterial and antifungal prophylaxis is not known. For further reading, please see {1}.

References: {1} van Tiel et al. Ann Oncol 2007, 18:1080-4 [PMID:17369599].