van Dalen, E.C., Mank, A., Leclercq, E., Mulder, R.L., Davies, M., Kersten, M.J., & van de Wetering, M.D. (2012). Low bacterial diet versus control diet to prevent infection in cancer patients treated with chemotherapy causing episodes of neutropenia. Cochrane Database of Systematic Reviews, 9, CD006247.

To determine the efficacy of a low bacterial diet (LBD) versus a control diet in preventing the occurrence of infection and reducing related mortality in patients with cancer receiving immunosuppressive chemotherapy.

Databases searched were the Cochrane Central Register (CENTRAL), DARE, PubMed, EMBASE, and CINAHL, as were conference proceedings from multiple professional groups.

Included in the study were patients with cancer receiving chemotherapy causing episodes of neutropenia, use of an LBD versus a control diet, with an LBD defined as any diet intended to reduce the ingestion of bacterial and fungal contaminants by exclusion of uncooked fruits and vegetables, cold cuts, undercooked eggs and meat, unsterilized water, unpasteurized milk products, and soft cheeses. The control diet was any other diet.

Children younger than 1 year were excluded from the study.
 

Six hundred nineteen total references were retrieved.

Risk of study bias was evaluated using the Cochrane Handbook for Systematic Reviews of Interventions.

• After exclusion, the final number of studies examined was three.
• The sample range across studies was 19 to 153.
• Samples included patients treated with remission induction chemotherapy for acute myeloid leukemia, high-risk myelodysplastic disease, or acute lymphoblastic leukemia.
• One study had a pediatric population.

• Patients were undergoing the active antitumor treatment phase of care.
• The study has clinical applicability for pediatrics.

Included studies had different definitions of infection rate and different outcomes measured and defined. Blinding and selection bias were problems in the study design, and only one study provided explicit data on the use of empirical antibiotics and antimycotics. Data could not be pooled for meta-analysis. In all three studies, there was no significant difference in outcomes between groups.

There is currently no strong evidence demonstrating the need or effectiveness of LBDs, and due to differing outcome measures, diets used and cointerventions for prophylaxis pooling of results was not possible. No firm conclusions can be drawn, and no recommendations for clinical practice are made.

• Small number of studies
• Various study design limitations and risks of study bias

The results suggest that no firm conclusions can be made about the usefulness of an LBD and that there is no strong evidence to show the effect. Additional well-designed research in this area would be helpful.