RESUMEN
BACKGROUND: Many haematology/oncology departments still provide a germ-free diet for neutropenic patients (neutropenic diet, ND) to minimise pathogen exposure, even though evidence on benefits is missing. We analysed the effects of a standard diet (SD) in neutropenic high-risk patients with cancer while focussing on infection-related outcomes. PATIENTS AND METHODS: Based on the Cologne Cohort of Neutropenic Patients, we conducted a propensity score-matched case-control study in haematological/oncological patients with a period of neutropenia longer than five days treated at our department between January 2004 and December 2012 (implementation of SD in January 2008). We assessed the association between an SD and selected infection-related end-points in an adjusted multivariable regression model and time-to-event analysis. RESULTS: In total, 2086 neutropenic episodes (1043 per diet group) were included into analysis. The median days of neutropenia were 9 (interquartile range 7-16). The adjusted multivariable model revealed no association between the SD and severity and persistence of fever, death within 28 days, antibiotic treatment and weight loss >3 kg and a non-significant adjusted association between SD and duration of antibiotic treatment and blood stream infections. There was a significant association between SD and incidence of diarrhoea (odds ratio [OR], 0.55; 95% confidence interval [CI], 0.45-0.68; P < 0.001), nausea (OR, 0.53; 95% CI, 0.43-0.66; P < 0.001) and weight loss >1 kg (OR, 0.93; 95% CI, 0.89-0.98; P = 0.002) with fewer events in SD than in the ND group. The hazard ratios of SD for the analysed end-points were non-significant. CONCLUSION: In our study, the implementation of an SD for high-risk neutropenic patients with cancer was safe regarding infection-related end-points.