Restriction of raw fruits and vegetables (neutropenic diet) is advised for patients receiving treatment for acute leukaemia in low-income and middle-income countries (LMICs) to reduce infections despite evidence to the contrary from high-income countries. We, therefore, conducted a randomised controlled trial to ascertain the efficacy of the neutropenic diet in an LMIC setting.

Patients aged 1-60 years receiving induction chemotherapy for acute leukaemia were randomised to a regular or neutropenic diet. The study's primary objective was to compare the incidence of major infections among patients receiving the two diets during induction chemotherapy. The secondary objectives were to compare stool microbial flora and induction mortality rates.

We randomised 200 patients, 98 patients to the regular diet arm and 102 to the neutropenic diet arm. Major infections occurred in 32 (32%) patients in the regular diet arm and 26 (25%) patients in the neutropenic diet arm (p=0.26). There were no statistically significant differences between patients receiving a regular diet versus neutropenic diet for blood culture positivity (n=6 vs 9), inotropic support (17 vs 12), mechanical ventilation (8 vs 5), third-line antibiotic use (28 vs 20), minor infections (12 vs 9), induction mortality (9 vs 4) and remission status (94% vs 94%). The stool culture on day 15 of induction grew multidrug-resistant bacteria in 38% of patients in the regular diet arm and 35% in the neutropenic diet arm (p=0.67).

A neutropenic diet did not prevent infections, reduce mortality or change stool microbial flora in patients with acute leukaemia.