The article by Fox and Freifeld presents a comprehensive history of the rationale for the neutropenic diet, along with important studies of this diet. At our institution, we initiated the protective environment in the 1960s for acute leukemia patients. It consisted of isolation tents and laminar air flow along with nonabsorbable antibiotics and sterilized food, which our patients did not find very palatable. Over the years the protective environment has evolved from a strict sterile environment to clean technique with isolation gowns, shoe covers, caps and masks, and extra HEPA filtration for all of our newly diagnosed AML (acute myelocytic leukemia) patients over 50 years of age and ALL (acute lymphocytic leukemia) patients older than 60 years. These precautions are still being used to reduce the incidence of, and mortality from, fungal infections and gram-negative bacteremia in induction patients.
The neutropenic diet study was conducted at our institution over a period of 3 years, from 2004 to 2007, and we had 153 randomized and 57 nonrandomized patients. We decided to collect data on nonrandomized patients because there was an adequate number of patients who requested to stay on the neutropenic diet, which was a requirement for all of our newly diagnosed leukemia patients. The study was conducted because there were numerous complaints from patients about not being able to eat fresh fruits or salads during chemotherapy, and many patients were experiencing nausea from their chemotherapy and only wanted fruits and vegetables. There was a great incentive for patients to go on the study because many of them wanted to be in the fresh-fruit group, and they knew that if they did not enroll, they would remain on the neutropenic diet—so by participating they at least had a 50% chance of being able to eat fresh fruits or vegetables.
Before and during the study, we did many food cultures of the food that was sent to our patients. We found that strawberries, salad, and grapes grew Enterobacter and Pseudomonas, and that cantaloupe grew Citrobacter. Concentrations of these bacteria were listed as varying from 10−2 to 10−6. The question was whether the prophylactic antibiotics were able to prevent patients from getting bacteremia from ingesting these foods. All fresh fruit and vegetables were washed thoroughly with cold water, always kept refrigerated, and delivered immediately to the patient’s room. The incidence of gram-negative bacteremia was similar in both groups, and the causative organisms were those that were cultured from the fresh fruit. Five patients in the raw group had bacteremia with alpha hemolytic streptococci, which were never cultured from any of the foods. Therefore, our practice was changed and now everyone is allowed a regular diet. In our study, we also recommended safe food handling for all our patients, whether they were on the neutropenic diet or the regular diet, and the recommendations provided by Fox and Freifeld are much more comprehensive than our guidelines. All of our patients were advised not to eat fast food, soft-serve ice cream, deli meats, raw fish or fresh salad dressing from the refrigerated section, and unpasteurized foods. Patients were also advised not to eat at buffets and not to leave food out of the refrigerator for several hours, as it would then be more likely to harbor microorganisms.
There is no question that salad, tomatoes, and fresh fruit do have microorganisms growing on them, as was reported in a 1976 article from Mercy Hospital in Pittsburgh in which patients were admonished not to eat salads. The problem is that it is difficult to measure compliance with this diet, and it is difficult to prove that the bacteremia was not resident in the patient’s gut prior to starting chemotherapy or that it occurred because of eating fresh fruit during chemotherapy. Many patients have told me they were following the neutropenic diet, but they were eating salads and fast food and really did not understand what they were supposed to eat—or maybe they just did not want to follow that restrictive diet.
One investigation not mentioned in the article by Fox and Freifeld has recently been published. It is a retrospective study of 726 consecutive hematopoietic stem cell transplant recipients; 363 received a neutropenic diet and 363 followed a general diet with safe food handling. The results of the study, which was conducted over a 4-year period, showed 135 microbiological confirmed infections in the neutropenic diet group and 106 microbiological infections in the general diet group. These infections included bacteremia and Clostridium difficile infection. The majority of the transplant patients received autologous transplants (n = 543), and these patients typically have a shorter duration of neutropenia (approximately 10 to 14 days). While this was not a randomized study, the sample size was large and there was no significant difference in microbiologically confirmed infections between groups.
We have found that patients frequently eat the fresh fruit they receive as part of their in-hospital meals, now that they are permitted to eat it. Our patients are very grateful that they are allowed to control more of their diet. Encouraging safe food handling, as Fox and Freifeld recommend, is improving the quality of life for oncology patients and will help to prevent life-threatening infections.
Financial Disclosure: The author has no significant financial interest or other relationship with the manufacturers of any products or providers of any service mentioned in this article.