For decades, the concept of a neutropenic diet has implied a strict limitation of foods allowed for consumption, as a presumptive means of reducing the risk of infection in cancer patients. The rationale was to limit the introduction of potentially harmful bacteria into the gastrointestinal tract by the restriction of certain foods that might harbor those organisms. However, this concept has not been substantiated with direct proof, and no universal definition of the neutropenic diet exists. Exactly which foods are restricted varies greatly by institution, but most notable is the restriction of fresh fruits and vegetables. Research evaluating potential benefits of a neutropenic diet is very limited, but the diet is still prescribed in many institutions with the hope that it will prevent foodborne infection and/or bacteremia in neutropenic patients. Review of the pathophysiology of foodborne illness and pertinent studies about the neutropenic diet lead to the conclusion that there is no clear benefit from the longstanding dietary restrictions that may be imposed during neutropenia. Instead, we propose adoption of standard safe food handling methods to allow for a more liberalized diet in the neutropenic patient.
Published Studies of the Neutropenic Diet
Studies evaluating the neutropenic diet have examined three main aspects of the diet: how and whether cancer treatment centers use neutropenic diets, factors influencing patient compliance, and the incidence of infection. Several of the more recent pertinent studies are reviewed here. In 2000, Smith and Besser mailed surveys assessing use of the neutropenic diet to 400 institutional members of the Association of Community Cancer Centers. The survey queried institutions about use of a specialized low-microbe diet among nontransplant hospitalized cancer patients, the degree of neutropenia that triggered the initiation of such a diet, and a list of foods that were typically restricted. Of 400 centers queried, 156 surveys were returned. A total of 120 respondents indicated that patients with neutropenia were, in fact, placed on restricted diets, while the remaining 36 centers did not institute such diets. The investigators reported that 34 hospitals initiated neutropenic diets when the ANC was less than 1000/μL, and another 36 started when the ANC was less than 500/μL. Ten hospitals initiated the diet at the start of cancer treatment. Most hospitals that started the diet for inpatients also instructed patients to restrict their diets at home, while a minority allowed patients to resume a regular diet at home. The most commonly restricted foods were fresh fruits, fresh fruit juice, and fresh vegetables. About two-thirds of the hospitals restricted raw eggs and rare- to medium-cooked meats, and nearly half restricted dried fruits, potato salad, and beer. These survey results show clearly the high inconsistency of dietary practices for neutropenic patients.
In 2001, French et al mailed a survey regarding use of the neutropenic diet to ten pediatric hospitals performing bone marrow transplantation. Seven of the ten hospitals replied to the survey. The survey included questions about whether or not a neutropenic diet was used and when restrictions were started and stopped. Five of the seven hospitals responding to the survey reported using neutropenic diet restrictions. Dietary restrictions ranged from preparing foods in separate kitchens using aseptic techniques, to focusing on safe food handling principles and avoiding foods associated with foodborne illness, to a modified in-house diet that only excluded fresh fruits and vegetables. The timing of initiation and discontinuance of the diet varied. Some hospitals initiated the diet upon hospital admission, whereas others did not begin restriction until transplant day or when the ANC was less than 500/μL. Criteria for discontinuing the neutropenic diet included specific neutrophil counts, a specific period of time following transplantation, or upon hospital discharge. Two of the five hospitals responding to the survey reported use of a step-down or transitional diet to gradually reintroduce restricted foods prior to resumption of a regular diet. During this time, patients were allowed to eat raw fresh fruits that had been washed and peeled. In 2006, DeMille et al conducted a pilot study to determine the effect of the neutropenic diet in the outpatient setting, excluding stem cell transplant and acute leukemia patients. Twenty-three patients between the ages of 33 and 67 years completed the 12-week study. Patients were educated about the neutropenic diet. Telephone interviews were conducted at weeks 6 and 12 to assess adherence to the diet. The study aimed to assess three outcomes: rates of compliance, febrile neutropenia, and bloodstream infections. Compliance with the diet was noted in 70% (16/23) of patients. Interestingly, four of the seven noncompliant patients perceived themselves as noncompliant. The investigators attributed the patients’ noncompliance to difficulty following the requirements of the diet because of side effects of the chemotherapy. The second aim was to determine whether the rate of febrile admissions would differ between compliant and noncompliant patients. There were five admissions for febrile neutropenia. Four of the 16 compliant patients (25%) were admitted, and 1 of the 7 noncompliant patients (14%) was admitted. The third aim of the study was to determine whether a difference existed in the rate of blood cultures positive for gram-negative rods between compliant and noncompliant patients. Three of the four compliant patients who were admitted for febrile neutropenia (75%) had blood cultures positive for gram-negative rods. The one noncompliant patient admitted for febrile neutropenia also had blood cultures positive for gram-negative rods.
Also in 2006, Moody et al randomized 19 pediatric patients and their parents either to receive education about a neutropenic diet or to follow a plan of modified food safety guidelines approved by the US Food and Drug Administration (FDA). Patients included those being treated with myelosuppressive chemotherapy for malignant brain tumor, acute lymphoblastic leukemia (ALL), or sarcoma. The mean number of days of neutropenia (ANC < 500/μL) was 5.9 days for the neutropenic diet arm (n = 10) and 9.2 days for the food safety arm (n = 9), which was a nonsignificant difference (P = .29). Four patients in each arm (approximately 40%) developed a fever during neutropenia. In the neutropenic diet arm, there was one episode of P. aeruginosa bloodstream infection, but none in the food safety diet arm. The rate of adherence to the neutropenic diet was 94.1%, and the rate of adherence to the food safety guidelines was 99.9%. Patients on the neutropenic diet reported more difficulty adhering to the list of food restrictions.
In 2007, Van Tiel et al randomized 20 adult patients receiving induction chemotherapy for acute leukemia to either a neutropenic (low-bacterial) diet or a normal hospital diet; all patients received antimicrobial prophylaxis and had neutropenic periods of about 2 to 3 weeks. Fecal samples were collected daily for evaluation of colonization with Candida species or gram-negative bacilli, quantitated as colony-forming units per gram of feces. No significant differences in the two groups were identified with regard to colonizing yeasts and gram-negative bacilli, number of days of fever, requirement for antimicrobial therapy, or rates of documented infections. One patient in each group had candidemia. Furthermore, calculated total medical costs were not different between the diet groups.
The largest study to date, published by Gardner et al in 2008, randomized 153 patients undergoing induction for acute myelogenous leukemia (AML) to either a diet containing raw fresh fruits and vegetables (raw diet, n = 75) or one that excluded those items (cooked diet, n = 78) during their hospitalization for chemotherapy and neutropenia, until neutrophil recovery and hospital discharge. All patients received both antibacterial and antifungal prophylaxis. The primary endpoint was the incidence of major infection (pneumonia, bacteremia, or fungemia) in each group. Among those in the cooked-diet group, major infections occurred in 29%, compared with 35% in the raw-diet group (P = .60; 95% confidence interval [CI], 11% to 21%). There appeared to be a trend toward fewer fevers of unknown origin in the cooked-diet group vs in the raw-diet group (51% vs 36%; P = .07). Probability of death did not differ between the groups (P = .36). The authors concluded that survival was as expected in newly diagnosed AML. While the overall incidence of bacteremia was higher in the raw-diet group (23% vs 9%, P = .03), there was no difference in isolation of organisms normally resident in the gastrointestinal tract (E coli, Enterococcus, Candida albicans, Enterobacter, Pseudomonas, and Klebsiella) from blood cultures in raw-diet and cooked-diet populations (15% vs 6%; P = .12). The incidence of pneumonia tended to be higher in the cooked-diet group (15% vs 5%, P = .06).
Taken together, these studies indicate that the neutropenic diet is defined very variably and is not clearly of significant benefit to patients. Significant questions about the neutropenic diet remain unanswered, including which foods should be included; when such a diet should be initiated; which food preparation techniques best eradicate bacteria and improve patient compliance; and, most importantly, which populations (if any) will benefit from these manipulations. So far, the studies reviewed give few answers and do not clearly point to any advantage for the neutropenic diet. In addition, the small sample sizes and numerous confounding variables (ie, prophylactic antimicrobials, growth factors, tumor types, degrees of mucositis and neutropenia, inpatient vs outpatient populations) limit a definitive interpretation.[10,12,20] It also has been suggested that awareness of the neutropenic diet by patients who have been assigned to a regular diet may result in study bias. Thus, the aggregate data so far do not provide proof that a restrictive “low microbial” or neutropenic diet will protect neutropenic patients from invasive bacterial infections.
Moving From Unanswered Questions to What We Do Know
Although it is clear that fresh fruits and vegetables may carry enteric gram-negative rods on their surfaces, it is not clear what—if any—role they play in the pathophysiology of bloodstream infections in neutropenic patients. At this time, no cause-and-effect relationship has been or can be established to prove that the consumption of commonly restricted foods, particularly fresh fruits and vegetables, increases the likelihood of cancer patients developing bloodstream infections as a result of bacterial translocation. One could agree that the inclusion of raw fresh fruits and vegetables may actually be of benefit, as they are an excellent source of dietary fiber. Fiber may be considered a useful dietary component for cancer patients. Insoluble fibers are associated with improved intestinal transit and reduced incidence of constipation, and soluble fibers may help in managing diarrhea. Another benefit of dietary fiber is enhanced immune function. In fact, in a review of studies of bacterial translocation induced by total parenteral nutrition, Deitch concluded that the addition of dietary fiber to low-residue oral diets improves the balance of intestinal flora and lessens bacterial translocation.
One important exception to the liberalized consumption of raw vegetables is raw sprouts, including alfalfa, bean, clover, and others. This is the one food group that the FDA recommends avoiding, for both healthy and immunocompromised individuals. Unlike other fresh produce, seeds and beans need warm and humid conditions to sprout and grow. The warm, moist conditions in which sprouts are produced are also an ideal nourishing environment for the growth of bacteria, including Salmonella, Listeria, and E coli. A single bacterial pathogen on the outside of a seed can easily grow to an infectious dose after it has sprouted. The bacteria in and on growing sprouts cannot be washed off. Even though sprout growers attempt to use practices to decontaminate seeds before sprouting, no method has proven completely effective.
Safe Food Handling Guidelines Should Replace the “Neutropenic Diet”
Safe food handling can be defined as “protecting food from contamination.” Currently, safe food handling is advocated by the US Department of Agriculture (USDA), the FDA, and the US Centers for Disease Control and Prevention (CDC) as the main method of preventing food-related illness in immunocompromised patients. Two USDA publications, the “Dietary Guidelines for Americans 2010” and “Food Safety for People with Cancer,”[25,26] do not make reference to a neutropenic diet at all. Instead, the emphasis is strongly on safe food handling and preparation. The “Dietary Guidelines for Americans 2010” advocate the need for following four core food safety principles: cleaning, separating, cooking, and chilling. Furthermore, in its “Nutrition Care Manual,” the American Dietetic Association recommends that patients with neutropenia follow the USDA’s food safety guidelines. Hospital food and nutrition services are required by their state’s health department to follow FDA food safety guidelines, ensuring the safety of foods provided to the most immunosuppressed individuals.
Foodborne illnesses are defined by the World Health Organization (WHO) as diseases that are either infectious or toxic in nature, caused by agents that enter the body through the ingestion of food. Signs and symptoms of foodborne illnesses can include fever, muscle aches, nausea, vomiting, abdominal cramping, and diarrhea. The onset of symptoms can vary greatly, occurring in just a few hours to several days after ingestion. Since these symptoms mimic some of the side effects seen in cancer treatments, identification of foodborne illness may be delayed. In the neutropenic patient, any delay in identifying and treating infection could be life-threatening. The goal, therefore, is to prevent the occurrence of foodborne illness by employing methods of food selection and preparation that will reduce the chances of transmission, as described in Table 2.
At The Nebraska Medical Center (TNMC), our review of the existing data has led us to abandon the old and restrictive neutropenic diet in favor of safe food handling techniques for immunosuppressed patients. There is a striking lack of evidence to support the view that a neutropenic diet will prevent infection in neutropenic patients. On the other hand, the anticipated benefits of a more palatable diet and the improved oral intake and better quality of life afforded by a liberalized diet are welcomed by caregivers and patients. At TNMC, our recently revised policy is to educate patients, caregivers, and staff about implementation of safe food handling as recommended by the USDA. In the inpatient setting, the importance of food safety is stressed not only in the hospital food services kitchens, but also on nursing units. Undercooked eggs, undercooked meats, unpasteurized dairy products, unpasteurized juices, and raw sprouts are not provided to any patient. Meal trays are not allowed to sit at the patient bedside for more than 1 hour. In the outpatient setting, patients and caregivers are educated about safe food handling, preparation, and storage at home as well as food safety guidelines for foods eaten outside the home. Table 1 lists the safe food handling guidelines adopted at our institution for outpatients.
In the setting of a diagnosis of cancer, with patients facing many fears, stressors, and an uncertain future, liberalization of neutropenic diet restrictions may make patients feel more empowered to gain control over a very essential aspect of their lives. Allowing patients to follow a more generous dietary pattern (with an emphasis on safe food handling) is likely to increase interest in food and therefore improve intake of calories and protein, resulting in less unintentional weight loss.
In summary, the available evidence does not implicate food as a direct cause of bacteremia in neutropenic patients. We suggest that until a cause-and-effect relationship can be proven, the neutropenic diet should not be imposed on cancer patients. Safe food handling guidelines are geared toward preventing acquisition of foodborne pathogens in the broader population. We recommend that these guidelines supplant the neutropenic diet restrictions previously imposed on cancer patients.
Financial Disclosure: The authors have no significant financial interest or other relationship with the manufacturers of any products or providers of any service mentioned in this article.
1. Freifeld A, Bow E, Sepkowitz K, et al. Clinical practice guidelines for the use of antimicrobial agents in neutropenic patients with cancer: 2010 update by the Infectious Diseases Society of America. Clin Infect Dis. 2011;52:56-93.
2. Deitch E, Winterton J, Berg R. Effect of starvation, malnutrition and trauma on the gastrointestinal tract flora and bacterial translocation. Arch Surg. 1987;122:1019-24.
3. Blijlevens N, Donnelly J, De Pauw B. Mucosal barrier injury: biology, pathology, clinical counterparts and consequences of intensive treatment for haematological malignancy: an overview. Bone Marrow Transplant. 2000;25:1269-78.
4. Klastersky J. Chemoprophylaxis of gram-negative infections in neutropenic patients. Eur Urol. 1990;17(Suppl 1):40-5.
5. Casewell M, Phillips I. Food as a source of Klebsiella species for colonization and infection of intensive care patients. J Clin Pathol. 1978;31:845-9.
6. Pizzo P, Purvis D, Waters C. Microbiologic evaluation of food items. J Am Diet Assoc. 1982;81:272-9.
7. Restau J, Clark A. The neutropenic diet: does the evidence support this intervention? Clin Nurse Spec. 2008;22:208-11.
8. DeMille D. Have you reviewed your neutropenic diet lately? Oncol Nutrit Connect. 2011;19:7-11.
9. Osterblad M, Pensala O, Peterzéns M, et al. Antimicrobial susceptibility of Enterobacteriaceae isolated from vegetables. J Antimicrob Chemother. 1999;43:503-9.
10. Gardner A, Mattiuzzi G, Faderl S, et al. Randomized comparison of cooked and noncooked diets in patients undergoing remission induction therapy for acute myeloid leukemia. J Clin Oncol. 2008;26:5684-8.
11. Moody K, Finlay J, Mancuso C, Charlson M. Feasibility and safety of a pilot randomized trial of infection rate: neutropenic diet versus standard food safety guidelines. J Pediatr Hematol Oncol. 2006; 28:126-32.
12. Van Tiel F, Harbers M, Terporten P, et al. Normal hospital and low-bacterial diet in patients with cytopenia after intensive chemotherapy for hematological malignancy: a study of safety. Ann Oncol. 2007; 18:1080-4.
13. US Food and Drug Administration, Center for Food Safety and Applied Nutrition. Bad Bug Book, ed 2. March 3, 2011. Available at: http://www.fda.gov/Food/FoodSafety/FoodborneIllness/FoodbornePathogensNaturalToxins/BadBugBook/default.htm. Accessed May 13, 2012.
14. Bodey G, Hart J, Freireich E, Frei E. Stu1dies of a patient isolator unit and prophylactic antibiotics in cancer chemotherapy. Cancer. 1968;22:1018-26.
15. Armstrong D. Protected environments are discomforting and expensive and do not offer meaningful protection. Am J Med. 1984;76:685-9.
16. Rizzo J, Vogelsang G, Krumm S, et al. Outpatient-based bone marrow transplantation for hematological malignancies: cost saving or cost shifting? J Clin Oncol. 1999;17:2811-18.
17. DeMille D, Deming P, Lupinacci P, Jacobs L. The effect of the neutropenic diet in the outpatient setting: a pilot study. Oncol Nurs Forum. 2006;33:337-43.
18. French M, Levy-Milne R, Zibrik D. A survey of the use of low microbial diets in pediatric bone marrow transplant programs. J Am Diet Assoc. 2001;101:
19. Smith L, Besser S. Dietary restrictions for patients with neutropenia: a survey of institutional practices. Oncol Nurs Forum. 2000;27:515-20.
20. Gafter-Gvili A, Fraser A, Paul M, Leibovici L. Meta-analysis: antibiotic prophylaxis reduces mortality in neutropenic patients. Ann Intern Med. 2005;142: 979-95.
21. Jubelirer S. The benefit of the neutropenic diet: fact or fiction? The Oncologist. 2011;16:704-7. Available at: http://theoncologist.alphamedpress.org/content/16/5/704. Accessed May 13, 2012.
22. Deitch E. Bacterial translocation: the influence of dietary variables. Gut. 1994;35(Suppl 1):S23-7.
23. US Food and Drug Administration. Hold the raw sprouts please. October 14, 2011. Available at: http://www.foodsafetynews.com/2011/10/hold-the-sprouts-please. Accessed May 13, 2012.
24. Arizona Department of Health Services. Food safety and environmental services: food processing and handling basics. August 24, 2011. Available at: http://www.azdhs.gov/phs/oeh/fses/fpfhb.htm. Accessed May 13, 2012.
25. United States Department of Agriculture Center for Nutrition Policy and Promotion. Dietary Guidelines for Americans 2010. Available at: http://www.cnpp.usda.gov/Publications/DietaryGuidelines/2010/PolicyDoc/PolicyDoc.pdf. Accessed May 13, 2012.
26. United States Department of Agriculture Food Safety and Inspection Service. Food safety for people with cancer. September 2006. Available at: http://www.fsis.usda.gov/PDF/Food_Safety_for_People_with_Cancer.pdf. Accessed May 13, 2012.
27. The American Dietetic Association. Food safety guidelines for neutropenia. The American Dietetic Association nutrition care manual. 2010. [Updated October 10, 2011.] Available at: http://nutritioncaremanual.org/docs/PE05_oncology_neutropenia.pdf. Accessed May 13, 2012.
28. United States Food and Drug Administration. State Retail and Food Service Codes by State. January 19, 2011. Available at: http://www.fda.gov/food/foodsafety/retailfoodprotection/federalstatecooperativeprograms/ucm122814.htm. Accessed May 13, 2012.
29. World Health Organization. Food safety and foodborne illness. March 2007. Available at: http://www.who.int/mediacentre/factsheets/fs237/en/#. Accessed May 13, 2012.