This article is more than 5 years old. Information may no longer be current.
Dietary, social restrictions do not reduce infection risk in children with AML
Click Here to Manage Email Alerts
Nonpharmacological anti-infective measures did not significantly reduce the risk for infection in children with acute myeloid leukemia and should be reconsidered, according to a secondary analysis of a randomized trial.
“Infectious complications are an important cause of morbidity and mortality in patients undergoing therapy for AML,” Thomas Lehrnbecher, MD, professor of pediatrics at Johann Wolfgang Goethe University in Frankfurt, Germany, and colleagues wrote. “The especially high risk for infections in this patient population is a result of the intensive chemotherapeutic regimen, which profoundly disrupts the different arms of the immune system. To decrease the risk [for] infection, nonpharmaologic anti-infective measures (eg, restriction of social contacts and food), pharmacologic supportive care (eg, antibacterial and antifungal compounds) and immunomodulation (eg, granulocyte colony–stimulating factor) are widely used.”
Although data supporting pharmacologic anti-infective measures are well established, little study has focused on the effectiveness of nonpharmacological measures.
Lehrnbecher and colleagues analyzed infectious complications in children receiving intensive treatment for AML as part of the randomized AML-BFM 2004 trial. They began by surveying treatment centers to ascertain institutional standards on nonpharmacological recommendations regarding pets in the home (five items listed), social contact and school attendance (six items listed), and dietary practices (eight items listed).
Respondents ranked each restriction-related question as zero (never restricted), one (restricted under certain circumstances) or two (always restricted).
The researchers sent surveys to 43 German pediatric hematology centers, of whom 86% (n = 37) responded. They accessed complete datasets on infectious complications from 339 children (39% aged between 11 years and 18 years; 59% diagnosed with high-risk disease).
A median of 11 patients (range, 4-28) per hospital received treatment according to the AML-BFM 2004 protocol, which included induction chemotherapy consisting of cytarabine, etoposide and anthracyclines for all patients; second induction with high-dose cytarabine and mitoxantrone; and intensification with high-dose cytarabine and etoposide.
A total of 277 children (81.7%) experienced a fever of unknown origin, with 174 children (51.3%) having at least one episode of bacteremia. One or more episodes of pneumonia occurred in 45 children (13.3%) and gastroenteritis occurred in 77 children (22.7%).
Restriction advice varied widely. More than 90% of patients received restrictions for attending kindergarten or school, and more than 80% were advised not to consume raw seafood or meat.
By contrast, less than one-third of children were advised to not allow friends to visit their homes, or to not consume takeout food.
In total, the institutions surveyed had a median recommendation score for social restrictions of nine (range, 7-12; highest possible score, 12). Pet restrictions had a median score of eight (range, 2-10; highest possible score, 10) and food restrictions had a median score of 14 (range, 0-16; highest possible score, 16).
Unadjusted multivariate regression analyses showed that higher social restriction scores appeared associated with an increased incidence of bacteremia (incidence rate ratio [IRR] = 1.21; 95% CI, 1.07-1.37), although higher pet restriction scores decreased the incidence of pneumonia (IRR = 0.86; 95% CI, 0.76-0.97).
However, upon adjustment for potential confounding factors — such as age, sex, weight group, risk stratification and antibiotic prophylaxis — no restriction scores significantly influenced the risk for fever of unknown origin, pneumonia, bacteremia or gastroenteritis.
Underweight children had an increased risk for fever compared with normal-weight children (adjusted IRR = 0.79; 95% CI, 0.65-0.97) and overweight children (IRR = 0.55; 95% CI, 0.38-0.8).
High-risk children had significantly higher risks for bacteremia (adjusted IRR = 1.59; 95% CI, 1.2-2.1) and pneumonia (adjusted IRR = 1.93; 95% CI, 1.03-3.6); however, the risk for bacteremia decreased with the prescription of antibiotic prophylaxis (adjusted IRR = 0.59; 95% CI, 0.43-0.8).
Older children exhibited a lower risk for gastroenteritis than younger children (adjusted IRR = 0.94; 95% CI, 0.9-0.98).
The researchers acknowledged study limitations, including the potential for bias inherent in its observational design. They further noted that nonadherence to supportive care recommendations could have confounded their analysis.
“Changing this strict policy could improve the patients’ quality of life without increasing the risk [for] infectious complications,” Lehrnbecher and colleagues wrote. “Future studies in children with all kinds of malignancies are needed to evaluate the impact of nonpharmacologic measures on the incidence of infections, the immune system, the quality of life, and overall clinical outcome.” – by Cameron Kelsall
Disclosure: Lehrnbecher reports research funding and travel expenses from, as well as consultant and speakers bureau roles with, Astellas Pharma, Basilea Pharmaceutica, Gilead Sciences, Merck and Pfizer. The other researchers report no relevant financial disclosures.
Click Here to Manage Email Alerts