Abstract
Malnutrition is prevalent in gastrointestinal (GI) cancer patients, possibly due to inflammation and altered fatty acids (FA). There is a lack of research describing nutritional decline in these patients during chemotherapy. We described changes in nutritional, inflammatory, and FA status over time and factors relating to change in nutritional status according to tumor presence in 41 GI cancer patients undergoing first-line treatment over four chemotherapy visits, using linear mixed effects models. At baseline, 53% of patients were malnourished. Over time, there was a decrease in the proportion of malnourished vs. well-nourished individuals (β= −0.564, p < 0.01). Median concentrations of plasma linoleic acid, arachidonic acid, eicosapentaenoic acid, docosahexaenoic acid, total n-3, total n-6 and total plasma phospholipid FA increased over time. Changes over time in nutritional status based on weight (p < 0.001), fat free mass (FFM) measured by bioelectrical impedance analysis (BIA, p = 0.02), and skinfold anthropometry (FSA, p = 0.04) were significantly dependent on tumor presence. There were positive associations between weight and total n-3 (β = 0.02, p < 0.01), FFM and IL-6 (BIA, β = 0.028, p = 0.02; FSA, β = 0.03, p = 0.02), and FFM and total n-6 (BIA, β = 0.003, p = 0.01). Changes in nutritional status during chemotherapy were negatively impacted by tumor presence, and were associated with increasing concentrations of cytokines and FA.
Acknowledgments
This study was presented in part in poster format at the Canadian Nutrition Society conference, Ottawa, 2016, and was presented in poster format at the Academy of Nutrition and Dietetics Food and Nutrition Conference and Expo, Washington, 2018. The authors thank the study participants, the staff of the Medical Day Care Unit at St. Michael’s Hospital, Maureen Lee and Dr. Phil Connelly at St. Michael’s Hospital for assistance with the cytokine analysis, and Kayla Hildebrand and Dr. Bazinet at the University of Toronto for lipid analysis.
Disclosure statement
The authors report no conflict of interest.