Wasting and short-term outcomes among children with cancer in resource-limited settings: A prospective study in Uganda

Abstract

Background Wasting contributes to poor treatment outcomes in children with cancer, especially in low-resource settings. In these settings, there is inadequate routine, systematic assessment of the wasting status of children with cancer. Wasting is diagnosed based on visual evidence, with a subjective bias for recognition. This study determined the prevalence of wasting at diagnosis among children with cancer at the Uganda Cancer Institute (UCI) and the diagnostic accuracy of “visible wasting” in identifying children with wasting as measured by anthropometric indices, and identified predictors of 6-months negative outcomes. Methods We assessed the wasting status at diagnosis, diagnostic accuracy of visible wasting, and 6-month outcomes of children newly diagnosed with cancer at the UCI (both ambulatory and hospitalized) between April 2022 and March 2023. Data were analyzed using SPSS version 26. Descriptive, bivariate, multivariate, and survival analyses were performed as appropriate. Statistical significance was determined at P-value<0.05. Results One hundred forty-four children with cancer, with a median age of 10.0 years (inter-quartile range [IQR] 4.0–14.0 years), were included. The majority, 89 (61.8%), had solid tumor, whereas 55 (38.2%) had hemato-lymphoid malignancies. Thirty-two PLOS One | https://doi.org/10.1371/journal.pone.0330107 August 7, 2025 2 / 17 (22.2%) of the participants had visible wasting, and 57 (39.6%) were wasted based on anthropometric measurements, 32 (56.1%) of whom showed no visible wasting. Visible wasting had a low sensitivity of 43.9% (95% CI 30.7–57.6) – ROC 0.32 (95% CI 0.23–0.42), with a false negative rate of 56.1%. Overall, visible wasting missed up to 80.6% (25/31) of children with moderate wasting and 26.9% (7/26) with severe wasting. Twenty-one (14.6%) of the patients died, 8 (38.1%) of whom were deemed to be wasted, and 15 (71.4%) had anthropometrically-defined wasting. Neutropenia occurred in 20.8% (n = 30) of the participants and sepsis in 13.9% (n = 20). In univariate analyses, wasted patients were more likely to develop neutropenia (OR 3.63; 95% CI 1.56–8.42; p = 0.003), sepsis (OR 4.50; 95% CI 1.65–12.29; p = 0.003), and die (OR 3.08; 95% CI 1.15–8.28; p = 0.026). Conclusion Wasting at diagnosis is a common problem among children with cancer in this resource-limited setting and is associated with increased risks of neutropenia, sepsis, and mortality. Reliance on visible wasting as a marker for wasting misses other wasted children, some of who may be malnourished and at risk of poor outcome. For accurate categorization of wasting, all patients should undergo a standard anthropometric evaluation.