Incidence, patterns of clinical presentation, and haematological characteristics of paediatric acute myeloid leukaemia in Uganda: a retrospective analysis

Abstract

Background: Current understanding of paediatric acute myeloid leukaemia (AML) in Africa is limited. This study investigated the incidence, presentation pattern, and haematological profiles of paediatric AML in Uganda. Methods: This retrospective cohort study examined the medical records of children under 18 years of age who were diagnosed with acute myeloid leukaemia (AML) at three cancer centres in Uganda from 2016 to 2022. Data included demographics, clinical features, and laboratory findings. Frequencies, bivariate analyses, and regression models were performed, with statistical significance set at p < 0.05. Results: The study included 159 children diagnosed with AML, with a median age of 9.0 years (IQR 3.0–12.0). The incidence was 7.0/million children 0–17 years. The most common presenting symptoms were fever (84.3%), weight loss (44.0%), fatigue (40.9%), bleeding (35.8%), and bone pain (28.9%). Clinical findings at diagnosis included splenomegaly (40.9%), lymphadenopathy (39.6%), myeloid sarcoma (39.6%), and hepatomegaly (37.7%). The median (IQR) white blood cell (WBC) count at diagnosis was 32.0 × 10⁹/L (0.3–81.8). The predominant FAB subtype was M7 (n = 22, 24.4%), followed by M5 (n = 20, 22.2%). Acute promyelocytic leukaemia was seen in 13 (8.2%) patients, while 11 (6.9%) patients had myeloid leukaemia of Down syndrome. Cytogenetic analysis was limited to only 10 patients. There were significant differences in clinical characteristics by age with respect to splenomegaly, central nervous system involvement, and FAB subtype. Conclusions: The incidence of paediatric AML in this study mirrors global patterns, with a higher-than-expected prevalence of the M7 subtype noted. In settings with limited resources, diagnosis relies mainly on clinical and morphological assessment. Broader access to cytogenetic and molecular testing could improve subtype identification and risk profiling.