Factors associated with occurrence of acute leukemias and myelodyplastic neoplasms as seen at the Kenyatta National Hospital, Nairobi, Kenya.

Abstract

e18502 Background: Acute leukaemias-acute myeloid leukaemia (AML) and acute lymphoblastic leukaemia (ALL) usually carry poor prognosis. Myelodysplastic neoplasms (MDNs) run variable courses. Depending on the risk category some tend to terminally transform into AML. Known risk factors include exposure to radiation, tobacco, chemotherapy and certain chemicals; age, rare congenital diseases, certain blood disorders, family history. Though rare, the apparent upsurge in number of cases in 2012 locally necessitated this study. Objective was to profile exposure factors associated with AML/MDS and ALL. Methods: A case-control study involving AML/MDN and ALL patients aged 13 years and above, double-matched with hospital and family controls. Hospital controls were patients from the same hospital, matched for sex and age within 5 years and above, but with normal haemogram. Family controls were healthy first-/second- degree relatives residing in the same environment. Consecutive sampling was applied. Variables included demographic and clinical details; occupation, water sources, shopping outlets, food. Primary endpoints were associated exposure factors. A sample size of 389 cases was calculated and matched with 2 controls per case (totalling 1167). Multivariable conditional logistic regression was used to assess associations of patient and exposure factor variables reporting adjusted odds ratios (aOR). P value of 0.05 or less was regarded as statistically significant. Results: In total 362 patients were included; males 185 (51.1%) and females 177 (48.9%). AML cases were 182, ALL 157 and MDN 23. Cases by age-group were 12-21 years - 120(31%), 22-45 years - 132 (36%), and those 46 years and above were 110 (31%). counties registering 5 or more cases were from cool highlands of Mount Kenya area or Kisii, though not tested against population densities. On multivariable analysis, alcohol consumption (aOR) = 1.89 [95% CI 1.15 - 3.11, (P =0.012)], family history of cancer (aOR = 4.57 [95% CI 3.06 - 6.77, (p < 0.001)] and occupations other than construction/casual labour [all aOR>1, (p=003)] were strongly associated. Conclusions: Exposure factors strongly associated with occurrence of AML/MDS and ALL together included alcohol consumption, family history of cancer and occupation. Residence in cool highlands was also an observed association but was not subjected to statistical discrimination.