Incidence and clinical significance of reactive thrombocytosis in children aged 1 to 24 months, hospitalized for community-acquired infections.

The aims of this study were to identify demographic, clinical and laboratory characteristics associated with reactive thrombocytosis useful for clinical management and to evaluate potential complications of this condition in a cohort of children selected for they young age as at high risk of reactive thrombocytosis. Retrospective analysis of medical records of 239 children among 902 aged 1-24 months, hospitalized during a 12-month period, and discharged with a diagnosis of infectious disease was performed. One hundred and nineteen children out of 239 (49.8%) presented thrombocytosis (>500 platelets x 10(9)/L; normal range 150-499 x 10(9)/L), 81/119 (68%) on admission. The incidences of thrombocytosis or extreme thrombocytosis (>1,000 x 10(9)/L) were 13.2% (119/902) and 0.8% (7/902). Thrombocytotic children had higher counts of white blood cells and had been treated more frequently with steroids (36/82, 43.9% vs. 5/53, 9.4%; p = 5 x 10(-5); relative risk 7.51, 95% confidence intervals 2.71-20.82). No significant difference was found in relation to sex, age, fever, C reactive protein level, diagnoses and antibiotic therapy. Two out of 239 (0.8%) enrolled children, both thrombocytotic and with other acquired risk factors, developed thrombosis. In conclusion, reactive thrombocytosis in children aged 1 up to 24 months is frequent and unrelated to markers of disease activity or degree of inflammation.