Acute sinusitis is a very common infection in childhood, but its management remains a controversial issue. Antibacterials may be effective in selected children, but direct evidence is limited. One randomized, placebo-controlled trial has shown that amoxicillin or amoxicillin/clavulanate are better than placebo for children with symptoms of nasal discharge and cough that are persistent (over 10 days) and not improving. However, another placebo-controlled trial of the same agents did not demonstrate any benefit from antibacterials in a patient population selected with a clinical diagnosis of sinusitis of moderate severity, based on a composite clinical symptom score. A systematic assessment of cure rates with various antibacterials shows no consistent differences between classes. Evidence on the use of ancillary measures and nasal corticosteroids is also limited. The only randomized, placebo-controlled trial of antihistamines and decongestants has shown no incremental benefit when given in addition to amoxicillin. Another placebo-controlled randomized trial showed some transient symptomatic improvement with the use of nasal corticosteroids. No randomized trials exist on the use of antral lavage in children with acute sinusitis. The current rates of antimicrobial resistance among commonly implicated pathogens should be considered in therapeutic decisions. However, there is no evidence from well-designed trials on specifically how to manage children at high risk of carrying resistant organisms. The inaccuracy of clinical signs and symptoms in documenting the diagnosis further complicates therapeutic decisions. Nevertheless, radiographic assessment does not meaningfully improve the accuracy of the diagnosis for uncomplicated cases, and it is not cost effective. In the absence of definitive evidence, treatment with amoxicillin 45 mg/kg/day in two divided doses may be used in selected patients with symptoms that are persistent and not improving. High doses (90 mg/kg in two divided doses) may also be considered, and amoxicillin/clavulanate may be a more appropriate choice when there is high risk of resistant pathogens, e.g. in a child attending a childcare center, or recent use of antibacterials. However, a considerable proportion of children, especially those with mild or improving symptoms, may not have to be treated at all.