Primary care doctors should be cautious in the diagnosis and treatment of sinusitis as acute bacterial sinusitis is currently over-diagnosed and over-treated in primary care practice. The clinical diagnosis of acute bacterial sinusitis is difficult in primary care practice; however, a history of purulent rhinorrhoea, purulent secretions in the nasal cavity on examination, tooth pain, worsening of symptoms following initial improvement, lack of effect of decongestants and an elevated erythrocyte sedimentation rate are supportive evidence of bacterial infection. Patients with symptoms for <7 days are not as likely to have bacterial infection. Acute sinusitis is over-treated in primary care practice for several reasons. Firstly, most cases of acute sinusitis are caused by viral infections and resolve without antibacterial treatment. Secondly, in clinical trials of antibacterial treatment, only about one-half of patients diagnosed with acute bacterial sinusitis by experienced primary care physicians have bacterial infection. Thirdly, antibacterial treatment of acute sinusitis is indicated only in patients with severe symptoms of sinusitis or in patients with moderate symptoms of >7 days duration. Symptomatic treatment is sufficient in patients with mild symptoms. Three recent meta-analyses have concluded that newer and broad-spectrum antibacterials are not significantly more effective than narrow-spectrum agents, such as amoxicillin or phenoxymethylpenicillin (penicillin V). However, because of the rapid increase in antibacterial resistance of Streptococcus pneumoniae and Haemophilus influenzae, treatment must take into account current recommendations for treating infections caused by these organisms. Fourthly, sinus imaging studies are not recommended in routine diagnosis but may be helpful in selected cases. Finally, other than pain medication, there is little evidence that use of adjunctive treatments, such as decongestants, is effective in symptom relief. However, a recent study in patients with recurrent sinusitis demonstrated that patients who received fluticasone propionate in addition to antibacterials had a higher rate of clinical success than did patients receiving placebo and antibacterials.