Purpose To investigate the clinical characteristics and outcomes of adults with occult bacteraemia and the clinical impact of appropriate antibiotics.

Methods A case-control study was conducted to retrospectively analyse the bacteraemic adults visiting the emergency department (ED) during the period between January 2005 and August 2006. The patients with occult bacteraemia were the case group. Two control groups (CGs) were selected for comparisons: CG I, those with bacteraemia and the same Pittsburgh bacteremia score who were admitted at the first ED visit temporally near a case patient; and CG II, those with bacteraemia admitted at their first ED visit, irrespective of the Pittsburgh bacteraemia score.

Results There were 119 adults composing of the case group, 119 matched adults as the CG I and 293 adults as the CG II. Demographic characteristics, clinical conditions and outcomes were retrieved from chart records. A lower 28-day death rate (5.0% vs 11.9%, p=0.03) and less critical illness (ie, Pittsburgh bacteremia score ≥4 points; 1.7% vs 22.2%; p<0.001) were noted among case patients compared with those in CG II. However, no difference in the 28-day death rate (5.0% vs 5.9%; p=0.77) between the case group and CG I was discovered. Among the case patients, thrombocytopenia (<100 000/mm³; OR, 8.87; p=0.03) and inappropriate antibiotic therapy at the second ED or outpatient-clinic visit (OR 7.59; p=0.045) were the independent factors of 28-day mortality in the multivariate analysis. Moreover, the survival curve revealed a significant difference in the survival rate between those with occult bacteraemia receiving inappropriate and appropriate antibiotic therapy after index bacteraemic-onset (p=0.02).

Conclusions For adults with occult bacteraemia, a lower severity of illness and death rate than those of bacteraemic patients hospitalised for ED visit could be demonstrated, demonstrating the importance of appropriate antibiotic therapy.