Theimpact ofbloodculture reporting andclinical liaison on theempiric treatmentofbacteraemia

Aims-To assess theimpactof blood culture results andearly clinical liaison on thetreatmentofpatients withbacteraemia. Methods-123patients withsignificant positive bloodcultures were followed over a ninemonthperiodina 620bedteaching hospital. Theimpactofearly bloodculture reporting andclinical liaison on thecost and appropriateness of treatmentwas assessed. Results-Empiric treatmentwas started beforetheGram stainresult in107(87%) patients. Treatmentwas altered on the basisoftheGram stain result in39(36%) ofthesepatients, and on cultureand sensitivity results in53(50%). The spectrum ofantibiotic treatmentwas narrowedin58(54%)ofthese; 20(19%)on Gram stain result alone. Thisresulted ina 42%reduction indailyantibiotic costsin patients who hadreceived empiric treatment. Empirictreatmentdidnot follow thehospital antibiotic policy in49(46%)of thepatients treated. In patients where empiric treatmentwas notinaccordance withhospital policy, 21 (44%)had an isolate resistant totheempiric treatment used;whilein patients who received agentsinaccordance withhospital policy onlyone (1.7%)had a resistant isolate (p< 0.05). Patients who died(11(9%)) were lesslikely tohavereceived empiric treatmentinaccordancewiththeantibiotic policy, although thisdidnotreach statistical significance (p= 0.1). Conclusion-Early reporting of Gram stainresults from bloodcultures, combinedwithearly clinical liaison, results in more rational and costeffective treatment. ( Clin Pathol 1997;50:1010-1012)