Inpatient utilization of blood cultures drawn in an urban ED

Am J Emerg Med. 2012 Jan;30(1):110-4. doi: 10.1016/j.ajem.2010.10.022. Epub 2010 Dec 3.


Bloodstream infections are now ranked as the 10th leading cause of death in the United States. Given the severity of bacteremia, physicians routinely order multiple sets of blood cultures in the emergency department. This is a retrospective chart review on 1124 patients admitted to the hospital for suspected bacteremia during calendar year 2004. The aims of the present investigation were to investigate the overall utility of blood cultures by the admitting services and to identify patient factors that might influence culture yield. Data were collected regarding patient demographics, comorbidities, vital signs, laboratory results, antibiotic use, blood culture results, and notation of blood culture results by admitting physicians. Increased age, elevated heart rate, use of chemotherapy, decreased sodium, and increased blood urea nitrogen significantly increased the likelihood of yielding a positive blood culture in our patient population. Culture results were noted in 517 patient charts by the primary medical team (46.0%) and were adjusted in 223 patients (43.3%). Of 1124 cultures, 10.3% were positive in at least 1 bottle for a pathogenic organism (true positive), and 6.3% were contaminants (false positive). In conclusion, cultures must be followed closely by the admitting physician after being obtained. Our data emphasize that blood cultures are currently not well used by the admitting physicians and that measures need to be taken to improve the overall utility of blood culture data by the admitting physician.

MeSH terms

  • Adolescent
  • Adult
  • Aged
  • Aged, 80 and over
  • Anti-Bacterial Agents / therapeutic use
  • Bacteremia / blood*
  • Bacteremia / diagnosis
  • Bacteremia / drug therapy
  • Blood Specimen Collection / statistics & numerical data*
  • Emergency Service, Hospital / statistics & numerical data*
  • Female
  • Hospitals, Urban / statistics & numerical data*
  • Humans
  • Logistic Models
  • Male
  • Middle Aged
  • Practice Patterns, Physicians' / statistics & numerical data
  • Retrospective Studies
  • Risk Factors
  • Young Adult


  • Anti-Bacterial Agents