Catheters, Microbes, Time, and Gold Standards

  1. Barry M. Farr, MD, MSc
  1. From University of Virginia Health System; Charlottesville, VA 22908.

    Vascular catheters became integral to medical care in the 1940s, and soon thereafter catheter-related bloodstream infection (CRBSI) became an important cause of morbidity and mortality. Catheter removal was considered imperative for diagnosis and therapy of CRBSI until a quarter century ago when Wing and colleagues proposed drawing paired quantitative blood cultures to diagnose CRBSI—1 set through the catheter and another from a peripheral vein (1). They reported a patient with a 400-fold higher concentration of 2 gram-negative bacilli (Enterobacter cloacae and Citrobacter freundii) in catheter-drawn blood and reasoned that the difference must mean that the catheter was the source of the infection. Although they did not verify this assertion with a gold-standard method for diagnosing CRBSI, nobody disputed their logic. In a subsequent study of experimentally induced Escherichia coli peritonitis with secondary bloodstream infection in rabbits, blood cultures from the superior vena cava had 1.14- to 3.54-fold higher E. coli concentrations than did femoral vein blood (2). The same authors found that catheter-drawn blood from 5 patients with CRBSI had 57- to 7100-fold higher concentrations of bacteria than did blood drawn from a peripheral vein. They concluded that a concentration at least 5-fold greater in catheter-drawn blood was a reasonable cut-point for diagnosing CRBSI. Multiple studies have since evaluated the accuracy of paired quantitative blood cultures, finding sensitivity around 90% and specificity over 99%. However, the study that reported the highest sensitivity incorporated the results of the test being evaluated into the gold standard, a flaw that inflates measured accuracy (3).

    Blot and colleagues noted that microbial inoculum was inversely proportional to the time to positivity for continuously monitored qualitative blood cultures (4) and suggested that results of catheter-drawn cultures that turn positive 2 or more hours before cultures drawn from a peripheral vein afford the same …

    « Previous | Next Article »Table of Contents