Diagnosis of IV Catheter-Related Bacteremia Without Catheter Removal
October 15, 1999 4 minutes read
Diagnosis of IV Catheter-Related Bacteremia Without Catheter Removal
Abstract & commentary
Synopsis: In patients with IV catheter-related bacteremia, a delay of two hours or longer to culture positivity of peripheral blood relative to that of blood from the catheter hub may be diagnostic.
Source: Blot F, et al. Diagnosis of catheter-related bacteremia: A prospective comparison of the time to positivity of hub-blood versus peripheral-blood cultures. Lancet 1999;354: 1071-1077.
Currently used methods for identifying catheter-related infection (CRI) require catheter removal. Blot and colleagues evaluated the differential time to positivity of simultaneously obtained peripheral and hub-blood cultures. All patients admitted to an intensive care unit (ICU) from whom an indwelling catheter was removed for suspicion of CRI were included. Blood was immediately inoculated into aerobic culture media and placed into an automatic blood culture system. Episodes of bacteremic CRI were strictly defined and required a positive (> 103 cfu/mL) cathetertip culture in addition to recovery of the same organism from simultaneous peripheral and hub-blood cultures. Patients with "probable" bacteremic CRI had lower colony counts in association with receipt of an antibiotic active against the blood isolate.
Cultures from 28 patients yielded the same micro-organism from peripheral and hub-blood cultures. Seventeen had bacteremic CRI (15 definite, 2 probable). In the remaining 11, CRI was excluded because the absence of the blood organism from quantitative catheter cultures (9 patients) or identification of another source of infection (2 patients). Three additional patients had CRI without bacteremia.
In all, 17 patients with bacteremic CRI, hub-blood cultures turned positive before peripheral cultures (range: 1 hour, 15 minutes to 73 hours, 30 minutes). All hub-blood cultures were positive within 24 hours. Using a time cutoff for differential time to positivity of 120 minutes, the method had a sensitivity of 94% and a specificity of 91%.
Comment by Robert Muder, MD
CRI is the most common cause of nosocomial bacteremia. The most frequently used method for identifying a vascular catheter as the source of bacteremia requires catheter removal and quantitative or semiquantitative culture of the catheter. The majority of catheters removed for suspicion of infection turn out to be uninfected. Catheter-related bacteremia can be accurately identified without catheter removal using quantitative blood cultures obtained from the catheter and peripheral blood. This method is rarely used, as it is labor-intensive.
The method used in this article relies on two observations. First, the time required for detectable growth of a blood culture is related to the size of the initial inoculum. Second, colony counts of bacteria in the blood within an infected catheter are typically much higher than that of peripheral blood in cases of catheter-related bacteremia. New automated blood culture systems detect bacterial growth at frequent intervals. The system used in this study, Vital bioMerieux (which I assume is the French equivalent of the VITEK system used in the United States), monitors for growth every 15 minutes.
A cutoff of two hours differential time to blood culture positivity appears to have reasonable sensitivity and specificity, and appears to be a reasonable method of identifying bacteremic CRI without the necessity of catheter removal. However, a few cautions are in order. The number of patients in this study was relatively small, so the sensitivity and specificity calculations have fairly wide confidence intervals. This study was performed in an oncology referral center, thus the catheters involved tended to be present for relatively long periods. Whether these results are applicable to more acutely inserted catheters is uncertain. A similar (albeit retrospective) study in a general hospital reported sensitivity of 73% and specificity of 69% using a differential time of two hours.1 Only two of the patients with bacteremic CRI in Blot et al’s study appear to have been receiving antibiotics active against the infecting organism; the use of the diagnostic method in such patients is uncertain. Although I suspect that similar results would be obtained using other automated blood culture systems, one can’t be absolutely certain of this. I would like to see confirmatory studies involving larger numbers of patients and using other automated culture systems in common use. Finally, none of the patients had fungal CRI, so the applicability of the method to management of fungemia is unclear.
In summary, I think that the differential time to positivity of simultaneously obtained hub-blood and peripheral cultures is a promising method of identifying catheter-related bacteremia. I do not believe that it should replace the current "gold standard" of catheter removal and culture until a larger study in a more general patient population is conducted. However, given that it is much easier to obtain simultaneous blood cultures than to remove and replace a catheter (or in some patients, multiple catheters), I suspect that some ICU physicians will adopt this diagnostic strategy. I would caution that the true negative predictive value of this method is not certain. One should still remove and culture a catheter from a patient with possible CRI, regardless of the differential time to blood culture positivity, if no other source of bacteremia is apparent and the patient has persistent fever, persistent bacteremia, or evidence of inflammation at the catheter site.
Reference
1. Mermel LA, et al. Diagnosis of catheter-related bloodstream infection (CRBSI) by differential growth rates of catheter-drawn and percutaneously drawn blood cultures. American Society for Microbiolgy 38th Annual ICAAC. San Diego, CA; Sept. 24-27, 1998: Abstract K-5.
Which of the following is correct with regard to patients with IV catheter-related bacteremia?
a. Blood cultures obtained from peripheral blood become positive prior to that obtained from the catheter hub.
b. Large bacterial inocula are associated with a shorter time to culture positivity when compared to small inocula.
c. Higher bacterial densities in peripheral blood than in IV catheter blood are characteristic of catheter-related bacteremia.
d. If a comparison of time to culture positivity of hub blood to peripheral blood indicates that the IV catheter is not the source of bacteremia, then the catheter need not be removed even if antibiotic therapy appears to be failing.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content