Bacterial Translocation Detected by PCR

ABSTRACT & COMMENTARY

In this pilot study performed to detect bacterial genes in the blood of critically ill patients, Kane and associates used the polymerase chain reaction (PCR) technique to amplify genes from Escherichia coli, Bacteroides fragilis, and a region of 16S ribosomal RNA incorporated in gram-positive and -negative bacteria. DNA fragments were extracted from the blood of 40 surgical patients and, as a control group, from 20 healthy volunteers.

Although none of the bacterial DNA genes were detected in the blood of the healthy volunteers, bacterial DNA genes were identified in all patients with positive bacterial blood cultures. Microbial genes were found in the blood of all eight transplant patients who received OKT3 therapy. Sixty-four percent of the critically ill patients had microbial DNA detected in their blood, although only three (14%) had positive blood cultures. When a primer pair for 16S ribosomal RNA genes was used, these genes were detected in 27%, 75%, and 100% of noninfected, infected, and septic patients in the ICU, respectively. Kane et al propose that this method may detect bacterial translocation from the gut and, thus, may be of help in understanding the development of multiple organ dysfunction syndrome caused by bacterial translocation. (Kane TD, et al. Ann Surg 1998;227:1-9.)

COMMENT BY JUN TAKEZAWA, MD

Identification of DNA fragments from various microbial organisms by PCR has been used to make the diagnosis of a specific pathogen in clinical specimens, such as bronchoalveolar lavage fluid and cerebrospinal fluid. This paper is the first to apply this method to detect previous and/or present existence of bacteria in the blood.

Detection of bacterial DNA fragments in the blood with this method is sensitive enough to allow one to make the diagnosis of the presence of invasive bacteria; Kane et al have reported that 10-100 microorganisms per 0.3 mL blood were detected in their previous animal study (J Surg Res 1996;63:59-63). This is far more sensitive than the blood culture test.

The 16S ribosomal RNA genes were detected more frequently with the increasing probability of the presence of infection, indicating that microbial genes are prevailing depending on the severity of infection. Therefore, this study looks promising in making the diagnosis of the presence of bacterial genes in the blood. However, bacterial DNA genes could directly come from either live or dead organisms, or they could be derived from the engulfed bacteria killed by phagocytes. It is, therefore, unknown whether identification of those genes reflects ongoing infection or killed bacteria from previous infection.

Additionally, although none of bacterial DNA genes were detected in healthy volunteers in this study, it is highly possible that they could have been detected if further cycles of PCR amplification were carried out. Although this method provides extremely high sensitivity for detecting microorganisms present in the blood, specificity may be conceptionally low for making the diagnosis of sepsis caused by an invasive microorganism. Therefore, all-or-none type detection of bacterial genes in the blood may inherit the limitation of detection of bacterial infection and/or bacterial translocation from the gut. To overcome this limitation, the alternative PCR method to quantitatively measure bacterial genes in the blood is mandatory for understanding bacterial translocation (if it occurs) with respect to the development of the multiple organ dysfunction syndrome (MODS).