Critical assessment of the methods used for detection of bacterial translocation.

AIM: Bacterial translocation (BT) can be demonstrated by blood and lymph node cultures and also by polymerase chain reaction (PCR) detection of DNA of enteric bacteria. Aiming at investigating BT after gastrointestinal operations we assessed it on two endpoints after ischemia-reperfusion (IR) or sham operation (SO).
METHODS: 2 groups of 200-g Brown Norway male rats were treated as follows: SO animals ( n=12) had laparotomy alone and IR animals ( n=12) had successively 15 min clamping of the portal vein and the mesenteric artery. Half the animals in each group were killed on postoperative (p.o.) day 2 the other half on p.o. day 7. Under sterile conditions regional lymph nodes and vena cava and portal vein blood samples were recovered and cultured for aerobes and anaerobes. Escherichia coli beta-galactosidase DNA was assessed in blood samples by PCR. The findings in the two groups were compared by means of chi(2) tests.
RESULTS: Post-hepatic (peripheral blood) BT was detected by cultures of gram-negative bacteria in 16% and 0% of SO and IR animals, respectively, on p.o. day 2 and in 16% and 50% on p.o. day 7. These differences were not significant (ns). E. coli DNA was found in one SO rat. Pre-hepatic BT (portal blood and/or lymph nodes) of gram-negative bacteria was found in 16% and 33%, respectively, on day 2 and in 16% and 16% on day 7 (ns). However, if gram-positive cultures were taken into account, the figures were 66% and 66% on day 2 and 66% and 83% on day 7 (ns). No anaerobes could be cultured.
CONCLUSIONS: (1) BT is frequent in surgically manipulated animals. (2) To limit the assessment of BT to Enterobacteriaceae is probably misleading, since consistent amounts of gram-positive bacteria are found in the pre-hepatic territory. (3) PCR tests limited to E. coli DNA alone are likely incomplete. (4) Short periods of vascular clamping do not increase BT on the two endpoints selected in comparison with SO animals.