Previous Abstract | Next Abstract
Printable Version
A-427
2003
Risk of Emergence of Pseudomonas aeruginosa Resistance to Antimicrobial Agents in Intensive Care Units
Bernard Georges, M.D., Maryse Archambaud, M.D., Ph.D., Sylvie Saivin, Pharm.D., Ph.D., Jean Marie Conil, M.D., Eric Bonnet, M.D.
Service d'Anesthesie Reanimation, CHU Rangueil Larrey, Toulouse, France.
Background : Emergence of P. aeruginosa resistance to antimicrobial agents is usual in Intensive Care Units (ICU) and could be correlated to the use of some specific agents. We attempted to find correlation between the use of various betalactams and the emergence of specific mechanisms of resistance.

Methods : We performed an open prospective study for a three-year period including all patients for which P. aeruginosa was isolated from one or more specimens : bronchial aspiration, blood cultures, catheters, urinary cultures.

Studied antibiotics were : amoxiclav, ticarcillin, piperacillin-tazobactam, cefotaxime, ceftazidime, cefepim and imipenem. Mechanisms of resistance that we studied were: production of penicillinase or cephalosporinase, non enzymatic mechanism and loss of porine OprD2. Khi-2 test was used with a level of significance at 0.05.

Results : 132 patients were included in the study. 82 strains emerged with various mechanisms of resistance during antibiotic treatment. When using multivariate analysis, no correlation was found between the use of amoxiclav, piperacillin-tazobactam, cefepim and any of the mechanisms. The use of cefotaxime was associated to the appearance of a non enzymatic mechanism (p = 0,005). No correlation was found with the use of ceftazidime, but if resistance appeared during ceftazidime treatment, it was due to secretion of cephalosporinase (p = 0,001). There was a significant relationship between the use of imipenem and emergence of resistance by loss of porine OprD2 (p = 0,00001). 36 strains from 9 patients, assayed with pulsed-field gel electrophoresis showed that for a same patient, all the strains were genetically linked together.Conclusion : Ours results show a high risk of selection of non enzymatic resistance by cefotaxime and emergence of resistance during treatment with imipenem.

Anesthesiology 2003; 99: A427
Logistic regression analysis
VARIABLES Odds ratio IC 95 % p Signifance
Age 0.228NS
Gravity Score 0.749NS
Ventilation duration1.021.002-1.0390.045Significant
Anterior hospitalisation 0.08NS
Delay of colonization 0.654NS
Infection2.321.09-4.960.03Significant
ANTIBIOTICS
Amoxiclav 0.477NS
Piperacillin-Tazobactam 0.469NS
Cefotaxime4.281.46-12.590.009Significant
Ceftazidime 0.941NS
Cefepim 0.454NS
Imipenem3.241.32-7.940.011Significant