Pseudomonas aeruginosa strains are generally resistant to many antibiotics, and nosocomial infections because of this species are one of the major problems in many hospitals. Molecular typing provides very useful information about origin and transmission of the strains. The aims of the present study were to investigate clinical and microbiologic characteristics of the nosocomial infections caused by P aeruginosa strains in a medical center and to bring up the cross-transmission level of this opportunistic pathogen in a university hospital by analyzing the clonal relationship among the isolates.

A total of 105 P aeruginosa strains had been identified among the 80 inpatients in a 1-year period from August 2003 to August 2004. Demographic, clinical, and epidemiologic data of the patients were prospectively recorded. The standardized disk-diffusion method was used to determine resistance of the strains to imipenem, ceftazidime, aztreonam, amikacin, gentamicin, mezlocillin, cefepime, tobramycin, meropenem, ceftriaxone, and ciprofloxacin. Clonal relatedness of the strains was investigated by pulsed-field gel electrophoresis (PFGE).

Of the 105 P aeruginosa strains identified, 45 (43%) were isolated from the patients hospitalized in intensive care units. Thirteen patients had repeated pseudomonas infection (total 38 infections/13 patients); 26 of these repeated infections in 9 patients showed the same localization. Half of the patients had at least 1 underlying disease such as burn (48%), chronic illness (32%), and malignancy (20%). Fifty-seven patients (71%) had urinary and/or other catheterization. Urinary tract infection (35%) was the most frequent infection encountered, followed by respiratory tract infection (34%) and sepsis (13%). Resistance to the antibiotics tested was in the 12% to 88% range; amikacin was the most effective and ceftriaxone was the least effective antibiotic. The PFGE typing method showed that 28 of the 80 patients’ isolates were clonally related, including 23 indistinguishable or closely related strains (29%), and 5 possibly related strains (6%). Epidemiologic data of the 16 patients (20% of the patients) confirmed a clonal relationship among the strains. Of the 26 isolates of the 9 patients having repeated infection in the same location, 18 (69%) were in the clonally related groups, whereas 11 of the 12 strains isolated from repeated infections on different body sites were clonally different.

Our results indicated that P aeruginosa infections in our hospital mainly affected the patients hospitalized in intensive care units and those having catheterization, burn, and/or chronic illness. Amikacin was the best antibiotic as far as bacterial resistance was considered. Although lack of major PFGE type confirmed no P aeruginosa outbreak, typing results showed that cross transmission and treatment failure are the 2 main problems, which should be consider together to prevent this bacterial infection in medical centers.