Epidemic Multiresistant Pseudomonas aeruginosa Causing Persistent Infections and High Mortality in a Hospital with AIDS-Predominant Admissions

Background: Relapse or recurrent infections caused by P. aeruginosa were described before in AIDS patients, but there were not multiresistant strains and had not significant impact on mortality. Persistence of strains was also described in AIDS patients with sequentially obtained non-multiresistant P. aeruginosa in respiratory infections. Objective: We present the clinical evolution of an epidemic multiresistant Pseudomonas aeruginosa (MR-PA) strain in a hospital with AIDS predominant admission. Methods: During April 2002 and February 2007 a prospective program for controlling nosocomial infection, which utilized an adaptation of the methodology developed for CDC-USA was performed in a 26-bed (2-4 intensive care beds) totally infectious disease research hospital in Rio de Janeiro, Brazil. This program included monitoring daily all clinical and surveillance cultures obtained from all admitted patients. Results: The incidence density of MR-PA acquisition was 2.80/1000 and 2.09/1000 patient-days during outbreaks against a base line of 0.53/1000. Thirty-three of 94 P. aeruginosa isolates (35%) had multiresistant profiles and were isolated from 13 MR-PA infected and three colonized patients. Pulsed field gel electrophoresis showed that all tested MR-PA strains which pertained to the index case, 5 of 11 patients from the outbreaks, and one patient from the inter-epidemic period had identical molecular profile. Most of the patients (93.75%) were immunocompromised with AIDS (10 patients) or HTLV-1 infection (5 patients).  Nineteen MR-PA infections (9 septicaemia associate to invasive dispositive in 6, 5 catheter-associated urinary tract infections, 3 ventilator associated pneumonia and 2 central line associated infections) were diagnosed in 13 patients. Excluding one patient who lost the follow up, 66.7% (8 patients) had one treatment failure; 66.7% (8 patients) persistent infections (average of 54.4 days between positive clinical sample, median 27, range 20-195 days) and 25% (3) recurrent infections. Fifty percent (6 patients) had severe infections and died before their last culture results, and the overall mortality was 91.6% (11) despite therapy and medical support. Conclusions: MR-PA emerges in severe and persistent infections resulting in adverse outcomes. Highly epidemiological clinical suspicion, early empiric treatment with combined therapy and higher dosage should be considered for AIDS patient during epidemics or in high MR-PA prevalence setting, as well as prolonged or suppressive antibiotic therapy, in view of the persistence characteristic and high mortality of MR-PA infection in these patients. Early multidisciplinary control interventions are essentials to reduce the burden caused by this agent as well as the development of better treatment armamentaria for MR-PA infection.