Surveillance for Carbapenem-Resistant Pseudomonas aeruginosa at Five United States Sites—2015

Abstract

Background. Pseudomonas aeruginosa causes approximately 51,000 healthcareassociated infections in the U.S. annually. P. aeruginosa can acquire resistance to carbapenems through chromosomal mutations, changes in gene expression, and acquisition of mobile elements harboring resistance genes. We piloted surveillance for carbapenem-resistant P. aeruginosa (CRPA). Methods. During July-October 2015, laboratory-based surveillance was conducted at sentinel facilities in Georgia, New Mexico, Oregon and Tennessee, and at all facilities in Rochester, New York. A case was isolation of P. aeruginosa resistant to imipenem, meropenem or doripenem from a clinical specimen (except rectal, peri-rectal, or nasal swabs); an incident case was the first case occurring in a patient in a 30-day period. Patient medical records were reviewed. The proportion resistant was determined by dividing the number of cases by the number of isolates tested for carbapenem susceptibility. Isolates were screened by PCR for carbapenemase-encoding genes. Results. During the 4-month period, 384 (9.1%) of 4243 isolates were carbapenem- resistant; the percent resistant ranged from 4.6% (26 of 560) in Oregon to 12% (68 of 566) in Georgia. We identified 294 incident cases among 274 patients. In New York, there were 8 incident cases per 100,000 population. Patients had a median age of 66 years (range: <1-98); 115 (42.0%) were female; the most common underlying condition was chronic pulmonary disease (98 of 272; 36%). Specimens were most commonly from respiratory sites (118; 40.1%) or urine (110; 37.4%); 64 (21.8%) were collected after the third hospital day. Only 23 (7.8%) cases were from patients without any identified healthcare risk factors; the most frequent healthcare risk factor was hospitalization in the prior year (208 of 253; 82.2%). Death occurred in 27 of 289 (9.3%) cases. Carbapenemase-encoding genes were identified in 2 of 131 (1.5%) isolates tested; both were from New Mexico and carried either blaVIM-2 or blaIMP-18. Conclusion. These data suggest that the incidence of CRPA might be greater than that of CR-Enterobacteriaceae and CR-Acinetobacter. Isolates with plasmid-mediated carbapenemases were rare. Multiple years of population-based CRPA rate data are needed to understand CRPA burden and the prevalence of carbapenemase-producing CRPA.