864 AJVR, Vol 62, No. 6, June 2001 P seudomonas aeruginosa is an aerobic gram-negative psychrotrophic bacterium associated with nosoco- mial infections in humans and occasionally with envi- ronmentally related mastitis in cattle, goats, and sheep. 1-3 Its prevalence has been related to its ability to survive in humid environments and food and its resis- tance to antimicrobials. 4 In association with mastitis in ruminants, P aeruginosa has been isolated from milk, water, hoses, and sanitizers that are used topically on the mammary gland. 5 In association with human noso- comial infections, P aeruginosa has been isolated from a variety of fomites (eg, ventilators) as well as body flu- ids and inflammatory exudates. 6 Immunocompromised patients (eg, those with severe burns or cystic fibrosis and those receiving cancer chemotherapy or intensive care) appear to have a greater risk of developing infec- tion with strains of P aeruginosa that are resistant to antimicrobial treatment, compared with other patients. 7 Several methods used to type P aeruginosa isolates have been examined. Phenotypic typing procedures yield poor typeability; for instance, serotyping may fail to type as many as 58.1% of isolates. 8 In contrast, geno- typing procedures such as analysis of ribosomal RNA yield higher typeability and discriminatory indices > 95%. 9,10 Pulsed-field gel electrophoresis (PFGE) has an index of discrimination as great as 0.998 in typing isolates of P aeruginosa. 11,12 However, even this method does not yield 100% typeability. 13 In addition, PFGE has poor interlaboratory reproducibility because of inadequate standardization. 14 Arbitrary primer poly- merase chain reaction (PCR) may achieve a greater degree of discrimination than chromosomal DNA via digestion with EcoRI (ribotyping). 15 Results of other studies indicate that conventional ribotyping is more discriminatory than ribo-PCR. 16 Evaluations of restric- tion endonucleases indicate that PvuII-based ribotyp- ing is more discriminatory than EcoRI-based meth- ods. 17,18 Ribotyping performed by use of ClaI digestion Received May 26, 2000. Accepted Aug 21, 2000. From the Departments of Population Medicine and Diagnostic Sciences (Rivas, González, Lein) and Biomedical Sciences (Quimby), College of Veterinary Medicine, and the Laboratory for Molecular Typing, Department of Food Science, College of Agriculture and Life Sciences, Cornell University, Ithaca, NY 14850 (Bodis, Batt); DuPont Qualicon, 3531 Silverside Rd, Wilmington, DE 19810 (Bruce); the Department of Farm Animal Health and Resource Management, College of Veterinary Medicine, North Carolina State University, Raleigh, NC 27606 (Anderson); and the Department of Pathology, The New York Hospital-Cornell Medical Center, New York, NY 10021 (Klein). Supported in part by the New York Department Board of Agriculture and Markets. The authors thank Phyllis Weybrecht, Juncal Arrobas, Ainhoa Arina, Argia Alba, Dr. Sharon Sickles, and Dr. Ozden Coksaygan for technical assistance. Molecular epidemiologic features and antimicrobial susceptibility profiles of various ribotypes of Pseudomonas aeruginosa isolated from humans and ruminants Ariel L. Rivas, DVM, PhD; Mary Bodis; James L. Bruce; Kevin L. Anderson, DVM, PhD; Renate F. Klein, PhD; Rubén N. González, DVM, PhD; Fred W. Quimby, VMD, PhD; Carl A. Batt, PhD; Donald H. Lein, DVM, PhD Objectives—To assess automated ribotyping for characterization of Pseudomonas aeruginosa isolates and to identify their type prevalence and geographic distribution. Sample Population—39 human and 56 ruminant P aeruginosa isolates. Procedures—Isolates were identified by use of bac- teriologic techniques and automated PvuII-based ribo- typing. Susceptibility to antimicrobials was tested in vitro. Data were analyzed for index of discrimination; prevalence ratio; geographic distribution of ribotypes found only in humans, only in cows, or only in goats (single-host ribotypes); and geographic distribution of ribotypes found in humans and ruminants (multihost ribotypes). Results—All isolates were typeable (45 ribotypes, 35 single-host ribotypes). Ribotyping index of discrimi- nation was 0.976. More isolates (45.3%) than expected yielded multihost ribotypes (22% of all ribotypes). Although 8.6% of single-host ribotypes were found in 4 or more isolates, 60% of multihost ribotypes were found in 4 or more isolates. Ninety percent of multihost ribotypes were isolated from different geographic areas, whereas 3.0% of single- host ribotypes were isolated from different geo- graphic areas. All ruminant isolates were susceptible to gentamicin and polymyxin B. In contrast, antibi- ogram profiles differed for human isolates from dif- ferent geographic areas. Susceptibility to antimicro- bials differentiated 6 isolates not distinguished by ribotyping. Conclusions and Clinical Relevance—Automated ribotyping with PvuII discriminated more isolates than in vitro antimicrobial susceptibility. In combination, both tests provided more information than either test alone. Given the greater prevalence and geographic distribution of multihost ribotypes, immunocompro- mised humans and lactating ruminants may have a greater risk for disease if exposed to multihost P aeruginosa ribotypes, compared with single-host ribo- types. (Am J Vet Res 2001;62:864–870)