SM is a nosocomial pathogen requiring strict control measures to prevent spread. The organism is multidrug resistant and can be devastating for compromised hosts such as high risk neonates. Risk factors for SM I/C in NICU have not been studied. Identification of risk factors in NICU would be critical in establishing effective control measures. We had an outbreak of SM in our NICU. We conducted an epidemiologic investigation, obtaining surveillance cultures from respiratory equipment, temperature probes and water faucets, as well as respiratory cultures from all patients in NICU at a certain time. SM isolates were DNA typed by Pulsed Field Gel Electrophoresis(PFGE) to confirm if they belonged to the same strain. We also conducted a case-control study to identify risks factors of SM I/C in NICU. Cases of SM I/C between 9/96 and 6/97 were matched to controls with regards to gestational age and period of hospitalization. Variables compared among cases and controls included type of mechanical ventilation, type and length of antibiotics, indwelling catheters, steroids and neutropenia. Surveillance environmental and respiratory cultures were all negative. PFGE done on five isolates revealed four different strains. The two patients with the same strain were admitted one week apart. Six cases of SM I/C were matched to 10 controls. Statistical analysis of potential risk factors using Pearson correlation coefficient revealed that patients pretreated with third generation cephalosporins and ventilated using High Frequency Oscillatory Ventilation showed a trend for higher risk of colonization/infection (p=0.09 and p=0.152 respectively). The small size of our sample limited our ability to find statistical significant risk factors. We conclude that although patient-to-patient transfer of SM may have occured, most of the acquisition occured independently. Host-centered risk factors for SM I/C in a larger NICU population deserves further investigation.