ORIGINAL ARTICLE Almond oil implicated in a Staphylococcus capitis outbreak in a neonatal intensive care unit C Gras-Le Guen 1,2 , S Fournier 1 , B Andre-Richet 3 , J Caillon 2,3,4 , C Chamoux 4 , E Espaze 4 , H Richet 3 , JC Roze 1 and D Lepelletier 2,3 1 Re´animation pe´diatrique et ne´onatale, CHU, Nantes, France; 2 Faculte´de me´decine, Laboratoire de The´rapeutiques cliniques et expe´rimentales des infections, EA UPRES 3826, Nantes, France; 3 Unite´d’hygie`ne hospitalie`re, CHU, Nantes, France and 4 Laboratoire de bacte´riologie, CHU, Nantes, France Objective: To develop an effective outbreak-control strategy by identifying the source and modes of transmission of Staphylococcus capitis in a 60-bed neonatal intensive care unit (NICU). Study Design: We conducted a study among neonates hospitalized during the outbreak (June 2000 through November 2003). All cases of S. capitis colonization or infection detected by clinical samples during the outbreak were included. The molecular analysis of the isolated was assessed by pulsed-field electrophoresis. We reported the description of the outbreak and the measures taken during this investigation. Result: Thirty-three patients were colonized or infected by S. capitis. Mean gestational age was 28.5±4.4 weeks of gestation, mean birth weight was 1068±637.3 g and the mean length of hospital stay was 77.9±35.9 days. We observed that positive S. capitis cultures were over-represented in six beds of the NICU. Because S. capitis is known to thrive in lipid media, we cultured samples from the almond oil bottles assigned to these beds. S. capitis strain recovered from one of the almond oil sample was genetically identical to the strain recovered from the cases. Conclusion: Almond oil is an unusual reservoir infection. Control policy allowed prompt institution of measures that were successful in ending the outbreak. Journal of Perinatology (2007) 27, 713 – 717; doi:10.1038/sj.jp.7211798; published online 6 September 2007 Keywords: nosocomial infection; prematurity; Gram positive Introduction Coagulase-negative staphylococci are the organisms most commonly isolated from critically ill neonates with hospital-acquired bacteremia. 1,2 Coagulase-negative staphylococci are commensals of the skin and mucous membranes, and although they occasionally cause disease in humans, their virulence is usually limited. 2,3 They have caused a few cases of bacteremia, most notably in infants with central venous catheters or respiratory infections. 4 Because premature infants are fragile, they can exhibit severe symptoms in response to coagulase-negative infection. Among the 30 coagulase-negative infection strains identified to date, Staphylococcus epidermidis and Staphylococcus haemolyticus have been the main causes of infection in humans. 5 Reported cases of Staphylococcus capitis infection in humans have been rare. 6 In July 2002, an outbreak of S. capitis infection occurred in the neonatal intensive care unit (NICU) of the Nantes Tertiary-Care Teaching Hospital. Genotype studies showed that the 33-affected infants carried organisms derived from a single clone. The aim of the present study was to develop an effective outbreak-control strategy by identifying the source and modes of transmission of the organism. Methods Definitions and case ascertainment The study patients were neonates admitted to the NICU between 17 June 2000 (identification of the first case of S. capitis-positive culture) and 20 November 2003. Our weekly surveillance of NICU microbiological results detected an unusual increase in S. capitis and the outbreak was formally identified in May 2002, and then the surveillance was heightened. A case was defined as a neonate in whom S. capitis isolates were recovered in cultures of blood, central venous catheters, tracheal fluid aspirates or abscess contents. Some patients with positive isolates were identified retrospectively by reviewing the microbiology laboratory records. Cases were then divided into patients with infection (symptoms consistent with S. capitis infection, with laboratory evidence of inflammation) and patients with colonization (no symptoms or laboratory evidence of inflammation). The study period extended from the day on which the outbreak was first reported to November 2003. Screening for Received 27 November 2006; revised 24 May 2007; accepted 19 June 2007; published online 6 September 2007 Correspondence: Dr C Gras-Le Guen, Neonatal Intensive Care Unit, Boulevard Jean Monnet, Ho ˆpital Me `re Enfant, Nantes 44 093, France. E-mail: christele.grasleguen@chu-nantes.fr Journal of Perinatology (2007) 27, 713–717 r 2007 Nature Publishing Group All rights reserved. 0743-8346/07 $30 www.nature.com/jp