Staphylococcus lugdunensis is purported to be the most virulent of the more than 20 clinically relevant coagulase-negative Staphylococcus (CNS) species1,2 and is a well-described cause of bacterial endocarditis,2 but its importance as a pediatric pathogen is largely unknown. As in adults, endocarditis in children has been described, with case reports of S. lugdunensis endocarditis in a 16 year-old boy with remote aortic valve angioplasty and in a 7-year-old boy with bicuspid aortic valve.3,4S. lugdunensis has also been reported as a cause of ventriculoperitoneal shunt infection and acute pyelonephritis in children.5,6
In addition to invasive infections such as endocarditis,S. lugdunensis has recently been suggested to be a common cause of skin and soft tissue infections. In a prospective study from Denmark that involved primarily adult patients, 13% of CNS from skin and soft tissue infections were identified as S. lugdunensis and 85% of those isolates were considered pathogens, with S. lugdunensis often found in pure culture suggesting a pathogenic role for the organism.7
It is possible that the importance of S. lugdunensis has been underestimated in the past due to difficulties in identifying it accurately in the clinical laboratory. For example, organism colony morphology may be variable with regard to colony size, pigmentation and hemolysis. Pure growth of S. lugdunensis infections can thus sometimes resemble mixed growth of Staphylococcus species, which potentially would be deemed as clinically insignificant.1 Conversely, it can potentially be misidentified as S. aureus, because S. lugdunensis may be positive by slide coagulase or latex agglutination tests in approximately 50% of cases because it possesses fibrinogen binding factor. S. lugdunensis is often identified by positive reactions for the biochemical tests for the enzymes pyrrolidonyl aminopeptidase (PYR) and ornithine decarboxylase (ODC) but the accuracy of using these 2 tests alone is uncertain.
Given the reported virulence of S. lugdunensis and because to our knowledge there have been no systematic studies of the prevalence and clinical significance of S. lugdunensis in a solely pediatric population, we undertook a prospective and retrospective study to look for S. lugdunensis at our pediatric microbiology laboratory, using molecular identification techniques to ensure accurate identification.
MATERIALS AND METHODS
The Children’s Hospital of Eastern Ontario (Ottawa, Ontario, Canada) has 167 inpatient beds and almost 60,000 emergency department visits each year. A prospective study was conducted for a 5-month period between mid-July and mid-December 2009. During this time, all sterile site isolates of CNS were studied, such as blood, cerebrospinal fluid, biopsies and other sterile fluids. Wound isolates were included in the study if they were hemolytic, pure or at least moderate in quantity on gram stain or culture. Urine CNS isolates were included only if the organism concentration was greater than 4×103 colony forming units/mL and no other pathogen was found. Finally, any CNS that were latex agglutination positive (Pastorex Staph-Plus, Biorad Laboratories, Montreal, Quebec, Canada) but tube coagulase negative were also included. We also retrospectively studied all sterile site CNS isolates recovered over a 1-year period from July 2008 to August 2009.
All isolates that met the above criteria were tested with PYR (PML Microbiologicals, Wilsonville, OR) for up to 3 minutes and, if PYR positive, were tested for ODC (Oxoid, Ottawa, Ontario, Canada) for over 16 hours. All PYR and ODC double positive isolates were tested with an automated identification system (VITEK 2, bioMerieux, Durham, NC) and sent to the National Microbiology Laboratory (Winnipeg, Manitoba, Canada) for definitive identification, using conventional biochemical,8 by nearly full 16S ribosomal RNA9 and rpoB sequencing.10
Determination of the clinical significance of the S. lugdunensis was done by the judgment of the 2 staff members (RS and FC). This was after reviewing clinical and microbiologic data (clinical presentation, laboratory values, treatment course if any and quantity of growth or pus cells).
Three hundred forty-seven CNS met the study inclusion criteria: 206 from the prospective study (181 from nonsterile sites and 25 from sterile sites); and 141 from the retrospective sterile site study (87 blood culture isolates, 23 cerebrospinal fluid isolates and 31 from other sterile sites).
Of these 347 isolates, 12 (3.5%) were positive with both the PYR and ODC tests. Seven of these 12 were identified asS. lugdunensis by molecular methods, corroborated by conventional biochemical testing, for an overall prevalence of 2.0% (7/347). The remaining 5 PYR-positive and ODC-positive isolates were identified by molecular methods as S. epidermidis (2 isolates) and S. warneri (2 isolates) and S. hominis (1 isolate). (Microbiologic results of the retrospective and prospective studies are summarized in Fig., Supplemental Digital Content 1, https://links.lww.com/INF/B361.)
Clinical information for the 7 S. lugdunensis isolates is shown in Table 1. Only 1 of the 7 S. lugdunensis isolates was considered to be possibly clinically significant. This isolate grew from a sterile site specimen (an intraoperative vertebral bone specimen from a patient who had undergone spinal fixation). The sterile site culture prevalence S. lugdunensis was therefore 0.6% (1/166).
The VITEK 2 correctly identified all 7 isolates as S. lugdunensis that were identified as that species using molecular reference methods. The VITEK 2 incorrectly identified 1 S. warneri isolate as S. lugdunensis.
S. lugdunensis was found infrequently in clinical specimens from pediatric patients at our hospital. As well, we did not find this organism in acute or aggressive infections as described in the literature, because only 1 of 7 isolates was judged to be possibly clinically significant. This isolate was recovered from a surgically obtained bone specimen from a patient with spinal fixation hardware. S. lugdunensis has been described as a cause of similar infections and was judged to be responsible for 7% of bone and joint infections in 1 study.11
The overall prevalence of 2.0% we observed was considerably lower than that reported in the prospective study of skin and soft tissue infections from Denmark which had an overall prevalence of 13%.7 The reasons for the lower prevalence observed at our site are uncertain. We used molecular sequencing to identify S. lugdunensis, leading to the exclusion of some isolates that would have been misidentified as S. lugdunensis if only biochemical screening with PYR and ODC was used. Of note, a nonmolecular method was used as the reference method in the prospective study by Bocher et al,7 but whether use of molecular identification would have affected the results of that study is unknown.
The low prevalence of S. lugdunensis in our setting magnified the importance of the relatively high number of false positives seen with use of PYR and ODC positivity to identify S. lugdunensis. Five of 12 isolates (41.7%) would have been incorrectly identified as S. lugdunensis with use of PYR and ODC alone. VITEK 2 testing of PYR+/ODC+ isolates appears to provide moderately accurate confirmation, with no false positives in 1 samples and 1 false negative of 5 samples.
There are some limitations to our study. First, we did not attempt to identify S. lugdunensis among all CNS isolates grown in our laboratory, because we included only those we felt might be clinically significant based upon amount and purity of growth. Thus, some S. lugdunensis may not have been detected due to the selection criteria used. Second, only PYR+/ODC+ isolates were investigated. Any PYR-negative or ODC-negative S. lugdunensis strains would also have been missed; however, such strains appear to be rare.1
We conclude that given the low prevalence and clinical significance of S. lugdunensis in pediatric specimens, routine testing of CNS to identify S. lugdunensis from nonsterile sites does not appear to be warranted in our setting. Identification from sterile site specimens such as blood may be justifiable, given the reported virulence of S. lugdunensis, despite the low prevalence we observed. We also recommend that molecular methods be used should definitive identification of S. lugdunensis be required.
1. Frank KL, Del Pozo JL, Patel R. From clinical microbiology to infection pathogenesis: how daring to be different works for Staphylococcus lugdunensis.
Clin Microbiol Rev. 2008;21:111–133
2. Liu PY, Huang YF, Tang CW, et al. Staphylococcus lugdunensis
infective endocarditis: a literature review and analysis of risk factors. J Microbiol Immunol Infect. 2010;43:478–484
3. Sotutu V, Carapetis J, Wilkinson J, et al. The “surreptitious Staphylococcus”: Staphylococcus lugdunensis
endocarditis in a child. Pediatr Infect Dis J. 2002;21:984–986
4. Jones RM, Jackson MA, Ong C, et al. Endocarditis caused by Staphylococcus lugdunensis
. Pediatr Infect Dis J. 2002;21:265–268
5. Elliott SP, Yogev R, Shulman ST. Staphylococcus lugdunensis
: an emerging cause of ventriculoperitoneal shunt infections. Pediatr Neurosurg. 2001;35:128–130
6. Casanova-Roman M, Sanchez-Porto A, Casanova-Bellido M. Urinary tract infection due to Staphylococcus lugdunensis
in a healthy child. Scand J Infect Dis. 2004;36:149–150
7. Böcher S, Tønning B, Skov RL, et al. Staphylococcus lugdunensis
, a common cause of skin and soft tissue infections in the community. J Clin Microbiol. 2009;47:946–950
8. Karsten B, von Eiff CVersalovic J, Carroll KC, Funke G. Staphylococcus, Micrococcus
and other catalase positive cocci. In: Manual of Clinical Microbiology. 2011;Vol. 110th ed Washington, DC ASM Press:308–330
9. Bernard KA, Shuttleworth L, Munro C, et al. Propionibacterium australiense
sp. nov., derived from granulomatous bovine lesions. Anaerobe. 2002;8:41–47
10. Drancourt M, Raoult D. rpoB gene sequence-based identification of Staphylococcus
species. J Clin Microbiol. 2002;40:1333–1338
11. Sivadon V, Rottman M, Chaverot S, et al. Use of genotypic identification by sodA sequencing in a prospective study to examine the distribution of coagulase-negative Staphylococcus
species among strains recovered during septic orthopedic surgery and evaluate their significance. J Clin Microbiol. 2005;43:2952–2954