LETTER TO THE EDITOR
“Probable Contaminants” No More: Rapid Identification of GramPositive Rods Leads to Improved Clinical Care Genevieve McKew, Bradley Watson, Raymond Chan, Sebastiaan J. van Hal Department of Infectious Diseases and Microbiology, Royal Prince Alfred Hospital, Camperdown, New South Wales, Australia
reviously, identification to the species level of “nonsignificant” bacteria (e.g., single specimens with Gram-positive rods or coagulase-negative staphylococci) was generally not performed in our laboratory, including for blood culture isolates. Reasons for this included the low pretest probability of these bacteria being true pathogens and the expense of identification methods. Methods used were API strips (bioMérieux, France) and 16S rRNA sequencing, with turnaround times of between 48 h and 2 weeks. Consequently, if an isolate was considered significant once the identity was confirmed, attempting to establish the source of bacteremia was difficult, as nonsterile specimens are kept for only 7 days. Our laboratory recently introduced the Bruker matrix-assisted laser desorption ionization–time of flight (MALDI-TOF) identification system utilizing the Biotyper database. Given the ease of species identification, this is now performed on all blood culture isolates, regardless of clinical significance, and all other significant isolates (i.e., from a sterile site or a pure/predominant growth from a nonsterile site). The utility of this new technology to patient care is demonstrated by a recent case. A 69-year-old man receiving radiotherapy for prostate cancer developed sepsis secondary to a urinary tract infection. Urine was cultured on 5% horse blood agar (HBA) and incubated in 5% CO2 only at 35°C. Three different organisms were isolated from urine at ⬎108 CFU/liter, namely, two colonial types of coagulase-negative staphylococci and a Gram-positive rod, which were deemed nonsignificant. According to usual laboratory protocols, in the presence of pyuria, positive culture plates are stored for 7 days. Anaerobic and aerobic blood culture bottles from the same day were flagged positive on days 2 and 4, respectively, for a short Gram-positive rod. Following inoculation of 5% HBA and incubation in 5% CO2 for 24 h, a pure organism was isolated. Growth, however, was sufficient to attempt identification only at 48 h. On day 4 postcollection, the isolate was identified as Actinobaculum schaalii by MALDI-TOF mass spectrometry, with a score of 2.3 (i.e., a reliable score to the species level) using the direct colony transfer method (1–3). The rod from his urine culture was retrieved and identified as the same organism, by the same method. Interestingly, a urine culture from the same patient a month earlier showed pyuria and Gram-positive rods. He had been treated empirically with ceftriaxone and improved. Following identification, his antibiotics were changed to amoxicillin for an extended duration, owing to reports of failures of shorter treatment courses for this organism (4) and resistance to commonly prescribed empirical treatments for culture-negative urinary tract infection, like trimethoprim and ciprofloxacin (5). Antimicrobial susceptibility testing was not pursued for this isolate, as the patient had improved and testing would have re-
May 2013 Volume 51 Number 5
quired referral to an external laboratory. On follow-up 2 months later, the pyuria had resolved, with no organisms isolated from the urine. A. schaalii is a slow-growing, often CO2-requiring organism that has emerged as a significant uropathogen. It is likely to have been underdiagnosed in the past, as it may not grow under routine conditions (4). Prior to the introduction of the MALDI-TOF system, identification would have taken, on average, 8 days using either 16S sequencing or phenotypic methods. Thus, this case illustrates the importance of real-time species-level identification in making the correct diagnosis and directing appropriate antimicrobial therapy. In addition, it is likely that other isolates previously regarded as contaminants may be pathogens. With the advent of these rapid and inexpensive technologies, we should review what is considered a significant pathogen. In conclusion, real-time species-level identification enables microbiologists to become familiar with novel pathogens and provide timely advice to clinicians. This will allow more-rapid directed antimicrobial therapy, which should assist with antimicrobial stewardship and may lead to improved clinical outcomes. REFERENCES 1. Farfour E, Leto J, Barritault M, Barberis C, Meyer J, Dauphin B, Le Guern AS, Leflèche A, Badell E, Guiso N, Leclercq A, Le Monnier A, Lecuit M, Rodriguez-Nava V, Bergeron E, Raymond J, Vimont S, Bille E, Carbonelle E, Guet-Revillet H, Lécuyer H, Beretti J-L, Vay C, Berche P, Ferroni A, Nassif X, Join-Lambert O. 2012. Evaluation of the Andromas matrix-assisted laser desorption ionization–time of flight mass spectrometry system for identification of aerobically growing Gram-positive bacilli. J. Clin. Microbiol. 50:2702–2707. 2. La Scola B, Fournier PE, Raoult D. 2011. Burden of emerging anaerobes in the MALDI-TOF and 16S rRNA gene sequencing era. Anaerobe 17:106 – 112. 3. Dusch H, Burren K, Hinrikson HP. 25 July 2010, posting date. MALDI-TOF mass spectrometry for routine identification of medically relevant bacteria. Unilabs Mittelland, Berne, Switzerland. http://www.unilabs.ch/Lists/Scientific Publication/Poster%20MALDI%20SSM%202010-HD1.pdf. 4. Cattoir V. 2012. Actinobaculum schaalii: review of an emerging uropathogen. J. Infect. 64:260 –267. 5. Cattoir V, Varac A, Greub G, Prod’hom G, Legrad P, Lienhard R. 2010. In vitro susceptibility of Actinobaculum schaalii to 12 antimicrobial agents and molecular analysis of fluoroquinolone resistance. J. Antimicrob. Chemother. 65:2514 –2517.
Published ahead of print 13 February 2013 Address correspondence to Genevieve McKew, [email protected]
Copyright © 2013, American Society for Microbiology. All Rights Reserved. doi:10.1128/JCM.03358-12
Journal of Clinical Microbiology