Abstract
The use of matrix-assisted laser desorption ionization–time of flight mass spectrometry (MALDI-TOF MS) for staphylococcal identification is now considered routine in laboratories compared with the conventional phenotypical methods previously used. We verified its microbiological relevance for identifying the main species of coagulase-negative staphylococci (CoNS) by randomly selecting 50 isolates. From 1 January 2007 to 31 August 2008, 12,479 staphylococci were isolated with phenotypic methods, of which 4,594 were identified as Staphylococcus aureus and 7,885 were coagulase negative staphylococci. Using MALDI-TOF MS from 1 January 2011 to 31 August 2012, 14,913 staphylococci were identified, with 5,066 as S. aureus and 9,847 as CoNS. MALDI-TOF MS allowed the identification of approximately 85% of the CoNS strains, whereas only 14% of the CoNS strains were identified to the species level with phenotypic methods because they were often considered contaminants. Furthermore, the use of MALDI-TOF MS revealed the occurrence of recently characterized Staphylococcus species, such as S. pettenkoferi, S. condimenti, and S. piscifermentans. Microbiological relevance analysis further revealed that some species displayed a high rate of microbiological significance, i.e., 40% of the S. lugdunensis strains included in the analysis were associated with infection risk. This retrospective microbiological study confirms the role of MALDI-TOF MS in clinical settings for the identification of staphylococci with clinical consequences. The species distribution reveals the occurrence of the recently identified species S. pettenkoferi and putative virulent species, including S. lugdunensis.
INTRODUCTION
Coagulase-negative staphylococci (CoNS) are normal inhabitants of the human skin that lately have revealed themselves as significant etiological agents causing nosocomial infections, particularly in medical devices (1–4). Until recently, the phenotypic identification of >40 currently recognized CoNS species remained a challenge, because the technique was labor-intensive and uncommon CoNS were usually not identified (5, 6). Until recently, CoNS identification at the species level was performed only for clinical isolates that were considered clinically significant according to clinical or microbiological criteria. Meanwhile, matrix-assisted laser desorption ionization–time of flight mass spectrometry (MALDI-TOF MS) has emerged as a reliable, fast, and cost-effective alternative for bacterial species identification and is now widely used in microbiology laboratories (7, 8). This method allows the identification of all staphylococci isolated in the clinical setting, thus rendering the use of well-known tests, such as free or bound coagulase tests, unnecessary. Previous reports have shown that MALDI-TOF MS is a reliable method to identify Staphylococcus at the species level; however, questions regarding the impact of such methods in clinical settings remain uncertain (9). Staphylococcus epidermidis is probably the major CoNS species causing nosocomial infections, although other CoNS, such as S. lugdunensis, S. schleiferi, and S. saprophyticus, have been reported as even more pathogenic in a range of both nosocomial and community-acquired infections (10–13).
The present study retrospectively compares the results of staphylococcal identifications in the routine workflow of a clinical laboratory before and after implementation of MALDI-TOF MS, details the methicillin susceptibility of the isolates, and compares the pathogenic roles of the various identified CoNS species based on microbiological data. Of the microbiological data, 14,913 staphylococci were identified by MALDI-TOF MS during a 16-month period from almost all clinical specimens without taking into account their possible involvement in a pathogenic process. Using this systematic method, an accurate representation of the various staphylococcal species was obtained in clinical specimens, and novel data for S. pettenkoferi and S. lugdunensis were acquired.
MATERIALS AND METHODS
Laboratory identification of CoNS.
All clinical samples were collected at the University Hospital of Strasbourg, France, and microbiological analyses were performed at the hospital laboratory. Our university hospital has 2,428 beds with 87,758 days of hospitalization in 2014 (13,649 days for patients <18 years). The number of day care center hospitalizations was 43,796 (6,594 days for patients <18 years). The median hospital stay is 708 days.
Phenotypic identification was performed for all specimens considered to be significant according to the 2007 Société Française de Microbiologie criteria (14). The strains isolated were first tested for free and bound coagulase. The tube coagulase test for the detection of free coagulase was performed with reconstituted rabbit plasma in EDTA (Bio-Rad Laboratories, Marne-La-Coquette, France). The slide agglutination test (combined detection of clumping factor, staphylococcal protein A, and capsular polysaccharides) was performed with the Pastorex Staph-Plus latex agglutination test (Bio-Rad Laboratories). Staphylococci positive for those two tests were considered to be S. aureus, and those negative for both tests or with discordant results were tested with Vitek 2. Analysis was performed from 1 January 2007 to 31 August 2008 using a Vitek 2 system with Gram-positive (GP) and ANC cards for S. saccharolyticus, which is anaerobic (bioMérieux, Marcy l'Etoile, France), according to the manufacturer's instructions.
In 2010, MALDI-TOF MS identification was implemented with a MALDI-TOF Microflex LT/LRF coupled to the MALDI Biotyper algorithm, as recommended by the manufacturer (Bruker Daltonics), which replaced phenotypic identification of Staphylococcus species. This identification was performed for all staphylococci except those obtained from throat and vaginal swabs that were examined only for beta-hemolytic streptococci according to the 2010 Société Française de Microbiologie criteria (15). Strains of CoNS were plated on Columbia agar with 5% sheep blood and incubated for 24 h at 37°C. One colony from each strain was placed on a 96-spot polished target plate and allowed to dry in room air. Next, 1 μl of matrix (3 mg/ml α-cyano-4-hydroxycinnamic acid in 50% [vol/vol] acetonitrile–2.5% trifluoroacetic acid [vol/vol]) was overlaid onto the dried sample and allowed to dry.
The acquisition of protein mass spectra was performed on the Microflex LT instrument using the flexControl 3.0 software (Bruker Daltonics), with a mass-to-charge ratio (m/z) range of 2 to 20 kDa. Automated data analysis of raw spectra was performed by the MALDI Biotyper RTC 3.0 software (Bruker Daltonics). During the study, the version 3.0.2.0 of the library (3,476 spectra, including 132 spectra from 38 Staphylococcus species) was successively updated to version 3.1.1.0 (3,740 spectra, including the same number of Staphylococcus spectra) and then to version 3.1.2.0 (3,955 spectra, including 139 spectra from the same 38 Staphylococcus species as those of the previous versions) (see Table S1 in the supplemental material).
An identification score of ≥2 was considered a high-confidence identification to the species level, whereas scores of 1.7 to 1.99 were considered intermediate confidence genus-level identification only. Scores of <1.7 were considered an unacceptable identification, according to the manufacturer's recommendations. The results of the identifications from 1 January 2011 to 31 August 2012 were retrospectively analyzed, and strains with an identification (ID) score of <2 were excluded from the analysis.
Antimicrobial susceptibility testing.
Antimicrobial susceptibility testing was performed with either Vitek 2 or agar diffusion, according to the recommendations of the European Committee on Antimicrobial Susceptibility Testing for the 2011 to 2012 study period (http://www.eucast.org/).
Assessment of microbiological relevance.
The microbiological relevance of 50 samples consecutively selected for each species of CoNS was evaluated according to the European manual of clinical microbiology (EMCM) criteria for interpretation (17) (Table 1). Analysis was started from the most recent sample to each previous sample, until reaching a total of 50 samples. For blood cultures, CoNS were not considered contaminants if there were at least two positive samples for the same bacterium (same species and same antimicrobial susceptibility) in two different sets of blood cultures. For the catheter samples, culture was significant at ≥1,000 CFU/ml or inferior provided it was associated with one blood culture positive for the same bacterium, according to the Cleri criteria (18) modified by Brun-Buisson (19). For urine samples, bacteriuria was significant at ≥105 CFU/ml, when associated with leukocyturia at ≥104/ml, or when ≥103 CFU/ml was detected for S. saprophyticus. For other specimens, the criteria previously reported for Corynebacterium were used, because CoNS are also inhabitants of skin and mucosal membranes (17). The doubtful category was considered to be not relevant, and those strains were considered to be contaminants in the analysis.
TABLE 1.
Criteria for microbiological relevance of bacteria isolated from clinical specimens according to the European Manual of Clinical Microbiology
| Specimen type | Culture result | Direct examination result | Associated flora | Considered significant |
|---|---|---|---|---|
| Blood culture | ≥2 positive bottles with same bacterium | Yes | ||
| 1 positive/1 performed | No, but considered doubtful if patient is coming from an oncology ward | |||
| 1 positive/≥2 performed | No | |||
| Urine | ≥105 CFU/ml | ≥104 leukocytes/ml | Yes | |
| Other results | No | |||
| Intravascular catheter | >1,000 CFU/ml | Yes | ||
| <1,000 CFU/ml plus positive blood culture with same bacterium | Yes | |||
| Other results | No | |||
| Superficial | ≥2 positive | Cocci Gram positive | Yes | |
| 1 positive/1 performed | Cocci Gram positive | 1 species | Yes | |
| Other results | No | |||
| Deep | ≥2 positive with same bacterium | Yes | ||
| 1 positive | Cocci Gram positive | 1 species | Yes | |
| 1 positive | Presence of neutrophil polynuclear cells | Doubtful | ||
| Other results | No |
The CoNS studied were S. capitis, S. caprae, S. cohnii, S. epidermidis, S. haemolyticus, S. hominis, S. lugdunensis, S. pettenkoferi, S. saprophyticus, S. schleiferi, S. simulans, and S. warneri. Only 45 samples were studied for S. cohnii and 42 samples for S. schleiferi. Representations of other species were insufficient to be included in this assessment.
Statistical analysis.
Statistical analysis was performed on the results of staphylococcal identification between periods 1 and 2 to determine whether staphylococcal identification with MALDI-TOF MS was significantly different from the phenotypic identification for each species. The two periods were compared using a chi-square test, with a P value of <0.05 considered statistically significant.
We also tested the microbiological relevance of 50 randomly selected clinical strains for statistical significance. Repartition between the two groups, relevant or not relevant, on microbiological criteria was tested using a chi-square test, with a P value of <0.05 considered statistically significant.
RESULTS
Laboratory identification of CoNS with phenotypic methods.
From 1 January 2007 to 31 August 2008 (period 1, Table 2), 12,479 staphylococci were isolated, of which 4,594 (36.8%) were identified as S. aureus and 7,885 were identified as coagulase negative. Identification of CoNS during this period was performed with phenotypic methods only. Among CoNS, 6,781 isolates were not identified to the species level because they were considered contaminants, according to the 2007 Société Française de Microbiologie criteria (14). Conversely, 1,104 (14%) isolates were considered significant and identified as follows: S. epidermidis (694, [62.9%]), S. capitis (52 [4.7%]), S. hominis (84 [7.6%]), S. haemolyticus (84 [7.6%]), and 190 others, comprised mainly of S. saprophyticus (80 [7.2%]), S. lugdunensis (25 [2.3%]), and S. warneri (25 [2.3%]).
TABLE 2.
Laboratory identification of staphylococci with phenotypic methods and MALDI-TOF MS in a clinical laboratory
| Organism(s) | No. (%) of strains in period: |
Comparison of periods 1 and 2 (all staphylococci) (P value) | ||
|---|---|---|---|---|
| 1, isolated from 2007 to 2008 by phenotypic methods (n = 12,479) | 2, isolated from 2011 to 2012 by MALDI-TOF MS identification (n = 14,913) |
|||
| All staphylococci | All staphylococci | Blood cultures | ||
| S. aureus | 4,594 (36.8) | 5,066 (34) | <0.001 | |
| Nonidentified CoNS | 6,781 (54.3) | 1,459 (9.8) | ||
| Identified CoNS | 1,104 (8.9) | 8,388 (56.2) | ||
| S. epidermidis | 694 (62.9) | 5,259 (62.7) | 1,388 (26.4) | 0.915 |
| S. hominis | 84 (7.6) | 698 (8.3) | 396 (56.7) | 0.418 |
| S. haemolyticus | 84 (7.6) | 975 (11.6) | 150 (15.4) | <0.001 |
| S. capitis | 52 (4.7) | 552 (6.6) | 194 (35.1) | 0.017 |
| S. warneri | 25 (2.3) | 278 (3.3) | 62 (22.3) | 0.064 |
| S. lugdunensis | 25 (2.3) | 205 (2.4) | 17 (8.3) | 0.716 |
| S. simulans | 7 (0.6) | 78 (0.9) | 2 (2.6) | 0.330 |
| S. saprophyticus | 80 (7.2) | 80 (1) | 3 (3.8) | <0.001 |
| S. caprae | 7 (0.6) | 56 (0.7) | 1 (1.8) | 0.897 |
| S. pettenkoferi | 0 | 54 (0.6) | 30 (55.6) | |
| S. schleiferi | 1 (0.1) | 30 (0.4) | 3 (10) | 0.176 |
| S. cohnii | 10 (0.9) | 35 (0.4) | 5 (14.3) | 0.030 |
| S. pasteuri | 14 (1.3) | 25 (0.3) | 9 (36) | <0.001 |
| S. intermedius | 3 (0.3) | 17 (0.2) | 0 | 0.639 |
| S. sciuri | 1 (0.1) | 14 (0.2) | 5 (35.7) | 0.555 |
| S. auricularis | 3 (0.3) | 9 (0.1) | 0.163 | |
| S. xylosus | 6 (0.5) | 9 (0.1) | 0.002 | |
| S. saccharolyticus | 5 (0.5) | 7 (0.08) | 0.004 | |
| S. condimenti | 0 | 3 (0.04) | ||
| S. piscifermentans | 0 | 3 (0.04) | ||
| S. carnosus | 0 | 1 (0.01) | ||
| S. vitulinus | 1 (0.1) | 0 | ||
| S. lentus | 1 (0.1) | 0 | ||
| S. chromogenes | 1 (0.1) | 0 | ||
Laboratory identification of CoNS with MALDI-TOF MS.
From 1 January 2011 to 31 August 2012 (period 2, Table 2), 14,913 staphylococci were isolated, of which 5,066 (33.9%) were identified as S. aureus and 9,847 were identified as CoNS. One-third of the isolation sites were superficial (cutaneous) or deep anatomical regions (osteoarticular, vascular, digestive, or cerebrospinal), followed by respiratory tract and blood (20% each), biomaterial/catheter (9%), and urine (7%) samples. Individually, each of the other type of samples (stool, ocular, and genital) represented <5% of the anatomical sites for Staphylococcus isolation. Identification of CoNS during this period was performed with MALDI-TOF MS only. Among the CoNS, 1,459 (14.8%) isolates that were not tested with MALDI-TOF MS were from throat and vaginal swabs that were examined only for beta-hemolytic streptococci, according to the 2010 Société Française de Microbiologie criteria (15). The 8,388 CoNS remaining were identified as follows: S. epidermidis (5,259 [62.7%]), S. haemolyticus (975 [11.6%]), S. hominis (698 [8.3%]), S. capitis (552 [6.6%]), and another 904. These other 904 CoNS were mainly S. warneri (278 [3.3%]), S. lugdunensis (205 [2.4%]), S. saprophyticus (80 [1%]), and S. simulans (78 [0.9%]).
Comparison between periods 1 and 2.
Statistical analysis revealed that S. aureus identification was significantly different for the 2 periods (P < 0.005). With regard to CoNS, the identification rates during period 2 were significantly different from those in period 1 for S. haemolyticus, S. capitis, S. saprophyticus, S. cohnii, S. pasteuri, S. xylosus, and S. saccharolyticus. Conversely, the identification rates were not significantly different for S. epidermidis, S. hominis, S. warneri, S. lugdunensis, S. simulans, or S. caprae. Some CoNS identified with MALDI-TOF MS, S. pettenkoferi, S. condimenti, S. carnosus, and S. piscifermentans, were never identified with phenotypic method.
Percentage of strains isolated from blood cultures.
The percentage of strains isolated from blood cultures (Bactec bottles incubated in Bactec 9240 instruments; Becton Dickinson) was determined from 1 January 2011 to 31 August 2012 and compared with that of all remaining sites of isolation after the removal of duplicate data (Table 2). The isolation rate in blood cultures was particularly high for S. hominis (396/698 [56.7%]) and S. pettenkoferi (30/54 [55.6%]). In contrast, some species, S. intermedius (0), S. caprae (1/56 [1.8%]), S. simulans (2/78 [2.6%]), S. saprophyticus (3/80 [3.8%]), and S. lugdunensis (17/205 [8.3%]), were rarely isolated in blood cultures.
Antimicrobial susceptibility testing.
Table 3 shows the results of antimicrobial susceptibility testing performed during the study period. S. haemolyticus displayed the highest rate of oxacillin resistance, at 87% among the 661 isolates tested; S. epidermidis was also resistant, at a high rate of 71% among 3,640 isolates tested, while S. hominis was 54.1% resistant among 525 isolates tested. None of the S. schleiferi isolates tested (n = 19) displayed oxacillin resistance, while resistance remained low for S. lugdunensis, (2% [n = 147]), S. simulans (2% [n = 48]), and S. caprae (5.7% [n = 35]). Resistance remained intermediary for S. warneri (23% among 204 strains), S. saprophyticus (32.5% among 77 strains), S. cohnii (36% among 25 strains), S. capitis (37% among 392 strains), and S. pettenkoferi (43.2% among 44 strains).
TABLE 3.
Oxacillin resistance and microbiological relevance of clinical strains of CoNS isolated in a clinical laboratory
| CoNS studied | Data for strains isolated during the 2007-2008 and 2011-2012 study periods |
|||||
|---|---|---|---|---|---|---|
| Oxacillin resistance data |
Microbiological relevance data |
|||||
| No. of antibiograms performed | No. (%) | No. of strains | No. (%) considered: |
|||
| Relevant | Doubtful | Not relevant | ||||
| S. epidermidis | 3,640 | 2,584 (71) | 50 | 7 (14) | 5 (10) | 38 (76) |
| S. hominis | 525 | 284 (54.1) | 50 | 11 (22) | 0 | 39 (78) |
| S. haemolyticus | 661 | 575 (87) | 50 | 11 (22) | 11 (22) | 28 (56) |
| S. capitis | 392 | 145 (37) | 50 | 9 (18) | 6 (12) | 35 (70) |
| S. warneri | 204 | 47 (23) | 50 | 9 (18) | 6 (12) | 35 (70) |
| S. lugdunensis | 147 | 3 (2) | 50 | 20 (40) | 13 (26) | 17 (34) |
| S. saprophyticus | 77 | 25 (32.5) | ||||
| S. simulans | 48 | 1 (2.1) | 50 | 11 (22) | 3 (6) | 36 (72) |
| S. pettenkoferi | 44 | 19 (43.2) | 50 | 1 (2) | 4 (8) | 45 (90) |
| S. caprae | 35 | 2 (5.7) | 50 | 11 (22) | 5 (10) | 34 (68) |
| S. cohnii | 25 | 9 (36) | 45 | 4 (9) | 7 (16) | 34 (76) |
| S. schleiferi | 19 | 0 | 42 | 11 (26) | 5 (12) | 26 (62) |
| S. pasteuri | 15 | 1 (6.7) | ||||
| S. intermedius | 15 | 2 (13.3) | ||||
| Total no. | 5,847 | 3,696 | 526 | |||
Microbiological relevance.
The putative pathogenic role of CoNS was determined as detailed above, consecutively selecting 50 isolates for each species while taking into consideration the microbiological criteria of significance, as detailed in Table 1 (17). The results displayed various rates of relevance. S. pettenkoferi was responsible for a putative clinical infectious process in only 2% of the case files studied, S. cohnii in 10%, and S. epidermidis in 14%. Conversely, S. lugdunensis displayed the highest rate of pathogenicity: 40% of the 50 strains selected were considered significant, based on microbiological criteria. Statistical analysis revealed that the repartition between relevant and nonrelevant groups was statistically significant, with a P value of <0.001.
DISCUSSION
This study assessed the implementation of mass spectrometry in a microbiological laboratory for the identification of CoNS. The results show that using MALDI-TOF MS in a hospital laboratory increased the identification of CoNS to approximately 85%, whereas only 14% of these staphylococci were identified to the species level prior to MALDI-TOF MS being available. This previous lack of identification was compounded by the high cost and time required for phenotypic tests, as well as the usual poor clinical significance of a majority of CoNS isolated from clinical specimens. Conversely, mass spectrometry is notably able to more accurately carry out S. aureus identification than when using slide or tube coagulase tests, and it is faster and cheaper than any current automated system using phenotypic tests (20, 21).
Some studies based on phenotypic identifications previously assessed the clinical significance and species distribution of CoNS in various clinical specimens. Kleeman et al. (22) studied the species distribution of 499 CoNS identified by phenotypic methods and found results similar to those reported here with regard to the isolation frequency for each species. S. epidermidis constituted 64.5% of all CoNS identified (versus 62.7% in our study), followed by S. haemolyticus (13.4% versus 11.6% in our study), S. hominis (7.4% versus 8.3% in our study), and S. warneri (4.0% versus 3.3% in our study). In our study, based on a greater number of isolates, we also found other species not identified in the Kleeman et al. (22) study, such as S. pettenkoferi, S. schleiferi, S. pasteuri, S. intermedius-S. pseudintermedius, and S. sciuri.
Interestingly, about 40% of the S. haemolyticus isolates in our study came from respiratory specimens, while only 4% of the S. hominis isolates did. Likewise, S. haemolyticus was found in urine specimens in 11% of the cases and S. hominis in <0.1%. Therefore, it could be postulated that S. haemolyticus is a normal inhabitant of mucosal membranes, such as those of the mouth or urinary tract, whereas S. hominis is a normal inhabitant of the skin.
Statistical analysis of staphylococcal distribution between periods 1 and 2 revealed that MALDI-TOF MS allowed a significantly higher rate of identification to be determined for S. aureus, S. haemolyticus, S. capitis, S. saprophyticus, S. cohnii, S. pasteuri, S. xylosus, and S. saccharolyticus. In addition, the use of MALDI-TOF MS revealed the further occurrence of certain species not well known until recently, such as S. pettenkoferi, S. condimenti, and S. piscifermentans. These species are not included in commercially available identification systems based on phenotypic features, and false identification might often occur with these systems (see Table S1 in the supplemental material) (23–25). A total of 54 S. pettenkoferi strains were identified during the study period, of which more than half originated from blood cultures. Of note, this species had never been isolated in our laboratory prior to the use of MALDI-TOF MS. From a clinical standpoint, it should be emphasized that this pathogen was described in several pathological instances, especially in immunocompromised patients, bloodstream infections, osteomyelitis, and wound infections (24–26). Its identification with the use of MALDI-TOF MS might therefore have valuable clinical consequences.
Based on phenotypic identifications, Tan, Ng, and Ng (27) also revealed some differences in clinical significance between 175 various CoNS. The authors notably found that 91% of the 53 strains of S. lugdunensis studied appeared to be clinically significant. They also found S. haemolyticus and S. capitis to be less virulent than S. epidermidis and S. warneri, which was not the case in our study. This may be explained by the smaller number of strains examined in that study.
Since its characterization in 1988 by Freney et al. (28), S. lugdunensis has been considered a potential emerging pathogen. Based on a retrospective study, Elamin et al. (29) showed that implementation of MALDI-TOF MS in 2010 for routine diagnostic use allowed the identification of 559 S. lugdunensis strains among 20,806 CoNS during a 24-month period, whereas 31 isolates were identified in the preceding 24 months using conventional methods (29). They also performed a clinical analysis for 478 individual isolates of S. lugdunensis and found no infectious diagnosis for 80 strains (16.7%), wound infection for 124 strains (25.9%), 99 abscesses (20.7%), and 12 significant bacteremias (2.5%), with 2 endocarditis cases (0.4%). In our study, S. lugdunensis was found to be microbiologically relevant in 40% of the cases, in contrast to the remaining CoNS, whose microbiological relevance was never >26%. The 50 S. lugdunensis isolates selected here for microbiological analysis originated from various independent sampling sites: superficial specimens (n = 20), deep-tissue specimens (n = 15), bone/synovial fluid/prosthetic joints (n = 10), intravascular catheters (n = 2), urine (n = 1), heart valve (n = 1), and blood culture (n = 1). Such associations of S. lugdunensis with severe infections were previously reported in various patient cases and series, including specimens from various origins, i.e. endocarditis, skin and soft tissue infections, arthritis, prosthetic joint infections, sepsis without identified gateway, pseudo-toxic shock syndrome, necrotizing sinusitis, and brain abscesses (30–35). S. lugdunensis is the second cause of endocarditis due to CoNS, with a poor prognosis in 40% of cases (36). Putative virulence factors have been described in S. lugdunensis that may explain its pathogenicity, including a fibrinogen binding protein, a von Willebrand factor binding protein, and possibly a delta-like hemolysin and a beta-hemolysin, although these hemolysins have not been correlated with any clinical type or severity of infection (37–39). The fibrinogen binding protein, also called the clumping factor, yields a positive result in slide coagulase and rapid latex agglutination tests, leading to its misidentification as S. aureus.
Interestingly, in our study, S. schleiferi was the second most important CoNS for its pathogenicity. This pathogen expresses a clumping factor, as does S. lugdunensis, and it also displays beta-hemolysin activity that may be responsible for its erroneous identification as S. aureus in the routine laboratory setting (37, 40). Clinical reports that describe infections with this pathogen remain scarce, although they appear unusual in their presentation and severity for a CoNS-mediated infection (40, 41). The production of a fibrinogen affinity factor might be involved in colonization of the vascular endothelium, one of the primary events in the pathogenesis of infective endocarditis, and its implication has been reported in experimental models of endocarditis (42).
Our study also reveals that >50% of the S. hominis and S. pettenkoferi isolates were recovered from blood cultures. There is no clear explanation for this observation. Previous studies reporting the diversity of CoNS in blood cultures remain sparse and mostly have described the presence of S. epidermidis and S. haemolyticus, likely because they are the two most frequently isolated CoNS in microbiology laboratories (43–46). The rate of isolation in blood cultures was not correlated with pathogenicity in our study, since S. hominis and S. pettenkoferi displayed a low rate of pathogenicity. Moreover, most of these isolates came from a single positive blood culture; therefore, this could be considered a probable contaminant.
The antimicrobial susceptibility testing revealed a global methicillin sensitivity of 66%, which is similar to the 70% rate reported in the literature (1, 4). Nevertheless, it is noteworthy to compare the rate of methicillin resistance for each strain to its pathogenicity. The present results show that some of the rates are inversely related. S. lugdunensis and S. schleiferi, CoNS species that have the highest rates of pathogenicity, at 40% and 26%, respectively, also remained methicillin sensitive in 98% and 100% of the cases examined, respectively. Conversely, S. haemolyticus and S. epidermidis were methicillin resistant in 87% and 71% of cases, respectively, but displayed low pathogenicity scores of about 22% and 14%, respectively. The impact of S. aureus methicillin resistance on clinical virulence is still a matter of debate. Pozzi et al. (47) recently published an in vitro study showing that methicillin resistance might attenuate S. aureus virulence (through repression of agr transcriptional activity and reduced protease production) and biofilm production in device-associated infections. Such observations have not been made in CoNS, although our data potentially support this hypothesis.
Several limitations of the present study should be acknowledged. Regarding the method described for phenotypic identification, it is well known that free coagulase is present in highly pathogenic species of staphylococci from animals (S. intermedius from dogs, S. delphini from dolphins, S. hyicus from pigs, and S. lutrae from otters), and S. lugdunensis, S. schleiferi, and S. saprophyticus can give positive results when tested with the slide agglutination test (48). Therefore, non-aureus Staphylococcus strains might have been misidentified as S. aureus. It also has to be noted that some deficient S. aureus strains lack both free coagulase and clumping factor, and some of these strains might have been misidentified in our study as coagulase-negative staphylococci. However, this bias should be very limited in our study, as both the free coagulase and slide agglutination tests were routinely performed for all isolates, and confirmation with Vitek 2 was systematically performed for deep-tissue samples or when discrepant results were obtained with tube and slide tests.
Another limitation in our study is that we selected for the MALDI-TOF analysis all the CoNS identified to the species level, meaning those with an identification score of ≥2. However, during this study period, it appears that 8% of the staphylococci tested with MALDI-TOF did not reach this threshold, were identified to genus level only, and were omitted from the analysis. Regarding the microbiological significance analysis, the main limitation is the methodology used to determine pathogenicity. Only microbiological criteria were examined and not clinical criteria. It could be postulated that while the microbiological criteria of significance were fulfilled according to EMCM guidelines in this study, it does not necessarily imply that the CoNS were involved in an ongoing infectious process. Nevertheless, the findings described here strongly correlate with previous data from the literature and may represent the first step toward a larger clinical/microbiological prospective analysis to confirm the role of certain CoNS species identified through the use of MALDI-TOF MS. This is especially true for S. lugdunensis, which appears to be one of the most pathogenic of the CoNS species. Another limitation regarding identification rate has to be pointed out. We compared the identification rates from periods 1 and 2, but in addition to using MALDI-TOF MS, more isolates were actually worked out in period 2 (1,104 CoNS in period 1 versus 8,388 in period 2). This difference might account for some of the differences in the ID rates between periods 1 and 2. Even if statistical analysis reveals significantly different ID rates, these data have to be read into carefully and confirmed.
In conclusion, the requirement of an accurate and cost-effective method, such as MALDI-TOF MS, to identify even rare CoNS cannot be overlooked at a time when the clinical significance of some CoNS appears to be very heterogenous (49). Exhaustive identification of CoNS by MALDI-TOF MS as well as microbiological criteria allow few suspect staphylococcal species to be identified as having a significant pathogenicity close to that of S. aureus, at least for S. lugdunensis. The heterogeneity of S. lugdunensis isolates is under investigation through a clinical and prospective study, as well as the possibility of a specific virulence content for the species.
Supplementary Material
ACKNOWLEDGMENTS
This work was made possible because of data available at the Plateau Technique de Microbiologie of the Strasbourg University Hospital. We thank P. Pothier for English editing assistance.
Footnotes
Supplemental material for this article may be found at http://dx.doi.org/10.1128/JCM.00177-15.
REFERENCES
- 1.Rogers KL, Fey PD, Rupp ME. 2009. Coagulase-negative staphylococcal infections. Infect Dis Clin North Am 23:73–98. doi: 10.1016/j.idc.2008.10.001. [DOI] [PubMed] [Google Scholar]
- 2.Piette A, Verschraegen G. 2009. Role of coagulase-negative staphylococci in human disease. Vet Microbiol 134:45–54. doi: 10.1016/j.vetmic.2008.09.009. [DOI] [PubMed] [Google Scholar]
- 3.von Eiff C, Peters G, Heilmann C. 2002. Pathogenesis of infections due to coagulase-negative staphylococci. Lancet Infect Dis 2:677–685. doi: 10.1016/S1473-3099(02)00438-3. [DOI] [PubMed] [Google Scholar]
- 4.Chu VH, Woods CW, Miro JM, Hoen B, Cabell CH, Pappas PA, Federspiel J, Athan E, Stryjewski ME, Nacinovich F, Marco F, Levine DP, Elliott TS, Fortes CQ, Tornos P, Gordon DL, Utili R, Delahaye F, Corey GR, Fowler VG Jr, International Collaboration on Endocarditis-Prospective Cohort Study Group. 2008. Emergence of coagulase-negative staphylococci as a cause of native valve endocarditis. Clin Infect Dis 46:232–242. doi: 10.1086/524666. [DOI] [PubMed] [Google Scholar]
- 5.Zadoks RN, Watts JL. 2009. Species identification of coagulase-negative staphylococci: genotyping is superior to phenotyping. Vet Microbiol 134:20–28. doi: 10.1016/j.vetmic.2008.09.012. [DOI] [PubMed] [Google Scholar]
- 6.Loonen AJM, Jansz AR, Bergland JNB, Valkenburg M, Wolffs PFG, van den Brule AJC. 2012. Comparative study using phenotypic, genotypic, and proteomics methods for identification of coagulase-negative staphylococci. J Clin Microbiol 50:1437–1439. doi: 10.1128/JCM.06746-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Patel R. 2013. Matrix-assisted laser desorption ionization-time of flight mass spectrometry in clinical microbiology. Clin Infect Dis 57:564–572. doi: 10.1093/cid/cit247. [DOI] [PubMed] [Google Scholar]
- 8.Patel R. 2013. MALDI-TOF mass spectrometry: transformative proteomics for clinical microbiology. Clin Chem 59:340–342. doi: 10.1373/clinchem.2012.183558. [DOI] [PubMed] [Google Scholar]
- 9.Dubois D, Leyssene D, Chacornac JP, Kostrzewa M, Schmit PO, Talon R, Bonnet R, Delmas J. 2010. Identification of a variety of Staphylococcus species by matrix-assisted laser desorption ionization–time of flight mass spectrometry. J Clin Microbiol 48:941–945. doi: 10.1128/JCM.00413-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Otto M. 2009. Staphylococcus epidermidis–the “accidental” pathogen. Nat Rev Microbiol 7:555–567. doi: 10.1038/nrmicro2182. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Otto M. 2011. Molecular basis of Staphylococcus epidermidis infections. Semin Immunopathol 34:201–214. doi: 10.1007/s00281-011-0296-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Frank KL, Del Pozo JL, Patel R. 2008. From clinical microbiology to infection pathogenesis: how daring to be different works for Staphylococcus lugdunensis. Clin Microbiol Rev 21:111–133. doi: 10.1128/CMR.00036-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Thibodeau E, Boucher H, Denofrio D, Pham DT, Snydman D. 2012. First report of a left ventricular assist device infection caused by Staphylococcus schleiferi subspecies coagulans: a coagulase-positive organism. Diagn Microbiol Infect Dis 74:68–69. doi: 10.1016/j.diagmicrobio.2012.05.027. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Sociéte Française de Microbiologie. 2007. Référentiel en microbiologie médicale, Sociéte Française de Microbiologie, Paris, France. [Google Scholar]
- 15.Société Française de Microbiologie. 2010. Référentiel en microbiologie médicale, 4th ed, Sociéte Française de Microbiologie, Paris, France. [Google Scholar]
- 16.Reference deleted.
- 17.European Society of Clinical Microbiology and Infectious Diseases. 2012. European manual of clinical microbiology, 1st ed ESCMID-SFM, Paris, France. [Google Scholar]
- 18.Cleri DJ, Corrado ML, Seligman SJ. 1980. Quantitative culture of intravenous catheters and other intravascular inserts. J Infect Dis 141:781–786. doi: 10.1093/infdis/141.6.781. [DOI] [PubMed] [Google Scholar]
- 19.Brun-Buisson C, Abrouk F, Legrand P, Huet Y, Larabi S, Rapin M. 1987. Diagnosis of central venous catheter-related sepsis. Critical level of quantitative tip cultures. Arch Intern Med 147:873–877. doi: 10.1001/archinte.1987.00370050069012. [DOI] [PubMed] [Google Scholar]
- 20.Rajakaruna L, Hallas G, Molenaar L, Dare D, Sutton H, Encheva V, Culak R, Innes I, Ball G, Sefton AM, Eydmann M, Kearns AM, Shah HN. 2009. High throughput identification of clinical isolates of Staphylococcus aureus using MALDI-TOF-MS of intact cells. Infect Genet Evol 9:507–513. doi: 10.1016/j.meegid.2009.01.012. [DOI] [PubMed] [Google Scholar]
- 21.Seng P, Rolain J-M, Fournier PE, La Scola B, Drancourt M, Raoult D. 2010. MALDI-TOF-mass spectrometry applications in clinical microbiology. Future Microbiol 5:1733–1754. doi: 10.2217/fmb.10.127. [DOI] [PubMed] [Google Scholar]
- 22.Kleeman KT, Bannerman TL, Kloos WE. 1993. Species distribution of coagulase-negative staphylococcal isolates at a community hospital and implications for selection of staphylococcal identification procedures. J Clin Microbiol 31:1318–1321. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Trülzsch K, Rinder H, Trcek J, Bader L, Wilhelm U, Heesemann J. 2002. “Staphylococcus pettenkoferi,” a novel staphylococcal species isolated from clinical specimens. Diagn Microbiol Infect Dis 43:175–182. doi: 10.1016/S0732-8893(02)00399-1. [DOI] [PubMed] [Google Scholar]
- 24.Loïez C, Wallet F, Pischedda P, Renaux E, Senneville E, Mehdi N, Courcol RJ. 2007. First case of osteomyelitis caused by “Staphylococcus pettenkoferi.” J Clin Microbiol 45:1069–1071. doi: 10.1128/JCM.02328-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Song SH, Park JS, Kwon HR, Kim SH, Kim HB, Chang HE, Park KU, Song J, Kim EC. 2009. Human bloodstream infection caused by Staphylococcus pettenkoferi. J Med Microbiol 58:270–272. doi: 10.1099/jmm.0.004697-0. [DOI] [PubMed] [Google Scholar]
- 26.Morfin-Otero R, Martínez-Vázquez MA, López D, Rodríguez-Noriega E, Garza-González E. 2012. Isolation of rare coagulase-negative isolates in immunocompromised patients: Staphylococcus gallinarum, Staphylococcus pettenkoferi and Staphylococcus pasteuri. Ann Clin Lab Sci 42:182–185. [PubMed] [Google Scholar]
- 27.Tan TY, Ng SY, Ng WX. 2006. Clinical significance of coagulase-negative staphylococci recovered from nonsterile sites. J Clin Microbiol 44:3413–3414. doi: 10.1128/JCM.00757-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Freney J, Brun Y, Bes M, Meugnier H, Grimont F, Grimont PAD, Nervi C, Fleurette J. 1988. Staphylococcus lugdunensis sp. nov. and Staphylococcus schleiferi sp. nov., two species from human clinical specimens. Int J Syst Bacteriol 38:168–172. [Google Scholar]
- 29.Elamin WF, Ball D, Millar M. 2015. Unbiased species-level identification of clinical isolates of coagulase-negative staphylococci: does it change the perspective on Staphylococcus lugdunensis? J Clin Microbiol 53:292–294. doi: 10.1128/JCM.02932-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Liu P-Y, Huang Y-F, Tang C-W, Chen Y-Y, Hsieh K-S, Ger L-P, Chen Y-S, Liu Y-C. 2010. Staphylococcus lugdunensis infective endocarditis: a literature review and analysis of risk factors. J Microbiol Immunol Infect 43:478–484. doi: 10.1016/S1684-1182(10)60074-6. [DOI] [PubMed] [Google Scholar]
- 31.Li YM, Blaskiewicz DJ, Hall WA. 2013. Shunt-related intracranial abscess caused by Staphylococcus lugdunensis in a hydranencephalic patient. World Neurosurg 80:e387–389. doi: 10.1016/j.wneu.2013.01.046. [DOI] [PubMed] [Google Scholar]
- 32.Matthews PC, Lazarus R, Protheroe A, Milford C, Bowler ICJW. 2011. Acute necrotizing sinusitis caused by Staphylococcus lugdunensis. J Clin Microbiol 49:2740–2742. doi: 10.1128/JCM.00722-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Shah NB, Osmon DR, Fadel H, Patel R, Kohner PC, Steckelberg JM, Mabry T, Berbari EF. 2010. Laboratory and clinical characteristics of Staphylococcus lugdunensis prosthetic joint infections. J Clin Microbiol 48:1600–1603. doi: 10.1128/JCM.01769-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Hung T, Zaghi S, Yousefzadeh J, Leibowitz M. 2012. Necrotizing fasciitis associated with Staphylococcus lugdunensis. Case Rep Infect Dis 2012:453685. doi: 10.1155/2012/453685. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Arias M, Tena D, Apellániz M, Asensio MP, Caballero P, Hernández C, Tejedor F, Bisquert J. 2010. Skin and soft tissue infections caused by Staphylococcus lugdunensis: report of 20 cases. Scand J Infect Dis 42:879–884. doi: 10.3109/00365548.2010.509332. [DOI] [PubMed] [Google Scholar]
- 36.Petti CA, Simmon KE, Miro JM, Hoen B, Marco F, Chu VH, Athan E, Bukovski S, Bouza E, Bradley S, Fowler VG, Giannitsioti E, Gordon D, Reinbott P, Korman T, Lang S, Garcia-de la-Maria C, Raglio A, Morris AJ, Plesiat P, Ryan S, Doco-Lecompte T, Tripodi F, Utili R, Wray D, Federspiel JJ, Boisson K, Reller LB, Murdoch DR, Woods CW, International Collaboration on Endocarditis–Microbiology Investigators. 2008. Genotypic diversity of coagulase-negative staphylococci causing endocarditis: a global perspective. J Clin Microbiol 46:1780–1784. doi: 10.1128/JCM.02405-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Hébert GA. 1990. Hemolysins and other characteristics that help differentiate and biotype Staphylococcus lugdunensis and Staphylococcus schleiferi. J Clin Microbiol 28:2425–2431. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Mitchell J, Tristan A, Foster TJ. 2004. Characterization of the fibrinogen-binding surface protein Fbl of Staphylococcus lugdunensis. Microbiology 150:3831–3841. doi: 10.1099/mic.0.27337-0. [DOI] [PubMed] [Google Scholar]
- 39.Szabados F, Nowotny Y, Marlinghaus L, Korte M, Neumann S, Kaase M, Gatermann SG. 2011. Occurrence of genes of putative fibrinogen binding proteins and hemolysins, as well as of their phenotypic correlates in isolates of S. lugdunensis of different origins. BMC Res Notes 4:113. doi: 10.1186/1756-0500-4-113. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Hernández JL, Calvo J, Sota R, Agüero J, Garcia-Palomo JD, Farinas MC. 2001. Clinical and microbiological characteristics of 28 patients with Staphylococcus schleiferi infection. Eur J Clin Microbiol Infect Dis 20:153–158. doi: 10.1007/PL00011251. [DOI] [PubMed] [Google Scholar]
- 41.Kumar D, Cawley JJ, Irizarry-Alvarado JM, Alvarez A, Alvarez S. 2007. Case of Staphylococcus schleiferi subspecies coagulans endocarditis and metastatic infection in an immune compromised host. Transpl Infect Dis 9:336–338. doi: 10.1111/j.1399-3062.2007.00222.x. [DOI] [PubMed] [Google Scholar]
- 42.Moreillon P, Que Y-A. 2004. Infective endocarditis. Lancet 363:139–149. doi: 10.1016/S0140-6736(03)15266-X. [DOI] [PubMed] [Google Scholar]
- 43.García-Vázquez E, Fernández-Rufete A, Hernández-Torres A, Canteras M, Ruiz J, Gómez J. 2013. When is coagulase-negative Staphylococcus bacteraemia clinically significant? Scand J Infect Dis 45:664–671. doi: 10.3109/00365548.2013.797599. [DOI] [PubMed] [Google Scholar]
- 44.Abdul Rahman Z, Hamzah SH, Hassan SA, Osman S, Md Noor SS. 2013. The significance of coagulase-negative staphylococci bacteremia in a low resource setting. J Infect Dev Ctries 7:448–452. doi: 10.3855/jidc.25345. [DOI] [PubMed] [Google Scholar]
- 45.Elzi L, Babouee B, Vögeli N, Laffer R, Dangel M, Frei R, Battegay M, Widmer AF. 2012. How to discriminate contamination from bloodstream infection due to coagulase-negative staphylococci: a prospective study with 654 patients. Clin Microbiol Infect 18:E355–E361. doi: 10.1111/j.1469-0691.2012.03964.x. [DOI] [PubMed] [Google Scholar]
- 46.Rahkonen M, Luttinen S, Koskela M, Hautala T. 2012. True bacteremias caused by coagulase negative Staphylococcus are difficult to distinguish from blood culture contaminants. Eur J Clin Microbiol Infect Dis 31:2639–2644. doi: 10.1007/s10096-012-1607-9. [DOI] [PubMed] [Google Scholar]
- 47.Pozzi C, Waters EM, Rudkin JK, Schaeffer CR, Lohan AJ, Tong P, Loftus BJ, Pier GB, Fey PD, Massey RC, O'Gara JP. 2012. Methicillin resistance alters the biofilm phenotype and attenuates virulence in Staphylococcus aureus device-associated infections. PLoS Pathog 8:e1002626. doi: 10.1371/journal.ppat.1002626. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Garcia LS, Isenberg HD. 2007. Clinical microbiology procedures handbook, 3rd ed ASM Press, Washington, DC. [Google Scholar]
- 49.Seng P, Drancourt M, Gouriet F, Scola BL, Fournier P-E, Rolain JM, Raoult D. 2009. Ongoing revolution in bacteriology: routine identification of bacteria by matrix-assisted laser desorption ionization time-of-flight mass spectrometry. Clin Infect Dis 49:543–551. doi: 10.1086/600885. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.