Introduction

Ralstonia pickettii, as a Gram-negative aerobic bacillus, was first isolated and ascribed to Burkholderia cepacian in 1973. Due to its phenotypic characteristics, cellular lipid and fatty acid analysis, rRNA–DNA hybridization and phylogenetic analysis of 16s rDNA nucleotide sequences, the genus was separated from Burkholderia spp and was ascribed to Ralstonia Spp in 1995 [1]. The most common clinically pathogenic pathogens in that species are Ralstonia pickettii (R. pickettii), Ralstonia mannitolilytica (R. mannitolilytica) and Ralstonia insidiosa (R. insidiosa). Ralstonia pickettii has been isolated from a wide variety of clinical specimens including blood, urine and cerebrospinal fluid [2]. The organism may be a more widespread pathogen and the types of infections are more invasive and severe than previously thought [3,4,5,6].

The existing commercial biochemical identification systems on the market, e.g. API @20 NE test system (BioMe ´rieux, Marcy l’Etoile, France) and Rap ID NE, do not always give the most dependable identification. Ralstonia pickettii has been shown to give variable results using the standard biochemical test kits, e.g. API 20NE test system [7]. Due to the similarity between Ralstonia pickettii and the B. cepacia complex, it is also thought that many cases of the B. cepacia complex may have been misidentified and are in fact Ralstonia pickettii. Ralstonia pickettii has been also associated with nosocomial outbreaks caused by contaminated solutions used for patient care and with pseudo epidemics caused by contaminated solutions in the diagnostic laboratory [8,9,10,11,12]. In addition, the treatment of Ralstonia pickettii infections is often challenging as this bacterium has been reported as being intrinsically resistant to many antimicrobial agents [13], which could be due to the presence of mobile genetic elements [14]. There is limited information on the surveillance and monitoring of the antibiotic resistance of these bacteria.

In this study we initially established a program for MLST to facilitate origin-tracing of infections and study of population structures and explored a sensitive way to identify Ralstonia pickettii. Meanwhile, we reported the antimicrobial susceptibility of Ralstonia pickettii isolates and analysed the clinical symptoms of Ralstonia pickettii infected patients, to provide reference for the treatment of clinical infection patients.

Materials and methods

Strains

48 non-repetitive Ralstonia pickettii strains were isolated from patients hospitalized from January 2008 to December 2013, which were identified by VITEK 2 system (BioMérieux, Marcy l’Étoile, France).

Bacteria identification

Commercial identification system

Strains were re-identifid according to the instructions of VITEK 2 system (BioMérieux, Marcy l’Étoile, France).

PCR experiment

Specific primer 1 for Ralstonia pickettii was developed in a previous study [15], while primer 2 was designed for R. mannitolilytica and primer 3 was designed for R. insidiosa (Table 1) [16]. Amplification was conducted in a programmable thermal cycler (Idaho Technologies Inc., Salt Lake City, Utah). After an initial 5-min denaturation at 95 °C, 30 amplification cycles were completed, each consisting of 40 s at 94 °C, 40 s at 55/57°C (the annealing temperatures used were 55 °C for the identification of R. insidiosa and Ralstonia pickettii while 57 °C for the identification of R. mannitolilytica), and 1 min at 72 °C, followed by an additional final extension of 5 min at 72 °C. Negative PCR controls, containing all reaction mixture components except template DNA, were included for every experiment.

Table 1 Primers for the identification of Ralstonia Spp. by PCR
Full size table

16 S rDNA

The consensus primers were used to perform 16 S rDNA experiment (Table 2) [17]. The amplification was carried out in a thermocycler (Idaho Technologies Inc., Salt Lake City, Utah). After an initial 5-min denaturation at 95 °C, 30 amplification cycles were completed, each consisting of 40 s at 94 °C, 40 s at 50 °C, and 1 min at 72 °C, followed by an additional final extension of 5 min at 72 °C. Perform bi-directional sequencing on amplification products and splice the determined sequence through EditSeq (DNA Star Inc.). Basic local alignment search tool (BLAST) on the spliced sequence was performed in NCBI database (www.ncbi.nlm.nih.gov/BLAST).

Table 2 Consensus Primer of 16 S rDNA Experiment
Full size table

MLST sequence typing

6 relatively conserved gene locus were chosen to design and synthesize 6 pairs of amplification primers which are presented in Table 3. The amplification was carried out in a thermocycler (Idaho Technologies Inc., Salt Lake City, Utah). After an initial 5-min denaturation at 95 °C, 30 amplification cycles were completed, each consisting of 40 s at 94 °C, 40 s at 56 °C, and 1 min at 72 °C, followed by an additional final extension of 5 min at 72 °C. For each of the selected genes, the sequences from all 42 isolates were compared and allele numbers were assigned to each unique sequence. The alleles present at the six loci define the allelic profile or sequence type (ST). Cluster analysis was performed using the unweighted pair group method with arithmetic mean (UPGMA) method to construct phylogenetic tree.

Table 3 Amplification primers of MLST gene locus
Full size table

Antibiotic susceptibility test

Disc diffusion tests and broth microdilution method were used to determine the antibiotic sensitivity of 34 Ralstonia pickettii strains identified by 16 S rDNA experiment. All tests were carried out according to Clinical and Laboratory Standards Institute (CLSI) standards (CLSI, 2015) [18]. As there were no CLSI susceptibility breakpoints available for Ralstonia pickettii, R. insidiosa or R. mannitolilytica, the antibiotic susceptibility results were interpreted using the CLSI criteria for Pseudomonas sp, BurkhoIderia cepacian and Acinetobacter spp (CLSI,2015). All results were found to be within recommended limits, demonstrating the validity Of the testing procedures used.

Clinical features and outcome of the infected patients

We performed a retrospective study on patients infected with Ralstonia pickettii. The patients’ data were collected by reviewing electronic medical records. We analyzed the clinical features, treatment and clinical outcome of the patients.

Results

The results of bacterial identification

All 48 strains tested were identified as Ralstonia pickettii by VITEK 2 system. There was a band of approximately 210 bp after the amplification of Ralstonia pickettii using primer (1) A band of 398 bp around was obtained after the amplification of R. mannitolilytica using primer 3, and two bands of about 210 bp and 403 bp were obtained after the amplification of R. insidiosa using primer 1 and (2) There was no visible band after the amplification of Burkholderia cepacian (See Fig. 1). 30 strains of Ralstonia pickettii, 1 strain of R. insidiosa, 8 strains of R. mannitolilytica were identified by the specific PCR, 9 strains had no target band. After 16 S rDNA identification, there were 34 strains of Ralstonia pickettii, 1 strain of R. insidiosa, 7 strains of R. mannitolilytica, 1 strain of Achromobacter xylosoxidans, 2 strains of Burkholderia cepacia and 3 strains of stenotrophomonas maltophilia.

Fig. 1
figure 1

The results of bacterial specific PCR assay. Notes: 1- Primer 1; 2-Primer 2; 3-Primer 3

Full size image

Evaluation of VITEK 2 system and PCR assays compared with the 16 S rDNA

Compared with the 16 S rDNA method, the sensitivity of VITEK 2 system was 100%, but the specificity was only 70.8%. As for the specific PCR method, the sensitivity and specificity were 85.3% and 85.7% for Ralstonia pickettii respectively,71.4% and 90.2% for R. mannitolilytica, both 100% for R. insidiosa (Table 4).

Table 4 Sensitivity and specificity of PCR assays for the identification of Ralstonia spp
Full size table

MLST sequence typing results

42 strains of Ralstonia Spp were analyzed and designated by MLST, resulting in 18 ST types ranging from ST1 to ST18. The most frequent ST type was ST9 with 22 strains which were considered as the advantage type. There are 2 strains were identified as ST2 and ST4, respectively. The result of genotyping by MLST of the 42 strains of Ralstonia is presented in Table 5. The result of typing by MLST is presented in Fig. 2. The 42 strains of Ralstonia were grouped into 2 groups (A and B) containing 18 genotypes with 8 genotypes in the group A and 10 in the group B respectively. The 8 genotypes in Group A contained 10 strains of Ralstonia, including 4 strains of Ralstonia pickettii and 6 strains of mannitolilytica. The 10 genotypes in group B contained 32 strains of Ralstonia, including 1 strain of R. mannitolilytica,1 train of R. insidiosa, and 30strains of Ralstonia pickettii.

Table 5 The result of genotyping by MLST of 42 strains of Ralstonia
Full size table
Fig. 2
figure 2

The result of the cluster analysis on 42 strains of Ralstonia by UPGMA method

Full size image

The results of antibiotic susceptibility test

Antibiotic sensitivity of Ralstonia pickettii by the disc diffusion tests

All of 48 strains were sensitive to Cefoperazone/sulbactam, Cefepime, Ciprofloxacin, Levofloxacin and Trimethoprim/Sulfamethoxazole. Followed by Piperacillin/tazobactam (94.1%), Minocycline (94.1%) and Ceftazidime (82.4%). However, all the strains were resistant to Aztreonam and Meropenem, while most of the strains were found to be less responsive to Cefoperazone (2.9%), Amikacin (14.7%%), Gentamicin (14.7%) and Polymyxin B (14.7%) (Table 6).

Table 6 The antibiotic sensitivity of Ralstonia pickettii by disc diffusion tests (%)
Full size table

Antibiotic sensitivity of Ralstonia pickettii by the broth microdilution method

All of 48 strains were sensitive to Cefuroxime, Ceftriaxone sodium and Trimethoprim/Sulfamethoxazole, followed by Ampicillin/Sulbactam (97.0%), Cefepime (97.0%), Levofloxacin (97.0%) and Cefuroxime axetil (93.9%). However, all the strains were resistant to Ampicillin, Aztreonam, Cefazolin, Cefotetan, Ceftazidime and Nitrofurantoin. while most of the strains were found to be less responsive to Imipenem (3.0%), Meropenem (3.0%) and Piperacillin (3.2%) (Table 7).

Table 7 The antibiotic sensitivity of Ralstonia pickettii by broth microdilution method (%)
Full size table

Clinical features, treatment and outcome of the infected patients

The average age of the 30 patients infected with Ralstonia pickettii was 88 years old, and 43.3% of the patients were above 90 years old. The most common primary diseases were Chronic Obstructive Pulmonary Disease and Coronary disease. The clinical characteristics were listed in Table 8. All the cases suffered from cough and sputum. Nearly 96% of the patients suffered from shortness of breath and pulmonary rales, followed by fever (76.7%) and chest pain (6.7%).

Table 8 The clinical manifestations and the positive rates of 30 patients
Full size table

As for laboratory examination results, only about half of patients had abnormal white blood cell (WBC) count results; however, around four-fifths had a higher neutrophil percentage (NEUT%) and an elevated C-reactive protein (CRP) concentration. The most common chest radiographic sign was bilateral pulmonary infiltration.

The antibiotics used empirically within 14 days before the positivity of Ralstonia pickttii culture were extensive. Nearly all the patients were treated with combination antibiotics. 73.3% of the patients used meropenem for the broad antibacterial spectrum, potent antibacterial properties and fewer adverse events, which characteristics were very important for elderly patients with pneumonia of unclear bacteria. 60% of the patients used cephalosporins, followed by quinolones(56.7%), which was thought to have more adverse reactions like QT prolongation. The results were shown in Table 9. After the results of bacteria identification and susceptibility tests returned, treatments were adjusted. Through clinical therapy, 73.3% of the patients obviously improved, while the mortality rate was 13.3% in thirty-days.

Table 9 Antibiotics used in patients within 14 days before the positivity of Ralstonia pickttii culture
Full size table

A two-stage MLST typing system was constructed based on16S rDNA

We have constructed a two-stage MLST typing system based on16S rDNA. Firstly, the strains were identified by 16 S rDNA, and then a typing of the strains within the species was performed to reach a high-resolution typing effect, thus achieving a quick origin-tracing of the infected strains inside the hospital. The current 16 S rDNA-MLST two-stage typing system had a relatively fine resolution (Figs. 3 and 4).

Fig. 3
figure 3

Typing with the species on Ralstonia pickettii by MLST

Full size image
Fig. 4
figure 4

Typing with the species on R. mannitolilytica by MLST

Full size image

Discussion

Ralstonia species are non-fermented Gram-negative bacillus which can cause a variety of infections and even outbreaks in hospitals [4, 19]. The Ralstonia pickettii, R. mannitolilytica and R. insidiosa have been reported to be the most common clinically pathogenic pathogens with mis-dentification in clinical laboratory [13]. In our study, only 34 of the 48 clinical strains identified to be Ralstonia pickettii with VITEK 2 system were confirmed by the 16 S rDNA method, with an accuracy of only 70.83%. This indicates that the identification results of Ralstonia pickettii by conventional methods are not wholly reliable and should be confirmed with molecular assays.

Tom Coenye et al. performed a PCR experiment to identify Ralstonia pickettii, R. mannitolilytica and R. insidiosa using designed primers Rp-F1/Rp-R1, Rm-F1/Rm-R1 and Rp-F1/R38-R1 [13, 16]. It showed that there were corresponding target bands after the amplification of Ralstonia pickettii and R. mannitolilytica with common primers. Due to the crossing between primers in the identification of Ralstonia pickettii and R. insidiosa, two target bands showed up in R. insidiosa. Our results were complied with that of Tom Coenye’s [15, 16]. Michael P Ryan’s [20] also got the same results. In addition, the 16 S rDNA method was also conducted as a standard method to evaluate the accuracy of the PCR assays. The sensitivity and specificity of the PCR assays about Ralstonia pickettii, R. mannitolilytica and R. insidiosa were 85.3% and 85.7%, 100% and 100%, 71.4% and 90.2%, respectively. In comparison with the previous studies, the sensitivity and specificity are not significantly increase. This is considered to correlate with the types of bacteria selected to evaluate the accuracy of the experiment. There were only 27 strains of Ralstonia pickettii and 34 strains of R. mannitolilytica out of those 152 strains selected for study by Tom Coenye et al [15]. The other 91 (59.9%) strains had relatively low species correlation with Ralstonia which were more easily and correctly to be identified. This study also showed that the 6 strains of non-Ralstonia were all identified correctly. Therefore, the PCR assays may be a reliable and simple method to identify Ralstonia pickettii, R. mannitolilytica and R. insidiosa.

Currently, there are fewer studies about the genotyping of Ralstonia [21]. Therefore, a standard MLST typing program of Ralstonia remains unestablished. Our study designed primers and determined the nucleotide sequence and performed an analysis of genotyping. The results showed that 42 strains of Ralstonia contained 18 STs and the most frequent ST was ST9 with 22 strains which are considered as the advantage type and the epidemic strain in our hospital. ST9 strains caused infections each year from 2008 to 2013. UPGMA results indicated that 42 strains of Ralstonia were grouped into 2 groups (A and B). Literature has reported identification difficulties of R. insidiosa and Ralstonia pickettii due to the close similarity [10]. As the cluster analysis shows, the R. insidiosa was in ST18, group B had a close genetic relationship with Ralstonia pickettii. But in this study, its ST code was 1-1-2-1-1-1 which was obviously different from those of the other strains in this study and consequently can be easily distinguished. However, the ST code of strain HXK003 and HXK042 were both 2-2-2-2-2, the identification result showed that they’re R. mannitolilytica and Ralstonia pickettii respectively. Our study failed to distinguish them. Due to the inconsistency between the identification result by 16 S rDNA and MLST results, it is considered to construct a two-stage MLST typing system based on 16 S rDNA. What’s more, because the strains of single source are always similar, additional gene locus are needed in the request for higher resolution, which should be further evaluated and screened.

Systematic antibiotic susceptibility data for Ralstonia Spp are scarce. Nevertheless, several studies and case reports suggested that most Ralstonia Spp are susceptible to commonly used antibiotics [22,23,24], but the result of the antibiotic susceptibility test using E-tests and broth microdilution method on 53 strains of Ralstonia pickettii by Ryan MP showed that the strains had high resistance to gentamicin and aztreonam, different degrees of resistance to ticarcillin/clavulanic acid and had sensitivity to antibiotics such as quinolones, tetracyclines, cephalosporins and sulfamido. Most of the strains were sensitive to Meropenem [22]. However, the results did not quite agree with our study. Almost all Ralstonia pickettii isolates were susceptible to cefepime, levofloxacin and Trimethoprim/Sulfamethoxazole and highly resistant to amikacin, gentamicin, aztreonam and meropenem both with the disc diffusion tests and broth microdilution method. However, poor correlation was found with piperacillin, piperacillin/tazobactam, imipenem and ciprofloxacin between MIC and disc diffusion results. Similar variation was found in the results of Ryan MP [22] and a comparable study that was carried out on Stenotrophomonas maltophilia [25, 26].

However, there was another interesting finding. As carbapenems, the susceptibilities of strains to imipenem and meropenem were strikingly different both with the disc diffusion tests and broth microdilution method. The isolates showed significantly high resistant to meropenem while much less resistant to imipenem. Especially for the disc diffusion tests, nearly half of the isolates were susceptible to imipenem, but none of the isolates were susceptible to meropenem, which might be attributed to the heavy use of meropenem in our hospital. That might indicate Ralstonia pickettii can acquire drug resistance in some way under the antibiotic pressure, which needs further studies.

In our study, we retrospectively analyzed the clinical data of 30 elderly patients infected with Ralstonia pickettii. According to our results, the most common features of patients were cough, sputum, shortness of breath, pulmonary rales, high blood NEUT%, CRP concentration and bilateral pulmonary infiltration. These findings were consistent with hospital acquired pneumonia (HAP) in elderly patients and did not have any specific characteristics. Empirical treatment for Ralstonia pickettii infection had not yet reached any consensus. For Ralstonia pickettii strains that were less responsive to multiple antibiotics in our study, the strategy of combination antibiotics was proven to be useful, but it might also give rise to the appearance of multidrug-resistant bacteria. Ralstonia pickettii is an opportunistic pathogen especially in the hospital setting. It is a waterborne microorganism that can survive in any kind of water source and tends to form and maintain biofilms. Most of the patients in this study were elderly and had a long hospital stay. Long-term medical interventions, including the use of broad-spectrum antibiotics, deep venous catheterization, dialysis and mechanical ventilation, increased the risk of infection with Ralstonia pickettii, especially in patients infected with multi-drug resistant bacteria. These factors may lead to infections and the difficulties of removing of Ralstonia pickettii. However, our results have some limitations. The incidence and drug-resistant of Ralstonia pickettii could vary among different countries, regions, or even hospitals, giving rise to significant bias. Moreover, the clinical data was collected retrospectively so that some details on patients’ symptoms could be lost. Therefore, further researches are still needed to discover more detailed features of Ralstonia pickettii infections.

The specific PCR experiment in this study was reliable and simple, therefore, it can be used as a method to identify the clinically pathogenic bacteria of Ralstonia and provide references for clinical diagnosis. This study tried to investigate the MLST typing on Ralstonia and preliminarily establish Ralstonia ‘s MLST system to provide references for its genotyping to facilitate Ralstonia ‘s origin-tracing of its infections, especially outbreaks inside the hospital, as well as the study on its population structure. There was no specific clinical manifestations for Ralstonia pickettii pneumonia, the antibiotics should be chosen in reason according to the result of the drug susceptibility test due to Ralstonia pickettii’s resistance to various antibiotics.