Rare non-fermenting bacteria are a serious epidemiological problem in hospital-acquired infection and cystic fibrosis (CF) because of the lack of knowledge on its existence and the restricted antimicrobial treatment options, which includes knowledge on Ochrobactrum anthropi. The genus Ochrobactrum are gram negative non-fermenting rods that at present comprises 17 species. The most isolated species in human clinical material are O. anthropi and O. intermedium. O. anthropi formerly was recognized among Achromobacter species with characteristic urease positive reaction. Ochrobactrum species are closely related to Brucella species. O. anthropi is oxidase-positive, indole-negative, motility test-positive, trypsin-positive, pyrrolidonyl aminopeptidase (PYR) positive and saccharolytic with rapid acidification of glucose and xylose. Can be distinguished from O. intermedius by colistin and netilmicin susceptibility¹.

Specifically regarding O. anthropi, a total of 335 studies are described in the database from the National Center for Biotechnology Information in several situations, such as: CF; sepsis; bacteremia due to the catheter contamination in parenteral nutrition; septicemia and pneumonia in preterm, which have been curiously observed in several isolated cases reports, fact associated with rarity of infection by O. anthropi.^2-7

Thus, we conducted a retrospective, cross-sectional, and descriptive study in a tertiary-level hospital, considering the diagnosis O. anthropi for different identification methods, namely:

(i) manual phenotypic method: (metabolism in carbohydrate) oxidation and fermentation (OF) of glucose [C.H₁₂O.], maltose OF [C_12.22.11], sucrose OF [C_12.22.11], lactose OF [C_12.22.11], xylose OF [C.H₁₀O.]; (Metabolism of amino acids) via Moeller decarboxylase of: lisin [C.H₁₄N.O.], ornithine [C.H₁₂N.O.], arginine [C.H₁₄N.O.], Moeller base control; Simmons citrate agar; aesculin hydrolysis [C_15.16O.]; Indole test; tube with broth of 6.5% NaCl; Mannitol broth [C.H₁₄O.]; tube of jelly; Imipenem disks [‎C_12.17N.O.S]; disk of Polimixin B [C_48.84.16.17S]; DNAse test; Pyrrolidonyl arylamidase test (PYR test); oxidase test; disks of ONPG (orto-nitrofenil-β-D-galactosidase, C_12.15NO.); motility test; urea test [H.N.O];

(ii) automated phenotypic method Vitek.2 (bioMeriéux, Jacarepaguá, Rio de Janeiro, Brazil) and Phoenix™ (Becton, Dickinson and Company, Loverson Circle, Sparks, USA);

(iii) polymerase chain reaction (PCR) (sense 5’ – GCA AGC TGG GTG TCG ATC TGG – 3’; antisense 5’– TTC TCG ACG ACA CCG GCC TTT A – 3’). For the PCR method, we used a standard strain by BCCM. LMG (ATCC49687) (Laboratorium voor Microbiologie, Universiteit Ghent, Ledeganckstraat, Ghent, Belgium) as positive control;

(iv) MALDI-TOF Vitek MS (MALDI-MS™) (bioMeriéux, Jacarepaguá, Rio de Janeiro, Brazil) and MALDI-TOF BioTyper™ (MALDI-BD™) (Becton, Dickinson and Company, Loverson Circle, Sparks, USA).

In this study, ten samples (seven of sputum culture in patients with CF and three of blood culture) previously identified as O. anthropi by Vitek.2, of high reliability level (of above 95% – very good) were subjected to different methods of bacterial identification. Compared with the result obtained with Phoenix™, there were discrepancies only for one sample, which was identified as Ralstonia pickettii.

In a comparative analysis, differently from Vitek.2, all samples were identified as Ochrobactrum sp by MALDI-MS™, with scores ranging from 84.1% to 99.9%. In MALDI-BD™, all analyses have had a score superior to two and only four bacteria have been identified as O. anthropi, three as Ochrobactrum tritici, one as Ochrobactrum intermedium and two as Ochrobactrum sp.. Briefly, there was a higher diversity of bacteria identified by MALDI-BD™, while MALDI-MS™ showed to be limited at genus level.

In biochemical tests, the identification of five samples such as O. anthropi,one as Burkholderia sp., one as Ralstonia insidiosa, and in three samples it was not possible to determine species and genus.

Finally, eight samples were evaluated by PCR, determining the final outcome of all samples as O. anthropi. Thus, besides the eight samples evaluated by PCR, Vitek.2, Phoenix™, MALDI-BD™, and biochemical tests identified eight, seven, six, and five, respectively, with the same result, namely O. anthropi. On the other hand, for MALDI-MS™, there was no sample identified at species level. Also, the major limitation from our work included the absence of 16S rDNA gene sequences use to study the bacterial phylogeny and taxonomy in our data. Data of the carried-out analysis are presented in table 1.

Table 1
Comparison between the results using different methods to identify Ochrobactrum anthropi in a reference center.

PCR, polymerase chain reaction; N/id, non-identified; O. anthropi, Ochrobactrum anthropi; R. pickettii, Ralstonia pickettii; R. insidiosa, Ralstonia insidiosa; O. tritici, Ochrobactrum tritici; O. intermedium; Ochrobactrum intermedium.

In the study, despite the limiting factor in having two samples in which it was not possible to obtain results using PCR, curiously we found the following results: (i) biochemical tests showed limitations to identify rare gram-negative non-fermenting such as O. anthropi, in addition to the identification of other species such as R. insidiosa and Burkholderia sp., as well as inconclusive results – which is possibly associated with results reading process, difficulty of interpretation and low prevalence of bacteria; (ii) Vitek.2 is the only method that showed the same result seen in PCR in all samples, followed by Phoenix™ (only one error); (iii) MALDI-MS™ failed at defining at species level, despite showing a high score in the analysis; (iv) MALDI-BD™ presented high diversity in identification, regardless of the small number of samples to be analyzed, showing to be a test of low specificity. Curiously, MALDI-MS™ and MALDI-BD™ might have methodological limitations, mainly associated with the available database, in the identification of rare bacteria.

Therefore, there is need for knowledge on rare bacteria and its identification, including the species O. anthropi, since we are still distant from a reality that favors the determining of species in a quick, cheap and effective way, which is of extreme importance in the health field.

REFERENCES

1. Vaneechoutte M, Nemec A, Kampfer P, et al. Acinetobacter, Chryseobacterium, Moraxella, and other Nonfermentative Gram-negative rods in: Jorgensen JH, Pfaller MA, Carroll KC, et al. 15^th Manual of Clinical Microbiology 2015, 2571p ASM Press Herndon VA USA, 813-37.

2. Ferreira ACM, Marson FAL, Cohen MA, et al. Hypertonic saline as a useful tool for sputum induction and pathogen detection in cystic fibrosis. Lung 2017; 195 (4): 431-9. doi: 10.1007/s00408-017-0008-3

3. Menezes FG, Abreu MG, Kawagoe JY, et al. Ochrobactrum anthropi bacteremia in a preterm infant with cystic fibrosis. Braz J Microbiol 2014; 45 (2): 559-61.

4. Ashraf F. A case of Ochrobactrum anthropi-induced septic shock and infective endocarditis. R I Med J (2013) 2016; 99 (7): 27-8.

5. Qasimyar H, Hoffman MA, Simonsen KA. Late-onset Ochrobactrum anthropi sepsis in a preterm neonate with congenital urinary tract abnormalities. J Perinatol 2014; 34 (6): 489-91. doi: 10.1038/jp.2014.31

6. Zakariya-Yousef I, Aller-García AI, Corzo-Delgado JE, et al. Catheter-associated bacteremia caused by Ochrobactrum anthropi in a patient on parenteral nutrition. Enferm Infecc Microbiol Clin 2014; 32 (8): 544-5. doi: 10.1016/j.eimc.2014.01.004

7. Kumar S, Kapoor S, Chadha S, et al. Ochrobactrum anthropi septicemia and pneumonia in a preterm, small for gestational age infant with multiple congenital anomalies. Indian J Pathol Microbiol 2013; 56 (3): 317-8. doi: 10.4103/0377-4929.120411