Value of molecular typing in the assessment of bacterial translocation during coagulase-negative staphylococcal bacteremia in preterm infants

One hundred fifty-seven preterm infants enrolled in the study were hospitalized between 2012-2014 at Amiens-Picardie University Hospital.Only 28 (17.8%) of these children who had experienced at least one episode of secondary Coagulase-negative Staphylococcal bacteremia with concomitant positive stool cultures were included in this study.The purpose of this study was to assess the rate of intestinal bacterial translocation associated with these infections. Blood cultures and stool cultures were performed in the context of this study. All isolates of Staphylococcus spp were examined by MALDI-TOF MS. Antibiotic susceptibility and genotyping were also performed. Sixteen resistance patterns were identified from blood and stool based on antibiotic susceptibility testing. Ten of the Coagulase-negative Staphylococcus strains isolated from blood samples exhibited R pattern e (35.7%) and eleven of the Coagulase-negative Staphylococcus strains isolated from stool samples exhibited R pattern e (39.2%). Blood culture results were concordant with stool culture results in 53.5% of cases and discordant in 46.5% of cases.Fifteen isolates exhibited three ERIC-2 (A, B, C) and three RAPD-PCR (D, E, F) patterns. ERIC-2 patterns comprised A ( S. epidermidis isolates); B ( S. haemolyticus isolates) and C ( unidentified Coagulase-negative Staphylococcus isolates). RAPD patterns consisted of D ( unidentified Coagulase-negative Staphylococcus isolates), E ( S. haemolyticus isolates), and F ( S. epidermidis isolates). kanamycin, gentamicin, tobramycin and netilmicin,and resistance to erythromycin, rifampin and ofloxacin.


Conclusion
Bacterial translocation from the intestinal tract was likely source of Coagulasenegative Staphylococcal bacteremia in hospitalized preterm infants.

Background
Bacterial translocation (BT) is defined as the passage of live bacteria, their products, or both, across the lamina propia to local mesenteric lymph nodes (MLN) and from there to extranodal sites [1][2][3]. The intestinal tract has multiple functions in the body apart from its primary function, the absorption of nutrients. It represents a barrier protecting the body from living microorganisms and antigens of the intestinal lumen.
Many physiological and pathological conditions, such as preterm infant, fasting, mesenteric ischemia-reperfusion, shock states, can alter intestinal functioning.
Alteration of the intestinal barrier often results in a systemic inflammatory process and,more rarely,potentially fatal multiple organ dysfunction syndrome (MODS).
Gastrointestinal tract (GIT) epithelial and immune cells play an essential role in initiation and resolution of the inflammatory process. Any adverse alteration of the intestinal barrier leads to disruption of this process, resulting in increased permeability of the intestinal barrier, promoting the passage of live bacteria, bacterial DNA or bacterial degradation products from the intestines to extraintestinal sites [4]. The route of BT has been examined in humans [1]. A study of human trauma patients [5] evaluated blood samples from portal vein (PV) catheters for evidence of translocating bacteria or endotoxins.The incidence of BT in humans undergoing emergency laparotomy has been reported to be 14 to 21% [6] and 15.4% [3].Active research has also been devoted to methods able to document the occurrence of BT.Numerous molecules have been evaluated as potential biomarkers for BT, including bacterial DNA, soluble CD14, lipopolysaccharide (LPS)/endotoxin, LPS binding protein (LBP), calprotectin [7].LPS and d-lactate have been identified as products of gut BT and systemic markers for increased gut permeability and BT [8]. BT has been proposed as the mechanism of bacteremia in clinical situations such as neutropenia in cancer patients [9] and hemorrhagic shock [10]. Translocation limited to MLN has also been described in patients with Crohn's disease, colorectal cancer [11], premature children and newborns [12].The incidence of BT and its relationship to sepsis and MODS in preterm newborns remain unclear. Coagulase-negative Staphylococcal (CoNS) bacteremia is common in neonates and is often associated with the presence of a catheter. The mechanism responsible for intestinal BT remains poorly elucidated. It would therefore be interesting to study the rate of BT responsible for CoNS bacteremia in a population of preterm infants.
Objectives of our study were to assess the rate of intestinal BT and its role in causing sepsis in clinical conditions, and to evaluate the correlation between CoNS isolated from blood cultures and stool cultures using molecular typing.

Study design and study population
During the study period, a total of 157 neonates (less than 28 days of life) born prematurely (<37 weeks of gestation-WG) were immediately hospitalized in the neonatal intensive care units (nICUs) and in the pediatric intensive care units (pICUs) of Amiens-Picardie University Hospital (APUH), Amiens (France) between 2012 and 2014. Twenty-eight (17.8%) of these infants had experienced at least one episode of CoNS bacteremia with a concomitant positive stool culture. Only these 28 children were included in the study. However, 129 (82.2%) children including full term infants (≥ 37 WG),preterm infants with positive stool culture without CoNS bacteremia, preterm infants with CoNS bacteremia without stool cuilture positive, and preterm infants corresponding to duplicate cases were all excluded in this study.

Data collection
All data were obtained from electronic medical records. Patient demographic characteristics, underlying conditions, and clinical and laboratory findings were collected.

Blood and stool samples
Blood was obtained for routine hematological, biochemical and bacteriological tests, including polymorphonuclear leukocyte and platelet counts and CRP. All blood cultures were performed for diagnostic purposes, and stool cultures were performed in the context of this study. Less than 0.5 mL of blood was inoculated into a Bactec Peds Plus F bottle and incubated in Bactec TM Becton Dickinson instrument (BD Diagnostic Systems, Spark, MD, USA). Subcultures of initial blood culture broth were seeded on sheep blood (5%) Columbia agar and mannitol salt agar (MSA) (Oxoid, France) were incubated aerobically at 37°C for 24 hours.Anal or rectal samples using sterile swabs were seeded on sheep blood (5%) Columbia agar and MSA and were incubated aerobically at 37°C for 48 hours.

Identification
Staphylococcus spp. were identified after examining all colonies. All isolates negative for mannitol and for bound coagulase (Pastorex Staph Plus-Bio-Rad, France) and positive for catalase and Gram staining were classified as CoNS.

MALDI-TOF-MS
All Staphylococcus strains isolated by routine tests were examined by Matrix-Assisted-Laser Desorption Ionization Time of Flight Mass Spectrometry (MALDI-TOF-MS) (Brucker Daltonics, Bremen, Germany) according to a previously described procedure [13,14].
In the presence of a negative blood culture, the corresponding stool cultures and blood cultures were discarded. Conversely, when a Staphylococcus sp. was isolated from the patient's blood culture, all morphologically distinct colonies detected by gross examination of stool cultures were identified by MALDI-TOF-MS.
Antibiotic susceptibility testing DNA isolation and molecular typing Total nucleic acid extraction was performed using the bioMérieux NucliSENS easy MAG platform (bioMérieux, Marcy l'Etoile, France) according to the manufacturer's instructions.ERIC-PCR was performed as previously described [16]. Extracted DNA (100 ng from each isolate) was amplified in a final volume of 50 µL of the ERI-C2 primer (5'-AAG-TAA-GTG-ACT-GGG-GTG-AGC-G-3') and 5 µL of CoralLoad PCR buffer.
PCR cycling consisted of 94°C for 7 min followed by 45 cycles of 94°C for 1 min, 45°C for 1 min and 72°C for 2 min, and 72°C for 7 min. RAPD-PCR was performed as previously described [17]. The primer used was, 5'-GGT-TGG-GTG-AGA-ATT-GCA-CG-3'. Amplification reactions were carried out with a final volume of 50 µL containing 25 µL of Top Taq Master MIX (QIAGEN, Les Ulis, France), 1 µL of RAPD1 primer, 2 µL of DNA as template, 5 µL of CoralLoad PCR buffer. The cycling conditions were 95°C for 3 min, followed by 35 cycles of 94°C for 1 min, 40°C for 1 min, and 72°C for 2 min, and 72°C for 10 min. After amplifications, PCR products were resolved by electrophoresis on 1.2% agarose gels at 90 V for 6 h, followed by ethidium bromide staining and were visualized under UV light. A photograph was also taken.

CoNS bacteremia
The following clinical, laboratory and treatment parameters listed in Table 7 were analyzed as potential RFs for BT in this study. These RFs were categorized in two groups according to molecular typing: group1 included patients in whom BT was indicated; group 2, comprised patients in whom BT was not indicated.

Statistical analysis
We calculated the relative risk (RR) of BT among preterm infants with and without documented BT. Data are expressed as means ± standard deviation (SD) for quantitative variables and frequencies for qualitative variables. We calculated the RR of BT with associated 95% confidence intervals (CIs) using Fisher's Exact test.
Comparisons between documented BT and undocumented BT groups were performed with the Wilcoxon-Mann-Whitney test for quantitative data and Fisher's Exact test for qualitative data. All tests of significance were two-sided and a p value of < 0.05 was considered to indicate statistical significance.

Patients characteristics
The characteristics of 28 preterm infants included in this study were as follows: gestational age (weeks):mean ± SD:29.  (Table 1) ;maternal age at birth (years) :mean ± SD :29.7 ± 6.2 ,median :29.5,range :19-45.  Antibiotics susceptibility of C o N S s t r a i n s The 28 CoNS strains isolated from both blood cultures and stool cultures were resistant to methicillin (cefoxitin-resistant strains) and kanamycin (Tables 3 and   5).One hundred percent of the 28 CoNS strains isolated from blood samples were resistant to penicillin, cefoxitin and kanamycin,and 96.4% of isolates were resistant to gentamicin,tobramycin and netilmicin.The resistance of these strains to other antibiotics tested are shown in Table 3.The distribution of resistance patterns of these isolates showed sixteen antimicrobial resistance patterns ( R patterns) a to p,and10 of these strains exhibited R pattern e [(35.7%) ( isolates 5-11,13-15)] (Table 4).
One hundred percent of the 28 CoNS strains isolated from stool samples were resistant to penicillin and cefoxitin, and 96.4% of isolates were resistant to kanamycin,gentamicin,tobramycin and netilmicin. Abbreviations: see Table 3  Table 5 Antimicrobial resistance patterns (R patterns) of the 28 strains isolated from blood samples Sixteen antimicrobial resistance patterns were observed, 10 of which exhibited R pattern e[(35.7%) (isolates 5-11, 13-15)].
One hundred percent of the 28 CoNS strains isolated from stool samples were resistant to penicillin, cefoxitin,and kanamycin,and 96.4% of isolates were resistant to gentamicin,tobramycin and netilmicin ( Table 5).The resistance of these isolates to other antibiotics tested are shown in Table 5.All these isolates were classified into sixteen R patterns a to p,and 11 of these strains exhibited R pattern e [(39.2%), (isolates 5-11,13-16)] (Table 6).
All isolates were classified into 16 antimicrobial resistance patterns designated R patterns (a to p), 11 of which exhibited R pattern e (39.2%).
The most common resistance patterns of the strains isolated combined heterogeneous resistance to methicillin, kanamycin, gentamicin, tobramycin and netilmicin,and resistance to erythromycin, rifampin and ofloxacin.

Molecular typing results
Phenotyping results suggested BT from the GIT to the circulatory system in 15 preterm infants. When the same Staphylococcus spp. were isolated from both stool and peripheral blood, and exhibited the same resistance pattern, they were further genotyped by ERIC-PCR and RAPD-PCR to confirm BT. Fifteen isolates were selected to obtain a diverse sample of patients, blood and stool samples, and R patterns. The three other uCoNS strains with resistance patterns e and f exhibited the CD genotype. Finally, nine S. haemolyticus phenotype e strains exhibited the BE genotype. This major epidemic BE profile included 60% of S. haemolyticus strains (9/15) isolated in both blood culture and stool culture. The remaining strains (three S. epidermidis and three uCoNS) exhibiting AF and CD genotypes,respectively, were considered to be sporadic cases. The BE genotype was identified in both units participating in this study.
Combined analysis of ERIC-2 and RAPD results identified three different genomic groups (gg): I to III. The strains isolated from blood culture and stool culture in each group were more similar to each other than to the other strains.

Bacterial Translocation results
Translocation from the GIT to the circulatory system was documented in 53.5% (15/28) of preterm infants. The same Staphylococcus spp. were not found in blood and stool in 46.5% (13/28) of preterm infants, strongly suggesting the absence of BT in these preterm infants, and that the intestinal tract would not constitute the only or direct source of bacteremia. In patient 18,blood culture was positive for S. epidermidis and stool culture was positive for S. haemolyticus, although culture of a nasopharyngeal sample taken prior to the onset of bacteremia isolated S. epidermidis, suggesting that the respiratory tract was the probable source of bacteremia in this child (this source was not included in this study).

Risk factors for the occurrence of BT in preterm infants with CoNS bacteremia
Comparison of documented (group1) and undocumented (group2) gastrointestinal BT is shown in

Discussion
In this study, S.haemolyticus and S.epidermidis were the CoNS species most isolated, with rates of 41.8% and 17.8% respectively. These pathogens are a major cause of nosocomial bacteremia and catheter infections in nICUs [18]. CoNS are the most common cause of late-reported sepsis [19]. Sepsis accounts for 45% of severe infections in neonatology units [20]. Our study also showed that BT in preterm infants with sepsis is responsible of secondary bacteremia and is driven by external factors such as length of stay in the nICUs or prolonged feeding by an enteral feeding tube. Such results have been observed by Jezioski E et al. [21].
According to the study conducted by Pappof P et al [2],prematurily appears to play a significant predisposing role by reducing mucosal barrier function.Other factors reported in the literature to influence BT are as follows:(i) bacterial overgrowth in small bowel, use of antibiotics, obstructive jaundice, intra-abdominal hypertension; (ii) damage to the gut barrier, systemic inflammatory response syndrome, or direct injury (abdominal surgery); (iii) systemic immunosuppression; (iv) immaturity of the intestinal barrier per se, and immaturity of host defense function [22,23].Various publications have identified BT in a wide group of diseases, such as acute pancreatitis, cirrhosis, malignancy, heart failure, aortic aneurysm repair, cardiopulmonary bypass and bowel transplant [22][23][24].
In this study, the presence of BT from the gut to the circulatory system in 15 of 28  [29] or when the host's immune defense mechanism is altered [9]. Under these conditions, micro-organisms from the GIT

Conclusions
This study clearly demonstrates that BT from the intestinal tract was the most likely source of CoNS bacteremia in hospitalized preterm infants.BT appears to be an  Representative RAPD-PCR types of 15 CoNS spp. Isolated from stool cultures. Isolate number