Intestinal loads of extended-spectrum beta-lactamase and Carbapenemase genes in critically ill pediatric patients

Abstract

Introduction: Intestinal colonization by Multi-Drug Resistant Organisms (MDROs) can pose a threat on the health of critically ill patients. The extent of colonization by these organisms is related to previous antibiotic treatments and their ability to cause infections among adult patients. The aim of this study is to determine the relationship between the intestinal Relative Loads (RLs) of selected antibiotic resistance genes, antibiotic consumption and extra-intestinal spread among critically ill pediatric patients.

Methods: RLs of bla _{CTX-M-1-Family}, bla _OXA-1, bla _OXA-48 and bla _VIM were determined in 382 rectal swabs obtained from 90 pediatric critically ill patients using qPCRs. The RLs were compared to the patients' demographics, antibiotic consumption, and detection of MDROs from extra-intestinal sites. 16SrDNA metagenomic sequencing was performed for 40 samples and clonality analyses were done for representative isolates.

Results and discussion: 76 (74.45%) patients from which 340 (89.01%) rectal swabs were collected had at least one swab that was positive for one of the tested genes. Routine cultures did not identify carbapenemases in 32 (45.1%) and 78 (58.2%) swabs that were positive by PCR for bla _OXA-48 and blaVIM, respectively. RLs of above 6.5% were associated with extra-intestinal spread of blaOXA-48-harboring MDROs. Consumption of carbapenems, non-carbapenem β-lactams, and glycopeptides were statistically associated with testing negative for bla _{CTX-M-1-Family} and bla _OXA-1 while the consumption of trimethoprim/sulfamethoxazole and aminoglycosides was associated with testing negative for blaOXA-48 (P<0.05). In conclusion, targeted qPCRs can be used to determine the extent of intestinal dominance by antibiotic resistant opportunistic pathogens and their potential to cause extra-intestinal infections among a critically ill pediatric population.

Keywords: antibiotic consumption; extra-intestinal multi-drug resistant organisms; intestinal dominance; microbiome; pediatric patients; qPCR; relative intestinal load; β-lactamase genes.

Figure 1

Violin plot showing the distribution of the relative loads of the samples that were positive by qPCR for bla _{CTX-M-1-Family}, bla _OXA-1, bla _OXA-48, and bla _VIM. The results are expressed in an inverse logarithmic scale where 0 is 100% of the bacterial population, -1 is 10%, -2 is 1%, and so on until -6 which is 0.0001% of the bacterial population, and our detection limit. “n” stands for the number of samples that were positive for the respective gene of antibiotic resistance.

Figure 2

Distribution of the relative loads of the samples that were positive by routine cultures and PCR for OXA-48 and VIM, compared to those that were only positive by PCR. The results are expressed in an inverse logarithmic scale where 0 is 100% of the bacterial population, -1 is 10%, -2 is 1%, and so on until -6 which is 0.0001% of the bacterial population, and our detection limit. *Implies statistically higher relative intestinal loads for samples that were positive by PCR and routine cultures as opposed to being positive by PCR alone (P<0.05).

Figure 3

The distribution of the intestinal relative loads of the bla _{CTX-M-1-Family}, bla _OXA-1, bla _OXA-48, and bla _VIM genes within 2 weeks of extra-intestinal isolation of an organism harbouring the same gene, as compared to the intestinal relative loads that do not correspond to extra-intestinal isolation of an organism harbouring these genes. * signifies statistically higher relative loads for the swabs corresponding with an extra-intestinal isolate compared to those that do not.

Figure 4

ROC analysis for the relative intestinal loads of (A) bla _{CTX-M-1-Family}, (B) bla _OXA-1, (C) bla _OXA-48, and (D) bla _VIM for samples that had extra-intestinal isolates harbouring the relative genes as compared to those that did not have extra-intestinal isolates harbouring these genes. The number of samples included in these analyses are the same as those in Figure 3 .

Figure 5

(A) 16sDNA sequencing results for samples S1 to S29 that were ordered according to decreasing relative intestinal loads to one or several of the genes tested for, and samples S30 to S40 that were negative to all genes tested for. (B) The same samples shown in Part A but where only Klebsiella spp. (in red) and Pseudomonas spp. (in green) are highlighted.