الفهرس 
TitleOnly 14 pages are availabe for public view
 |  | from | 143 |  |  |
| | | from | 143 | | |
Abstract
K. pneumoniae is one of the common causative agents of HCAI worldwide. The ability of
K. pneumoniae to spread easily together with the rise in the occurrence of multidrug resistant
strains makes this bacterium an important public health problem.
Understanding the distribution and the relatedness between K. pneumoniae isolates is
essential for determining the epidemiology of K. pneumoniae HCAI, studying the population
dynamics of this pathogen, and hence planning for successful infection control measures. Since
the beginning of the genomics era, bacterial strain typing by molecular methods has widely
replaced the traditional phenotyping methods.
VNTR analysis is a PCR-based molecular typing method that can distinguish different
strains of K. pneumoniae based on the variability in the number of multiple tandem repeat
regions in their genomes. Molecular typing of K. pneumoniae strains can also be done by
identifying the polymorphism in the sequence of wzi gene, which is one of the conserved genes
within the cps gene cluster responsible for the capsular phenotype displayed by this organism.
The aim of this study was to compare the molecular patterns of MDR K. pneumoniae
isolated from clinical cases in Alexandria university hospitals using two molecular typing
methods: VNTR analysis and molecular capsular typing by wzi gene sequencing.
Sixty biochemically identified K. pneumoniae subsp. pneumoniae isolates (50 MDR and
10 non-MDR) were included in this study. The identity of the isolates as K. pneumoniae subsp.
pneumoniae was verified by PCR amplification of K. pneumoniae-specific 16S-23S internal
transcribed spacer ITS region.
The isolates were collected from clinical specimens obtained from hospitalized patients
in two of Alexandria University hospitals; the MRI hospital (20 MDR and 10 non-MDR
isolates), and Elshatby pediatric hospital (30 MDR isolates) over the period from August 2014 to
December 2016. Clinical data about the patients were collected from the hospital records.
The most predominant age category among patients from Elshatby pediatric hospital was
1-2 years old, and the risk factor mostly observed among them was the previous administration
of antibiotics. Patients from MRI hospital were mostly below 65 years of age, and the mostly
observed risk factors among them were the previous administration of antibiotics, followed by
malignancy and Diabetes Mellitus, respectively. No significant difference was found in the
occurrence of any of the risk factors between the patients infected with the MDR and the non-
MDR K. pneumoniae isolates.
Most of the MDR K. pneumoniae were isolated from respiratory specimens (48%),
whereas 50% of the non-MDR isolates were from Urine, followed by sputum (30%).
All of the 50 MDR K. pneumoniae isolates were resistant, by disc diffusion method, to all
the tested β-lactam drugs (amoxicillin-clavulanate, ceftriaxone, cefotaxime, cefepime, and
aztreonam), with the exception of ceftazidime, cefoxitin and imipenem which were resistant in
98%, 72% and 66% respectively. The resistance to other classes of antimicrobials
(sulphamethoxazole-trimethoprim, fluoroquinolones and aminoglycosides) ranged from 54% to
gentamicin up to 92% to sulphamethoxazole-trimethoprim. Resistance to colisin and tigecycline
by broth microdilution test was 14% and 2%, respectively.
Fifteen isolates were positive for ESBL-production by DDST; 13 from the MRI hospital,
and 2 from Elshatby pediatric hospital. The 50 MDR isolates showed 25 different antibiotic
resistance patterns (A-Y), with the pattern F, denoting resistance to all tested antibiotics except
colistin and tigecycline, being the most predominant (24%). The 10 non-MDR isolates showed 4
different resistance patterns (Z1-Z4), and the pattern Z4 (No resistance) was the most
predominant (60%).
Twenty three K. pneumoniae isolates (15 MDR from Elshatby pediatric hospital, 5
MDR, and 3 non-MDR from MRI hospital) were randomly selected for PCR amplification and
sequencing of the wzi gene fragment. The obtained sequences were analyzed by construction of 2
UPGMA dendrograms; one using our 23 wzi-sequences alone, and another using our sequences
in addition to 135 previously published reference wzi-sequences.
Our 23 K. pneumoniae isolates displayed 14 different wzi alleles, 4 of which were shared
by more than one isolate, while the remaining ten were unique. Three of the 4 shared wzi-alleles
corresponded to the published wzi-alleles wzi137K17, wzi64K14, and wzi2K2, while 7/10
unique wzi alleles corresponded to wzi79K0, wzi95K20, wzi121K43, wzi115K54,
wzi50K15K17K50K51K52, wzi66K54, and wzi44.
PCR amplification of the 8 VNTR loci (A, E, H, J, K, D, I, L) in the 60 K. pneumoniae
isolates was done. Non-amplification was encountered mostly with locus K (35/60 cases).
Assessment of the discriminatory power of each of the 8 VNTR loci revealed that locus J
displayed the highest degree of polymorphism (0.863/ 12 alleles), followed by the loci H, L, A,
E, and K respectively. The 2 loci D and I displayed the least degree of polymorphism (2 alleles),
and their diversity indices were 0.235, 0.155, respectively.
The 60 K. pneumoniae isolates displayed 43 distinct VNTR genotypes. Six of these
genotypes were shared by more than one isolate, with the most predominant including 9 isolates
from Elshatby pediatric hospital. The other 5 shared genotypes included 2-4 isolates. The
remaining 37 genotypes were represented by a single isolate.
The 23 wzi-sequenced isolates displayed 20 VNTR genotypes, and 14 wzi-alleles. The
global discriminatory indices for the 2 methods were 0.90 for wzi-sequence typing versus 0.98
for VNTR typing. Comparison between the genotyping results of these isolates by VNTR
analysis versus capsular typing via wzi gene sequencing revealed that the 2 methods were
successful in typing of the 13/15 isolates from Elshatby pediatric hospital in a similar way. The
remaining 2 isolates showed the same genotype (wzi2-K2) but had different VNTR genotypes
Sharing of certain genotypes among some isolates from Elshatby pediatric hospital,
particularly when confirmed by the 2 molecular typing methods, could suggest probable crosstransmission
of these pathogens within the hospital. This was not encountered among isolates
from MRI, which could be explained by their fewer number, and the larger diversity of
departments from which they were collected. The totally different genotypes of the 10 non-MDR
isolates, including the 3 typed by both methods, supports the high discriminatory power of the
investigated molecular typing methods in distinguishing the MDR K. pneumoniae isolates from
the unrelated non-MDR isolates.from the present work we concluded that:
- There is a high degree of resistance to beta-lactam drugs among clinical isolates of
K. pneumoniae.
- Resistance to gentamicin is much lower than resistance to sulphamethoxazoletrimethoprim,
and floroquinolones among clinical isolates of K. pneumoniae.
- Resistance to imipenem is rapidly rising among clinical isolates of K. pneumoniae .
- Resistance to colistin among clinical isolates of K. pneumoniae is higher than resistance
to tigecycline.
- The wzi-gene sequencing of K. pneumoniae isolates is a useful molecular typing method
with a high discriminatory power; however, it is rather expensive.
- Molecular typing of K. pneumoniae isolates by VNTR analysis using the 8 loci (A, E, H,
J, K, D, I, L) has a high discriminatory power comparable to that of wzi-gene sequencing,
but the method is much more laborious and time consuming.
- The VNTR locus J is the most helpful with the highest degree of polymorphism.
- The 2 VNTR loci D and I are less helpful than the other loci for their low degree of
polymorphism.
- The 2 molecular typing methods were successful in typing of 13/15 isolates from
Elshatby pediatric hospital in a similar way.
- The 2 molecular typing methods had the ability to discriminate the 10 non-MDR
K. pneumoniae isolates from each other and from the 50 MDR isolates.
- Despite the high ability of the 2 molecular typing methods to discriminate the MDR
isolates, this did not correlate with their phenotypic discrimination by antibiotic
resistance patterns.