Bacterial isolates
In this descriptive cross-sectional study, 100 urinary K. pneumoniae isolates were obtained from hospitalized patients in Milad Hospital, Tehran, Iran, from December 2016 to October 2017. The isolates were identified as K. pneumoniae by colony morphology, gram staining and standard biochemical tests (12). The ethical approval of the present study was provided by the Ethics Committee of Islamic Azad University of Tehran Medical Branch (No: IR.IAU. TMU.REC.1396.278).

Antimicrobial susceptibility testing
All K. pneumoniae isolates were examined for their antibiotic resistance profile using Kirby Bauer’s disk diffusion method according to the instructions of the Clinical and Laboratory Standard Institute (CLSI) (13).
The antibiotic disks used were ceftriaxone (30 μg), ceftazidime (30 μg), cefoxitin (30 μg), cefepime (30 μg), gentamicin (30 μg), ciprofloxacin (30 μg), levofloxacin (5 μg), amikacin (30 μg), imipenem (10 μg), meropenem (30 μg), piperacillin (30 μg) and aztreonam (30 μg) (Mast Diagnostics, UK). E. coli ATCC 25922 was used as a reference (13).
Multidrug resistant (MDR) was estimated according to previously described definitions (14).

Screening of AmpC beta‑lactamase‑producing strains
All the isolates were tested for AmpC beta-lactamase production using discs of cefoxitin (30 μg) alone and in combination with boronic acid (400 µg). For this purpose, each isolate was inoculated on a Mueller–Hinton agar plate (Himedia, India). The discs were then placed on the surface of the plate and incubated overnight at 37°C.  An increase of ≥ 5 mm in zone diameter around the cefoxitin disc in combination with boronic acid compare to that of cefoxitin disc alone was considered positive (15).

DNA extraction and PCR assay
The DNA extraction was carried out by the boiling method as explained by Perez-Perez and Hanson (16). Six families of plasmid-mediated AmpC beta-lactamases, including DHA, MOX, ACC, EBC, CIT and FOX were amplified by a polymerase chain reaction (PCR) using the primers shown in Table 1. PCR reaction (50 μl) contained 50 mM KCl, 20 mM Tris-HCl (pH 8.4), 1.5 mM MgCl2, 0.2 mM each deoxynucleoside triphosphate, 0.5 μM of each primers, 100 ng of extracted DNA and 1.25 U of Taq DNA polymerase (Ampliqon, Denmark). PCR reaction was carried out as follows: initial denaturation at 94°C for 3 min followed by denaturation at 94°C for 30s, annealing at 64°C for 30s, extension at 72°C for 1 min (25 cycles) and a final extension at 72°C for 7 min.
 

 Table 1. Comparison of rat weights in the studied groups after 7 and 14 days. Data is expressed as mean±SEM. (n=6)

Molecular typing of AmpC-producing isolates by ERIC-PCR
The clonal relationships among the AmpC-producing K. pnumoniae isolates were determined by ERIC-PCR using the ERIC2 primer as previously described (17).
Briefly, 2 μl of the DNA template was added to 12.5 μl master mix (Ampliqon, Denmark), 1 μl primer (10 pmol), and 9.5 μl H₂O. A PCR reaction was performed under the following conditions: initial denaturation at 94°C for 15 min followed by denaturation at 94°C for 1 min, annealing at 37°C for 1 min, extension at 72°C for 1 min (40 cycles), and a final extension at 72°C for 8 min. The resulting products were analyzed on 1.5% agarose gels. Then, the presence and absence of the bands were scored as 1 and 0, respectively, and the data were analyzed by the NTSYS program (NTSYSpc version 2.10e). Finally, a cluster analysis was performed, and a dendrogram was constructed using an unweighted pair group method with arithmetic averages (UPGMA). To identify clonally related isolates, the similarity cut-off level was set at 90% (18).

Statistical analysis
Data were analyzed with SPSS version 20. Differences between antibiotic resistance among AmpC-positive and negative isolates were statistically analyzed by chi-square tests. A P-value < 0.05 was considered significant.

One hundred K. pneumoniae isolates were obtained from the urine samples of patients with UTIs at Milad Hospital during the aforementioned study period. Of these, 50 isolates (50%) were obtained from females, and 50 (50%) were obtained from males. The mean age of patients was 46.95 ± 23 years. The highest rates of resistance were observed against amikacin and levofloxacin (65% and 64%, respectively). Moreover, the highest susceptibility was demonstrated in relation to aztreonam and imipenem (97% and 83%, respectively). In this study, more than 50% of the isolates were resistant to gentimicin, cefepime, ceftazidime, and piperacillin. Among the 100 K. pneumoniae isolates, 49 (49%) produced AmpC beta-lactamases. The results of antibiotic susceptibility testing are shown in Table 2.
 

Table 2. Antimicrobial susceptibilities of the K. pneumoniae isolates (n=100).


 

Among the 100 K. pneumoniae isolates studied, the PCR revealed that PMABL genes were present in 49 (49%) isolates. Of these, 34 isolates harbored only one AmpC gene group, including MOX (n=11), EBC (n=8), ACC (n=7), CIT (n=4), FOX (n=2), and DHA (n=2). The 15 remaining PMABL-containing isolates harbored at least two AmpC gene groups as follows: DHA, CIT, and MOX in 1 isolate; CIT and MOX in 2 isolates; CIT and ACC in 2 isolates; MOX and ACC in 2 isolates; FOX and DHA in 1 isolate; DHA and ACC in 2 isolates; DHA and MOX in 2 isolates; DHA and CIT in 2 isolates; and EBC and ACC in 1 isolate. Figure 1 displays the electrophoretic pattern of the AmpC genes.
 

Figure 1. PCR amplification of the AmpC genes. Lane 1-6 positive results for AmpC genes, M: 100 bp DNA ladder.

The antimicrobial susceptibility pattern of the 49 AmpC-producers showed resistance levels of 49% (n=24) to piperacillin and gentamicin, 69.4% (n=34) to amikacin, 63.3% (n=31) to levofloxacin, 44.9% (n=22) to ciprofloxacin, 55% (n=27) to ceftazidime, 46.9% (n=23) to cefepime, 38.8% (n=19) to ceftriaxone, 28.6% (n=14) to meropenem, and 10.2% (n=5) to imipenem. There was a significant association (P0.05) between AmpC gene carriage and resistance to cefoxitin and levofloxacin. In this study, aztreonam (98% susceptibility) was found to be the most active antibiotic against AmpC-producing isolates.
Multidrug drug resistance was detected in 30 (61.2%) of the AmpC beta-lactamase producers. These isolates were distributed into 24 antimicrobial resistance patterns, dominated by resistance to gentamicin/amikacin/meropenem/ceftriaxone/cefoxitin (GM/AK/MER/CRO/FOX; 3/30, 10%), followed by amikacin/cefepime/ceftazidime /ciprofloxacin/levofloxacin/piperacillin (AK/CPM/CZA/CIP/LEV/PRL; 2/30, 6.7%). The profile of antimicrobial sensitivity in MDR isolates and the prevalence of AmpC beta-lactamase genes are reported in Table 3.
 

Table 3. Antimicrobial resistance patterns of multidrug resistant K. pneumoniae isolates and frequency of genes coding for MDR AmpC beta-lactamase.

Enterobacterial repetitive intergenic consensus analyses revealed 12 distinct patterns of AmpC-producing K. pneumoniae isolates with a similarity of above 90% (Figures 2 and 3). The 49 AmpC-producing isolates were divided into four groups (A, B, C, and D), among which group D, with 7 clonal types and 28 isolates, was the most dominant. As shown in the dendrogram, among 12 clonal types, types I, V, and XII were the predominant types, with 8, 8, and 7 isolates, respectively (Figure 3).
 

Figure 2. Example of DNA banding patterns obtained for AmpC producing K. pneumoniae isolates by ERIC-PCR fingerprinting. 1: Negative Control, 2-13: Twelve clonal types of K. pneumoniae isolates, 14: 1Kb Ladder).

Figure 3. Dendrogram generated by NTSYS software of ERIC-PCR patterns from the 49 AmpC producing isolates. The vertical line represents the similarity cut-off level of 90%.

Discussion

In The present study demonstrated that the most common organisms isolated from both community and hospital-acquired infections in pediatrics patients were E. coli and S. aurous with a prevalence of 67.6% and 30.2%, respectively. These findings were inconsistent with the studies conducted in India (14) and Turkey (15) which found that E. coli and S. aurous were most prevalent. Moreover, the highest resistance was observed against Co-trimoxazole in E. coli which caused both community-acquired and hospital-acquired infections. Therefore, due to its high resistance, Co-trimoxazole is not recommended as the mainstay of treatment for empirical therapy in both types of infections. This could be due to indiscriminate prescribing of these antibiotics to pediatric outpatients.
 

Table 1. Oligonucleotide primers used for RT-PCR


Table 2. Results of susceptibility test in gram negative organisms based on the infections types



Table 3. Results of susceptibility testing in gram positive organisms based on the infections

In The present study demonstrated that the most common organisms isolated from both community and hospital-acquired infections in pediatrics patients were E. coli and S. aurous with a prevalence of 67.6% and 30.2%, respectively. These findings were inconsistent with the studies conducted in India (14) and Turkey (15) which found that E. coli and S. aurous were most prevalent. Moreover, the highest resistance was observed against Co-trimoxazole in E. coli which caused both community-acquired and hospital-acquired infections. Therefore, due to its high resistance, Co-trimoxazole is not recommended as the mainstay of treatment for empirical therapy in both types of infections. This could be due to indiscriminate prescribing of these antibiotics to pediatric outpatients.
As evidenced by the obtained results, no resistance was found against Colistin and Meropenem in Gram-negative microorganisms which caused the community-acquired infections, followed by Imipenem and Cefepime with the prevalence of 2.2%, and 8.9%, respectively. Consequently, Meropenem can be used as the first-line treatment empirically in critically ill patients with HA-sepsis until obtaining the results of culture and susceptibility test. However, in CA-infections, cefepime can be used as an acceptable antibiotic in the empirical regime. In A. baumannii, the minimum of resistance was reported against ciprofloxacin which was probably owing to the no or at least use of this antibiotic in the pediatric ward. In CA and HA-infections caused by Gram-positive cocci, no resistance was observed against vancomycin; however, 36.4% of CA-S. aurous and 42.6% of HA-S. aurous were resistant to cefoxitin (MRSA).In addition, the highest resistance in S. aurous was reported against clindamycin which indicates that these medicines should not be prescribed in empirical therapy.
There exists a high resistance of methicillin in MRSA strains with a high percentage of prevalence in hospitalized patients suspected to Gram-positive infectious agents. In this case, vancomycin empirical prescription and utilization should be undertaken until getting the results of culture and susceptibility testing. In community-acquired infections which are caused by E. coli, the highest resistance was to Co-trimoxazole with a frequency of 64.4%, followed by Ceftriaxone (27.6%). In line with these results, a study conducted in Tehran in 2007 found that the most common organism was E. coli with a prevalence of 51.4%, and the most resistance was against Ampicillin and the least one was to ciprofloxacin, ofloxacin, and amikacin (16). Whereas, in a study performed in Bahrami Hospital, Tehran, the most common reported isolates were P. aeruginosa (24.3%) and Klebsiella pneumonia (18.6%) (17). The prevalence of pathogenesis is roughly similar in the distinct regions all around the world despite the fact that antimicrobial resistance patterns are markedly different.
This difference in antibiotic resistance patterns in various parts of the world is attributed to the immoderate and infelicitous antibiotics utilization, geographical difference, and the difference in prescribing the antibiotic regimes (15). E. coli was recognized as the most prevalent organism in UTI which is consistent with the results of previous studies (18,19). Moreover, our findings revealed that E. coli isolated from UTI in both CA and HA infections had the same antimicrobial pattern since Imipenem was the most active agent against E. coli and the most resistant one to Co-trimoxazole. In the present study, 100% of K. pneumonia isolated from community-acquired infections were resistant to Co-trimoxazole. Whereas, all isolates were susceptible to almost all antibiotics in acquired-hospital infections. In accordance with these results, a study conducted by Yuksel et al. (2006) conducted on UTI in Turkey revealed that the most common causative agent was E. coli, followed by K. pneumonia. Moreover, resistance to ampicillin and Co-trimoxazole in all isolates was another noteworthy result of the mentioned study (20). In community-acquired UTI, we reported 100% resistance against Co-trimoxazole; nonetheless, it should be omitted from the drug regime in the treatment of children-UTI. Instead, the low level of resistance to Imipenem and Meropenem can be a reasonable alternative in the treatment of UTI (20, 21).
In community-acquired bloodstream infection (CA-BSI) caused by Gram-negative bacteria, the highest resistance was reported against ceftriaxone with a prevalence of 31.6%, while no resistance was described against Colistin. Whereas in hospital-acquired bloodstream infection (HA-BSI), the most resistance was against ceftriaxone with a prevalence of 37.5% and the lowest was against Colistin/gentamicin and tazocin with a frequency of 12.5%. Consequently, the prescription of gentamicin or Piperacillin/Tazobactam rather than ceftriaxone can be the proper choice in BSI cases. Furthermore, in Gram-positive organisms which caused CA-BSI, the least resistance was against vancomycin (0%), while the highest was to clindamycin (50%). Nevertheless, among Gram-positive organisms which developed HA-BSI, the most resistance was to cefoxitin with a prevalence of 46.2% and the least resistance to vancomycin (0%). Therefore, in patients with HA-BSI, vancomycin can be a suitable choice for empirical treatment. In the current study, MRSA which caused community and hospital-acquired bloodstream infections was reported as 16.7% and 46.2%, respectively. However, the reported prevalence of MRSA was 79% in a study (five-year evaluation) conducted by Pourakbari et al. in Iran within 2001-2005. This prevalence was very higher than the value obtained in the current study; moreover, the mentioned study was performed on the patients who suffered from BSI (22).
As mentioned earlier, the antimicrobial pattern of BSI caused by gram-negative isolates in community-acquired infections was consistent with the hospital-acquired one. The best choice for antibiotic therapy was Imipenem or Meropenem, whereas ceftriaxone seemed to be ineffective. In Gram-positive cocci, the most resistance was observed to clindamycin and erythromycin and the most susceptibility was detected to vancomycin. In a study carried out in Nepal within 2006-2008 on neonatal sepsis, the most frequent organisms isolated from blood culture samples were Klebsiella, followed by S. aurous. Moreover, the most effective antibiotic against Gram-negative organisms was cefoperazone/sulbactam and vancomycin in Gram-positive bacteria (23). As illustrated by the obtained results, a number of factors have increased the resistance against antimicrobial drugs. They include adequate staff, state health laws, hygiene in medical processes, control measures, stringent hospital policy regarding diagnosis and treatment of hospital-acquired infections, ongoing training and education, limited medical methods and medical device use, proper application of disinfectants-washers, and antimicrobial medicines (24). Geographic variations in the prevalence of pathogens and profiles of antibiotic susceptibility necessitate continuous monitoring and lead to the selection of best therapeutic options (25).
Taking everything into account, the present study pointed to the high resistance of Gram-negative isolates against Cotrimoxazole and ceftriaxone, as well as clindamycin in Gram-positive isolates. Consequently, it can be used as a good guide for describing the empirical treatment in CA and HA infections. Moreover, the results suggested that Cefepime can be a promising choice for empirical treatment in UTI cases. Furthermore, a combination of gentamycin and vancomycin can be used as an empirical regime in hospital-sepsis.

 

Conclusion

In addition, the findings of the current study were indicative of the high resistance to several antibiotics that can be used as excellent choices in the treatment of both CA and HA-infections. On a final note, it is recommended that region-specific monitoring studies be carried out in order to assist the clinician to select the accurate empirical therapy.

Acknowledgements

The authors thank all those who helped them writing this paper.

Conflicts of Interest

Authors declared no conflict of interests.