Abstract
Background:
Klebsiella pneumoniae is a major etiological agent of pediatric sepsis. The emergence and dissemination of carbapenem-resistant K. pneumoniae (CRKP) and hypervirulent strains (hvKp) represent an escalating global health concern. However, data regarding the coexistence of carbapenem resistance and hypervirulence in pediatric populations, especially in low- and middle-income countries, remain limited.
Objectives:
This study aimed to determine the prevalence of carbapenem resistance and hypervirulence-associated genes among pediatric K. pneumoniae bloodstream infections in Egypt and to assess their association.
Methods:
A cross-sectional study was conducted involving 100 pediatric patients with culture-confirmed K. pneumoniae sepsis at Mansoura University Children’s Hospital between January 2023 and January 2025. Clinical and laboratory data were collected. Antimicrobial susceptibility testing was performed using standard disc diffusion and carbapenemase inhibition assays. Polymerase chain reaction (PCR) was employed to detect carbapenemase genes (including blaNDM for New Delhi metallo-β-lactamase, blaOXA-48 for oxacillinase-48, blaKPC for K. pneumoniae carbapenemase, blaVIM for Verona integron-encoded metallo-β-lactamase, and blaIMP for imipenemase). Hypervirulence genes tested included iucA (aerobactin siderophore synthesis), iroN and iroB (salmochelin siderophore cluster), and peg-344 (putative transporter). Genotypic hvKp was defined as the presence of two or more hypervirulence-associated genes.
Results:
Carbapenem resistance was identified in 64% of isolates, with blaNDM (New Delhi metallo-β-lactamase) and blaOXA-48-like (oxacillinase-48) genes being the most prevalent carbapenemase determinants. Hypervirulence-associated genes were detected in 86% of isolates, which were classified as genotypic hvKp, most frequently iucA (aerobactin system) and peg-344 (putative transporter). No significant demographic or inflammatory differences were observed between CRKP and carbapenem-susceptible groups. Hypervirulence was present at similarly high frequencies in CRKP (87%) and carbapenem-susceptible K. pneumoniae (84%) isolates, with no significant association between these traits (p = 0.782). Logistic regression analysis did not identify any clinical predictors of hypervirulent infection.
Conclusion:
The detection of multidrug-resistant K. pneumoniae isolates harboring multiple hypervirulence-associated genes highlights the potential convergence of resistance and virulence determinants. However, further phenotypic and functional studies are required to confirm the hypervirulent phenotype and assess its clinical significance.
Introduction
Klebsiella pneumoniae is a Gram-negative, encapsulated bacterium of global significance that causes a range of infections, including pneumonia, urinary tract infections, bloodstream infections, and meningitis. Its ability to acquire resistance and virulence genes makes it a formidable clinical pathogen.1,2 Classical strains mainly affect hospitalized or immunocompromised children, while hypervirulent strains, defined by enhanced virulence genes (such as rmpA/rmpA2, siderophores, peg-344, and kfu), also cause severe disease in otherwise healthy individuals.3,4
Carbapenem-resistant (CR) K. pneumoniae (CRKP) rates are rising, especially in low- and middle-income countries. These strains carry carbapenemase genes, leading to limited treatment options and higher mortality.5,6 Pediatric studies in Egypt report high rates, particularly in intensive care units.7,8
Convergence of carbapenem resistance and hypervirulence in K. pneumoniae is increasingly reported. These strains cause severe, hard-to-treat infections, raising the risk of hospital outbreaks.4,9
Global studies confirm that CR hypervirulent K. pneumoniae (CR-hvKp) is emerging, with international spread facilitated by plasmids. CR-hvKp has been linked to increased mortality, though the criteria for defining hypervirulence remain variable.
Recent work shows hypervirulent strains increasingly harbor multidrug resistance, particularly in resource-limited hospitals. Genomic findings underline that coexistence of resistance and virulence is now common, echoing our high prevalence results and highlighting the urgent need for surveillance and infection control.
Despite recognition of this convergence, pediatric-specific data are limited. The frequency of carbapenemase and hvKp-associated genes among pediatric sepsis cases and their association, are incompletely understood. Clarifying these patterns is crucial for guiding empirical treatment and infection control.
This study investigated the prevalence of carbapenem resistance and the distribution of key carbapenemase and virulence genes among pediatric patients with sepsis at Mansoura University Children’s Hospital, Egypt. Additionally, the association between carbapenem resistance and hypervirulent genotypes was examined from January 2023 to January 2025.
Materials and Methods
Study design and setting
The study was a cross-sectional observational study conducted at Mansoura University Children’s Hospital, Egypt, between January 2023 and January 2025. The hospital is a tertiary-care pediatric center providing emergency, intensive care, and inpatient services. All eligible pediatric patients who met the inclusion criteria were consecutively enrolled without randomization.
Ethics approval and consent to participate
All methods were performed in accordance with the ethical standards as laid down in the Declaration of Helsinki and its subsequent amendments, or with comparable ethical standards. The ethical approval for the study was obtained from the Ethics Committee of Mansoura Faculty of Medicine (R.25.11.3466), and written informed consent was obtained from the parents of each child.
Study population
The study included 100 pediatric patients diagnosed with sepsis and having culture-confirmed K. pneumoniae isolates. Patients were recruited from emergency units, neonatal intensive care units, pediatric intensive care units (PICU), and inpatient wards. Pediatric sepsis was defined according to the Phoenix Sepsis Criteria (2024) as a suspected or confirmed infection accompanied by life-threatening organ dysfunction, as indicated by a Phoenix Sepsis Score of ≥2 points. The Phoenix score incorporates dysfunction across four organ systems (respiratory, cardiovascular, coagulation, and neurological), with higher scores indicating greater severity of organ dysfunction. Septic shock was defined as sepsis with cardiovascular dysfunction resulting in a Phoenix Cardiovascular Score of ≥1 point. Patients included in the present study fulfilled the criteria for pediatric sepsis according to these contemporary consensus definitions.
Inclusion and exclusion criteria
Inclusion criteria:
Pediatric patients aged 0–18 years with confirmed sepsis. Positive blood cultures for K. pneumoniae. Availability of complete clinical and laboratory data.
Exclusion criteria:
Polymicrobial infections. Incomplete data.
Clinical data collected from patients’ medical records included age, sex, admission ward, comorbidities, prior antibiotic exposure, invasive device use, duration of hospitalization, and clinical outcomes.
Sample size calculation
A minimum required sample size of 92 participants was calculated using OpenEpi version 3.01, based on an expected carbapenem resistance prevalence of 50%, a 95% confidence level, and an absolute precision of 10%. To improve statistical power, 100 A consecutive sampling technique was employed, enrolling all eligible patients who met the inclusion criteria during the study period.
Laboratory methods
Blood cultures were processed using the automated BACT/ALERT blood culture system. Positive samples were subcultured on MacConkey and blood agar plates and incubated at 37°C. Identification of K. pneumoniae was achieved using standard biochemical tests and the automated VITEK 2 system.
Antimicrobial susceptibility testing
Antimicrobial susceptibility testing was performed using the Kirby–Bauer disc diffusion method following CLSI 2023 guidelines (Oxoid, Thermo Fisher, Wade Road, Basingstoke, Hampshire, RG24 8PW, United Kingdom). Antibiotic discs were used with the following potencies: meropenem 10 µg, imipenem 10 µg, ertapenem 10 µg, ceftriaxone 30 µg, ceftazidime 30 µg, piperacillin–tazobactam 100/10 µg, amikacin 30 µg, gentamicin 10 µg, ciprofloxacin 5 µg, and trimethoprim–sulfamethoxazole 1.25/23.75 µg.
Laboratory biomarkers of inflammation
The laboratory parameters analyzed included inflammatory markers such as white blood cell count (WBC), C-reactive protein (CRP), and procalcitonin, as well as platelet count and serum creatinine. Complete blood count parameters were measured using the Sysmex XN-1000 hematology analyzer (Sysmex Corporation, Kobe, Japan). Serum CRP levels were determined using a latex-enhanced immunoturbidimetric assay on the Cobas c311 chemistry analyzer (Roche Diagnostics, Mannheim, Germany). Procalcitonin levels were measured using commercially available ELISA kits (SunRed Biotechnology Company, Shanghai, China) according to the manufacturer’s instructions.
Phenotypic inhibitory methods for carbapenemase detection
For metallo-lactamase (MBL) detection, imipenem and meropenem discs were supplemented with EDTA. A ≥7 mm increase in the inhibition zone indicated MBL production. KPC detection used boronic acid synergy; ≥5 mm increase indicated positivity. OXA-48 was detected using a cefotaxime disc with a zone diameter <10 mm, indicating positivity.
Although CLSI currently recommends the modified carbapenem inactivation method (mCIM) and EDTA-modified carbapenem inactivation method (eCIM) for phenotypic carbapenemase detection, EDTA-based disc synergy and boronic acid inhibition assays were used in this study as screening methods because they are routinely available and cost-effective in our laboratory setting. Molecular confirmation was subsequently performed by polymerase chain reaction (PCR) detection of carbapenemase genes, which was considered the definitive method for characterization.
PCR detection of resistance and virulence genes
PCR was performed for carbapenemase genes (blaNDM, blaOXA-48, blaKPC, blaVIM, and blaIMP) and virulence genes (iucA, rmpA, rmpA2, iroN, iroB, peg-344, ybtS, and kfuA). DNA was extracted from K. pneumoniae colonies using a standard boiling lysis method. The primers used are listed in Table 1. Two multiplex PCR assays were performed. The first multiplex assay targeted carbapenemase genes (blaKPC, blaNDM, blaOXA-48, blaVIM, and blaIMP), whereas the second multiplex assay targeted virulence-associated genes (iucA, rmpA, rmpA2, iroB, iroN, peg-344, ybtS, and kfuA). Primer combinations were selected following in-house optimization to minimize nonspecific amplification and amplicon size overlap. A common annealing temperature of 58°C was used, as preliminary optimization experiments demonstrated acceptable amplification efficiency across all targets. PCR amplification was performed with an initial denaturation at 95°C for 5 minutes, followed by 35 cycles of denaturation at 95°C for 30 seconds, annealing at 58°C for 30 seconds, extension at 72°C for 45 seconds, and a final extension at 72°C for 7 minutes.10–18
Primer Sequences, Amplicon Sizes, and References
Genotypic hypervirulent K. pneumoniae (hvKp) was defined as the presence of two or more virulence-associated genes detected by PCR, according to previously published criteria that use the accumulation of virulence determinants as a molecular marker of hypervirulence. 19
Statistical analysis
All statistical analyses were performed using IBM SPSS Statistics software (version 26; IBM Corp., Armonk, NY, USA). Data were checked for completeness and normality using the Shapiro–Wilk test. Continuous variables were expressed as median and interquartile range (IQR) due to nonnormal distribution and were compared using the Mann–Whitney U test. Categorical variables were presented as frequencies and percentages and were compared using the Chi-square test or Fisher’s exact test, as appropriate.
A binary logistic regression analysis was performed to identify independent predictors of hypervirulence. Variables entered the model included clinically relevant parameters such as WBC, C-reactive protein (CRP), procalcitonin, hospital stay duration, and carbapenem resistance status. Odds ratios (ORs) with 95% confidence intervals (CIs) were calculated. A p value < 0.05 was considered statistically significant. Given the limited number of nonhypervirulent isolates (n = 14), the multivariable logistic regression model was restricted to a small number of clinically relevant variables to minimize overfitting and preserve model stability. Additional demographic and clinical variables were assessed in univariate analyses and were considered during model construction.
RESULTS
Demographic and clinical characteristics
A total of 100 pediatric patients with culture-confirmed K. pneumoniae sepsis were included in the study. The mean age was 36.6 ± 26.9 months, and males represented 55% of cases. Most patients were admitted to the PICU (45%), followed by general wards (28%) and neonatal intensive care units (24%). Comorbidities were present in 55% of patients, with prematurity being the most common (23%), followed by malignancy (15%), congenital heart disease (10%), and chronic lung disease (7%). Prior antibiotic exposure was reported in 60% of patients. Invasive devices were frequently used, including central venous catheters (26%), urinary catheters (25%), and both devices simultaneously (15%). The overall mortality rate was 18%, while 82% of patients survived (Table 2).
Demographic and Clinical Characteristics of the Studied Children
PICU, pediatric intensive care unit.
Carbapenem resistance profile
Carbapenem resistance was identified in 64% of K. pneumoniae isolates, while 36% of isolates were carbapenem-susceptible (CS).
Molecular study results
Molecular analysis demonstrated a high prevalence of carbapenemase and virulence-associated genes among the studied K. pneumoniae isolates. Among carbapenemase genes, blaNDM was the most frequently detected gene (32%), followed by blaOXA-48–like (27%) and blaKPC (22%), whereas blaVIM and blaIMP were detected at lower frequencies (11% and 9%, respectively).
Regarding virulence-associated genes, iucA was the most prevalent marker (46%), followed by rmpA (35%), iroN (31%), rmpA2 and ybtS (30% each), peg-344 (27%), kfuA (26%), and iroB (23%). Overall, 86% of isolates were classified as genotypic hypervirulent K. pneumoniae (genotypic hvKp) based on the presence of two or more hypervirulence-associated genes (Table 3).
Molecular Detection Results
Percentages represent the prevalence of individual genes among the studied isolates. Because individual isolates may harbor multiple carbapenemase and/or virulence-associated genes, percentages are not mutually exclusive and therefore do not sum to 100%.
Representative agarose gel images of multiplex PCR products showed successful amplification of the investigated carbapenemase and virulence genes in the K. pneumoniae isolates studied. Distinct bands corresponding to the expected amplicon sizes were observed for blaKPC (798 bp), blaNDM (621 bp), blaOXA-48 (438 bp), blaIMP (232 bp), blaVIM (390 bp), iucA (300 bp), rmpA (535 bp), rmpA2 (500 bp), iroB (300 bp), iroN (202 bp), peg-344 (300 bp), ybtS (242 bp), and kfuA (799 bp), confirming the specificity of the multiplex PCR assays (Fig. 1).
Multiplex PCR detection for carbapenem and virulence genes. PCR, Polymerase chain reaction.
Comparison of laboratory parameters between CR and CS isolates
Importantly, no significant clinical or laboratory differences were observed between CRKP and CS K. pneumoniae (CSKP) infections. Both groups exhibited comparable inflammatory responses, organ function, and hospitalization duration, suggesting that carbapenem resistance does not necessarily correlate with increased clinical severity (Table 4).
Comparison Between Carbapenem-Resistant and Carbapenem-Susceptible Isolates
p < 0.05 was considered statistically significant.
CI, confidence interval; CRP, C-reactive protein; OR, odds ratio; WBC, white blood cell count.
Univariate analysis of factors associated with genotypic hypervirulence
Univariate analysis was performed to evaluate potential demographic and clinical factors associated with genotypic hypervirulent K. pneumoniae (hvKp). No statistically significant differences were observed between hvKp-positive and hvKp-negative isolates. The median age was comparable between the two groups (33 vs. 40 months, p = 0.310). Similarly, sex distribution did not differ significantly, with males accounting for 55.8% of hvKp-positive cases and 50.0% of hvKp-negative cases (p = 0.776).
Prior antibiotic exposure was reported in 60.5% of patients with hvKp-positive isolates and 57.1% of those with hvKp-negative isolates (p = 1.000). Admission to the PICU was also not significantly associated with hypervirulence (46.5% vs. 57.1%, p = 0.568). Likewise, carbapenem resistance was detected at similar frequencies among hvKp-positive and hvKp-negative isolates (65.1% vs. 57.1%, p = 0.563).
Because all patients had documented comorbidities and invasive device use, these variables showed no variability and could not discriminate between hypervirulent and nonhypervirulent isolates. Overall, no demographic or clinical factor was found to be significantly associated with genotypic hypervirulence in the studied cohort (Table 5).
Relationship between resistance and hypervirulence
Hypervirulence-associated genotypes were detected at similarly high frequencies among both CR and CS isolates. Genotypic hvKp was identified in 87% (56/64) of CR isolates and 84% (30/36) of CS isolates, with no statistically significant association between carbapenem resistance and hypervirulence (p = 0.782).
Antibiotic resistance patterns
The antimicrobial susceptibility profile demonstrated considerable resistance among the studied K. pneumoniae isolates. The highest resistance rate was observed against amikacin (46%), followed by levofloxacin (38%), imipenem (36%), ceftriaxone (35%), and ceftazidime (34%). Moderate resistance rates were detected for gentamicin (33%) and meropenem (32%), while lower resistance rates were observed for ertapenem and ciprofloxacin (29% each). Piperacillin–tazobactam showed the lowest resistance rate (28%). Overall, the findings indicate widespread multidrug resistance among the isolates (Fig. 2).
Antibiotic resistance rates (%) among the studied isolates.
When comparing CRKP and CSKP isolates, CRKP strains generally exhibited higher resistance rates to most antimicrobial agents; however, none of the observed differences reached statistical significance (all p > 0.05). Resistance to piperacillin–tazobactam was higher among CRKP isolates (32.8%) compared with CSKP isolates (19.4%), although the difference was not statistically significant (p = 0.231). Overall, the findings indicate widespread multidrug resistance among both CRKP and CSKP isolates (Table 6).
Antimicrobial Resistance Profile of Carbapenem-Resistant K. pneumoniae and Carbapenem-Susceptible Klebsiella pneumoniae Isolates
Because 10 pairwise comparisons were performed across antimicrobial agents, a Bonferroni correction was applied to control the family-wise Type I error rate. The adjusted significance threshold was p < 0.005 (0.05/10). None of the observed comparisons remained statistically significant after correction.
CRKP, carbapenem-resistant Klebsiella pneumoniae; CSKP, carbapenem-susceptible Klebsiella pneumoniae.
Univariate Analysis of Factors Associated with Genotypic Hypervirulence
p < 0.05 was considered statistically significant.
Footnote: Because only 14 isolates were classified as nonhypervirulent, multivariable logistic regression was restricted to avoid model overfitting and unstable parameter estimates. Therefore, potential predictors were initially evaluated using univariate analyses.
IQR, interquartile range.
Discussion
This study provides insight into the molecular epidemiology of K. pneumoniae sepsis among pediatric patients in Egypt. The findings revealed a high prevalence of both carbapenem resistance (64%) and hypervirulence-associated genes (86%), highlighting the substantial burden of multidrug-resistant and potentially hypervirulent K. pneumoniae in this vulnerable population. These findings are consistent with reports from low- and middle-income countries, where increasing antimicrobial resistance among K. pneumoniae has emerged as a major public health concern. 20
The predominance of blaNDM and blaOXA-48-like carbapenemases in the present study is consistent with surveillance data from the Middle East and North Africa, where these enzymes are the major mechanisms of carbapenem resistance.18,19 Several studies from Egypt have similarly documented the increasing prevalence of NDM- and OXA-48-producing K. pneumoniae isolates, particularly in neonatal and pediatric intensive care settings.21–25 This regional epidemiology underscores the need for continuous monitoring of carbapenemase-producing organisms and for implementing effective infection control measures. A notable finding was the widespread detection of virulence-associated genes, particularly iucA, rmpA, rmpA2, iroN, and peg-344. These markers are commonly associated with hypervirulent K. pneumoniae and have been linked to enhanced invasive potential, including severe invasive infections.26–30 However, hypervirulence in the current study was defined solely on genotypic criteria, and no phenotypic confirmation assays were performed. Therefore, the isolates should be regarded as carrying hypervirulence-associated genotypes rather than representing definitively confirmed hypervirulent phenotypes.
Importantly, no significant association was observed between carbapenem resistance and hypervirulence-associated genotypes. Hypervirulence was highly prevalent in both CR and CS isolates (87% vs. 84%), suggesting that resistance and virulence determinants may disseminate independently within the studied population. Similar observations have been reported in studies from Asia and the Middle East, supporting the concept that these traits have largely evolved separately and only occasionally converge through acquisition of hybrid plasmids carrying both resistance and virulence determinants.24,26,29,31,32 Nevertheless, given the limited sample size and the small difference observed between groups, a weak association cannot be completely excluded.
Another important observation was the absence of identifiable clinical or laboratory predictors of hypervirulence. Neither inflammatory markers nor demographic and clinical characteristics were significantly associated with the presence of hypervirulence-associated genes. These findings suggest that routine clinical assessment alone may be insufficient to identify isolates carrying virulence determinants and support the value of molecular approaches for accurate characterization of K. pneumoniae strains. 33 The antimicrobial susceptibility profile demonstrated widespread multidrug resistance, with high resistance rates across several commonly used antimicrobial agents. This finding is consistent with the predominance of carbapenemase-producing isolates and reflects the limited therapeutic options available for severe K. pneumoniae infections. The observed resistance patterns are comparable to those reported for NDM- and OXA-48-producing strains in previous studies. 34 Although newer -lactamase inhibitor combinations may provide additional treatment options, their availability and use in pediatric populations remain restricted in many resource-limited settings. 35
The findings of this study have important clinical implications. The coexistence of high rates of carbapenem resistance and hypervirulence-associated genes highlights the importance of strengthening laboratory diagnostic capacity, antimicrobial stewardship programs, and infection prevention strategies. In resource-limited settings, targeted molecular testing for major resistance and virulence determinants may represent a practical approach for surveillance and risk assessment.
Several limitations should be acknowledged. First, hypervirulence was inferred solely from the presence of virulence-associated genes, with no phenotypic confirmation using serum resistance assays, hypermucoviscosity testing, phagocytosis assays, or animal infection models. Second, the sample size calculation was based on estimating the prevalence of carbapenem resistance rather than detecting differences in hypervirulence between CR and CS isolates. Consequently, the study may have lacked sufficient power to detect modest associations between these characteristics. Third, phenotypic carbapenemase detection relied on EDTA-based disc synergy and boronic acid inhibition assays rather than the currently CLSI-recommended mCIM/eCIM methods. However, all phenotypic findings were subsequently confirmed by PCR, reducing the likelihood of carbapenemase misclassification.
Conclusion
Pediatric K. pneumoniae isolates in our setting exhibited a high prevalence of both carbapenem resistance and hypervirulence-associated genes. Although these characteristics appeared to circulate independently, their coexistence within the same clinical environment raises concern regarding the potential emergence of fully convergent CR hypervirulent strains.28,36,37 Future multicenter studies incorporating phenotypic virulence testing and whole-genome sequencing are warranted to further elucidate the clinical and epidemiological significance of these findings.
Authors’ Contributions
D.M.A.-H. contributed to clinical data collection, the preparation of the article draft, data analysis of the study, and revision of the article draft. M.A.N. contributed to the preparation of the article draft, data analysis for the study, and the revision of the draft. M.E.S.Z. contributed to the laboratory study and to the preparation of the article draft. E.H.S. shared the laboratory study, the draft of the article, and the data analysis. K.A.M. contributed to the laboratory study and to the preparation of the article’s draft. All authors have read and approved the final article.
Footnotes
Availability of Data and Materials
The datasets generated and analyzed during the current study are available on demand from the authors.
Disclosure Statement
There is no conflicts of interest for any of the authors.
Funding Information
The study was self-funded.