Bloodstream infections associated with Serratia Marcescens in children

Abstract

Serratia marcescens is an important pathogen increasingly associated with nosocomial infections, particularly in neonates and immunocompromised children. This study aimed to evaluate the clinical features, antimicrobial resistance, and mortality risk factors of pediatric patients with S. marcescens bacteremia. A retrospective analysis was conducted on 77 pediatric patients (0-18 years) diagnosed with S. marcescens bacteremia between January 2017 and December 2021. Demographics, underlying conditions, antimicrobial susceptibility, treatments, and mortality-related factors were assessed. The median age was 3 months (1–15.5 months), and 32.5% were female. Overall mortality was 31.2%. Prematurity was observed in 58.5% of cases. Chronic diseases were significantly more common among deceased patients (p = 0.025). The most frequent comorbidities were neurometabolic disorders (37.7%), surgical conditions (29.3%), and chronic lung diseases (12.1%). Empirical therapy included meropenem in 57.1% and amikacin in 29.9% of patients. Antibiotic regimens were modified in 65.7% due to resistance. Resistance rates were 50% for meropenem, 38.9% for piperacillin-tazobactam, 2.5% for amikacin, and 9% for tigecycline. A substantial proportion of isolates exhibited multidrug resistance. Serratia marcescens bacteremia in children is associated with high mortality and alarming antimicrobial resistance, especially in intensive care settings. Careful empirical therapy selection, early recognition, and risk factor management are essential to improve clinical outcomes.

INTRODUCTION

In the 1950s, Serratia was initially classified as a harmless saprophyte but was later identified as a causative agent in sporadic cases. Over the past fifty years, it has been definitively recognized as an opportunistic pathogen responsible for nosocomial infections. Serratia is a genus of Gram-negative bacteria within the order Enterobacterales [1]. To date, 23 species of the Serratia genus have been identified, of which six are known to cause infections in humans. The most frequently isolated species is Serratia marcescens [2]. This microorganism is a motile bacterium capable of growing at temperatures between 5°C and 40°C and pH levels ranging from 5 to 9. Common habitats for S. marcescens include water, soil, animals, insects, and plants [3]. Serratia marcescens is typically isolated from blood and can be cultured on blood agar or selective media such as MacConkey or chromogenic agars [4].

Serratia marcescens is known to cause a wide range of infections, including septicemia. It is considered a primary entry point for infections associated with urinary catheters, intubation, and central venous catheters [4]. Serratia marcescens bacteremia is particularly associated with significant morbidity and mortality in premature infants. Key risk factors include prematurity and low birth weight, prolonged hospital stay, exposure to invasive devices such as central or percutaneous venous catheters, and prior antibiotic use [5-7].

Clinical outcomes of S. marcescens infections vary from localized infections to sepsis and death, particularly among neonates and critically ill pediatric patients. Several recent studies have emphasized the high mortality rate and treatment challenges caused by multidrug-resistant strains in intensive care units [8-10].

Serratia marcescens often acquires antibiotic resistance by producing enzymes such as deoxyribonuclease, lipase, and gelatinase [4]. It exhibits resistance to a wide range of antibiotics, including penicillins, cephalosporins, tetracyclines, macrolides, nitrofurantoin, and colistin. In the past, antibiotics such as fluoroquinolones, aminoglycosides, and third-generation cephalosporins formed the basis of treatment for S. marcescens infections. However, many clinical isolates of S. marcescens now exhibit multidrug resistance to these antibiotics [4]. Emerging resistance to carbapenems, especially meropenem, has been reported in multiple centers, limiting therapeutic options and leading to treatment failures [11-13].

In this study, we aimed to emphasize the clinical significance of S. marcescens, increasingly responsible for hospital-acquired infections, by evaluating intensive care admissions, antibiotic susceptibility results, and mortality risk factors in pediatric patients from whom S. marcescens was isolated at our hospital.

MATERIALS AND METHODS

In this study, pediatric patients aged 0–18 years with S. marcescens bacteremia between January 2017 and December 2021 were retrospectively evaluated. The demographic characteristics of the patients, coexisting systemic diseases, antimicrobial susceptibility of S. marcescens strains, length of hospital stay, treatments received, responses to treatment, and mortality risk factors were assessed.

The study was conducted at Necmettin Erbakan University Meram Faculty of Medicine, a tertiary care referral hospital with a 40-bed neonatal intensive care unit (NICU) and a 10-bed pediatric intensive care unit (PICU).

Blood cultures were analyzed using automated blood culture systems (DL-Bt240, Zhuhai Biotech Co. Ltd, China, and BACTEC Fx Top, Becton Dickinson, USA). Gram staining was performed on samples that showed growth signals, and identification was carried out using conventional methods and automated systems (VITEK MS MALDI-TOF, bioMérieux, Marcy l’Etoile, France, and BACTEC Fx Top, Becton Dickinson, USA). Antibiotic susceptibility testing was performed using an automated system (BACTEC Fx Top, Becton Dickinson, USA) and evaluated according to EUCAST criteria.

Not all S. marcescens isolates were tested against the full antibiotic panel due to variations in laboratory workflow and clinical prioritization. This limitation has been acknowledged in the discussion. Multidrug resistance (MDR) was defined as resistance to at least one agent in three or more antimicrobial categories, according to standard international definitions.

Data entry and statistical analyses were performed using SPSS for Windows version 18.0 (SPSS Inc., Chicago, IL, USA). The normality of data distribution was evaluated using visual (histograms and probability plots) and analytical methods (Shapiro–Wilk test). Numerical data were summarized with mean, standard deviation, and median (1st–3rd quartiles), while categorical data were summarized with frequency distributions and percentages. Non-normally distributed numerical data and categorical data were compared using the Mann–Whitney U test, and categorical comparisons were conducted using the Chi-square (χ²) test. The diagnostic predictive properties of C-reactive protein (CRP) for mortality were examined using receiver operating characteristics (ROC) curve analysis. Statistical significance was accepted at p < 0.05.

RESULTS

Seventy-seven pediatric patients with confirmed S. marcescens bacteremia were included. The median age was 3.00 (1.00-15.50) months. Of the patients, 32.5% (n = 25) were female and 67.5% (n = 52) were male.

Overall, 31.2% (n = 24) of patients died during hospitalization. Chronic disease prevalence was significantly higher in deceased patients compared with survivors (p = 0.025). In total, 55 patients (71.4%) had one or more comorbidities (Table 1).

Table 1.

Comparison of demographic and disease characteristics of patients.

Characteristics	All patients(n = 58)	Deceased patients(n = 24)	Surviving patients(n = 53)	p
	n (%)	n (%)	n (%)
Total parenteral nutrition
No	45 (58.4)	16 (66.7)	29 (54.7)	0.232*
Yes	32 (41.6)	8 (33.3)	24 (45.3)
Chronic disease
No	19 (24.7)	2 (8.3)	17 (32.1)	0.025*
Yes	58 (75.3)	22 (91.7)	36 (67.9)
Neutropenia
No	70 (90.9)	22 (91.7)	48 (90.6)	0.623*
Yes	7 (9.1)	2 (8.3)	5 (9.4)
Thrombocytopenia
No	49 (63.6)	16 (66.7)	33 (62.3)	0.710*
Yes	28 (36.4)	8 (33.3)	20 (37.7)
	Median (1st-3rd Quartile)	Median (1st-3rd Quartile)	Median (1st-3rd Quartile)
Age (months)	3.00 (1.00-15.50)	4.00 (2.00-29.25)	3.00 (1.00-15.50)	0.691**
Neutrophil (mm3)	5900.00(3265.00-9135.00)	70005.00(3530.00-11542.50)	5780.00(2130.00-7925.00)	0.149**
Platelets (mm3)	235.00(144.50-319.50)	195.00 (115.25-249.25)	245.65(170.00-335.00)	0.081**
C-reactive protein (mg/l)	43.90(10.55-83.55)	48.90 (14.70-100.62)	36.00 (9.50-63.00)	0.370**

*Chi-square test

**Mann Whitney U test

Of these comorbidities, 29.3% (n = 17) were surgical, 10.3% (n = 6) congenital heart diseases, 37.7% (n = 23) neurometabolic diseases, 12.1% (n = 7) chronic lung diseases, 5.2% (n = 3) immune deficiencies, and 3.4% (n = 2) chronic kidney diseases. Among the 22 deceased patients with chronic disease, 50.0% (n = 11) had neurometabolic disorders. Among the 36 survivors with chronic disease, 33.3% (n = 12) had neurometabolic conditions and 33.3% (n = 12) had surgical conditions.

A total of 44 patients (57.1%) were admitted to the NICU, while 33 patients (42.9%) were followed in the PICU. In the NICU group, 70.4% (n = 31) of patients were premature. Overall, 58.5% (n = 45) of the cohort were premature infants.

The median time to blood culture positivity after admission was 25.00 (12.50-40.00) days. The median hospital stay was 64.00 (35.00-128.50) days.

Empirical antibiotic therapy included meropenem in 57.1% (n = 44) and amikacin in 29.9% (n = 23) of cases. Based on susceptibility results and clinical response, treatment regimens were changed in 65.7% (n = 44), while no change was made in 34.3% (n = 23).

In NICU patients, meropenem resistance was detected in 88.6% (n = 39), and piperacillin–tazobactam resistance in 68.1% (n = 30).

The diagnostic performance of C-reactive protein (CRP) in predicting mortality was evaluated by ROC analysis. CRP showed limited predictive value (AUC: 0.564, 95% CI: 0.419-0.709, p = 0.370) (Figure 1).

Figure 1.

ROC curve analysis for C-reactive protein (AUC: 0.564).

DISCUSSION

Hospital-acquired infections due to S. marcescens have become increasingly significant, especially in neonatal and pediatric intensive care units. This pathogen is now recognized as one of the top three causes of outbreak-related bacteremia in NICUs [8-10].

S. marcescens bacteremia is associated with high mortality, ranging between 10% and 30%, depending on host factors and antimicrobial treatment [11,12]. Arslan et al. reported a NICU sepsis outbreak due to S. marcescens, resulting in the death of a premature infant [12]. In our cohort, 57.1% of patients were treated in the NICU and 42.9% in the PICU.

The high proportion of premature infants and long hospital stays likely contributed to bloodstream infections in our population. In line with Friedman et al., we observed that hospitalization beyond 30 days is a strong risk factor for S. marcescens acquisition in NICUs [13]. We also found a strong association between neurometabolic diseases and mortality (p = 0.025). Consistent with previous literature, which indicates that underlying chronic conditions are significant predictors of poor outcomes [14].

Our study highlights an alarming rate of antimicrobial resistance, especially among NICU patients. Meropenem resistance exceeded 80% in this group, which is notably higher than previously reported national figures [6-13,15-17].

S. marcescens produces an inducible chromosomal β-lactamase, contributing to resistance against aminopenicillins and early-generation cephalosporins [16]. However, increasing resistance to third-generation cephalosporins and carbapenems has also been observed in recent years [15-17].

In our study, multidrug resistance was identified in a substantial portion of isolates (11.7%), which further complicates empirical therapy and may negatively affect patient outcomes.

We were unable to perform molecular analyses to characterize resistance genes, which is a limitation of this study. Additionally, not all isolates were tested for the full antibiotic panel due to laboratory constraints. Another limitation is the absence of a control group with infections caused by other pathogens. This restricts our ability to determine whether the observed risk factors are specific to S. marcescens or generally associated with mortality.

Nevertheless, our study provides new insights into the clinical profiles, treatment responses, and mortality factors in children with S. marcescens bacteremia. We believe that the inclusion of resistance trends and patient-level outcomes adds value despite the limited cohort size.

Future multicenter studies with standardized testing protocols and molecular investigations are necessary to deepen our understanding of this important pathogen.

CONCLUSION

Serratia marcescens is a clinically significant nosocomial pathogen, especially in neonatal and pediatric intensive care settings. The increasing prevalence of multidrug resistance, particularly to carbapenems, underscores the importance of local antibiogram surveillance and rational empirical antibiotic selection.

In our study, prematurity, prolonged hospitalization, and underlying neurometabolic diseases were identified as key mortality risk factors. Understanding these factors is essential for early risk stratification and improved clinical management.

Efforts to reduce the incidence and impact of S. marcescens infections must include infection control measures, antimicrobial stewardship, and broader multicenter surveillance initiatives.

CONFLICTS OF INTEREST

The authors declared no conflict of interest.

FUNDING

The authors have no relevant financial or non-financial interests to disclose.

ETHICS APPROVAL

The study protocol was approved by the Local Ethics Committee of Necmettin Erbakan University Meram Faculty of Medicine on May 13, 2022 (Decision No: 3778). The study was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice guidelines. The data analysis was conducted retrospectively, and confidentiality was maintained at all levels.