Prevalence and risk factors of Acinetobacter baumannii infection in Pediatric Intensive Care Unit at Thammasat University Hospital

Introduction

Acinetobacter baumannii causes significant problematic issues in healthcare-associated infections (HAIs). A. baumannii usually causes various infections, including sepsis, pneumonia, meningitis, urinary tract infection, skin and soft tissue infection.^1–3 The critical factors contributing to the global challenge of A. baumannii in healthcare settings include the pathogen’s capacity to rapidly evolve mechanisms of drug resistance, its ability to persist in the environment, leading to widespread transmission, and the high morbidity and mortality rates of infections.^4–7 The emergence of pan-antibiotic-resistant A. baumannii and the lack of effective treatments have led to its classification as a serious public health concern. A. baumannii has been designated as an urgent threat, necessitating the development of enhanced therapeutic strategies.^8,9

Previous studies demonstrated the emergence of A. baumannii infections (ABI) with extremely high rates of carbapenem resistance among adults in Thailand, ranging between 70% to 80%, and mortality rates exceeding 60%.^10–12 Consequently, ABI is estimated to result in over 15,000 deaths each year.¹³ However, there is limited research on A. baumannii in pediatric populations, particularly within intensive care unit (ICU) settings.

Identifying risk factors for ABI are essential for developing targeted prevention strategies. Previous studies, predominantly in adult populations and with limited data from pediatric populations, identified several potential risk factors, including invasive procedures, parenteral nutrition, prior exposure to broad-spectrum antibiotics, age, comorbidities, and ICU length of stay, with reported odds ratios ranging from 1.42 to 10.01.^14–22 Nevertheless, many of these studies lacked clarity regarding the duration of exposure to these risk factors, leading to inconsistent findings. Therefore, this study aimed to identify the prevalence and potential risk factors for ABI, focusing on defining the duration of exposure to these risk factors. This aspect remains underexplored, particularly in the pediatric population.

Methods

Results

Throughout the 8-year duration of the study, a total of 1,635 PICU admission were recorded. Eighty-two cases of ABI were identified giving the overall prevalence of 5.02%. Twelve cases were excluded (7 cases with A. baumannii identified within the first 48 hours, 3 cases had a PICU stay of less than 5 days, and 2 cases had incomplete medical data). The remaining ABI were 70 cases. Among them, the percentage of CRAB reached 94.26%. One hundred forty patients with negative A. baumannii specimens who met the matching criteria were included as controls.

Upon analyzing the annual prevalence of ABI in the PICU, the rate exhibited a fluctuating pattern from 2014 to 2022. From 2014 to 2016, the rate remained stable at 3.00, followed by a noticeable increase in 2017 to 4.14. The prevalence peaked in 2018 at 5.94, before slightly declining to 5.22 in 2019. A more significant decrease was observed in 2020 and 2021, with rates dropping to 3.97 and 3.78, respectively. However, in 2022, the prevalence rebounded to 5.4, approaching the 2018 peak. This overall trend suggests periodic fluctuations in ABI, with notable peaks occurring in 2018 and 2022 ( Figure 1).

Figure 1. Prevalence of Acinetobacter baumannii infection (ABI) in PICU.

Patients’ clinical characteristics in each group are shown in Table 1. The ABI cases exhibited a younger age, a greater frequency of chronic respiratory disease, and prior mechanical ventilation. Additionally, the use of piperacillin/tazobactam and lengthy hospitalization prior to the onset of ABI tended to be more frequent in this group. Patients with ABI experienced prolonged hospitalization after the onset of infection and demonstrated an increasing mortality rate compared to patients without ABI.

Among patients with ABI, there were 30 cases (42.85%) identified as ventilator-associated pneumonia, 8 cases (11.43%) as bloodstream infections, 6 cases (8.57%) as catheter-related bloodstream infections, 5 cases (7.15%) as skin and soft tissue infections, 2 cases (2.86%) as catheter-associated urinary tract infections, 2 cases (2.86%) as ventricular shunt infections, and 17 cases (24.28%) were identified as having a combination of multiple infection types.

In our study, ABI was associated with a mortality rate of 15.71%. All fatal cases involved patients with at least one comorbidity or a history of major surgical intervention. Of these, two patients had acute leukemia, two had chronic respiratory disease, two had an underlying genetic disorder with airway anomalies, two had congenital heart disease, two had a recent car accident and had undergone major surgery in the past 10 days, and one had a brain tumor with craniotomy in the past two weeks. All cases were complicated by septic shock and multiorgan failure, resulting in fatal outcomes.

Cultures were obtained from 70 cases of ABI. Of these 185 isolates, 142 were from respiratory samples (76.76%), 34 were from blood (18.38%), 5 were from pus (2.70%), and 4 were from urine (2.16%). High co-resistance rates were observed for meropenem (94.26%), imipenem (91.78%), piperacillin-tazobactam (89.46%), ciprofloxacin (88.59%), ceftazidime (88.72%), amikacin (83.45%), and cefoperazone-sulbactam (82.71%). In contrast, low resistance rates were observed for tigecycline (3.48%). All the isolates were susceptible to colistin according to the EUCAST guidelines,²⁶ while exhibiting intermediate susceptibility to colistin based on the CLSI guidelines.²⁵ According to the criteria established by Magiorakos et al,²⁷ most of the isolates were classified as MDRAB (n = 174, 94.05%) and 84.32% were classified as XDRAB (n = 156).

Factors associated with ABI, as determined by both univariable and multivariable conditional logistic regression analysis, are presented in Table 2. In the univariable analysis, the risk factors associated with ABI included being under 12 months of age, chronic respiratory disease, and PICU stay longer than 7 days were significantly associated with ABI. Furthermore, within the last 2 weeks prior to the onset of ABI, mechanical ventilation lasting 5 days or more, and using piperacillin/tazobactam significantly increased the risk of ABI.

The final multivariable conditional logistic regression analysis identified chronic respiratory disease, mechanical ventilation for 5 days or more within the preceding 2 weeks, and the use of piperacillin/tazobactam within the last 2 weeks as independent factors associated with ABI.

Discussion

A. baumannii species are emerging as a significant public health threat, frequently exhibiting multidrug resistance and being associated with high mortality rates. An analysis of data collected over eight consecutive years revealed that the overall prevalence of ABI in the PICU was approximately 5.02%. This prevalence is lower than previous studies, which ranged from 8.30% to 15.30%.^28–30

The prevalence of ABI significantly declined between 2020 and 2021, a period coinciding with the global COVID-19 pandemic. Several studies have reported reductions in ABI rates during this time, with decreases ranging from 1.00% to 3.00%.^31–34 This reduction is likely associated with widespread lifestyle changes and the emphasis for more stringent practices of infection control aimed at timing the transmission of SARS-CoV-2. These measures contributed to a decline in respiratory infections among children, reduced ICU admissions for non-COVID-19 cases, a reduction in hospitalization rates, and fewer pediatric patients with severe SARS-CoV-2 infections who required PICU admission.

Similar to adults, the major factors related to the emergence and spread of ABI in children, as demonstrated in previous studies, included misguided antimicrobial treatment and prior invasive procedure such as mechanical ventilation, central venous catheterization, and urinary catheters.^1,35–37 Our study found that the use of piperacillin-tazobactam, prolonged mechanical ventilation for a minimum of 5 days within the last 2 weeks, and chronic respiratory disease were significantly associated with an increased risk of ABI.

Nowadays, where carbapenems are frequently administered, particularly for the treatment of severe cases, the selective pressure may encourage the emergence and spread of CRAB. As a result, the widespread administration of carbapenems is a potential risk factor of ABI.^6,38–40 Differing from previous studies, our study identified piperacillin-tazobactam as a significant risk factor. This difference was attributed to the antibiotic prescribing practices at Thammasat University Hospital.

Piperacillin-tazobactam was widely available and could be prescribed by physicians across various specialties without restriction. Conversely, the use of carbapenem was tightly regulated, requiring authorization from a pediatric infectious disease specialist prior to prescription. As a result, physicians commonly chose piperacillin-tazobactam, a broad-spectrum antibiotic with a similar efficacy profile to carbapenems, particularly for critically ill patients in the PICU. This prescribing pattern likely contributed to the observed association between piperacillin-tazobactam and an increased risk of ABI, which was statistically significant in this study.

Our study is the first to identify chronic respiratory disease, including bronchopulmonary dysplasia (BPD) and asthma, as significant risk factors for ABI. In these patients, the compromised pulmonary architecture and function facilitated bacterial colonization and subsequent infection. Furthermore, these patients often had a history of recurrent hospitalizations and a potential need for mechanical ventilation, both of which contribute to an increased risk of hospital-acquired ABI. Research data in adults demonstrated that chronic obstructive pulmonary disease (COPD) was a significant risk factor for acquiring ABI.^41–44 These findings underscore the significance of chronic lung conditions in predisposing individuals to ABI across different age groups.

Our study demonstrated that mechanical ventilation lasting 5 days or more within the last 2 weeks was an independent factor associated with ABI. Encouraging non-invasive mechanical ventilation and gradually weaning respiratory support while assessing the necessity for endotracheal intubation,⁴⁵ combined with the implementation of targeted antibiotic stewardship program and the reinforcement of rigorous infection control protocols within healthcare setting,⁵ can reduce the hospital-acquired pneumonia/ventilator-associated pneumonia (HAP/VAP) and substantially mitigate the risk of ABI.

The rise of CRAB poses a significant challenge to treatment, as therapeutic options become markedly constrained. These resistant strains are associated with higher morbidity and mortality rates and are increasingly recognized as among the most problematic bacterial pathogens.^{1,3,6,7,40,46} In our study, the prevalence of CRAB reached 94.26%. The presence of CRAB strains is frequently associated with outbreaks in hospital ICU and presents significant challenges to infection control within healthcare settings.^5,47,48 To minimize the risk, carbapenems should be used judiciously, and infection control measures should be strictly implemented.

The high mortality rate associated with ABI may be related to the extent of antimicrobial resistance, the efficacy of empirical therapy, and the availability of targeted therapeutic options. The overall mortality rate in our study was 15.71%, consistent with finding from previous studies.^3,39,49 However, other studies reported higher mortality rate, ranging from 28.3% to 48.65%. This variation is likely attributable to the inclusion of patients with A. baumannii sepsis, with or without septic shock, and delayed in initiating appropriate antibiotic regimens effective against A. baumannii once the infection was established.^1,4,6,40,50

A. baumannii is an important threat to patients of all ages, including children. Therefore, ongoing surveillance of ABI in pediatric populations and continuous assessment of effective prevention strategies for at-risk children are essential. The widespread prevalence of antibiotic-resistant A. baumannii strains, combined with the scarcity of new antimicrobial agents, restricts the available therapeutic options. Prioritizing vaccine development against A. baumannii may be regarded as a potential strategy to combat the challenge of antimicrobial resistance in this pathogen moving forward, which is currently under investigation in research studies.^51–54

Conclusion

This study highlights the significant burden of ABI, particularly carbapenem-resistant strains, in the PICU setting. Prolonged mechanical ventilation for at least 5 days, underlying chronic respiratory disease, and the administration of piperacillin/tazobactam contribute to a higher risk of developing ABI. The findings of our study underscore the importance of the stringent practices of infection control and antibiotic stewardship measures, alongside advocating the use of non-invasive mechanical ventilation in lieu of endotracheal intubation, which can substantially mitigate the risk of ABI. Although the study identifies key risk factors and provides valuable insights into the epidemiology of A. baumannii in the PICU, further research and enhanced infection prevention strategies are necessary to reduce morbidity and mortality related to ABI in critically ill pediatric populations.