Intensive care units (ICUs) are among the most common hospital areas where antibiotics are administered. They also play a crucial role in detecting and controlling multidrug-resistant organisms (MDROs), particularly bacteria (Versporten et al., 2018). Many antibiotics are administered improperly in ICUs, with estimates ranging from 30% to 60% (Trivedi et al., 2020). Epidemiological studies have consistently demonstrated a strong link between the irrational use of antibiotics and the emergence and spread of drug-resistant bacteria (Trivedi et al., 2020; National Health Commission of the People’s Republic of China, 2018). In light of these findings, implementing Antimicrobial Stewardship (AMS) Programs is highly recommended in ICUs. AMS aims to promote the rational use of antibiotics and improve patient outcomes (Barlam et al., 2016; Pickens and Wunderink, 2019). A well-executed AMS can have several positive therapeutic effects, including reducing antibacterial therapy duration, antibiotic consumption, hospital stay, healthcare costs, and the incidence of healthcare-associated infections (Ntagiopoulos et al., 2007; Feazel et al., 2014; Lee et al., 2019).

The neurosurgical ICU at our hospital is a semi-enclosed ICU with 22 beds, primarily serving patients who have undergone surgery in the neurosurgical department. Post-neurosurgical meningitis is showing an upward trend, and the increasing prevalence of MDRO affects outcomes following nosocomial meningitis (15van de Beek et al., 2010). Pulmonary infection is also a common complication after neurosurgery. According to relevant studies, the incidence of pneumonia with MDRO also exceeds 10% in patients with mechanical ventilation in the neurosurgical ICU (Teng et al., 2022). In 2021, the neurosurgical ICU had a markedly higher detection rate of MDRO compared to other ICUs within our hospital, with a rate of 18.48%. This alarming finding has prompted the hospital’s AMS team to take action. The occurrence of MDRO in the neurosurgical ICU is primarily attributed to bacterial variation and excessive use of antibiotics in clinical practice. To address this problem, our task force has proposed several critical interventions based on previous studies that assessed AMS’s effectiveness in acute care settings. These interventions include order review and feedback, multidisciplinary management, and whole-process monitoring. Although individual interventions have been studied in previous research, there is a lack of comprehensive assessments that evaluate the impact of a comprehensive AMS program (Lee et al., 2019), especially in the neurosurgical ICU setting.

Therefore, our study aimed to implement a comprehensive AMS program in the neurosurgical ICU and evaluate changes in antibiotic use, bacterial resistance patterns, and patient outcomes before and after implementing the program. By collecting and analyzing these data, we aimed to provide evidence that supports the effectiveness of AMS in the neurosurgical ICU setting and contributes to improving patient care.

The neurosurgical ICU accepts only post-neurosurgical patients. Most of these patients had significant intraoperative blood loss, received blood transfusions, and remained unconscious for a long time after operation (Sachdeva et al., 2017). These conditions are independent risk factors for pulmonary infection. Meanwhile, meningitis is one of the main complications following neurosurgery (Li et al., 2016; Iwasaki et al., 2017). According to reports, the incidence ranges from 0.3% to 6.6% (Chen et al., 2014; Ma et al., 2019) and is particularly high in developing countries. Even the incidence of catheter-associated urinary tract infections in the neurosurgical ICU was the third highest among all ICUs, reaching 5.28% (Ma et al., 2019). All of them result in prolonged hospital stays and increased antibiotic usage, which eventually give rise to MDRO. As with all neurosurgical ICUs, our neurosurgical ICU is staffed not only by intensivists, but also by neurosurgeons, who are not good at postoperative management and using antibiotics. Therefore, it is necessary to implemente a comprehensive AMS program in the neurosurgical ICU and there are few related studies.

This study examined the impact of a comprehensive AMS program led by clinical pharmacists on antibiotic utilization, bacterial resistance, and patient outcomes in a neurosurgical ICU. Implementing this comprehensive AMS program demonstrated significant benefits, including reducing broad-spectrum antibiotics (APBLs) use within the neurosurgical ICU setting. Furthermore, it led to a decrease in the detection rate of multi-drug-resistant bacteria.

The use of antibiotics in ICUs is considerable, and studies indicate that up to 70% of ICU patients receive antibiotics (Gruenberg et al., 2006; Park et al., 2017). The average consumption of antibiotics per 1,000 days of hospital stays in ICUs is reported to be 1,563 defined daily doses, three times higher than in normal wards (Bitter et al., 2016). Alarmingly, more than 50% of antibiotics administered in the ICU are deemed inappropriate (Rhodes et al., 2017). Furthermore, over one-third of non-infected patients receive unnecessary antibiotics (Lam et al., 2018; van Someren et al., 2018). Implementing an AMS program has demonstrated the potential to reduce antimicrobial usage in ICUs without increasing mortality rates (Lindsay et al., 2019). These findings are consistent with the conclusions drawn from our study, which revealed a significant decrease in the proportion of empiric treatment with broad-spectrum antibiotics (APBLs) in patients while maintaining a similar mortality rate before and after AMS implementation, despite the difference of the patient severity. Early and aggressive intervention seemed to contribute to disease severity of patients during the AMS period.

The emergence and spread of antimicrobial resistance can be attributed to various factors. However, antibiotic usage is essential in bacterial resistance, particularly when highly potent antibacterial agents with broad-spectrum activity are used over extended periods. Research has consistently demonstrated a correlation between increased piperacillin/tazobactam use and an 8% increased risk of multi-drug-resistant bacteria (Teshome et al., 2019). Furthermore, the use of imipenem has been associated with the development of substantial resistance to bacterial drugs (Armand-Lefèvre et al., 2013). Our study underscores the importance of promoting appropriate de-escalation practices, which can significantly reduce the detection of MDROs and the resistance rate of common clinically relevant Gram-negative bacteria such as Klebsiella pneumoniae, Escherichia coli, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter cloacae, against antimicrobial agents. By effectively administering antibiotics and evaluating their efficacy, we can effectively delay the progression of bacterial resistance. And this improvement continued, half a year after the end of the study, the detection rate of MDRO in neurosurgical ICU was 14.71%. Due to the outbreak of COVID-19 infection in China from January to March 2023, MDRO infections increased. We also collected the detection rate of MDRO in neurosurgical ICU from April to July 2023 was 11.11%, indicating that the detection rate of MDRO had decreased significantly and remained stable. During the same period, GNB maintained a high sensitivity to APBLs (Ceftazidime: 67.57%; Piperacillin/tazobatam: 85.56%; cefoperazone/sulbactam: 77.48%; Meropenem: 89.19%; Imipenem: 89.19%).

Implementing an AMS program in ICUs poses specific challenges. Clinical pharmacists play a crucial role as integral members of the AMS team, and in our case, the AMS director is a clinical pharmacist. Their expertise in pharmacokinetics and pharmacodynamics is essential to accurately adjust antibiotic dosages in ICU patients. Factors such as molecular weight and fat-soluble protein binding properties of drugs influence their distribution within the body and at the site of infection. Individualized administration of antibacterial agents must consider the patient’s organ function, disease severity, source of infection, and other relevant factors (Chua et al., 2022). A study by MacLaren et al. revealed that a shortage of clinical pharmacists in the ICU contributes to increased mortality among infected patients (Maclaren et al., 2008). This emphasizes the importance of having adequate clinical pharmacists available in ICU settings. In our study, pharmacists assisted physicians in making more prescribing adjustments. It indicates that pharmacists can timely assist physicians to adjust individualized drug administration regimen according to the efficacy of anti-infection treatment, tNGS results and TDM results.

Our neurosurgical ICU exhibited relatively lower mortality rates in both groups compared to some previous studys (Lindsay et al., 2019). It is important to note that the neurosurgical ICU exclusively admits patients undergoing neurosurgery. Unlike general ICUs, sepsis, identified as one of the leading causes of mortality (Kashiouris et al., 2019), is rare in the neurosurgical ICU. Therefore, the lower mortality rates observed in our ICU can be attributed to the different disease profiles between the two settings. However, it should be noted that the neurosurgical ICU experiences an increased proportion of patients with intracranial infections, leading to higher overall use of antibacterial agents, as reflected in the DOT metric. Intracranial infection patients typically require a longer antibacterial treatment than those with other infections (Tamma et al., 2021).

Implementing a comprehensive AMS program with limited resources can be a burden. In this study, team members, including clinical pharmacists, dedicated much time to completing the necessary tasks. Clinical pharmacists had to allocate half a day to monitor patients in the neurosurgical ICU, and the time commitment was even more significant when regular meetings were scheduled. A Japanese study has suggested that a once-weekly intervention can effectively reduce the use of carbapenem antibiotics in the ICU (Hagiwara et al., 2018). Therefore, more research is required to assess the optimal frequency of implementing different interventions within AMS, achieving a balance between improving antimicrobial usage and minimizing the workload of medical staff. Furthermore, we only collected information on antimicrobial resistance after AMS to assess the persistence of the AMS program. More data of antimicrobial application should be collected to assess the sustainability of the AMS program.

Implementing a comprehensive pharmacist-led AMS program in a neurosurgical ICU yielded positive outcomes. The program successfully reduced the use of broad-spectrum antibiotics (APBLs) and increased the susceptibility of Gram-negative bacteria to antimicrobial agents. This achievement highlights the effectiveness of a collaborative approach involving a multidisciplinary team in promoting prudent antimicrobial use and improving patient outcomes within the neurosurgical ICU setting.