Evaluating antimicrobial appropriateness in a tertiary care pediatric ICU in Saudi Arabia: a retrospective cohort study

Antibiotics are the most common medications prescribed in pediatric intensive care units (PICUs), with up to 50–100% of patients receiving an antibiotic prescription [1‐5]. Patients in a PICU setting have a high prevalence of community and hospital-acquired infections and an overall high risk of morbidity and mortality [6‐8]. However, antibiotics are not justified in every case and their prescription may signify a lack of judicious use or improper reasoning. It has been found that almost 20–50% of all antibiotics prescribed in a pediatric critical care setting are inappropriate [2, 9]. Such inappropriate antibiotic utilization contributes to emergence of antimicrobial resistance (AMR), adverse drug reactions, and additional morbidity and hospitalization costs [10‐12]. It has been estimated that AMR will result in 10 million deaths by 2050 if no global action is taken [13]. In 2015, the World Health Organization (WHO) introduced a global strategy to address AMR that incorporated several interventions, including the reduction of inappropriate antimicrobial utilization [14, 15]. With increasing reports of emerging AMR and declining pathogen susceptibility in ICU patients, Saudi Arabia joined the WHO global action plan for Containment of Antimicrobial Resistance [16‐18].

Antimicrobial stewardship programs (ASPs) are recommended to decrease inappropriate antibiotic use and mitigate its effects [19, 20]. This requires an understanding of current antibiotic prescription practices and rationale. Data on antibiotic prescription in our institute have been published without detailing indications, types of infections or the appropriateness of antibiotic prescriptions [21]. Such information helps to quantify the problem and identify the main areas that require attention and modification through an ASP strategy. The objectives of this study were to characterize antibiotic utilization, determine the appropriateness of antibiotic use by applying the Center for Disease Control and Prevention (CDC) 12-step Strategy to prevent antimicrobial resistance among hospitalized children, and study factors associated with inappropriate antibiotic prescription in a tertiary care PICU.

During the study period, 497 patients were admitted to the PICU, accounting for 3009 patient days. After excluding 238 patients based on the study criteria, 259 patients were included in the review. Table 2 shows the characteristics of patients in the cohort. One hundred thirty-two (50.8%) patients were male, and the median age was 22 months (IQR: 5–65 months).

Overall, 259 children received 274 antibiotic courses, resulting in 2553 DOTs. The median duration of an antibiotic course was 8 days (IQR, 5–13). The reason for antibiotic initiation was empiric in 187 courses (68.2%), prophylactic in 61 courses (22.3%), and definitive in 26 courses (9.5%). The most common clinical indications for antibiotic initiation were community-acquired pneumonia (19.3%), hospital-acquired sepsis (10.6%), community-acquired sepsis (9.9%) and bronchiolitis (9.5%).

Compared to appropriate antibiotic courses, inappropriate courses were more likely in surgical admissions and in patients receiving antibiotics prophylactically (P < 0.05). A higher proportion of inappropriate courses occurred in patients receiving antibiotics post CNS procedures (P = 0.001). The duration of antibiotic courses was significantly higher in children with inappropriate antibiotic use than in children who had appropriate antibiotic use (median of 10 vs. 7.0, P ≤ 0.001). For empiric antibiotics, no action taken at 72 h was associated with a higher percentage of inappropriate antibiotic courses, 20.8% versus 9.1% (P = 0.031).

Table 3 shows the distribution of nonadherent courses based on the CDC steps violated. Out of 274 courses, 133 (48.5%) were found to be nonadherent to at least 1 of the five CDC steps. Step 6 (Practice antimicrobial control) was the most prevalent step violated, followed by step 4 (Target the pathogen).

Figure 1 lists the ten most frequently consumed antibiotics in the PICU and the number of inappropriate DOT per antibiotic. In our center, third generation cephalosporins, vancomycin, and cefazolin were the most frequently used antibiotics. Cefazolin and vancomycin, were the most inappropriately used antibiotic therapies.

The main objectives of this study were to describe antibiotic consumption in our setting, investigate appropriateness of antibiotic utilization by assessing adherence to the CDC’s 12-steps and uncover factors associated with inappropriate antibiotic utilization. The findings suggest a high percentage of inappropriate antibiotics courses. The most common violated CDC steps were “target the pathogen” and “practice antimicrobial control”. There was a higher percentage of inappropriate courses in surgical admissions and with prophylactic courses. Cefazolin and vancomycin had the highest number of inappropriate DOTs.

Total antibiotic consumption was 848.5 DOTs per 1000 patient days, indicating a rising increment from 708.3 DOTs per 1000 patient-days previously reported between 2012 and 2015 [21]. This difference may be attributable in part, to seasonal variation and an increase in prescriptions during winter months in our study period [27]. Our study uncovered a high percentage of inappropriate antibiotic courses (n = 133 out of 274 courses, 48.5%). Previous studies evaluating appropriateness of antibiotics have shown similar percentages in Turkey (50%), Canada (16–60%), and Switzerland (33%) [2, 4, 5].

The most common reason for non-adherence was violation of step 6 (target the pathogen) or step 4 (practice antimicrobial control). Nonadherence to step 4 includes inappropriate empiric or definitive antibiotic choices. Improving empirical and definitive choices would require a better understanding of common pathogens and local susceptibility patterns, and better utilization of diagnostics, in addition to reassessment and taking action after 72 h of antibiotic initiation. Within our cohort, for empiric antibiotics, a lack of action taken at 72 h was associated with a higher percentage of inappropriate antibiotic use (P = 0.031). In view of this finding, conducting an antibiotic ‘time-out’ would be a suitable and effective intervention [26]. A time-out is an ASP intervention recommended by the CDC that standardizes the review of clinical and laboratory results at a predefined time post antibiotic initiation to identify opportunities for discontinuing or deescalating empiric antibiotics [28].

Nonadherence to step 6 suggests an inappropriate prophylactic choice or duration, inappropriate antibiotic combinations, or route of administration. In our cohort 45 out of 61 (74%) prophylactic courses were deemed inappropriate. This may explain why the percentage of inappropriate antibiotics was higher in surgical patients (P < 0.001). Unnecessary prolongation of prophylactic treatment for specific procedures or for the presence of drains may be a contributing factor. Guidelines for perioperative prophylaxis recommend durations not extending beyond 24 h from initiation irrespective of the surgical procedure or the presence of drains or catheters [22, 23].

It is noted that nonadherence to a combination of the CDC steps, and in particular combinations including steps 4, 9, and 10 have contributed to 30.7% of inappropriate DOTs. These steps are related to empiric and definitive antibiotic choices, durations, and avoiding the unnecessary use of broad spectrum agents. Antibiotic treatment decisions in intensive care are challenging and influenced by a difficulty to differentiate between bacterial and viral infections, a lack of gold standard to diagnose pneumonia and an uncertainty in sepsis etiology [29‐32]; for instance, in our setting cultures were positive in only 38% of patients who met severe sepsis or septic shock criteria [33]. An international survey on antibiotics decision determinants found significant variability across countries, reflecting cultural and contextual factors [34]. The issue of nonadherence to multiple CDC steps found in the presented study and available literature emphasizes the need to support clinicians and PICU units with adequate diagnostics and local antibiotic guidelines to rationalize choices and durations of treatment.

Cefazolin and vancomycin contributed to one third of inappropriate antibiotic utilization (301 of 801 inappropriate DOTs, 36.6%). The finding that cefazolin had the highest inappropriate utilization supports the finding that CDC step 6 was the most violated step in nonadherent courses and that a higher percentage of inappropriate courses were seen in surgical patients, as it is the drug of choice for peri-operative prophylaxis. Inappropriate vancomycin utilization in our centre corresponds with studies in both adult and pediatric critical care settings [35, 36]. Vancomycin prescription has been increasing in critical care settings, due to the risk of community and hospital acquired methicillin-resistant Staphylococcus aureus (MRSA) [37, 38]. Several strategies to control vancomycin use in critical care units have been proposed; negative MRSA surveillance swabs were found to have a high negative predictive value for subsequent MRSA infections [39]. Automating vancomycin prescriptions based on controlled approved indications and vancomycin targeted time-out were associated with reduction of inappropriate vancomycin utilization [26, 35].

This study has several limitations. First, data were collected retrospectively; thus, some information influencing decision making could have been missed. Second, this is a single center report, which limits its generalizability. Third, decisions regarding antibiotic appropriateness are prone to bias, despite assessing the appropriateness in this study by two independent infectious disease physicians based on the unified CDC 12 step classification. Additionally, due to the retrospective nature of the study, the accuracy of diagnoses was not confirmed, and the authors accepted all diagnoses made by treating physicians. This meant that we could not assess nonadherence to step 8 (treat infection but not contamination or colonization).

We were able to confirm a high prevalence of antibacterial utilization in the PICU with evidence of inappropriate prescription practices, mainly due to inappropriate perioperative prophylaxis durations and inappropriate empiric and definitive antibiotic choices. Based on our findings, the main areas for stewardship interventions include the development of empiric antibiotic guidelines (addresses step 4 and 9), re-education on surgical prophylaxis guidelines (addresses step 6), and the introduction of ‘time-out’ moments post-antibiotic initiation for empiric antibiotics (addresses step 10). Different approaches to antimicrobial stewardship need to be individualized based on identified concerns. Decisions regarding the appropriateness of antibiotic use could be facilitated by the development of local antibiotic guidelines and the utilization of tools such as the CDC steps to both standardize the process and limit bias and subjectivity.