Risk factors for multidrug-resistant bacteria in critically ill children and MDR score development.

Antimicrobial resistance and healthcare-associated infections (HAIs) are major health concerns in the pediatric intensive care unit (PICU). Device-associated HAIs (DA-HAIs) produced by multidrug-resistant (MDR) bacteria are especially worrying, as they can lead to an inappropriate empirical antibiotic therapy, worsened outcomes and increased mortality. The MDR score was designed to enable the prompt identification of patients at high risk of developing an MDR infection. This was a single-center, prospective, observational study, conducted between January 2015 and December 2022, including PICU patients with a microbiologically confirmed DA-HAI. Demographic, clinical characteristics and outcomes were compared between patients with a DA-HAI caused by MDR and non-MDR-associated DA-HAI, and a risk score for multi-resistance was designed. In total, 257 DA-HAI cases were included, 86 (33.46%) caused by an MDR microbe. In the univariate analysis, comorbidity (p = 0.002), previous MDR colonization (p < 0.001), previous surgery (p = 0.018), and previous antibiotic therapy (p = 0.009) were more frequent among MDR-associated DA-HAI (MDR DA-HAI). In addition, days from device insertion to infection and from PICU admission (p < 0.005) to infection were longer in patients with MDR. In the multivariate analysis, previous comorbidity (OR 2.201), previous MDR colonization (OR 5.149), and PICU length of stay longer than 9 days (OR 1.782) were independently associated with MDR-DA-HAI. Using these three independent risk factors for MDR, a risk score was created: the MDR score. Three risk groups were obtained: low risk (0-2 points), intermediate risk (3-7 points), and high risk (8-12 points). Seventy-one patients with MDR-DA-HAI (82.6%) were classified in the intermediate or high-risk group, with a global sensitivity of 82.6%. The specificity in the high-risk group was 91.8%, and 81.0% of patients who were stratified into the low-risk group had non-MDR-associated infections, so they were correctly classified. Conclusions: The MDR score can be a useful tool to stratify patients in risk groups for MDR-DA-HAI. It may help to guide the choice of empirical therapy, leading to early optimization and avoiding delays in establishing appropriate treatment. This study reinforces the importance of stratifying patients based on their individual risk profile for MDR infection.