Our study confirms the large variability in voriconazole trough plasma levels in children and a trend to non-linear pharmacokinetics in older patients. In addition, doses significantly higher than those recommended in younger children seem warranted and a significant relationship between plasma voriconazole above the normal range and some adverse events is confirmed.

There is an urgent need to reconsider the empirical therapy used in neonatal EOS, particularly in VLBW newborns. What is Known: • E. coli early-onset sepsis (EOS) and E. coli resistant strains have been described as overall stable but increasing in VLBW neonates (< 1.500 g) in previous studies. What is New: • Our study shows an increasing incidence of E. coli EOS in all age groups, overruling group B Streptoccocus for the last 10 years. E. coli resistant strains also increased equally in all age groups, with high resistance rates to our first line antibiotics (ampicillin and gentamicin). • Empiric antibiotic therapy of EOS, mainly in VLBW newborns, should be adapted to this new scenario.

The incidence of SAB tripled during the years studied but remained stable in the last period. Antimicrobial resistances did not increase. Although a decrease in mortality was documented, approximately half the 30-day cumulative mortality was caused by SAB.

What is known and objective Tocilizumab is an IL‐6 receptor inhibitor agent which has been proposed as a candidate to stop the inflammatory phase of infection by the severe acute respiratory syndrome coronavirus‐2 (SARS‐CoV‐2). However, safety data of tocilizumab in pregnant women and their newborn are scarce. We aimed to describe maternal and neonatal safety outcomes associated with tocilizumab treatment in pregnant women with severe COVID‐19. Methods This is a retrospective study of severe COVID‐19 pregnant women, treated with tocilizumab in two Spanish hospitals between 1 March and 31 April 2020. Demographics, medical history, clinical and radiologic findings, treatment information and laboratory data of mothers and their newborns were collected from electronic medical records. Results and discussion A total of 12 pregnant women were identified to have received tocilizumab during pregnancy in the two hospitals. Median gestational age at admission was 27.7 weeks (interquartile range, 18.0–36.4). Most of them received lopinavir/ritonavir, azithromycin and hydroxychloroquine, two patients received corticosteroids and one received interferon beta 1B. All 12 pregnancies resulted in live births. Somatometric values were normal for all newborns, and evolution at 14 and 28 days was favourable for all of them. Hepatotoxicity was observed in 2 patients, which improved or resolved at discharge. Cytomegalovirus reactivation was detected in another patient who had also received corticosteroids for 15 days, causing a congenital infection in her newborn. Both hepatotoxicity and viral reactivation adverse events were classified as possibly related to tocilizumab administration according to Naranjo's causality algorithm. What is new and conclusions It does not appear that tocilizumab has detrimental effects for the mother and newborn. Close monitoring of infections should be considered, especially if other immunosuppressive agents are used.