Background: Combination therapy in the treatment of sepsis, especially the value of combining a beta-Lactam antibiotic with an aminoglycoside, has been discussed. This retrospective cohort study including patients with sepsis or septic shock aimed to investigate whether one single dose of gentamicin at admittance (SGA) added to beta-Lactam antibiotic could result in a lower risk of mortality than beta-Lactam monotherapy, without exposing the patient to the risk of nephrotoxicity.

Methods and findings: All patients with positive blood cultures were evaluated for participation (n = 1318). After retrospective medical chart review, a group of patients with community-acquired sepsis with positive blood cultures who received beta-Lactam antibiotic with or without the addition of SGA (n = 399) were included for the analysis. Mean age was 74.6 yrs. (range 19-98) with 216 (54%) males. Sequential Organ Failure Assessment score (SOFA score) median was 3 (interquartile range [IQR] 2-5) and the median Charlson Comorbidity Index for the whole group was 2 (IQR 1-3). Sixty-seven (67) patients (17%) had septic shock. The 28-day mortality in the combination therapy group was 10% (20 of 197) and in the monotherapy group 22% (45 of 202), adjusted HR 3.5 (95% CI (1.9-6.2),p= < 0.001. No significant difference in incidence of acute kidney injury (AKI) was detected.

Conclusion: This retrospective observational study including patients with community-acquired sepsis or septic shock and positive blood cultures, who meet Sepsis-3 criteria, shows that the addition of one single dose of gentamicin to beta-lactam treatment at admittance was associated with a decreased risk of mortality and was not associated with AKI. This antibiotic regime may be an alternative to broad-spectrum antibiotic treatment of community-acquired sepsis. Further prospective studies are warranted to confirm these results.

Blood stream infections are associated with high mortality and morbidity. NEWS2 is a quick scoring system including bedside measurable vital signs. This study aimed to investigate the ability of NEWS2 ≥ 5p to identify sepsis, per Sepsis-3 criteria, among adult patients with community-acquired infection and positive blood cultures. It also explored if NEWS2 ≥ 5p could indicate infection etiology based on bacterial species in blood culture. This retrospective study included 555 patients with positive blood cultures. 425 of 555 (76.6%) patients had sepsis. The sensitivity of NEWS2 ≥ 5p for detecting sepsis was 86.6% and was not statistically associated with infection etiology. Patients with S. pneumoniae had a higher median NEWS2 score than those with other bacterial species. The 28-day mortality rate was 12.1%, and the sensitivity of NEWS2 ≥ 5p for detecting 28-day mortality was 91.0%. NEWS2 ≥ 5p was detected in a high proportion of sepsis cases among patients with blood stream infections, independent of bacterial species, and is a quick tool for identifying high sepsis likelihood in the emergency department.

OBJECTIVES: Optimal antibiotic treatment is important in the treatment of sepsis. However, patients with sepsis are at risk of suboptimal antibiotic concentrations. This study aimed to evaluate β-lactam antibiotic concentrations during the first 48 h in patients with community-acquired sepsis admitted to the ICU, and to identify variables associated with antibiotic concentrations that were too low or too high.

METHODS: This prospective, observational, single-centre study included patients aged ≥18 years with a high likelihood of infection, a SOFA score of ≥2p, planned β-lactam antibiotic treatment, and ICU admission. The exclusion criteria were ongoing antibiotic treatment and/or nosocomial infections. β-Lactam concentrations were measured up to seven times during the first 48 h. The estimated trough concentrations were divided by the predetermined MIC to generate MIC-multiples for comparison. Patients were allocated to three groups based on the MIC-multiple (MIC× < 1, 1-8 or >8).

RESULTS: Fifty patients were included, with a median of seven samples per patient (257 samples). The group with MIC-multiples of <1 (n = 16) was associated with younger age, lower Charlson comorbidity index, Simplified Acute Physiology Score 3, creatinine concentration, and need for noradrenaline. The group with MIC-multiples of >8 (n = 15) had higher creatinine and noradrenaline levels.

CONCLUSIONS: ICU patients with sepsis are at risk of either too low or too high antibiotic concentrations, and specific patient characteristics may be predictable. Therapeutic drug monitoring in combination with model-informed precision dosing may also help to optimize antibiotic dosing in the early phase of community-acquired sepsis to prevent treatment failure and toxicity.