Prehospital antibiotics for sepsis: beyond mortality?

Impact of Prehospital Antibiotic Therapy on Septic Shock Mortality

BACKGROUND

Septic shock (SS) is associated with high morbidity and mortality rate. Early antibiotic therapy administration in septic patients was shown to reduce mortality but its impact on mortality in a prehospital setting is still under debate. To clarify this point, we performed a retrospective analysis on patients with septic shock who received antibiotics in a prehospital setting. Methods: From April 15th, 2017 to March 1st, 2020, patients with septic shock requiring Mobile Intensive Care Unit (MICU) intervention were retrospectively analyzed to assess the impact of prehospital antibiotic therapy administration on a 30-day mortality. Results: Three-hundred-eight patients with septic shock requiring MICU intervention in the prehospital setting were analyzed. The mean age of the study population was 70 ± 15 years. Presumed origin of SS was mainly pulmonary (44%), digestive (21%) or urinary (19%) infection. Overall 30-day mortality was 29%. Ninety-eight (32%) patients received antibiotic therapy. Using Cox regression analysis, we showed that prehospital antibiotic therapy significantly reduces 30-day mortality for patients with septic shock (hazard ratio = 0.56, 95%CI [0.35-0.89], p = 0.016). Conclusion: In this retrospective study, prehospital antibiotic therapy reduces 30-day mortality of septic shock patients cared for by MICU. Further studies will be needed to confirm the beneficial effect of prehospital antibiotic therapy in association or not with prehospital hemodynamic optimization to improve the survival of septic shock patients.

Timeliness of antibiotics for patients with sepsis and septic shock

For many years, sepsis guidelines have focused on early administration of antibiotics. While this practice may benefit some patients, for others it might have detrimental consequences. The increasingly shortened timeframes in which administration of antibiotics is recommended, have forced physicians to sacrifice diagnostic accuracy for speed, encouraging the overuse of antibiotics. The evidence supporting this practice is based on retrospective data, with all the limitations attached, while the only randomized trial on this subject does not show a mortality benefit from early administration of antibiotics in a population of patients with sepsis as often seen in the emergency department (ED). Physicians are challenged to treat patients suspected of having sepsis within a short period of time, while the real challenge should be to identify patients who would not be harmed by withholding treatment with antibiotics until the diagnosis of infection with a bacterial origin is confirmed and the appropriateness of a course of antibiotics can be evaluated more adequately. Therefore, in the general population of patients with sepsis, taking the time to gather additional data to confirm the diagnosis should be encouraged without a specific timeframe, although physicians should be encouraged to perform an adequate work-up as soon as possible. Patients with suspected sepsis and signs of shock should immediately be treated with antibiotics, as there is no margin for error.

Clinical applications of artificial intelligence in sepsis: A narrative review

Many studies have been published on a variety of clinical applications of artificial intelligence (AI) for sepsis, while there is no overview of the literature. The aim of this review is to give an overview of the literature and thereby identify knowledge gaps and prioritize areas with high priority for further research. A literature search was conducted in PubMed from inception to February 2019. Search terms related to AI were combined with terms regarding sepsis. Articles were included when they reported an area under the receiver operator characteristics curve (AUROC) as outcome measure. Fifteen articles on diagnosis of sepsis with AI models were included. The best performing model reached an AUROC of 0.97. There were also seven articles on prognosis, predicting mortality over time with an AUROC of up to 0.895. Finally, there were three articles on assistance of treatment of sepsis, where the use of AI was associated with the lowest mortality rates. Of the articles, twenty-two were judged to be at high risk of bias or had major concerns regarding applicability. This was mostly because predictor variables in these models, such as blood pressure, were also part of the definition of sepsis, which led to overestimation of the performance. We conclude that AI models have great potential for improving early identification of patients who may benefit from administration of antibiotics. Current AI prediction models to diagnose sepsis are at major risks of bias when the diagnosis criteria are part of the predictor variables in the model. Furthermore, generalizability of these models is poor due to overfitting and a lack of standardized protocols for the construction and validation of the models. Until these problems have been resolved, a large gap remains between the creation of an AI algorithm and its implementation in clinical practice.