Evaluation of the Performance of Sepsis Predictive Scores on the Elderly Population in the Emergency Department.

The main objective of the research is to evaluate the prognostic performance of qSOFA, NEWS and MEDS scores in patients over 65 years of age presenting to the emergency department with a diagnosis of infection made by the emergency physician in charge.

The main judgment criterion is intra-hospital mortality. Secondary criteria for judgement are admission to intensive care or intensive care (continuous monitoring unit), length of hospital stay, length of stay in the emergency room, door-to-antibiotic time. A comparison of score performance will be carried out between the population aged over 65 years old and the rest of the population.

1. EXPECTED FALLOUT. Early recognition of patients with potentially severe sepsis and adaptation to monitoring appropriate to severity will enable patients to be identified and managed more quickly from the time of assessment by the emergency reception and referral nurse. In this way, these patients can be integrated into a dedicated line of care enabling early antibiotic therapy, vascular filling and support by vasopressor amines if necessary. Older people present a particularly fragile population with physiological peculiarities. It is necessary to validate these prognostic scores in this segment of the population.
2. EXPERIMENTAL PLAN. 4.1 Design of the study. This study will be a monocentric, prospective, non-interventional study. The patients recruited will be those who's the emergency doctor in charge suspects an infection, either through clinical examination or complementary examinations (radiological, microbiological). The diagnosis of bacterial infection will be re-evaluated remotely by 2 experts once the follow-up phase is completed. In case of discrepancies, a consensus will be reached between the 2 experts. Patients with no confirmed infection will be excluded from the study. Investigators also exclude pregnant women, minors, patients under legal protection, or patients refusing to participate, prisoners and patients with localized infections without general repercussions (abscess, cystitis, angina). For each patient included, the reception and orientation nurse assisted by the emergency doctor will have to fill in the parameters needed to calculate qSOFA score, the NEWS score and the MEDS score on admission. The main judgment criterion is intra-hospital mortality at 28 days. Patients still hospitalized after D28 will be considered still alive. Secondary endpoints will be admission to intensive care (resuscitation/continuous monitoring unit) for more than 72 hours, length of hospital stay, length of stay in intensive care, door-to-antibiotic time and a composite endpoint of mortality or admission to intensive care for more than 72 hours. 4.2 Criteria for inclusion: Adult patient presenting to the adult emergency department with suspected infection by the reception and orientation nurse or diagnosed by the emergency physician. One subgroup is composed of patients aged 65 years and over, a second subgroup is composed of patients aged 18 to 64 years. 4.3 Exclusion criteria: Underage patients, vulnerable persons (legal protection) Pregnant women Prisoner Infection not confirmed after review of the files by the experts Localized infection without general repercussions: abscess, angina, cystitis. Patient's opposition to participating in the study Inclusion period: 6 months Follow-up period: 28 days Place of study: emergency department of the Tours CHRU (University Hospital Centre)
3. BENEFIT/RISK The study provides a reminder to practitioners of the international recommendations for sepsis management based on the Surviving Sepsis Campaign. It can thus be expected to improve the time taken by patients to receive antibiotic therapy and potentially improve the triage of these patients. There is no risk expected by this study since there is no change in practices. The objective of this study is to evaluate prognostic performance of scores without intervention on patient management.
4. STATISTICAL ANALYSES

Numerical variables will be expressed as average and 95% confidence interval, and categorical variables will be expressed as number and percentage. In the case of missing variables for scores, investigators assume that the values are within normal values. To achieve the performance of the qSOFA, NEWS and MEDS scores in predicting the main judgement criterion, investigators will calculate the diagnostic performance (sensitivity, specificity, positive and negative predictive values) for a qSOFA greater than or equal to 2, a NEWS ≥ 5 and a MEDS greater than or equal to 12. Investigators will construct a ROC curve and calculate an area under the curve for each score. The scores will be compared with each other in the population over 65 years old. Each score will also be compared between the population over 65 and the population 18-64.

A p value of less than 0.05 is considered significant. The analyses are performed using R software version 3.6.2.