This observational study aims to evaluate the prognostic significance of the combination of a clinical score (specifically, a score obtained through the measurement of multiple parameters such as blood pressure, body temperature, etc., typically assessed during triage), called NEWS2, with certain laboratory tests in patients presenting with a suspected infection to our Emergency Department. We decided to conduct this study because the presentation symptoms of an infection are varied and sometimes subtle, making it very difficult to immediately recognize an infection in patients arriving at the Emergency Department. It is even more challenging to identify those at higher risk for negative clinical outcomes. However, it has been shown that rapid identification and early, targeted therapy significantly improve prognosis, increasing the chances of patient recovery.

The study aims to identify a tool that can help physicians to early recognize patients with infection, particularly those at higher risk for an unfavorable course. To this end, we are evaluating the effectiveness of combining a clinical score, called NEWS2, with a laboratory parameter (procalcitonin).

It is anticipated that approximately 1,000-1,200 people will participate in this study over the course of one year.

The study plans to involve all patients presenting to the Emergency Department with a suspected infection. The diagnostic and therapeutic process will remain unchanged from the standard clinical practice applied according to international guidelines, and no follow-up assessments will be required in the subsequent months

In the literature, so far, for patients presenting to the Emergency Department with suspected infection, no clinical score for risk stratification has proven to be significantly superior to others, although NEWS2 seems to have the best performance. Additionally, no clinical study has evaluated the role of combining this score with procalcitonin. In the oncological subpopulation, the role of procalcitonin (PCT) is less investigated and more debated compared to the general population. This study aims to identify a valid tool for the early recognition of patients arriving at the Emergency Department with suspected infection who are at higher risk of negative clinical outcomes, both in the general population and in the oncological subpopulation.

During the Emergency Department visit, the NEWS2 clinical score will be calculated, and the patient will be treated according to the usual diagnostic and therapeutic protocol, which also includes procalcitonin measurement and SOFA score calculation (which, if ≥ 2, allows for the identification of sepsis, i.e., infection with organ damage). The correlation of the NEWS2 + procalcitonin association with in-hospital mortality, the percentage of admissions to intensive/sub-intensive care, and the length of hospital stay will then be assessed. These analyses will be conducted by subdividing the population based on the presence or absence of active neoplasm. Summary tables will report and describe the characteristics of the patients enrolled in the study for both the total cohort and the sub-cohorts (oncological and non-oncological patients).

Descriptive statistical analyses will be used to summarize the collected data: continuous variables will be summarized using means, standard deviations, minimum and maximum values, and percentiles; discrete or nominal variables will be reported using absolute frequencies and relative percentages. The distribution of continuous variables will be assessed through graphical visualization and the Kolmogorov-Smirnov correction test. Depending on the characteristics of the distribution, Chi-Square tests, Student's T-test, Mann-Whitney U test, and Wilcoxon tests will be used.

To address the primary objectives of the study, the following methodologies will be employed: the correlation between serum procalcitonin levels and NEWS2 will be examined using box plots. To assess whether procalcitonin may have independent incremental prognostic value from NEWS2 variables, a logistic regression model will be utilized. Procalcitonin values will be entered into the model as predefined ordinal scales to facilitate clinical application. We will then compare the predictive value of NEWS2 and NEWS2 + procalcitonin ordinal class. The predictive value of each model will first be calculated using the C-statistic or AUROC. Kaplan-Meier curves will also be used to evaluate the performance of procalcitonin and NEWS2.

To address the secondary objectives of the study, ROC analysis and Kaplan-Meier curves will be utilized. Data will be processed using IBM SPSS statistical software (version 28.0), and a p-value < 0.05 will be considered statistically significant.

Inclusion Criteria:

• Patients presenting to the general Emergency Department of the IRCCS AOU of Bologna - S. Orsola Polyclinic for fever and/or other signs or symptoms of infection (for a period of 12 months following the date of study approval by the ethics committee and subsequent authorization by the General Director)
• Age ≥ 18 years - Obtaining informed consent (if possible, as specified in the protocol)

Exclusion Criteria:

• Patients with confirmed or suspected viral infections
• Pregnancy

Patients presenting to the general Emergency Department for fever and/or other signs or symptoms of infection