Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Jan 30:12:1503868.
doi: 10.3389/fmed.2025.1503868. eCollection 2025.

Sepsis in Internal Medicine: blood culture-based subtypes, hospital outcomes, and predictive biomarkers

Gaetano Zizzo  1 , Gabriele Guazzardi  2 , Daniela Bompane  1 , Francesco Di Terlizzi  1 , Giorgio Rotola  1 , Ilario Stefani  1 , Michela Medugno  3 , Mario Bucalo  2 , Antonino Mazzone  1
Affiliations

Sepsis in Internal Medicine: blood culture-based subtypes, hospital outcomes, and predictive biomarkers

Gaetano Zizzo et al. Front Med (Lausanne). .

Abstract

Background: Sepsis is a challenging condition increasingly managed in medical wards, however literature and clinical evidence in this hospital setting are lacking.

Methods: Using the computational i2b2 framework, we retrospectively analyzed data from patients admitted to internal medicine units of four hospitals in Lombardy (Italy) between January 2012 and December 2023, with a discharge diagnosis of sepsis, septic shock, or septicemia.

Results: A total of 4,375 patients were recruited. Median length of stay (LOS) was 14 days, and mean ward-to-intensive care unit (ICU) transfer and in-hospital mortality rates were 11 and 26%, respectively; significant differences were observed over the years, with LOS peaks preceding mortality peaks by 1 year. Blood culture-negative sepses showed shorter stays and higher mortality (acute kidney injury and fast deterioration) compared to culture-positive ones; polymicrobial sepses showed higher ICU transfer rates (acute respiratory distress); while multidrug-resistant (MDR+) and/or polymicrobial sepses showed longer stays and higher mortality (complicated course) compared to drug-sensitive or monomicrobial ones. C-reactive protein elevation predicted rapidly evolving culture-negative sepsis, whereas lower leukocyte counts predicted prolonged hospitalization; higher fractions of inspired oxygen predicted polymicrobial sepsis, while lactate elevation predicted ICU transfer; ferritin elevation and increased leukocyte counts predicted MDR+ sepsis, while further ferritin elevation and decreased platelet counts predicted death. From 2016 to 2023, MDR+ sepsis frequency declined, due to decreased resistance to several antibiotic classes, such as cephalosporins, fluoroquinolones, and aminoglycosides; however, carbapenemase- and extended-spectrum beta-lactamase-producing Gram-negative bacteria, as well as vancomycin-resistant enterococci, increased, as did the frequency of polymicrobial sepsis following the COVID-19 outbreak.

Conclusion: This work provides novel insights into sepsis management in internal medicine units, highlighting the need for validated biomarkers and implemented therapies in this scenario.

Keywords: COVID-19; Sepsis; biomarkers; culture-negative; internal medicine; multidrug-resistant (MDR); outcomes; polymicrobial.

PubMed Disclaimer

Conflict of interest statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Figures

Figure 1
Figure 1
Hospital outcomes for sepsis in medical wards between 2012 and 2023 (n = 4,375 patients). (A) Length of stay (LOS). **From 2012 to 2023, p = 0.0017 (Kruskal–Wallis test). ***2015 vs. 2012, p = 0.0001. *2015 vs. 2022, p = 0.0449. **2017 vs. 2012, p = 0.0044. **2021 vs. 2012, p = 0.0021 (Dunn’s multiple comparisons). (B) Ward-to-ICU transfer. **From 2012 to 2023, p = 0.0423 (chi-square). ***2016 vs. 2013, p = 0.0003. **2016 vs. 2019, p = 0.0010. **2016 vs. 2014, p = 0.0048. *2016 vs. 2012, p = 0.0168. *2016 vs. 2020, p = 0.0239. *2016 vs. 2015, p = 0.0257. *2016 vs. 2021, p = 0.0293. *2016 vs. 2023, p = 0.0471. *2016 vs. 2018, p = 0.0491. *2022 vs. 2013, p = 0.0254 (Fisher’s exact test). (C) In-hospital mortality. **From 2012 to 2023, p = 0.0031 (chi-square). ***2016 vs. 2019, p = 0.0009. **2016 vs. 2021, p = 0.0031. **2016 vs. 2023, p = 0.0071. *2016 vs. 2017, p = 0.0249. **2018 vs. 2019, p = 0.0030. *2018 vs. 2021, p = 0.0101. *2018 vs. 2023, p = 0.0252. **2013 vs. 2019, p = 0.0038. *2013 vs. 2021, p = 0.0111. *2013 vs. 2023, p = 0.0257. **2012 vs. 2019, p = 0.0050. *2012 vs. 2021, p = 0.0116. *2012 vs. 2023, p = 0.0323. *2022 vs. 2019, p = 0.0239. *2022 vs. 2021, p = 0.0453 (Fisher’s exact test).
Figure 2
Figure 2
Differences in sepsis types and hospital outcomes for sepsis between before and after the outbreak of the COVID-19 pandemic (i.e., 2016–2019 vs. 2020–2023) (n = 2,907 patients). *p < 0.05, ***p < 0.001, and ****p < 0.0001; ns, not significant.
Figure 3
Figure 3
Antibiotic resistance trends in blood-isolated Gram-negative bacteria and enterococci. (A) Cephalosporin (ceftazidime) resistance (n = 368 antibiograms). *2016 vs. 2020, p = 0.0453. 2016 vs. 2022, p = 0.0513. 2018 vs. 2020, p = 0.0742. 2018 vs. 2022, p = 0.0790 (Fisher’s exact test). (B) Fluoroquinolone (ciprofloxacin) resistance (n = 324). *From 2016 to 2023, p = 0.0203 (chi-square). *2017 vs. 2020, p = 0.0153. *2017 vs. 2023, p = 0.0161. 2017 vs. 2022, p = 0.0528. ***2016–2019 vs. 2020–2023, p = 0.0007 (Fisher’s exact test). (C) Aminoglycoside (gentamicin) resistance (n = 351). *From 2016 to 2023, p = 0.0140 (chi-square). ***2016 vs. 2022, p = 0.0008. *2016 vs. 2020, p = 0.0138. *2016 vs. 2018, p = 0.0420. *2016 vs. 2019, p = 0.0485. *2021 vs. 2022, p = 0.0244. 2019 vs. 2022, p = 0.0544. 2017 vs. 2022, p = 0.0690. *2016–2019 vs. 2020–2023, p = 0.0484 (Fisher’s exact test). (D) Ureidopenicillin + β-lactamase (piperacillin/tazobactam) resistance (including extended spectrum β-lactamase producing bacteria or ESBL+) (n = 367). 2023 vs. 2020, p = 0.0568. 2023 vs. 2019, p = 0.0828 (Fisher’s exact test). (E) Carbapenem (ertapenem) resistance (including Klebsiella pneumoniae carbapenemase-producing bacteria or KPC) (n = 336). *2023 vs. 2022, p = 0.0400. 2023 vs. 2019, p = 0.0842. 2016 vs. 2022, p = 0.0558 (Fisher’s exact test). (F) Glicopeptyde (vancomycin) resistant enterococci (VRE) (n = 177).
See this image and copyright information in PMC

References

    1. Reinhart K, Daniels R, Kissoon N, Machado FR, Schachter RD, Finfer S. (2017). Recognizing sepsis as a global health priority—a WHO resolution. N Engl J Med 377:414–7. doi: 10.1056/NEJMp1707170, PMID: - DOI - PubMed
    1. Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M, et al. (2016). The third international consensus definitions for sepsis and septic shock (sepsis-3). JAMA 315:801–10. doi: 10.1001/jama.2016.0287, PMID: - DOI - PMC - PubMed
    1. Suarez De La Rica A, Gilsanz F, Maseda E. (2016). Epidemiologic trends of sepsis in western countries. Ann Transl Med 4:325. doi: 10.21037/atm.2016.08.59, PMID: - DOI - PMC - PubMed
    1. Fleischmann C, Scherag A, Adhikari NK, Hartog CS, Tsaganos T, Schlattmann P, et al. (2016). Assessment of global incidence and mortality of hospital-treated sepsis. Current estimates and limitations. Am J Respir Crit Care Med 193:259–72. doi: 10.1164/rccm.201504-0781OC, PMID: - DOI - PubMed
    1. World Health Organization . Global report on the epidemiology and burden of sepsis: current evidence, identifying gaps and future directions. Geneva: World Health Organization; (2020).

LinkOut - more resources