Identifying Predictors Associated with Risk of Death or Admission to Intensive Care Unit in Internal Medicine Patients with Sepsis: A Comparison of Statistical Models and Machine Learning Algorithms

Mirijello, Antonio; Fontana, Andrea; Greco, Alessandra; Tosoni, Alberto; D’Agruma, Angelo; Labonia, Maria; Copetti, Massimiliano; Cosmo, Salvatore De

doi:10.3390/antibiotics12050925

Cited by 6 publications

(3 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Having shown that MR-proADM correlated with organ damage, we also tested its correlation with prognosis, which was significant. This reinforces data already described in the literature [6,7,[17][18][19][20][21][22][23]49,[54][55][56][57][58][59][60][61][62].…”

Section: Mr-proadm Levels and Organ Damage And Outcomesupporting

confidence: 91%

Mid-Regional Pro-Adrenomedullin Can Predict Organ Failure and Prognosis in Sepsis?

Spoto,

Basili,

Cangemi

et al. 2023

IJMS

View full text Add to dashboard Cite

Sepsis causes immune dysregulation and endotheliitis, with an increase in mid-regional pro-adrenomedullin (MR-proADM). The aim of the study is to determine an MR-proADM value that, in addition to clinical diagnosis, can identify patients with localized infection or those with sepsis/septic shock, with specific organ damage or with the need for intensive care unit (ICU) transfer and prognosis. The secondary aim is to correlate the MR-proADM value with the length of stay (LOS). In total, 301 subjects with sepsis (124/301 with septic shock) and 126 with localized infection were retrospectively included. In sepsis, MR-proADM ≥ 3.39 ng/mL identified acute kidney injury (AKI); ≥2.99 ng/mL acute respiratory distress syndrome (ARDS); ≥2.28 ng/mL acute heart failure (AHF); ≥2.55 ng/mL Glascow Coma Scale (GCS) < 15; ≥3.38 multi-organ involvement; ≥3.33 need for ICU transfer; ≥2.0 Sequential Organ Failure Assessment (SOFA) score ≥ 2; and ≥3.15 ng/mL non-survivors. The multivariate analysis showed that MR-proADM ≥ 2 ng/mL correlates with AKI, anemia and SOFA score ≥ 2, and MR-proADM ≥ 3 ng/mL correlates with AKI, GCS < 15 and SOFA score ≥ 2. A correlation between mortality and AKI, GCS < 15, ICU transfer and cathecolamine administration was found. In localized infection, MR-proADM at admission ≥ 1.44 ng/mL identified patients with AKI; ≥1.0 ng/mL with AHF; and ≥1.44 ng/mL with anemia and SOFA score ≥ 2. In the multivariate analysis, MR-proADM ≥ 1.44 ng/mL correlated with AKI, anemia, SOFA score ≥ 2 and AHF. MR-proADM is a marker of oxidative stress due to an infection, reflecting severity proportionally to organ damage.

show abstract

Section: Mr-proadm Levels and Organ Damage And Outcomesupporting

confidence: 91%

Mid-Regional Pro-Adrenomedullin Can Predict Organ Failure and Prognosis in Sepsis?

Spoto,

Basili,

Cangemi

et al. 2023

IJMS

View full text Add to dashboard Cite

show abstract

“…Moreover, it has been witnessed that there is virtually no benefit of using more complex machine learning methods over logistic regression ( 37 ). In particular, logistic regression has shown good calibration compared to other approaches ( 38 ).…”

Section: Methods and Datamentioning

confidence: 99%

Dissecting contributions of individual systemic inflammatory response syndrome criteria from a prospective algorithm to the prediction and diagnosis of sepsis in a polytrauma cohort

2023

View full text Add to dashboard Cite

BackgroundSepsis is the leading cause of death in intensive care units (ICUs), and its timely detection and treatment improve clinical outcome and survival. Systemic inflammatory response syndrome (SIRS) refers to the concurrent fulfillment of at least two out of the following four clinical criteria: tachycardia, tachypnea, abnormal body temperature, and abnormal leukocyte count. While SIRS was controversially abandoned from the current sepsis definition, a dynamic SIRS representation still has potential for sepsis prediction and diagnosis.ObjectiveWe retrospectively elucidate the individual contributions of the SIRS criteria in a polytrauma cohort from the post-surgical ICU of University Medical Center Mannheim (Germany).MethodsWe used a dynamic and prospective SIRS algorithm tailored to the ICU setting by accounting for catecholamine therapy and mechanical ventilation. Two clinically relevant tasks are considered: (i) sepsis prediction using the first 24 h after admission to our ICU, and (ii) sepsis diagnosis using the last 24 h before sepsis onset and a time point of comparable ICU treatment duration for controls, respectively. We determine the importance of individual SIRS criteria by systematically varying criteria weights when summarizing the SIRS algorithm output with SIRS descriptors and assessing the classification performance of the resulting logistic regression models using a specifically developed ranking score.ResultsOur models perform better for the diagnosis than the prediction task (maximum AUROC 0.816 vs. 0.693). Risk models containing only the SIRS level average mostly show reasonable performance across criteria weights, with prediction and diagnosis AUROCs ranging from 0.455 (weight on leukocyte criterion only) to 0.693 and 0.619 to 0.800, respectively. For sepsis prediction, temperature and tachypnea are the most important SIRS criteria, whereas the leukocytes criterion is least important and potentially even counterproductive. For sepsis diagnosis, all SIRS criteria are relevant, with the temperature criterion being most influential.ConclusionSIRS is relevant for sepsis prediction and diagnosis in polytrauma, and no criterion should a priori be omitted. Hence, the original expert-defined SIRS criteria are valid, capturing important sepsis risk determinants. Our prospective SIRS algorithm provides dynamic determination of SIRS criteria and descriptors, allowing their integration in sepsis risk models also in other settings.

show abstract

“…Furthermore, MR-proADM acts as a biomarker for sepsis, reflecting the degree of oxidative stress and indicating the severity of the condition in line with organ damage and prognosis [ 120 , 124 ]. From a clinical perspective, MR-pro-ADM can accurately detect patients with one or multiple organ failures, particularly showing a stronger correlation for certain variables such as acute kidney injury (AKI), GCS < 15, and the requirement for intensive care unit (ICU) transfer indicating a more severe prognosis [ 19 , 20 , 120 , 125 , 126 , 127 , 128 , 129 ].…”

Section: Pathophysiology Of Endothelitis: Adrenomedullin Expressionmentioning

confidence: 99%

A Focus on the Pathophysiology of Adrenomedullin Expression: Endothelitis and Organ Damage in Severe Viral and Bacterial Infections

Spoto,

Basili,

Cangemi

et al. 2024

Cells

View full text Add to dashboard Cite

Adrenomedullin (ADM) is a peptide hormone produced primarily in the adrenal glands, playing a crucial role in various physiological processes. As well as improving vascular integrity and decreasing vascular permeability, ADM acts as a vasodilator, positive inotrope, diuretic, natriuretic and bronchodilator, antagonizing angiotensin II by inhibiting aldosterone secretion. ADM also has antihypertrophic, anti-apoptotic, antifibrotic, antioxidant, angiogenic and immunoregulatory effects and antimicrobial properties. ADM expression is upregulated by hypoxia, inflammation-inducing cytokines, viral or bacterial substances, strength of shear stress, and leakage of blood vessels. These pathological conditions are established during systemic inflammation that can result from infections, surgery, trauma/accidents or burns. The ability to rapidly identify infections and the prognostic, predictive power makes it a valuable tool in severe viral and bacterial infections burdened by high incidence and mortality. This review sheds light on the pathophysiological processes that in severe viral or bacterial infections cause endothelitis up to the development of organ damage, the resulting increase in ADM levels dosed through its more stable peptide mid-regional proadrenomedullin (MR-proADM), the most significant studies that attest to its diagnostic and prognostic accuracy in highlighting the severity of viral or bacterial infections and appropriate therapeutic insights.

show abstract

Identifying Predictors Associated with Risk of Death or Admission to Intensive Care Unit in Internal Medicine Patients with Sepsis: A Comparison of Statistical Models and Machine Learning Algorithms

Cited by 6 publications

References 41 publications

Mid-Regional Pro-Adrenomedullin Can Predict Organ Failure and Prognosis in Sepsis?

Mid-Regional Pro-Adrenomedullin Can Predict Organ Failure and Prognosis in Sepsis?

Dissecting contributions of individual systemic inflammatory response syndrome criteria from a prospective algorithm to the prediction and diagnosis of sepsis in a polytrauma cohort

A Focus on the Pathophysiology of Adrenomedullin Expression: Endothelitis and Organ Damage in Severe Viral and Bacterial Infections

Contact Info

Product

Resources

About