Early sepsis prediction in elderly patients with urinary tract infections: a machine learning

Background: Sepsis remains a leading cause of morbidity and mortality, particularly among elderly patients, for whom urinary tract infections (UTIs) are a significant trigger. This study applies machine learning methods to identify early predictors of sepsis in elderly patients presenting with UTIs in the emergency department (ED). Methods: This retrospective study analyzed elderly patients with UTIs over five years, excluding those with sepsis at presentation, secondary infections or incomplete records. Logistic regression, Generalized Additive Modeling (GAM), Least Absolute Shrinkage and Selection Operator (LASSO) regression, and a Decision Tree model were evaluated for sepsis prediction within 72 hours. Model performance was assessed using area under the curve (AUC), Brier scores, and Net Reclassification Improvement. Results: Of 1176 patients, 139 (11.8%) developed sepsis within 72 hours. Independent predictors included age (adjusted odds ratio (aOR) 1.10, 95% confidence interval (CI) 1.05–1.14), blood urea nitrogen (BUN) (aOR 1.03, 95% CI 1.00–1.05), C-reactive protein (CRP) (aOR 1.04, 95% CI 1.03–1.05), creatinine (aOR 1.79, 95% CI 1.11–3.14), respiratory rate (aOR 1.10, 95% CI 1.05–1.16), temperature (aOR 2.52, 95% CI 1.65–4.38), and lower systolic blood pressure (aOR 0.95, 95% CI 0.92–0.97). GAM (AUC 0.954) and LASSO (AUC 0.942) outperformed logistic regression (AUC 0.792, p < 0.001). GAM showed superior discrimination over the Decision Tree (AUC 0.915, p = 0.046). Conclusions: This study highlights that clinical parameters including age, BUN, CRP, creatinine, respiratory rate, body temperature and systolic blood pressure are independent risk factors for early sepsis in elderly patients with UTIs. These factors should be carefully considered when assessing elderly patients presenting to the ED for sepsis risk.

Sepsis prediction; Machine learning; Artificial intelligence; Urinary tract infections; Elderly patients; Biomarkers; Logistic regression

[1] Chiscano-Camón L, Ruiz-Sanmartin A, Bajaña I, Bastidas J, Lopez-Martinez R, Franco-Jarava C, et al. Current perspectives in the management of sepsis and septic shock. Frontiers in Medicine. 2024; 11: 1431791.

[2] Kule A, Stassen W, Flores GE, Djarv T, Singletary E, Kule A, et al. Recognition and awareness of sepsis by first-aid providers in adults with suspected infection: a scoping review. Cureus. 2024; 16: e61612.

[3] Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M, et al. The third international consensus definitions for sepsis and septic shock (Sepsis-3). JAMA. 2016; 315: 801–810.

[4] Ibarz M, Haas LEM, Ceccato A, Artigas A. The critically ill older patient with sepsis: a narrative review. Annals of Intensive Care. 2024; 14: 6.

[5] Wang X, Yang C, Jiang J, Hu Y, Hao Y, Dong JY. Polypharmacy, chronic kidney disease, and mortality among older adults: a prospective study of National Health and nutrition examination survey, 1999–2018. Frontiers in Public Health. 2023; 11: 1116583.

[6] Andaluz D, Ferrer R. SIRS, qSOFA, and organ failure for assessing sepsis at the emergency department. Journal of Thoracic Disease. 2017; 9: 1459–1462.

[7] Dutta C, Pasha K, Paul S, Abbas MS, Nassar ST, Tasha T, et al. Urinary tract infection induced delirium in elderly patients: a systematic review. Cureus. 2022; 14: e32321.

[8] Zhen J, Li J, Liao F, Zhang J, Liu C, Xie H, et al. Development and validation of machine learning models for young-onset colorectal cancer risk stratification. Npj Precision Oncology. 2024; 8: 1–14.

[9] Mittas N, Chatzopoulou F, Kyritsis KA, Papagiannopoulos CI, Theodoroula NF, Papazoglou AS, et al. A risk-stratification machine learning framework for the prediction of coronary artery disease severity: insights from the GESS trial. Frontiers in Cardiovascular Medicine. 2022; 8: 812182.

[10] van Doorn WPTM, Helmich F, van Dam PMEL, Jacobs LHJ, Stassen PM, Bekers O, et al. Explainable machine learning models for rapid risk stratification in the emergency department: a multicenter study. The Journal of Applied Laboratory Medicine. 2024; 9: 212–222.

[11] Lawati HA, Blair BM, Larnard J. Urinary tract infections: core curriculum 2024. American Journal of Kidney Diseases. 2024; 83: 90–100.

[12] Vincent JL, Moreno R, Takala J, Willatts S, De Mendonça A, Bruining H, et al. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. Intensive Care Medicine. 1996; 22: 707–710.

[13] Bujang MA, Sa’at N, Sidik TMITAB, Joo LC. Sample size guidelines for logistic regression from observational studies with large population: emphasis on the accuracy between statistics and parameters based on real life clinical data. The Malaysian Journal of Medical Sciences. 2018; 25: 122.

[14] He Y. Missing data analysis using multiple imputation: getting to the heart of the matter. Circulation Cardiovascular Quality and Outcomes. 2010; 3: 98–105.

[15] Liu A, Bui T, Van Nguyen H, Ong B, Shen Q, Kamalasena D. Serum C-reactive protein as a biomarker for early detection of bacterial infection in the older patient. Age and Ageing. 2010; 39: 559–565.

[16] Tang Y, Jiang J, Zhao Y, Du D. Aging and chronic kidney disease: epidemiology, therapy, management and the role of immunity. Clinical Kidney Journal. 2024; 17: sfae235.

[17] Francis A, Harhay MN, Ong ACM, Tummalapalli SL, Ortiz A, Fogo AB, et al. Chronic kidney disease and the global public health agenda: an international consensus. Nature Reviews Nephrology. 2024; 20: 473–485.

[18] Maringhini S, Zoccali C. Chronic kidney disease progression—a challenge. Biomedicines. 2024; 12: 2203.

[19] Jiang X, Zhang C, Pan Y, Cheng X, Zhang W. Effects of C-reactive protein trajectories of critically ill patients with sepsis on in-hospital mortality rate. Scientific Reports. 2023; 13: 15223.

[20] Zincircioğlu Ç, Rollas K, Güldoğan IK, Sarıtaş A, Özkarakaş H, Ersan G, et al. Diagnostic value of procalcitonin and C reactive protein for infection and sepsis inelderly patients. Turkish Journal of Medical Sciences. 2021; 51: 2649–2656.

[21] Nasa P, Juneja D, Singh O. Severe sepsis and septic shock in the elderly: an overview. World Journal of Critical Care Medicine. 2012; 1: 23.

[22] Rowe TA, McKoy JM. Sepsis in older adults. Infectious Disease Clinics. 2017; 31: 731–742.

[23] Shimazui T, Nakada T, Walley KR, Oshima T, Abe T, Ogura H, et al. Significance of body temperature in elderly patients with sepsis. Critical Care. 2020; 24: 387.

[24] Rumbus Z, Matics R, Hegyi P, Zsiboras C, Szabo I, Illes A, et al. Fever is associated with reduced, hypothermia with increased mortality in septic patients: a meta-analysis of clinical trials. PLOS ONE. 2017; 12: e0170152.

[25] Doman M, Thy M, Dessajan J, Dlela M, Do Rego H, Cariou E, et al. Temperature control in sepsis. Frontiers in Medicine. 2023; 10: 1292468.

[26] Riley RD, Ensor J, Snell KIE, Harrell FE, Martin GP, Reitsma JB, et al. Calculating the sample size required for developing a clinical prediction model. The BMJ. 2020; 368: m441.