Plasma presepsin as a predictor of mortality in patients with UTI-related severe infection: a single-center retrospective study Authors: Ji Won Lee, Daun Song, Se-Hee Yoon, Sung-Ro Yun, Won Min Hwang, Yong Kyun Kim, Seung Ho Hur, Jae Kwang Lee, Sun Jung Kwon, Jieun Shin, Yohan Park Published: 2025 ABSTRACT AbstractBackgroundThis study aimed to evaluate whether plasma presepsin levels are a more reliable predictor of mortality than traditional infection biomarkers in patients with urinary tract infection (UTI)-related severe infection.MethodsThis single-center retrospective study evaluated 69 patients with UTI-related severe infection admitted to the emergency department between May 2022 and August 2023. Data on vital signs, plasma presepsin levels, procalcitonin (PCT) levels, C-reactive protein (CRP) levels, white blood cell (WBC) count, and other laboratory values at admission were collected. The values of presepsin, PCT, CRP, and WBC count for predicting 28-day mortality were analyzed. Survivors and non-survivors were propensity score-matched in a 2:1 ratio based on age, sex, and estimated glomerular filtration rate.ResultsIn the overall cohort, presepsin showed the highest area under the receiver operating characteristic curve (AUROC) of 0.716 for predicting 28-day mortality, surpassing that of PCT (0.641), CRP (0.488), and WBC count (0.433). Presepsin levels were independently associated with increased 28-day mortality risk (hazard ratio, 1.026; p = 0.008). In the matched cohort, presepsin levels showed the highest AUROC (0.660), followed by PCT (0.584), CRP (0.496), and WBC count (0.436).ConclusionsPresepsin levels were independently associated with increased 28-day mortality risk; this may be a valuable biomarker for identifying high-risk patients with UTI-related severe infection.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12894-025-01997-z.Keywords: Presepsin, Sepsis, Urinary tract infection, Mortality Background Urinary tract infections (UTIs) are among the most common infectious diseases encountered in both community and hospital settings [1]. UTI presents with a broad range of clinical symptoms, ranging from uncomplicated bacteriuria to life-threatening septic shock [2]. Prompt and accurate diagnosis is crucial because inadequate treatment can lead to significant morbidity and mortality due to severe complications. Rapid identification and evaluation of UTI sepsis are essential for determining the need for intensive care unit (ICU) admission and guiding appropriate antibiotic therapy [3]. During the initial evaluation, inflammatory markers such as white blood cell (WBC) count and C-reactive protein (CRP) and procalcitonin (PCT) levels are frequently used to assess the severity of infection. However, these markers have certain limitations. Although the WBC count is a standard indicator of infection, it has low specificity because it can be elevated in both infectious and noninfectious conditions [4]. Moreover, the WBC count does not effectively differentiate between lower and upper UTIs and does not correlate with the severity of kidney involvement [5]. CRP also has limited specificity, as its levels can be elevated in various inflammatory states and is unreliable for differentiating between lower and upper UTIs [6]. Although PCT is correlated with the degree of kidney involvement and its levels are commonly elevated in severe infections, it may be affected by non-infectious inflammatory conditions [7–9]. Presepsin is a relatively novel biomarker that has attracted attention owing to its potential use in the early detection of sepsis [10]. Presepsin is a soluble subtype of CD14 (sCD14-ST) generated through the proteolytic cleavage of CD14, a receptor expressed on macrophages and monocytes, and plays a crucial role in the immune response to bacterial infections [11]. Presepsin offers several advantages over traditional biomarkers, including rapid measurement and high sensitivity and specificity in the early stages of infection [11–13]. Moreover, recent studies suggest that presepsin has been shown to distinguish between upper and lower UTIs [14]. However, the role of presepsin in UTIs remains underexplored, and research on the relationship between presepsin levels and patient outcomes, such as mortality, is lacking. Thus, this study aimed to evaluate the utility of plasma presepsin levels as a predictor of mortality in patients with UTI-related severe infection. To this end, we compared the predictive value of plasma presepsin with that of traditional biomarkers, namely WBC count, CRP, and PCT, in patients with UTI-related severe infection who presented to the emergency department. Methods Study design and patients This was a single-center, retrospective study of patients who presented to the emergency department of our institution between May 1, 2022, and August 31, 2023. An initial cohort of 251 adult patients who underwent an infection work-up and had a serum presepsin sample collected was identified for screening. Inclusion criteria were age ≥ 18 years, emergency department presentation within the study window, and availability of presepsin and PCT measured at arrival, with UTI subsequently confirmed per prespecified criteria (see Definition of UTI and UTI-related severe infection). Exclusion criteria were absence of presepsin or PCT at arrival, identification of a non-urinary source during adjudication, or end-stage renal disease on maintenance dialysis. Of the 251 suspected infection encounters initially identified, 34 were excluded owing to the absence of presepsin or PCT measurements at admission, leaving 217 patients for analysis. In total, 106 patients without a confirmed UTI diagnosis were excluded. Of the remaining 111 patients with UTI, 39 who did not meet the criteria for UTI-related severe infection were excluded, along with 3 patients with end-stage renal disease on maintenance dialysis. Ultimately, 69 patients were included in the final analysis (Fig. 1). All patients in the final cohort received broad-spectrum antibiotics within one hour of arrival at the emergency department. The choice of initial empirical antibiotics was determined following consultation between the attending emergency medicine specialist and at least one nephrologist. Subsequent antibiotic therapy was then tailored according to the antimicrobial susceptibility of the isolated pathogens. In cases involving multidrug-resistant organisms, an infectious disease specialist was consulted to ensure appropriate treatment was administered (Additional file 1). Patient selection flowchart. In total, 251 patients admitted to the emergency department between May 2022 and August 2023 were screened for inclusion. After excluding 34 patients without presepsin or procalcitonin measurements and 106 patients without a confirmed UTI diagnosis, 111 patients remained. Of these, 39 who did not meet the UTI-related severe infection criteria and 3 with end-stage kidney disease were excluded. Finally, 69 patients were included in the study. UTI, urinary tract infection This study was approved by the Institutional Review Board (IRB) of Konyang University Hospital, College of Medicine, Konyang University (No. KYUH 2024-12-013). The IRB waived the need to obtain informed patient consent because of the retrospective nature of the study. Definition of UTI and UTI-related severe infection For this study, we defined and analyzed a cohort of patients with UTI-related severe infection. Inclusion in this cohort required patients to meet criteria for a UTI combined with evidence of a severe systemic response. A UTI was diagnosed based on any of the following: the identification of a causative pathogen in a urine culture; the presence of urinary tract symptoms accompanied by pyuria or bacteuria on urinalysis; or clinical findings suggestive of a UTI upon physical examination by an emergency medicine specialist or nephrologist [15]. Evidence of a severe systemic response was defined by the presence of at least one of the following criteria at the time of emergency department presentation: 1) A quick Sequential Organ Failure Assessment (qSOFA) score of 2 or greater according to the Sepsis-3 definitions. The qSOFA criteria include: systolic blood pressure ≤ 100 mmHg, respiratory rate ≥ 22 breaths per minute, or altered mental status (Glasgow Coma Scale [GCS] < 15), and 2) A requirement for vasopressor therapy to maintain a mean arterial pressure of 65mmHg or higher [16]. Additionally, patients with a confirmed UTI who also had bacteremia with a urinary source pathogen were included in this category. This definition was chosen to capture the full clinical spectrum of severe infection originating from a UTI, thereby including patients with sepsis as well as those with septic shock who require ICU care. Data collection and primary and secondary outcomes Baseline clinicopathological data, including age, sex, and comorbidities, were collected. Laboratory test results at the time of emergency department admission were collected, including complete blood cell count, arterial blood gas, electrolytes, liver enzymes, total protein, albumin, bilirubin, creatinine, arterial pH, CRP, and PCT levels. The mean arterial pressure, heart rate, respiratory rate, body temperature, partial pressure of oxygen in the arterial blood/fraction of inspired oxygen ratio, and GCS at admission were also recorded. For intubated patients, the GCS was assessed based on the approach described by Brennan et al. [17]. The estimated glomerular filtration rate (eGFR) was measured using the Chronic Kidney Disease Epidemiology Collaboration 2021 equation [18]. The primary outcome was 28-day mortality, defined as death within 28 days of hospital admission. The secondary outcome was overall mortality during the whole study period. Survival time was defined as the time from the date of admission to the date of death or the last follow-up. Patients who were alive at the end of the study period were censored at the last follow-up date. Measurement of plasma presepsin levels Plasma presepsin levels were quantified using the PATHFAST Presepsin assay (LSI Science Corp.). Blood samples were collected in tubes containing ethylenediaminetetraacetic acid and maintained at room temperature. Plasma presepsin measurements were performed within 4 h of sample collection in accordance with the manufacturer’s protocol. The assay has a detection range of 20–20,000 pg/mL with a coefficient of variation of 4%–5%. Statistical analysis All continuous variables are expressed as medians (interquartile ranges). Normally and non-normally distributed continuous variables were compared using the Student’s t test and Mann–Whitney U test, respectively. Categorical variables are expressed as proportions and were compared using the chi-square test or Fisher exact test, as appropriate. The predictive values of plasma presepsin, PCT, CRP, and WBC count for 28-day mortality were analyzed using receiver operating characteristic (ROC) curve analysis. Cutoff values were determined to identify the highest simultaneous sensitivities and specificities. Survival curves were generated using the Kaplan–Meier method. Survival was then compared between patients with high and low presepsin, PCT, and CRP levels and with high and low WBC count according to the cutoff value for each variable using the log-rank test. Univariate and multivariate Cox proportional hazards regression analyses were performed to identify the risk factors for 28-day mortality. Variables for the multivariate Cox proportional hazards model were selected based on a multi-step rationale. We included 1) all variables that showed a statistically significant difference (p < 0.05) in the baseline comparison between the survivor and non-survivor group as shown in Table 1; 2) key inflammatory markers (WBC count, CRP and PCT) to allow for direct comparison with presepsin; and 3) variables of known clinical importance. Specifically, eGFR was included due to is well-established influence on presepsin levels. Given the strong influence of kidney function on presepsin levels, we employed two distinct statistical strategies to control for the confounding effect of the eGFR. First, in the primary multivariate analysis of the overall cohort, we addressed the potential non-linear effect of eGFR by modeling it as a restricted cubic spline with four knots placed at clinically established thresholds for chronic kidney disease stages (eGFR of 15, 30, 60, and 90 mL/min/1.73 m²). This allowed for a flexible, non-linear adjustment for renal function. Table 1Comparison of baseline characteristics between survivor and non-survivor groups in the overall cohortCharacteristicSurvivor (n = 56)Non-survivor (n = 13)p-valueAge, years (IQR)80.00 (73.00–83.75)80.00 (76.00–84.50)0.782Body mass index, kg/m2 (IQR)21.43 (19.63–23.86)19.50 (18.08–24.00)0.424Male (%)26 (46.4)3 (23.1)0.212Mean arterial pressure, mmHg (IQR)78.83 (67.25–96.67)70.00 (59.83–85.83)0.145Heart rate, bpm (IQR)90 (78–108)98 (81–112)0.614Respiratory rate, rpm (IQR)20 (18–22)22 (18–26)0.510Body temperature, °C (IQR)37.2 (36.7–37.9)36.7 (36.3–38.0)0.359Lactate, mmol/L (IQR)1.90 (1.00–3.03)4.50 (1.65–5.85)0.038*Serum Na, mEq/L (IQR)138.0 (135.0–142.0)138.0 (133.0–148.5)0.888Serum K, mEq/L (IQR)3.93 (3.43–4.61)3.94 (2.99–4.69)0.956Serum creatinine, mg/dL (IQR)1.45 (0.91–2.18)2.65 (1.10–3.30)0.084eGFR, ml/min/1.73 m2 (IQR)48.96 (29.89–86.29)23.63 (18.00–70.01)0.079Hematocrit, % (IQR)32.45 (28.30–38.98)34.90 (29.10–38.90)0.679WBC count, × 103/µL (IQR)10.2 (7.5–15.6)7.9 (3.4–20.8)0.459Platelet count, × 103/µL (IQR)155 (107–215)96 (66–142)0.021*Total bilirubin, mg/dL (IQR)0.86 (0.57–1.57)0.91 (0.78–1.38)0.506Total protein, g/dL (IQR)6.18 (5.52–6.69)6.51 (5.58–6.95)0.276Albumin, g/dL (IQR)3.15 (2.69–3.45)3.05 (2.42–3.27)0.239Glasgow coma scale (IQR)14 (11–15)13 (11–14.5)0.443PaO2/FiO2 ratio (IQR)259.31 (162.35–374.76)215.31 (136.96–406.80)0.702Mechanical ventilation (%)7 (12.5)2 (15.4)0.674Inotrope administration (%)24 (42.9)10 (76.9)0.034*Corticosteroid administration (%)4 (7.1)0 (0)1.000Blood culture (%)33 (82.5)8 (61.5)1.000Presepsin, pg/mL (IQR)736 (467–1487)1,850 (765–3923)0.015*Procalcitonin, ng/mL (IQR)5.8 (0.8–46.6)21.1 (3.2–95.5)0.117CRP, mg/dL (IQR)13.5 (5.1–26.8)16.7 (3.1–25.2)0.900Comorbidities Diabetes mellitus (%)24 (42.9)7 (43.9)0.545 Hypertension (%)39 (69.6)11 (84.6)0.491 Chronic kidney disease (%)15 (26.8)6 (46.2)0.322 Structural heart disease (MI + CHF) (%)11 (19.6)6 (46.2)0.071 Cerebrovascular disease (%)22 (39.3)4 (30.8)0.753 Dementia (%)21 (37.5)6 (46.2)0.753 Malignancy (%)9 (16.1)0 (0)0.191Charlson’s comorbidity index (IQR)3 (1–4)3 (2–4)0.399Continuous variables are expressed as or median (IQR). Categorical variables are expressed as numbers (percentages)CRP C-reactive protein, eGFR estimated glomerular filtration rate, IQR interquartile range, K potassium, Na sodium, PaO2/FiO2 partial pressure of oxygen in arterial blood/fraction of inspired oxygen, WBC white blood cell*indicates a p-value < 0.05 Comparison of baseline characteristics between survivor and non-survivor groups in the overall cohort Continuous variables are expressed as or median (IQR). Categorical variables are expressed as numbers (percentages) CRP C-reactive protein, eGFR estimated glomerular filtration rate, IQR interquartile range, K potassium, Na sodium, PaO2/FiO2 partial pressure of oxygen in arterial blood/fraction of inspired oxygen, WBC white blood cell *indicates a p-value < 0.05 Second, in a separate sensitivity analysis, we used propensity score matching to create a well-balanced cohort. We performed 2:1 matching between the survivor and non-survivor groups based on key confounders: age, sex, and eGFR. Nearest neighbor matching was applied using the ‘MatchIt’ package in R. After matching, a multivariate Cox regression analysis was performed on the matched cohort to re-evaluate the prognostic value of presepsin. All statistical analyses were conducted using IBM SPSS Statistics version 24 (IBM Corp., Armonk, NY, USA) and R software version 4.4.2 (R Foundation for Statistical Computing, Vienna, Austria). Statistical significance was set at p < 0.05. Results Comparison of baseline characteristics between the survivor and non-survivor groups Among the 69 patients, 56 survived and 13 did not. Table 1 shows the baseline characteristics of the survivor and non-survivor groups. Lactate was significantly lower in the survivor than in the non-survivor group (1.90 mmol/L vs. 4.50 mmol/L, p = 0.038). The platelet count was significantly higher in the survivor than in the non-survivor group (155 × 10³/µL vs. 96 × 10³/µL, p = 0.021). The administration of inotropes was significantly higher in the non-survivor group (42.9% vs. 76.9%, p = 0.034). However, comorbidities and the Charlson Comorbidity Index were not significantly different between the groups. Plasma presepsin levels were significantly lower in the survivor than in the non-survivor group (736 pg/mL vs. 1,850 pg/mL, p = 0.015). PCT (5.8 ng/mL vs. 21.1 ng/mL), and CRP (13.5 mg/dL vs. 16.7 mg/dL) levels tended to be higher in the non-survivor group, although the differences were not significant. Similarly, although the WBC count was higher in the survivor group (10.2 × 103/µL vs. 7.9 × 103/µL), the difference was not significant. Values of WBC count and plasma CRP, PCT, and presepsin levels for predicting survival In the overall cohort, presepsin had the highest area under the ROC curve (AUROC) (0.716; 95% confidence interval [CI], 0.547–0.884), followed by PCT (0.641; 95% CI, 0.479–0.803), CRP (0.488; 95% CI, 0.312–0.665), and WBC count (0.433; 95% CI, 0.210–0.656) (Fig. 2a). The optimal cutoff values were determined using Youden’s index as 1,318 pg/mL, 16.6 ng/mL, 16.6 mg/dL, and 16.1 × 103/µL, respectively. When the patients were stratified into high and low presepsin groups based on a cutoff value of 1,318 pg/mL, the survival rate was significantly lower in the high than in the low presepsin group (log-rank p < 0.05; Fig. 3a). The mean survival time was shorter in the high-presepsin group than in the low-presepsin group (20.0 vs. 26.5 days). Additionally, Kaplan–Meier survival analysis of overall mortality during the whole follow-up period demonstrated a significantly shorter mean survival time in the high-presepsin group than in the low-presepsin group (6.9 vs. 11.3 months; log-rank p < 0.05; Fig. 3b). ROC curve analysis for predictive biomarkers of 28-day mortality in overall and propensity score-matched cohorts. ROC curves for presepsin, PCT, CRP, and WBC count are shown. a In the overall cohort, presepsin showed the highest AUROC (0.716), followed by PCT (0.641), CRP (0.448), and WBC count (0.433). b In the propensity score-matched cohort, Presepsin showed the highest AUROC (0.660), followed by PCT (0.584), CRP (0.496), and WBC count (0.436). All values were within the 95% confidence intervals. a The optimal cutoff values in the overall cohort were 1318 pg/mL for presepsin, 16.6 ng/mL for PCT, 16.6 mg/dL for CRP, and 16.1 × 10 ³/µL for WBC count. b In the propensity score-matched cohort, the cutoff values were 1,310 pg/mL for presepsin, 14.9 ng/mL for PCT, 21.7 mg/dL for CRP, and 16.0 × 10 ³/µL for WBC count. ROC, receiver operating characteristic; AUROC, area under the receiver operating characteristic curve; PCT, procalcitonin; CRP, C-reactive protein; WBC, white blood cell Kaplan–Meier survival curves based on presepsin levels in the overall cohort.(a) Survival analysis for 28-day mortality and (b) overall mortality during the whole follow-up period. Patients were stratified by a presepsin cutoff of 1,318 pg/mL. In both analyses, the high-presepsin group (red line) showed significantly lower survival rates than the low-presepsin group (blue line) (log-rank p < 0.05) Risk factors of 28-day mortality Table 2 presents the results of the Cox proportional hazards regression analysis conducted to identify the risk factors associated with 28-day mortality. The analysis included sex, age, lactate, eGFR, PLT, inotrope administration, WBC count, CRP, PCT, and presepsin. Univariate Cox proportional hazards regression analysis was performed for all variables. Lactate, PLT, and presepsin were significant in the univariate analysis, with univariate hazard ratios (HRs) of 1.228 (95% CI, 1.042–1.448), 0.990 (95% CI, 0.981–0.999), 1.021 (95% CI, 1.005–1.037), respectively. After adjusting for multiple confounding factors, lactate, PLT, and presepsin were identified as independent risk factors for 28-day mortality, with multivariate HRs of 1.255 (95% CI, 1.059–1.488), 0.991 (95% CI, 0.982–1.000) and 1.026 (95% CI, 1.007–1.045), respectively. Table 2Multivariate Cox regression analysis for 28-day mortality in the overall cohortVariableUnivariate analysisMultivariate analysisp-valueHR95% CIp-valueHR95% CISex (Ref. Male)0.1462.6070.716–9.490---Age (per 1 year)0.4211.0280.961–1.101---WBC count (per 10 3 /µL)0.8280.9910.911–1.077---CRP (per 1 mg/dL)0.7060.9910.948–1.037---PCT (per 1 ng/mL)0.2391.0080.994–1.022---Lactate (per 1 mmol/L)0.014*1.2281.042–1.4480.009*1.2551.059–1.488eGFR (per 1 ml/min/1.73 m 2 )0.1240.9850.966–1.004---PLT (per 10 3 /µL)0.036*0.9900.981–0.9990.040*0.9910.982–1.000Inotrope administration (Ref. No)0.0543.5470.976–12.891---Presepsin (per 100 pg/mL)0.011*1.0211.005–1.0370.008*1.0261.007–1.045Multivariate regression model was adjusted for known significant factors and those that showed statistical differences between the survivor and non-survivor groups. All participants (100%) were included in the regression model. The following parameters were used: sex, age, WBC count, CRP, PCT, Lactate, eGFR, PLT, inotrope administration and presepsinCI confidence interval, CRP C-reactive protein, eGFR estimated glomerular filtration rate, HR hazard ratio, PCT procalcitonin, PLT platelet, WBC white blood cell*indicates a p-value < 0.05 Multivariate Cox regression analysis for 28-day mortality in the overall cohort Multivariate regression model was adjusted for known significant factors and those that showed statistical differences between the survivor and non-survivor groups. All participants (100%) were included in the regression model. The following parameters were used: sex, age, WBC count, CRP, PCT, Lactate, eGFR, PLT, inotrope administration and presepsin CI confidence interval, CRP C-reactive protein, eGFR estimated glomerular filtration rate, HR hazard ratio, PCT procalcitonin, PLT platelet, WBC white blood cell *indicates a p-value < 0.05 Value of plasma presepsin, PCT, and CRP levels and WBC count for predicting 28-day mortality in the propensity score-matched cohort In total, 26 survivors and 13 non-survivors were enrolled in the propensity score-matched cohort. Presepsin demonstrated the highest AUROC value (0.660; 95% CI, 0.464–0.855), followed by PCT (0.584; 95% CI, 0.397–0.772), CRP (0.496; 95% CI, 0.293–0.698) and WBC count (0.436; 95% CI, 0.209–0.664) (Fig. 2b). Table 3 shows the results of the Cox proportional hazard regression analysis performed to identify the risk factors influencing 28-day mortality in the propensity score-matched cohort. The analysis included age, lactate, PLT, inotrope administration, WBC count, CRP, PCT, and presepsin. Lactate and inotrope administration were significant factor in the univariate analysis, with a HR of 1.350 (95% CI, 1.091–1.671) and 4.072 (1.119–14.914). After adjusting for multiple confounding variables, lactate and presepsin were identified as independent risk factors for 28-day mortality, with multivariate HRs of 1.415 (95% CI, 1.131–1.771) and 1.019 (95% CI, 1.001–1.037), respectively. Table 3Multivariate Cox regression analysis for 28-day mortality in the propensity score-matched cohortVariableUnivariate analysisMultivariate analysisp-valueHR95% CIp-valueHR95% CIAge (per 1 years)0.8950.9950.925–1.070---WBC count (per 103/µL)0.7710.9880.910–1.072---CRP (per 1 mg/dL)0.6550.9890.945–1.036---PCT (per 1 ng/mL)0.7591.0020.988–1.016---Lactate (per 1 mmol/L)0.006*1.3501.091–1.6710.002*1.4151.131–1.771PLT (per 103/µL)0.0670.9930.985–1.001---Inotrope administration (Ref. No)0.033*4.0721.119–14.914---Presepsin (per 100 pg/mL)0.1381.0130.996–1.0300.038*1.0191.001–1.037Multivariate regression model was adjusted for known significant factors and those that showed statistical differences between the survivor and non-survivor groups. All participants (100%) were included in the regression model. The following parameters were used: age, WBC count, CRP, PCT, lactate, PLT, inotropes administration and presepsinCI confidence interval, CRP C-reactive protein, HR hazard ratio, PCT procalcitonin, PLT platelet, WBC white blood cell*indicates a p-value < 0.05 Multivariate Cox regression analysis for 28-day mortality in the propensity score-matched cohort Multivariate regression model was adjusted for known significant factors and those that showed statistical differences between the survivor and non-survivor groups. All participants (100%) were included in the regression model. The following parameters were used: age, WBC count, CRP, PCT, lactate, PLT, inotropes administration and presepsin CI confidence interval, CRP C-reactive protein, HR hazard ratio, PCT procalcitonin, PLT platelet, WBC white blood cell *indicates a p-value < 0.05 Discussion The current study showed that serum presepsin levels at the time of admission were a risk factor for 28-day and overall mortality during the whole follow-up period in patients with UTI-related severe infection. Furthermore, compared to traditional markers such as PCT, CRP, and WBC count, presepsin demonstrated superior predictive performance for mortality. These findings extend beyond the initial hypothesis that presepsin can serve as an indicator of severe sepsis in patients with UTIs, highlighting its potential role in mortality-risk stratification. In the current study, 13 of the 69 patients died, yielding a mortality rate of 18.8%. This rate is consistent with previously reported mortality rates associated with UTI and sepsis [19]. In the analysis of baseline characteristics of the overall cohort, lactate, platelet count, and inotrope administration were significantly different between the survivor and non-survivor groups, in addition to inflammatory markers. Lactate is a well-established component in the definition of septic shock and, similar to other inflammatory markers, plays a major role in prognostic assessment. Platelet count and inotrope administration are integral components of established prognostic scoring systems, including the Sequential Organ Failure Assessment (SOFA). The significantly higher rate of inotrope administration in the non-survivor group indicates a greater baseline illness severity and a higher prevalence of septic shock in these patients. To address the potential for this to confound our results, we included inotropes administration as a covariate in our multivariate Cox regression model. Notably, even after adjusting for this critical indicator of severity, presepsin remained a strong and independent predictor of 28-day mortality. This finding reinforces the robustness of presepsin as a prognostic biomarker, suggesting its predictive value extends beyond simply reflecting the patient’s initial hemodynamic instability. Although our findings are largely consistent with those of previous studies that utilized various scoring systems [20, 21], a direct analysis using the full SOFA, Acute Physiology and Chronic Health Evaluation II (APACHE II), and National early warning score was not conducted. The qSOFA score was adopted as the primary tool for cohort identification due to its reliability and suitability for rapid clinical application in settings like the emergency department. Compared to the full SOFA score, which is the official diagnostic standard for sepsis, qSOFA is more easily calculated and is effective for quickly identifying patients at high risk of severe infection upon admission. Furthermore, since the “UTI-related severe infection” category included patients who did not require ICU care, data for calculating full SOFA and APACHE II scores were not collected for all patients, precluding their analysis. Our study demonstrated the utility of presepsin as a novel biomarker for predicting mortality in patients with UTI-related severe infection in the emergency department, providing practical and clinically relevant information. This result is supported by the findings of a previous study that evaluated the use of presepsin as a prognostic factor in the emergency department. The prognostic utility of presepsin for UTI-related severe infection in the emergency department is particularly significant, given the rapidly progressive nature of UTI sepsis [22–24]. In our study, serum presepsin levels were identified as independent risk factors for mortality and were significantly associated with survival rates. Previous studies have reported presepsin cutoff values of 400–600 pg/mL for the diagnosis of sepsis [25, 26]. However, our findings indicate a higher threshold of 1,318 pg/mL in the overall cohort. Presepsin levels tend to be elevated in patients with kidney dysfunction under similar clinical conditions [27, 28]. This phenomenon is attributed to the fact that presepsin is primarily eliminated by the kidneys; thus, when kidney function declines, serum presepsin levels may accumulate. In our study, the mean serum creatinine level in the overall cohort was 2.07 mg/dL. This result is higher than the mean serum creatinine level of 1.2 mg/dL reported in a previous study, in which a presepsin cutoff value of 492 ng/mL was identified as a prognostic factor for UTI sepsis [25]. In addition, while previous studies utilized presepsin cutoff values for the diagnosis of sepsis, this study applied the presepsin cutoff value to predict 28-day mortality. Therefore, it is important to consider that the primary outcomes of this study differ from those of previous studies. Previous studies have examined the diagnostic value of presepsin as a predictor of the progression of UTI to septic shock, its role in predicting bacteremia, and the need for hospitalization in patients with acute pyelonephritis [25, 29]. To the best of our knowledge, only one study has evaluated the use of presepsin as a predictor of mortality in UTI sepsis [30]. However, the prior study did not directly account for renal function using eGFR, although serum creatinine was included as a variable in the Cox proportional hazard regression analysis. Given that serum presepsin levels are significantly influenced by kidney function, eGFR may serve as a confounding variable when assessing the suitability of presepsin as a prognostic factor. Our study was notable for controlling confounding factors, including eGFR, through propensity score matching (Additional File 2 has a table that shows the comparison of baseline characteristics between survivor and non-survivor groups in the propensity score-matched cohort). Despite comparable kidney function between the two groups, serum presepsin levels remained a significant predictor of 28-day mortality. In the multivariate Cox regression analysis of the propensity score-matched cohort, presepsin was independently associated with an increased risk of mortality, consistent with the results observed in the overall cohort. This study has several limitations. First and foremost, the small sample size, which was further reduced after propensity score matching, is a significant limitation. While propensity score matching was used to mitigate potential biases by adjusting for confounders like kidney function, age, and sex, the resulting small cohort limits the statistical power of our analysis. Second, as this was a retrospective, single-center study, our findings may lack generalizability to other populations and settings. Third, due to the short half-life of presepsin, sequential measurements after admission could have provided valuable prognostic information, but such data were not collected. Further studies with larger cohorts, including multicenter clinical trials, are required to confirm the utility of plasma presepsin as a predictor of mortality in UTI and to establish its generalizability to the broader population. Conclusions Plasma presepsin is a superior and independent predictor of 28-day mortality in patients with UTI-related severe infections compared to WBC count, CRP levels, and PCT levels. As an early and objective indicator of severity, presepsin can aid in crucial clinical decisions in the emergency department, such as prioritizing high-risk patients for ICU admission and optimizing the allocation of healthcare resources, potentially improving patient outcomes. Supplementary Information Additional file 1: Distribution of antibiotic susceptibility patterns and administered antibiotics by isolated organism. Additional file 2: Comparison of baseline characteristics between survivor and non-survivor groups in the propensity score-matched cohort. Acknowledgements Not applicable. Abbreviations Acute Physiology and Chronic Health Evaluation II Area under the receiver operating characteristic curve C-reactive protein Estimated glomerular filtration rate Glasgow Coma Scale Hazard ratios Intensive care unit Institutional Review Board Procalcitonin Receiver operating characteristic Soluble subtype of CD14 Sequential Organ Failure Assessment Urinary tract infection White blood cell Authors’ contributions JWL and YP were responsible for conceptualization and methodology. JWL, DS, YKK, and SHH were responsible for data curation. JWL and JS were responsible for the formal analysis. JWL and YP were responsible for investigation, validation, visualization and review and editing of the manuscript. YP were responsible for funding acquisition. JWL were responsible for writing the original draft. SHY, SRY, WMH, JKL, and SJK were responsible for supervision. Each author contributed important intellectual content during manuscript drafting or revision and agrees to be personally accountable for the individual’s own contributions and to ensure that questions pertaining to the accuracy or integrity of any portion of the work, even those for which the author was not directly involved, are appropriately investigated and resolved, including with literature documentation if appropriate. Funding This research was supported by a grant of the Korean Health Technology R&D Project through the Korean Healthy Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (grant number: RS-2025-02215809). This study is an investigator-led project with no specific participation by funders. Data availability The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request. Declarations Ethics approval and consent to participate This study was approved by the Institutional Review Board (IRB) of Konyang University Hospital, College of Medicine, Konyang University (No. KYUH 2024-12-013). The need to obtain informed patient consent was waived by the IRB owing to the retrospective nature of the study. This study was conducted in accordance with the principles of the Declaration of Helsinki. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Additional file 1: Distribution of antibiotic susceptibility patterns and administered antibiotics by isolated organism. Additional file 2: Comparison of baseline characteristics between survivor and non-survivor groups in the propensity score-matched cohort. Data Availability Statement The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.