Blog

Introduction

Bloodstream infections (BSIs) are a major global health problem in the elderly as age-related immune defects, multiple comorbid conditions and diminished physiological reserves predispose these patients to high morbidity, prolonged hospitalization and increased mortality.^1,2 Although Escherichia coli (E.coli) is the predominant pathogen causing BSIs, non-E.coli organisms such as Pseudomonas aeruginosa and Klebsiella pneumoniae associated with BSIs are known to be more severe, and result in higher mortality and complex treatment requirements necessitating a longer hospital stay.³ The elderly are at particular risk for complicated clinical courses due to late diagnosis which may limit prompt initiation of appropriate treatment, thereby emphasizing the need for pathogen-specific risk stratification approach.

Neutrophil-to-platelet ratio (NPAR) also reflects both systemic inflammation and thrombotic risk.⁴ Elevated NPAR is associated with poor prognosis in sepsis and other bacterial infections, but the role of NPAR in predicting E.coli bloodstream infection (BSI) among elderly patients was not well studied.⁵ With development of microfluidics and novel phage-based detection systems, combining NPAR with pathogen specific diagnostic tools holds promise in the early management of E.coli BSI. Though evidence suggests that patients experience higher mortality in non-E.coli compared to E.coli infection causing BSIs, there is currently no study that has directly compared risk of mortality between E.coli and non- E.coli BSIs among elderly population. Furthermore, the utility of NPAR in predicting E.coli BSI among elderly patients remains unexplored.

This study aimed to compare mortality between E.coli versus non-E.coli BSIs among elderly inpatients, explore potential utility of NPAR as a diagnostic biomarker to predict E.coli BSI and its prognostic implications among them.

Methods

Study Design and Participants

This single-center, retrospective cohort study encompassed 527 elderly patients diagnosed with BSIs between December 2011 and February 2024 at the Second Medical Center of the Chinese PLA General Hospital through the hospital’s infection information system. The inclusion criteria were: (1) age greater than 65 years; and (2) availability of complete medical records. The exclusion criteria were as follows: (1) Incomplete medical records. The study protocol was reviewed and approved by the Chinese PLA Hospital Ethical Committee (Approval No.NO. S2024-359-02) and complied with the Declaration of Helsinki. Due to the retrospective design, informed consent was waived.

Data Collection

Data on NPAR levels and other covariates, including demographic and clinical factors, were collected at baseline. NPAR was measured as both a continuous variable and categorized into tertiles (T1, T2, and T3). The outcome of interest was the occurrence of E.coli BSI, which was confirmed by blood culture. A BACT/ALERT 3D automatic blood culture instrument (bioMérieux, France) was used for blood culture, a Vitek2 Compact automatic microbiological identification and antimicrobial susceptibility analysis system (bioMérieux, France) was used for strain identification and antimicrobial susceptibility testing. Baseline data were collected on a range of demographic and clinical characteristics, including age, gender, department, smoking status, comorbidities (eg, diabetes mellitus, hypertension, coronary disease), and clinical interventions (eg, number of operations, use of ventilator, central venous catheter, urinary catheter, chemotherapy, radiotherapy, blood transfusion, polypharmacy regimens). The hospitalization duration was also recorded as a continuous variable.

Statistical Analysis

Descriptive statistics were used to summarize the baseline characteristics of the participants. Continuous variables were expressed as means with standard deviations, and categorical variables were reported as frequencies and percentages. Categorical variables were compared using the chi-square or Fisher’s exact test; continuous variables were analyzed using the Student’s t-test, Mann–Whitney U-test or Kruskal–Wallis test as appropriate; logistic regression was used to assess the predictive value of NPAR for E. coli BSI; Cox proportional hazards models were applied for survival analysis; and the Kaplan–Meier method with Log rank test was used to compare mortality between groups. To evaluate the association between NPAR and E. coli BSI, we constructed three models in our analysis: Model 1: Unadjusted crude model. Model 2: Adjusted for age and sex. Model 3: Based on Model 2, we further adjusted for variables that showed statistical significance in the univariate analysis, as well as through reverse adjustment, including department, coronary disease, and combination of drug. The odds ratios (ORs) and 95% confidence intervals (CIs) were calculated for both continuous NPAR and NPAR tertiles (T1 as the reference group). The dose-response relationship for continuous NPAR was assessed, and the trend across NPAR tertiles was evaluated using a likelihood ratio test. The non-linearity of the dose-response relationship was examined using a non-linear regression model. The p-value for trend was calculated to assess the monotonicity of the dose-response across tertiles of NPAR. Statistical significance was set at P < 0.05 for all analyses. All statistical analyses were performed using R (version 3.6.3) statistical software.

Results

Baseline Characteristics

A cohort of 510 elderly participants (mean age 89.9 ± 8.5 years) meeting inclusion and exclusion criteria was stratified into E.coli BSI (n=92, 18.2%) and non-E.coli BSIs groups (n=418, 81.8%) based on the causative agent of bloodstream infection. The baseline characteristics of the overall cohort and the two groups are summarized in Table 1. No intergroup differences in age (P=0.978), gender (P=0.229), smoking status (P=0.193), diabetes (P=0.614), or hypertension (P=1.0). Higher coronary disease prevalence in non-E.coli group (57.2% vs 32.3%, P<0.001) and comparable surgical frequency (P=0.441).

Clinical Interventions and Outcomes

As shown in Table 2, 49.1% (n = 510) patients did not require ventilator support, with a higher proportion of patients in the E. coli group (61.3%) compared to the non-E.coli group (46.4%) (P = 0.045). Similarly, the duration of central venous catheter use was significantly longer in the E.coli group, with 25.8% of patients requiring it for more than 90 days, compared to 14.1% in the non-E.coli group (P = 0.023). Blood transfusion was more frequently administered in the E.coli group (73.1% vs 58.1%, P = 0.01). The length of hospital stay did not differ significantly between the two groups (P = 0.563), with a median length of stay of 90.0 days (IQR: 65.4, 96.0) in the overall cohort.

Table 2 Clinical Interventions and Outcomes of Enrolled Bloodstream Infection Patients

Departments and Infection Source Distribution

There was a statistically significant difference between E.coli group and non-E.coli groups regarding the departments in which patients were hospitalized (P = 0.046) (Table 1). The distribution of departments of patients with E.coli bloodstream infections were as follows: Cardiology (16.30%), Endocrinology (3.26%), Gastroenterology (35.87%), Hematology (1.09%), ICU (4.35%), Nephrology (4.35%), Neurology (5.43%), Oncology (2.17%), and Respiratory Medicine (10.87%) (Figure 1). The distribution of infection sources among patients with E.coli bloodstream infections were as follows: Biliary infection source (28.26%), non-biliary intra-abdominal infection (3.26%), Pulmonary infection source (16.30%), Unidentified infection source (23.92%), and Urinary tract infection source (28.26%) (Figure 1). Also, we found that non-E. coli pathogens were the primary contributors. Among these, Staphylococcus species were the most prevalent, accounting for 125 isolates out of 510 total samples (Supplementary Table 1).

Figure 1 Distribution of infection sources and departments in patients with Escherichia coli bloodstream infections. (A) Distribution of infection sources in patients with Escherichia coli bloodstream infections. (B) Distribution of infection sources in internal medicine bloodstream Infection patients with Escherichia coli. (C) Distribution of infection sources in surgeon bloodstream infection patients with Escherichia coli. (D) Distribution of departments of patients with Escherichia coli bloodstream infections.

NPAR Associations with E.coli BSI

The association between NPAR and E.coli BSI was evaluated using three models (Table 3). All models demonstrated a statistically significant inverse relationship between continuous NPAR and E.coli BSI risk: Model 1, OR=0.88 (95% CI: 0.84, 0.93; P < 0.001), Model 2, OR =0.88 (95% CI: 0.84, 0.92; P < 0.001), and in Model 3, OR=0.89 (95% CI: 0.84, 0.94; P < 0.001). Using Tertile 1 (T1) as the reference group, in Tertile 2 (T2), the odds of infection were significantly reduced, with Model 1(OR=0.53; 95% CI: 0.32, 0.89; P = 0.017), Model 2(OR= 0.50; 95% CI: 0.29, 0.84; P = 0.01) and Model 3(OR =0.46; 95% CI: 0.26, 0.81; P = 0.008). In Tertile 3 (T3), the odds of infection were even more substantially reduced, with Model 1(OR =0.21; 95% CI: 0.11, 0.40; P < 0.001), Model 2 (OR=0.21; 95% CI: 0.11, 0.39; P < 0.001), and Model 3 (OR= 0.23; 95% CI: 0.11, 0.46; P < 0.001). Furthermore, a statistically significant trend towards decreasing odds of infection across increasing tertiles of NPAR was observed in all models (P for trend < 0.001). These findings suggest that higher NPAR levels correlate with a reduced likelihood of E.coli BSI, with the risk decreasing progressively across tertiles. This aligns with NPAR’s role as a composite inflammatory marker, where elevated values may reflect an attenuated susceptibility to systemic infection.

Table 3 Associations of NPAR with Escherichia coli Bloodstream Infections in Elder Patients

Dose-Response Relationship Between NPAR and E.coli BSI

A linear relationship between NPAR and E.coli BSI was statistically significant (P < 0.05), while a non-linear relationship was not evident (P = 0.424, Cutoff value =19.4) (Figure 2). Therefore, as NPAR increases, there is a linear decrease in E.coli BSI risk. Hence NPAR may serve as a continuous protective marker rather than having a threshold at a particular level.

Figure 2 The dose-response relationship between NPAR and Escherichia coli bloodstream infection. (NAPR: neutrophil-to-platelet ratio).

Survival Analysis

The Kaplan-Meier survival plots of patients with E.coli and non-E.coli BSIs are shown in Figure 3, which revealed higher survival probability of patients with E coli compared with non-E coli counterparts (HR=0.43; 95% CI 0.21, 0.88, P=0.021). Notably, the survival probability for patients with non-E.coli infections dropped more rapidly over time, whereas the E.coli group exhibited a more gradual decline in survival.

Figure 3 Kaplan-Meier curve analysis of patients in Escherichia coli and non-Escherichia coli bloodstream infection groups.

Antimicrobial Resistance Pattern

As shown in Table 4 and Figure 4, almost 97.1% of E.coli isolates were resistant to Ampicillin (AMP), 77.1% were resistant to Ciprofloxacin (CIP), 74.3% were resistant to Cefazolin (CEZ), 72.9% were resistant to Levofloxacin (LVX), and 72.7% were resistant to Ampicillin/Sulbactam (AMP/SUL). Overall, E.coli isolates exhibited no resistance to the following antibiotics: Amikacin (AMK), Ertapenem (ETP), Tigecycline (TGC), and Cefotetan (CET). The antibiotic with the highest resistance rate in E.coli isolates from bloodstream infection patients in the respiratory department were AMP (100%), Ceftriaxone (CTX, 100%), CEZ (100%). AMP (100%), Doxycycline (DXY, 100%), Gentamicin (GEN, 100%). In contrast, the general surgery department had a lower resistance rate for antibiotics like Ampicillin/Sulbactam (60.0%) and Gentamicin (66.7%). Notably, antibiotics like Amikacin, Ertapenem, and Tigecycline showed a resistance rate of 0% across all departments. Resistance to Cefotetan was absent across all departments tested.

Table 4 Antibiotic Resistance Rate of Escherichia coli in Bloodstream Infection Patients

Figure 4 Overall antibiotic resistance rate of Escherichia coli.

Discussion

Our study provides several important implications to be addressed clinically for BSIs in extremely elderly patients. First, we demonstrated that non-E. coli BSIs had a significantly higher risk of mortality compared to E. coli BSI in extremely elderly inpatients with the mean age of 89.9 ± 8.5 years, indicating the clinical importance of identifying the pathogen causing infection. We also provided that low NPAR was inversely associated with the presence of E. coli BSI which may be useful early identification and risk stratification in elderly, leading to a more tailored and early intervention to be initiated.

Among 30,923 cases of E.coli bloodstream infections, 2961 cases of 30-day mortality were observed, resulting in an overall 30-day mortality of 9.6% (2961/30,923).⁶ Hospital-acquired or third-generation-cephalosporin-resistant E.coli BSI showed significantly higher mortality rates compared to community-acquired or third-generation-cephalosporin-susceptible E.coli BSI.⁶ In our cohort of elderly patients, non-E.coli BSIs were associated with higher mortality compared to E.coli BSIs, even after adjustment for demographics and clinical factors, which is consistent with previous studies. Various studies have reported that the reasons why high mortality is associated with non-E.coli infections include difficulty in diagnosis, limited treatment options, and increased infection severity due to multidrug-resistant organisms or high virulence.^6,7 As elderly patients have weak immunity and often suffer from multiple underlying diseases, there would be a greater concern about their course and response to treatment.^8,9 Our findings indicate that it is vital to early detection and proper management for non-E.coli BSIs. Previous studies have reported that E.coli BSI in elderly patients predominantly originate from urinary tract infections,¹⁰ which is consistent with the findings of our study, where urinary and biliary tract infections were identified as the leading sources. These infections typically elicit a relatively mild systemic inflammatory response, characterized by only a modest elevation in neutrophil counts and minimal alterations in platelet levels. In contrast, non-E. coli BSIs – such as those caused by Staphylococcus aureus, Klebsiella pneumoniae, or Enterococcus faecalis – are more often associated with catheter-related infections, non-biliary intra-abdominal infections, and respiratory tract infections, which usually trigger a more severe systemic inflammatory response.¹¹ It is reported that thrombocytopenia was independently associated with mortality among patients with BSIs.¹² This aligns with prior studies showing NPAR and related indices (eg, neutrophil-to-lymphocyte ratio, platelet-to-lymphocyte ratio) as strong, independent predictors of sepsis severity and mortality.¹³ A high NPAR captures a dual-risk profile: amplified inflammatory response and impaired hemostatic balance, both of which have been independently linked to increased mortality in bloodstream infections.¹⁴ In our multivariable Cox regression models, high NPAR remained significantly associated with mortality even after adjusting for age, comorbidities, and infection source, indicating its robust prognostic value. Clinically, low NPAR may serve as an early indicator of E. coli BSI and help clinicians stratify patients who are likely to have more benign infection courses, potentially guiding early empirical therapy decisions and resource allocation. Conversely, persistently elevated NPAR should prompt vigilance for non-E. coli pathogens or complicated infection sources.

The prevalence of E.coli producing extended-spectrum beta-lactamases (ESBL) among BSI patients was 40.98%. E.coli isolates were generally sensitive to carbapenems and β-lactam/β-lactamase inhibitor combinations. Hospital-acquired infections, biliary tract infections, gastric tube insertion procedures, and prior cephalosporin administration were identified as independent risk factors for the isolation of ESBL-producing strains. ESBL positivity, hospital-acquired infections, and cancer were independent risk factors for mortality.¹⁵ Meta-analysis results indicate that it is necessary to shift current treatment practices from antibiotic escalation strategies that delay appropriate therapy to early, relatively aggressive, and comprehensive antibiotic treatment, especially in patients with BSIs caused by Klebsiella pneumoniae or E.coli.¹⁶ Choi et al found that E.coli is the most common pathogenic microorganism in BSIs, accounting for 32.3%, and the adjusted hazard ratio (aHR) for 30-day mortality and subsequent medical costs for E.coli BSI was lower compared to other microorganisms causing BSI. E.coli-BSI resulted in lower mortality rates during the first 7 days and from days 8 to 30 compared to BSIs caused by other microorganisms.¹⁷ Our study found that the proportion of internal medicine patients was higher in the non-E.coli group, while E.coli infections were more common in the surgical department. The incidence of coronary artery disease was lower in the E.coli group, whereas it was higher in the non-E.coli group. There were significant differences in the duration of ventilator use and central venous catheter use between the E.coli and non-E.coli groups, with patients in the non-E.coli group having a longer durations of use.

NPAR is a simple, readily accessible measure derived from routine blood tests, and it has been proposed as a marker for various infectious and inflammatory conditions.^18–20 As shown in Figure 2, the OR for E.coli BSI decreases with increasing NPAR values. The analysis indicates a significant overall association between NPAR and the risk of E.coli bloodstream infection. Overall P-value: <0.001, indicating a strong statistical significance for the association between NPAR and the risk of E.coli bloodstream infection. In our analysis, the non-linear regression yielded a p-value of 0.424, indicating no significant evidence of a non-linear relationship between NPAR and infection risk. Therefore, we conclude that the relationship is best modeled as linear, suggesting a consistent, proportional association between NPAR and infection risk. OR (95% CI): The odds ratio decreases progressively with higher NPAR levels, approaching 1, indicating that higher NPAR values are associated with a reduced risk of E.coli bloodstream infection. Our findings suggest that low NPAR values are strongly associated with an increased risk of E.coli infection, with patients in the lowest NPAR tertile having substantially higher odds of having an E.coli infection compared to those in the highest tertile. This association remained consistent across various analytical models, further reinforcing the evidence for NPAR as a predictor of E.coli infections. Although the precise mechanisms linking NPAR to infection risk are not fully understood, it is believed that NPAR reflects the balance between inflammatory and immune responses.^21–23 During E. coli BSI, lipopolysaccharides (LPS) derived from the bacterial cell wall activate macrophages and other immune cells via Toll-like receptor 4 (TLR4) and related signaling pathways.²⁴ This activation triggers a cascade release of proinflammatory cytokines, including interleukin (IL)-6, IL-1, and TNF-α. Among these, IL-6 plays a central role in the IL-6–liver axis by markedly stimulating hepatic synthesis of thrombopoietin (TPO), the key regulator of megakaryocyte proliferation and differentiation.²⁵ Elevated TPO levels subsequently enhance platelet production, resulting in reactive thrombocytosis.²⁵ In addition, several inflammatory cytokines, such as IL-6, IL-11, and granulocyte-macrophage colony-stimulating factor (GM-CSF), may directly or indirectly act on hematopoietic stem and progenitor cells to promote megakaryocyte maturation and platelet release. In contrast, non-E. coli BSIs like S. aureus can induce platelet aggregation and clearance through α-toxin- and ClfA-mediated mechanisms, thereby promoting thrombocytopenia and contributing to an elevated NPAR.²⁶ Together, these inflammatory responses provide a plausible explanation for the increased platelet counts frequently observed in E. coli BSI and may partially underlie the association between a low NPAR and disease progression.

The antibiotic resistance profiles of E.coli isolates in this study reveal significant variability across different clinical departments. High resistance rates to Ampicillin, Ampicillin/Sulbactam, and Cefazolin are consistent with previous reports of widespread beta-lactam resistance.^27,28 However, the absence of resistance to Amikacin and Ertapenem is encouraging, as these antibiotics are vital for treating multidrug-resistant infections. The elevated resistance rates to Ciprofloxacin and Levofloxacin in the respiratory and cardiology departments may be linked to the frequent use of these antibiotics in those specialties. The lack of resistance to Tigecycline and Ertapenem across all departments suggests that these antibiotics could serve as effective treatment options for E.coli bloodstream infections. The relatively low resistance rates to Imipenem/Cilastatin and Meropenem in most departments further highlight the importance of carbapenems in managing severe E.coli infections. The significant resistance to Ticarcillin/Clavulanate, particularly in the general surgery department, underscores the need for more judicious use of this antibiotic to prevent the development of further resistance. Doctors should emphasize careful monitoring of antibiotic use, restricting the use of broad-spectrum antibiotics, and promoting individualized treatment guided by sensitivity testing to reduce the spread of resistance.^29,30

Conclusion

The NPAR, as demonstrated in our study, holds significant potential as a simple, cost-effective, and globally applicable biomarker for early identifying and targeted managing E. coli BSI in elderly patients. Low NPAR is associated with an increased likelihood of E. coli BSI and can help clinicians identify high-risk patients who may benefit from early therapeutic interventions. In future research, the role of NPAR as a predictive and prognostic biomarker for E.coli BSI could be further extended to populations across different age groups, with subsequent studies needed to explore longitudinal NPAR trends during treatment as a monitoring biomarker, and further elucidate the underlying biological mechanisms linking NPAR with infection susceptibility and clinical outcomes in E. coli BSI.

Data Confidentiality Statement

All patient data were handled in strict compliance with confidentiality regulations. The data were anonymized prior to analysis, and no identifiable personal information was disclosed or shared outside the research team.

Data Sharing Statement

The data are available from the corresponding author on reasonable request.

Consent to Participate

This research was approved and waived the consent by the Ethics Committee of Chinese PLA General Hospital (NO. S2020-25601). Informed consent was not required due to the retrospective nature of the study design. All authors confirm this study adheres to the Declaration of Helsinki.

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Funding

This retrospective study was supported by National Clinical Research Center for Geriatric Diseases Open Project: NCRCG-PLAGH-2022012 and NCRCG-PLAGH-2023004; Beijing Natural Science Foundation: L222014.

Disclosure

All authors in this study declare no competing conflicts.

References

1. Leibovici-Weissman Y, Tau N, Yahav D. Bloodstream infections in the elderly: what is the real goal? Aging Clin Exp Res. 2021;33(4):1101–1112. PMID: 31486996. doi:10.1007/s40520-019-01337-w

2. Lin H, Gao Y, Qiu Y, et al. The prognostic factors of bloodstream infection in immunosuppressed elderly patients: a retrospective, single-center, five-year cohort study. Clin Interv Aging. 2022;17:1647–1656. PMID: 36425478; PMCID: PMC9680683. doi:10.2147/CIA.S386922

3. Mondal U, Warren E, Bookstaver PB, et al. Incidence and predictors of complications in Gram-negative bloodstream infection. Infection. 2024;52(5):1725–1731. PMID: 38436912; PMCID: PMC11499525. doi:10.1007/s15010-024-02202-3

4. Gong Y, Li D, Cheng B, et al. Increased neutrophil percentage-to-albumin ratio is associated with all-cause mortality in patients with severe sepsis or septic shock. Epidemiol Infect. 2020;148:e87. PMID: 32238212; PMCID: PMC7189348. doi:10.1017/S0950268820000771

5. Mousa N, Salah M, Elbaz S, et al. Neutrophil percentage-to-albumin ratio is a new diagnostic marker for spontaneous bacterial peritonitis: a prospective multicenter study. Gut Pathog. 2024;16(1):18. PMID: 38561807; PMCID: PMC10985869. doi:10.1186/s13099-024-00610-2

6. MacKinnon MC, McEwen SA, Pearl DL, et al. Mortality in Escherichia coli bloodstream infections: a multinational population-based cohort study. BMC Infect Dis. 2021;21(1):606. PMID: 34172003; PMCID: PMC8229717. doi:10.1186/s12879-021-06326-x

7. Leistner R, Gürntke S, Sakellariou C, et al. Bloodstream infection due to extended-spectrum beta-lactamase (ESBL)-positive K. pneumoniae and E. coli: an analysis of the disease burden in a large cohort. Infection. 2014;42(6):991–997. PMID: 25100555. doi:10.1007/s15010-014-0670-9

8. Chen Q, Ma G, Cao H, et al. Risk factors and diagnostic markers for Escherichia coli bloodstream infection in older patients. Arch Gerontol Geriatr. 2021;93:104315. PMID: 33310397. doi:10.1016/j.archger.2020.104315

9. Giannella M, Pascale R, Toschi A, et al. Treatment duration for Escherichia coli bloodstream infection and outcomes: retrospective single-centre study. Clin Microbiol Infect. 2018;24(10):1077–1083. PMID: 29371138. doi:10.1016/j.cmi.2018.01.013

10. Choi HJ, Jeong SH, Shin KS, et al. Characteristics of Escherichia coli urine isolates and risk factors for secondary bloodstream infections in patients with urinary tract infections. Microbiol Spectr. 2022;10(4):e0166022. PMID: 35862950; PMCID: PMC9430824. doi:10.1128/spectrum.01660-22

11. Timsit JF, Ruppé E, Barbier F, et al. Bloodstream infections in critically ill patients: an expert statement. Intensive Care Med. 2020;46(2):266–284. PMID: 32047941; PMCID: PMC7223992. doi:10.1007/s00134-020-05950-6

12. Adelman MW, Casarin S, Kurian J, et al. Platelets and mortality in bloodstream infection: a multicenter cohort study. Clin Microbiol Infect. 2025;31(10):1733–1736. PMID: 40744277; PMCID: PMC12377423. doi:10.1016/j.cmi.2025.07.021

13. Gao L, Shi Q, Li H, et al. Prognostic value of the combined variability of mean platelet volume and neutrophil percentage for short-term clinical outcomes of sepsis patients. Postgrad Med. 2021;133(6):604–612. PMID: 32912023. doi:10.1080/00325481.2020.1823137

14. Cecconi M, Evans L, Levy M, et al. Sepsis and septic shock. Lancet. 2018;392(10141):75–87. PMID: 29937192. doi:10.1016/S0140-6736(18)30696-2

15. Zhao S, Wu Y, Dai Z, et al. Risk factors for antibiotic resistance and mortality in patients with bloodstream infection of Escherichia coli. Eur J Clin Microbiol Infect Dis. 2022;41(5):713–721. doi:10.1007/s10096-022-04423-6

16. Lodise TP, Zhao Q, Fahrbach K, et al. A systematic review of the association between delayed appropriate therapy and mortality among patients hospitalized with infections due to Klebsiella pneumoniae or Escherichia coli: how long is too long? BMC Infect Dis. 2018;18(1):625. doi:10.1186/s12879-018-3524-8

17. Choi MH, Kim D, Kim J, et al. Shift in risk factors for mortality by period of the bloodstream infection timeline. J Microbiol Immunol Infect. 2024;57(1):97–106. doi:10.1016/j.jmii.2023.11.008

18. Lan CC, Su WL, Yang MC, et al. Predictive role of neutrophil-percentage-to-albumin, neutrophil-to-lymphocyte and eosinophil-to-lymphocyte ratios for mortality in patients with COPD: evidence from NHANES 2011–2018. Respirology. 2023;28(12):1136–1146. doi:10.1111/resp.14589

19. Ji W, Li H, Qi Y, et al. Association between neutrophil-percentage-to-albumin ratio (NPAR) and metabolic syndrome risk: insights from a large US population-based study. Sci Rep. 2024;14(1):26646. PMID: 39496695; PMCID: PMC11535182. doi:10.1038/s41598-024-77802-y

20. Dong K, Zheng Y, Wang Y, et al. Predictive role of neutrophil percentage-to-albumin ratio, neutrophil-to-lymphocyte ratio, and systemic immune-inflammation index for mortality in patients with MASLD. Sci Rep. 2024;14(1):30403. PMID: 39638820; PMCID: PMC11621551. doi:10.1038/s41598-024-80801-8

21. Ding W, La R, Wang S, et al. Associations between neutrophil percentage to albumin ratio and rheumatoid arthritis versus osteoarthritis: a comprehensive analysis utilizing the NHANES database. Front Immunol. 2025;16:1436311. PMID: 39917306; PMCID: PMC11799277. doi:10.3389/fimmu.2025.1436311

22. Yang F, Dong R, Wang Y, et al. Prediction of pulmonary infection in patients with severe myelitis by NPAR combined with spinal cord lesion segments. Front Neurol. 2024;15:1364108. PMID: 38481940; PMCID: PMC10932995. doi:10.3389/fneur.2024.1364108

23. Zawiah M, Khan AH, Abu Farha R, et al. Predictors of stroke-associated pneumonia and the predictive value of neutrophil percentage-to-albumin ratio. Postgrad Med. 2023;135(7):681–689. PMID: 37756038. doi:10.1080/00325481.2023.2261354

24. Luo R, Yao Y, Chen Z, et al. An examination of the LPS-TLR4 immune response through the analysis of molecular structures and protein-protein interactions. Cell Commun Signal. 2025;23(1):142. PMID: 40102851; PMCID: PMC11921546. doi:10.1186/s12964-025-02149-4

25. Tsutsumi N, Masoumi Z, James SC, et al. Structure of the thrombopoietin-MPL receptor complex is a blueprint for biasing hematopoiesis. Cell. 2023;186(19):4189–4203.e22. PMID: 37633268; PMCID: PMC10528194. doi:10.1016/j.cell.2023.07.037

26. Xia Y, Sun C, Zhou K, et al. Platelet glycoprotein Ibα cytoplasmic tail exacerbates thrombosis during bacterial sepsis. Int J Mol Sci. 2024;25(21):11548. PMID: 39519103; PMCID: PMC11546206. doi:10.3390/ijms252111548

27. Nasrollahian S, Graham JP, Halaji M. A review of the mechanisms that confer antibiotic resistance in pathotypes of E. coli. Front Cell Infect Microbiol. 2024;14:1387497. PMID: 38638826; PMCID: PMC11024256. doi:10.3389/fcimb.2024.1387497

28. Sundaramoorthy NS, Shankaran P, Gopalan V, et al. New tools to mitigate drug resistance in Enterobacteriaceae – escherichia coli and Klebsiella pneumoniae. Crit Rev Microbiol. 2023;49(4):435–454. PMID: 35649163. doi:10.1080/1040841X.2022.2080525

29. Katip W, Rayanakorn A, Oberdorfer P, et al. Short versus long course of colistin treatment for carbapenem-resistant A. baumannii in critically ill patients: a propensity score matching study. J Infect Public Health. 2023;16(8):1249–1255. PMID: 37295057. doi:10.1016/j.jiph.2023.05.024

30. Katip W, Rayanakorn A, Sornsuvit C, et al. High-loading-dose colistin with nebulized administration for carbapenem-resistant acinetobacter baumannii pneumonia in critically ill patients: a retrospective cohort study. Antibiotics. 2024;13(3):287. PMID: 38534721; PMCID: PMC10967279. doi:10.3390/antibiotics13030287