Predicting outcomes in undifferentiated prolonged fever presenting to the emergency department: A prospective study from North India

Abstract

OBJECTIVES: Undifferentiated prolonged fever presenting to the emergency department (ED) poses significant diagnostic and prognostic challenges, particularly in low and middle income countries. We aimed to describe the clinical spectrum and in hospital outcomes of adults presenting to the ED with undifferentiated prolonged fever and to identify the baseline predictors of mortality.

METHODS: In this prospective observational study, consecutive patients aged≥13years who fulfilled Durack and Street’s criteria for fever of unknown origin were enrolled between March 2022 and September 2024 at the ED of a tertiary care center in North India. Patients with immunosuppressive states or nosocomial fever were excluded. All patients underwent a standardized diagnostic evaluation guided by potentially diagnostic clues. Baseline demographic, clinical, and laboratory variables recorded at ED presentation, including complete blood count derived inflammatory indices, were evaluated as potential predictors of in hospital mortality using receiver operating characteristic (ROC) analysis and multivariable logistic regression.

RESULTS: A total of 100 patients were enrolled (mean age: 39 years; 57% male), with a median fever duration of 45 days (interquartile range: 30–90). Infections were the most common etiology (37%), with tuberculosis predominating, followed by malignancies (21%), predominantly advanced hematological malignancies, and noninfectious inflammatory diseases (19%), which frequently presented with multisystem involvement, most commonly systemic lupus erythematosus and Still’s disease. Despite extensive evaluation, 19 patients (19%) remained undiagnosed. Overall, 25 patients (25%) died during hospitalization, with the highest mortality observed among undiagnosed cases (8/19, 42%). In the univariate analysis, neutrophil percentage, neutrophil to lymphocyte ratio(NLR), aspartate transaminase, and lactate dehydrogenase levels were significantly higher in the mortality group, whereas total protein and albumin levels were significantly lower. ROC analysis demonstrated modest discriminatory ability for NLR (area under the curve: 0.651). In multivariable analysis, admission NLR emerged as the only independent baseline predictor of in hospital mortality (odds ratio: 1.089 per one unit increase; 95% confidence interval: 1.006–1.178; P = 0.034).

CONCLUSION: Undifferentiated prolonged fever presenting to the ED is associated with substantial in hospital mortality and diagnostic uncertainty. Admission NLR was independently associated with mortality, but discrimination was modest; it may have adjunctive value for early risk stratification pending external validation.

Introduction

Undifferentiated prolonged fever remains one of the most challenging clinical presentations in acute and emergency care. Traditionally described within the framework of fever of unknown origin (FUO), it is defined as a fever ≥38.3°C lasting at least 3 weeks without a diagnosis despite structured evaluation.[1-3] It encompasses a broad and evolving spectrum of infections, noninfectious inflammatory diseases (NIIDs), malignancies, and miscellaneous conditions, influenced by regional epidemiology, host immune status, and the availability of diagnostic resources.[1-5]

Most existing FUO literature originates from inpatient wards or outpatient clinics, where evaluation is sequential, and patients are relatively stable. [1-5] However, a clinically significant subset of patients with undifferentiated prolonged fever first present to the emergency department (ED), often with acute systemic illness, organ dysfunction, or sepsis like features, and without prior definitive diagnostic evaluation. In this setting, ED physicians must balance early resuscitation, risk stratification, and disposition decisions under significant time and resource constraints, often before the etiology of fever is established. Despite these challenges, undifferentiated prolonged fever presenting to the ED remains poorly characterized. In resource constrained settings, where the ED frequently serves as the primary point of care, delayed referral pathways and limited access to advanced diagnostics may further influence disease severity and prognosis at presentation.

Classically, FUO has been considered a relatively benign condition with a good prognosis, including spontaneous resolution of fever in a substantial proportion of undiagnosed cases without specific therapy.[6] Limited available evidence, however, suggests that ED presentations of FUO may not follow this benign course. A retrospective Australian study conducted over two decades ago reported that although such presentations were relatively uncommon, they were associated with disproportionately high rates of bacteremia and poor clinical outcomes.[7] However, prospective data evaluating the outcomes among patients presenting to the ED with undifferentiated prolonged fever are lacking.

To address this gap, we conducted a prospective observational study to evaluate the baseline predictors of in hospital mortality among adult patients presenting to the ED with undifferentiated prolonged fever. We hypothesized that baseline laboratory markers reflecting systemic inflammatory burden – particularly complete blood count derived indices – would be independently associated with in hospital mortality. As a secondary objective, we aimed to describe the clinical spectrum and etiological distribution of undifferentiated prolonged fever in this ED cohort at a large tertiary care center in North India.

Material and Methods

Study design and setting

This prospective observational study was conducted from March 2022 to September 2024 in the Acute Care and Emergency Medicine (ACEM) Unit of the Department of Internal Medicine at the Postgraduate Institute of Medical Education and Research (PGIMER), Chandigarh, India. PGIMER, Chandigarh is a large tertiary care academic referral center serving patients from multiple states in North India. The ACEM unit manages a high volume of emergency admissions and includes emergency rooms, observation units, and a high dependency unit providing care for critically ill medical patients. The study was approved by the Institutional Ethics Committee of PGIMER, Chandigarh (INT/IEC/2022/00648, dated March 16, 2022). Written informed consent was obtained from all participants or their legally authorized representatives. This study was designed and reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.

Study population

Consecutive patients aged ≥13 years admitted to the ACEM unit with undifferentiated prolonged fever who fulfilled Durack and Street’s criteria for classic FUO were prospectively assessed for eligibility. Classic FUO was defined as fever ≥38.3°C on multiple occasions for at least 3 weeks, with no established diagnosis after three outpatient visits or 3days of inpatient evaluation.[3] Patients with known immunosuppressive states (including human immunodeficiency virus [HIV] infection, posttransplant status, or chemotherapy induced neutropenia) and those with nosocomial FUO were excluded.

Diagnostic evaluation and patient management

All enrolled patients were evaluated using a standardized diagnostic protocol for FUO guided by potentially diagnostic clues, defined as localizing clinical features or laboratory abnormalities suggestive of a specific etiology.[2,5] Baseline evaluation included detailed clinical assessment and laboratory investigations, including complete blood count (with peripheral smear), biochemical profile, coagulation parameters, blood and other body fluid cultures, urinalysis, C reactive protein, HIV testing, chest radiography, and ultrasonography of the abdomen.

Cross sectional imaging with contrast enhanced computed tomography (CT) and ¹⁸F fluorodeoxyglucose positron emission tomography combined with CT (FDG PET/CT) was performed when clinically indicated. Image guided biopsies and bone marrow examinations were undertaken based on clinical or radiological findings. Final diagnoses were established using the standard diagnostic criteria and guideline recommended definitions.

Following enrollment, patients were admitted to the observation unit or high dependency unit, depending on clinical severity and bed availability. Patients were followed throughout hospitalization until discharge, death, or leaving against medical advice.

Outcomes

In hospital outcomes – including discharge, death, or leaving against medical advice – were recorded. Baseline demographic, clinical, and laboratory variables recorded at ED presentation were evaluated as potential predictors of in hospital mortality, with particular emphasis on complete blood count derived inflammatory indices, including the neutrophil to lymphocyte ratio (NLR; neutrophil count/lymphocyte count), monocyte–lymphocyte ratio (monocyte count/ lymphocyte count), platelet–lymphocyte ratio (platelet count/lymphocyte count), systemic inflammatory response index (neutrophil count × monocyte count/ lymphocyte count), and systemic immune–inflammation index (platelet count × neutrophil count/lymphocyte count). Patients who left against medical advice were excluded from the analyses of mortality predictors.

Statistical analysis

Data were analyzed using the Statistical Package for the Social Sciences (SPSS) software version 25.0 (IBM Corp., Armonk, NY, USA). The categorical variables were summarized as frequencies and percentages, and continuous variables were expressed as mean ± standard deviation (SD) or median with interquartile range (IQR), depending on whether the data were normally distributed. Normality of continuous variables was assessed using the Shapiro–Wilk test. Group comparisons were performed using the chi square test for categorical variables and the Mann–Whitney U test or Student’s t test for continuous variables, depending on data distribution.

Variables demonstrating significant association on univariate analysis were evaluated for their discriminatory ability for in hospital mortality using receiver operating characteristic (ROC) curve analysis with the calculation of the area under the curve (AUC). Variables demonstrating fair discriminatory ability and considered clinically relevant were entered into multivariable logistic regression to identify the independent predictors of mortality. The number of predictors included in the model was limited to minimize overfitting, given the limited number of mortality events. Model calibration was assessed using the Hosmer– Lemeshow goodness of fit test. Multicollinearity among candidate predictors was evaluated using correlation matrices.

A two sided P < 0.05 was considered statistically significant. As this was a prospective cohort study, all consecutive eligible patients presenting during the study period were enrolled; a formal a priori sample size calculation was not performed.

Results

Baseline characteristics

During the study period, 103 consecutive patients fulfilled criteria for FUO. Three patients were excluded due to underlying HIV infection, and 100 patients were included in the final analysis [Figure 1]. The mean age was 38.8 ± 17.6 years, and 57 were male. The median duration of fever at ED presentation was 45 days (IQR: 30–90), and 77 patients had a documented temperature ≥ 39°C at ED admission. The common presenting symptoms included weight loss (71%), cough (49%), dyspnea (33%), and abdominal pain (17%). On examination, hepatosplenomegaly was present in 60% of patients, lymphadenopathy in 36%, and skin rash in 18%.

Figure 1. Flow diagram of eligibility assessment and inclusion. ED: Emergency department, FUO: Fever of unknown origin, HIV: Human immunodeficiency virus

Diagnostic investigations

Whole body imaging was performed in 46 patients (contrast enhanced CT in 38, FDG PET/CT in 19, and both in 11), yielding potentially diagnostic clues in 36 patients (78.3%). Imaging guided tissue sampling was undertaken in 29 patients and established the final diagnosis in 21 (72.4%). Bone marrow examination was performed in 26 patients and contributed to the final diagnosis in 14 (53.8%). Blood cultures were obtained at admission in all patients and were diagnostically contributory in four cases.

Etiological spectrum

The etiological distribution is summarized in Table 1. Infections were the most common cause (n = 37).

Table 1. Etiological spectrum of undifferentiated prolonged fever (n=100)

Tuberculosis was the single most frequent diagnosis (n = 15), including disseminated (n = 10) and extrapulmonary disease (n = 5). Diagnosis was established by acid fast bacilli staining (n = 6), GeneXpert(n = 2), histopathology (n = 2), or characteristic clinicoradiological features (n = 5). Thirteen patients improved on antituberculous therapy; two died due to advanced disease.

Tropical infections (n = 9) included enteric fever (n = 3), scrub typhus (n = 2), malaria (n = 1), dengue (n = 1), and undifferentiated tropical fevers (n = 2). Malaria was diagnosed on repeat peripheral smear examination. Other tropical infections (except undifferentiated cases) were serologically confirmed. Thrombocytopenia and hepatosplenic involvement were common, and all patients improved.

Three patients with complicated urinary tract infections presented with acute kidney injury. Diagnosis was established based on repeat urinalysis and CT imaging; one had Escherichia coli growth on repeat cultures. One patient died due to sepsis and multiorgan failure.

Two patients had nonresolving pneumonia with multiorgan dysfunction. Diagnosis was established through repeated respiratory sampling and thoracic CT. One patient with dual E. coli–Aspergillus infection died due to progressive respiratory failure with multiorgan dysfunction.

Infective endocarditis was culture proven (n = 2; Streptococcus species and Pseudomonas aeruginosa complicating a coronary stent pseudoaneurysm). One patient had Klebsiella pneumoniae bacteremia without an identifiable primary focus and died due to septic shock. Additional infectious causes included esophageal candidiasis diagnosed on endoscopy (n = 1) and an infected pancreatic collection identified on whole body imaging (n = 1). Visceral leishmaniasis (n = 1) was diagnosed by rK39 serology and polymerase chain reaction and successfully treated with amphotericin B. Progressive disseminated histoplasmosis (n = 1) was confirmed on lymph node biopsy; this patient died due to advanced disease.

Malignancy (n = 21) included 19 hematological and 2 solid organ cancers. Lymphoid malignancies predominated (n=13), including non Hodgkin lymphoma (n=8; diffuse large B cell, n=5; angioimmunoblastic T cell, n = 2; anaplastic large cell, n = 1) and Hodgkin lymphoma (n = 5). All lymphomas were Stage IV at presentation. Acute leukemia (n = 5) and myeloproliferative neoplasm (n = 1; primary myelofibrosis) required bone marrow examination for the diagnosis due to absent peripheral blasts and noncontributory initial imaging. Solid tumors included hepatic angiosarcoma (n = 1) and metastatic squamous cell carcinoma (n = 1), diagnosed by liver and retroperitoneal biopsies, respectively. Three patients with malignancy (two with acute leukemia and one with Hodgkin lymphoma) died due to advanced or progressive disease.

Among noninfectious inflammatory diseases (n = 19), systemic lupus erythematosus (SLE; n = 8) was the most common diagnosis and uniformly presented with multisystem involvement. Three patients with SLE died (two due to multiorgan dysfunction and one due to superadded fungal sepsis). Patients with Still’s disease (n = 5) presented with neutrophilic leukocytosis, marked hyperferritinaemia (>2000 ng/mL), and serositis. As diagnosis required exclusion of alternative causes, extensive evaluation was performed, including CT (n=5), FDG PET/CT (n = 2), bone marrow examination (n = 4), and skin biopsy (n = 2). One patient with Still’s disease died due to a progressive hyperinflammatory state.

Two patients with anti-melanoma differentiationassociated gene 5 (MDA5) dermatomyositis developed rapidly progressive interstitial lung disease and died of respiratory failure. Both cases of vasculitis responded to immunosuppressive therapy. Sarcoidosis was confirmed by endobronchial ultrasound guided biopsy. Silicosis associated pulmonary alveolar proteinosis was confirmed by open lung biopsy, and the patient died of progressive respiratory failure.

Miscellaneous causes included hemophagocytic lymphohistiocytosis (HLH) without identifiable triggers (n = 3) and thrombotic microangiopathy (n = 1). HLH was diagnosed using HLH 2004 criteria; one patient died due to progressive hyperinflammatory syndrome. Thrombotic microangiopathy was diagnosed based on schistocytes on repeat peripheral smear and resolved with therapeutic plasma exchange.

Despite extensive evaluation, no definitive etiology was identified in 19 patients. Among these, only four underwent whole body imaging, which did not yield potentially diagnostic clues. Eight of the 19 undiagnosed patients died, predominantly due to progressive multiorgan dysfunction.

Outcomes

The median time from ED admission to definitive diagnosis was 6.5 days (IQR 3–15). Twenty six patients required intensive care unit admission, of whom 15 required mechanical ventilatory support. The median hospital stay was 10 days (IQR: 5–16), shorter in the undiagnosed group (7 days, IQR: 3.5–12) than in diagnosed cases (11 days, IQR: 6–18).

Overall, 25 patients (25%) died during hospitalization, and five left against medical advice. Mortality was higher among undiagnosed cases (8/19, 42%) compared with those with established diagnoses (17/81, 21%).

Predictors of in hospital mortality

In the univariate analysis, neutrophil percentage, NLR, aspartate transaminase (AST), and lactate dehydrogenase (LDH) levels were significantly higher in the mortality group, whereas total protein and albumin levels were significantly lower [Table 2]. ROC analysis demonstrated modest discriminatory ability for LDH (AUC: 0.727; 95% confidence interval [CI]: 0.595–0.860; P = 0.002), AST (AUC: 0.677; 95% CI: 0.523– 0.830; P = 0.018), NLR (AUC: 0.651; 95% CI: 0.506–0.796; P = 0.043), and neutrophil percentage (AUC: 0.648; 95% CI: 0.506–0.790; P = 0.048). Albumin and total protein showed inverse, nonsignificant discrimination.

Table 2. Univariate analysis of baseline predictors of in‑hospital mortality

Multivariable logistic regression analysis, including NLR, albumin, and LDH, showed that admission NLR remained the only independent predictor of in hospital mortality (odds ratio: 1.089 per one unit increase; 95% CI: 1.006–1.178; P = 0.034), whereas albumin and LDH did not retain statistical significance [Table 3]. The model demonstrated acceptable calibration (Hosmer– Lemeshow P = 0.272) with modest explanatory power (Nagelkerke R² = 0.213). In an alternative model that included AST in place of albumin, NLR remained the only independent predictor, whereas AST and LDH did not retain statistical significance.

Table 3. Multivariate analysis of baseline predictors of in‑hospital mortality

Discussion

This prospective study demonstrates that undifferentiated prolonged fever presenting to the ED represents a high risk clinical entity. In contrast to traditional FUO cohorts – often derived from ward or clinic settings – our ED cohort was characterized by acute multisystem illness, advanced disease at presentation, and high in hospital mortality. A substantial proportion of patients remained without a definitive diagnosis despite extensive evaluation. Admission NLR was independently associated with in hospital mortality; however, its discriminatory performance was modest.

Although infections remained the most common etiological category, their relative contribution differed from many classical FUO series from low and middle income countries, where indolent infections, particularly tuberculosis, often predominate.[1,4,5,8] While tuberculosis remained frequent in our cohort, its proportional dominance was attenuated, with a broader spectrum of acute bacterial and endemic infections becoming more prevalent. This pattern suggests presentation during a more dynamic and clinically unstable phase of infectious illness in the ED, in which classical diagnostic features may be incomplete or evolving.[9] Similarly, malignant and NIID etiologies frequently presented with advanced or rapidly progressive multisystem involvement mimicking sepsis.[8,10] Together, these overlapping infectious, inflammatory, and malignant phenotypes underscore the diagnostic complexity of undifferentiated prolonged fever in the ED.

Notably, nearly one fifth of patients remained undiagnosed, and this group experienced the highest mortality. This likely reflects both early disease evolution and the practical constraints inherent to ED based care. Advanced diagnostic modalities such as FDG PET/CT and image guided biopsies, while valuable in classical FUO algorithms, may not be immediately feasible in unstable patients or within the time limited ED environment.[1,2,4] Consequently, undifferentiated prolonged fever presenting to the ED cannot be approached using the serial, stepwise diagnostic pathways designed for stable inpatient or outpatient populations.[1,2,5,11] Instead, ED evaluation must prioritize simultaneous physiological stabilization and early source localization, with expedited imaging and targeted tissue sampling when feasible, alongside careful reassessment to avoid diagnostic anchoring.

Classical FUO series have largely emphasized etiological classification and diagnostic yield, with limited focus on short term outcomes. In many such cohorts, not only do diagnosed cases have a favorable prognosis, but undiagnosed cases often experience spontaneous resolution of fever.[4,6,12] In contrast, our findings highlight that FUO presenting to the ED is associated with substantial mortality. This divergence underscores the importance of recognizing ED presentations of prolonged fever as a prognostically distinct entity, in which early outcome assessment may be as critical as etiological diagnosis. As emphasized by Durack and Street, FUO must be interpreted within its clinical context, as prognosis, diagnostic priorities, and urgency vary across patient groups.[3] Consistent with this principle, the outcomes observed in our emergency department cohort underscore the need for early risk stratification before etiological clarity is achieved.

In this context, admission NLR was independently associated with in hospital mortality. Prior emergency and critical care studies have consistently demonstrated associations between elevated NLR and adverse outcomes across heterogeneous conditions, including severe infections, inflammatory disease flares, and malignancy associated hyperinflammatory states.[13-20] Notably, reported “high risk” thresholds vary considerably, underscoring the context dependent nature of NLR rather than a universal cut off. Accordingly, NLR was analyzed as a continuous variable in our cohort. Although the per unit effect size was modest, increasing NLR values were associated with higher odds of mortality. NLR may reflect the degree of early systemic inflammatory response at ED presentation. Given its availability and low cost, it may have adjunctive value in early ED assessment of undifferentiated prolonged fever, supporting closer monitoring and expedited diagnostic evaluation. However, its modest discriminatory performance indicates that it should not be used as a standalone prognostic tool. These findings are hypothesis generating and warrant validation in larger, multicenter cohorts before routine clinical application.

Limitations

This study has limitations. First, it was conducted at a single tertiary care referral ED, which may limit the generalizability to other emergency care settings with differing patient acuity, referral patterns, and resource availability. Second, advanced investigations, such as FDG PET/CT, image guided tissue sampling, and metagenomic testing, could not be performed uniformly across all patients due to clinical instability or logistical constraints inherent to emergency care, potentially reducing the etiological yield, particularly in undiagnosed cases. Third, postdischarge follow up was limited, and some diagnoses may have been established after hospitalization. Fourth, the sample size was determined by patient accrual over the study period and may have limited statistical power to detect the smaller associations. Given the limited number of mortality events, the number of variables included in multivariable modeling was restricted to minimize overfitting. Only admission clinical and laboratory parameters were analyzed, consistent with the primary objective of evaluating early predictors of outcome. Time dependent factors such as disease progression, treatment escalation, intensive care interventions, and mechanical ventilation were not incorporated into the regression models; therefore, residual confounding by underlying disease severity cannot be entirely excluded, as is common in observational prognostic studies conducted in acute care settings.

Conclusion

Undifferentiated prolonged fever presenting to the ED represents a high risk clinical entity distinct from classical FUO, characterized by diagnostic uncertainty and substantial in hospital mortality, and requiring early prognostication. Admission NLR was independently associated with in hospital mortality; however, its discriminatory performance was modest. As a readily available and inexpensive marker, NLR may have adjunctive value in early ED evaluation, but it should not be considered a standalone prognostic tool. These findings warrant validation in larger, multicenter cohorts before routine clinical implementation.

† These authors contributed equally to the manuscript and share first authorship.

How to cite this article: Pannu AK, Kanneganti V, Selvam S, Gupta P, Kumar R, Dhibar DP, etal. Predicting outcomes in undifferentiated prolonged fever presenting to the emergency department: A prospective study from North India. Turk J Emerg Med 2026;26:306-13.

The Institutional Ethics Committee of Postgraduate Institute of Medical Education and Research, Chandigarh (India) approved the study (No.: INT/IEC/2022/00648, date March 16, 2022).

AKP: Conceptualization (lead); Methodology (lead); Formal analysis (lead); Writing original draft (supporting), Writing review and editing (lead)
VK: Data curation (lead); Writing original draft (lead)
SS: Data curation (supporting); Writing original draft (supporting)
PG: Data curation (supporting); Writing original draft (supporting)
RK: Data curation (supporting); Writing review and editing (supporting)
DPD: Data curation (supporting); Writing review and editing (supporting)
NS: Data curation (supporting); Writing - review and editing (supporting)
Ashok Kumar Pannu and Vineetha Kanneganti contributed equally to the manuscript and are the joint first authors.
The corresponding author is responsible for ensuring that the descriptions are accurate and agreed upon by all authors.

None.

References

1. Haidar G, Singh N. Fever of unknown origin. N Engl J Med 2022;386:463 77.
2. Wright WF, Durso SC, Forry C, Rovers CP. Fever of unknown origin. BMJ 2025;388:e080847.
3. Durack DT, Street AC. Fever of unknown origin – Reexamined and redefined. Curr Clin Top Infect Dis 1991;11:35 51.
4. Pannu AK, Golla R, Kumari S, Suri V, Gupta P, Kumar R. Aetiology of pyrexia of unknown origin in North India. Trop Doct 2021;51:34 40.
5. Pannu AK. Approach to pyrexia of unknown origin in low and middle income countries. Trop Doct 2025;55:364 7.
6. Mulders Manders CM, Engwerda C, Simon A, van der Meer JW, Bleeker Rovers CP. Long term prognosis, treatment, and outcome of patients with fever of unknown origin in whom no diagnosis was made despite extensive investigation: A questionnaire based study. Medicine (Baltimore) 2018; 97:e11241.
7. Ingarfield SL, Celenza A, Jacobs IG, Riley TV. Outcomes in patients with an emergency department diagnosis of fever of unknown origin. Emerg Med Australas 2007;19:105 12.
8. Yadav BK, Pannu AK, Kumar R, Rohilla M, Kumari S. Fever of unknown origin in older adults: A prospective observational study from North India. J Assoc Physicians India 2021; 69:11 2.
9. Hernandez Trujillo HS, Dalberg T, Feder H Jr., Smith SR. A fever of unknown origin workup in the emergency department reveals an unusual pathogen. Pediatr Emerg Care 2009;25:684 6.
10. Pannu AK, Singla V, Suri V, Kumar R, Mathur Y, Mohindra R, et al. Adult onset still’s disease and fever of unknown origin in India. Clin Exp Med 2023;23:1659 66.
11. WrightWF, Betz JF, Auwaerter PG. Prospective studies comparing structured versus nonstructured diagnostic protocol evaluations among patients with fever of unknown origin: A systematic review and meta analysis. JAMA Netw Open 2022;5:e2215000.
12. Vanderschueren S, Eyckmans T, De Munter P, Knockaert D. Mortality in patients presenting with fever of unknown origin. Acta Clin Belg 2014;69:12 6.
13. Lionte C, Bologa C, Sorodoc V, Petris OR, Puha G, Stoica A, et al. Biomarkers of inflammation and inflammation related indexes upon emergency department admission are predictive for the risk of intensive care unit hospitalization and mortality in acute poisoning: A 6 year prospective observational study. Dis Markers 2021;2021:4696156.
14. CheblRB, AssafM, Kattouf N, Haidar S, KhamisM, Abdeldaem K, et al. The association between the neutrophil to lymphocyte ratio and in hospital mortality among sepsis patients: A prospective study. Medicine (Baltimore) 2022;101:e29343.
15. Ho TH, Lin JW, Chi YC, How CK, Chen C T . Neutrophil to lymphocyte ratio as a predictor for outcomes in patients with short term emergency department revisits. J Chin Med Assoc 2024;87:782 8.
16. Boga DE. Relationship between neutrophil to lymphocyte ratio and sepsis and mortality in patients admitted to the emergency department: A retrospective observational study. Eastern J Med Sci 2025;10:43 7.
17. Gao Z, Wang X, Wang W, Kang Z, Chen X. Association between neutrophil to lymphocyte ratio and the mortality of patients with sepsis: An update systematic review and meta analysis. Front Med (Lausanne) 2025;12:1637365.
18. Shukla R, Silvanus V, Dahal S, Sitaula P, Niraula Z, Gupta CK, et al. Complete blood count (CBC) derived indices at admission and short term outcomes in ICU patients: A retrospective study. Ann Med 2025;57:2522963.
19. Wen X, Zhang Y, Xu J, Song C, Shang Y, Yuan S, et al. The early predictive roles of NLR and NE% in in hospital mortality of septic patients. Heliyon 2024;10:e26563.
20. Galardo G, Crisanti L, Gentile A, Cornacchia M, Iatomasi F, Egiddi I, et al. Neutrophil to lymphocyte ratio (NLR) and short term mortality risk in elderly acute medical patients admitted to a university hospital emergency department. Intern Emerg Med 2025;20:553 62.

The authors thank Mrs. Sunaina Verma for her help in statistics.