European Journal of Cardiovascular Medicine

Sepsis is a life-threatening condition that continues to be a major global health burden, significantly contributing to morbidity and mortality. In 2017, it was responsible for approximately 11 million deaths, accounting for 19.7% of all global fatalities.¹ Annually, an estimated 49 million people develop sepsis worldwide, with mortality rates approaching 20%.¹ In China, the standardized incidence of hospitalized sepsis was reported as 421.85 per 100,000 individuals in 2019.² ICU patients with sepsis experience particularly poor outcomes, with studies reporting mortality rates of up to 41.9%, and in septic shock cases, this can rise as high as 60%. Sepsis is characterized by a dysregulated host response to infection, resulting in systemic inflammation, oxidative stress, and multi-organ dysfunction.³

Accurate and timely prognostication in sepsis is essential for effective management, but existing tools such as laboratory markers and scoring systems like APACHE have limited predictive accuracy and are time-consuming.⁴ Early recognition remains a challenge, especially in milder presentations where life-saving interventions may be delayed. Hence, simpler and more cost-effective prognostic markers are needed to improve outcomes.⁵

Serum albumin is a multifunctional protein involved in maintaining oncotic pressure, transporting various molecules, and providing antioxidant defense. Hypoalbuminemia is associated with increased disease severity and poor prognosis in sepsis.⁶ When combined with hematological markers like lactic acid or neutrophil percentages, albumin enhances prognostic accuracy. Fibrinogen, a coagulation factor, is another important marker, with lower levels linked to higher mortality in both adult and neonatal sepsis.⁷

RDW, a measure of red blood cell size variability, is influenced by inflammation and oxidative stress and is associated with poor outcomes in sepsis and other critical illnesses. The RDW-to-albumin ratio (RAR) is a novel marker showing prognostic relevance in conditions such as ARDS and stroke,⁸ but its role in sepsis remains underexplored. Similarly, the neutrophil-to-lymphocyte count ratio (NLCR) has demonstrated superior predictive value for bacteremia and sepsis compared to conventional markers.⁹

In resource-limited settings like India, where ICU access and diagnostic capabilities are constrained, these biomarkers offer a practical and efficient alternative for risk assessment. This study aims to evaluate the prognostic significance of serum albumin, fibrinogen, RDW, and NLCR in sepsis patients.

This was an observational, hospital-based study conducted over 18 months (from IEC approval to August 2024) in the Department of General Medicine, Mahatma Gandhi Medical College & Hospital, Jaipur. After obtaining IEC approval and informed written consent, consecutive patients admitted with fever and suspected sepsis were enrolled.

Inclusion Criteria:

1. Suspected sepsis cases admitted to ICU/HDU
2. Age between 18–65 years

Exclusion Criteria:

1. Unwilling patients
2. Pregnant women
3. Post-operative patients
4. Patients with malignancy or chemotherapy in the past 6 months
5. Patients on immunosuppressants or RBC morphology-altering drugs

All eligible patients underwent detailed clinical evaluation, including history, comorbidities, and physical examination. Blood samples were collected on admission and after 48 hours for CBC, biochemistry, coagulation profile, and relevant sepsis investigations. SOFA scores were recorded at both time points. Microbiological cultures from suspected infection sources were obtained on admission. Patients were followed throughout their hospital stay to assess clinical outcomes.

Statistical Analysis:

Data were entered in Microsoft Excel and analyzed using SPSS v2022. Descriptive statistics (means, SD, frequencies) summarized baseline data. Chi-square test, Student’s t-test, and ANOVA were used as appropriate. A p-value < 0.05 was considered statistically significant

Table 1: Comparison of Biomarkers Between Survivors and Non-Survivors in Sepsis Patients.

On admission, the mean NLR was significantly higher in non-survivors (17.92) than survivors (13.02) (p = 0.02), and rose further at 48 hours (26.88 vs. 10.43; p < 0.0001). Serum albumin was lower in non-survivors at both admission (2.71 vs. 3.06; p = 0.007) and 48 hours (2.40 vs. 2.76; p = 0.01). RDW was also higher in non-survivors at admission (17.54 vs. 16.80; p = 0.02) and at 48 hours (18.63 vs. 16.24; p < 0.0001). Plasma fibrinogen levels showed no significant difference.

SOFA scores were similar at admission (p = 0.42) but significantly higher in non-survivors at 48 hours (14.85 vs. 7.86; p < 0.0001). Patients with ≤7-day hospital stay had higher SOFA at 48 hours (10.01 vs. 6.52; p = 0.009). Out of 120 patients, 73 (52.14%) had positive cultures, with no significant difference between survivors and non-survivors (p = 0.125).

Table 2: Distribution of cases according to Mean Arterial Pressure.

On admission, mean MAP was significantly higher in survivors (86.44 mmHg) than non-survivors (73.23 mmHg; p < 0.0001). At 48 hours, MAP remained higher in survivors but without statistical significance (p = 0.16). GCS was significantly better in survivors at admission (12.65 vs. 10.78; p = 0.01) and improved further at 48 hours (12.95 vs. 6.33; p < 0.0001). Platelet counts showed no significant differences between survivors and non-survivors at either time point.

Table 3: Distribution of the cases according to PaO2/FiO2.

On admission, the PaO₂/FiO₂ ratio was higher in survivors (268.09 mmHg) than non-survivors (223.78 mmHg), though not statistically significant (p = 0.06). At 48 hours, survivors improved significantly (309.9 mmHg) while non-survivors declined sharply (118.11 mmHg; p < 0.0001). Bilirubin levels showed no difference at admission but were significantly higher in non-survivors at 48 hours (2.49 vs. 1.53 mg/dL; p = 0.006). Creatinine was unexpectedly higher in survivors on admission (3.47 vs. 1.45 mg/dL; p = 0.0003), but showed no difference at 48 hours (p = 0.94).

Figure 1: ROC Analysis of NLR as a prognostic marker of outcome.

The ROC analysis showed NLR had moderate predictive value at admission (AUC 0.668, cut-off 11) with high sensitivity but low specificity, improving significantly at 48 hours (AUC 0.848, cut-off 11.1) with better specificity. In contrast, GCS had poor predictive value both at admission (AUC 0.327) and 48 hours (AUC 0.02), with low sensitivity and specificity. Thus, NLR at 48 hours is a more reliable mortality predictor than GCS.

Figure 2: ROC Analysis of S. Albumin as a prognostic marker of outcome

The ROC analysis showed that serum albumin had low predictive value both at admission (AUC 0.326, cut-off 2.25, sensitivity 75%, specificity 18.4%) and at 48 hours (AUC 0.35, cut-off 2, sensitivity 70.8%, specificity 10.5%). In contrast, SOFA scores had poor prediction at admission (AUC 0.451, cut-off 7.5) but significantly improved at 48 hours (AUC 0.94, cut-off 12.5) with high sensitivity (87.5%) and specificity (85.5%).

Figure 3: ROC Analysis of Plasma Fibrinogen as a prognostic marker of outcome

Plasma fibrinogen showed no significant predictive value at admission (AUC 0.533, p = 0.622) with a cut-off of 312.3, sensitivity of 87.5%, and low specificity (27.6%). At 48 hours, its predictive ability declined further (AUC 0.492, p = 0.91) with a lower cut-off of 206, sensitivity of 75%, and very low specificity (15.8%).

Figure 4: ROC Analysis of RDW as a prognostic marker of outcome

RDW showed good predictive value at admission with an AUC of 0.717 (p = 0.005), using a cut-off of 17.6 that yielded 83.3% sensitivity and 68.8% specificity. At 48 hours, its predictive accuracy improved further with an AUC of 0.877 (p < 0.0001), a cut-off of 17.8, maintaining 83.3% sensitivity and increasing specificity to 85.9%.

Sepsis is a life-threatening immunologic response from the host to an infection, resulting in multiple organ dysfunctions. Timely recognition and intervention are crucial for improving survival rates and prevention of progression to septic shock, which carries ≥ 40% mortality rate.¹⁰ While microbiological culture has traditionally been the gold standard for diagnosing sepsis, its efficacy is limited by time constraints and reduced yield due to prior antibiotic use and potential contamination during inoculation. Given these limitations, there is ongoing research to identify new, effective markers for diagnosing sepsis with better diagnostic and prognostic value. Additionally, numerous studies are exploring prognostic factors to decrease sepsis-related mortality.¹¹ However, complete identification of prognostic factors in sepsis patients remains elusive.

NLCR AS A PROGNOSTIC MARKER

In this study, NLR was significantly higher in non-survivors than survivors at admission (17.92 vs. 13.02; p = 0.02) and markedly elevated at 48 hours (26.88 vs. 10.43; p < 0.0001). ROC analysis showed NLR’s predictive value improved from admission (AUC 0.668, cut-off 11, sensitivity 87.5%, specificity 38.2%) to 48 hours (AUC 0.848, cut-off 11.1, sensitivity 87.5%, specificity 59.2%).

Kaushik R et al12 found that the the Neutrophil-to-Lymphocyte Ratio Index (NLRI) was found to be significantly higher in cases (8.5 ± 5.95) compared to controls (2.0 ± 0.94), with a statistical significance of p < 0.001. This indicates a strong association between elevated NLRI and the condition under study. De Jager et al.9 noted that NLCR serves as a superior predictor of severity and outcome in bacteraemia when compared to conventional markers. Zahorec identified NLCR as easily measurable and showing a stronger correlation with sepsis severity and 28-day mortality compared to neutrophilia, lymphocytopenia, and clinical course severity.31

In a study by Hota P K et al13 an NLCR less than 7 was found in twenty-one cases (~42%), while more than 7 was seen in twenty-nine cases (~58%). They reported an 86.2% sensitivity and 85.7% specificity for the neutrophil-lymphocyte ratio, suggesting its efficacy as a prognostic marker for predicting sepsis. Similarly, in a study by Yong Xia et al14, NLR demonstrated a sensitivity of 40.91% and specificity of 93.22%, indicating its utility as a prognostic marker in sepsis.

1. ALBUMIN AS A PROGNOSTIC MARKER

In our study, serum albumin was significantly lower in non-survivors than survivors at admission (2.71 vs. 3.06; p = 0.007) and further declined at 48 hours (2.4 vs. 2.76; p = 0.01). ROC analysis showed low predictive value for albumin at admission (AUC 0.326, cut-off 2.25, sensitivity 75%, specificity 18.4%) and at 48 hours (AUC 0.35, cut-off 2, sensitivity 70.8%, specificity 10.5%).

Shan X et al15 found thatAmong 402 patients, the median albumin level was 2.71 g/dL (IQR: 2.30–3.11). Survivors (n = 244) had a higher median albumin level of 2.83 g/dL (IQR: 2.48–3.26) compared to non-survivors (n = 158), who had a lower median level of 2.43 g/dL (IQR: 2.10–2.79). The difference was statistically significant (p < 0.001).

Ahmed A et al16 found that theSerum albumin levels were also significantly higher in survivors (4.41 ± 1.033 g/dL) than in non-survivors (3.11 ± 0.691 g/dL), with p < 0.001.

 A study by Saravanakumar G et al17 found that the mean serum albumin levels on day one was around 3.7 g/dL in the survivor group, whereas in the non-survivor group it was around  3.1 g/dL. The difference in mean serum albumin on day one was statistically significant with a p-value of less than 0.001. In a study conducted by Nirmala et al. (2015), a little higher serum albumin levels were observed in survivors compared to non-survivors on day one (3.4 ± 0.25 vs. 3.4 ± 0.30), but this difference was not significant  statistically.18               

PLASMA FIBRINOGEN AS A PROGNOSTIC MARKER

In our study, plasma fibrinogen levels showed no significant difference between survivors and non-survivors at admission (463.96 vs. 437.15; p = 0.59) or at 48 hours (414.78 vs. 421.01; p = 0.91). ROC analysis confirmed its poor predictive value with AUCs of 0.533 at admission and 0.492 at 48 hours, both non-significant, despite high sensitivity but low specificity at respective cut-offs.

Contrary to these findings, Tang X et al19 suggested that p.fibrinogen is a great prognostic marker for pediatric sepsis patients, with an AUC for hospital mortality of 0.780. They identified a cutoff level of fibrinogen for hospital mortality at 2.46 g/L upon admission to pediatric ICU, achieving a sensitivity of 66.67% and specificity of 82.00%, which aligns with our study. Shan X et al15 found thatAmong 402 patients, the median fibrinogen level was 5.10 g/L (IQR: 3.85–6.56). Survivors (n = 244) had a higher median fibrinogen level of 5.26 g/L (IQR: 4.28–6.60) compared to non-survivors (n = 158), who had a lower median level of 4.74 g/L (IQR: 3.44–6.45). The difference was statistically significant (p = 0.010).

Chalodiya K et al20 found that the mean plasma fibrinogen on day one and day seven was significantly higher in the expired group of patients compared with the survived group of patients. The mean fibrinogen on day seven did not differ significantly compared with the mean fibrinogen on day one in the expired group of patients, whereas the mean fibrinogen on day seven was significantly higher compared with the mean fibrinogen on day one in the survived group of patients.

RDW AS A PROGNOSTIC MARKER

Red Blood Cell Distribution Width (RDW) reflects variation in red blood cell size and increases with accelerated RBC production, often influenced by inflammatory cytokines like TNF-alpha, IFN-gamma, IL-1β, and IL-6, which impair erythropoiesis. In our study, RDW was significantly higher in non-survivors than survivors at admission (17.54 vs. 16.8; p = 0.02) and rose further at 48 hours (18.63 vs. 16.24; p < 0.0001). ROC analysis showed good predictive value for RDW, with an admission AUC of 0.717 (cut-off 17.6, sensitivity 83.3%, specificity 68.8%) that improved at 48 hours to an AUC of 0.877 (cut-off 17.8, sensitivity 83.3%, specificity 85.9%).

Shan X et al15 found thatAmong 402 patients, the median red blood cell distribution width (RDW) was 13.80% (IQR: 12.60–15.60). Non-survivors (n = 158) had a significantly higher median RDW of 15.10% (IQR: 13.40–17.05) compared to survivors (n = 244), who had a median RDW of 13.10% (IQR: 12.30–14.60). The difference was statistically significant (p < 0.001). Uffen J W et al21 found that among 1,046 patients, the median red blood cell distribution width (RDW) was 13.42% (IQR: 12.30–15.11). Non-survivors (n = 61) had a significantly higher median RDW of 14.74% (IQR: 13.23–16.71) compared to survivors (n = 985), who had a median RDW of 13.38% (IQR: 12.27–14.99). The difference was statistically significant (p < 0.001).

Jaindial A et al22 found that ROC curve analyses showed marginal discriminatory power of RDW (AUC 0.606) for prediction of 30-days mortality when compared to other markers. Furthermore, a study by Chalodiya K et al20 found statistically insignificant difference in the median duration of ICU and hospital stay according to the severity of sepsis, consistent with the findings of Jaindial A et al22 where the mean duration of hospital stay was 10 ± 9 days.

ROC

Kaushik R et al12 found that the ROC for NLR on day 1 (NLR1) showed area under the ROC curve (AUC) of 0.911 with significance of P < 0.001 with sensitivity of 87.5% and specificity of 90% at a value of ≥3.3. Also, Yoo et al. reported that the area under the ROC curve for the RDW/albumin ratio was higher than that for the RDW alone (0.681 vs. 0.576, P = 0.002) in predicting the prognosis of patients with acute respiratory distress syn drome, which is consistent with our findings.23

Jain K et al24 found that the diagnostic accuracy of outcome prediction, RDW showed a fair area under the ROC curve 0.852, CI of 95% (0.796-0.909). The AUC between the RDW and SOFA score with a reference curve was 0.852, CI 95% (0.796-0.909) and 0.950, CI 95% (0.918-0.981). Ahmed A et al16 found that the ROC curve (receiver operating curve) results shown that Serum albumin as a prognostic markerbased on admission day value gives that sensitivity (93.44) and specificity (64.10). ROC curve (receiveroperating curve) results shown that SOFA score as a prognostic marker based on admission day value gives that sensitivity (93.44) and specificity (89.74).

Sepsis is a major cause of hospital mortality and morbidity, highlighting the need for reliable prognostic markers. RDW and NLR ratio, simple and cost-effective measures, show promise in predicting sepsis severity and outcomes. However, larger studies are needed to validate their effectiveness across diverse populations and explore their use in noninfective inflammatory conditions. Robust evidence will help establish their role in improving clinical decisions and patient outcomes in sepsis management.

1. Rudd KE, Johnson SC, Agesa KM, Shackelford KA, Tsoi D, Kievlan DR, et al. Global, regional, and national sepsis incidence and mortality, 1990–2017: analysis for the global burden of Disease Study. Lancet. 2020;395:200–11.
2. Weng L, Xu Y, Yin P, Wang Y, Chen Y, Liu W, et al. National incidence and mortality of hospitalized sepsis in China. Crit Care. 2023;27:84
3. Chousterman BG, Swirski FK, Weber GF. Cytokine storm and sepsis disease pathogenesis. Semin Immunopathol. 2017 Jul;39(5): 517–28.
4. Afessa B, Keegan MT, Mohammad Z, Finkielman JD, Peters SG. Identifying potentially ineffective care in the sickest critically ill patients on the third ICU day. Chest 2004;126:1905‑9.
5. Pierrakos C, Velissaris D, Bisdorff M, et al. Biomarkers of sepsis: time for a reappraisal. Crit Care 2020;24:287.
6. Rungsakulkij N, Vassanasiri W, Tangtawee P, et al. Preoperative serum albumin is associated with intraabdominal infection following major hepatectomy. J Hepatobiliary Pancreat Sci 2019;26:479-89.
7. Matsubara, K. Yamakawa, Y. Umemura et al., “Significance of plasma fibrinogen level and antithrombin activity in sepsis: a multicenter cohort study using a cubic spline model,” Thrombosis Research 2019; 181:17–23.
8. Jeong JH, Heo M, Lee SJ, et al. Clinical Usefulness of Red Cell Distribution Width/Albumin Ratio to Discriminate 28-Day Mortality in Critically Ill Patients with Pneumonia Receiving Invasive Mechanical Ventilation, Compared with Lacate/Albumin Ratio: A Retrospective Cohort Study. Diagnostics (Basel) 2021;11:2344.
9. de Jager CP, van Wijk PT, Mathoera RB, de Jongh‑Leuvenink J, van der Poll T, Wever PC, et al. Lymphocytopenia and neutrophil‑lymphocyte count ratio predict bacteremia better than conventional infection markers in an emergency care unit. Crit Care 2010;14:R192.
10. Cecconi, L. Evans, M. Levy, and A. Rhodes, “Sepsis and septic shock,” Lancet 2018;392(10141):75–87.
11. The SAFE Study Investigators, “Impact of albumin compared to saline on organ function and mortality of patients with severe sepsis,” Intensive Care Medicine 2011;37(1):86– 96.
12. Kaushik R, Gupta M, Sharma M, Jash D, Jain N, Sinha N, et al. Diagnostic and prognostic role of neutrophil-to-lymphocyte ratio in early and late phase of sepsis. Indian J Crit Care Med 2018;22:660-3.
13. Hota PK, Reddy BG. Role of eosinophil count and neutrophil: lymphocyte count ratio as prognostic markers in patients with sepsis. Int Surg J 2017;4:2243-6.
14. Xia Y, Guo XG, Ji TX, Chen Q. Neutrophil count to lymphocyte count ratio is a potential diagnostic index for bacteremia in adult. Life Sci J. 2014;11(1):172-7.
15. Shan X, Li Z, Jiang J, Li W et al. Prognostic value of red blood cell distribution width to albumin ratio for predicting mortality in adult patients meeting sepsis-3 criteria in intensive care units. BMC Anesthesiology (2024);24:208.
16. Ahmed A, Rekha K B, Ruknuddin S, Ahmed Z. Serum Albumin as a Prognostic Marker in Critically Ill Patients. Int J Recent Sci Res. (2021);12(09):43020-43024.
17. Saravanakumar G, Balasubramaniyan S, Paari N. Estimation of Serial Serum Albumin Level as a Prognostic Marker in Sepsis Patients Admitted in Intensive Care Units. JMSCR(2020);8(11): 2455-0450.
18. Nirmala A.C., Namratha and Avinash B.H. A Study of Serial Estimations Of Serum Albumin As A Progonostic Marker. International Journal of Basic and Applied Medical Sciences ISSN: 2277-2103 (2015); 5 (1):228- 234.
19. Tang X, Shao L, Dou J, Zhou Y. Fibrinogen as a Prognostic Predictor in Pediatric Patients with Sepsis: A Database Study. Mediators of Inflammation (2020); Article ID 9153620, 10.
20. Chalodiya K, Bahulikar A, Khadke V, Phalgune D. Correlation of plasma fibrinogen and mean platelet volume in patients of sepsis to the severity of sepsis and its outcome. Int J Adv Med 2020;7:1830-5.
21. Uffen JW, Oomen P, Regt MD, Oosterheert JJ, Kaasjager K. The prognostic value of red blood cell distribution width in patients with suspected infection in the emergency department. BMC Emergency Medicine (2019);19:76.
22. Jandial A, Kumar S, Bhalla A, Sharma N, Varma N, Varma S. Elevated red cell distribution width as a prognostic marker in severe sepsis: A prospective observational study. Indian J Crit Care Med 2017;21:552-62.
23. Yoo JW, Ju S, Lee SJ, Cho YJ, Lee JD, Kim HC. Red cell distribution width/albumin ratio is associated with 60-day mortality in patients with acute respiratory distress syndrome. Infect Dis (Lond). 2020;52:266–70.
24. Jain K, Sharma D, Patidar M, Nandedkar S. Red Cell Distribution Width as a Predictor of Mortality in Patients With Clinical Sepsis: Experience From a Single Rural Center in Central India. Clinical Pathology Volume 2022;15:1–7.