Comparison between the Accuracy of TLC and Neutrophil-to-Lymphocyte ratio as Predictive Biomarker for the Development of Sepsis

The recognition of sepsis is still a significant issue in ED general surgery and intensive care settings as diagnosis is essential to improving outcomes [1]. Sepsis according to the Sepsis-3 consensus definition is a life-threatening organ dysfunction caused by a dysregulated host response to suspected or proven infection [5]. Patients with acute surgical diseases like perforated hollow viscus, necrotizing soft tissue infections, complicated appendicitis, and severe ascending cholangitis are especially susceptible to rapid clinical deterioration with localized disease potentially degrading into a systemic inflammatory dysfunction very quickly. If sepsis is not identified quickly enough, then appropriate antimicrobial, and definitive source control, which are strongly associated with reduced morbidity and mortality, may be delayed [1,5]. Hence, identification of simple, readily available biomarkers, which can lead to early stratification of risks in surgical practice, become an important priority [2,4].

The Total Leukocyte Count (TLC) which is part of the basic complete blood count (CBC) has long been considered a starting point laboratory indicator of systemic inflammation/infection. One of the main criteria of the Systemic Inflammatory Response Syndrome (SIRS) was traditionally leukocytosis (TLC>12.0 × 10^9/L) or leukopenia (TLC<4.0 ×10^9/L). However, there are significant limitations in the use of TLC for diagnostic purposes, especially in the context of acute surgical conditions, in which a high leukocytic count may exist without infection [4,6]. Circulating leukocytes also can be increased independently by surgical stress, tissue injury, ischemia, trauma and corticosteroid therapy, making the specificity as an indicator of true sepsis less [3]. Thus, the use of TLC alone could not always be used to differentiate von infectious systemic inflammation from a physiological stress response.

Inflammatory markers based on the differential leukocyte count have gained increasing interest in order to improve the accuracy of the diagnosis, including the Neutrophil-to-Lymphocyte Ratio (NLR) [1,2]. The NLR is the ratio of the absolute value of the neutrophils to the lymphocytes, and reflects the balance of innate and adaptive immunity [1]. In the early stages of sepsis inflammatory cytokines induce a rapid release of neutrophils from the bone marrow and induce apoptosis and depletion of circulating lymphocytes, particularly CD4+ cells and CD8+ T cells [5,8]. All the above-mentioned immune alterations occur concurrently and result in a significant rise in the NLR level, which could make the NLR a more useful parameter of the immune imbalance of the body rather than the total leucocyte count alone [1,8].

While biomarkers like procalcitonin (PCT) and C-reactive protein (CRP) show promise for the early diagnosis of sepsis, their cost and availability are limited and they are not infrequently used in LMICs, including Pakistan, especially in low-resource settings [4]. In contrast, both are easily available from a routine complete blood count done under automation (CBC), which does not require extra testing, making them low cost and easily utilized in the clinical routine evaluation [1,6]. However, direct comparative data on the predictive ability of the two biomarkers in acute surgical patients within the context of the Pakistani health-care system is scarce. Hence, the present study was conducted to compare the diagnostic accuracy of baseline TLC and NLR for prediction of sepsis in patients admitted in the surgical admission ward with acute surgical diseases in a tertiary care hospital of Lahore, Pakistan.

Design and setting of the studies

This is a prospective comparative cohort study done in Department of General Surgery, Combined Military Hospital (CMH), Lahore, Pakistan. CMH Lahore is a Tertiary Care Teaching Hospital which offers emergency and elective surgical facilities to the Military and the general public of Punjab.The events lasted for a total of eight months from March 2025 to October 2025. The study protocol was approved by the Institutional Review Board and Ethics Committee of Combined Military Hospital, Lahore (IRB No. CMH/GS/2024-045) and registered in Pakistan National Trial Registry (PNTR202401000456). All procedures performed in the studies were in line with the principles of the Declaration of Helsinki. All the individuals who were eligible for the study or their legal representative if applicable, were given a detailed description of the objective of the study before enrolment and were asked to sign an informed consent form. All patient records were anonymised with alphanumeric ID numbers for the purpose of keeping the information confidential.

Sample Size Calculation

The Buderer formula for diagnostic accuracy studies was used to calculate the required sample size. It was based on the assumed sensitivity of 85% and 65% for Neutrophil-to-Lymphocyte Ratio (NLR) and Total Leukocyte Count (TLC) respectively in diagnosis of sepsis with the determined margin of error as 10, 95% confidence level and the estimated prevalence of sepsis as 22% among the patients admitted with acute surgical condition [1,5]. The smallest calculated sample size was 165 participants. Considering the possible losses due to poor laboratory documentation, absence of follow-up data or early transfer of patients, the sample size was expanded to 180 consecutive eligible patients.

Selection Criteria

Inclusion Criteria:

•         Male and female adult patients (ages 18-75) years.

•         Patients who are admitted to the emergency department or outpatient clinics with acute surgical conditions of various types, which need surgical exploration or close inpatient monitoring such as complicated acute appendicitis, acute cholecystitis, mechanical intestinal obstruction, hollow viscus perforation, intra-abdominal abscess or necrotising fasciitis [3,6,7].

•         Prompt availability of a FBC that includes an automated DBC within two hours of presentation to the hospital and prior to the start of empirical antimicrobial therapy [1].

Exclusion Criteria

•         Those who, at admission, have known sepsis or septic shock [1].

•         A history of haematological malignancy, bone marrow disorders, currently on chemotherapy or high dose corticosteroids/immunosuppressive agents [5].

•         Patients in end stage chronic kidney disease on maintenance hemodialysis or in advanced liver cirrhosis (Child – Pugh Class C).

•         Patients who died within 24 hours of admission due to catastrophic haemorrhagic shock, or non-infectious polytrauma.

Operational Definitions

1.        Total Leukocyte Count (TLC): Count the number of leukocytes in the venous blood as ×10^9/L using automated hematology analyzer.

2.        Neutrophil-to-Lymphocyte Ratio (NLR): This is a ratio index which is calculated from the differential count of leukocytes in the following manner [1,2]

NLR refers to Absolute Neutrophil Count / Absolute Lymphocyte Count

3.        Primary Outcome (Development of Sepsis): Primary outcome was development of sepsis during 7 days follow up period. Sepsis was diagnosed using the consensus-based definition of sepsis (Sepsis-3): an acute. increase of ≥2 points in Sequential Organ Failure Assessment (SOFA) score due to suspected or proved infection [5]. The SOFA score assesses failure in each of six organ systems – respiratory, coagulation, hepatic, cardiovascular, neurological and renal.

Data collection and laboratory analysis

The principal investigator recorded demographic characteristics at admission (age/sex), presenting complaints and baseline vital signs (heart rate, respiratory rate, blood pressure, temperature) and provisional clinical diagnosis on a data collection sheet. A 3 mL venous blood sample was obtained in an ethylenediaminetetraacetic acid (EDTA) tube within 2 hours of hospital presentation and was analyzed within the institutional pathology laboratory using an automated an automated hematology analyzer in regard to total leukocyte count, absolute neutrophil count and absolute lymphocyte count [1].

In both groups, all patients were managed with standard surgery along with judicious fluid resuscitation and culture directed antibiotics or empirical antibiotic therapy, and definitive source control (laparotomy, laparoscopy and/or wound debridement) was performed when indicated [3]. Organ dysfunction and sepsis were assessed on 7th day after admission in the patients. Every day clinical and biochemical parameters (arterial blood gas analysis, platelet count, serum bilirubin, serum creatinine, Glasgow Coma Scale (GCS) score and urine output) were recorded to calculate the SOFA score and determine the occurrence of sepsis [5].

Statistical Analysis

SPSS Statistics 29.0 for Windows, IBM Corp., Armonk, NY, USA, was used for data analysis. The Shapiro–Wilk test was used to explore the distribution of continuous variables. All continuous variables were normally distributed and presented as mean ± standard deviation (SD), and were statistically compared using independent Student's t-test. Categorical data were presented as percentages and numbers; comparison was done with the Pearson chi-square and Fisher's exact test as appropriate.

The predictive performance of TLC and NLR was tested by Receiver Operating Characteristic (ROC) curve analysis [4] Area under the curve (AUC) and corresponding 95% confidence intervals (CI) were computed for both biomarkers [4,5]. Youden's index (J = Sensitivity + Specificity − 1) was used to obtain each marker's optimal cut-off value. At these cut-off values the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) were calculated. A two-sided P-value of ≤0.05 was considered statistically significant.

Diagnostic Parameter	Neutrophil-to-Lymphocyte Ratio (NLR)	Total Leukocyte Count (TLC)
Optimal Cut-off Value	5.2	11.5 ×10^9/L
Area Under the Curve (AUC)	0.884 (95% CI: 0.824–0.944)	0.742 (95% CI: 0.658–0.826)
P-value	<0.001	<0.001
Sensitivity (%)	85.7	71.4
Specificity (%)	81.2	64.5
Positive Predictive Value (PPV) (%)	58.1	38.0
Negative Predictive Value (NPV) (%)	94.9	88.1

 

Our study demonstrated that the neutrophil to lymphocyte ratio (NLR) at admission had better diagnostic performance to predict sepsis in acute surgical patients than the total leukocyte count (TLC) [1,4]. This is consistent with the NLR being a readily available inflammatory indicator that expresses neutrophilia together with relative lymphopenia in the setting of systemic infection [1,5]. While TLC is commonly used to assess inflammation, it is not a reliable way to determine whether the inflammatory response is local or a dysregulated systemic inflammatory process seen in sepsis [4]. NLR on the other hand does include the change in neutrophils and lymphocytes level which is a more sensitive indicator of severe bacterial infection [5,8].

The results of the present study showed that NLR had a high diagnostic value with Area Under the Curve 0.884 as compared to TLC of 0.742. This has clinical significance in general surgical practice at the emergency department where early recognition of patients at risk of sepsis may allow timely intervention [3,6]. A high sensitivity was accompanied with a negative predictive value of 94.9% at an admission NLR cut off value of 5.2 (based on Youden's index). The results of this study indicate that patients having admission NLR value less than this threshold value are less likely to develop sepsis infection, and may have stratification of reduced risk settings when taken together with the routine clinical assessment.

On the other hand, TLC demonstrated relatively low specificity (64.5%) and positive predictive value (38.0%) with the best cutoff value of 11.5 × 10^9/L; leucocytosis induced by surgical trauma and/or tissue injury without systemic infection is common with a low diagnostic value of TLC alone in emergency surgical patients [3,4]. Thus, the use of TLC as a single parameter raises the possibility of false positive clinical suspicion of sepsis.

The findings of our study are in agreement with previous reports. Rehman et al.  found NLR had higher performance characteristics when compared to traditional WBC parameters for early detection of sepsis [1]. In another study, Sarin et al also observed that alterations in differential WBC and high NLR value was correlated with developing sepsis after major abdominal surgery [2]. Giacometti et al. also showed that a higher admission NLR was linked to extensive intraabdominal disease such as bowel ischemia and hence the results found in this study [3].

Patients who developed sepsis always had higher NLR values on admission than those who did not , irrespective of the main surgical diagnosis. If this finding proves to be correct, this could be a good low-cost screening tool for a wide array of emergency surgery cases [6,7]. As it is derived from a CBC, NLR can be easily included in the assessment process in the ED, at no extra cost, and without any specialized equipment [4].

There are several strengths to this study, being a prospective study with consecutive patient enrollment, the use of Sepsis-3 criteria, and the generation of data from an emergency general surgery population in a tertiary care hospital in Pakistan. Several limitations should, however, be recognized. First, it was a single center study limiting generalization of the results. Second, only NLR and TLC levels were assessed at admission and not serially to determine how these changed over time. Third, unfortunately due to resources and finances, comparison with other markers (such as procalcitonin or interleukin-6) was not possible.

The usefulness of the findings needs to be confirmed in further multicenter studies using larger and more heterogeneous patient groups in order to define optimal diagnostic cut-off values for different healthcare settings [4]. In addition, serial evaluation of NLR during the treatment process should also be conducted in future studies and associations of NLR with infection control, clinical outcomes, length of intensive care unit stay and survival should also be analyzed.

 Overall, the NLR at admission exhibits superior predictive and diagnostic values for predicting sepsis onset in acute surgical patients when compared to TLC in acute surgical patients. NLR has a good sensitivity and specificity when used with a cut-off value of 5.2, and has high NPV, which could be useful for clinical risk stratification. The NLR is a practical, available screening tool that comes from a commonly ordered blood test, doesn't require any extra health care costs, can complement and augment clinical judgment, and can assist in making triage decisions and identify patients early in any condition.

ACKNOWLEDGMENT

The authors are grateful for the helpful assistance extended by the laboratory staff, nursing staff, and surgical residents of Combined Military Hospital (CMH) Lahore, Pakistan, in samples processing, data collection and patient management during the entire study.

CONFLICT OF INTEREST

None

FUNDING

None

ETHICAL APPROVAL

The study protocol was approved by the IRB No. CMH/GS/2024-045 Combined Military Hospital Lahore (IRB & EC) Institutional Review Board and Ethics Committee. The study has been registered in Pakistan National Trial Registry (PNTR202401000456). The study procedures were performed according to the Declaration of Helsinki and written informed consent was derived from all participants prior to their consent to participate.

1.      Rehman FU, Khan A, Aziz A, Iqbal M, Mahmood SB, Ali N. Neutrophil to lymphocyte ratio: earliest and efficacious marker of sepsis. Cureus. 2020;12(10):e11060. doi:10.7759/cureus.11060.

2.      Sarin S, Pamecha V, Sinha PK, Patil N, Mahapatra N. NLR can predict Septic evolution after ALD LT. J Clin Exp Hepatol. 2022;12(4):1142-1149. doi:10.1016/j.jceh.2021.11.008.

3.      Giacometti M, Costantini A, Tosi F, et al. Neutrophil-to-lymphocyte ratio predict bowel ischemia in non-strangulated adhesive small bowel occlusions: a retrospective analysis from an acute care surgical service. BMC Surg. 2024;24(1):182. doi:10.1186/s12893-024-02490-x.

4.      Arif S, Zahoor S, Ahmed A. Comparative diagnostic accuracy of composite leukocyte indices as compared to systemic inflammatory markers in emergency general surgery. Journal of Coll Physicians Surg Pak. 2023;33(5):512-517. doi: 10.29271/jcpsp.2023.05.512.

5.      Li T, Wang J. Predictive value of the neutrophil-to-lymphocyte ratio in the prognosis and risk of death in adult patients with sepsis: a meta-analysis. Front Immunol. 2024;15:1374520. doi:10.3389/fimmu.2024.1374520.

6.      Soliman MM, Hegazi MA. Application of the complete blood count (CBC) ratios as single parameters in the prediction of intensive care admission and tissue hypoperfusion in secondary peritonitis. Egypt Heart J. 2021;73(1):54. doi:10.1186/s43044-021-00178-5.

7.      Özkurt S, Dönmez H. Systemic inflammatory response syndrome is correlated with clinical outcomes in acute cholecystitis, and the value of neutrophil-to-lymphocyte ratio in their prediction is investigated. Med Sci Monit. 2015;21:148-154.  doi: 10.1111/anec.12187.

8.       Zhang R, Chen S, Zhao Q, et al. Prognostic value of the lymphocyte-to-monocyte-ratio and the neutrophil-to-lymphocyte-ratio in patients with severe intra-abdominal sepsis. Exp Ther Med. 2017;14(3):2131-2138. doi:10.3892/etm.2017.4391.