Community-acquired pneumonia is an acute infection of the lung parenchyma that develops outside hospitals or other healthcare facilities. It continues to be a significant cause of morbidity, hospitalisation and mortality, especially in older people and patients with chronic conditions. Clinical manifestations may be mild with only respiratory symptoms that can be managed in an out-patient setting, to severe infection that necessitates intensive care, mechanical ventilation and circulatory support. Severe disease may be undiagnosed and result in respiratory failure, sepsis, extended hospital stays and mortality. Therefore an early and reliable determination of severity of the disease is important in order to determine the appropriate site of care and to begin treatment early enough [1-3].

There are a few clinical prediction tools that have been created for evaluating the severity and prognosis of community-acquired pneumonia. Some of the most widely used systems are the Pneumonia Severity Index and the CURB-65 score [4-6]. Confusion, raised blood urea, raised respiratory rate, low blood pressure and age of 65 years or over, are the basis of CURB-65. While these scoring systems are useful, they may need to incorporate multiple clinical and laboratory parameters and may not necessarily account for the extent of the systemic inflammatory response. Thus, the search for simple, easy-to-measure, and readily available markers to complement clinical severity scores to better stratify risk early is ongoing [4, 7, 8].

Neutrophil to lymphocyte ratio (NLR) is derived from absolute counts of neutrophils and lymphocytes measured by a standard complete blood count (CBC). In acute bacterial infection, the production and release of neutrophils rises, as part of the innate immune response, and physiological stress and inflammatory mediators can lead to redistribution and apoptosis of lymphocytes. This rise in NLR is driven by both neutrophilia and a relative decrease in lymphocytes, and could be a more comprehensive assessment of systemic inflammation than a simple leukocyte count. Elevated NLR has been reported in previous studies to be correlated with the severity of pneumonia, admission to the intensive care unit, treatment failure and death. Reported cut-off values, however, have been inconsistent in different populations and in different clinical settings [6, 9, 10].

In resource-limited healthcare facilities, advanced biomarkers such as procalcitonin may not always be available, whereas complete blood count testing is widely accessible and routinely performed at admission. Determining the relationship between NLR and established severity measures may therefore provide a practical method for identifying patients who require closer monitoring or more intensive treatment. The present study was conducted to evaluate the association between neutrophil-to-lymphocyte ratio and the severity of community-acquired pneumonia among adult patients. It also assessed the relationship of NLR with CURB-65 score, intensive care admission, mechanical ventilation, sepsis, length of hospital stay and in-hospital mortality.

This hospital-based cross-sectional study was conducted at a Tertiary Care Hospital of Islamabad from February 2025 to February 2026 to determine the association between the neutrophil-to-lymphocyte ratio and the severity of community-acquired pneumonia. A total of 89 adult patients diagnosed with community-acquired pneumonia were enrolled through consecutive non-probability sampling. Community-acquired pneumonia was defined as an acute infection of the lung parenchyma acquired outside a hospital setting, supported by compatible clinical features and a new pulmonary infiltrate on chest radiography. Patients were recruited from the emergency department, medical wards, and intensive care unit after confirmation of eligibility by the treating physician. Patients aged 18 years or older with a fever, cough, sputum, dyspnoea, pleuritic chest pain, or other symptoms suggestive of LRTI, who had radiological evidence of pneumonia, were included. Excluded patients included those with hospital-acquired pneumonia, ventilator-associated pneumonia, pulmonary tuberculosis, active malignancy, haematological disorders, autoimmune disease, chronic immunosuppressive therapy, recent chemotherapy, and known infection elsewhere in the body. In addition, patients who received systemic corticosteroids or antibiotics more than 48 hours prior to admission were excluded due to the possibility of altering inflammatory response and leukocyte differential count. Patients with incomplete clinical data or with absent complete blood count analysis were not analysed. Venous blood samples were obtained at admission prior to initiation of inpatient antibiotic therapy, if obtained, prior to inguinal surgery. Complete blood count (CBC) and differential leucocyte count (DLC) was done in the hospital laboratory with an automated haematology analyser as per standard operating procedure. The following haematological parameters were recorded: total leucocyte count, absolute neutrophil count, absolute lymphocyte count, haemoglobin level and platelet count. Neutrophil to lymphocyte ratio was computed as a ratio of absolute neutrophil count to absolute lymphocyte count. Other laboratory variables such as C-reactive protein, serum urea, serum creatinine, serum sodium, serum albumin and blood glucose were collected if available. In-hospital mortality, duration of hospital stay, sepsis/septic shock, mechanical ventilation, admission to intensive care unit and clinical outcome also were recorded. IBM SPSS Statistics 26 was used to enter and analyse data. All continuous variables were tested for normality with the Shapiro–Wilk test and were reported as either mean SD or median interquartile range, depending on the test results. Frequencies and percentages were used to express categorical variables. One-way analysis of variance (ANOVA) was used for normally distributed data and Kruskal–Wallis test was used for non-normally distributed data to compare the neutrophil-to-lymphocyte ratio between the three pneumonia groups. Chi-square test or Fisher's exact test was used to test the association between categorical variables. Spearman’s rank correlation was used to determine the relationship between NLR and CURB-65 score. Logistic regression was done to calculate the discriminatory power and the optimum cut-off value of NLR in predicting severe pneumonia by receiver operating characteristic curve (ROC) analysis. Variables that were clinically important and those with a p-value < 0.20 by univariable analyses were included in a multivariable logistic regression model to determine independent predictors of severe community-acquired pneumonia. Odds ratios with 95% confidence intervals were presented and a p value less than < 0.05 was deemed statistically significant.

The study included a total of 89 patients with community-acquired pneumonia (CAP) diagnosis. The mean age of the participants was 56.84 ± 15.27 years (range 19-84 years). Of the total patients, 53 (59.6%) were male and 36 (40.4%) were female. Forty-one patients (46.1%) were current or former smokers. Most common comorbid condition was hypertension in 37 (41.6%) patients, diabetes mellitus (34.8%), chronic obstructive pulmonary disease (21.3%), ischaemic heart disease (15.7%) and chronic kidney disease in 9.0% of patients.

Table 1. Demographic and clinical characteristics of the study participants

The most common presenting symptom reported by 83 (93.3%) patients was cough. Fever was present in 76 (85.4%), shortness of breath in 69 (77.5%), sputum production in 58 (65.2%), and pleuritic chest pain in 29 (32.6%) patients. Eleven (12.4%) patients had altered mental status noted. The mean respiratory rate was 27.83 ± 6.14 breaths per minute, and the mean oxygen saturations on admission was 89.71 ± 5.82%. The mean total leukocyte count was 13.62 ± 4.81 × 10⁹/L. The mean absolute neutrophil count was 10.71 ± 4.27 × 10⁹/L, whereas the mean absolute lymphocyte count was 1.39 ± 0.63 × 10⁹/L. The overall median NLR was 7.12 (range 4.35–11.68). The mean C-reactive protein levels were 78.46 ± 45.73 mg/L; 27 (30.3%) cases showed multilobar involvement and 16 (18.0%) cases had a pleural effusion.

Table 2. Laboratory and radiological findings

The CURB-65 score was used to categorize pneumonia severity: 38 (42.7%) patients had mild pneumonia, 28 (31.5%) had moderate pneumonia, and 23 (25.8%) had severe pneumonia. There was a progressive increase in the NLR as severity of pneumonia increased. The median NLR was 4.18 (3.12–5.76) among patients with mild pneumonia, 7.64 (5.82–9.73) among those with moderate pneumonia, and 13.27 (9.84–17.92) among patients with severe pneumonia. The three severity groups were found to be statistically different in terms of NLR (p < 0.001).

Table 3. Comparison of neutrophil-to-lymphocyte ratio according to pneumonia severity

A significant positive correlation was found between the neutrophil-to-lymphocyte ratio and CURB-65 score (Spearman’s r = 0.67, p < 0.001). NLR also showed a positive correlation with C-reactive protein levels (r = 0.51, p < 0.001), respiratory rate (r = 0.42, p < 0.001), and length of hospital stay (r = 0.46, p < 0.001). In contrast, NLR had a significant negative correlation with oxygen saturation at admission (r = −0.48, p < 0.001) and serum albumin level (r = −0.36, p = 0.001). Patients were separated into the low and high NLR groups with a NLR cutoff value of 8.0 for further analysis. Forty-nine patients (55.1%) had an NLR below 8.0, while 40 (44.9%) had an NLR of 8.0 or above. Patients with a high NLR had a higher incidence of severe pneumonia than did those with a low NLR (18/40 (45.0%) vs 5/49 (10.2%), respectively, p < 0.001). In a similar fashion, ICU admission, mechanical ventilation, sepsis, and in-hospital mortality were significantly higher in those with a high NLR.

Table 4. Association between NLR category and clinical outcomes

Overall, 19 patients (21.3%) needed admission to intensive care unit (ICU), 11 (12.4%) needed invasive mechanical ventilation and 15 (16.9%) developed sepsis or septic shock. The hospital stay was 6.61 ± 3.29 days, on average. In-hospital mortality was 9.0% (8 patients died during hospital stay). Patients who died had a significantly higher median NLR than those who survived, 15.42 (11.37–20.18) versus 6.61 (4.12–10.24), respectively (p = 0.002). When analysed by receiver operating characteristic curve, good discriminatory capacity was obtained for the diagnosis of severe CAP by NLR. The area under the curve was 0.84 (95% CI: 0.75–0.93, p < 0.001). The sensitivity and specificity to predict severe pneumonia were 78.3% and 66.7%, respectively, at an NLR cutoff value of 8.0. The positive predictive value was 0.45, while the negative predictive value was 0.89.

Table 5. Diagnostic performance of NLR for predicting severe CAP

 Multivariate logistic regression analysis showed that the NLR of ≥8.0 was an independent predictor of severe CAP after adjustment for age, sex, diabetes mellitus, COPD, serum albumin, and multilobar involvement. Patients with an NLR > 8.0 had about four times the odds of severe pneumonia than those with NLR < 8.0 (adjusted OR: 4.18; 95% CI: 1.39–12.57; p = 0.011). Multilobar involvement and low serum albumin were also independently associated with severe disease.

Table 6. Multivariable logistic regression analysis for predictors of severe CAP

These findings demonstrated a clear association between an elevated neutrophil-to-lymphocyte ratio and increasing severity of community-acquired pneumonia. Higher NLR values were also associated with ICU admission, mechanical ventilation, prolonged hospitalization, sepsis, and in-hospital mortality.

Figure 1. Median neutrophil-to-lymphocyte ratio according to the severity of community-acquired pneumonia. The median NLR increased progressively from mild disease (4.18) to moderate disease (7.64) and severe disease (13.27), with a statistically significant difference among the three CURB-65 severity groups (p < 0.001).

The present study demonstrated a significant association between the neutrophil-to-lymphocyte ratio and the severity of community-acquired pneumonia. The median NLR increased progressively from 4.18 among patients with mild pneumonia to 7.64 in moderate pneumonia and 13.27 in severe pneumonia. This difference was statistically significant, while NLR also showed a strong positive correlation with the CURB-65 score. These findings suggest that the inflammatory response becomes increasingly pronounced as pneumonia severity increases. Neutrophilia reflects activation of the innate immune response during bacterial infection, whereas lymphopenia may result from physiological stress, redistribution of lymphocytes and increased lymphocyte apoptosis. Therefore, NLR combines two complementary components of the immune response and may provide more clinically useful information than either neutrophil or lymphocyte count alone. Earlier work by de Jager et al. also showed that the neutrophil–lymphocyte count ratio was useful in assessing disease severity and prognosis among patients with community-acquired pneumonia [11-13]. Secondly, the trend of progressive increase in NLR with the increasing categories of CURB-65 in this study corroborates with previous studies. A high NLR was correlated with poor clinical outcomes in adult patients with community-acquired pneumonia (CAP) according to Yan et al. [14]. Likewise, Kartal and Kartal determined higher NLR levels in pneumonia patients and demonstrated that NLR levels can be used to help differentiate patients suitable for hospitalisation from those suitable for outpatient treatment [15]. Cataudella et al. (195 older patients with CAP) found NLR to be an independent predictor of 30-day mortality and that NLR outperformed WBC, CRP, and CURB-65 in their population [16]. This observation is consistent with the current data that the NLR is a marker of the abnormalities present in CURB-65: tachypnoea, elevated urea, hypotension, confusion and advanced age. The present study revealed that patients with an NLR ≥ 8.0 significantly had higher rates of severe pneumonia, multilobar pneumonia, intensive care unit admission, mechanical ventilation, sepsis and in-hospital mortality. The median NLR was also significantly elevated in patients with death compared to those who survived. These findings are similar to Feng et al.'s CAP study in elderly patients, in which the authors showed a combined NLR–CURB-65 score was useful in predicting in-hospital mortality [7]. A meta-analysis conducted in 2021 from 10 studies and 5220 patients with community-acquired pneumonia also found that patients with elevated NLR had an increased risk of death [17]. A rise in NLR is associated with increased incidence of adverse outcomes, which could be attributed to an exaggerated systemic inflammatory response, extensive pulmonary tissue damage and reduced gas exchange and increased risk of organ dysfunction. The area under the curve (AUC) of this receiver operating characteristic (ROC) analysis was good for identifying severe CAP (0.84). At a cutoff of 8.0, NLR provided a sensitivity of 78.3%, specificity of 66.7% and a high negative predictive value of 89.8%. This indicates that low NLR could be useful in assessing patients less likely to be severely ill [9, 18, 19]. However, there are a variety of published NLR cutoffs due to differing age ranges, disease spectrum, time of blood sampling, and the definition of adverse outcomes and comorbidities. Some studies adopted different cut-off points of NLR to predict clinical instability or mortality [16], but Cataudella et al. noted a significant rise in mortality at NLR > ~11.Cataudella et al. reported that mortality increased significantly at NLR > ~11, while other studies had different cut-off points for clinical instability/mortality [16]. Lee et al. also showed that serial NLR levels gave information on treatment response and outcomes in hospitalized patients with CAP and that NLR values at admission might not be as informative as changes in NLR during the hospital stay [20]. A multivariable analysis revealed that NLR ≥ 8.0 was independently associated with severe pneumonia after adjusting for demographic characteristics, comorbidities and laboratory findings. Low serum albumin and multilobar involvement were also independent predictors, suggesting that severity of CAP is related to both systemic inflammatory response and degree of nutritional and pulmonary compromise. NLR is cheap and easy to calculate, given the routine blood count, and especially well suited for use in the emergency department and other clinical settings where more sophisticated biomarkers, like procalcitonin, are not always available. However, NLR should not be used in isolation of the currently available clinical assessment instruments like CURB-65 and the Pneumonia Severity Index. The study was constrained by being single-centre, a relatively small sample size, cross-sectional evaluation, and a single admission NLR measurement. Leucocyte counts may have been influenced by residual confounding due to undiagnosed inflammatory conditions, drugs, and the causative pathogen. Further multicentre prospective studies are called for to assess the evolution of NLR over time, and to test the pathogen-specific differences, and the standardisation of NLR cut-off values in larger and more diverse populations.

The neutrophil-to-lymphocyte ratio was significantly associated with the severity of community-acquired pneumonia. NLR increased progressively across mild, moderate and severe CURB-65 groups and was positively correlated with the CURB-65 score. Patients with an elevated NLR had higher risks of intensive care admission, mechanical ventilation, sepsis, prolonged hospitalization and in-hospital mortality. An NLR cutoff of 8.0 showed good ability to identify severe pneumonia and retained an independent association with disease severity after adjustment for relevant confounding factors. NLR may therefore serve as a simple, rapid and inexpensive supplementary marker for early risk stratification in patients with community-acquired pneumonia. It should be interpreted together with clinical findings, radiological features and established pneumonia severity scores.

1. Nirgude, D., et al., Study of Absolute Neutrophil Absolute Lymphocyte count ratio in Community Acquired Pneumonia patients as a prognostic indicator at a tertiary care centre. 2021. 12(9): p. 69-74.
2. Elassal, G.M., M.A. Elsayed, and A.M.J.Q.A.I.J.o.M. Shehata, Value of neutrophil to lymphocyte ratio in prognosis of elderly patients with community acquired pneumonia compared to CRP level. 2021. 114(Supplement_1): p. hcab087. 001.
3. Cury, V.F., et al., Developing the pneumonia-optimized ratio for community-acquired pneumonia: an easy, inexpensive and accurate prognostic biomarker. 2021. 16(3): p. e0248897.
4. Altas, O.F. and M.J.M.m.j. Kizilkaya, The effects of neutrophil-lymphocyte ratio, platelet-lymphocyte ratio and prognostic markers in determining the mortality in patients diagnosed with pneumonia in intensive care. 2021. 36(2): p. 130.
5. Zahorec, R.J.B.L.L., Neutrophil-to-lymphocyte ratio, past, present and future perspectives. 2021. 122(7): p. 474-488.
6. Zheng, N., D. Zhu, and Y.J.B.P.M. Han, Procalcitonin and C-reactive protein perform better than the neutrophil/lymphocyte count ratio in evaluating hospital acquired pneumonia. 2020. 20(1): p. 166.
7. Feng, D.-Y., et al., Combined neutrophil-to-lymphocyte ratio and CURB-65 score as an accurate predictor of mortality for community-acquired pneumonia in the elderly. 2021: p. 1133-1139.
8. Qun, S., et al., Neutrophil-to-lymphocyte ratios are closely associated with the severity and course of non-mild COVID-19. 2020. 11: p. 563543.
9. Bekis Bozkurt, H.J.T.J.o.B., Is there any relationship between C-reactive protein/albumin ratio and clinical severity of childhood community-acquired pneumonia. 2021. 46(6): p. 647-653.
10. Motamed, H., et al., Association between community-acquired pneumonia and platelet indices: A case-control study. 2021. 10(1): p. 23-27.
11. de Jager, C.P., et al., The neutrophil-lymphocyte count ratio in patients with community-acquired pneumonia. 2012.
12. Akan, O.Y. and O.J.E. Bilgir, Effects of neutrophil/monocyte, neutrophil/lymphocyte, neutrophil/platelet ratios and c-reactive protein levels on the mortality and intensive care need of the patients diagnosed with Covid-19. 2021. 5(1): p. 21-5.
13. Acar, E., Comparison of inflammation markers with prediction scores in patients with community-acquired pneumonia. 2021.
14. Ge, Y.L., et al., Neutrophil-to-Lymphocyte Ratio in Adult Community-Acquired Pneumonia Patients Correlates with Unfavorable Clinical Outcomes. 2019. 65(5).
15. Kartal, O. and A.J.B.M.J. Kartal, Value of neutrophil to lymphocyte and platelet to lymphocyte ratios in pneumonia. 2017. 118(9): p. 513-516.
16. Cataudella, E., et al., Neutrophil‐to‐lymphocyte ratio: an emerging marker predicting prognosis in elderly adults with community‐acquired pneumonia. 2017. 65(8): p. 1796-1801.
17. Alzoubi, O. and A.J.M.A.f.C.D. Khanfar, Association between neutrophil to lymphocyte ratio and mortality among community acquired pneumonia patients: a meta-analysis. 2022. 92(3).
18. Qiu, Y., et al., Neutrophil-to-lymphocyte ratio predicts mortality in adult renal transplant recipients with severe community-acquired pneumonia. 2020. 9(11): p. 913.
19. Beyaz, A., et al., The effect of C-reactive protein, procalcytonine, nuetrophil/lymphosite levels on mortality and duration of hospital stay in pneumonia. 2021. 3(1): p. 15-19.
20. Lee, H., et al., Prognostic value of serial neutrophil-to-lymphocyte ratio measurements in hospitalized community-acquired pneumonia. 2021. 16(4): p. e0250067.