Neutrophil-to-lymphocyte and platelet-to-lymphocyte ratios in bacterial infections: contributions to diagnostic strategies in a tertiary care hospital in Tunisia

Background Bacterial infections continue to pose a global health challenge, driven by antibiotic resistance and septicemia. This study aimed to assess the diagnostic utility of neutrophil-to-lymphocyte ratio (NLR) and platelet-to-lymphocyte ratio (PLR) in bacterial infections versus non-infectious causes of inflammation. Methods A prospective study included 164 adult patients who were divided into two groups: a group of patients with confirmed bacterial infections and a second group of patients with other diagnoses (inflammatory pathologies, neoplasms, venous thromboembolic diseases, etc.). NLR and PLR values were compared between the bacterial infection group and the non-infectious causes group and the diagnostic performances of NLR and PLR for detecting bacterial infections were evaluated in comparison with other infection markers. Results NLR and PLR were significantly higher in bacterial infections (p < 10 ^-6), and NLR was correlated positively with inflammation markers. NLR and PLR demonstrated significant potential in diagnosing bacterial infections, with an AUC of 0.72 and 0.60, respectively, using the following cutoff values: 4.3 for NLR and 183 for PLR. Conclusion These findings underscore the importance of NLR and PLR as adjunctive tools for bacterial infection diagnosis.


Methods
A prospective study included 164 adult patients who were divided into two groups: a group of patients with confirmed bacterial infections and a second group of patients with other diagnoses (inflammatory pathologies, neoplasms, venous thromboembolic diseases, etc.).NLR and PLR values were compared between the bacterial infection group and the non-infectious causes group and the diagnostic performances of NLR and PLR for detecting bacterial infections were evaluated in comparison with other infection markers.

Introduction
Bacterial infections are common conditions affecting millions of people worldwide.Due to the rise in antibiotic resistance and septicemia, bacterial infections remain one of the leading causes of death on a global scale.In 2019, deaths associated with bacterial infections accounted for 13.6% of all deaths. 1 Some infections are relatively benign, while others can be much more serious, requiring urgent medical intervention, even hospitalization.This presents a significant challenge for healthcare professionals.
The gold standard for diagnosing a bacterial infection is isolating the pathogenic microorganism.However, this can be challenging in routine practice.Despite the availability of numerous diagnostic tools, optimizing patient management is crucial for effective treatment.Culturing bacteria can take days, delaying diagnosis and treatment initiation.Additionally, some bacteria are difficult to culture, leading to false negatives.Inflammatory markers such as C-reactive protein (CRP), white blood cell count, and procalcitonin (PCT) have proven effective in diagnosing, prognosing, and assessing treatment response. 2owing interest is emerging in new markers such as the neutrophil-to-lymphocyte ratio (NLR) and platelet-tolymphocyte ratio (PLR).Their relevance stems from changes observed in the number of neutrophilic granulocytes, platelets, and lymphocytes during the inflammatory response. 3,4 response to a bacterial infection, the number of neutrophilic granulocytes and platelets increases, while the number of lymphocytes decreases due to redistribution and increased apoptosis. 4[7] The aim of this study was to investigate the diagnostic utility of NLR and PLR in bacterial infections compared to other non-infectious causes of inflammation.

Study population
A prospective comparative study was conducted within the internal medicine department.A total of 164 patients were admitted between January 1st and July 31st, 2023.

Inclusion criteria
• Adults aged 18 and older admitted to the internal medicine department between January 1st and July 31st, 2023.

REVISED Amendments from Version 1
The revised manuscript incorporates minor changes based on the feedback from the reviewers.While the core content and findings remain largely unchanged, we have made several refinements to improve clarity and address specific points raised by the reviewers: 1. Enhanced context: We expanded the introduction to provide more context on the limitations of existing diagnostic methods of bacterial infections.

Methodology clarification:
The methods section now includes more detailed information on the control and reliability of the analytical process for all laboratory measurements.

Results elaboration:
We have added more detailed statistical analysis, specifically the 95% confidence interval of r in the Fisher correlation.
4. Discussion enhancement: The discussion has been modestly expanded to address the study's limitations and provide practical recommendations.
These modifications have improved the manuscript's clarity and scientific rigour without altering the study's fundamental content or conclusions.We believe these changes address the reviewers' concerns while maintaining the original focus and findings of our research.
Any further responses from the reviewers can be found at the end of the article

Exclusion criteria
• Previous antibiotic treatment before admission (one week before admission).

Laboratory values and clinical assessment
For each patient, the following parameters were recorded upon admission: • Demographic data, medical history, and clinical symptoms.
• Results of laboratory analyses, including complete blood count (absolute neutrophil, platelet, and lymphocyte counts), and other inflammatory markers such as CRP and PCT (CRP with normal range < 5mg/L and PCT with normal range < 0.1 ng/ml); We considered PCT positive for infectious diseases when it was > 0.25 ng/ml).
NLR and PLR were calculated by dividing the number of neutrophils by the number of lymphocytes and the number of platelets by the number of lymphocytes, respectively.The normal range of neutrophil-to-lymphocyte ratio (NLR) and platelet-to-lymphocyte ratio (PLR) varies from 1 to 3 and 90 to 210, accordingly. 8NLR and PLR values were compared between the bacterial infection group and the non-infectious causes group.
White blood cell (WBC) and differential cell counts were obtained using the DxH 900 hematology system by Beckman Coulter, California, USA.
A daily passage of internal quality controls was performed to ensure good control and reliability of the analytical process for all laboratory measurements.
Bacterial infections were confirmed through a combination of clinical assessment, laboratory tests, and microbiological cultures.Specifically, blood cultures, urine cultures, and other site-specific cultures were performed as clinically indicated.Positive culture results, along with corresponding clinical symptoms and elevated inflammatory markers (CRP, PCT), were used to confirm the presence of bacterial infections.In cases where cultures were negative but clinical suspicion was high, a combination of radiological findings, response to antibiotic therapy, and expert clinical judgment was used to classify cases as bacterial infections.The significance level for statistical analysis was set at p < 0.05, denoting that differences between groups and the strength of parameter correlations were assessed against this predetermined threshold.

Statistical analysis
The correlation coefficient (r) was employed to measure the strength and direction of linear relationships between parameters.An r-value close to 1 indicated a strong positive correlation, while an r-value close to -1 indicated a strong negative correlation.An r-value close to 0 suggested a weak or non-existent correlation between parameters.
The significance levels were indicated as follows: a p < 0.05 b p < 0.01 c p < 0.001 ROC curves were plotted to compare the diagnostic accuracy of NLR, PLR, CRP, and PCT in the context of bacterial infections versus non-infectious causes.The diagnostic performances of NLR, CRP, and PCT for detecting bacterial infections were evaluated by calculating sensitivity and specificity.NLR: neutrophil-to-lymphocyte ratio; PLR: platelet-to-lymphocyte ratio; CRP: C-reactive protein; PCT: procalcitonin.
Ethical considerations -The prior agreement of the departments' chiefs concerned with the study (the internal medicine department and the hematology laboratory) was obtained, as well as the initial approval from the ethics committee of Sahloul University Hospital of Sousse on May 11, 2022, under the number HS 22-2022, this approval confirms that there are no ethical issues concerning the work.
-Written and verbal consent were obtained from all participants before their involvement in the study, ensuring that their rights, privacy, and confidentiality were rigorously upheld throughout the research process.

Results
A total of 164 patients were included in the study.The study population was divided into two groups: a group of patients with confirmed bacterial infections and a second group of patients with other diagnoses (inflammatory pathologies, neoplasms, venous thromboembolic diseases, etc.).The median age within this study population was 50 years (ranging from 37 to 66 years), with a male-to-female ratio of 0.76.The epidemiological characteristics and the mean rates of NLR, PLR, PCT and CRP are described in Table 1.
Our findings revealed that bacterial infections are linked to a notable increase in the NLR to 9.7 (normal range from 1 to 3) and the PLR to 218.14 (normal range from 90 to 210).Additionally, patients with neoplasms also demonstrate elevated PLR and NLR.
Our results revealed significant variations in NLR based on gender.Indeed, men generally had slightly higher NLR values than women, with means of 7.2 in men and 4.7 in women (p<10 -6 ).In addition to gender-based variations, we also observed significant differences in NLR based on age.Specifically, older subjects had higher NLR values, with a mean of 8.8 compared to 4.3 in younger individuals (p=0.001).
In the group of patients with bacterial infections, there were significantly higher levels of NLR, PLR, PCT, CRP, and absolute neutrophil count (ANC) compared to those with non-infectious diseases (p<10 -6 ) (see Table 2).
Similarly, there was a significant difference between groups with inflammatory diseases, venous thromboembolic diseases, and neoplasms (Table 3).
A significant positive correlation was found between NLR and PLR and inflammation markers (CRP and PCT) (see Table 4).
The analysis of ROC curves showed that the NLR had a sensitivity of 65.38% and a specificity of 76.47% in predicting bacterial infections in patients with clinical symptoms, with an optimal threshold value of 4.3 and an area under the curve (AUC) of 0.72 (95% CI, [0.65-0.8].The PLR also showed prediction performance with a sensitivity of 55.3%, a specificity of 70.5%, an optimal threshold value of 183, and an AUC of 0.6 (95% CI, [0.52-0.68]).
The diagnostic performance evaluation of the four markers, namely CRP, PCT, NLR, and PLR, was conducted to predict bacterial infections in patients with clinical symptoms.CRP had the highest sensitivity and specificity with 69.23% and 88.35% respectively.The results are summarized in Table 5.The area under the ROC curve for NLR was 0.72 (95% CI, [0.65-0.8]),indicating a good diagnostic performance.

Discussion
Bacterial infections are a significant public health problem due to their severity and associated complications.Healthcare professionals are turning more frequently to markers of inflammation to quickly assess these infections and guide diagnostic strategies, regardless of the initial symptoms.Bacterial infections are often accompanied by a systemic inflammatory response, leading to changes in the levels of certain blood cells, including neutrophils, lymphocytes, and platelets.The increase in NLR and PLR during bacterial infections is mainly explained by the body's inflammatory response to the infection.Neutrophils are the major immune cells involved in the defense against bacterial infections.When a bacterial infection occurs, the number of neutrophils in the blood increases rapidly in response to the bacterial invasion.Lymphocytes, on the other hand, are immune cells that play an essential role in regulating the immune response.During a bacterial infection, the number of lymphocytes may temporarily fall as some of them are mobilized to the site of infection to help fight the bacteria.The NLR therefore increases as a result of the increase in neutrophils and the temporary decrease in lymphocytes.Similarly, the PLR ratio increases because platelets, although initially designed for blood coagulation, can also be mobilized in response to inflammation.This increase in platelets can be observed during bacterial infections as a result of the body's general inflammatory response. 4r study analyzed a group of 164 participants aged 19 to 92 years, with a male-to-female ratio of 0.76.The study population was divided into two groups: a group with confirmed bacterial infections and a group with other diagnoses.The primary objective of this study was to describe the diagnostic utility of NLR and PLR ratios in bacterial infections compared to other non-infectious causes of inflammation.
In addition, significant positive correlations were observed between NLR, PLR, and markers of inflammation.
ROC curves confirmed that NLR and PLR were good diagnostic indicators of bacterial infections, with AUC values of 0.72 and 0.6, respectively.These results highlight the importance of these ratios in the management of patients with bacterial infections.Clinical studies, such as the one conducted by Zahorec and al. on patients with confirmed bacterial infections, have demonstrated a significant increase in the NLR compared to reference values, suggesting its utility as a reliable indicator. 9imilarly, Wang and al. have investigated the potential of the PLR as an indicator of systemic inflammation, with a specific focus on sepsis cases.Their findings suggested that PLR could indeed serve as a valuable prognostic indicator for sepsis and a significant association between elevated PLR values and unfavorable outcomes in sepsis patients was demonstrated. 10e use of NLR and PLR markers is crucial for a rapid identification of bacterial infections, particularly when initial clinical manifestations are inconspicuous.These markers offer an interesting insight into the patient's immune and inflammatory status at the beginning of their admission, allowing earlier and more accurate management.Identifying substantial deviations from reference values enables healthcare professionals to promptly suspect an underlying bacterial infection, even with mild symptoms.This proactive approach helps managing ulterior investigations and monitoring the effectiveness of treatments.In consequence, clinicians could predict severe complications and enhance clinical outcomes. 11,12r results revealed significant variations in NLR based on gender.Indeed, men generally had slightly higher NLR values than women, with means of 7.2 in men and 4.7 in women (p < 10 -6 ), consistent with trends observed in other studies, including the one conducted by Fors and al.In this large-scale study, NLR values were on average higher in men (median of 2.69) than in women (median of 2.35) with a p value < 0.001. 13Additionally, our identification of age-related differences, with older individuals exhibiting elevated NLR levels (mean of 8.8) compared to their younger counterparts, is in agreement with the broader literature on age-related changes in inflammatory markers. 14is observation of NLR variations by gender underscores the importance of considering this difference when interpreting results and evaluating the diagnosis of bacterial infections.
Furthermore, the results of our study indicate that these markers have an important diagnostic utility in bacterial infections.[17][18] To illustrate, in our study, the means values were as follows: PCT, 3.2 ng/mL; CRP, 152 mg/L; NLR, 9.7; PLR, 218.These results are consistent with several other recent studies on bacterial infections.In another study by Li and colleagues, the median PCT in patients with bacterial infections was similar, with a value of 0.67 ng/mL. 19Additionally, Ding and al also found comparable results for NLR, with a median of 8.7 in patients with bacterial infections. 2 Similarly, Yang and colleagues reported a mean PLR of 169.5 in patients with bacterial infections, which corroborates our findings. 20wever, it is important to note that in the non-infectious pathology group, we observed significantly lower levels, with mean values of PCT, NLR, and PLR of 0.21 ng/mL, 3.9, and 174, respectively.
In the context of our study, we also evaluated the diagnostic performance of NLR, PLR, and CRP in predicting bacterial infections using the Receiver Operating Characteristic (ROC) curve areas (AUC).Our results showed that NLR had an AUC of 0.72, while PLR had an AUC of 0.6.Although these values indicate a reasonable ability of these markers to differentiate bacterial infections from other pathologies, CRP showed the highest AUC of 0.82, highlighting its superior effectiveness in predicting bacterial infections.
These findings are in line with previous works.For example, Cai and colleagues' study in 2017 which evaluated NLR in predicting bacterial infections in patients with decompensated cirrhosis, confirmed a significant AUC value of approximately 0.824.In this retrospective study involving 2066 decompensated cirrhotic patients, the incidence of hospitalacquired (HA) bacterial infections was approximately 35.87% in the training cohort and 31.05% in the validation cohort.Multivariate analysis identified total bilirubin, albumin, white blood cell count, and NLR as independent predictors of HA bacterial infections.These results reinforce the importance of NLR as a valuable and noninvasive marker for early detection and prediction of hospital-acquired bacterial infections in individuals with decompensated cirrhosis. 21imilarly, Thill and colleagues' study (2019) examined the diagnostic value of NLR in the context of fever and inflammatory syndromes in internal medicine, reinforcing our findings regarding NLR. 4 In this prospective study of 184 hospitalized patients in internal medicine, the NLR was assessed as a diagnostic marker to differentiate bacterial infections from other causes of fever and inflammation.The study revealed that NLR, with an optimal threshold of seven, showed significance in the presence of 82 bacterial infection cases, and its Area Under the ROC Curve (AUC) was 0.64, comparable to CRP and PCT in this clinical context.
It is important to note that CRP has consistently demonstrated superior diagnostic performance in the context of bacterial infections.In this regard, Legrain and al. (2017) also contributed to the understanding of CRP's effectiveness as a distinct indicator of bacterial infections. 22 contrast, another study conducted by Shanshan Ding et al. yielded contradictory results to ours.This retrospective clinical study aimed to differentiate infectious fever from tumor fever (TF) and assess outcomes in non-neutropenic lung cancer patients (NNLCPs) by evaluating PCT, CRP, and NLR as markers of inflammation. 2The study included 588 febrile NNLCPs, of which 311 had bacterial infections, and 277 had TF.Inflammatory markers (PCT, CRP, white blood cells, neutrophils, and NLR) were significantly higher in bacterial infection cases (p < 0.0001).PCT demonstrated the highest predictive value for bacterial infections (AUC = 0.874), followed by CRP (AUC = 0.855) and NLR (AUC = 0.792) (p < 0.0001).In conclusion, PCT outperformed CRP and NLR in diagnosing bacterial infections in febrile patients and proved useful in assessing clinical outcomes and cancer progression in NNLCPs.
In summary, our results on the AUCs for NLR, PLR, and CRP align with the conclusions of previous studies, confirming the variable performance of these markers in predicting bacterial infections.CRP remains the most performant marker, while NLR and PLR retain reasonable diagnostic utility.This understanding can guide clinical decision-making for the precise diagnosis of bacterial infections, taking into account the context and individual patient characteristics.
However, it is important to note that this study has certain limitations.The relatively small sample size may restrict the generalizability of the results to larger populations.It's important to acknowledge that this study has some limitations.While the sample size was relatively small, it's an opportunity for future research to expand and validate the findings in larger populations.Healthcare facilities can leverage this insight to establish tailored reference ranges based on their patient population, considering the variability in NLR and PLR values.Although there isn't a universally accepted cut-off value for these ratios in diagnosing bacterial infections, this presents an opportunity for further research and standardization.By considering the patient's medical history, including chronic inflammatory conditions, clinical characteristics, and the specific type of bacterial infections, healthcare providers can refine the interpretation of baseline values.
Additionally, the fluctuation of these ratios throughout an infection underscores the importance of multiple measurements over time for effective monitoring of treatment response.

Conclusion
In conclusion, this study has demonstrated that NLR and PLR could have a significant diagnostic role in bacterial infections.These markers could contribute to improving the accuracy of diagnosis, assessing the severity of infection, and guiding therapeutic decisions.However, further research and larger studies are needed to confirm and explore these promising results.

Ethics and consent
Ethical approval was obtained from the ethics committee of Sahloul University Hospital of Sousse on May 11, 2022, under the number HS 22-2022, this approval confirms that there are no ethical issues concerning the work.Written and verbal consent were obtained from all participants before their involvement in the study, ensuring that their rights, privacy, and confidentiality were rigorously upheld throughout the research process.

Data availability
Underlying data Dryad: Data from: The interest of inflammatory biomarkers in the diagnostic approach, https://doi.org/10.5061/dryad.n02v6wx3d. 23is project contains following dataset: Farhat Hached University Hospital of Sousse (Ringgold ID: 280368), Internal Medicine department, Sousse, Sousse, Tunisia I congratulate all authors for their strong choice of subject and the quality of the manuscript.I have just some remarks to make better your work : 1.I suggest to add "Bacteria" or "bacterial infection" as a keyword.
2. There are 2 abbreviations that were explained for the first time in the discussion but were present in the abstract (AUC) and in results for (ROC) so put the explanations on their first appearance.
3. You should add in excluded criteria the exclusion of viral infections and parasitic.4. exclude the number of the population from the methodology because its a result " a prospective comparative study was conducted within the internal medicine department on patients admitted between January 1st and July 31st, 2023." These remarks will just improve your work and do not decrease in any way its rigour and quality.

Aya Chakroun
La Rabta University Hospital, Tunis, Tunisia I congratulate you on the choice of subject and the quality of the manuscript.However, there are a few improvements that should be made: In the "methods" section: ○ -Please remove '' To address the research question'' -Please remove the expression "The study included a total of 164 patients " .
-Please add to the section "Laboratory values and clinical assessment", an indication of good control of the analytical process through the daily passage of internal quality controls.
In the "results" section: ○ -please add "footnotes" to the tables, explaining abbreviations and add confidence intervals of r.
I congratulate you for your scientific rigour and the quality of the discussion of the relevant scientific literature.

Is the work clearly and accurately presented and does it cite the current literature? Yes
Is the study design appropriate and is the work technically sound?Yes

Are sufficient details of methods and analysis provided to allow replication by others? Yes
If applicable, is the statistical analysis and its interpretation appropriate?Yes Are all the source data underlying the results available to ensure full reproducibility?
No source data required

Are the conclusions drawn adequately supported by the results? Yes
Competing Interests: No competing interests were disclosed.
Reviewer Expertise: areas of expertise: benign and malignant hematologyHematology analyzer technologiesmedical educationStatistical analyses and biomedical research I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.

Acceptance Statement:
After thorough analysis, I recommend this article for publication, with minor revisions.Its findings are compelling and make a valuable contribution to the field.Expanding the discussion and addressing confounding factors will enhance its overall impact.

Is the work clearly and accurately presented and does it cite the current literature? Yes
Is the study design appropriate and is the work technically sound?Yes

Are sufficient details of methods and analysis provided to allow replication by others? Yes
If applicable, is the statistical analysis and its interpretation appropriate?Yes Are all the source data underlying the results available to ensure full reproducibility?Yes

Are the conclusions drawn adequately supported by the results? Yes
Competing Interests: No competing interests were disclosed.
Reviewer Expertise: Hematology and immunogenetics I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.
The benefits of publishing with F1000Research: Your article is published within days, with no editorial bias • You can publish traditional articles, null/negative results, case reports, data notes and more • The peer review process is transparent and collaborative • Your article is indexed in PubMed after passing peer review • Dedicated customer support at every stage • For pre-submission enquiries, contact research@f1000.com NLR: neutrophil-to-lymphocyte ratio; PLR: platelet-to-lymphocyte ratio; CRP: C-reactive protein; PCT: procalcitonin.Note: Probability values (p) indicate the degree of significance of the correlation between parameters.p-values are associated with each correlation coefficient (r).A p-value less than 0.05 is considered statistically significant.The correlation coefficient (r) measures the strength and direction of the linear relationship between parameters.An r value close to 1 indicates a strong positive correlation, while an r value close to -1 indicates a strong negative correlation.An r value close to 0 suggests a weak or non-existent correlation between the parameters.*95% Confidence Interval of r.

Figure 1 .
Figure 1.Comparison of Areas under the ROC curve for prediction of bacterial infections by CRP, PCT, NLR, and PLR markers.
Data are available under the terms of the Creative Commons Zero "No rights reserved" data waiver (CC0 1.0 Public domain dedication).

Table 1 .
The epidemiological characteristics and the median rates of NLR, PLR, PCT and CRP of the Study Population by Disease Type.

Table 5 .
Comparison of Diagnostic Performances between CRP, PCT, NLR, and PLR for Predicting Bacterial Infections.

the work clearly and accurately presented and does it cite the current literature? Yes Is the study design appropriate and is the work technically sound? Yes Are sufficient details of methods and analysis provided to allow replication by others? Yes If applicable, is the statistical analysis and its interpretation appropriate? Yes Are all the source data underlying the results available to ensure full reproducibility? Yes Are the conclusions drawn adequately supported by the results? Yes Competing Interests:
No competing interests were disclosed.

have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.
This is an open access peer review report distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.