The diagnostic implications of the pleural fluid cell differentiation

Introduction

The current British Thoracic Society (BTS) guideline recommends a diagnostic pleural aspiration as the initial investigation of an undiagnosed unilateral pleural effusion (1). A pleural fluid cell differentiation is recommended alongside routine biochemical, microbiological, and cytological tests. The literature reports that lymphocytic pleural effusions are most associated with malignancy, tuberculosis, and congestive heart failure (2-5). In contrast, neutrophilic predominance in pleural fluid is typically indicative of acute inflammatory processes, particularly infection (1,6). Eosinophilic effusions meanwhile are most frequently linked to the presence of air or blood within the pleural space (7,8). However, the sensitivity and specificity of cell differentiation from the initial pleural aspiration have not been reported. In this prospective cohort study, we aimed to establish the value of pleural fluid cell differential in the investigation of undiagnosed pleural effusions. We present this article in accordance with the STARD reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1017/rc).

Methods

Study design

Between 2008 and 2016, patients presenting to a UK tertiary pleural service with undiagnosed pleural effusions requiring a diagnostic thoracentesis were prospectively recruited to an observational study.

Ethics

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the South West Regional Ethics Committee (IRAS ethics number 08/H0102/11). Informed consent was taken from all the patients.

Setting

This study was based at Southmead Hospital (North Bristol NHS Trust) between December 2008 and December 2016.

Participants

All patients within this time were eligible if they had a new undiagnosed unilateral pleural effusion identified on a chest radiograph requiring thoracentesis. Participants were followed up at 12 months, with two consultant respiratory physicians independently ascertaining the cause of the effusion upon review of their clinical history, radiological imaging and laboratory results. Exclusion criteria included instances where a consensus diagnosis was never reached, and those without cytology results. All patients consented at the point of thoracentesis, which was indicated as per normal clinical practise.

Variables and data collection

Samples were collected and analysed as per local trust practise. Cell differentiation was categorised as per the BTS guidelines. Cytology from pleural aspiration was categorised as ‘neutrophilic’ in the presence of >50% neutrophil count and as ‘lymphocytic’ when lymphocytes accounted for >50% of the differential cell counts. The effusion was defined as ‘eosinophilic’ in the presence of >10% eosinophils. If an effusion did not meet any of the above criteria, it was classified as ‘mixed’. Some effusions were too heavily blood stained for a formal differential cell count and were classified as ‘blood’. After these five groups had been defined, the ratios of each cell type for a given sample were not considered for analysis. Data regarding patient characteristics was collected retrospectively and input into a standardised spreadsheet.

Bias

Recruitment-related selection biases were minimised through a broad inclusion and narrowly defined exclusion criteria. To mitigate performance bias, diagnosing consultants were blinded to the study’s data handling process.

Statistical analysis

Descriptive statistics were used to compare and analyse patient demographics and pleural fluid cytology. Sensitivities, specificities and accuracies were calculated with likelihood ratios and 95% confidence intervals (CIs) were calculated for lymphocytic, neutrophilic and eosinophilic effusions. The analysis was conducted using Microsoft Excel and SPSS v27.

Results

Participants

A total of 1,023 patients presenting with new pleural effusions were recruited between 2008 and 2016. Eighteen patients were excluded as they did not have pleural fluid cell counts reported, and eight patients had an unclear 12-month diagnosis despite extensive investigation leaving 997 cases with full data for analysis.

Descriptive data

The median age of this cohort was 73 (95% CI: 71–73) years in the group with malignancy and 72 (95% CI: 69–73) years in the non-malignant group, with a male predominance (n=639, 64.1%). Biochemical analysis revealed that 84.75% (n=845) of the cohort had an exudative effusion. At the 12-month diagnosis, 532 (53.4%) patients were diagnosed with a malignant aetiology and 465 (46.7%) with a benign aetiology, of which 182 (18.3%) were infectious.

Of the 532 patients with malignant disease, 139 (26.1%) had a primary lung cancer, 69 (13.0%) had mesothelioma, and 68 (12.8%) had a primary breast cancer. Of the 182 patients with infective disease, 20 (11.0%) had tuberculosis, 68 (37.4%) had a simple parapneumonic effusion (PPE) and 61 (33.5%) had a complicated PPE. Of the 283 patients with non-malignant, non-infective disease, 111 (39.2%) had congestive heart failure, 34 (12.0%) had benign asbestos pleural effusion (BAPE) and 12 (4.2%) had renal failure.

Cell differential

Cell differentials in the 997 patients were classified as follows: 423 were mixed (42.4%), 352 were lymphocytic (35.3%), 95 were neutrophilic (9.5%), 72 were eosinophilic (7.2%), and 55 were heavily blood-stained making differential counts impossible (5.5%) (Table 1).

Over half of the lymphocytic effusions, 53.7% (n=189), were found to be of malignant origin. Of the remaining 163 patients, 51 (31.3%) were diagnosed with congestive heart failure, 17 (10.4%) had BAPE, 26 (16.0%) had PPE, and 15 (9.2%) had tuberculous pleurisy.

Of the 95 neutrophilic effusions, 80% (n=76) had a primarily infectious aetiology. Thirteen patients with neutrophilic effusions (13.7%) were found to have a malignant aetiology, representing 2.4% (13 out of 532) of the malignant group. Only six patients with a neutrophilic effusion had neither a malignant nor infectious diagnosis.

The prevalence of malignancy among eosinophilic pleural effusions was 43.0% (31 out of 72 patients). Other aetiologies included ten (13.9%) PPE, seven (9.7%) effusions were asbestos-related, and five (6.9%) had cardiac failure. Among the 55 heavily blood-stained effusions, malignancies were diagnosed in 49.1% (n=27).

The majority of patients (42.4%) did not have a predominant cell type and were classified as mixed. Among the 423 patients with mixed pleural effusions, 64.3% (n=272) had a malignant aetiology, 10.9% (n=46) had an infective aetiology, and 24.8% (n=105) had a non-malignant, non-infective pleural effusion. Malignancy was more prevalent in the mixed effusion group than the overall cohort (64.3% vs. 53.4%).

The distribution of diagnoses across all effusion types is summarised in Table 1.

Diagnostic accuracy analysis

Figure 1 details the diagnostic accuracy information for each of the three main effusion categories with their respective diagnoses. The highlighted data represents the accuracy information for the diagnosis most associated with each predominant cell differential.

Figure 1 Diagnostic performance metrics (sensitivity, specificity, accuracy and positive likelihood ratio with 95% CIs) of individual cell differentials across the three primary diagnoses. CI, confidence intervals.

Discussion

Reporting cell differentiation within pleural fluid is advocated by clinical guidelines (1). It is a time-consuming reporting process for a cytopathologist, with a scant evidence base to support this practice. This study represents the largest prospective assessment of the diagnostic utility of pleural fluid cell differentiation in patients presenting with undiagnosed pleural effusions. While lymphocytic effusions inferred a malignant aetiology, neutrophilic effusions inferred an infective aetiology and eosinophilic effusions inferred an aetiology which was neither malignant nor infective, they lacked the diagnostic accuracy to be used as either a rule-in or rule-out test.

Current literature suggests malignancy, tuberculosis and cardiac failure are the commonest causes of lymphocytic effusions (2-5). In our study, the common aetiologies were malignant pleural effusion (53.7%), congestive heart failure (14.5%), BAPE (4.8%), and PPE (7.7%). The UK has a low incidence of pleural tuberculosis (9), explaining the low rates in lymphocytic effusions compared to other prospective cohorts (15 of 349, 4%). Lymphocytic effusions had a poor sensitivity (35.5%) and specificity (65.0%) for diagnosing malignancy, demonstrating their limited clinical utility.

While neutrophilic effusions had a high specificity (97.7%) for diagnosing infection, they had a very poor sensitivity (46.9%). In addition, nearly one in seven patients with neutrophilic effusions had an underlying malignant aetiology, the value as a rule-in test is limited given the consequences of missing this diagnosis. Furthermore, given the plethora of investigations which markedly increase the pre-test probability of infection (including biochemical analysis, clinical findings and chest radiographs), the necessity for clinicians to use pleural cytology for ‘ruling in’ the diagnosis is much less.

Eosinophilia is classically associated with the presence of air and blood in the pleural space (7,8). In this study, only one out of nine cases of haemothorax was noted in eosinophilic effusions. While there was a correlation between eosinophilic effusions and an underlying malignant aetiology (>40%), the likelihood ratio was <1 (0.66).

Among the 997 patients included in this study, more than 40% had pleural effusions had no predominant cell type, according to BTS criteria, and were classified as mixed. The high prevalence of mixed-cell effusions, together with the scarce evidence regarding their clinical significance, highlights the restricted utility of cytological differentiation in real-world practice. The inability to classify pleural fluid by a predominant cell type contributes to the poor diagnostic accuracy observed across the three major effusion types. Although excluding mixed or blood-stained effusions from the analysis could have improved diagnostic precision, such exclusion would have limited the generalisability and clinical relevance of our findings.

This study is single centre which limits some generalisability of its findings. The incidence of certain conditions causing pleural effusions varies significantly internationally (10). The prevalence of different disease in our population (i.e., 53% malignancy) reflects the more elderly, western demographic of patients (11). This includes a high incidence of mesothelioma (accounting for 13% of all malignancy in our population), given the history of asbestos industry in the UK (12). In this study, we did not explore the delay in processing or the method of sample handling, which could influence specimen interpretation due to cellular degeneration. However, all samples were handled as per normal clinical care.

Conclusions

In conclusion, this is the largest prospective study of cell differential reporting in undiagnosed pleural effusions. This study demonstrates that cell counts in pleural aspirates carry limited diagnostic value. Whilst they might guide towards the likely diagnosis when combined with other diagnostic indicators (history, radiology, etc.), they lack the diagnostic accuracy as a rule-in/out test. Physicians should therefore be cognisant of these findings when interpreting pleural fluid results.

Acknowledgments

The authors would like to recognise that an undeveloped abstract of this paper was presented at the 2023 European Respiratory Society International Congress, in session “Inflammatory endotyping: the macrophage across disease areas”. This has been published in the European Respiratory Journal 2023; 62: Suppl. 67, PA3298. This work was carried out by the same authorship group.

Footnote

Reporting Checklist: The authors have completed the STARD reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1017/rc

Data Sharing Statement: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1017/dss

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1017/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1017/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the South West Regional Ethics Committee (IRAS ethics number 08/H0102/11). Informed consent was taken from all the patients.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Roberts ME, Rahman NM, Maskell NA, et al. British Thoracic Society Guideline for pleural disease. Thorax 2023;78:s1-s42. [Crossref] [PubMed]
2. Pettersson T, Riska H. Diagnostic value of total and differential leukocyte counts in pleural effusions. Acta Med Scand 1981;210:129-35. [Crossref] [PubMed]
3. Li C, Kazzaz FI, Scoon JM, et al. Lymphocyte predominant exudative pleural effusions: a narrative review. Shanghai Chest 2022;6:5.
4. Mercer RM, Corcoran JP, Porcel JM, et al. Interpreting pleural fluid results . Clin Med (Lond) 2019;19:213-7. [Crossref] [PubMed]
5. Dooey D, William L, Cheng A, et al. Lymphocytic pleural effusions: aetiology and frequency in a UK district general hospital. Eur Respir J 2019;54:PA3089.
6. Roy B, Shak HJ, Lee YCG. Pleural fluid investigations for pleural infections. J Lab Precis Med 2021;6:12.
7. Adelman M, Albelda SM, Gottlieb J, et al. Diagnostic utility of pleural fluid eosinophilia. Am J Med 1984;77:915-20. [Crossref] [PubMed]
8. Kalomenidis I, Light RW. Pathogenesis of the eosinophilic pleural effusions. Curr Opin Pulm Med 2004;10:289-93. [Crossref] [PubMed]
9. Arnold DT, Bhatnagar R, Fairbanks LD, et al. Pleural fluid adenosine deaminase (pfADA) in the diagnosis of tuberculous effusions in a low incidence population. PLoS One 2015;10:e0113047. [Crossref] [PubMed]
10. Porcel JM, Esquerda A, Vives M, et al. Etiology of pleural effusions: analysis of more than 3,000 consecutive thoracenteses. Arch Bronconeumol 2014;50:161-5. [Crossref] [PubMed]
11. Prathap R, Kirubha S, Rajan AT, et al. The increasing prevalence of cancer in the elderly: An investigation of epidemiological trends. Aging Med (Milton) 2024;7:516-27. [Crossref] [PubMed]
12. Gilham C, Rake C, Hodgson J, et al. Past and current asbestos exposure and future mesothelioma risks in Britain: The Inhaled Particles Study (TIPS). Int J Epidemiol 2018;47:1745-56. [Crossref] [PubMed]