Background
Pleural effusion is the most common manifestation of pleural disease and can develop as a result of over 50 different pleuropulmonary or systemic disorders [
1]. Although the annual incidence of pleural effusions is difficult to assess, since pleural effusions are usually the result of an underlying disease, there are an estimated 1.5 million cases per year in the United States alone [
2]. The most common causes for pleural effusions are congestive heart failure, infection (e.g., pneumonia), malignancy and pulmonary embolism. In general, the prevalence of pleural effusions in industrialized countries is approximately 320 cases per 100,000 residents, and is directly related to the prevalence of the underlying diseases [
3]. For example, among countries with a high incidence of tuberculosis, the most frequent cause of pleural effusions is tuberculosis [
4].
A major clinical challenge in the diagnosis and management of pleural effusions remains the differentiation between malignant and infectious effusions, using the right laboratory test leading to an accurate diagnosis, due to their different outcome and management [
5]. Thus, the need for biomarkers that may help in this differentiation, in conjunction with the ones previously suggested by our group (IL-1A, IL-6, TNF) [
6], is imperative.
The YKL-40 or Chitinase 3-like 1 protein is a growth factor for chondrocytes and fibroblasts. The precise role of this factor has not been clearly defined, but it seems that YKL-40 promotes fibroblast growth and is expressed by many cell types, such as synovial, smooth muscle cells, granulocytes, macrophages, liver and cancer cells, possessing a role in growth, tissue remodeling and inflammation [
7]. YKL-40 levels increase during inflammation, since it plays an important role in chemiotaxis and in the accumulation and activation of cells associated with inflammation [
7,
8]. While high levels in serum and tissues have been observed in many lung diseases [
9,
10], only three studies have assessed this factor as a diagnostic tool in pleural effusions [
11‐
13].
Macrophage inflammatory protein (MIP-1a) is a cytokine belonging to the subgroup of CCL chemokines. Chemokines are low molecular weight proteins that act as mediators in chemotactic migration of leukocytes. The synthesis of chemokines is inducted by several cells after the activation of inflammation. CCL chemokines are chemotactic for mononuclear cells, neutrophils and other granulocytes. MIP-1a plays a significant role in the chemotactic activity of monocytes and of mononuclear phagocytes. In addition, MIP-1a has a different effect on chemotactic T-lymphocytes, natural killer cells, cytotoxic T-cells, B-cells, basophils and eosinophils. In general, MIP-1a is induced by YKL-40 in lung inflammatory diseases [
10], and is expressed at the stages of both acute and chronic inflammation [
14,
15]. However, few studies have examined the role of this protein in pleural effusions [
16‐
18].
The purpose of this research was to measure YKL-40 and MIP-1a in conjunction, in both pleural fluids and in the serum of patients with well-defined causes of pleural effusion, in order to ascertain their use in the differential diagnosis among the underlying diseases.
Discussion
In the present study we measured the protein levels of YKL-40 and MIP-1a in pleural fluids, in order to demonstrate the correlation with their circulating levels in peripheral blood, and to determine the diagnostic value of these molecules in the differential diagnosis of pleural effusions, especially between pleural effusions associated with lung cancer and tuberculosis.
The levels of YKL-40 in pleural effusions were similar among all examined groups, without any statistical differences between them. On the contrary, YKL-40 values in the peripheral blood of patients with tuberculosis were statistical significantly lower in comparison with all the other patient categories, even after age correction. YKL-40 is expressed in the lung and serum of patients with bronchial asthma, chronic obstructive pulmonary disease, pulmonary fibrosis, sarcoidosis, lung cancer, respiratory infections, tuberculosis and cystic fibrosis [
21‐
24].
The biological activity of YKL-40 is still largely unknown, and although a specific cell receptor for YKL-40 has not yet been found, it seems to be associated with collagen type I, II and III [
25]. YKL-40 activates intracellular pathways through the cell membrane [
7,
8], while acting like chitin sensor directing the macrophages and activating the anti-inflammatory response to infection [
26]. Additionally, YKL-40 promotes the migration of endothelial cells and contributes to the diversification of the morphology of the endothelium. It is also found in special granules of neutrophil and mast cells [
8,
27]. Nevertheless, YKL-40 inhibits oxidative damage in the lung and increases the Th2 immune response, regulates apoptosis, activates macrophages and contributes to fibrosis and rehabilitation of tissue injury [
28,
29]. The expression of YKL-40 seems to be affected by IFN-γ, an important cytokine to Th1 immune response [
7], while it is also activated from cytokines IL-6, IL-13, IL-17 and IL-18, which play an important role in inflammation [
7,
29].
YKL-40 has not been extensively studied in pleural effusions. Kim
et al measured the levels of YKL-40 in pleural fluid and serum of patients with tuberculosis, malignant effusions, parapneumonic effusions and transudates due to congestive heart failure [
11]. Their results suggest that YKL-40 levels were higher in pleural fluids from exudates versus transudates. A similar finding was reported by Kayhan
et al [
12]. Both studies are not in accordance with our observations, in which YKL-40 levels were similar in both exudates and transudates. The high levels of YKL-40 in our transudates could be attributed to the existence of fluid in interstitial lung space, to the increased pressure in the pleural capillaries and to endothelial vessel damage, factors that can lead to increasing levels of YKL-40 [
7]. In addition, patients in our study with congestive heart failure and transudates had a considerable amount of co-morbidities, such as atherosclerotic coronary artery disease, type II diabetes and smoking habit, which can also contribute to the increasing levels of YKL-40 [
10,
22,
30,
31].
Moreover, in the study by Kim
et al, YKL-40 levels were higher in pleural fluids from tuberculous pleural effusions and lower in malignant effusions [
11]. In the present study the levels of YKL-40 in pleural fluids were similar among tuberculosis, lung cancer and metastatic cancer of non-lung origin. Our study, however, agrees with Kim
et al study’s findings regarding the ratio of YKL-40 in tuberculous pleural fluid compared to that of the serum, since in both studies this percentage was higher than in the other groups [
11]. This observation accounts for an important finding of our study, given that it differentiates tuberculous pleural effusions from the other exudates subgroups, with high sensitivity (91 %) and specificity (100 %), despite the significant age difference of TB patients. Although recent studies have shown that YKL-40 serum levels could be utilized in the diagnosis of endometrial carcinoma (with 74 % sensitivity and 87 % specificity) [
32], and of esophageal squamous cell carcinomas (with 73 % sensitivity and 84 % specificity) [
33], or could provide information regarding the response to chemotherapy and overall survival in patients with small cell lung cancer [
34], this research provides evidence for the first time that YKL-40 could also be used for the differential diagnosis of tuberculosis from other pleural effusions.
The diagnostic performance of YKL-40, in comparison to already established markers like C-reactive protein (CRP) and ADA, is extremely promising. CRP had 100 % sensitivity and only 46 % specificity when distinguishing TB from malignant effusions [
35], while in another study its sensitivity was 74 % and its specificity 77 %, respectively [
36], findings that were verified by a third study, in which CRP performed poorly (AUC = 0.57 vs. 0.86 for YKL-40 in ours) and only ADA performed extremely well (AUC = 0.94) [
37]. Another study, in which ADA levels were utilized to distinguish tuberculous from malignant effusions, had 89 % sensitivity and 70 % specificity [
38], while two more studies in which TB was compared to effusions of all other origins, ADA had 87–88 % sensitivity and 92 % specificity [
39,
40]. Based on the above, we can deduce that YKL-40 is superior as a diagnostic marker in distinguishing tuberculous from malignant effusions than CRP, and that is has a similar if not a better performance when compared to ADA.
Yu
et al have shown that the immunological environment of a tuberculous pleural effusion is characterized by distinct biomarkers and in different concentrations in comparison with the serum [
17]. CCL1, CCL21 factors and IL-6 are also elevated in tuberculous pleural effusions and have a specific antigenic reaction in their expression. Following the antigenic stimulation from the Mycobacterium tuberculosis, these factors are secreted in large amounts from the mononuclear cells of pleural fluid, activating YKL-40.
The anti-inflammatory protein of macrophage MIP-1a (CCL3) belongs to the cytokines family and in the subgroup of MIP-1 CC chemokines [
41]. MIP-1 chemokines are produced by many cells, especially from T and B lymphocytes, neutrophils, dendritic cells, osteoblasts, astrocytes, epithelial cells of the lower airways, alveolar macrophages, eosinophils, fibroblasts and natural killer cells. The production of MIP-1 is caused by various proinflammatory factors and cytokines, such as viral infection, Gram positive bacteria, TNF-a, IFN-γ, IFN7, IL-1 α/β, IL-13 and many others [
16,
41,
42]. MIP-1 chemokines act through surface receptors and while having strong chemotactic action, they play a significant action in the activation of inflammation and hemostasis [
41]. Their actions include target cells via chemotaxis, degranulation, phagocytosis and mediator synthesis [
41,
42]. MIP-1 chemokines play an important role in both acute and chronic inflammation, primarily with the recruitment of proinflammatory cells. Their role is particularly important in chemotaxis of T lymphocytes in the inflammatory tissues, but also in the migration of monocytes, dendritic cells and natural killer cells [
42].
.It seems that this group of chemokines, and especially MCP-1 (CCL2) whose levels increase significantly in malignant pleural effusions [
16,
43‐
46], plays an important role in inflammatory lung diseases such as asthma, sarcoidosis, pulmonary fibrosis, but also in tuberculosis, pleural effusions, pneumonia, acute espiratory distress syndrome (ARDS) and tumors development [
42,
47‐
49].
In our study, MIP-1a protein levels were high in all patient groups, both in the pleural fluid and the peripheral blood, with the exception of patients with malignant effusions associated with lung cancer. Their values were statistically significant lower in comparison with all the other categories (tuberculous, parapneumonic, transudates and malignant effusions associated with metastatic cancer of non-lung origin), both in the pleural fluid and the peripheral blood. In the study by Mohammed
et al, the levels of MIP-1a were elevated in patients with complicated and non-complicated parapneumonic pleural effusions, while they were low in malignant pleural effusions and even lower in transudates associated with congestive heart failure [
16]. According to that study, the chemotactic activity of MIP-1a was reduced in malignant pleural effusions compared with parapneumonic pleural effusions. In another study by Yuan
et al, MIP-1a along with its receptor CCR1, facilitate the migration of malignant hepatoma cells through Ca
2+ ion channels, thus playing a significant role in hepatocellular carcinoma invasion and metastasis [
50]. These findings are in accordance with ours, since MIP-1a levels were higher in effusions associated with metastatic cancers of non-lung origin, compared to effusions associated with lung cancer. Our findings also suggest that MIP-1a could also be used for the differential diagnosis of lung cancer from tuberculosis (sensitivity 100 %/specificity 70 %), from parapneumonic effusions (sensitivity 80 %/specificity 100 %) and especially from metastatic tumors of non-lung origin (sensitivity 67 %/specificity 100 %), which are more difficult to distinguish. MIP-1a has not been widely used for diagnostic purposes, apart from malignant gliomas, in which MIP-1a levels provided 100 % sensitivity and 88 % specificity for the diagnosis of this tumor type versus controls [
51].
YKL-40 and MIP-1a belong to the same biochemical pathway, and MIP-1a is induced by YKL-40 in lung inflammatory diseases. Letuve
et al demonstrated that YKL-40 causes the release of three chemokines from the alveolar macrophages of smokers with or without chronic obstructive pulmonary disease (COPD). Precisely IL-8, MCP-1 and MIP-1a seem to be associated with the pathogenesis of COPD through tissue inflammation and fibrosis [
10]. In the study of Sutherland
et al, the researchers observed the inhibitory action of acidic mammalian chitinase (AMC) in the recruitment of neutrophils through MIP-1a action. They speculated that AMC and chitinase like proteins (CLP), which YKL-40 is a member of, have cross-regulatory actions. The increased expression of CLPs leads to neutrophils recruitment and causes increased secretion of MIP-1a [
52].
Competing interest
The authors declare that they have no conflict of interests.
Authors’ contribution
TA and NMS conceived the study and participated in its design along with NS, EN and AG. Data acquisition was performed by NS, EN, KB and AP. Data was analyzed by NS and SZ and interpreted by NS, NMS and SES. TA and NS drafted the manuscript, which was revised by TA, NS, NMS and SES. All authors read and approved the final manuscript.