Abstract
Heparin-binding protein is a serine protease that is mobilized rapidly from emigrating polymorphonuclear leukocytes that acts as a chemoattractant activator of monocyte and macrophages. We investigated the potential role and efficacy of serum and cerebrospinal fluid heparin binding protein in differentiating bacterial meningitis from tuberculosis and viral meningitis. A case diagnosed with acute bacterial meningitis (n:37), viral meningitis (n:30) and tuberculous meningitis (n:30) was included in this study. The diagnosis was based on history, clinical criteria, cerebrospinal fluid examination, latex agglutination and culture, and response to therapy. Heparin-binding protein was measured using enzyme-linked immunosorbent technique in both cerebrospinal fluid and serum. Cerebrospinal fluid heparin-binding protein levels were 7.81 ± 0.23 ng/mL in bacterial meningitis, 6.11 ± 0.3 ng/mL in tuberculosis meningitis and 5.75 ± 0.1 ng/mL in viral meningitis. The mean serum level was 14.98 ± 1.1 ng/mL in bacterial meningitis, 6.89 ± 0.4 ng/mL in tuberculosis meningitis, and 6.02 ± 0.4 ng/mL in viral meningitis. Both heparin-binding protein levels were significantly higher in patients with bacterial meningitis. We found that serum and cerebrospinal fluid heparin binding protein is a useful marker for differentiating bacterial meningitis from non-bacterial meningitis.
Keywords: Heparin binding protein, Bacterial meningitis, Viral meningitis, Tuberculosis meningitis
Introduction
Meningitis is the most common infectious central nervous system disease. Although meningitis is a relatively less common disease in developed countries, the incidence of meningitis is high in developing countries such as Turkey.
Bacterial meningitis (BM) is a life-threatening disease with high mortality rates and notorious neurologic sequelae ratios, especially in cases where diagnosis and antibiotic administration are delayed. It is one of the top 10 causes of infection-related death worldwide and 30–50% of its survivors have permanent neurological sequelae [1]. Hence, BM represents a medical emergency and early antibacterial therapy is essential for a good prognosis. However, early diagnosis of BM remains a great challenge because symptoms such as fever, headache, vomiting and so on, are not specific [2]. Therefore, a laboratory test, quick and easy to perform with high sensitivity and specificity, is required to identify all patients with BM on admission; delayed treatment of false negative cases may be fatal. The conventional diagnostic method of BM using the cerebrospinal fluid (CSF) white blood cell (WBC) count, lactate, protein, glucose, and plasma C-reactive protein (CRP) levels are often not discriminative enough in the early phase of the disease, since if they have low sensitivity and specificity for the diagnosis. According to WHO guidelines, (WHO, 2011) bacterial culture is the gold standard method for BM confirmation but misclassification is a possible risk in cases where a negative culture is obtained. CSF culture is a method with high specificity, but approximately 50% of suspected cases are not culture approved. Also, the result of culture takes days to achieve [1, 3–5].
Thus, easy and rapid predictors for diagnosing BM are needed. New potential biomarkers are being investigated to differentiate BM from non-bacterial.
Heparin-binding protein (HBP), also known as azurocidin or cationic antimicrobial protein of 37 KDa (CAP 37) is a member of the serine proteinase. It is secreted from neutrophil leukocytes early stages of inflammation and plays a central role in early capillary leakage and extravasation of neutrophils. First, this protein attracted attention for its antimicrobial properties, and later it was found that HBP acts as a multifunctional mediator in infection and inflammation. HBP is pre-produced in polymorphonuclear leukocytes and is rapidly released after stimulation by various bacterial structures, cytokines, inflammatory factors, and chemotactic factors [6, 7]. Also, it can exert significant subsequent effects on the immune system. Because of these properties, HBP is seen as a promising candidate that can be used in the detection of early infection. Although several studies related to HBP levels in severe infections exist [8–12], its role in patients with meningitis has not been investigated sufficiently.
This study aims to evaluate the role and effectiveness of serum and CSF heparin binding protein in the differential diagnosis of BM.
Material and Methods
Study Setting and Population
This study was performed on patients who were admitted to the Department of Infectious Diseases and Clinical Microbiology Department of Gaziantep University Medical Faculty, between January 2018 and June 2020 with a pre-diagnosis of meningitis. Thirty-seven BM, 30 tuberculosis meningitis (TM) and 30 viral meningitis (VM) patients were included in the study.
The protocol was approved by the Clinical Research Ethics Committee of Gaziantep University with the protocol number (No: 2015/363) and the research was conducted in compliance with the Declaration of Helsinki (version 2008).
The purpose and procedures of the study were explained and written informed consent was received from each participant or their guardians before participation.
Inclusion criteria: Patients with meningitis admitted to University of Gaziantep Medicine Faculty Hospital.
Exclusion criteria: Patients who were hospitalized with a prediagnosis of meningitis, cerebrovascular event, malignant infiltration of the meninx, immune compromised patients for any reason, pregnant women, and those lacking laboratory data used for routine diagnosis were excluded from the study.
Diagnosis of Meningitis
The diagnosis was made according to the patient's clinic and CSF examination criteria. Lumbar puncture (LP) was performed if clinical signs and symptoms suggest meningitis. LP can be safely performed in the absence of increased intracranial pressure, focal neurological findings and/or papillary edema. The diagnosis of bacterial meningitis is based on a course of clinical history and laboratory experiments. Clinical features were such as the acute onset of headache, fever, and signs of meningeal irritation. Laboratory diagnosis of acute bacterial meningitis (ABM) was made by CSF examination. Positive CSF findings were pleocytosis (≥ 5/mm3, mainly neutrophilic), elevated protein concentration (≥ 45 mg/dL) a reduced ratio of CSF glucose to serum glucose (≤ 0.60) additionally a positive CSF culture, smear, or PCR for bacterial pathogens or good specific response to antibacterial therapy.
Viral meningitis cases had clinical features such as, the acute onset of headache, fever, and signs of meningeal irritation. In addition, there were no signs of cortical involvement such as altered consciousness, aphasia, or seizures. Viral meningitis CSF findings were also pleocytosis (≥ 5/mm3, mainly lymphocytic), a negative CSF stain, culture, or PCR for bacteria, mycobacteria, fungi and a positive PCR for viral pathogens or full recovery without any specific treatments including antibacterial or antituberculosis therapy.
Diagnosis of TM clinical symptom, CSF criteria, cerebral imaging criteria and it was placed based on other evidence of tuberculosis. If Mycobacterium tuberculosis was detected in CSF of cases a definite diagnosis of TM was made.
Measurements
Blood and CSF were taken for analysis before starting treatment. The serum was separated with a standard centrifugation procedure and immediately divided into portions that were kept tightly closed at − 70 °C until analysis.
Measurement of Heparin Binding Protein
Heparin Binding Protein concentrations in serum and CSF samples were determined by the enzyme-linked immunosorbent assay (ELISA) using commercial kits (Human Heparin Binding Protein ELISA kits, Rel Assay Diagnostics, catalog number: MBS739465) following the manufacturer’s guidelines. The minimum detection limits in the assays were 0.01 ng/ml for HBP. Absorbance was measured at 450 nm as a reference wavelength by an automated ELISA plate reader (Rel Assay Diagnostics, Gaziantep, Turkey). Heparin Binding Protein concentrations were determined by calculating the mean optical densities of the duplicates, which were correlated to the results from the standard curve. The day-to-day variation of the assay had a coefficient of variance of 10%. Serum and CSF concentrations of HBP were presented as ng/ml.
Statistical Analyses
All statistical analyses were performed using PASW, version 18.0 (SPSS Inc., Chicago, IL, USA) for Windows. Continuous variables are presented as mean ± SEM and categorical variables are presented as n of patients (%). The Kolmogorov–Smirnov test was used to test the normality of the distribution of continuous variables. Statistical analysis of data between two groups was performed using an unpaired t-test for parametric data and Mann–Whitney U-test for nonparametric data. Correlations were tested with Spearman’s rank correlation coefficient. χ2-test was used for categorical variables.
To perform evaluate the sensitivity and specificity for distinguishing BM from other meningitis, we analyzed the clear cutoff values using the receiver operating characteristic (ROC) curve. The data obtained using ROC curves are presented as 95% confidence interval (CI) and standard error (SE) values. A two-tailed p-value < 0.05 was considered statistically significant.
Results
A comparison of demographic and laboratory profiles of cases and HBP levels in CSF and serum are illustrated in Table 1.
Table 1.
Comparison of demographic and laboratory profiles of cases with bacterial, tuberculosis and viral meningitis
| Parameters | Bacterial meningitis (n:37) | Tuberculosis meningitis (n:30) | Viral meningitis (n:30) | p |
|---|---|---|---|---|
| Demographic profile | ||||
| Men, n (%) | 18(%48.6) | 17(%56.7) | 19(%63.3) | > 0.05 |
| Age*, years | 40.19 ± 2.73 | 42.60 ± 3.31 | 45.40 ± 2.56 | > 0.05 |
| Blood profile | ||||
| WBC† × 106/L | 9.42 (2.35–26.21) | 8.80 (3.61–17.82) | 7.91 (4.60–20.80) | > 0.05 |
| Neutrophil† (%) | 75.50 (46.0–91.0) | 76.0 (46.0 -90.0) | 75.0 (51.0–92.0) | > 0.05 |
| ESH† (mm/hour) | 35.0 (2–103) | 56.0 (2–120) | 29.0 (2–96) | > 0.05b; < 0.05a,c |
| CRP† (mg/L) | 55.0 (0.5–328.0) | 55.0 (2–250) | 24.0 (0.78–145.0) | > 0,05a; < 0,05b, c |
| CSF profile | ||||
| WBC†/mm3 | 450 (70–1100) | 250 (40–850) | 90 (20–450) | > 0,05a; < 0,05b, c |
| Lymphocyte percentage† | 40 (10–75) | 87.5 (20–90) | 90 (60–100) | < 0,05a, b; > 0,05c |
| Protein† (mg/dl) | 153 (26–1361) | 647 (34–1640) | 90 (34–215) | < 0,05 |
| Glucose ratio *(CSF/Blood) | 0.41 ± 0,06 | 0.30 ± 0.04 | 0.59 ± 0.03 | < 0,05a.c; > 0,05b |
| Clinical profile | ||||
| Systolic Blood Pressure† (mmHg) | 120 (90–171) | 120 (100–206) | 119 (90–168) | > 0.05 |
| Diastolic Blood Pressure†(mmHg) | 70 (50–96) | 70 (47–99) | 70 (60–86) | > 0.05 |
| Death, n (%) | 10 (%27) | 4 (% 10) | 1 (% 3.3) | < 0,05 |
| Sequelae, n (%) | 5 (%13.5) | 2 (% 6.7) | 1 (% 3.3) | < 0,05 |
| Serum HBP*(ng/ml) | 14,98 ± 1,1 | 6,89 ± 0,4 | 6,02 ± 0,4 | < 0,001a,b > 0,05c |
| CSF HBP*(ng/ml) | 7,81 ± 0,2 | 6,11 ± 0,3 | 5,75 ± 0,1 | < 0,001a,b > 0,05c |
†Median (interquartile range) *Mean ± SEMCRP: C-reactive protein, CSF: cerebrospinal fluid, WBC: white blood cell
aBM vs. TM; bBM vs. VM; cTM vs. VM
The demographic profile including age; the blood profile including the WBC count, percentage of neutrophils and clinical profile including systolic/diastolic blood pressure did not differ significantly among the patients with bacterial, tuberculosis and viral meningitis (p > 0.05). The patients with VM had lower blood CRP levels and CSF WBC count than those with bacterial and TM (p < 0.05). Glucose ratio (CSF/Blood) and erythrocyte sedimentation rate (ESR) values were found to be statistically significantly different in patients with TM from other case groups (p < 0.05). The CSF lymphocyte percentage was found lower in BM than in tuberculosis and VM groups (p < 0.05). CSF protein values were found to be statistically significantly different from each other in all 3 groups (p < 0.05).
The highest mortality and sequelae rates were found in the BM group (27% and 13.5%, respectively). CSF HBP and serum HBP levels were both significantly higher in patients with acute BM than in patients with tuberculosis and VM (p < 0.001). We have also shown that there was also a statistically significant positive correlation between serum HBP and CSF HBP values in the cases (r: 0.502; p:0,00).
Table 2 and Fig. 1 show the accuracy measures (at the cutoff point with the highest sensitivity and specificity) of some conventional CSF parameters and HBP as a new marker. For diagnosis of BM, the overall accuracy of CSF protein at a cutoff of 137 mg/dl, CSF glucose at a cutoff of 58.45 mg/dl and serum CRP at a cutoff of 170 mg/dl were: 43.7%, 48.5% and 55.2% respectively.
Table 2.
Accuracy measures with the cut off points of HBP and old markers in the prediction of bacterial meningitis
| Biomarker | Cutoff | Sensitivity | Specificity | Overall accuracy (area under the curve 95% C.I) | p value |
|---|---|---|---|---|---|
| Serum HBP | 9.03 ng/ml | 87.9% | 98.2% | 93.8% (88.5–99.1%) | 0.00 |
| CSF HBP | 6.99 ng/ml | 68.6% | 89.5% | 85.6% (77.8–93.3%) | 0.00 |
| CSF Protein | 137 mg/dl | 54.3% | 52.6% | 43.4% (31.3–55.5%) | 0.29 |
| CSF Glucose | 58.45 mg/dl | 45.7% | 61.4% | 48.5% (36.1–61.0%) | 0.816 |
| Serum CRP | 170 mg/l | 25.7% | 96.5% | 55.2% (42.3- 68.1%) | 0.403 |
Fig. 1.
Receiver operator characteristics curve of serum heparin binding protein, CRP and cerebrospinal fluid HBP, protein and glucose in the prediction of bacterial meningitis
A cutoff point of CSF HBP level at 6.99 ng/ml showed 68.6% sensitivity, 89.5% specificity; and an overall accuracy of 85.6%. At a cutoff point of 9.03 ng/ml serum HBP also showed 87.9% sensitivity, 98.2% specificity and overall accuracy of 93.8%. The area under the curve for both CSF and serum HBP was found to be higher than the other parameters investigated (Fig. 1 and Table. 2).
Discussion
Bacterial meningitis continues to be a major health problem as a cause of widespread morbidity and mortality. Early diagnosis and proper treatment of BM are crucial for a successful outcome. Therefore, clinical and laboratory tools are needed to distinguish between bacterial and non-bacterial meningitis. Kandil et al. previously investigated serum and CSF HBP levels in the differentiation of BM from VM in early diagnosis [1]. However, tuberculous meningitis cases have not been investigated. This study is the first study investigating the usefulness of serum and CSF HBP levels as a diagnostic factor in the early diagnosis of BM and in differentiating it from tuberculosis and VM.
We found that the level of HBP in the blood and CSF at admission was higher in patients with BM than in those with viral and TM (p < 0.001). These results suggest that serum and CSF HBP is a useful markers for the diagnosis prediction of BM at the initial stage.
We found that the CSF protein levels were significantly different between patients with tuberculosis, viral and patients with BM (p < 0.05). This finding is consistent with those of previous studies [13]. In addition, the blood CRP and the CSF WBC count were found significantly lower in a viral group than in tuberculosis and BM (p < 0.05).
Cerebrospinal fluid bacterial culture is the gold standard for confirmation of acute bacterial meningitis, but many cases cannot confirmed with culture. In this study of the 37 patients with BM, 11 had a positive CSF culture, smear, or PCR for bacterial pathogens, comprising 2 with Streptococcus species, 7 with Staphylococcus species, and 2 with other species. Only 5 of the 30 patients with TM had a positive CSF culture, smear, or PCR for Mycobacterium tuberculosis, and they were classified as definite TM.
The mortality rate in acute bacterial meningitis varies between 5 and 40% depending on the type, age, underlying presence and the time of starting treatment. In survivors, the rate of sequelae can reach up to 30% [14, 15]. In this study, the highest rate of death and sequelae were found in the group with BM (respectively 27% and 13.5%). In tuberculosis and VM group, mortality and sequelae were found to be lower (10%; 6.7% and 1%; 1%, respectively).
It has been reported that HBP has a typical alarm characteristic released from PMN during tissue injury and infection. Due to the immune stimulating properties of HBP, recombinant forms may be useful in the treatment of bacterial infections [6]. HBP has an important role in the pathophysiology of severe bacterial infections and therefore is considered to be a potential diagnostic marker and a target for treatment [16]. It has been suggested that HBP may be a marker for early detection of those at risk of developing severe sepsis and septic shock in emergency department patients with a bacterial infection. In the studies; HBP levels in the blood of patients with severe sepsis and septic shock were significantly higher than HBP levels in critically ill non-infectious patients. Thus, there are many studies demonstrating the role of HBP as a serum biomarker in bacterial infections [9, 17–19].
Previous studies found that HBP level was higher in BM than in viral meningitis [1, 3]. However, no previous study has investigated differences in the serum and CSF HBP levels between patients with tuberculosis and viral meningitis.
Linder et al. concluded that CSF HBP concentrations were significantly higher (p < 0.01) in patients with ABM than in patients with VM (41 BM patients and 10 VM patients) and control patients with normal CSF cell count [3].
Kandil et al. found that CSF HBP levels averaged 0.82 ± 0.3 ng/mL in controls, 3.3 ± 1.7 ng/mL in viral and 174.8 ± 46.7 ng/mL in BM [1]. In addition, have been detected mean serum HBP level was 0.84 ± 0.3 ng/mL in the controls, 3.7 ± 1.9 ng/mL in viral, and 192.2 ± 56.6 ng/mL in BM. As a result, they stated that both HBP levels in patients with BM were significantly higher than in VM and control groups.
In this present study was detected that serum HBP correlates also between CSF HBP in cases of meningitis. This finding agrees with those of previous studies [1].
We found that the area under the ROC curve for serum and CSF HBP were 0.938 and 0.856 respectively, which were higher than that for the other investigated parameters. According to these results, serum HBP has higher sensitivity and specificity than BOS HBP. In this case, early diagnosis can be made with serum HBP levels without LP. Therefore, assisting in the early diagnosis of bacterial meningitis with a less invasive sample (as blood serum rather than CSF) would be of great benefit not only to patients but also to physicians. We think that serum HBP, together with other markers such as lactate and procalcitonin, may be an auxiliary diagnostic marker in the detection of BM.
Study limitations included small sample size and the use of a single medical center. There is a need for further studies that include covering a large number of patients,
Author’s Contribution
ESN, MN, IK—research concept and design; ESN, MN, EE—collection and/or assembly of data; ESN, MN, EE—data analysis and interpretation; ESN, MN—writing the article; ESN, MN, IK—critical revision of the article; ESN, MN—final approval of article.
Funding
This study was financially supported by the Scientific Research Projects Governing Unit of Gaziantep University (Protocol number: SBF.17.03).
Declarations
Conflict of interest
The authors declare no conflicts of interest.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Kandil M, Khalil G, El-Attar E, Shehata G, Hassan S. Accuracy of heparin binding protein: as a new marker in prediction of acute bacterial meningitis. Braz J Microbiol. 2018;49(1):213–219. doi: 10.1016/j.bjm.2018.05.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Scheld WM, Whitley RJ, Marra MC. Infections of the central nervous system. 4. Philadelphia: Wolters Kluwer Health, United States; 2014. [Google Scholar]
- 3.Linder A, Åkesson P, Brink M. Heparin-binding protein: diagnostic marker of acute bacterial meningitis. Crit Care Med. 2011;39:812–817. doi: 10.1097/CCM.0b013e318206c396. [DOI] [PubMed] [Google Scholar]
- 4.Carbonnelle E. Laboratory diagnosis of bacterial meningitis: usefulness of various tests for the determination of the etiological agent. Med Mal Infect. 2009;39(7–8):581–605. doi: 10.1016/j.medmal.2009.02.017. [DOI] [PubMed] [Google Scholar]
- 5.Bartt R. Acute bacterial and viral meningitis. Continuum Minneap Minn. 2012;18(6):1255–1270. doi: 10.1212/01.CON.0000423846.40147.4f. [DOI] [PubMed] [Google Scholar]
- 6.Soehnlein O, Lindbom L. Neutrophil-derived azurocidin alarms the immune system. J Leukoc Biol. 2009;85(3):344–351. doi: 10.1189/jlb.0808495. [DOI] [PubMed] [Google Scholar]
- 7.Yang Y, Liu G, He Q, Shen J, Xu L, Zhu P, et al. A promising candidate: heparin-binding protein steps onto the stage of sepsis prediction. J Immunol Res. 2019;16:1–11. doi: 10.1155/2019/7515346. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Linder A, Soehnlein O, Åkesson P. Roles of heparin-binding protein in bacterial infections. J Innate Immunol. 2010;2:431–438. doi: 10.1159/000314853. [DOI] [PubMed] [Google Scholar]
- 9.Linder A, Arnold R, Boyd JH, Zindovic M, Zindovic I, Lange A, et al. Heparin-binding protein measurement improves the prediction of severe infection with organ dysfunction in the emergency department. Crit Care Med. 2015;43(11):2378–86. doi: 10.1097/CCM.0000000000001265. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Kjölvmark C, Påhlman LI, Åkesson P, Linder A. Heparin-binding protein (HBP)—a diagnostic biomarker of urinary tract infection in adults. Open Forum Infect Dis. 2014;23:4–10. doi: 10.1093/ofid/ofu004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Llewelyn MJ, Berger M, Gregory M, Ramaiah R, Taylor AL, Curdt I. Sepsis biomarkers in unselected patients on admission to intensive or high-dependency care. Crit Care. 2013;17:60–65. doi: 10.1186/cc12588. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Mariscalco MM. Heparin-binding protein: another neutrophil granule protein. another new biomarker? Crit Care Med. 2011;39(4):910–911. doi: 10.1097/CCM.0b013e31820e6a43. [DOI] [PubMed] [Google Scholar]
- 13.Kim J, Kim SE, Park BS, Shin KJ, Ha SY, Park J, et al. Procalcitonin as diagnostic and prognostic factor for tuberculosis meningitis. J Clin Neurol. 2016;12(3):332–339. doi: 10.3988/jcn.2016.12.3.332. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Ozen M, Kanra G, Kara A, Secmeer G, Cengiz AB. Long-term beneficial effects of dexamethasone on intellectual and neuropsychological outcome of children with pneumococcal meningitis. Scand J Infec Dis. 2006;38(2):104–109. doi: 10.1080/00365540500276005. [DOI] [PubMed] [Google Scholar]
- 15.Ulutan F. Acute bacterial meningitis. Turkiye Klinikleri J Int Med Sci. 2006;2(28):69–74. [Google Scholar]
- 16.Fisher J, Linder A. Heparin-binding protein: a key player in the pathophysiology of organ dysfunction in sepsis. J Int Med. 2017;281(6):562–574. doi: 10.1111/joim.12604. [DOI] [PubMed] [Google Scholar]
- 17.Linder A, Åkesson P, Inghamma M, Treutiger CJ, Linnér A, Sundén-Cullberg J. Elevated plasma levels of heparin-binding protein in intensive care unit patients with severe sepsis and septic shock. Crit Care. 2012;16:90–96. doi: 10.1186/cc11353. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Holub M, Beran M. Should heparin-binding protein levels be routinely monitored in patients with severe sepsis and septic shock? Crit Care. 2012;16:133–138. doi: 10.1186/cc11379. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Bentzer P, Fisher J, Kong HJ, Mörgelin M, Boyd JH, Walley KR, et al. Heparin-binding protein is important for vascular leak in sepsis. Intensive Care Med Exp. 2016;4(1):33–49. doi: 10.1186/s40635-016-0104-3. [DOI] [PMC free article] [PubMed] [Google Scholar]