Characteristics of Serum Endocan Levels in Infection
Characteristics of Serum Endocan Levels in Infection
Kazunori Seo 0
Takatoshi Kitazawa 0
Yusuke Yoshino 0
Ichiro Koga 0
Yasuo Ota 0
0 Department of Internal Medicine, Teikyo University , Tokyo , Japan
A total of 78 patients were enrolled in this study. The median endocan level decreased by
only 23% during infection, whereas both serum CRP and PCT levels decreased by more
than 80%. Endocan levels were correlated to neither CRP levels nor PCT levels in each
period. Endocan levels at day 0 in patients with bacteremia were higher than those without
bacteremia (1.09 ng/mL vs 0.82 ng/mL, P=0.002), but neither CRP levels nor PCT levels at
day 0 were different between the two groups. Areas under the receiver operator
characteristic (ROC) curves of endocan, CRP, and PCT at day 0 were 0.662, 0.343, and 0.563,
respectively. Positive blood cultures tended to be related to high endocan levels, but not
significantly (odds ratio: 4.24, 95% CI: 0.99-10.34, P=0.05).
In bacteremic cases, serum endocan levels in bacteremia tended to be higher than in
nonbacteremic cases. Although endocan level was not identified as a prognostic factor of
bacteremia, further prospective study concerning the relationship between serum endocan
level and bacteremia would be needed.
Bacteremia is potentially life-threatening and thus requires early diagnosis and prompt
administration of antibiotics to reduce mortality related to multiple organ failure. During the course
of bacteremia, the time at which adequate antibiotic treatment is started is a key factor affecting
survival. However, bacteremia is identified in only about 30% of patients with sepsis .
C-reactive protein (CRP) and procalcitonin (PCT) are currently used as clinical indicators
of inflammation and infection including bacteremia, and several other biochemical markers
have been investigated for their ability to detect sepsis in an early, reversible phase .
Nonetheless, identification of an ideal biomarker capable of making a clear distinction between
sepsis and SIRS is imperative .
Endocan is a proteoglycan that can be detected in human blood and is expressed on the
surface of endothelial cells of lungs and kidneys . Endocan is a marker of angiogenesis
previously studied in various types of cancer [6,7,8,9,10,11,12]. Synthesis and release of this molecule
are promoted by proinflammatory cytokines such as tumor necrosis factor (TNF)- and
interleukin (IL)-1 [5,13,14]. A few studies have shown that endocan can be acknowledged as a
good marker of endothelial dysfunction and multiorgan failure in sepsis, and it can be accepted
as a good marker of survival prognosis in sepsis . In clinical practice, however, the
usefulness and characteristics of endocan in infection have not been investigated. Changes in serum
endocan levels through infection have not been analyzed.
In this study, the time course of serum endocan levels during infection and the associations
between serum endocan levels and clinical factors, including blood culture-positivity,
This prospective study was conducted in the Department of Internal Medicine of Teikyo
University Hospital between June 2013 and March 2014. The subjects were patients aged 20
years admitted due to infection. Infection was defined as (1) systemic manifestations of
infection (e.g., fever, chills and/or hypotension) with positive results of microbiological or
serological tests, or (2) focal manifestations of infection with positive microbiological or radiological
tests. The onset of infection was defined as the time of admission. Patients with pre-existing
malignancy were excluded to avoid measurement inconsistencies. This study was approved by
the Teikyo University ethics committee (the approval number:13032). Written informed
consent was obtained from all participants in this study.
Biomarkers were measured during the four following periods: day 0, on the admission day; day
12, between 1 and 2 days after onset of infection; day 35, between 3 and 5 days after onset of
infection; and day 610, between 6 and 10 days after onset of infection. Within one hour,
samples were centrifuged at 3000 rpm for 10 minutes, and sera were kept at -20C until assayed. To
measure serum endocan and PCT concentrations, enzyme-linked immunosorbent assays were
used (LUNGINNOV Systems, Lille, France; CUSABIO, Wuhan, P.R. China). CRP
concentrations were measured by latex agglutination assays (LSI Medience Corporation, Tokyo, Japan).
In order to collect demographic information and clinical information, medical records
Relationships between categorical variables were analyzed using Pearsons chi-square test.
Fishers exact test was used when the expected count was less than 5. Continuous variables
were compared using the Mann-Whitney U test because the data did not follow a normal
distribution. Kendalls rank correlation coefficients were determined in order to evaluate the
possible relationships among biomarkers. Logistic regression analysis was carried out to estimate
odds ratios (ORs) and 95% confidence intervals (CIs). All p values were two-sided, and
p < 0.05 was considered significant. All analyses were performed using PASW software for
A total of 78 patients were enrolled in this study. The patients demographic data are shown in
Table 1. Of these 78 patients, 40 were male (51%) and 38 were female (49%). The median age
of the patients was 79 years (range, 2397 years). As for pre-existing conditions, diabetes was
the most common comorbidity (26% of patients). Primary infection sites were identified in 70
Total (n = 78)
Numbers and percentages of each case are displayed. Abbreviations: IQR, interquartile range.
cases. The most frequent site was the respiratory tract (45%), and the second most frequent site
was the urinary tract (21%). In all patients, the Pitt Bacteremia score was 4. Pathogens were
detected in 47 cases. Only one patient died of meningitis with Streptococcus pneumoniae. There
were 52 patients who had sepsis and 26 patients who did not have sepsis according to the
consensus sepsis definition.
Changes in biomarker levels during infection and correlations with each
Changes in in biomarkers during infection were examined, and the serum levels of endocan,
CRP, and PCT are shown in Fig 1. During the infection periods, a significant reduction of
endocan levels was observed from day 12 to day 35 (median 1.03 ng/mL to 0.81 ng/mL,
p<0.001). The reduction in the median endocan level during the whole infection period was
22.5%. On the other hand, significant reductions of both CRP and PCT levels were observed in
two periods, from day 12 to day 35, and from day 35 to day 610. The reduction rates of
the median CRP and PCT levels during the whole infection period were 82.0% and almost
Endocan levels at day 0 were not correlated with CRP at day 0 (r2 = -0.042; p = 0.59), or
with PCT at day 0 (r2 = 0.119; p = 0.16) (Fig 2). In the other periods, endocan levels were
correlated with neither CRP nor PCT (data not shown). These results indicate that the reduction
rate in endocan levels during infection was smaller than that of CRP and PCT, and endocan
was a biomarker that was independent of the CRP and PCT levels during infection.
Association of biomarkers with blood culture positivity
Blood cultures were examined in all 78 patients. Blood cultures were positive in 29 cases and
negative in 49 cases. Changes in endocan, CRP, and PCT levels in the positive blood culture
groups and negative blood culture groups are shown in Fig 3. Endocan levels at day 0 and at
day 12 were significantly higher in the positive blood culture group than in the negative
culture group (median 1.09 ng/mL vs 0.82 ng/mL, p = 0.003; 1.73 ng/mL vs 0.92 ng/mL, p = 0.02,
respectively). CRP did not differ between the two groups during infection. PCT levels at day
35 were significantly higher in the positive blood culture group than in the negative culture
group (median 0.02 ng/mL vs 0.00 ng/mL, p = 0.03), although PCT levels in other periods were
not significantly different between the two groups. Using biomarkers at day 0, receiver operator
characteristic (ROC) analysis showed that areas under the ROC curve (AUC) of endocan,
CRP, and PCT at day 0 were 0.662, 0343, and 0.563, respectively (P = 0.02, 0.02, 0.35,
respectively) (Fig 4). The serum endocan level at day 0 had a specificity of 85.7%, a sensitivity of
41.3%, positive predictive value of 63.2%, and negative predictive value of 76.3% at the cutoff
level of 1.70 ng/mL. On the other hand, the serum PCT level at day 0 had a specificity of 83.7%,
a sensitivity of 27.6%, positive predictive value of 50.0%, and negative predictive value of 66.1%
at the cutoff level of 0.12 pg/mL. Using PCT and endocan, the sensitivity to blood culture
positivity was 55.2%.
Comparisons of clinical factors between patients with high endocan
levels and patients with low endocan levels at the onset of infection
According to their serum endocan levels at day 0 with the cutoff level of 1.70 ng/mL, all
patients were divided into two groups. As shown in Table 2, there were no significant differences
in age, sex, and comorbidities. However, with respect to the primary infection site, the rate of
catheter-related blood stream infections was higher in the high endocan group than in the low
Fig 1. Trends of endocan (A), C-reactive protein (CRP)(B), procalcitonin (PCT)(C) levels in patients
during infection. The boxes represent the interquartile range, and the transverse lines in the boxes
represent the medians of each group. The vertical lines represent 1.5 of the interquartile range. The circles
indicate outliers of each group. The y axes for endocan in (A) and for PCT in (C), but not for CRP in (B) are in
log values.* P<0.001.
Fig 2. Correlations between the results for C-reactive protein (CRP) and endocan (A), and those for
procalcitonin (PCT) and endocan (B). The black and white circles represent the levels of the markers in the
blood culture-positive (n = 29) and-negative (n = 49) groups, respectively. The y axes for endocan in (A) and
(B), and the x axis for PCT in (B) are in log values.
endocan group (21% vs 3%, P = 0.03). The rate of positive blood cultures was also higher in the
high endocan group than in the low endocan group (63% vs 29%, P = 0.01). However, there
were no significant differences on multivariate logistic regression analyses, although positive
Fig 3. Blood culture positivity and biomarkers during infection. The boxes represent the interquartile
ranges, and the transverse lines in the boxes represent the medians of each group. The vertical lines
represent 1.5 of the interquartile range. The circles indicate outliers of each group. The y axes for endocan in
(A) and for PCT in (C), but not for CRP in (B) are in log values.* P = 0.003, # P = 0.02, P = 0.03
Fig 4. Receiver operating characteristic (ROC) curves for predicting the prognosis of blood culture
positivity based on biomarkers levels at day 0. Abbreviations: CRP, C-reactive protein; PCT, procalcitonin
blood culture tended to be related to high endocan levels (odds ratio: 4.24, 95% CI: 0.9910.34,
P = 0.05) (Table 3).
Widely used infection biomarkers, such as CRP or PCT, have been investigated for their
usefulness for distinguishing sepsis from other inflammatory diseases, assessing severity of illness,
monitoring effectiveness of treatment, and predicting outcome in many previous studies.
Endocan, a newly recognized biomarker, has been investigated only with respect to its
relationship with sepsis. Changes in endocan levels during infection and the associations of endocan
levels with clinical factors have not been previously investigated.
The first characteristic of endocan identified in the present study was that changes in serum
levels during infection were smaller than those of CRP or PCT. PCT and CRP have been
recognized as biomarkers with wide ranges in the short course of infection. There is increasing
evidence that a PCT-based algorithm is a useful tool to reduce antibiotic duration . The
reduction in endocan levels during infection in this study was 22.5%. The serum endocan level
has been shown to be related to progression of various cancers, although serum endocan levels
were continuously high in these chronic diseases [7,17,18]. The serum endocan level has also
been shown to be related to acute diseases other than sepsis, including acute lung injury and
acute respiratory distress [19,20], but changes in its serum levels in the acute phase of these
diseases have not been investigated. In an in vitro study, endocan was continuously produced in
the presence of TNF- in primary cultured HUVECs from 6 to 72 hours, but production over
a longer period was not investigated . The process of regulating endocan in vivo has not
been clarified. From the present results of the dynamics of serum endocan levels during
infection, we speculated that the serum endocan level might not only reflect the degree of
inflammation, but also the healing process of vascular damage in the late phase of infection.
Serum endocan level at onset
< 1.70 ng/mL (n = 59)
1.70 ng/mL (n = 19)
Numbers and percentages of each case are displayed. Abbreviations: IQR, interquartile range.
Biomarker levels with a wide range can be more affected by the timing of sampling or
measurement errors. In the Survival Sepsis Campaign Guideline, no recommendation was given
for the use of these markers to distinguish between severe infection and other acute
inflammatory states . Serum endocan levels might be inappropriate to monitor the effectiveness of
treatment, but the mortality in the present study was 1.3%. All the patients in the present study
had low Pitt Bacteremia Scores, which means that severe septic cases were not included. In
Abbreviations: OR, odds ratio; CI, confidence interval.
order to investigate whether monitoring serum endocan levels would be useful in assessing the
clinical course both in severe cases and in mild cases, more cases of various severities would
The second characteristic of endocan was that its serum level was not correlated with that of
CRP or PCT. In the present study, there were few cases in which both serum levels of endocan
and CRP were high. In a previous study of sepsis, endocan was not correlated to CRP and PCT
, which is compatible with the present results. In clinical settings, some cases of infection
can be misdiagnosed because of a low CRP level. The primary production site of CRP is the
liver in response to IL-6 , and that of PCT is the liver and leukocytes . Endocan was
expressed mainly in the vascular endothelial cells, which is different from PCT or CRP. Several
cases of catheter-related bloodstream infection showed high endocan levels in the present
study. From these results, when the endocan level is high in patients, even if CRP and PCT
levels are not high at the onset of the infection, the presence of bacteremia might be predicted.
The third characteristic of endocan was that its serum level was related to bacteremia. For
the prediction of bacteremia, PCT has been the most promising parameter. PCT had negative
predictive values of 99% and 96% when the PCT cut-off values were 0.5 ng/mL and 0.5 ng/mL,
respectively . In a prospective cohort study of SIRS patients, ROC analysis revealed an
AUC value of 0.675 for PCT (95% CI 0.6360.714) for differentiating patients with blood
stream infections from those without . In the present study, the PCT level at day 0 was
not related to bacteremia. In the ROC analysis of the present study, the AUC value of serum
endocan at the onset of the infection was higher than that of PCT for differentiating patients
with blood stream infections from those without. When using endocan and PCT for predicting
positivity of blood culture, the sensitivity was higher than that of both the biomarkers. There
has been no study about usefulness of endocan in bacteremia, so the endocan cutoff value of
1.7ng/ml was determined retrospectively in this study. In order to confirm the cutoff value of
endocan, a subsequent prospective study would be necessary.
As previously mentioned, the specific production sites of endocan are endothelial cells. The
expression of endocan is up-regulated in response to TNF- or IL1- in vitro, although the
precise mechanism of endocan expression in infection has not been elucidated [14,26]. LPS also
induces endocan release in endothelial cells [15,27]. Inflammatory cytokines were released
independent of infection sites. From these results, it appears that, in patients with bacteremia,
direct stimulation with pathogenic molecules or inflammatory cytokines in endothelial cells
might induce high serum endocan levels in the blood.
Patients with bacteremia were occasionally found to have no focal or systemic
manifestations of infection . In previous studies, it has been emphasized that serum endocan levels
are related to severity in sepsis. The present results suggest that with measurement of endocan
levels at the onset of infections other than sepsis it might be possible to predict bacteremia.
In clinical practice, including cancer management, endocan has not been widely used. It
took about three hours to measure the serum endocan level by ELISA assay in the present
study. By shortening the measurement time and reducing the cost, endocan might become
widely used and recognized as superior to PCT as a screening tool for bacteremia.
There were several limitations in this study. This prospective study was conducted for the
purpose of investigating the characteristics of endocan during infection, so that the numbers of
enrolled cases may be too small for comparisons of primary infection sites or pathogens. The
patients in the present study were admitted to the general ward in our hospital, and the severity
of illness was mild or moderate in many cases. Since biomarkers levels including endocan
before infection were not measured, endocan levels were not determined at the acute period of
infection. It would be necessary to investigate the relationship between each biomarker levels,
culture results, and clinical manifestations in hospital-acquired infection. Further study would
be needed to assess the whole time course of serum endocan levels and the relationship
between endocan levels and bacteremia in severe infection cases.
In conclusion, the serum endocan level was elevated at the onset of infection, and it then
decreased slightly during infection. The endocan level was not correlated with the CRP level or
the PCT level. In bacteremic cases, serum endocan levels in bacteremia tended to be higher
than in non-bacteremic cases. Although endocan level was not identified as a prognostic factor
of bacteremia, further prospective study concerning the relationship between serum endocan
level and bacteremia would be needed.
This work was supported by JSPS KAKENHI Grant Number 25462833.
Conceived and designed the experiments: TK YY IK YO. Performed the experiments: KS.
Analyzed the data: KS TK. Contributed reagents/materials/analysis tools: KS TK. Wrote the paper:
1. Mathevon T , Souweine B , Traore O , Aublet B , Caillaud D ( 2002 ) ICU-acquired nosocomial infection: impact of delay of adequate antibiotic treatment . Scand J Infect Dis 34 : 831 - 835 . PMID: 12578154
2. Bates DW , Sands K , Miller E , Lanken PN , Hibberd PL , Graman PS , et al. ( 1997 ) Predicting bacteremia in patients with sepsis syndrome . Academic Medical Center Consortium Sepsis Project Working Group. J Infect Dis 176 : 1538 - 1551 . PMID: 9395366
3. Luzzani A , Polati E , Dorizzi R , Rungatscher A , Pavan R , Merlini A ( 2003 ) Comparison of procalcitonin and C-reactive protein as markers of sepsis . Crit Care Med 31 : 1737 - 1741 . PMID: 12794413
4. Dellinger RP , Levy MM , Rhodes A , Annane D , Gerlach H , Opal SM , et al. Surviving sepsis campaign: international guidelines for management of severe sepsis and septic shock: 2012 . Crit Care Med 41 : 580 - 637 . doi: 10.1097/CCM.0b013e31827e83af PMID: 23353941
5. Bechard D , Gentina T , Delehedde M , Scherpereel A , Lyon M , Aumercier M , et al. ( 2001 ) Endocan is a novel chondroitin sulfate/dermatan sulfate proteoglycan that promotes hepatocyte growth factor/scatter factor mitogenic activity . J Biol Chem 276 : 48341 - 48349 . PMID: 11590178
6. Maurage CA , Adam E , Mineo JF , Sarrazin S , Debunne M , Siminski RM , et al. ( 2009 ) Endocan expression and localization in human glioblastomas . J Neuropathol Exp Neurol 68 : 633 - 641 . doi: 10.1097/ NEN.0b013e3181a52a7f PMID: 19458546
7. Grigoriu BD , Depontieu F , Scherpereel A , Gourcerol D , Devos P , Ouatas T , et al. ( 2006 ) Endocan expression and relationship with survival in human non-small cell lung cancer . Clin Cancer Res 12 : 4575 - 4582 . PMID: 16899604
8. Chen LY , Liu X , Wang SL , Qin CY Over-expression of the Endocan gene in endothelial cells from hepatocellular carcinoma is associated with angiogenesis and tumour invasion . J Int Med Res 38 : 498 - 510 . PMID: 20515564
9. Kim JH , Park MY , Kim CN , Kim KH , Kang HB , Kim KD , et al. Expression of endothelial cell-specific molecule-1 regulated by hypoxia inducible factor-1alpha in human colon carcinoma: impact of ESM-1 on prognosis and its correlation with clinicopathological features . Oncol Rep 28 : 1701 - 1708 . doi: 10. 3892/or. 2012 . 2012 PMID: 22948784
10. Leroy X , Aubert S , Zini L , Franquet H , Kervoaze G , Villers A , et al. Vascular endocan (ESM-1) is markedly overexpressed in clear cell renal cell carcinoma . Histopathology 56 : 180 - 187 . doi: 10.1111/j. 1365- 2559 . 2009 .03458.x PMID: 20102396
11. Roudnicky F , Poyet C , Wild P , Krampitz S , Negrini F , Huggenberger R , et al. Endocan is upregulated on tumor vessels in invasive bladder cancer where it mediates VEGF-A-induced angiogenesis . Cancer Res 73 : 1097 - 1106 . doi: 10.1158/ 0008 - 5472 . CAN-12-1855 PMID: 23243026
12. El Behery MM , Seksaka MA , Ibrahiem MA , Saleh HS , El Alfy Y Clinicopathological correlation of endocan expression and survival in epithelial ovarian cancer . Arch Gynecol Obstet 288 : 1371 - 1376 . doi: 10.1007/s00404- 013 - 2863 - 3 PMID: 23708323
13. Bechard D , Meignin V , Scherpereel A , Oudin S , Kervoaze G , Bertheau P , et al. ( 2000 ) Characterization of the secreted form of endothelial-cell-specific molecule 1 by specific monoclonal antibodies . J Vasc Res 37 : 417 - 425 . PMID: 11025405
14. Lassalle P , Molet S , Janin A , Heyden JV , Tavernier J , Fiers W , et al. ( 1996 ) ESM-1 is a novel human endothelial cell-specific molecule expressed in lung and regulated by cytokines . J Biol Chem 271 : 20458 - 20464 . PMID: 8702785
15. Scherpereel A , Depontieu F , Grigoriu B , Cavestri B , Tsicopoulos A , Gentina T , et al. ( 2006 ) Endocan, a new endothelial marker in human sepsis . Crit Care Med 34 : 532 - 537 . PMID: 16424738
16. Hohn A , Schroeder S , Gehrt A , Bernhardt K , Bein B , Wegscheider K , et al. Procalcitonin-guided algorithm to reduce length of antibiotic therapy in patients with severe sepsis and septic shock . BMC Infect Dis 13 : 158. doi: 10.1186/ 1471 - 2334 - 13 -158 PMID: 23547790
17. Hatfield KJ , Lassalle P , Leiva RA , Lindas R , Wendelboe O , Bruserud O Serum levels of endotheliumderived endocan are increased in patients with untreated acute myeloid leukemia . Hematology 16 : 351 - 356 . doi: 10.1179/102453311X13127324303434 PMID: 22183069
18. Voiosu T , Balanescu P , Bengus A , Voiosu A , Baicus CR , Barbu M , et al. Serum endocan levels are increased in patients with inflammatory bowel disease . Clin Lab 60 : 505 - 510 . PMID: 24697130
19. Mikkelsen ME , Shah CV , Scherpereel A , Lanken PN , Lassalle P , Bellamy SL , et al. Lower serum endocan levels are associated with the development of acute lung injury after major trauma . J Crit Care 27 : 522 e511- 527 .
20. Tang L , Zhao Y , Wang D , Deng W , Li C , Li Q , et al. Endocan levels in peripheral blood predict outcomes of acute respiratory distress syndrome . Mediators Inflamm 2014 : 625180. doi: 10.1155/2014/625180 PMID: 25132734
21. Su YH , Shu KH , Hu CP , Cheng CH , Wu MJ , Yu TM , et al. Serum Endocan correlated with stage of chronic kidney disease and deterioration in renal transplant recipients . Transplant Proc 46 : 323 - 327 . doi: 10.1016/j.transproceed. 2013 . 10.057 PMID: 24655954
22. Hurlimann J , Thorbecke GJ , Hochwald GM ( 1966 ) The liver as the site of C-reactive protein formation . J Exp Med 123 : 365 - 378 . PMID: 4379352
23. Meisner M ( 2002 ) Pathobiochemistry and clinical use of procalcitonin . Clin Chim Acta 323 : 17 - 29 . PMID: 12135804
24. Liaudat S , Dayer E , Praz G , Bille J , Troillet N ( 2001 ) Usefulness of procalcitonin serum level for the diagnosis of bacteremia . Eur J Clin Microbiol Infect Dis 20 : 524 - 527 . PMID: 11681430
25. Hoenigl M , Raggam RB , Wagner J , Prueller F , Grisold AJ , Leitner E , et al. Procalcitonin fails to predict bacteremia in SIRS patients: a cohort study . Int J Clin Pract.
26. Sarrazin S , Adam E , Lyon M , Depontieu F , Motte V , Landolfi C , et al. ( 2006 ) Endocan or endothelial cell specific molecule-1 (ESM-1): a potential novel endothelial cell marker and a new target for cancer therapy . Biochim Biophys Acta 1765 : 25 - 37 . PMID: 16168566
27. Lee W , Ku SK , Kim SW , Bae JS Endocan elicits severe vascular inflammatory responses in vitro and in vivo . J Cell Physiol 229 : 620 - 630 . doi: 10.1002/jcp.24485 PMID: 24446198
28. Trinchet JC , Hornstein M , Jupeau A , Beaugrand M , Hecht Y , Ferrier JP ( 1983 ) [Do latent bacteremia and endotoxinemia exist in cirrhotics? A study of 48 patients] . Gastroenterol Clin Biol 7 : 518 - 522 . PMID: 6873564