- Email: [email protected]
YKL-40 is elevated in patients with peripheral arterial disease and diabetes or pre-diabetes
Atherosclerosis 222 (2012) 557–563
Contents lists available at SciVerse ScienceDirect
Atherosclerosis journal homepage: www.elsevier.com/locate/atherosclerosis
YKL-40 is elevated in patients with peripheral arterial disease and diabetes or pre-diabetes Klaudija Batinic a,b , Clemens Höbaus b , Milan Grujicic b , Angelika Steffan b , Finka Jelic b , David Lorant b,c , Thomas Hörtenhuber b,d , Florian Hoellerl b,e , Johanna-Maria Brix e , Guntram Schernthaner e , Renate Koppensteiner b , Gerit-Holger Schernthaner b,∗ a
Division of Pediatric Cardiology, University Zurich, Switzerland Medical University of Vienna, Department of Medicine II, Vienna, Austria c Medical University of Vienna, Department of Anesthesiology, Vienna, Austria d Medical University of Vienna, Department of Pediatrics, Vienna, Austria e Rudolfstiftung Hospital, Department of Medicine I, Vienna, Austria b
a r t i c l e
i n f o
Article history: Received 25 November 2011 Received in revised form 21 March 2012 Accepted 28 March 2012 Available online 4 April 2012 Keywords: : YKL-40 Peripheral artery disease Type 2 diabetes mellitus
a b s t r a c t Objective: YKL-40 is secreted by macrophages in atherosclerotic lesions and involved in plaque rupture. YKL-40 is elevated in coronary artery disease, and predicts cardiovascular mortality. Experimental in vivo and in vitro data suggest a role of YKL-40 in tissue remodeling. A disease modulating potency of YKL-40 was not investigated in peripheral arterial disease (PAD). Methods: We measured YKL-40 in 460 subjects: 316 PAD: 71 normal glucose metabolism (PAD-NGM), 90 pre-diabetes (PAD-PREDM) and 155 diabetes (PAD-DM); 20 diabetes with atherosclerosis but without PAD (AS-DM); 85 diabetes without macro-vascular complications (DM) and 39 healthy controls (CO). Results: YKL-40 is higher in PAD vs. CO (median [25–75 percentile]: 103 [69–159] vs. 43 [30–80] ng/ml; p < 0.001). In addition, YKL-40 is elevated in DM (p < 0.001), PAD-NGM (p = 0.001), PAD-PREDM (p < 0.001), PAD-DM (p < 0.001) and AS-DM (p = 0.002) compared to CO. Among PAD, YKL-40 is increased in PADPREDM (p = 0.001) and PAD-DM (p = 0.01) vs. PAD-NGM. By multivariate regression YKL-40 is significantly associated with age (beta = 0.272), triglycerides (beta = 0.216), aspartate-amino-transferase (beta = 0.177) and c-reactive-protein (beta = 0.178). Underpinning its role YKL-40 was found to be associated with micro-/macroalbuminuria (p = 0.014/p = 008) – a strong remodeling inducer. In addition, YKL-40 was elevated in existence of mediasclerosis (p = 0.008), a remodeling process. Conclusion: We are first to show that YKL-40 is higher in subjects with peripheral arterial disease. YKL-40 was higher in PAD patients with pre-/diabetes. In addition, YKL-40 was associated with the “severity” of generalized atherosclerosis estimated by affected vascular beds. All our findings point towards a role of YKL-40 in the progression/prognosis of patients with PAD and concomitant diabetes. © 2012 Elsevier Ireland Ltd. All rights reserved.
1. Introduction YKL-40, a chitin binding protein named after its three Nterminal amino acids tyrosine (Y), lysine (K), leucine (L) and its molecular mass of 40 kDa was first identified in 1985 under tissue remodeling and subsequently inflammatory conditions [1,2]. YKL40 – described to be involved in acute and chronic inflammatory reactions – is secreted in large amounts by vascular smooth muscle cells (VSMCs) [2], synovial cells, articular chondrocytes, activated neutrophils [3], cancer cells [4], and in fibrogenesis process [5].
∗ Corresponding author at: Medical University & General Hospital of Vienna, Department of Medicine II, Waehringer Guertel 18-20, A-1090 Vienna, Austria. Tel.: +43 1 40400 4671; fax: +43 1 40400 4665. E-mail address: [email protected] (G.-H. Schernthaner). 0021-9150/$ – see front matter © 2012 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.atherosclerosis.2012.03.034
Macrophages during late stage of differentiation are a major source of YKL-40 [6]. A subpopulation of activated macrophages shows a strong induction of YKL-40 expression in vivo, especially in early stage of atherosclerotic lesion development [7]. In addition, in vitro studies have shown that analogues of YKL-40 induce the proliferation and migration of VSMCs [8]. Thus, YKL-40 is involved in at least two key events of ontogenesis of atherosclerosis as well as plaque generation. Following animal in vivo and human in vitro studies, clinical investigations found elevated YKL-40 in patients with coronary artery disease (CAD) compared to healthy subjects [9–12]. YKL-40 is associated with the severity of CAD expressed by the number of affected coronary artery vessels as well as atherosclerotic lesion progression [9,13]. A recent study could not confirm YKL-40’s association with the number of affected vessels [12]. However, this discrepancy might be caused by the anti-inflammatory effect of
558
K. Batinic et al. / Atherosclerosis 222 (2012) 557–563
statins that lower YKL-40 levels [14]. YKL-40 is elevated in acute myocardial infarction (AMI) as a consequence of acute plaque rupture [10,11]. During AMI, YKL-40 correlates with C-reactive protein (CRP). One month after AMI this association diminishes as CRP normalizes and YKL-40 remains elevated. This makes a role of YKL40 in myocardial tissue remodeling plausible [10]. In stable CAD, YKL-40 correlates with CRP [10,13]. Recently, YKL-40 was shown as independent predictor of all-cause and cardiovascular mortality in CAD patients [15]. Moreover, YKL-40 has been identified as predictor of ischemic stroke, thus being relevant for another, the cerebrovascular bed [16]. However, YKL-40 is not only associated with consequence (endpoints) of CAD–AMI, but also its cardiovascular risk factors. YKL-40 is elevated in patients with type 2 diabetes mellitus (T2DM) and type 1 diabetes mellitus (T1DM) [17–20]. The association of YKL-40 with body mass index (BMI)/obesity with or without diabetes and vice versa is controversial [18,20,21]. In childhood T1DM the elevation of YKL-40 compared to healthy controls is driven by obesity and components of the metabolic syndrome [22]. Diabetic patients with albuminuria have a high cardiovascular disease burden. Interestingly, YKL-40 is correlated with the degree of albuminuria in diabetes [19,23] pointing towards a premature and accelerated atherosclerosis. In contrast to clinical and experimental data in the coronary vascular bed, a disease modulating or at least event predicting potency of YKL-40 was not investigated in other sites of atherosclerosis such as aneurysms, peripheral arterial disease (PAD) or microvascular diseases so far. PAD is characterized by a generalized systemic atherosclerosis of arteries with a frequent co-occurrence of atherosclerosis in other parts of the vascular bed and associated worse prognosis compared to CAD [24]. The fatal combination of diabetes and PAD results in an even more severe and accelerated form of atherosclerosis leading to a three to five times higher risk for cardiovascular mortality in contrast to both morbidities alone [25]. Up to date, cardiovascular biomarkers have failed to improve predictability of prognosis of patients with PAD with and without diabetes compared to each other or to patients with CAD with and without diabetes. Thus, we investigated the potential role of YKL-40 in a cohort study of patients suffering from PAD with and without diabetes and control subjects. 2. Patients and methods 2.1. Patients After approval of the ethics committee of the Medical University of Vienna and written informed consent from all participants, 460 subjects were enrolled at the Division of Angiology of the Department of Medicine II at the General Hospital Vienna and at the Department of Medicine I at the Rudolfstiftung Hospital: 39 healthy subjects (CO), 85 with DM without known macro- and microvascular complications, 20 with DM and cerebro- or cardiovascular disease but without PAD (AS-DM), 316 with PAD subdivided by standardized oral glucose tolerance tests (oGTT) [26] in 71 normal glucose metabolism (PAD-NGM), 90 pre-DM (PAD-PREDM) having either impaired fasting glucose or impaired glucose tolerance status, and 155 with concomitant DM (PAD-DM). PAD stage was classified according to the Fontaine classification system and its adaption as used in TASC II [24] on the basis of anamnesis of pain-free walking distance, clinical examination, ankle-brachial index (ABI) and oscillometric measurements. PAD stage I (n = 124) was defined by ABI values <0.9, whereas PAD stage IIa was diagnosed in existence of claudicatio intermittens and pain-free walking distance >200 m (n = 106); and stage IIb with pain-free walking distance <200 m (n = 52). In 34 subjects it was not
possible to differ between stages I and II because of comorbidities such as vertebrostenosis and others that also affect walking distance. Mönckeberg’s mediasclerosis was assessed in appearance of ABI >1.4 in non-compressible arteries (as described in a previous study) [27]. Exclusion criteria were acute/inflammatory states of PAD such as acute ischemia (PAD stage III) or ulcer (PAD stage IV), critical illness within the last six months, cancer, renal failure (serum creatinine >3 mg/dl), connective tissue diseases, and hormone replacement therapy. Other sites of atherosclerosis diagnosed were recorded as well: in summary, of our 460 subjects, 124 did not have any diagnosed atherosclerosis, 179 had PAD alone, 15 CAD alone, and 3 cerebrovascular disease alone. Combinations of different atherosclerosis sites were also frequently represented: 87 had PAD and CAD, 36 PAD and cerebrovascular disease, 2 CAD and cerebrovascular disease and 14 had PAD/CAD/cerebrovascular disease. 2.2. Methods After overnight fast, venous blood specimens were obtained and immediately centrifuged, aliquoted, and frozen at −80 ◦ C. Urine samples were collected for calculation of urine albumin/creatinine ratio (UACR). Renal function was estimated using creatinine clearance (ml/min) and estimated glomerular filtration rate (eGFR) ml/min/1.72 m2 . Total cholesterol, low-density-lipoprotein cholesterol (LDL), triglycerides, aspartate-amino-transferase (ASAT), alanine-amino-transferase (ALAT), fasting glucose, fasting insulin, and HbA1c were determined. Insulin resistance was estimated on the homeostasis model assessment for insulin resistance formula (HOMA-IR). Metabolic syndrome (MetS) was diagnosed according to the WHO classification system [26]. Assessment of smoking history, medication, family history and cardiovascular risk factors were obtained by questionnaire and patients’ medical carts. C-reactive-protein (CRP) was measured by an immunonephelometric assay in serum. Serum YKL-40 levels were measured by a sandwich ELISA (Quidel Corporation, San Diego, CA) with an intra-assay and inter-assay coefficient of variation of 6.8% and 10.9%. 2.3. Statistics All statistical analyses were performed with the statistical software package SPSS 15.0 (SPSS Inc., Chicago, IL). Data are presented as mean ± standard deviation, or median plus percentile [25th/75th] or percentages [%]. YKL-40 showed a non-Gaussian leftskewed distribution and was log10 transformed. CRP also exhibited a non-Gaussian left-skewed distribution, but was used without mathematical transformation, following the strategy of Ridker et al. [28]. Statistics included student’s unpaired t-test, ANOVA, 2 -test, univariate and multivariate regression analysis as appropriate. Multivariate modeling was based on the prior establishment of a multivariate model of all univariate significant determinants of YKL-40 levels. After a confounding analysis following the recommendations of a cut off of 10% beta change by Rothman et al. [29], all resulting independent parameters for YKL-40 were included in a stepwise backward regression analysis. After obtaining the presumptuous final model, the model was challenged by multiple multivariate adjustments. An alpha-level of p < 0.05 (two-fold) was considered statistically significant. 3. Results 3.1. Patients’ characteristics Baseline data are described in Table 1. DM, PAD-PREDM and PAD-DM had >50% prevalence of MetS. In addition, PAD-DM
K. Batinic et al. / Atherosclerosis 222 (2012) 557–563
559
Table 1 Baseline characteristics. Group number Group name
1 CO
2 DM
3 PAD-NGM
4 PAD-PREDM
5 PAD-DM
6 AS-DM
p All groups
n Age (years) Gender (m %) BMI (kg/m2 ) Cardiovascular events Mediasclerosis MetS (WHO) BPs (mmHg) BPd (mmHg) Creatinine clearance (ml/min) eGFR (ml/min/m2 ) Creatinine (mol/L) UACR (mg/g) Triglyceride (mmol/L) HDL-C (mmol/L) Total cholesterol (mmol/L) Lipoprotein a (mol/L) LDL-C (mmol/L) Bilirubin (mol/L) AP (U/L) ASAT (U/L) ALAT (U/L) CRP (mg/L) Glucose (mmol/L) HbA1c (rel. %)
39 42.1 ± 13.4 17 (44%) 26.5 ± 4.9 – – – n.a. n.a. 106.2 ± 30.8 86.3 ± 13.2 79.9 ± 15.1 n.a. 2.0 ± 1.3 1.4 ± 0.4 5.5 ± 1.2 n.a 3.2 ± 1.1 10.6 ± 6.2 64.3 ± 24.9 27.0 ± 16.5 25.6 ± 16.4 3.3 ± 4.8 4.8 ± 0.8 5.4 ± 0.4
85 60.9 ± 9.8 46 (54%) 30.9 ± 5.5 – – 20 (50%) 144 ± 21 86 ± 12 96.1 ± 48.2 75.8 ± 37.4 100.1 ± 40.8 171 ± 405.4 2.2 ± 1.4 1.3 ± 0.4 4.8 ± 1.1 n.a 2.6 ± 0.9 11.5 ± 6.8 79.4 ± 26.7 26.7 ± 8.7 27.5 ± 12.6 4.1 ± 6.6 9.6 ± 4.0 8.1 ± 1.7
71 68.3 ± 10.7 39 (55%) 25.8 ± 3.7 10 10 (19.6%) 0 138 ± 20 78 ± 14 71.8 ± 25.5 71.8 ± 16.9 90.2 ± 20.8 46.7 ± 181.2 1.8 ± 0.9 1.5 ± 0.3 5.3 ± 1.2 2.0 ± 2.1 3.1 ± 1.1 10.9 ± 3.7 79.2 ± 24.1 25.2 ± 6.2 23.7 ± 10.0 3.2 ± 2.3 5.0 ± 0.4 5.7 ± 0.3
90 68.5 ± 11.2 59 (66%) 27.7 ± 3.8 10 13 (21.7%) 50 (55.5%) 140 ± 21 80 ± 10 75.9 ± 28.6 69.9 ± 17.6 95.0 ± 22.1 35.6 ± 133.9 2.0 ± 1.2 1.4 ± 0.5 5.1 ± 1.1 2.1 ± 2.7 2.8 ± 0.9 11.9 ± 4.3 77.6 ± 23.8 26.9 ± 8.2 27.2 ± 14.1 5.3 ± 8.5 5.9 ± 0.4 5.8 ± 0.4
155 70.1 ± 9.6 108 (70%) 29.0 ± 4.8 25 47 (48%) 81 (64.8%) 146 ± 23 79 ± 11 73.7 ± 32.4 66.1 ± 22.4 104.4 ± 32.1 156.8 ± 385.1 2.2 ± 1.5 1.3 ± 0.3 4.8 ± 1.3 1.6 ± 2.0 2.6 ± 1.0 11.5 ± 4.3 85.7 ± 49.6 25.6 ± 9.3 27.9 ± 12.9 5.0 ± 5.5 8.1 ± 2.3 7.1 ± 1.2
20 61.5 ± 8.1 14 (70%) 30.8 ± 4.6 – – – 150 ± 23 86 ± 14 80.9 ± 38.9 61.1 ± 29.2 119.9 ± 40.8 314.1 ± 354.4 2.4 ± 1.0 1.2 ± 0.4 4.6 ± 0.6 n.a. 2.4 ± 0.7 n.a. n.a. n.a. n.a. 5.7 ± 6.5 n.a. 7.8 ± 1.3
<0.001* 0.015 <0.001* 0.191 <0.001* <0.001* 0.034 <0.001* <0.001* <0.001* <0.001* <0.001* 0.166 <0.001* 0.003 0.227 <0.001* 0.608 0.041 0.712 0.260 0.191 <0.001* <0.001*
Data are mean ± STD, or number (n) of individuals and percentages (%). PAD-NGM, peripheral arterial disease with a normal glucose metabolism; PAD-PREDM, PAD with impaired glucose tolerance or impaired fasting glucose; PAD-DM, PAD with type 2 diabetes mellitus; AS-DM, cerebrovascular or cardiovascular disease with DM but without PAD; na, not available; BMI, body-mass-index; MetS (WHO), metabolic syndrome according to the WHO classification; BPs, systolic blood pressure; BPd, diastolic blood pressure; eGFR, estimated glomerular filtration rate; UACR, urine albumin/creatinine-ratio; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; AP, alkaline phosphatase; ASAT, aspartate-aminotransferase; ALAT, alanine-aminotransferase; CRP, C-reactive protein; HbA1c, glycosylated haemoglobin. A p-value <0.05 was considered significant. * Indicates sustained significance (<0.05) after Bonferroni correction.
patients showed higher BMI (p < 0.001), systolic blood pressure (p = 0.011), albuminuria (p = 0.005), triglycerides (p = 0.006), CRP (p = 0.002), glucose and HbA1 C (p < 0.001) but lower HDL (p < 0.001), compared to PAD-NGM.
3.2. Serum YKL-40 levels in comparison to the different subject groups YKL-40 was elevated in DM patients (median [25–75 percentile]: 96 [57–171]; in PAD-NGM: 84 [53–128]; in PAD-PREDM: 114 [79–164], p < 0.001; in PAD-DM: 113 [71–157], p < 0.001; and in AS-DM: 100 [58–170], p = 0.002; compared to CO: 43 [30–80] ng/ml (Fig. 1). In PAD, YKL-40 was higher in patients with PREDM (p = 0.001) and DM (p = 0.01) compared to patients without glycemic dysregulation (Fig. 1). However between PAD-PREDM and PAD-DM group YKL-40 was not different (p = 0.311) (Fig. 1). PAD patients exhibited (independently of glucose status) higher YKL-40 values compared to healthy subjects: 103 [69–159] vs. 43 [30–80] ng/ml, p < 0.001 (Fig. 2a). In addition, YKL-40 was higher in patients with diabetes (irrespective of atherosclerosis) compared to non-diabetics: 100 [66–165] vs. 68 [39–114] ng/ml, p < 0.001 (Fig. 2b). Upon division of all patients in those with and without PAD, YKL-40 was higher in patients with PAD: 103 [69–159] vs. 85 [47–159] ng/ml, p = 0.003 (Fig. 2c). Next, we investigated whether YKL-40 is associated with wide spreading of atherosclerosis: Clearly YKL-40 was different in patients with no atherosclerosis vs. either PAD or CAD or cerebrovascular disease vs. dual combination of the latter vs. the triple combination: none vs. one vs. two vs. three affected vessel beds: 79 [44–154] vs. 98 [62–162] vs. 108 [70–156] vs. 145 [78–201] ng/ml, p = 0.004 (independent of glucose status).
Fig. 1. Depicted are log YKL-40 values in the respective groups: CO, controls; DM, type 2 diabetes mellitus; PAD-NGM, peripheral arterial disease with normal glucose metabolism; PAD-PREDM, peripheral arterial disease with pre-diabetes; PAD-DM, peripheral arterial disease with diabetes; AS-DM, atherosclerosis without PAD but with DM. All p-values shown are in between each group differences. Only the upper line resembles the in between all group difference.
3.3. YKL-40 and C-reactive protein YKL-40 and C-reactive protein (CRP) are both inflammatory proteins, the last being a classic hallmark of atherosclerosis. As published previously, PAD-DM (p = 0.002) and PAD-PREDM (p = 0.026) express higher CRP compared to PAD-NGM. We failed to obtain a difference between PAD-DM and PAD-PREDM (p = 0.708). YKL40 followed a similar pattern: whereas YKL-40 was elevated in
560
K. Batinic et al. / Atherosclerosis 222 (2012) 557–563
Fig. 2. Depicted are log YKL-40 values in the respective groups: CO, controls; PAD, peripheral arterial disease; AS, sites of atherosclerosis. All p-values shown are in between each group differences.
PAD-DM (p = 0.01) and PAD-PREDM (p = 0.001) compared to PADNGM, YKL-40 did not differ between PAD-DM and PAD-PREDM (Fig. 1). In addition, we report a correlation of YKL-40 and CRP (Table 2). Even more, CRP is an independent predictor of YKL-40 levels in the total study population (not shown) as well as in the PAD cohort (Table 3). 3.3.1. Univariate correlation analysis for identification of associates of YKL-40 To identify predictors of YKL-40, univariate correlation analysis was done (Table 2). Age is associated with YKL-40 in all groups except PAD-PREDM and AS-DM. Furthermore triglycerides correlated with YKL-40 in all patients, in CO, in PAD and in the subgroup PAD-PREDM, but not in patients with diabetes without atherosclerosis. In addition, an association of YKL-40 and aspartateaminotransferase in all patients, control, in all PAD patients, in PAD-DM subgroup, in non-diabetes, diabetes and non-PAD was observed. 3.3.2. Multivariate regression analysis for predictors of YKL-40 in PAD patients All variables significantly associated with YKL-40 in univariate fashion in PAD patients (n = 316) (Table 2: PAD), were taken into a multivariate stepwise backward regression analysis (Table 3): Model A (unadjusted) revealed age (beta = 0.230, p < 0.001), triglycerides (beta = 0.157, p = 0.006), ASAT (beta = 0.192, p = 0.001) and CRP (beta = 0.159, p = 0.001) as presumptuous independent predictors of YKL-40 (Table 3). Adjustment for gender (Model B) did not change latter associations with YKL-40. Further adjustment for renal parameters (creatinine clearance, UACR) (Model C), did not change associations of triglycerides, ASAT, and CRP, but the association of YKL-40 and age was attenuated (beta = 0.183, p = 0.026) (more than 10% beta change). Additional adjustment
for lipid parameters (HDL-C, total cholesterol, LDL-C, lipoprotein a) (Model D) improved the association of YKL-40 with age (beta = 0.208, p = 0.012) and triglycerides (beta = 177, p = 0.049) but did not change the association with ASAT (beta = 0.187, p = 0.001) and CRP (beta = 0.182, p = 0.002). Adjustment for liver parameters (bilirubin, AP, ALAT) (Model E) did have no impact on the associations. Moreover, additional adjustment for HOMA (Model F) did not alter the association of CRP with YKL-40, but significantly improved the association of age and triglycerides. In contrast, the association of ASAT with YKL-40 was not attenuated. Further adjustment for blood pressure (Model G) and BMI (Model H) did not alter the associations of our four explanatory variables with YKL-40 apart of a strengthening of age. Thus, we are able to state that in our cohort, after multivariate analysis and multiple adjustments, age, triglycerides, ASAT and CRP were identified as most independent predictors of YKL-40. 4. Discussion We are the first to demonstrate elevated levels of YKL-40 in patients with severe and generalized atherosclerosis. YKL-40 is significantly elevated in our cohort of patients with peripheral arterial disease (PAD), independently of the glucose status. Among patients with PAD, YKL-40 levels differed between patients with pre-diabetes or diabetes compared to normal glucose metabolism. In addition, we confirm that patients with diabetes without atherosclerosis exhibit higher YKL-40 compared to healthy controls [17–19]. Furthermore we are able to add, that in a small group of patients with diabetes and atherosclerosis (=coronary artery or cerebrovascular disease) but without PAD, YKL-40 levels are augmented as well. A previous study could only demonstrate a trend – most presumably because of small sample size [30]. Interestingly, in our study cohort, YKL-40 was correlated with the number of the
K. Batinic et al. / Atherosclerosis 222 (2012) 557–563 Table 3 Multivariate regression for predictors of YKL-40 in all PAD patients.
Table 2 Univariate associations of YKL-40 in the different patient groups. Group
Variable
R
p
All
Age Creatinine clearance Estimated glomerular filtration rate Urine albumin/creatinine-ratio Creatinine Triglyceride Lipoprotein a Alkaline phosphatase Aspartate-aminotransferase C-reactive protein Glucose Glycosylated haemoglobin Gender
0.338 −0.203 −0.184 0.152 0.183 0.144 −0.129 0.172 0.194 0.167 0.201 0.099 0.091
<0.001* <0.001* <0.001* 0.003* <0.001* 0.003* 0.032 0.001* <0.001* 0.002* <0.001* 0.038 0.050
Age Triglyceride Alkaline phosphatase Aspartate-aminotransferase Alanine-aminotransferase C-reactive protein Glucose Glycosylated haemoglobin Gender
0.508 0.443 0.479 0.438 0.517 0.368 0.566 0.516 0.352
0.001* 0.016 0.005* 0.014 0.003* 0.042 0.001* 0.004* 0.028
Age Urine albumin/creatinine-ratio
0.277 0.360
0.010* 0.011*
0.332 −0.425 −0.407 0.275 0.377
0.005* <0.001* <0.001* 0.023 0.001*
CO
DM
PAD-NGM
Age Creatinine clearance Estimated glomerular filtration rate Urine albumin/creatinine-ratio Creatinine
561
PAD-PREDM
Triglyceride C-reactive protein
0.263 0.259
0.013* 0.014*
PAD-DM
Age Aspartate-aminotransferase
0.179 0.258
0.026* 0.004*
AS-DM
Creatinine clearance Estimated glomerular filtration rate Creatinine
−0.447 −0.517 0.582
0.048 0.020* 0.007*
Non-diabetes
Age Creatinine clearance Estimated glomerular filtration rate Urine albumin/creatinine-ratio Creatinine Alkaline phosphatase Aspartate-aminotransferase Glucose Glycosylated haemoglobin
0.488 −0.402 −0.447 0.275 0.406 0.260 0.228 0.279 0.221
<0.001* <0.001* <0.001* 0.023 <0.001* 0.008 0.022 0.005* 0.029
Diabetes
Age Creatinine clearance Urine albumin/creatinine-ratio Aspartate-aminotransferase Alanine-aminotransferase
0.193 −0.146 0.192 0.261 0.184
0.002* 0.021 0.004* 0.001* 0.020
Non-PAD
Age Urine albumin/creatinine-ratio Creatinine Alkaline phosphatase Aspartate-aminotransferase Alanine-aminotransferase Glucose Glycosylated haemoglobin Body-mass-index
0.457 0.360 0.203 0.420 0.284 0.373 0.514 0.199 0.208
<0.001* 0.002* 0.021 <0.001* 0.023 0.002* <0.001* 0.024 0.014
PAD
Age Creatinine clearance Estimated glomerular filtration rate Urine albumin/creatinine-ratio Creatinine Triglyceride Lipoprotein a Aspartate-aminotransferase C-reactive protein
0.206 −0.194 −0.173 0.115 0.174 0.149 −0.129 0.164 0.154
<0.001* 0.001* 0.002* 0.046 0.002* 0.009 0.032 0.006* 0.008
CO, healthy subjects; DM, type 2 diabetes mellitus; PAD-NGM, peripheral arterial disease with a normal glucose metabolism; PAD-PREDM, PAD with impaired glucose tolerance or impaired fasting glucose; PAD-DM, PAD with type 2 diabetes mellitus. A p-value <0.05 was considered significant. * Indicates sustained significance (<0.05) after Bonferroni correction.
Models
Variable
Beta
p
Model A Unadjusted
Age Triglyceride Aspartate-aminotransferase C-reactive protein
0.230 0.157 0.192 0.159
<0.001 0.006 0.001 0.006
Model B + Adjusted for gender
Age Triglyceride Aspartate-aminotransferase C-reactive protein
0.239 0.151 0.189 0.165
<0.001 0.009 0.001 0.004
Model C + Adjusted for renal parameters
Age Triglyceride Aspartate-aminotransferase C-reactive protein
0.183 0.155 0.208 0.167
0.026 0.008 <0.001 0.004
Model D + Adjusted for lipid parameters
Age Triglyceride Aspartate-aminotransferase C-reactive protein
0.208 0.177 0.187 0.182
0.012 0.049 0.001 0.002
Model E + Adjusted for liver parameters
Age Triglyceride Aspartate-aminotransferase C-reactive protein
0.209 0.181 0.190 0.172
0.015 0.048 0.030 0.005
Model F + Adjusted for HOMA
Age Triglyceride Aspartate-aminotransferase C-reactive protein
0.245 0.221 0.181 0.174
0.005 0.018 0.038 0.005
Model G + Adjusted for blood pressure
Age Triglyceride Aspartate-aminotransferase C-reactive protein
0.252 0.221 0.176 0.174
0.005 0.018 0.046 0.005
Model H + Adjusted for BMI
Age Triglyceride Aspartate-aminotransferase C-reactive protein
0.272 0.216 0.177 0.178
0.004 0.021 0.045 0.005
HOMA, homeostasis model assessment; BMI, body-mass-index; p-value <0.05 was considered significant.
affected vessel beds. All those findings argue for a strong involvement of YKL-40 not only in diabetes and CAD but also in both, and potentially even more important, in PAD, which exhibits the worst prognosis of all clinical manifestations of atherosclerosis. Since our study is cross-sectional by nature we cannot elucidate whether YKL-40 is cause or consequence or even – but most unlikely – an innocent bystander of atherosclerosis. However, multivariate regression analysis revealed that YKL-40 was predicted by the four most independent predictors age, triglycerides, ASAT and CRP between 24.2% and 21.2% in the total study population (unadjusted; final adjusted model) as well as in PAD patients (14.0%/18.3%). Since all four variables are associated with different associates of atherosclerosis (age, lipid metabolism, nonalcoholic steatotic hepatitis, vascular inflammation) the possibility of chance is rendered unlikely. Recent data have demonstrated that YKL-40 is elevated in patients with CAD compared to controls [9]. Even more the extension of affected coronary artery vessels and progression of atherosclerosis are attended by higher YKL-40 values [9,13]. In our study we confirm that extension/severity of atherosclerosis is associated with higher YKL-40 levels. Whether YKL-40 is cause or consequence is subject of ongoing studies, however, activated macrophages in acute as well as in chronic diseases secrete large amounts of YKL-40 [6,7]. Moreover, the induction of release of YKL40 from macrophages was shown in early lesions of atherosclerosis [7]. In those in vitro studies YKL-40 was most highly concentrated deeply in the centre core of atherosclerotic lesions and the elevated secretion was most pronounced during foam cell development [31].
562
K. Batinic et al. / Atherosclerosis 222 (2012) 557–563
One of the major cardiovascular risk factors – diabetes – is associated with YKL-40: Rathcke et al. have demonstrated that YKL-40 is elevated in T1DM and T2DM in adults [17,19]. Possibly YKL-40 is already involved in the early development of disease, since an elevation of YKL-40 already in children with T1DM has been shown [22]. Although diabetes and obesity are often jointly observed as diabesity, an association of BMI and YKL-40 [18,21] is rare except gestational diabetes [20]. Herein we endorse the association of YKL40 and diabetes. Until now no data was available on YKL-40 in patients with a cooccurrence of diabetes and generalized atherosclerosis. We show for the first time that YKL-40 is elevated in patients with a combination of diabetes and a generalized form of atherosclerosis (PAD) compared to controls. Furthermore YKL-40 in patients with diabetes and atherosclerosis is also more elevated than in patients with atherosclerosis alone (PAD). Although the association of YKL-40 and atherosclerosis is so far stringent, a plausible relationship of YKL-40 to prediabetes has not been investigated. In our study, PAD patients with an impaired glucose metabolism have comparable elevations of YKL-40 levels as PAD patients with diabetes. Since impaired fasting glucose and impaired glucose tolerance have been demonstrated to be associated with an intermediate cardiovascular hazard ratio of 1.6/1.5 (95% CI: 1.0–2.4/1.1–2.0) between newly diagnosed diabetes and known diabetes [32], an intermediate augmentation of YKL-40 from no glucose disturbance to diabetes in our patients with prediabetes mimics the cardiovascular risk increase of those patients. Nevertheless our study is cross-sectional at the time being, and we cannot delineate an increase of cardiovascular risk. Atherosclerosis, pre-diabetes and diabetes are inflammatory conditions with elevations of responsible proteins such as CRP and IL-6 [33]. YKL-40 is elevated in diabetes and also in atherosclerosis. We add that YKL-40 is elevated in patients with atherosclerosis and concomitant diabetes. Furthermore, we first found an elevation of YKL-40 in pre-diabetes. In our cohort, YKL-40 and CRP follow a similar pattern: both are elevated in PAD-DM and PAD-PREDM vs. PAD-NGM; however we failed to obtain a difference in both proteins in between PAD-PREDM and PAD-DM. This suggests that both proteins increase with disease, but are most presumable differentially involved in different diseases. In accordance with our data, YKL-40 is described to be associated with CRP levels in CAD [9,10]. In contrary, no association of YKL-40 and CRP was found in patients with DM alone [17]. Recent studies demonstrated an association of YKL-40 and albuminuria in patients with DM [23], which is already known in patients with T1DM [19]. We can extend this observation to patients with atherosclerotic disease: PAD patients with albuminuria exhibited higher YKL-40. Bearing in mind, the in vivo and in vitro data for an involvement of YKL-40 in tissue remodeling, an association of YKL-40 and albuminuria – a strong remodeling inducer [34] – merits further investigation. Mediasclerosis is a tissue remodeling process related to vascular calcification [35]. Mediasclerosis is known to be a common appearance in DM patients. We confirm that mediasclerosis is more frequent in PAD-DM than in PAD-NGM (p = 0.001) or PADPREDM (p = 0.001). In our study we found an association of YKL-40 and mediasclerosis by demonstrating that patients with mediasclerosis have higher YKL-40 levels compared to patients without mediasclerosis: 135.1 [76.5–171.1 vs. 93.8 [53.2–153.9], p = 0.010 (median [25th–75th percentile]). Linear regression revealed an unadjusted association of YKL-40 with mediasclerosis (beta = 0.218, p = 0.008), which was attenuated after adjustment for diabetes (beta = 0.180, p = 0.032). Whether this independent 3.24% influence on the variation coefficient of YKL-40 is biologically relevant and involved in the development of mediasclerosis cannot be answered by our study. Nevertheless, YKL-40 is involved in tissue remodeling and fibrogenesis [5].
The limitation of our study is that for the time being, we could only examine our cohort study cross-sectionally. However, we found that YKL-40 is higher in PAD patients. In addition, patients with diabetes but also prediabetes exhibited higher values. Furthermore our data suggest that YKL-40 might be involved in mediasclerosis, a tissue remodeling process, whose causes are rarely understood. References [1] Millis AJ, Hoyle M, Reich E, Mann DM. Isolation and characterization of a Mr = 38.000 protein from differentiating smooth muscle cells. J Biol Chem 1985;260:3754–61. [2] Vos K, Steenbakkers P, Miltenburg AM, et al. Raised human cartilage glycoprotein-39 plasma levels in patients with rheumatoid arthritis and other inflammatory conditions. Ann Rheum Dis 2000;59:544–8. [3] Volck B, Price PA, Johansen JS, et al. YKL-40, a mammalian member of the bacterial chitinase family, is a matrix protein of specific granules in human neutrophils. Proc Assoc Am Physicians 1998;110:351–60. [4] Johansen JS, Bojesen SE, Mylin AK, et al. Elevated plasma YKL-40 predicts increased risk of gastrointestinal cancer and decreased survival after any cancer diagnosis in the general population. J Clin Oncol 2009;27:572–8. [5] Johansen JS, Christoffersen P, Møller S, et al. Serum YKL-40 is increased in patients with hepatic fibrosis. J Hepatol 2000;32:911–20. [6] Rehli M, Niller HH, Ammon C, et al. Transcriptional regulation of CHI3L1, a marker for late stages of macrophage differentiation. J Biol Chem 2003;278:44058–67. [7] Boot RG, van Achterberg TA, van Aken BE, et al. Strong induction of members of the chitinase family of proteins in atherosclerosis: chitotriosidase and cartilage gp-39 expression in lesion macrophages. Arterioscler Thromb Vasc Biol 1999;19:687–94. [8] Nishikawa KC, Millis AJ. gp38k (CHI3L1) is a novel adhesion and migration factor for vascular cells. Exp Cell Res 2003;287:79–87. [9] Kucur M, Isman FK, Karadag B, Vural VA, Tavsanoglu S. Serum YKL-40 levels in patients with coronary artery disease. Coron Artery Dis 2007;18:391–6. [10] Wang Y, Ripa RS, Johansen JS, et al. YKL-40 a new biomarker in patients with acute coronary syndrome or stable coronary artery disease. Scand Cardiovasc J 2008;42:295–302. [11] Nøjgaard C, Høst NB, Christensen IJ, et al. Serum levels of YKL-40 increases in patients with acute myocardial infarction. Coron Artery Dis 2008;19:257–63. [12] Mathiasen AB, Harutyunyan MJ, Jørgensen E, et al. Plasma YKL-40 in relation to the degree of coronary artery disease in patients with stable ischemic heart disease. Scand J Clin Lab Invest 2011;71:439–44. [13] Zheng JL, Lu L, Hu J, et al. Increased serum YKL-40 and C-reactive protein levels are associated with angiographic lesion progression in patients with coronary artery disease. Atherosclerosis 2010;21:590–5. [14] Mygind ND, Harutyunyan MJ, Mathiasen AB, et al. The influence of statin treatment on the inflammatory biomarkers YKL-40 and HsCRP in patients with stable coronary artery disease. Inflamm Res 2011;60:281–7. [15] Kastrup J, Johansen JS, Winkel P, et al. High serum YKL-40 concentration is associated with cardiovascular and all-cause mortality in patients with stable coronary artery disease. Eur Heart J 2009;30:1066–72. [16] Kjaergaard AD, Bojesen SE, Johansen JS, Nordestgaard BG. Elevated plasma YKL-40 levels and ischemic stroke in the general population. Ann Neurol 2010;68:672–80. [17] Rathcke CN, Johansen JS, Vestergaard H. YKL-40, a biomarker of inflammation, is elevated in patient with type 2 diabetes and is related to insulin resistance. Inflamm Res 2006;55:53–9. [18] Nielsen AR, Erikstrup C, Johansen JS, et al. Plasma YKL-40: a BMI-independent marker of type 2 diabetes. Diabetes 2008;57:3078–82. [19] Rathcke CN, Persson F, Tarnow L, Rossing P, Vestergaard H. YKL-40, a marker of inflammation and endothelial dysfunction, is elevated in patients with type 1 diabetes and increases with levels of albuminuria. Diabetes Care 2009;32:323–8. [20] Schaller G, Brix J, Placher-Sorko G, et al. YKL-40 concentrations are not elevated in gestational diabetes. Eur J Clin Invest 2010;40(4):339–43. [21] Hempen M, Kopp HP, Elhenicky M, et al. YKL-40 is elevated in morbidly obese patients and declines after weight loss. Obes Surg 2009;19:1557–63. [22] Hoertenhuber T, Hoellerl F, Hoebaus C, et al. YKL-40 is elevated in type 1 diabetic children which show components of metabolic syndrome compared to those without, as well as to healthy control subjects. Diabetologia 2009;52:S1–345. [23] Brix JM, Hoellerl F, Koppensteiner R, Schernthaner G, Schernthaner GH. YKL-40 in type 2 diabetic patients with different levels of albuminuria. Eur J Clin Invest 2011;41:589–96. [24] Norgren L, Hiatt WR, Dormandy JA, et al. Inter-society consensus for the management of peripheral arterial disease (TASC II). Eur J Vasc Endovasc Surg 2007;33(Suppl 1):S1–75. [25] Kallio M, Forsblom C, Groop PH, Groop L, Lepantalo M. Development of new peripheral arterial occlusive disease in patients with type 2 diabetes during a mean follow-up of 11 years. Diabetes Care 2003;26:1241–5. [26] Report of a WHO consultation. In: Alwan A, King H, editors. Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1: diagnosis and classification of diabetes mellitus. Geneva: World Health
K. Batinic et al. / Atherosclerosis 222 (2012) 557–563
[27]
[28]
[29] [30]
Organization, Department of Noncommunicable Disease Surveillance; 1999. p. 1–59. http://www.who.int. Lorant DP, Grujicic M, Hoebaus C, et al. Fetuin-A levels are increased in patients with type 2 diabetes and peripheral arterial disease. Diabetes Care 2011;34(Jan (1)):156–61. Ridker PM, Hennekens CH, Buring J, Rifai N. C-reactive protein and other markers of inflammation in the prediction of cardiovascular disease in women. N Engl J Med 2000;342:836–43. Rothman K, Greenland S. Modern epidemiology. Philadelphia: Lippincott Williams & Wilkins; 1998. Thomsen SB, Rathcke CN, Zerahn B, Vestergaard H. Increased levels of the calcification marker matrix Gla Protein and the inflammatory markers YKL-40 and CRP in patients with type 2 diabetes and ischemic heart disease. Cardiovasc Diabetol 2010;9:86.
563
[31] Fach EM, Garulacan LA, Gao J, et al. In vitro biomarker discovery for atherosclerosis by proteomics. Mol Cell Proteomics 2004;3:1200–10. [32] Barr E, Zimmet P, Welborn T, et al. Risk of cardiovascular and all-cause mortality in individuals with diabetes mellitus, impaired fasting glucose, and impaired glucose tolerance. Circulation 2007;116:151–7. [33] Pradhan AD, Manson JE, Rifai N, Buring JE, Ridker PM. C-reactive protein, interleukin 6, and risk of developing type 2 diabetes mellitus. JAMA 2001;286:327–34. [34] Hermans MM, Henry RM, Dekker JM, et al. Albuminuria, but not estimated glomerular filtration rate, is associated with maladaptive arterial remodeling: the Hoorn Study. J Hypertens 2008 Apr;26(4):791–7. [35] Demer LL, Tintut Y. Vascular calcification: pathobiology of a multifaceted disease. Circulation 2008;117(22):2938–48.
Contents lists available at SciVerse ScienceDirect
Atherosclerosis journal homepage: www.elsevier.com/locate/atherosclerosis
YKL-40 is elevated in patients with peripheral arterial disease and diabetes or pre-diabetes Klaudija Batinic a,b , Clemens Höbaus b , Milan Grujicic b , Angelika Steffan b , Finka Jelic b , David Lorant b,c , Thomas Hörtenhuber b,d , Florian Hoellerl b,e , Johanna-Maria Brix e , Guntram Schernthaner e , Renate Koppensteiner b , Gerit-Holger Schernthaner b,∗ a
Division of Pediatric Cardiology, University Zurich, Switzerland Medical University of Vienna, Department of Medicine II, Vienna, Austria c Medical University of Vienna, Department of Anesthesiology, Vienna, Austria d Medical University of Vienna, Department of Pediatrics, Vienna, Austria e Rudolfstiftung Hospital, Department of Medicine I, Vienna, Austria b
a r t i c l e
i n f o
Article history: Received 25 November 2011 Received in revised form 21 March 2012 Accepted 28 March 2012 Available online 4 April 2012 Keywords: : YKL-40 Peripheral artery disease Type 2 diabetes mellitus
a b s t r a c t Objective: YKL-40 is secreted by macrophages in atherosclerotic lesions and involved in plaque rupture. YKL-40 is elevated in coronary artery disease, and predicts cardiovascular mortality. Experimental in vivo and in vitro data suggest a role of YKL-40 in tissue remodeling. A disease modulating potency of YKL-40 was not investigated in peripheral arterial disease (PAD). Methods: We measured YKL-40 in 460 subjects: 316 PAD: 71 normal glucose metabolism (PAD-NGM), 90 pre-diabetes (PAD-PREDM) and 155 diabetes (PAD-DM); 20 diabetes with atherosclerosis but without PAD (AS-DM); 85 diabetes without macro-vascular complications (DM) and 39 healthy controls (CO). Results: YKL-40 is higher in PAD vs. CO (median [25–75 percentile]: 103 [69–159] vs. 43 [30–80] ng/ml; p < 0.001). In addition, YKL-40 is elevated in DM (p < 0.001), PAD-NGM (p = 0.001), PAD-PREDM (p < 0.001), PAD-DM (p < 0.001) and AS-DM (p = 0.002) compared to CO. Among PAD, YKL-40 is increased in PADPREDM (p = 0.001) and PAD-DM (p = 0.01) vs. PAD-NGM. By multivariate regression YKL-40 is significantly associated with age (beta = 0.272), triglycerides (beta = 0.216), aspartate-amino-transferase (beta = 0.177) and c-reactive-protein (beta = 0.178). Underpinning its role YKL-40 was found to be associated with micro-/macroalbuminuria (p = 0.014/p = 008) – a strong remodeling inducer. In addition, YKL-40 was elevated in existence of mediasclerosis (p = 0.008), a remodeling process. Conclusion: We are first to show that YKL-40 is higher in subjects with peripheral arterial disease. YKL-40 was higher in PAD patients with pre-/diabetes. In addition, YKL-40 was associated with the “severity” of generalized atherosclerosis estimated by affected vascular beds. All our findings point towards a role of YKL-40 in the progression/prognosis of patients with PAD and concomitant diabetes. © 2012 Elsevier Ireland Ltd. All rights reserved.
1. Introduction YKL-40, a chitin binding protein named after its three Nterminal amino acids tyrosine (Y), lysine (K), leucine (L) and its molecular mass of 40 kDa was first identified in 1985 under tissue remodeling and subsequently inflammatory conditions [1,2]. YKL40 – described to be involved in acute and chronic inflammatory reactions – is secreted in large amounts by vascular smooth muscle cells (VSMCs) [2], synovial cells, articular chondrocytes, activated neutrophils [3], cancer cells [4], and in fibrogenesis process [5].
∗ Corresponding author at: Medical University & General Hospital of Vienna, Department of Medicine II, Waehringer Guertel 18-20, A-1090 Vienna, Austria. Tel.: +43 1 40400 4671; fax: +43 1 40400 4665. E-mail address: [email protected] (G.-H. Schernthaner). 0021-9150/$ – see front matter © 2012 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.atherosclerosis.2012.03.034
Macrophages during late stage of differentiation are a major source of YKL-40 [6]. A subpopulation of activated macrophages shows a strong induction of YKL-40 expression in vivo, especially in early stage of atherosclerotic lesion development [7]. In addition, in vitro studies have shown that analogues of YKL-40 induce the proliferation and migration of VSMCs [8]. Thus, YKL-40 is involved in at least two key events of ontogenesis of atherosclerosis as well as plaque generation. Following animal in vivo and human in vitro studies, clinical investigations found elevated YKL-40 in patients with coronary artery disease (CAD) compared to healthy subjects [9–12]. YKL-40 is associated with the severity of CAD expressed by the number of affected coronary artery vessels as well as atherosclerotic lesion progression [9,13]. A recent study could not confirm YKL-40’s association with the number of affected vessels [12]. However, this discrepancy might be caused by the anti-inflammatory effect of
558
K. Batinic et al. / Atherosclerosis 222 (2012) 557–563
statins that lower YKL-40 levels [14]. YKL-40 is elevated in acute myocardial infarction (AMI) as a consequence of acute plaque rupture [10,11]. During AMI, YKL-40 correlates with C-reactive protein (CRP). One month after AMI this association diminishes as CRP normalizes and YKL-40 remains elevated. This makes a role of YKL40 in myocardial tissue remodeling plausible [10]. In stable CAD, YKL-40 correlates with CRP [10,13]. Recently, YKL-40 was shown as independent predictor of all-cause and cardiovascular mortality in CAD patients [15]. Moreover, YKL-40 has been identified as predictor of ischemic stroke, thus being relevant for another, the cerebrovascular bed [16]. However, YKL-40 is not only associated with consequence (endpoints) of CAD–AMI, but also its cardiovascular risk factors. YKL-40 is elevated in patients with type 2 diabetes mellitus (T2DM) and type 1 diabetes mellitus (T1DM) [17–20]. The association of YKL-40 with body mass index (BMI)/obesity with or without diabetes and vice versa is controversial [18,20,21]. In childhood T1DM the elevation of YKL-40 compared to healthy controls is driven by obesity and components of the metabolic syndrome [22]. Diabetic patients with albuminuria have a high cardiovascular disease burden. Interestingly, YKL-40 is correlated with the degree of albuminuria in diabetes [19,23] pointing towards a premature and accelerated atherosclerosis. In contrast to clinical and experimental data in the coronary vascular bed, a disease modulating or at least event predicting potency of YKL-40 was not investigated in other sites of atherosclerosis such as aneurysms, peripheral arterial disease (PAD) or microvascular diseases so far. PAD is characterized by a generalized systemic atherosclerosis of arteries with a frequent co-occurrence of atherosclerosis in other parts of the vascular bed and associated worse prognosis compared to CAD [24]. The fatal combination of diabetes and PAD results in an even more severe and accelerated form of atherosclerosis leading to a three to five times higher risk for cardiovascular mortality in contrast to both morbidities alone [25]. Up to date, cardiovascular biomarkers have failed to improve predictability of prognosis of patients with PAD with and without diabetes compared to each other or to patients with CAD with and without diabetes. Thus, we investigated the potential role of YKL-40 in a cohort study of patients suffering from PAD with and without diabetes and control subjects. 2. Patients and methods 2.1. Patients After approval of the ethics committee of the Medical University of Vienna and written informed consent from all participants, 460 subjects were enrolled at the Division of Angiology of the Department of Medicine II at the General Hospital Vienna and at the Department of Medicine I at the Rudolfstiftung Hospital: 39 healthy subjects (CO), 85 with DM without known macro- and microvascular complications, 20 with DM and cerebro- or cardiovascular disease but without PAD (AS-DM), 316 with PAD subdivided by standardized oral glucose tolerance tests (oGTT) [26] in 71 normal glucose metabolism (PAD-NGM), 90 pre-DM (PAD-PREDM) having either impaired fasting glucose or impaired glucose tolerance status, and 155 with concomitant DM (PAD-DM). PAD stage was classified according to the Fontaine classification system and its adaption as used in TASC II [24] on the basis of anamnesis of pain-free walking distance, clinical examination, ankle-brachial index (ABI) and oscillometric measurements. PAD stage I (n = 124) was defined by ABI values <0.9, whereas PAD stage IIa was diagnosed in existence of claudicatio intermittens and pain-free walking distance >200 m (n = 106); and stage IIb with pain-free walking distance <200 m (n = 52). In 34 subjects it was not
possible to differ between stages I and II because of comorbidities such as vertebrostenosis and others that also affect walking distance. Mönckeberg’s mediasclerosis was assessed in appearance of ABI >1.4 in non-compressible arteries (as described in a previous study) [27]. Exclusion criteria were acute/inflammatory states of PAD such as acute ischemia (PAD stage III) or ulcer (PAD stage IV), critical illness within the last six months, cancer, renal failure (serum creatinine >3 mg/dl), connective tissue diseases, and hormone replacement therapy. Other sites of atherosclerosis diagnosed were recorded as well: in summary, of our 460 subjects, 124 did not have any diagnosed atherosclerosis, 179 had PAD alone, 15 CAD alone, and 3 cerebrovascular disease alone. Combinations of different atherosclerosis sites were also frequently represented: 87 had PAD and CAD, 36 PAD and cerebrovascular disease, 2 CAD and cerebrovascular disease and 14 had PAD/CAD/cerebrovascular disease. 2.2. Methods After overnight fast, venous blood specimens were obtained and immediately centrifuged, aliquoted, and frozen at −80 ◦ C. Urine samples were collected for calculation of urine albumin/creatinine ratio (UACR). Renal function was estimated using creatinine clearance (ml/min) and estimated glomerular filtration rate (eGFR) ml/min/1.72 m2 . Total cholesterol, low-density-lipoprotein cholesterol (LDL), triglycerides, aspartate-amino-transferase (ASAT), alanine-amino-transferase (ALAT), fasting glucose, fasting insulin, and HbA1c were determined. Insulin resistance was estimated on the homeostasis model assessment for insulin resistance formula (HOMA-IR). Metabolic syndrome (MetS) was diagnosed according to the WHO classification system [26]. Assessment of smoking history, medication, family history and cardiovascular risk factors were obtained by questionnaire and patients’ medical carts. C-reactive-protein (CRP) was measured by an immunonephelometric assay in serum. Serum YKL-40 levels were measured by a sandwich ELISA (Quidel Corporation, San Diego, CA) with an intra-assay and inter-assay coefficient of variation of 6.8% and 10.9%. 2.3. Statistics All statistical analyses were performed with the statistical software package SPSS 15.0 (SPSS Inc., Chicago, IL). Data are presented as mean ± standard deviation, or median plus percentile [25th/75th] or percentages [%]. YKL-40 showed a non-Gaussian leftskewed distribution and was log10 transformed. CRP also exhibited a non-Gaussian left-skewed distribution, but was used without mathematical transformation, following the strategy of Ridker et al. [28]. Statistics included student’s unpaired t-test, ANOVA, 2 -test, univariate and multivariate regression analysis as appropriate. Multivariate modeling was based on the prior establishment of a multivariate model of all univariate significant determinants of YKL-40 levels. After a confounding analysis following the recommendations of a cut off of 10% beta change by Rothman et al. [29], all resulting independent parameters for YKL-40 were included in a stepwise backward regression analysis. After obtaining the presumptuous final model, the model was challenged by multiple multivariate adjustments. An alpha-level of p < 0.05 (two-fold) was considered statistically significant. 3. Results 3.1. Patients’ characteristics Baseline data are described in Table 1. DM, PAD-PREDM and PAD-DM had >50% prevalence of MetS. In addition, PAD-DM
K. Batinic et al. / Atherosclerosis 222 (2012) 557–563
559
Table 1 Baseline characteristics. Group number Group name
1 CO
2 DM
3 PAD-NGM
4 PAD-PREDM
5 PAD-DM
6 AS-DM
p All groups
n Age (years) Gender (m %) BMI (kg/m2 ) Cardiovascular events Mediasclerosis MetS (WHO) BPs (mmHg) BPd (mmHg) Creatinine clearance (ml/min) eGFR (ml/min/m2 ) Creatinine (mol/L) UACR (mg/g) Triglyceride (mmol/L) HDL-C (mmol/L) Total cholesterol (mmol/L) Lipoprotein a (mol/L) LDL-C (mmol/L) Bilirubin (mol/L) AP (U/L) ASAT (U/L) ALAT (U/L) CRP (mg/L) Glucose (mmol/L) HbA1c (rel. %)
39 42.1 ± 13.4 17 (44%) 26.5 ± 4.9 – – – n.a. n.a. 106.2 ± 30.8 86.3 ± 13.2 79.9 ± 15.1 n.a. 2.0 ± 1.3 1.4 ± 0.4 5.5 ± 1.2 n.a 3.2 ± 1.1 10.6 ± 6.2 64.3 ± 24.9 27.0 ± 16.5 25.6 ± 16.4 3.3 ± 4.8 4.8 ± 0.8 5.4 ± 0.4
85 60.9 ± 9.8 46 (54%) 30.9 ± 5.5 – – 20 (50%) 144 ± 21 86 ± 12 96.1 ± 48.2 75.8 ± 37.4 100.1 ± 40.8 171 ± 405.4 2.2 ± 1.4 1.3 ± 0.4 4.8 ± 1.1 n.a 2.6 ± 0.9 11.5 ± 6.8 79.4 ± 26.7 26.7 ± 8.7 27.5 ± 12.6 4.1 ± 6.6 9.6 ± 4.0 8.1 ± 1.7
71 68.3 ± 10.7 39 (55%) 25.8 ± 3.7 10 10 (19.6%) 0 138 ± 20 78 ± 14 71.8 ± 25.5 71.8 ± 16.9 90.2 ± 20.8 46.7 ± 181.2 1.8 ± 0.9 1.5 ± 0.3 5.3 ± 1.2 2.0 ± 2.1 3.1 ± 1.1 10.9 ± 3.7 79.2 ± 24.1 25.2 ± 6.2 23.7 ± 10.0 3.2 ± 2.3 5.0 ± 0.4 5.7 ± 0.3
90 68.5 ± 11.2 59 (66%) 27.7 ± 3.8 10 13 (21.7%) 50 (55.5%) 140 ± 21 80 ± 10 75.9 ± 28.6 69.9 ± 17.6 95.0 ± 22.1 35.6 ± 133.9 2.0 ± 1.2 1.4 ± 0.5 5.1 ± 1.1 2.1 ± 2.7 2.8 ± 0.9 11.9 ± 4.3 77.6 ± 23.8 26.9 ± 8.2 27.2 ± 14.1 5.3 ± 8.5 5.9 ± 0.4 5.8 ± 0.4
155 70.1 ± 9.6 108 (70%) 29.0 ± 4.8 25 47 (48%) 81 (64.8%) 146 ± 23 79 ± 11 73.7 ± 32.4 66.1 ± 22.4 104.4 ± 32.1 156.8 ± 385.1 2.2 ± 1.5 1.3 ± 0.3 4.8 ± 1.3 1.6 ± 2.0 2.6 ± 1.0 11.5 ± 4.3 85.7 ± 49.6 25.6 ± 9.3 27.9 ± 12.9 5.0 ± 5.5 8.1 ± 2.3 7.1 ± 1.2
20 61.5 ± 8.1 14 (70%) 30.8 ± 4.6 – – – 150 ± 23 86 ± 14 80.9 ± 38.9 61.1 ± 29.2 119.9 ± 40.8 314.1 ± 354.4 2.4 ± 1.0 1.2 ± 0.4 4.6 ± 0.6 n.a. 2.4 ± 0.7 n.a. n.a. n.a. n.a. 5.7 ± 6.5 n.a. 7.8 ± 1.3
<0.001* 0.015 <0.001* 0.191 <0.001* <0.001* 0.034 <0.001* <0.001* <0.001* <0.001* <0.001* 0.166 <0.001* 0.003 0.227 <0.001* 0.608 0.041 0.712 0.260 0.191 <0.001* <0.001*
Data are mean ± STD, or number (n) of individuals and percentages (%). PAD-NGM, peripheral arterial disease with a normal glucose metabolism; PAD-PREDM, PAD with impaired glucose tolerance or impaired fasting glucose; PAD-DM, PAD with type 2 diabetes mellitus; AS-DM, cerebrovascular or cardiovascular disease with DM but without PAD; na, not available; BMI, body-mass-index; MetS (WHO), metabolic syndrome according to the WHO classification; BPs, systolic blood pressure; BPd, diastolic blood pressure; eGFR, estimated glomerular filtration rate; UACR, urine albumin/creatinine-ratio; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; AP, alkaline phosphatase; ASAT, aspartate-aminotransferase; ALAT, alanine-aminotransferase; CRP, C-reactive protein; HbA1c, glycosylated haemoglobin. A p-value <0.05 was considered significant. * Indicates sustained significance (<0.05) after Bonferroni correction.
patients showed higher BMI (p < 0.001), systolic blood pressure (p = 0.011), albuminuria (p = 0.005), triglycerides (p = 0.006), CRP (p = 0.002), glucose and HbA1 C (p < 0.001) but lower HDL (p < 0.001), compared to PAD-NGM.
3.2. Serum YKL-40 levels in comparison to the different subject groups YKL-40 was elevated in DM patients (median [25–75 percentile]: 96 [57–171]; in PAD-NGM: 84 [53–128]; in PAD-PREDM: 114 [79–164], p < 0.001; in PAD-DM: 113 [71–157], p < 0.001; and in AS-DM: 100 [58–170], p = 0.002; compared to CO: 43 [30–80] ng/ml (Fig. 1). In PAD, YKL-40 was higher in patients with PREDM (p = 0.001) and DM (p = 0.01) compared to patients without glycemic dysregulation (Fig. 1). However between PAD-PREDM and PAD-DM group YKL-40 was not different (p = 0.311) (Fig. 1). PAD patients exhibited (independently of glucose status) higher YKL-40 values compared to healthy subjects: 103 [69–159] vs. 43 [30–80] ng/ml, p < 0.001 (Fig. 2a). In addition, YKL-40 was higher in patients with diabetes (irrespective of atherosclerosis) compared to non-diabetics: 100 [66–165] vs. 68 [39–114] ng/ml, p < 0.001 (Fig. 2b). Upon division of all patients in those with and without PAD, YKL-40 was higher in patients with PAD: 103 [69–159] vs. 85 [47–159] ng/ml, p = 0.003 (Fig. 2c). Next, we investigated whether YKL-40 is associated with wide spreading of atherosclerosis: Clearly YKL-40 was different in patients with no atherosclerosis vs. either PAD or CAD or cerebrovascular disease vs. dual combination of the latter vs. the triple combination: none vs. one vs. two vs. three affected vessel beds: 79 [44–154] vs. 98 [62–162] vs. 108 [70–156] vs. 145 [78–201] ng/ml, p = 0.004 (independent of glucose status).
Fig. 1. Depicted are log YKL-40 values in the respective groups: CO, controls; DM, type 2 diabetes mellitus; PAD-NGM, peripheral arterial disease with normal glucose metabolism; PAD-PREDM, peripheral arterial disease with pre-diabetes; PAD-DM, peripheral arterial disease with diabetes; AS-DM, atherosclerosis without PAD but with DM. All p-values shown are in between each group differences. Only the upper line resembles the in between all group difference.
3.3. YKL-40 and C-reactive protein YKL-40 and C-reactive protein (CRP) are both inflammatory proteins, the last being a classic hallmark of atherosclerosis. As published previously, PAD-DM (p = 0.002) and PAD-PREDM (p = 0.026) express higher CRP compared to PAD-NGM. We failed to obtain a difference between PAD-DM and PAD-PREDM (p = 0.708). YKL40 followed a similar pattern: whereas YKL-40 was elevated in
560
K. Batinic et al. / Atherosclerosis 222 (2012) 557–563
Fig. 2. Depicted are log YKL-40 values in the respective groups: CO, controls; PAD, peripheral arterial disease; AS, sites of atherosclerosis. All p-values shown are in between each group differences.
PAD-DM (p = 0.01) and PAD-PREDM (p = 0.001) compared to PADNGM, YKL-40 did not differ between PAD-DM and PAD-PREDM (Fig. 1). In addition, we report a correlation of YKL-40 and CRP (Table 2). Even more, CRP is an independent predictor of YKL-40 levels in the total study population (not shown) as well as in the PAD cohort (Table 3). 3.3.1. Univariate correlation analysis for identification of associates of YKL-40 To identify predictors of YKL-40, univariate correlation analysis was done (Table 2). Age is associated with YKL-40 in all groups except PAD-PREDM and AS-DM. Furthermore triglycerides correlated with YKL-40 in all patients, in CO, in PAD and in the subgroup PAD-PREDM, but not in patients with diabetes without atherosclerosis. In addition, an association of YKL-40 and aspartateaminotransferase in all patients, control, in all PAD patients, in PAD-DM subgroup, in non-diabetes, diabetes and non-PAD was observed. 3.3.2. Multivariate regression analysis for predictors of YKL-40 in PAD patients All variables significantly associated with YKL-40 in univariate fashion in PAD patients (n = 316) (Table 2: PAD), were taken into a multivariate stepwise backward regression analysis (Table 3): Model A (unadjusted) revealed age (beta = 0.230, p < 0.001), triglycerides (beta = 0.157, p = 0.006), ASAT (beta = 0.192, p = 0.001) and CRP (beta = 0.159, p = 0.001) as presumptuous independent predictors of YKL-40 (Table 3). Adjustment for gender (Model B) did not change latter associations with YKL-40. Further adjustment for renal parameters (creatinine clearance, UACR) (Model C), did not change associations of triglycerides, ASAT, and CRP, but the association of YKL-40 and age was attenuated (beta = 0.183, p = 0.026) (more than 10% beta change). Additional adjustment
for lipid parameters (HDL-C, total cholesterol, LDL-C, lipoprotein a) (Model D) improved the association of YKL-40 with age (beta = 0.208, p = 0.012) and triglycerides (beta = 177, p = 0.049) but did not change the association with ASAT (beta = 0.187, p = 0.001) and CRP (beta = 0.182, p = 0.002). Adjustment for liver parameters (bilirubin, AP, ALAT) (Model E) did have no impact on the associations. Moreover, additional adjustment for HOMA (Model F) did not alter the association of CRP with YKL-40, but significantly improved the association of age and triglycerides. In contrast, the association of ASAT with YKL-40 was not attenuated. Further adjustment for blood pressure (Model G) and BMI (Model H) did not alter the associations of our four explanatory variables with YKL-40 apart of a strengthening of age. Thus, we are able to state that in our cohort, after multivariate analysis and multiple adjustments, age, triglycerides, ASAT and CRP were identified as most independent predictors of YKL-40. 4. Discussion We are the first to demonstrate elevated levels of YKL-40 in patients with severe and generalized atherosclerosis. YKL-40 is significantly elevated in our cohort of patients with peripheral arterial disease (PAD), independently of the glucose status. Among patients with PAD, YKL-40 levels differed between patients with pre-diabetes or diabetes compared to normal glucose metabolism. In addition, we confirm that patients with diabetes without atherosclerosis exhibit higher YKL-40 compared to healthy controls [17–19]. Furthermore we are able to add, that in a small group of patients with diabetes and atherosclerosis (=coronary artery or cerebrovascular disease) but without PAD, YKL-40 levels are augmented as well. A previous study could only demonstrate a trend – most presumably because of small sample size [30]. Interestingly, in our study cohort, YKL-40 was correlated with the number of the
K. Batinic et al. / Atherosclerosis 222 (2012) 557–563 Table 3 Multivariate regression for predictors of YKL-40 in all PAD patients.
Table 2 Univariate associations of YKL-40 in the different patient groups. Group
Variable
R
p
All
Age Creatinine clearance Estimated glomerular filtration rate Urine albumin/creatinine-ratio Creatinine Triglyceride Lipoprotein a Alkaline phosphatase Aspartate-aminotransferase C-reactive protein Glucose Glycosylated haemoglobin Gender
0.338 −0.203 −0.184 0.152 0.183 0.144 −0.129 0.172 0.194 0.167 0.201 0.099 0.091
<0.001* <0.001* <0.001* 0.003* <0.001* 0.003* 0.032 0.001* <0.001* 0.002* <0.001* 0.038 0.050
Age Triglyceride Alkaline phosphatase Aspartate-aminotransferase Alanine-aminotransferase C-reactive protein Glucose Glycosylated haemoglobin Gender
0.508 0.443 0.479 0.438 0.517 0.368 0.566 0.516 0.352
0.001* 0.016 0.005* 0.014 0.003* 0.042 0.001* 0.004* 0.028
Age Urine albumin/creatinine-ratio
0.277 0.360
0.010* 0.011*
0.332 −0.425 −0.407 0.275 0.377
0.005* <0.001* <0.001* 0.023 0.001*
CO
DM
PAD-NGM
Age Creatinine clearance Estimated glomerular filtration rate Urine albumin/creatinine-ratio Creatinine
561
PAD-PREDM
Triglyceride C-reactive protein
0.263 0.259
0.013* 0.014*
PAD-DM
Age Aspartate-aminotransferase
0.179 0.258
0.026* 0.004*
AS-DM
Creatinine clearance Estimated glomerular filtration rate Creatinine
−0.447 −0.517 0.582
0.048 0.020* 0.007*
Non-diabetes
Age Creatinine clearance Estimated glomerular filtration rate Urine albumin/creatinine-ratio Creatinine Alkaline phosphatase Aspartate-aminotransferase Glucose Glycosylated haemoglobin
0.488 −0.402 −0.447 0.275 0.406 0.260 0.228 0.279 0.221
<0.001* <0.001* <0.001* 0.023 <0.001* 0.008 0.022 0.005* 0.029
Diabetes
Age Creatinine clearance Urine albumin/creatinine-ratio Aspartate-aminotransferase Alanine-aminotransferase
0.193 −0.146 0.192 0.261 0.184
0.002* 0.021 0.004* 0.001* 0.020
Non-PAD
Age Urine albumin/creatinine-ratio Creatinine Alkaline phosphatase Aspartate-aminotransferase Alanine-aminotransferase Glucose Glycosylated haemoglobin Body-mass-index
0.457 0.360 0.203 0.420 0.284 0.373 0.514 0.199 0.208
<0.001* 0.002* 0.021 <0.001* 0.023 0.002* <0.001* 0.024 0.014
PAD
Age Creatinine clearance Estimated glomerular filtration rate Urine albumin/creatinine-ratio Creatinine Triglyceride Lipoprotein a Aspartate-aminotransferase C-reactive protein
0.206 −0.194 −0.173 0.115 0.174 0.149 −0.129 0.164 0.154
<0.001* 0.001* 0.002* 0.046 0.002* 0.009 0.032 0.006* 0.008
CO, healthy subjects; DM, type 2 diabetes mellitus; PAD-NGM, peripheral arterial disease with a normal glucose metabolism; PAD-PREDM, PAD with impaired glucose tolerance or impaired fasting glucose; PAD-DM, PAD with type 2 diabetes mellitus. A p-value <0.05 was considered significant. * Indicates sustained significance (<0.05) after Bonferroni correction.
Models
Variable
Beta
p
Model A Unadjusted
Age Triglyceride Aspartate-aminotransferase C-reactive protein
0.230 0.157 0.192 0.159
<0.001 0.006 0.001 0.006
Model B + Adjusted for gender
Age Triglyceride Aspartate-aminotransferase C-reactive protein
0.239 0.151 0.189 0.165
<0.001 0.009 0.001 0.004
Model C + Adjusted for renal parameters
Age Triglyceride Aspartate-aminotransferase C-reactive protein
0.183 0.155 0.208 0.167
0.026 0.008 <0.001 0.004
Model D + Adjusted for lipid parameters
Age Triglyceride Aspartate-aminotransferase C-reactive protein
0.208 0.177 0.187 0.182
0.012 0.049 0.001 0.002
Model E + Adjusted for liver parameters
Age Triglyceride Aspartate-aminotransferase C-reactive protein
0.209 0.181 0.190 0.172
0.015 0.048 0.030 0.005
Model F + Adjusted for HOMA
Age Triglyceride Aspartate-aminotransferase C-reactive protein
0.245 0.221 0.181 0.174
0.005 0.018 0.038 0.005
Model G + Adjusted for blood pressure
Age Triglyceride Aspartate-aminotransferase C-reactive protein
0.252 0.221 0.176 0.174
0.005 0.018 0.046 0.005
Model H + Adjusted for BMI
Age Triglyceride Aspartate-aminotransferase C-reactive protein
0.272 0.216 0.177 0.178
0.004 0.021 0.045 0.005
HOMA, homeostasis model assessment; BMI, body-mass-index; p-value <0.05 was considered significant.
affected vessel beds. All those findings argue for a strong involvement of YKL-40 not only in diabetes and CAD but also in both, and potentially even more important, in PAD, which exhibits the worst prognosis of all clinical manifestations of atherosclerosis. Since our study is cross-sectional by nature we cannot elucidate whether YKL-40 is cause or consequence or even – but most unlikely – an innocent bystander of atherosclerosis. However, multivariate regression analysis revealed that YKL-40 was predicted by the four most independent predictors age, triglycerides, ASAT and CRP between 24.2% and 21.2% in the total study population (unadjusted; final adjusted model) as well as in PAD patients (14.0%/18.3%). Since all four variables are associated with different associates of atherosclerosis (age, lipid metabolism, nonalcoholic steatotic hepatitis, vascular inflammation) the possibility of chance is rendered unlikely. Recent data have demonstrated that YKL-40 is elevated in patients with CAD compared to controls [9]. Even more the extension of affected coronary artery vessels and progression of atherosclerosis are attended by higher YKL-40 values [9,13]. In our study we confirm that extension/severity of atherosclerosis is associated with higher YKL-40 levels. Whether YKL-40 is cause or consequence is subject of ongoing studies, however, activated macrophages in acute as well as in chronic diseases secrete large amounts of YKL-40 [6,7]. Moreover, the induction of release of YKL40 from macrophages was shown in early lesions of atherosclerosis [7]. In those in vitro studies YKL-40 was most highly concentrated deeply in the centre core of atherosclerotic lesions and the elevated secretion was most pronounced during foam cell development [31].
562
K. Batinic et al. / Atherosclerosis 222 (2012) 557–563
One of the major cardiovascular risk factors – diabetes – is associated with YKL-40: Rathcke et al. have demonstrated that YKL-40 is elevated in T1DM and T2DM in adults [17,19]. Possibly YKL-40 is already involved in the early development of disease, since an elevation of YKL-40 already in children with T1DM has been shown [22]. Although diabetes and obesity are often jointly observed as diabesity, an association of BMI and YKL-40 [18,21] is rare except gestational diabetes [20]. Herein we endorse the association of YKL40 and diabetes. Until now no data was available on YKL-40 in patients with a cooccurrence of diabetes and generalized atherosclerosis. We show for the first time that YKL-40 is elevated in patients with a combination of diabetes and a generalized form of atherosclerosis (PAD) compared to controls. Furthermore YKL-40 in patients with diabetes and atherosclerosis is also more elevated than in patients with atherosclerosis alone (PAD). Although the association of YKL-40 and atherosclerosis is so far stringent, a plausible relationship of YKL-40 to prediabetes has not been investigated. In our study, PAD patients with an impaired glucose metabolism have comparable elevations of YKL-40 levels as PAD patients with diabetes. Since impaired fasting glucose and impaired glucose tolerance have been demonstrated to be associated with an intermediate cardiovascular hazard ratio of 1.6/1.5 (95% CI: 1.0–2.4/1.1–2.0) between newly diagnosed diabetes and known diabetes [32], an intermediate augmentation of YKL-40 from no glucose disturbance to diabetes in our patients with prediabetes mimics the cardiovascular risk increase of those patients. Nevertheless our study is cross-sectional at the time being, and we cannot delineate an increase of cardiovascular risk. Atherosclerosis, pre-diabetes and diabetes are inflammatory conditions with elevations of responsible proteins such as CRP and IL-6 [33]. YKL-40 is elevated in diabetes and also in atherosclerosis. We add that YKL-40 is elevated in patients with atherosclerosis and concomitant diabetes. Furthermore, we first found an elevation of YKL-40 in pre-diabetes. In our cohort, YKL-40 and CRP follow a similar pattern: both are elevated in PAD-DM and PAD-PREDM vs. PAD-NGM; however we failed to obtain a difference in both proteins in between PAD-PREDM and PAD-DM. This suggests that both proteins increase with disease, but are most presumable differentially involved in different diseases. In accordance with our data, YKL-40 is described to be associated with CRP levels in CAD [9,10]. In contrary, no association of YKL-40 and CRP was found in patients with DM alone [17]. Recent studies demonstrated an association of YKL-40 and albuminuria in patients with DM [23], which is already known in patients with T1DM [19]. We can extend this observation to patients with atherosclerotic disease: PAD patients with albuminuria exhibited higher YKL-40. Bearing in mind, the in vivo and in vitro data for an involvement of YKL-40 in tissue remodeling, an association of YKL-40 and albuminuria – a strong remodeling inducer [34] – merits further investigation. Mediasclerosis is a tissue remodeling process related to vascular calcification [35]. Mediasclerosis is known to be a common appearance in DM patients. We confirm that mediasclerosis is more frequent in PAD-DM than in PAD-NGM (p = 0.001) or PADPREDM (p = 0.001). In our study we found an association of YKL-40 and mediasclerosis by demonstrating that patients with mediasclerosis have higher YKL-40 levels compared to patients without mediasclerosis: 135.1 [76.5–171.1 vs. 93.8 [53.2–153.9], p = 0.010 (median [25th–75th percentile]). Linear regression revealed an unadjusted association of YKL-40 with mediasclerosis (beta = 0.218, p = 0.008), which was attenuated after adjustment for diabetes (beta = 0.180, p = 0.032). Whether this independent 3.24% influence on the variation coefficient of YKL-40 is biologically relevant and involved in the development of mediasclerosis cannot be answered by our study. Nevertheless, YKL-40 is involved in tissue remodeling and fibrogenesis [5].
The limitation of our study is that for the time being, we could only examine our cohort study cross-sectionally. However, we found that YKL-40 is higher in PAD patients. In addition, patients with diabetes but also prediabetes exhibited higher values. Furthermore our data suggest that YKL-40 might be involved in mediasclerosis, a tissue remodeling process, whose causes are rarely understood. References [1] Millis AJ, Hoyle M, Reich E, Mann DM. Isolation and characterization of a Mr = 38.000 protein from differentiating smooth muscle cells. J Biol Chem 1985;260:3754–61. [2] Vos K, Steenbakkers P, Miltenburg AM, et al. Raised human cartilage glycoprotein-39 plasma levels in patients with rheumatoid arthritis and other inflammatory conditions. Ann Rheum Dis 2000;59:544–8. [3] Volck B, Price PA, Johansen JS, et al. YKL-40, a mammalian member of the bacterial chitinase family, is a matrix protein of specific granules in human neutrophils. Proc Assoc Am Physicians 1998;110:351–60. [4] Johansen JS, Bojesen SE, Mylin AK, et al. Elevated plasma YKL-40 predicts increased risk of gastrointestinal cancer and decreased survival after any cancer diagnosis in the general population. J Clin Oncol 2009;27:572–8. [5] Johansen JS, Christoffersen P, Møller S, et al. Serum YKL-40 is increased in patients with hepatic fibrosis. J Hepatol 2000;32:911–20. [6] Rehli M, Niller HH, Ammon C, et al. Transcriptional regulation of CHI3L1, a marker for late stages of macrophage differentiation. J Biol Chem 2003;278:44058–67. [7] Boot RG, van Achterberg TA, van Aken BE, et al. Strong induction of members of the chitinase family of proteins in atherosclerosis: chitotriosidase and cartilage gp-39 expression in lesion macrophages. Arterioscler Thromb Vasc Biol 1999;19:687–94. [8] Nishikawa KC, Millis AJ. gp38k (CHI3L1) is a novel adhesion and migration factor for vascular cells. Exp Cell Res 2003;287:79–87. [9] Kucur M, Isman FK, Karadag B, Vural VA, Tavsanoglu S. Serum YKL-40 levels in patients with coronary artery disease. Coron Artery Dis 2007;18:391–6. [10] Wang Y, Ripa RS, Johansen JS, et al. YKL-40 a new biomarker in patients with acute coronary syndrome or stable coronary artery disease. Scand Cardiovasc J 2008;42:295–302. [11] Nøjgaard C, Høst NB, Christensen IJ, et al. Serum levels of YKL-40 increases in patients with acute myocardial infarction. Coron Artery Dis 2008;19:257–63. [12] Mathiasen AB, Harutyunyan MJ, Jørgensen E, et al. Plasma YKL-40 in relation to the degree of coronary artery disease in patients with stable ischemic heart disease. Scand J Clin Lab Invest 2011;71:439–44. [13] Zheng JL, Lu L, Hu J, et al. Increased serum YKL-40 and C-reactive protein levels are associated with angiographic lesion progression in patients with coronary artery disease. Atherosclerosis 2010;21:590–5. [14] Mygind ND, Harutyunyan MJ, Mathiasen AB, et al. The influence of statin treatment on the inflammatory biomarkers YKL-40 and HsCRP in patients with stable coronary artery disease. Inflamm Res 2011;60:281–7. [15] Kastrup J, Johansen JS, Winkel P, et al. High serum YKL-40 concentration is associated with cardiovascular and all-cause mortality in patients with stable coronary artery disease. Eur Heart J 2009;30:1066–72. [16] Kjaergaard AD, Bojesen SE, Johansen JS, Nordestgaard BG. Elevated plasma YKL-40 levels and ischemic stroke in the general population. Ann Neurol 2010;68:672–80. [17] Rathcke CN, Johansen JS, Vestergaard H. YKL-40, a biomarker of inflammation, is elevated in patient with type 2 diabetes and is related to insulin resistance. Inflamm Res 2006;55:53–9. [18] Nielsen AR, Erikstrup C, Johansen JS, et al. Plasma YKL-40: a BMI-independent marker of type 2 diabetes. Diabetes 2008;57:3078–82. [19] Rathcke CN, Persson F, Tarnow L, Rossing P, Vestergaard H. YKL-40, a marker of inflammation and endothelial dysfunction, is elevated in patients with type 1 diabetes and increases with levels of albuminuria. Diabetes Care 2009;32:323–8. [20] Schaller G, Brix J, Placher-Sorko G, et al. YKL-40 concentrations are not elevated in gestational diabetes. Eur J Clin Invest 2010;40(4):339–43. [21] Hempen M, Kopp HP, Elhenicky M, et al. YKL-40 is elevated in morbidly obese patients and declines after weight loss. Obes Surg 2009;19:1557–63. [22] Hoertenhuber T, Hoellerl F, Hoebaus C, et al. YKL-40 is elevated in type 1 diabetic children which show components of metabolic syndrome compared to those without, as well as to healthy control subjects. Diabetologia 2009;52:S1–345. [23] Brix JM, Hoellerl F, Koppensteiner R, Schernthaner G, Schernthaner GH. YKL-40 in type 2 diabetic patients with different levels of albuminuria. Eur J Clin Invest 2011;41:589–96. [24] Norgren L, Hiatt WR, Dormandy JA, et al. Inter-society consensus for the management of peripheral arterial disease (TASC II). Eur J Vasc Endovasc Surg 2007;33(Suppl 1):S1–75. [25] Kallio M, Forsblom C, Groop PH, Groop L, Lepantalo M. Development of new peripheral arterial occlusive disease in patients with type 2 diabetes during a mean follow-up of 11 years. Diabetes Care 2003;26:1241–5. [26] Report of a WHO consultation. In: Alwan A, King H, editors. Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1: diagnosis and classification of diabetes mellitus. Geneva: World Health
K. Batinic et al. / Atherosclerosis 222 (2012) 557–563
[27]
[28]
[29] [30]
Organization, Department of Noncommunicable Disease Surveillance; 1999. p. 1–59. http://www.who.int. Lorant DP, Grujicic M, Hoebaus C, et al. Fetuin-A levels are increased in patients with type 2 diabetes and peripheral arterial disease. Diabetes Care 2011;34(Jan (1)):156–61. Ridker PM, Hennekens CH, Buring J, Rifai N. C-reactive protein and other markers of inflammation in the prediction of cardiovascular disease in women. N Engl J Med 2000;342:836–43. Rothman K, Greenland S. Modern epidemiology. Philadelphia: Lippincott Williams & Wilkins; 1998. Thomsen SB, Rathcke CN, Zerahn B, Vestergaard H. Increased levels of the calcification marker matrix Gla Protein and the inflammatory markers YKL-40 and CRP in patients with type 2 diabetes and ischemic heart disease. Cardiovasc Diabetol 2010;9:86.
563
[31] Fach EM, Garulacan LA, Gao J, et al. In vitro biomarker discovery for atherosclerosis by proteomics. Mol Cell Proteomics 2004;3:1200–10. [32] Barr E, Zimmet P, Welborn T, et al. Risk of cardiovascular and all-cause mortality in individuals with diabetes mellitus, impaired fasting glucose, and impaired glucose tolerance. Circulation 2007;116:151–7. [33] Pradhan AD, Manson JE, Rifai N, Buring JE, Ridker PM. C-reactive protein, interleukin 6, and risk of developing type 2 diabetes mellitus. JAMA 2001;286:327–34. [34] Hermans MM, Henry RM, Dekker JM, et al. Albuminuria, but not estimated glomerular filtration rate, is associated with maladaptive arterial remodeling: the Hoorn Study. J Hypertens 2008 Apr;26(4):791–7. [35] Demer LL, Tintut Y. Vascular calcification: pathobiology of a multifaceted disease. Circulation 2008;117(22):2938–48.