Osteopontin as a novel prognostic marker in stable ischaemic heart disease: a 3-year follow-up study


Panagiota Georgiadou, Attikon University General Hospital, 1 Rimini Str., 12462 Athens, Greece. Tel.: +30 2109493372; fax: +30 210 9493373; e-mail: pg171@hlk.forthnet.gr


Eur J Clin Invest 2010; 40 (4): 288–293


Objectives  Osteopontin (OPN) is a glycoprotein, which may play a major role in the regulation of biological phenomena. Increased levels of OPN have been linked to the presence and to the severity of atherosclerosis. This study was undertaken to assess the prognostic significance of plasma OPN levels in patients with stable ischaemic heart disease (IHD).

Methods  In 101 patients with stable IHD and angiographically documented significant coronary artery stenosis, plasma OPN levels were measured at baseline (time of coronary arteriography). Patients were prospectively followed for a median time of 3 years (minimum 2·25, maximum 3·9 years). The primary study endpoint was the composite of cardiovascular death, non-fatal myocardial infarction, need for revascularization and hospitalization for cardiovascular reasons.

Results  Baseline lnOPN levels were directly related to age (r = 0·27, P < 0·001) and inversely to left ventricular ejection fraction (r = −0·32, P < 0·01). Left ventricular ejection fraction was an independent predictor of plasma OPN levels after adjustment for age and gender (β = −0·013, P = 0·02). Median OPN value was 55 ng mL−1. In the univariate Cox-regression analysis, OPN levels > 55 ng mL−1 (n = 50) were significantly related to adverse cardiac outcome (HR = 2·40, 95% CI: 1·11–5·23, P = 0·027). In multivariate model, OPN levels > 55 ng mL−1 remained statistically significant independent predictor of adverse outcome after adjustment for age, gender, left ventricular ejection fraction and the number of diseased coronary arteries (HR = 2·88, 95% CI: 1·09–7·58, P = 0·032).

Conclusion  OPN may provide significant prognostic information independent of other traditional prognostic markers in patients with stable IHD.


Angina pectoris is a common symptom in patients with ischaemic heart disease (IHD) and constitutes a major symptom for those patients referred for myocardial percutaneous or surgical revascularization [1,2]. In contrast to patients with acute ischaemic syndromes, there is a lack of biomarkers in patients with stable IHD; thus, there is a lack of biomarkers to better stratify those patients [3,4].

Osteopontin (OPN), a highly acidic multifunctional glycoprotein, is a key player in essential biological phenomena, such as inflammation, autoimmune disease progression, bone remodelling and cell metastasis [5,6]. Although OPN was first isolated from mineralized bone matrix, it can also be synthesized by several other types of cells, including cardiac myocytes, vascular endothelial cells and fibroblasts. OPN exists either as an immobilized extracellular matrix molecule or as a soluble cytokine [7]. OPN has been found in human atherosclerotic plaques in the aorta, in the carotid and in the coronary arteries, and it has been shown to be implicated in the development and progression of atherosclerosis [8]. Increased levels of OPN have been related to the presence and extent of coronary artery disease and to restenosis after percutaneous coronary revascularization [9–12].

Based on available information, the present study was undertaken to investigate prospectively the prognostic significance of plasma OPN levels in patients with stable IHD.


Patient evaluation

Among the 334 consecutive patients who underwent diagnostic coronary angiography during the 4 months of recruitment – from 1 April 2004 to 31 July 2004 – 232 patients with stable IHD were screened for the study. Ninety-one patients were excluded (see exclusion criteria) and 40 refused to participate. The remaining 101 patients (86 men and 15 women; mean age: 67 ± 10) were prospectively investigated. Stable IHD was defined as the presence of stable angina pectoris with angiographic confirmation of significant coronary artery stenosis. The diagnosis of stable angina was based on the presence of chest pain that did not change its pattern during the preceding 6 months. Left ventricular ejection fraction was > 40% in all patients and was measured by contrast ventriculography or echocardiography.

Exclusion criteria were acute ischaemic syndrome, heart failure, cardiomyopathy, acute or chronic inflammatory disease, immunologic disease, major surgical procedure, osteoporosis and/or administration of any vitamin supplement for at least 6 months prior to the study [13,14]. The study was approved by the Ethics Committee of the Hospital; informed consent was obtained from all patients prior to entry into the study.

Diabetes mellitus was defined as fasting glucose level ≥ 126 mg dL−1 (7·0 mmol L−1) or if the patient was treated with insulin or oral hypoglycaemic agents. Hypertension was defined as blood pressures 140/90 mmHg or current use of antihypertensive drugs. Hypercholesterolaemia was defined as total cholesterol level > 240 mg dL−1 (6·2 mmol L−1), and/or current use of lipid-lowering agents. Smokers were defined as those currently smoking any tobacco product. A family history of IHD was defined as documented evidence of premature coronary artery disease in a close relative (men < 55 and women < 65 years of age). History was obtained and physical examination was performed in all patients. All patients completed a questionnaire that provided information about past medical history, risk factors of IHD and current medication use.

Coronary arteriography was performed using the Judkings technique and angiograms were recorded by the cineangiogram system, Philips, Netherlands. Significant coronary artery stenosis was considered, when a diameter narrowing > 50% was present. Left ventricular ejection fraction was estimated by left ventriculogaphy in 73 patients and by echocardiography in 28 patients; echocardiography was performed within the last 3 months prior to the study.

Biochemical evaluation was performed with usual standard techniques.

Osteopontin measurements

In the morning (the same day when coronary angiography was performed), fasting blood samples were collected prior to angiography and promptly centrifuged at 1400 g at 4 °C for 7 min within 30 min of the collection. Plasma aliquots were stored at −80 °C until analysis. OPN levels were determined by an enzyme-linked immunoabsorbent assay (ELISA) using the quantikine, human OPN kit (from R&D Systems, Wiesbaden-Nordenstadt, Germany), and are expressed in ng mL−1. The average of duplicate readings for each standard, control and samples was used. The standard curve was created using a four parameter logistic curve fit. Samples were usually diluted 25 times. The minimum detectable level of OPN was 0·011 ng mL−1. The intra- and inter-assay coefficients of variations were 4% and 7% respectively.

Follow-up and study end points

Follow-up consisted of telephonic interview or review of medical records. Median follow-up was 3 years (minimum 2·25, maximum 3·9 years). The time of cardiovascular event was considered the end of follow-up for the patients who had such an event. The primary study endpoint was the composite of cardiovascular death, non-fatal myocardial infarction, need for revascularization and hospitalization for cardiovascular reasons. The diagnosis of myocardial infraction was based on the presence of typical chest pain, diagnostic electrocardiographic changes, and/or increase of cardiac enzymes. Revascularization was defined as coronary bypass surgery or percutaneous revascularization performed during the follow-up period because of symptoms or signs of myocardial ischaemia.

Statistical analysis

Mean values were compared using the Student’s t-test; the assumption of t-test was evaluated by the Shapiro–Wilk test for normality. OPN levels were natural log-transformed because their distribution was not normal. Pearson correlation coefficient r or multivariate linear regression was performed to identify the linear correlations between continuous variables. The Hazard ratios (HR) and the 95% confidence intervals (CI) of continuous or categorical variables were calculated by univariate and multivariate Cox regression analysis. The multivariate Cox regression model based on the sample of 101 patients achieved 81·3% power at 0·05 significance level to detect a HR equal to 2·88. Survival analysis was performed by applying the Kaplan–Meier method. A P-value < 0·05 was considered statistically significant. Data were analysed using STATA 9·1 College Station, TX, USA.


The baseline demographic and clinical characteristics of the study population are shown in Table 1. Plasma OPN levels in the study cohort ranged from 21 to 204 ng mL−1 with a median value of 55 ng mL−1. There was no difference in baseline OPN levels between smokers and non-smokers (58·2 ± 25·3 vs. 67·7 ± 37·4, P = 0·21). Statins, the type of statin or BMI had no influence on OPN levels (P = 0·84, P = 0·31 and P = 0·64 respectively). Baseline lnOPN levels were correlated directly with age (r = 0·27, P < 0·001) and inversely with ejection fraction (r = −0·32, P < 0·01). Multiple linear regression analysis revealed that ejection fraction (β = −0·013, P = 0·02) was an independent predictor of baseline plasma OPN levels after adjustment for age and gender.

Table 1.   Baseline demographic and clinical characteristics of the study population Thumbnail image of

Seventy per cent of the patients were in class II or III according to the Canadian Cardiovascular Society (CCS) and only 8% of the patients were in class I. CCS classification had no influence on OPN levels (P = 0·49). Eighty-eight per cent of the patients had a positive stress test, which also had no influence on OPN levels (P = 0·83), either.

During the follow-up period, the composite study endpoint was reached in 31 of the 101 patients (31%), (Fig. 1). Three patients died, five had myocardial infarction, 18 had revascularization (surgical or percutaneous) and 21 were hospitalized because of acute ischaemic syndrome.

Figure 1.

 Kaplan–Meier survival plots for adverse cardiovascular events according to the median value of OPN levels.

The median value of baseline OPN (55 ng mL−1) was used to categorize patient population. In the univariate Cox regression analysis, OPN levels > 55 ng mL−1 (n = 50) were significantly related to adverse cardiac outcome (HR = 2·40, 95% CI: 1·11–5·23, P = 0·027) (Table 2). In multivariate Cox regression analysis, OPN levels > 55 ng mL−1 remained a statistically significant independent predictor of adverse outcome after adjustment for age, gender, ejection fraction and the number of significant coronary artery stenoses (HR = 2·88, 95% CI: 1·09–7·58, P = 0·032).

Table 2.   Univariate Cox regression analysis of future adverse cardiovascular events
 HR95% CIP-value
  1. HR, Hazard ratio; CI, confidence interval; IHD, ischaemic heart disease; ACE, angiotensin-converting enzyme; OPN, osteopontin.

  2. *P < 0·05 indicates statistical significance.

Age (years)1·010·98–1·050·49
Male gender0·920·35–2·410·87
Family history of IHD0·980·47–2·030·96
Previous myocardial infarction1·180·52–2·690·68
Number of diseased vessels1·581·00–2·500·047*
Ejection fraction (%)0·960·93–1·000·07
 OPN > 55 ng mL−12·401·10–5·230·027*


The present study has shown that plasma OPN levels constitute a prognostic indicator in patients with stable IHD, independent of other established classical prognostic markers including age, gender, left ventricular function and the number of coronary artery stenosis. Furthermore, the results suggested that age and left ventricular ejection fraction were independent predictors of baseline plasma OPN levels.

Despite the small number of patients, OPN proved to be an independent prognostic indicator. The small number of patients, however, did not allow focusing only on hard endpoints such as cardiovascular death and/or myocardial infarction. For this reason, need for revascularization and hospitalization for cardiovascular reasons were included in the analysis. The underlying mechanisms relative to plasma OPN levels and future cardiovascular events in patients with stable IHD are not completely understood. It is known that OPN stimulates proliferation of smooth muscle cells, migration of endothelial cells and recruitment of macrophages [15], all of which are directly related to atherosclerotic process. Experimental studies have shown that in the setting of enhanced oxidative stress, OPN is upregulated, while high OPN expression enhances oxidative stress resulting in vascular lesion formation [16,17]. Furthermore, an association was found between OPN and malondialdehyde levels in low-risk patients with stable IHD, independent of traditional vascular risk factors such as age, hypertension and diabetes mellitus [18]. Previous studies have shown that OPN may also play a role in the restenosis post-percutaneous coronary intervention injury [10,19]. These data indicate that OPN may be involved in inflammatory process, which promotes atherogenesis and instability in atherosclerotic plaques.

Studies have shown that in experimental animal models with heart failure, OPN is upregulated [20–22]. In addition, Rosenberg et al., have shown that circulating OPN plasma levels provided significant prognostic information in patients with chronic heart failure [12]. In agreement with these data are the findings from the present study, where OPN levels were significantly and inversely correlated with left ventricular ejection fraction. However, even though left ventricular systolic function is a predictor of cardiac death, the results of this study have shown that OPN levels were associated with cardiovascular events independent of the left ventricular ejection fraction. Thus, OPN may provide additional information beyond that given by the left ventricular function in patients with stable IHD. Furthermore, multivariate analysis model indicated that OPN did not diminish the predictive value of the left ventricular ejection fraction, suggesting that these two parameters are independent prognostic indicators in stable IHD.

The OPN may reflect the role of coronary calcification during the progression of atherosclerosis. OPN belongs to the family of bone mediators, like osteoprotegerin, which were shown to act as bridge between cells and minerals, promoting the formation of dystrophic calcification within the vessel walls [23]. Coronary calcification has been associated with atherosclerotic lesions of increasing cardiovascular risk [24]. These results support the particular exciting hypothesis that vascular calcification is an actively regulated process and it is associated to inflammation.

The results of this study are in agreement with that reported by Minoretti et al., who have shown a statistically significant trend towards higher cardiovascular event rates in those patients with higher OPN, when patients were subdivided according to OPN quartiles [11]. In this study, the prognostic significance of OPN was independent of C-reactive protein, whereas in our study, OPN provides strong and incremental prognostic information independent of the conventional risk factors of age, gender, left ventricular function and the number of coronary artery stenosis [11]. In our study, there was no association between OPN levels and the functional assessment of the patients suggesting that the mechanisms of the myocardial inducible ischaemia may not be involved in OPN expression. We also did not observe any association between the use of statins or any of their type and the levels of OPN probably because of a small sample size.

It should be emphasized that OPN is not specific for the heart and it has been shown to be elevated in other conditions [25,26]. Thus, it will be critical to interpret elevated OPN levels in the appropriate clinical context and in concert with other clinical biomarkers and laboratory findings [27]. In our study, the predictive value of OPN was independent from the number of diseased vessels; however, it is important to confirm this result using other means of evaluating coronary lesion complexity such as the Syntax score. Although OPN levels in healthy subjects had been previously reported, control data would have allowed a better interpretation mainly of the baseline OPN results. OPN reported concentrations may vary depending on ethnic diversities, different clinical cohorts studied or different methodologies used. Differences in analytic performance of the available OPN assays may also have affected the results. However, different assays appear to have a similar capacity to rank patients according to plasma OPN levels [28].

In conclusion, circulating levels of plasma OPN may provide useful prognostic information, independent of other traditional risk prognostic indicators in patients with stable IHD and documented significant coronary artery stenosis. Larger studies with follow up are needed to confirm the prognostic power of OPN and its potential in advancing the clinical management of this cohort of patients.


This work was supported by EPAN grant YB_22 from the Secretariat for R&D.


2nd University Department of Cardiology, Medical School, Attikon University General Hospital, Athens, Greece (P. Georgiadou, E. K. Iliodromitis, F. Kolokathis, C. Varounis, V. Gizas, D. T. Kremastinos); Center of Basic Research, Biomedical Research Foundation, Academy of Athens, Athens, Greece (M. Mavroidis, Y. Capetanaki); Center of Clinical Research, Biomedical Research Foundation, Academy of Athens, Athens, Greece (H. Boudoulas).