Osteoprotegerin levels are increased in patients with venous thromboembolic disease

In the last few years a series of clinical studies showed a high prevalence of atherosclerotic disease and increased risk of arterial events in patients with a history of unprovoked venous thromboembolism (VTE) [1,2]. However, the mechanisms underlying the association between arterial and venous disease are far frombeingclarified.Theosteoprotegerin(OPG)/receptor activator of nuclear factor-kappa B (RANK)/RANK ligand (RANKL) axis was first identified for its central role during bone remodeling. In particular OPG, acting as a decoy receptor for RANKL, prevents the RANKL–RANK interaction which drives osteoclasts s maturation and bone resorption. More recently a series of clinical and basic science studies highlighted the involvement of the OPG/RANK/RANKL triad during vascular disease progression. In particular, OPG emerged both as inhibitor of vascular calcification and a potent predictor of atherosclerotic disease [3,4]. To date, no clinical information is available about a possible connection between the OPG/ RANK/RANKL axis and unprovoked VTE events. For this reason, we designed a case–control study and compared OPG serum levels observed in a well-characterized cohort of VTE patientswith thosemeasured inagroupofage-andsex-matched controls (selected fromhealthyvolunteers, health carepersonnel of the Hospital, blood donors and referred patients without evidence/historyofVTE). Inparticular,we enrolled110patients who were referred between October 2010 and July 2011 to the Thrombosis Unit of the Cà Foncello Hospital in Treviso and had had at least one objective diagnosed episode of a deep venous thrombosis (DVT) and/or a pulmonary embolism (PE). We excluded from the study patients with a provokedVTE and patientswithaVTE<3 monthsapart fromtheenrollmentvisit. Demographic data of the study population are summarized in Table 1. Subject s characteristics are presented as mean ± standard deviation (SD) or median ± interquartile range (IQR) depending on parameter distribution. Comparison between groups for continuous variables was performed using Student s t-test for normally distributed parameters and Mann–Whitney U-test for variables without normal distribution. The chi-square test was used for comparison of categorical variables. All statistical analyzes were performed using PAWS statistics 18.0 (SPSS Inc., Chicago, IL, USA). Patients with previous objectively documented VTE and controls showed similar lipid profile, systolic blood pressure (SBP), diastolic blood pressure (DBP), glomerular filtration rate (eGFR) (calculated throughMDRDequation) and blood glucose (Table 1). Compared with controls, VTE patients had a significantly greater mean body mass index (BMI) (26.8 ± 4.6 vs. 25.4 ± 3.8, P = 0.018). In detail, obesity was more prevalent in patients with VTE without reaching significant difference (P = 0.12), overweight subjects were equally distributed between cases and controls, whereas individuals with a normal BMI (<25 kg m) were more prevalent in the control group (50% vs. 36.4%, P = 0.041). Active smoking habit and prevalence of hypertension were similar between the two groups (P = 0.38 and P = 0.58 respectively), whereas the frequency of diabetes in cases and controls was low and modestly higher in patients with VTE compared with controls (P = 0.08). No differences were observed between cases and controls in high sensitivity Creactive protein (hs-CRP) serum levels (IMMULITE, Siemens Healthcare Diagnostics Products Ltd., Llanberis, Gwynedd, UK). Viceversa, a significant increase in OPG serum levels (R&DSystem Inc.,Minneapolis,MN,USA)was documented inpatientswithVTEcomparedwith controls. Inparticular, the median OPG levels were 1100.4 (IQR 888.5–1342.9) pg mL in cases and 800.5 (IQR 646.1–1026.8) pg mL in control subjects (P < 0.0001) (all samples were measured in duplicate and averaged). The coefficient of variation for OPG, as judged from duplicate measurements, was 7.9%. A correlation analysis conducted on the whole population demonstratedapositive, significant correlationbetweenOPG levels, age (r = 0.256, P < 0.001), glycemia (r = 0.148, P = 0.029) and HDL-C (r = 0.175, P = 0.01). However, after adjustment for BMI, lipid levels, blood pressure levels and blood glucose, only age remained significantly correlated withOPGlevels (P = 0.003).Wealso analyzedOPG levels in different subgroups of patients with VTE. In particular, we Correspondence: Marcello Rattazzi, Department of Medicine, University of Padova Medicina Interna I^, Ospedale Ca Foncello, Via Ospedale, 31100 Treviso, Italy. Tel.: +39 498211867; fax: +39 498754179. E-mail: marcello.rattazzi@unipd.it