Microalbuminuria Is Associated with An Increased Risk of Venous Thromboembolism. A Novel Risk Marker for Venous Thromboembolism

Introduction: Microalbuminuria (albuminuria 30–300 mg/24h) is a well known risk marker for arterial thromboembolism (ATE). It has been postulated that microalbuminuria is a reflection of generalized endothelial dysfunction that is associated with, among others, increased levels of several coagulation factors. The impact of coagulation disorders is more evident in the pathogenesis of venous thromboembolism (VTE), than in the pathogenesis of ATE. Therefore, we hypothesized that microalbuminuria might be an independent risk factor for VTE.

Methods: In 1997, all inhabitants of the city of Groningen, aged between 28 and 75 years (n=85,421) were sent a postal questionnaire collecting information about risk factors for cardiovascular disease and a vial to collect a first morning urine sample for measurement of urinary albumin concentration (UAC). Of responded subjects (40,856), a cohort (8,592 subjects) enriched for higher levels of UAC was formed that completed screening at an outpatient clinic. Screening data were collected on albuminuria (assessed in two 24-h urine collections), and risk factors for atherosclerosis and renal disease. To identify subjects with VTE, we used the databases of the regional anticoagulation clinic, the national registry of hospital discharge diagnoses and the national registry of death certificates. Of identified subjects patient charts were drawn. Only symptomatic and objectively verified venous thromboembolic events were considered established. We evaluated the association between albuminuria and the incidence of VTE, using Cox-regression models.

Results: Of 8592 subjects, 18 subjects were excluded because of missing data on albuminuria. Of the remaining 8574 subjects (mean age±SD, 49±13 years; 50% male), 129 subjects experienced VTE during a follow-up period of 8.6±1.8 years. In univariate analyses albuminuria, age, hypertension, smoking, body mass index (BMI), eGFR, triglycerides, C-reactive protein, tissue plasminogen activator, tissue plasminogen activator inhibitor 1 and cancer were associated with VTE (P<0.05). After adjustment for the aforementioned factors, albuminuria of 10–19 (n=2,134), 20–29 (n=608), 30–300 (n=1,144) and >300 mg/24h (n=134) conferred hazard ratios of 1.35 (95% CI, 0.83–2.19, P=0.23), 1.90 (1.02–3.53, P=0.04), 2.49 (1.56–3.99, P<0.0001) and 3.27 (1.367–7.85, P=0.008), respectively, as compared to albuminuria <10 mg/24h (n=4,554). Besides albuminuria, age and BMI were independently associated with VTE.

Conclusions: For the first time, we report that microalbuminuria and even high normal albuminuria (i.e. 20–29 mg/24h) are independently associated with an increased risk for VTE. Given the high prevalence of microalbuminuria in the general population (7%), the general population attributable risk of VTE associated with microalbuminuria is probably comparable to the risk conferred by factor V Leiden mutation. In addition to cardiovascular and renal risk conveyed by microalbuminuria, our findings give further credence to population-based screening for microalbuminuria and its subsequent treatment.