Aims: Congestion is a cardinal feature of untreated heart failure (HF) and might be detected by ultrasound (US) before overt clinical signs appear. Methods and Results: We investigated the prevalence and clinical associations of subclinical congestion in 238 patients with at least one clinical risk factor for HF (diabetes, ischaemic heart disease or hypertension) using three US variables: 1) inferior vena cava (IVC) diameter; 2) jugular vein distensibility (JVD) ratio (the ratio of the jugular vein diameter during Valsalva to that at rest); 3) the number of B-lines from a 28-point lung US. US congestion was defined as IVC diameter >2.0 cm, JVD ratio <4.0 or B-lines count >14. The prevalence of subclinical congestion (defined as at least one positive US marker of congestion) was 30% (13% by IVC diameter, 9% by JVD ratio and 13% by B-line quantification). Compared to patients with no congestion on US, those with at least one marker had larger left atria and higher plasma concentrations of natriuretic peptide (NP). Patients with raised plasma NT-proBNP/BNP had a lower JVD ratio (7.69 vs. 8.80; P=0.05) and more often had at least one lung B-line (74% vs. 63%; P=0.05). However, plasma NP concentrations were more closely related to left atrial volume than other US measures of congestion. Conclusions: Subclinical evidence of congestion by ultrasound is common in patients with clinical risk factors for heart failure. Whether these measurements provide additional value for predicting the development of HF and its prevention deserves consideration.