Aims Jugular venous distension is a classical sign of heart failure (HF) but it can be difficult to assess clinically.

Methods and results Outpatients with HF and control subjects were assessed. Internal jugular vein diameter (JVD) was measured using a linear high-frequency ultrasound probe (10 MHz) at rest, after a Valsalva manoeuvre and during deep inspiration. JVD ratio was calculated as the maximum diameter during Valsalva to that measured at rest. 311 patients (mean age 71 years, mean left ventricular EF 42%, median (IQR) amino-terminal pro-brain natriuretic peptide 979 (441–2007) ng/L) and 66 controls were included. JVD (median and IQR range) at rest was smaller in controls (0.16 (0.14–0.20) cm) than in patients with HF (0.23 (0.17–0.33) cm; p<0.001) but similar during Valsalva (1.03 (0.90–1.16) cm vs 1.08 (0.90–1.25) cm; p=0.28). Consequently, JVD ratio was greater in controls (6.3 (4.9–7.6)) than in patients (4.5 (2.9–6.1); p<0.001). During a median follow-up of 516 (IQR 335–622) days, 48 patients (15%) with HF died or were hospitalised for HF. In multivariable models, among clinical, echocardiographic or biochemical variables, only increasing NT-proBNP and ultrasound assessment of internal jugular vein were independently associated with prognosis. Comparing top and bottom tertiles of JVD ratio (2.3 (IQR 1.7–2.9) versus 6.8 (6.1–7.7)), the tertile with lower values had a 10-fold greater risk of an adverse event (HR 10.05, 95% CI 3.07 to 32.93).

Conclusions Ultrasound assessment of the internal jugular vein identifies outpatients with HF who have a higher risk of an adverse outcome.

Clinical trial registration NCT01872299.