Objectives: To compare the use of an unenhanced high-resolution time-of-flight MR angiography sequence (Hr-TOF MRA) with fat-suppressed axial/coronal T1-weighted images and contrast-enhanced angiography (standard MRI) for the diagnosis of cervical artery dissection (cDISS). Methods: Twenty consecutive patients (9 women, 11 men, aged 24-66 years) with proven cDISS on standard MRI underwent Hr-TOF MRA at 3.0 T using dedicated surface coils. Sensitivity (SE), specificity (SP), positive and negative predictive values (PPV, NPV), Cohen's kappa (k) and accuracy of Hr-TOF MRA were calculated using the standard protocol as the gold standard. Image quality and diagnostic confidence were assessed on a four-point scale. Results: Image quality was rated better for standard MRI (P = 0.02), whereas diagnostic confidence did not differ significantly (P = 0.27). There was good agreement between Hr-TOF images and the standard protocol for the presence/absence of cDISS, with k = 0.95 for reader 1 and k = 0.89 for reader 2 (P < 0.001). This resulted in SE, SP, PPV, NPV and accuracy of 97 %, 98 %, 97 %, 98 % and 97 % for reader 1 and 93 %, 96 %, 93 %, 96 % and 95 % for reader 2. Conclusions: Hr-TOF MRA can be used to diagnose cDISS with excellent agreement compared with the standard protocol. This might be useful in patients with renal insufficiency or if contrast-enhanced MR angiography is of insufficient image quality. Key Points: • New magnetic resonance angiography sequences are increasingly used for vertebral artery assessment. • A high-resolution time-of-flight sequence allows the diagnosis of cervical artery dissection. • This technique allows the diagnosis without intravenous contrast medium. • It could help in renal insufficiency or when contrast-enhanced MRA fails.