Objective. Invasively measured arterial blood pressure (ABP) is associated with complications, while non-invasively measured ABP is generally considered risk-free. This study aimed to investigate the reliability of non-invasive ABP measured using finger-cuff volume-clamp device compared to invasive ABP measured by an arterial catheter in patients with aneurysmal subarachnoid haemorrhage (SAH). Approach. In 30 patients admitted for neurointensive care with SAH, invasive and non-invasive ABP were recorded simultaneously. Reliability was assessed for mean, diastolic and systolic ABP separately using intraclass correlation coefficient (ICC) agreement for each full period and each 3 s average. Main results. A median of 3 (IQR: 2-3) periods were included for each participant. The full periods (n = 81) showed an ICC of 0.34 (95% CI: 0.14-0.52), 0.31 (95% CI: 0.10-0.49), and 0.20 (95% CI: 0.00-0.39) for mean, diastolic, and systolic ABP, respectively. Three-second averages (n = 33 786) for mean (ICC: 0.35; 95% CI: 0.33-0.36), diastolic (ICC: 0.25; 95% CI: 0.25-0.28), and systolic ABP (ICC: 0.26; 95% CI: 0.18-0.33) yielded similar findings. Pearson's correlation coefficient showed an R 2 of 0.15 (p < 0.001), 0.15 (p < 0.001), 0.06 (p = 0.027) for mean, diastolic and systolic ABP, respectively. Significance. In patients with SAH, non-invasive measurement of ABP using the widely used Nano system from Finapres Medical Systems - a finger-cuff volume-clamp device (Finapres, Chennai, India) showed poor reliability and therefore cannot be used interchangeably with invasively measured ABP.