Objective: To assess the eligibility for replacement of invasive blood pressure as measured "within" the arterial vessel (IBP) with non-invasive continuous arterial blood pressure (cNIP) monitoring during total intravenous anaesthesia (TIVA), the ability of cNiP to track fast blood pressure changes needs to be quantified. A new method of statistical data analysis is developed for this purpose. Methods: In a pilot study on patients undergoing neurosurgical anaesthesia, mean arterial pressure MAP_IBP measured with IBP was compared to MAP_CNP measured by the CNAP Monitor 500 in ten patients (age: 63±13 a). Correlation analysis of changes of device differences ΔeMAP=ΔMAP_CNP-ΔMAP_IBP with changes of MAP_IBP (ΔMAP_IBP) during intervals of vasoactivity was conducted. An innovative technique, of linear trend analysis (LTA) applied to two signals, is described to perform this analysis without a priori knowledge of intervals of vasoactivity. Results: Analysis of ΔeMAP during vasoactivity revealed that ΔMAP_CNP systematically underestimated ΔMAP_IBP by 37%. This was confirmed in the complete data set using LTA technique showing a systematic, yet patient specific, underestimation in tracking ΔMAP_IBP (16...120%). Conclusion: The proposed LTA technique is able to detect systematic errors in tracking short-term blood pressure changes otherwise masked by established analysis. LTA may thus be a useful tool to assess the eligibility of cNIP to replace IBP during TIVA.