Résumé
Background: the COVID-19 crisis has shown that we need to optimize sedation and analgesia management for patients treated with Neuro-Muscular Blocking Agents (NMBA) to ensure optimal patient’s comfort without wasting sedative and analgesic drugs in Intensive Care Units (ICU) in a context of drugs shortage. Also, optimizing sedation/analgesia management is mandatory to prevent drugs overuse, delayed awakening and prolonged utilization of ICU resources. There is no recommendation regarding the assessment of pain in ICU patients receiving NMBA. This study evaluated the validity of different physiological parameters to better assess pain in pharmacologically paralyzed ICU patients.Methods: four electrophysiological parameters (instant-Analgesia Nociception Index (ANI), Bispectralindex (BIS), Heart Rate (HR) and Mean Arterial Blood Pressure (MAP)) and one clinical physiological parameter (Behavioural Pain Scale (BPS)) were recorded before, during and after tracheal suctioning in all consecutive patients admitted in a medical-surgical ICU who required a sustained use of cisatracurium, as well as just after the complete paralysis recovery measured by the Train-of-Four ratio using an automatized accelometer. The validity of the five pain-related parameters was assessed comparing the values recorded during different situations (at rest before, during, and after thenociceptive procedure) (discriminant validation method), and the effect of paralysis was assessed comparing values obtained during and after paralysis (parameters’ reliability).Results: twenty patients were included for analysis. ANI, BIS and HR significantly changed during the nociceptive procedure in both paralyzed and non paralyzed patients, while BPS changed only in non paralyzed patients. ANI and HR were the only two parameters that were not significantly modified by paralysis, contrary to BIS and BPS (mixed effect model for repeated measurements). ANI showed the highest discriminant validity with values (min 0, max 100) decreasing from 71 [48-89] at rest to 41 [25-72] during the procedure in paralyzed patients and from 71 [53-85] at rest to 40 [31-52] in non paralyzed patients.Conclusions: ANI is the most accurate parameter to detect pain in both paralyzed and non paralyzed sedated ICU patients. Its decrease during tracheal suctioning is significant, clinically relevant, and not modified by the use of NMBA in ICU patients.