Rapid Induction of COOLing in Stroke Patients (iCOOL1): a randomised pilot study comparing cold infusions with nasopharyngeal cooling

INTRODUCTION Induction methods for therapeutic cooling are under investigated. We compared the effectiveness and safety of cold infusions (CI) and nasopharyngeal cooling (NPC) for cooling induction in stroke patients. METHODS A prospective, open-label, randomised (1:1), single-centre pilot trial with partially blinded safety endpoint assessment was conducted at the neurointensive care unit of Heidelberg University. Intubated stroke patients with an indication for therapeutic cooling and an intracranial pressure (ICP)/temperature brain probe were randomly assigned to CI (4°C, 2L at 4L/h) or NPC (60L/min for 1 h). Previous data suggested a maximum decrease of tympanic temperature for CI (2.1L within 35 min) after 52 min. Therefore the study period was 1 hour (15 min subperiods I-IV). The brain temperature course was the primary endpoint. Secondary measures included continuous monitoring of neurovital parameters and extracerebral temperatures. Statistical analysis based on repeated-measures analysis of variance. RESULTS Of 221 patients screened, 20 were randomized within 5 months. Infusion time of 2L CI was 33 ± 4 min in 10 patients and 10 patients received NPC for 60 min. During active treatment (first 30 min), brain temperature decreased faster with CI than during NPC (I: -0.31 ± 0.2 versus -0.12 ± 0.1°C, P = 0.008; II: -1.0 ± 0.3 versus -0.49 ± 0.3°C, P = 0.001). In the CI-group, after the infusion was finished, the intervention no longer decreased brain temperature, which increased after 3.5 ± 3.3 min. Oesophageal temperature correlated best with brain temperature during CI and NPC. Tympanic temperature reacted similarly to relative changes of brain temperature during CI, but absolute values slightly differed. CI provoked three severe adverse events during subperiods II-IV (two systolic arterial pressure (SAP), one shivering) compared with four in the NPC-group, all during subperiod I (three SAP, one ICP). Classified as possibly intervention-related, two cases of ventilator failure occurred during NPC. CONCLUSIONS In intubated stroke patients, brain cooling is faster during CI than during NPC. Importantly, contrary to previous expectations, brain cooling stopped soon after CI cessation. Oesophageal but neither bladder nor rectal temperature is suited as surrogate for brain temperature during CI and NPC. Several severe adverse events in CI and in NPC demand further studying of safety. TRIAL REGISTRATION ClinicalTrials.gov NCT01573117. Registered 31 March 2012.