COOLIST (Cooling for Ischemic Stroke Trial)
A Multicenter, Open, Randomized, Phase II, Clinical Trial
Background and Purpose—Animal studies suggest that cooling improves outcome after ischemic stroke. We assessed the feasibility and safety of surface cooling to different target temperatures in awake patients with acute ischemic stroke.
Methods—A multicenter, randomized, open, phase II, clinical trial, comparing standard treatment with surface cooling to 34.0°C, 34.5°C, or 35.0°C in awake patients with acute ischemic stroke and an National Institutes of Health Stroke Scale score of ≥6, initiated within 4.5 hours after symptom onset and maintained for 24 hours. The primary outcome was feasibility, defined as the proportion of patients who had successfully completed the assigned treatment. Safety was a secondary outcome.
Results—Inclusion was terminated after 22 patients because of slow recruitment. Five patients were randomized to 34.0°C, 6 to 34.5°C, 5 to 35.0°C (cooling was initiated in 4), and 6 to standard care. No (0%), 1 (17%), and 3 (75%) patients, respectively, completed the assigned treatment (P=0.03). No (0%), 2 (33%), and 4 (100%) patients reached the target temperature (P=0.01). Pneumonia occurred in 8 cooled patients but not in controls (absolute risk increase, 53%; 95% confidence interval, 28–79%; P=0.002).
Conclusions—In awake patients with acute ischemic stroke, surface cooling is feasible to 35.0°C, but not to 34.5°C and 34.0°C. Cooling is associated with an increased risk of pneumonia.
Clinical Trial Registration—URL: http://www.trialregister.nl. Unique identifier: NTR2616.
Cooling is a promising new treatment for patients with acute ischemic stroke.1 In animal studies of acute ischemic stroke, the benefit of hypothermia was inversely related to the temperature reduction achieved, but infarct size was still reduced by 30% with cooling to 35.0°C.2
Few randomized, phase II, clinical trials have assessed the feasibility of cooling in awake patients with acute ischemic stroke,3–6 and cooling strategies varied considerably across these studies. The feasibility of different target temperatures has not been compared.
We assessed the feasibility and safety of surface cooling to 34.0°C, 34.5°C, and 35.0°C for 24 hours in awake patients with acute ischemic stroke on a stroke unit. The surface cooling protocol was comparable to that in the ongoing phase III study EuroHYP-1.7
This was a prospective, randomized, open, phase II, clinical trial with blinded end point assessment. The study design and complete eligibility criteria are included in the online-only Data Supplement. In brief, adult patients with a clinical diagnosis of acute ischemic stroke, a possibility to initiate cooling within 4.5 hours after stroke onset, and a National Institutes of Health Stroke Scale (NIHSS) score of ≥6 were randomized to standard care or to cooling to 34.0°C, 34.5°C, or 35.0°C in a 1:1:1:1 ratio.
Cooling was started as soon as possible and within 4.5 hours after stroke onset by means of intravenous infusion of 20 mL/kg cooled normal saline (4°C) over 30 to 60 minutes. Surface cooling was initiated immediately after the start of the infusion and continued for 24 hours. The detailed cooling and rewarming strategies are included in the online-only Data Supplement.
Shivering was assessed with the 4-point Columbia Shivering Scale. The antishivering regime is included in the online-only Data Supplement.
The primary outcome measure was feasibility of cooling, defined as the proportion of patients who completed the 24 hours of cooling on the assigned target temperature. Secondary outcome measures included time to target temperature and safety. Tertiary outcome measures were functional outcome at 3 months and tolerability.
The χ2 test, independent t test, ANOVA, Mann–Whitney U test, or Kruskal–Wallis was used to compare continuous data between the treatment groups where appropriate. Analyses were by intention to treat (baseline and functional outcome) or per protocol (cooling parameters and adverse events).
We included 22 patients in 3 centers (Table I in the online-only Data Supplement); mean age, 63 years (SD 12); 19 (86%) male; median score on the NIHSS, 13 (range 7–23). Five patients were randomized to 34.0°C, 6 to 34.5°C, 5 to 35.0°C, and 6 to standard care. Baseline characteristics are shown in Table II in the online-only Data Supplement. Patients randomized to 35.0°C had a higher median score on the NIHSS than other cooled patients. One patient who was allocated to cooling to 35.0°C completely recovered after randomization but before initiation of cooling. Cooling was therefore not started. All patients were followed-up until 90 days. The trial was stopped in January 2015 because of slow recruitment.
No (0%) patients randomized to 34.0°C, 1 (17%) to 34.5°C, and 3 (75%) to 35.0°C in whom cooling was initiated, successfully completed the assigned treatment (P=0.03; Table). Completeness of assigned treatment was different between 35.0°C on the one hand and 34.5°C or 34.0°C on the other (P=0.04 and P=0.003, respectively; Figure).
No (0%) patients randomized to 34.0°C, 2 (33%) to 34.5°C, and 4 (100%) to 35.0°C in whom cooling was initiated, reached their respective target temperatures (P=0.01; Table). Reaching target temperature was different between 35.0°C and 34.0°C (P=0.02; Figure).
Median times to target temperature varied widely between patients and did not differ between groups (Table).
Adverse events are presented in Table III in the online-only Data Supplement. Shivering occurred in all cooled patients, mainly between 2 and 6 hours after initiating of cooling (Figure I in the online-only Data Supplement). Pneumonia occurred in 8 cooled patients and in none of the patients with standard treatment (absolute risk increase, 53%; 95% confidence interval, 28%–79%) and was diagnosed at a mean interval of 43 hours (SD 37) after stroke onset. The occurrence of pneumonia was not associated with NIHSS at admission.
There was no difference in functional outcome or mortality at 3 months between the groups (Table IV and Figure II in the online-only Data Supplement). Tolerability results are presented in Table V in the online-only Data Supplement.
This study suggests that surface cooling to 35.0°C for 24 hours is feasible in awake patients with acute ischemic stroke, whereas cooling to 34.0°C or 34.5°C is not. Cooling is associated with an increased risk of pneumonia. Our results may have implications for the ongoing phase III trial EuroHYP-1.7
COOLIST (Cooling for Ischemic Stroke Trial) is the only study that has compared different target temperatures in a randomized fashion. Previous studies have shown the feasibility of surface cooling in awake patients with acute ischemic stroke with target temperatures to 35.5°C for 6 hours3 and to 35.0°C for 12 hours.6
Shivering occurred in all our patients and was a common side effect in all previous cooling trials involving awake patients,3–6 despite antishivering regimes. Shivering could be the most important reason for not reaching target temperature and might partially explain the differences between the treatment groups.
Pneumonia is a common side effect in patients treated with hypothermia for any indication,8 possibly because of a detrimental effect on inflammatory mechanisms such as secretion of proinflammatory cytokines, leukocyte migration, and phagocytosis. Further, awake stroke patients treated with hypothermia might be prone to pneumonia because the antishivering regime entails nausea and mild sedation, leading to an increased risk of dysphagia and (micro)aspiration. Although pneumonia did not affect functional outcome after 3 months in our series, infections in stroke have been associated with a poor functional outcome.9 Therefore, therapeutic benefits of hypothermia might be decreased by the increased risk on pneumonia.
Our analyses are limited by the premature termination of the study and by the small number of patients recruited. One reason for slow recruitment was a competing trial in all but one of the participating centers. Second, the institutional review board of the initiating center (University Medical Center Utrecht) had significant concerns before approval, mostly on the presumed risks of arrhythmia and respiratory insufficiency. Therefore, the approval of the trial was granted under specific conditions, including the continuous presence of a trial physician in the hospital during the 24 hours of cooling and one-on-two nursing. These conditions could often not be met.
Because we included patients with an NIHSS score of ≥6, our results do not apply to patients with a less severe stroke. The study was not designed to assess efficacy outcomes, and the results of the ongoing large phase III trial EuroHYP-17 have to be awaited.
We thank P. Greebe, B. Zweedijk, and E. Poromaa, research-nurses, for functional outcome assessments; C. McGill for hosting the online randomization service; and the members of the Data Safety Monitoring Board (P.M.W. Bath [chair, Nottingham, United Kingdom], J. Horn [Amsterdam, The Netherlands], and C. Weir [Edinburgh, United Kingdom]) for their supervision of the safety of the trial.
Sources of Funding
Drs Geurts and van der Worp are supported by the Dutch Heart Foundation (2010B239 and 2010T075). COOLIST (Cooling for Ischemic Stroke Trial) was supported by the Dutch Heart Foundation (2010B239). Medivance (later BARD) provided Arctic Sun systems for the duration of the trial at no costs and cooling pads at reduced costs. Medivance and BARD were not involved in the design of the study, analyses, or reporting of the results.
Guest Editor for this article was James C. Grotta, MD.
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.116.014757/-/DC1.
- Received July 12, 2016.
- Revision received October 1, 2016.
- Accepted October 10, 2016.
- © 2016 American Heart Association, Inc.
- van der Worp HB,
- Macleod MR,
- Kollmar R
- van der Worp HB,
- Sena ES,
- Donnan GA,
- Howells DW,
- Macleod MR
- Kammersgaard LP,
- Rasmussen BH,
- Jørgensen HS,
- Reith J,
- Weber U,
- Olsen TS
- Hemmen TM,
- Raman R,
- Guluma KZ,
- Meyer BC,
- Gomes JA,
- Cruz-Flores S,
- et al
- Piironen K,
- Tiainen M,
- Mustanoja S,
- Kaukonen KM,
- Meretoja A,
- Tatlisumak T,
- et al
- van der Worp HB,
- Macleod MR,
- Bath PM,
- Demotes J,
- Durand-Zaleski I,
- Gebhardt B,
- et al