Thermogard XP for therapeutic hypothermia after cardiac arrest

Study component

Description

Objectives/hypotheses

To compare the neurological outcomes, effectiveness and adverse events of surface and endovascular cooling techniques in cardiac arrest patients.

Study design

Retrospective, observational cohort study with propensity‑matched analysis.

Setting

24 teaching hospitals across South Korea contributing data to the Korean Hypothermia Network registry, from 2007 to 2012.

Inclusion/exclusion criteria

Inclusion criteria: all adult cardiac arrest patients treated with therapeutic hypothermia using surface or endovascular cooling techniques.

Exclusion criteria: traumatic cardiac arrest, patients receiving both surface and endovascular cooling techniques in combination.

To adjust for differences in the baseline characteristics of patients undergoing each cooling method, 1 to 1 matching was performed using the propensity score.

Primary outcomes

Proportion of patients with poor neurological outcome at discharge, defined as Cerebral Performance Categories 3 to 5.

Statistical methods

Categorical variables were compared using the chi‑square test or Fisher's exact test, where appropriate. The normality of the continuous variables was verified by the Shapiro–Wilk test. Thereafter, variables were expressed as the mean ± standard deviation and were compared using Student's t test, after confirming homogeneity of variance with Levene's test.

For propensity score matching, multivariate logistic regression was used to model the dichotomous outcome of the surface or the endovascular cooling group for 803 patients in the study. The logistic model demonstrated a sufficient ability to differentiate between the two groups (c statistic=0.8). One‑to‑one matching with the propensity score used the Greedy‑matching macro^a.

After propensity score matching, the success of the propensity score modelling was assessed by the standardised difference, and the balance of the two groups was evaluated using Student's t test for continuous variables and the Chi‑square test or Fisher's exact test for categorical variables.

After estimating the propensity scores, a logistic regression analysis was performed, to determine the prognosis factor (mortality, poor outcome).

A probability value of p<0.05 was considered significant.

Patients included

803 patients were included in the analysis: 559 in the surface cooling group and 244 in the Thermogard XP cooling group.

After propensity score matching, 379 patients were excluded from the surface cooling group and 64 patients were excluded from the Thermogard XP cooling group, leaving 180 in each arm for the matched analysis.

Results

With regard to the overall cohort patient baseline characteristics, the authors reported no differences in age, sex or underlying disease between the two study groups.

After propensity score matching, there was no significant difference in neurological outcome or hospital mortality between the surface cooling and Thermogard XP groups (OR: 1.26, 95% CI: 0.81 to 1.96, p=0.31 and OR: 0.85, 95% CI: 0.55 to 1.30, p=0.44 respectively).

With regard to cooling efficacy, both groups attained the target temperature in a similar amount of time.

The rates of some adverse events were significantly increased in the surface cooling group compared with the Thermogard XP cooling group.

Conclusions

In the matched patient cohort, no significant differences in the rates of poor neurological outcomes and hospital mortality were observed between the surface cooling and Thermogard XP cooling groups.

Overcooling, rebound hyperthermia, rewarming‑related hypoglycaemia and rewarming‑related hypotension were significantly increased in the surface cooling group compared with the Thermogard XP cooling group, although these complications were not associated with the surface cooling method using hydrogel pads (which employ automatic temperature feedback regulation).

Abbreviations: CI, confidence interval; n, number of patients; OR, odds ratio.

^a Parsons LS. (2001) Reducing bias in a propensity score matched-pair sample using Greedy matching techniques. Accessed 6 Jan 2015.

Thermogard XP

Various surface cooling devices with automatic temperature feedback systems

Analysis

Randomised

Not applicable

Not applicable

Efficacy

n=180/244

n=180/559

Primary outcome: Poor neurological outcome

65.0% (117/180)

70.0% (126/180)

OR 1.26 (95% CI 0.81 to 1.96)

p=0.31

Selected secondary outcomes:

Hospital mortality

38.3% (69/180)

34.4% (62/180)

OR 0.85 (95% CI 0.55 to 1.30)

p=0.44

Induction time to target temperature (minutes)

209.4 ± 15.4

235.3 ± 18.0

OR 1.13 (95% CI 0.79 to 1.62)

p=0.51

Safety:

Overcooling

7.8% (14/180)

17.8% (32/180)

OR 2.56 (95% CI 1.32 to 4.99)

p=0.01

Safety:

Bradycardia

8.3% (15/180)

9.0% (16/180)

OR 1.09 (95% CI 0.52 to 2.27)

p=0.83

Patients reporting serious adverse events:

Sepsis in critical care

8.4% (15/180)

7.2% (13/180)

OR 0.85 (95% CI 0.39 to 1.84)

p=0.68

Pneumonia in critical care

35.4% (63/180)

31.8% (57/180)

OR 0.85 (95% CI 0.55 to 1.32)

p=0.48

Abbreviations: CI, confidence interval; n, number of patients; OR, odds ratio.

Study component

Description

Objectives/hypotheses

The aim of this study was to investigate the efficacy and safety of endovascular cooling in consecutively admitted survivors of cardiac arrest.

Study design

Retrospective, comparative cohort study with Bayesian approach.

Setting

Single tertiary teaching hospital in Austria. Retrospective clinical data from August 1991 to November 2004 were prospectively collated in a registry format.

Inclusion/exclusion criteria

Eligibility criteria:

Exclusion criteria:

The hypothermia group consisted of patients treated with CoolGard 3000 with the Icy catheter, with or without administration of infused cold fluids in the induction phase of hypothermia.

All other patients served as controls.

For the safety analysis, adverse events for the CoolGard 3000 group were predefined and recorded prospectively. Controls were selected by frequency matching. Matching criteria were witnessed out‑of‑hospital ventricular fibrillation cardiac arrest of presumed cardiac cause with duration of cardiac arrest longer than 1 minute. In the control group, a chart review was conducted to identify all adverse events. Arrhythmias and bleeding events were recorded between admission and 32 hours. Pneumonia, sepsis, acute renal failure and pancreatitis were recorded within the first 7 days.

Primary outcomes

Survival at 30 days after cardiac arrest.

Neurological performance at 30 days using the cerebral performance category (CPC) score.

Statistical methods

Continuous variables were given as mean ± SD if they were normally distributed or as median and the range between the 25th and 75th quartile if they were not. Nominal data were given as counts and percentage of total number.

Univariate comparisons between groups were conducted with a χ2 test or Fisher's exact test when indicated.

For safety analyses, contingency tables and Fisher's exact tests were used.

Patients included

1038 patients met the inclusion criteria, 97 were treated with CoolGard 3000 and 941 served as controls.

Of the 97 patients who were cooled with CoolGard 3000, 41 received additional cold Ringer's lactate during induction of hypothermia. As this might have influenced the calculated cooling rate of CoolGard 3000, the authors restricted the cooling efficacy analysis to patients who received endovascular cooling alone. Therefore, data from 56/97 CoolGard 3000 patients contributed to the analysis of cooling rate.

In the matching process for the safety analysis, there were 62/97 patients in the CoolGard 3000 group and 104/941 in the control group.

Results

Regardless of treatment, 51% (533/1038) of patients survived for 30 days and 36% (371/1038) of patients survived for 30 days and had favourable neurological recovery (CPC 1 or 2).

In the univariate analysis, patients in the CoolGard 3000 group had 2‑fold increased odds of survival (67/97 patients versus 466/941 patients; odds ratio 2.28, 95% CI, 1.45 to 3.57; p<0.001).

After adjustment for baseline imbalances in a multivariate model, the odds ratio was only slightly attenuated (odds ratio 1.96, 95% CI, 1.19 to 3.23; p=0.008).

The model had an acceptable fit (Hosmer‑Lemeshow χ2=7.39, df=8; p=0.495). When discounting the observational data in a Bayesian analysis by using a sceptical prior the posterior odds ratio was 1.61 (95% credible interval, 1.06 to 2.44).

Conclusions

Treatment of unselected patients after resuscitation from cardiac arrest with CoolGard 3000 significantly reduced mortality and improved favourable neurological recovery at 30 days or discharge as compared with controls from a retrospective resuscitation database.

There was no statistically significant difference in adverse events between the CoolGard 3000 group and the control group, except for transient bradycardia.

The cooling rate of 1.2°C/hour in this study was faster than external cooling with a water‑filled cooling blanket or with cold air, based on previous studies in the literature.

Abbreviations: °C/h, degrees Centigrade per hour; CI, confidence interval; CPC, cerebral performance category score; df, degrees of freedom; n, number of patients; OR, odds ratio; SD, standard deviation; χ2, Chi‑square test.

CoolGard 3000 with Icy catheter

Standard post‑resuscitation therapy

Multivariate Analysis

Randomised

Not applicable

Not applicable

Efficacy

n=97

n=56/97^a

n=941

n=466/941^b

Primary outcome: Survival to 30 days

69% (67/97)

50% (466/941)

OR 1.96 (95% CI 1.19 to 3.23)

p=0.008

Selected secondary outcomes:

Survival to 30 days, or until discharge, with Good Neurological Recovery^b

53% (51/97)

34% (320/941)

OR 2.56 (95% CI 1.57 to 4.17)

p<0.001

Cooling rate

(°C/h) [Interquartile range, IQR]

1.2 [0.7–1.5]

Not applicable

-

Safety

n=62

n=104

-

Patients reporting serious adverse events

Bradycardia

15% (9/62)

3% (2/104)

p=0.025

Abbreviations: °C/h, degrees Centigrade per hour; CI, confidence interval; CPC, cerebral performance category score; n, number of patients; OR, odds ratio; χ2, Chi‑square test.

^a. n=97 for reported primary clinical outcomes; n=56 for reported cooling rate.

^b. Good neurological recovery in this study was defined as patients with a Cerebral Performance Category (CPC) score of 1 or 2 (the patient is at least alert and has sufficient cerebral function to live independently and work part‑time).

Study component

Description

Objectives/hypotheses

The authors hypothesised that the clinical benefit of mild therapeutic hypothermia is greater when achieved more rapidly.

Study design

Prospective case series.

Setting

Single academic, regional, tertiary care hospital in Germany. Recruitment dates from June 2004 to February 2006.

Inclusion/exclusion criteria

Eligibility criteria:

CA was defined as the absence of a palpable pulse and spontaneous respiration with ventricular fibrillation (VF), sustained pulseless ventricular tachycardia (VT) or asystole as the initial cardiac rhythm.

Patients were excluded if they had malignancy or another terminal illness, pre‑existing brain damage or a severe neurologic deficit (dementia, previous stroke, advanced/severe Parkinson's syndrome).

Primary outcomes

The primary endpoint was neurological outcome, based on defined 'cerebral performance categories:' Good outcomes related to no/mild cerebral disability (cerebral performance categories 1 and 2) and poor outcomes included severe disability, coma/vegetative state and brain death (cerebral performance categories 3–5).

Association of the following time intervals from cardiac arrest to mild therapeutic hypothermia with neurological outcomes:

Statistical methods

Stepwise forward logistic regression with calculation of odds ratios (OR) to examine the effects of the cooling time variables on the primary endpoint (good neurological outcome).

In addition, the following variables were included to control for potential confounding factors:

Comparisons of continuous variables were made using the Mann–Whitney U‑test. Categorical variables were compared by the use of a χ2 (Chi square) test or Fisher's exact test.

All data are presented as median [interquartile range]. All tests are two‑tailed.

A probability value of p<0.05 was considered statistically significant.

Patients included

49 consecutive patients meeting the eligibility criteria were enrolled.

Results

Mild therapeutic hypothermia was maintained adequately for 24 hours in all patients.

Complications such as sepsis, bleeding disorder, pneumonia, or thrombocytopenia did not occur in any patient during CoolGard 3000 cooling. There were no deaths during the CoolGard 3000 use.

The time to cooling ranged between 35 minutes and 6 hours.

Both time to target temperature (TTT) and time to coldest temperature (TCT) ranged between 2 hours 15 minutes and 23 hours 35 minutes.

TCT was the only cooling variable with predictive value for the neurological outcome (OR for good outcome: 0.73, 95% CI 0.45–0.98; p=0.013).

The target T of 33°C could not be achieved in 11 patients.

Conclusions

Early achievement of mild therapeutic hypothermia is a determinant of the final neurological outcome and thus measures to speed up the initiation of cooling therapy after cardiac arrest appear warranted.

Abbreviations: BMI, body mass index; CA, cardiac arrest; CI, confidence interval; CPR, cardiopulmonary resuscitation; n, number of patients; OR, odds ratio; T, temperature; TCT, time to coldest temperature; TTT, time to target temperature; VF, ventricular fibrillation; VT, ventricular tachycardia; χ2, Chi‑square test.

^a Brain Resuscitation Clinical Trial I Study Group. (1986) A randomized clinical study of cardiopulmonary-cerebral resuscitation: design, methods and patient characteristics. American Journal of Emergency Medicine 4: 72–86.

CoolGard 3000

Good neurological outcome

(No/mild cerebral disability)

CoolGard 3000

Poor neurological outcome

(Severe disability, coma/vegetative state, brain death)

Analysis

Randomised

Not applicable

Not applicable

Efficacy

n=28

n=21

Primary outcome: Time to cooling [interquartile range] (minutes)

150 [114–199]

150 [98–198]

p=0.762

Primary outcome: Time to target temperature (TTT) [interquartile range] (minutes)

334 [250–498]

450 [322–674]

p=0.071

Primary outcome: Time to coldest temperature (TCT) [interquartile range] (minutes)

443 [280–543]

555 [425–985]

p=0.035

Selected secondary outcomes:

Maximum serum neurone specific enolase (NSE) as a biochemical marker of brain damage [interquartile range] (µg/l)

18.7
[15.7–25.7]

58.8
[32.5–184.2]

p<0.001

Safety

Not applicable

Not applicable

-

Patients reporting serious adverse events

No events reported

No events reported

-

Abbreviations: n, number of patients; NSE, neurone specific enolase; TCT, time to coldest temperature; TTT, time to target temperature.

CoolGard

Various surface cooling techniques

Analysis

Study 1 – Intravascular versus surface cooling speed and stability after cardiopulmonary resuscitation (de Waard et al. 2014)

Design

Retrospective comparative study comparing the efficacy of CoolGard to non‑invasive water‑circulating body wraps (Medi‑Therm, with automatic temperature feedback control)

Randomised (if applicable)

Not applicable

Not applicable

Efficacy

n=97

n=76

Primary outcomes:

Cooling efficacy

Mean cooling speed [interquartile range] (°C/h)

0.65
[0.45–0.83]

0.7
[0.34–0.90]

p=0.95

Mean temperature (°C)

33.1 ± 0.3

32.5 ± 0.5

p<0.0001

Mean CV temperature [interquartile range] (%)

0.35
[0.18–0.60]

0.85
[0.50–1.29]

p<0.0001

Selected secondary outcomes:

Glasgow Coma Scale at discharge from ICU

15
[3–15]

10
[4–13]

p=0.008

Length of stay (LOS) in ICU (hours)

144
[96–192]

172
[102–304]

p=0.08

Safety

n=97

n=76

–

Patients experiencing serious adverse events

None reported

None reported

–

Study 2 – Invasive versus non‑invasive cooling after in- and out‑of‑hospital cardiac arrest: a randomized trial (Pittl et al. 2013)

Design

Prospective, randomized, single centre study comparing the efficacy of CoolGard to the non‑invasive ArcticSun surface cooling system (hydrogel pads with automatic temperature feedback control)

Randomised

n=40

n=40

Efficacy

n=39

n=39

Primary outcomes:

Neurological function

Neurological outcome at 72 hours assessed by NSE values [interquartile range] (ng/ml)

19.0
[11.0–42.0]

16.5
[11.8–46.5]

p<0.0001

Survival with good neurological outcome during hospitalisation

35.9% (14/39)

35.9% (14/39)

p=0.99

Selected secondary outcomes:

Cooling efficacy

Cooling starting after sudden cardiac arrest [interquartile range] (mins)

242
[165–275]

180
[155–245]

p=0.13

Cooling rate [interquartile range] (°C/h)

1.3 [0.7–1.6]

1.0 [0.6–1.3]

p=0.29

Time to target temperature after the start of mild induced hypothermia (mins)

180
[150–330]

240
[180–390]

p=0.29

Target temperature maintenance (°C)

33.0
[32.9–33.0]

32.7
[32.4–32.9]

p<0.001

Safety

Not reported

Not reported

–

Patients experiencing serious adverse events

None reported

None reported

Other safety endpoints, such as infections or therapy requiring arrhythmia did not differ significantly.

Bleeding complication

17.9% (7/39)

43.6% (17/39)

p=0.03

Study 3 – Comparison of intravascular and conventional hypothermia after cardiac arrest (Knapik et al. 2011)

Design

Prospective, non‑randomised comparative study, comparing the efficacy of CoolGard with a traditional cooling using uncontrolled surface cooling, ice‑cold intravenous fluids and ice‑cold gastric lavage (without automatic temperature feedback control).

Randomised

Not applicable

Not applicable

Efficacy

n=20

n=21

Primary outcomes:

Cooling efficacy

Urinary bladder temperature of <34°C reached

95.0% (19/20)

52.4% (11/21)

p=0.004

Mean time to achieve temperature of <34°C (hours)

4.0 ± 3.2

6.3 ± 4.3

Not significant

Stable temperature (32–34°C) during hypothermia reached

80% (16/20)

14.3% (3/20)

p<0.001

Selected secondary outcomes:

Periods of inadequate cooling above 34°C

20% (6/20)

71.4% (15/21)

p=0.013

Periods of temperature overshoot below 32°C

0% (0/20)

19.1% (4/21)

Not significant

Safety

Not reported

Not reported

–

Patients experiencing serious adverse events

None reported

None reported

–

Study 4 – A comparison of intravascular and surface cooling techniques in comatose cardiac arrest survivors (Tømte et al. 2011)

Design

Single centre observational comparative study, comparing the efficacy of CoolGard to the non‑invasive ArcticSun surface cooling system (hydrogel pads with automatic temperature feedback control)

Randomised

Not applicable

Not applicable

Efficacy

n=75

n=92

Primary outcomes:

Cooling efficacy

Median initiation of cooling to 34°C [interquartile range] (mins)

188 [86–256]

170 [83–240]

p=0.391

Median time for maintained hypothermia [interquartile range] (hours)

24 [24–24]

24 [24–24]

p=0.431

Selected secondary outcomes:

Clinical outcomes

Survival to final hospital discharge with good neurological function (CPC 1‑2)

45% (34/75)

38% (34/90)

p=0.326

Median time in intensive care unit [interquartile range] (hours)

130 [74–213]

156 [74–213]

p=0.431

Safety

Not reported

Not reported

–

Patients experiencing serious adverse events

Sustained hyperglycaemia (>8mmol/l) of >4 consecutive hours

48% (36/75)

70% (64/92)

p=0.005

Low serum magnesium (<0.7 mmol/l)

37% (27/74)

18% (16/87)

p=0.010

Study 5 – Therapeutic hypothermia after cardiac arrest: A retrospective comparison of surface and endovascular techniques (Gillies et al. 2010)

Design

Retrospective comparative study comparing the efficacy of CoolGard to that of surface cooling systems including conventional ice packs, a cooling mattress and tent (Theracool, 2 patients, without automatic temperature feedback control), and a cold water recirculation system (CritiCool, 1 patient, with automatic temperature feedback control).

Randomised

Not applicable

Not applicable

Efficacy

n=42

n=41

Primary outcomes:

Cooling efficacy

Mean time to target temperature (hours)

5.2 ± 3.3

6.1 ± 4.8

p=0.29

Mean time at target temperature (hours)

22.4 ± 6.1

17.5 ± 12.3

p=0.02

Average temperature fluctuation (°C) over 10–20 hours of maintenance phase

1.0 ± 0.8

1.7 ± 1.3

p=0.003

Selected secondary outcomes:

Clinical outcomes

ICU mortality

38.1% (16/42)

51.2% (21/41)

p=0.27

Poor neurological outcome

57.1% (24/42)

61.0% (25/41)

p=0.82

Safety

Not reported

Not reported

–

Patients experiencing serious adverse events

91% (38/42)

85% (35/41)

p=0.52

Overcooling

10% (4/42)

27% (11/41)

p=0.049

Target temperature not reached

7% (3/42)

24% (10/41)

p=0.04

Bleeding

14% (6/42)

2% (1/41)

p=0.11

Study 6 – Efficacy of and tolerance to mild induced hypothermia after out‑of‑hospital cardiac arrest using an endovascular cooling system (Pichon et al. 2007)

Design

Prospective single‑arm cohort study evaluating the efficacy of CoolGard.

Randomised

Not applicable

Not applicable

Efficacy

n=40

Not applicable

Primary outcome:

Cooling efficacy

Time to target temperature of 33°C after initiation of MIH [interquartile range] (mins)

187 ± 119 [30–600]

Not applicable

–

Patients in which hypothermia was maintained

91% (31/34)*

Not applicable

*6 patients died during mild induced hypothermia

Selected secondary outcomes:

Safety

Not reported

Not applicable

–

Patients experiencing serious adverse events

Not reported

Not applicable

–

Bleeding of any severity

8% (3/36)

Not applicable

–

Complication at catheter site

8% (3/36)

Not applicable

–

Nosocomial infection – septicaemia

13% (5/36)

Not applicable

–

Study 7 – Comparison of external and intravascular cooling to induce hypothermia in patients after CPR (Flemming et al. 2006)

Design

Non-randomised comparative study comparing the efficacy of CoolGard to that of conventional cooling (Theracool, without automatic temperature feedback control, and additional use of cooling blankets and cold infusions when necessary).

Randomised

Not applicable

Not applicable

Efficacy

n=31

n=49

Primary outcome:

Cooling efficacy

Achieved target temperature of 33°C

100% (31/31)

9% (4/49)

Not reported

Time to achieve target temperature (hours)

3.48 ± 0.6

9.2 ± 1.2*

Not reported

* In this group a mean minimum temperature of 34.8°C was achieved after the onset of cooling

Selected secondary outcomes:

Clinical outcomes

Length of hospital stay (days)

13.7 ± 1.4

16.5 ± 1.6

p=0.17

Hospital mortality

26% (8/31)

22% (11/49)

p=0.2

Safety

Not reported

Not reported

–

Patients experiencing serious adverse events

None reported

None reported

–

Abbreviations: CPC, cerebral performance category; GCS, Glasgow Coma Scale; ICU, intensive care unit; IQR, interquartile range; LOS, length of stay; MIH, mild induced hypothermia; mmol/l, millimole per litre; n, number of patients; ng/ml, nanogram per millilitre; NSE, neuron‑specific enolase; °C/h, degrees Centigrade per hour.