TTM 2 Trial

TTM 2 Trial 

Background 

ILCOR guidelines strongly recommend targeted temperature management with a constant target between 32 C and 36 C however, the overall evidence behind this recommendation is of low certainty.

Hypothesis 

At 6 months, the incidence of death would be lower in the hypothermia group than in the normothermia group.

Methods 

Population: An international multicenter study involving 14 countries, 61 study sites. Adult patients who suffered out of hospital cardiac arrest due to cardiac or unknown cause, irrespective of the rhythm. Patients were more than 20 minutes of spontaneous circulation after resuscitation. They were unconscious, with no response to verbal commands. Randomized in a 1:1 ratio to hypothermia or control group. Web-based randomization permuted blocks and stratified by centre. Blinding was not carried out; however, assessors were unaware of trial assignment

Excluded: If randomization could not be done within 180 minutes of ROSC. Unwitnessed arrest with asystole as the initial rhythm. Patients with prespecified limitations in care.

Intervention

Hypothermia was induced by the surface or intravascular cooling. Patients were cooled to 33 C for 28 hour after randomization. Rewarming by one-third of a degree per hour – took approximately 12 hours; hence a total duration of intervention of 40 hours.

Control

Normothermia was targeted, defined as a temperature of 37.5 C or less. If temperature more than 37.8, active cooling by surface or intravascular cooling.

Common management, assessment

Sedation until the end of the intervention period. After the period of intervention, the normothermic range between 36.5 to 37.7 was maintained until 72-hour post-randomization for patients who remained comatose or sedated.

Neurological assessment at 96 hours post-randomization. There were definitive criteria for neurological prognostication. Withdrawal of life-sustaining measures due to predicted poor neurological outcome.

Power

A sample size of 1862 patients demonstrating 15% RR reduction and 7.5% AR reduction for 90% power and a 2-sided alpha level of 0.05. To allow for loss to follow-up and consent withdrawal, a sample size of 1900 was determined.

Characteristics

Mean age: 64 years, 80% male, bystander CPR in 82% vs. 78%, initial shockable rhythm in 72 vs. 75%, time to sustained ROSC: 25 minutes

Time to cooling to 34 C: 3 h; 53/930 (6%) rewarmed due to cardiovascular instability.

40 vs. 38% underwent PCI

Primary outcome

All-cause mortality at 6 months: 465/925 (50%) vs. 446/925 (48%); RR: 1.04 (0.94–1.14) P = 0.37

No difference is prespecified subgroup analysis: gender, age >65, time to ROSC >25 min, initial shockable vs. non-shockable rhythm, shock on admission

Secondary outcomes

Poor functional outcome at 6 months: score on the Modified Rankin Scale of 4–6

(1: no disability, 2: mild disability, 3: moderate disability, able to walk without assistance, 4: unable to walk, unable to attend to bodily needs without help, 5: bedridden, 6: death

488/881 (55%) vs. 479/886 (55%)

16 vs. 17% patients did not have any limitation (Rankin score of 0)

Binary assessment of functional outcome: 54% of each group with poor outcome

The results were similar across all subgroups

Health-related quality of life with a VAS was similar between groups

Adverse events

Arrhythmias: 24 vs. 16% (p < 0.001)

No difference in bleeding, skin complications, pneumonia, sepsis

Summary

In patients who suffered out of hospital cardiac arrest, targeted hypothermia to 33 C within the first 3 hours of the event compared to targeted normothermia of 37.5 C or less  did not reduce 6-month mortality

Neurological outcomes at 6 months were similar between groups 

Strengths  

1. Multicentre RCT with very good internal validity
2. Answered the important question around temperature management in a post-ROSC setting.
3. The pragmatic design of the trial
4. Independent clinician to assess the prognostication.
5. Objective and standardised criteria used for neuro prognostication

Limitations  

1. Questionable generalizability – conducted in the first world settings, immediate bystander CPR in 92% of patients in each group.
2. 96 hours of mandatory ICU stay – questionable feasibility in resource-poor settings.
3. Targeted temperature management vs. no temperature management at all – the answer is still unknown. 
4. No blinding
5. Whether hypothermia is beneficial in different subgroups of out-of-hospital cardiac arrest such as drug overdose, hanging, drowning is not tested in this trial.

Our recommendations  

1. Continue targeted temperature management in patients who develop out of hospital cardiac arrest.
2. Insufficient evidence to abandon temperature management or fever control at this stage.,
3. Independent clinician assessment using standardised objective criteria for neuro prognostication is the way forward.