快樂小藥師 Im pharmacist nichts glücklich

Trial Design

The Target Temperature Management 33°C versus 36°C after Out-of-Hospital Cardiac Arrest (TTM) trial was a randomized clinical trial recruiting patients in 36 intensive care units (ICUs) in Europe and Australia. The rationale for and design of the trial, as well as the statistical analysis plan, have been published previously.12,13 The protocol (available with the full text of this article at NEJM.org) was approved by the ethics committees in each participating country and institution. An independent data and safety monitoring committee reviewed the data and performed one prespecified, blinded interim analysis. The steering group (see the Supplementary Appendix, available at NEJM.org) vouches for the accuracy and completeness of the data and analysis and for the adherence of this report to the trial protocol.

Patients

We consecutively screened patients 18 years of age or older who were unconscious (a score of <8 on the Glasgow Coma Scale [on which scores range from 3 to 15, with lower scores indicating reduced levels of consciousness]) on admission to the hospital after out-of-hospital cardiac arrest of presumed cardiac cause, irrespective of the initial rhythm. Eligible patients had more than 20 consecutive minutes of spontaneous circulation after resuscitation.14 The main exclusion criteria were an interval from the return of spontaneous circulation to screening of more than 240 minutes, unwitnessed arrest with asystole as the initial rhythm, suspected or known acute intracranial hemorrhage or stroke, and a body temperature of less than 30°C. A full list of exclusion criteria is provided in the Supplementary Appendix. In accordance with national requirements and the principles of the Declaration of Helsinki, written informed consent was waived, delayed, or obtained from a legal surrogate, depending on the circumstances, and was obtained from each patient who regained mental capacity.15

Randomization and Trial Intervention

After being screened for eligibility, patients were randomly assigned in a 1:1 ratio to targeted temperature management with a target body temperature of either 33°C or 36°C. Randomization was performed centrally with the use of a computer-generated assignment sequence. Intervention assignments were made in permuted blocks of varying size and were stratified according to site.

Health care professionals caring for the trial patients were aware of the intervention assignments because of inherent problems with blinding of body temperature. Physicians performing neurologic prognostication, assessors of neurologic follow-up and final outcome, study administrators, statisticians, and the authors were unaware of the intervention assignments. During the analysis phase, the intervention groups were identified only as 0 and 1, and the manuscript was written and approved by all the authors before the randomization code was broken.16

The intervention period of 36 hours commenced at the time of randomization. Sedation was mandated in both groups until the end of the intervention period. The goal was to achieve the assigned temperature as rapidly as possible with the use of ice-cold fluids, ice packs, and intravascular or surface temperature-management devices at the discretion of the sites. Details of the trial interventions, including the management of an initial body temperature below the assigned target, are provided in the Supplementary Appendix.

After 28 hours, gradual rewarming to 37°C in hourly increments of 0.5°C was commenced in both groups. At 36 hours, mandatory sedation was discontinued or tapered. After the intervention period, the intention was to maintain the body temperature for unconscious patients below 37.5°C until 72 hours after the cardiac arrest, with the use of fever-control measures at the discretion of the sites.

Neurologic Prognostication and Withdrawal of Life-Sustaining Therapies

A physician who was unaware of the intervention assignments performed a neurologic evaluation 72 hours after the end of the intervention for patients who remained unconscious and issued a recommendation for the continuation or withdrawal of therapy. The trial protocol established prespecified criteria for withdrawal of life-sustaining therapy12 (see the Supplementary Appendix). All clinical decisions remained at the discretion of the treating team.

Follow-up and Outcomes

All surviving patients were followed until 180 days after the enrollment of the last patient. The primary outcome was all-cause mortality through the end of the trial. The main secondary outcome was a composite of poor neurologic function or death, defined as a Cerebral Performance Category17,18 (CPC) of 3 to 5 and a score of 4 to 6 on the modified Rankin scale,19,20 at or around 180 days. The CPC scale ranges from 1 to 5, with 1 representing good cerebral performance or minor disability, 2 moderate disability, 3 severe disability, 4 coma or vegetative state, and 5 brain death. Scores on the modified Rankin scale range from 0 to 6, with 0 representing no symptoms, 1 no clinically significant disability, 2 slight disability, 3 moderate disability, 4 moderately severe disability, 5 severe disability, and 6 death. Mortality at 180 days and individual neurologic scores were also analyzed separately. Other secondary outcomes were the CPC at discharge from the ICU and from the hospital and the best (numerically lowest) reported CPC during the trial period. Predefined serious adverse events21 were recorded up to day 7 in the ICU. Data collection and verification for all trial data and for the outcome measures are described in the Supplementary Appendix.

Statistical Analysis

We estimated that a sample of 900 patients would provide 90% power to detect a 20% reduction in the hazard ratio for death in the 33°C group as compared with the 36°C group, at a two-sided alpha level of 0.05. Alternatively, to detect a relative risk reduction of 20%, with the assumption of a mortality of 44% in the 33°C group versus 55% in the 36°C group, a sample of 850 patients would be needed. On the basis of these assumptions, a sample of 950 patients was chosen, to allow for a loss to follow-up of 50 patients.

The principal trial analyses were performed in the modified intention-to-treat population, defined as all randomly assigned patients except those withdrawing consent for use of all trial data and those not fulfilling inclusion criteria and never receiving the intervention.22 Additional analyses were performed in the intention-to-treat population, which included all randomly assigned patients except those withdrawing consent, and in the per-protocol population, which excluded patients with one or more major protocol violations (listed in the Supplementary Appendix).

The Wilcoxon signed-rank test was used to compare distributions of continuous outcome measures. Kaplan–Meier survival curves were compared between the intervention groups with the use of the log-rank test. Relative risks were compared with the use of Cochran–Mantel–Haenszel statistics. Trends were assessed with the use of the Cochran–Armitage test. Logistic-regression and Cox analyses were performed as appropriate, with adjustment for site and for five baseline variables: age, sex, presence or absence of shockable rhythm, presence or absence of circulatory shock on admission, and the time from cardiac arrest (or from the emergency call for unwitnessed cardiac arrests) to the return of spontaneous circulation. Odds ratios were converted to relative risks.23 All primary analyses were adjusted for site.24 Temperature data were analyzed with the use of a mixed model with repeated measures. The effect of time was modeled with the use of a polynomial; the use of compound symmetry and first-order autoregressive covariance structures was compared, and the better-fitting model was used. SAS software, version 9.3, and SPSS software, version 17.1, were used for all analyses. All tests were two-sided and adjusted for multiple comparisons. A P value of 0.05 or less was considered to indicate statistical significance.

Patients

A total of 950 patients were enrolled between November 2010 and January 2013; of these patients, 476 were randomly assigned to the 33°C group and 474 to the 36°C group. The modified intention-to-treat population (the primary-analysis population) consisted of 473 patients assigned to 33°C and 466 assigned to 36°C (Fig. S1 in the Supplementary Appendix). The two groups had similar prerandomization characteristics (Table 1
).

Glasgow Coma Scale scores on admission, cardiovascular Sequential Organ Failure Assessment scores, and details of diagnostic procedures, interventions, and the use of health services are provided in Tables S1, S2, and S3, respectively, in theSupplementary Appendix.

Temperature Intervention

The mean values of the initial recorded body temperature (tympanic) were 35.2°C and 35.3°C in the 33°C and 36°C groups, respectively. Temperature was managed with an intravascular cooling catheter in 24% of patients and with a surface cooling system in 76% of patients in both groups. The temperature curves are depicted in Figure 1
 (P<0.001 for separation of the curves).

Three patients in the 33°C group and four in the 36°C group did not receive the assigned intervention (Table S4 in the Supplementary Appendix). Sixteen patients assigned to the 33°C group were rewarmed before reaching the intended time point of 28 hours after randomization, at the discretion of the treating physician and as allowed by the protocol (Table S5 in theSupplementary Appendix). Additional information regarding shivering and fever is available in the Supplementary Appendix.

Withdrawal of Life-Sustaining Therapy

During the first 7 days of hospitalization, life-sustaining therapy was withdrawn in 247 patients (132 in the 33°C group and 115 in the 36°C group). Reasons for withdrawal of life-sustaining therapy included brain death, multiorgan failure, and ethical concerns (Table S7 in the Supplementary Appendix). A protocol-defined approach to neurologic prognostication was used to make recommendations regarding the continuation or withdrawal of life-sustaining therapy (Table S8 in the Supplementary Appendix).

Follow-up and Outcomes

Follow-up was obtained by means of a face-to-face interview with the patient (for 86% of patients), a structured telephone interview with the patient (6%), a telephone call to the patient or a relative (5%), or a telephone call to a proxy provider of information (i.e., a staff member of a nursing home or a general practitioner) (3%). The last follow-up assessment was performed on July 9, 2013. The mean period of follow-up for all patients was 256 days.

At the end of the trial, 235 of 473 patients in the 33°C group (50%) and 225 of 466 patients in the 36°C group (48%) had died (hazard ratio in the 33°C group, 1.06; 95% confidence interval [CI], 0.89 to 1.28; P=0.51) (Table 2
 and Figure 2).

The groups did not differ significantly with respect to the composite outcome of death or poor neurologic function at 180 days with the use of either the CPC or the modified Rankin scale score (risk ratio for a CPC of 3 to 5 in the 33°C group, 1.02; 95% CI, 0.88 to 1.16; P=0.78; and risk ratio for a score of 4 to 6 on the modified Rankin scale in the 33°C group, 1.01; 95% CI, 0.89 to 1.14; P=0.87) (Table 2). The neurologic scores on both scales are shown in Table 3
 and in Table S9 in the Supplementary Appendix. There were no significant differences in the distribution of CPCs or modified Rankin scale scores between the two groups (P=0.85 and P=0.67 for trend, respectively). With the use of the best reported CPC during the trial (Table 3), the relative risk of death or poor neurologic function in the 33°C group was 1.04 (95% CI, 0.89 to 1.17; P=0.67).

Similar results were obtained in adjusted analyses and in the intention-to-treat and per-protocol populations (see the Supplementary Appendix, including Tables S10 and S11). The effect of the intervention was consistent across predefined subgroups (Fig. S2 in the Supplementary Appendix).

One or more serious adverse events occurred in 439 of 472 patients in the 33°C group (93%) as compared with 417 of 464 patients in the 36°C group (90%) (risk ratio, 1.03; 95% CI, 1.00 to 1.08; P=0.09). Hypokalemia was more frequent in the 33°C group (19%, vs. 13% in the 36°C group, P=0.02). For the full list of serious adverse events, see Table S12 in theSupplementary Appendix. The presumed causes of death as assessed by the trial investigators were similar in the two groups (Table S13 in theSupplementary Appendix).