Resuscitation Outcomes Consortium (ROC) PRIMED cardiac arrest trial methods

Resuscitation (2008) 78, 186—195

available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/resuscitation

CLINICAL PAPER

Resuscitation Outcomes Consortium (ROC) PRIMED cardiac arrest trial methods Part 2: Rationale and methodology for ‘‘Analyze Later vs. Analyze Early’’ protocol夽 Ian G. Stiell a,∗, Clif Callaway b, Dan Davis c, Tom Terndrup d, Judy Powell e, Andrea Cook e, Peter J. Kudenchuk e, Mohamud Daya f, Richard Kerber g, Ahamed Idris h, Laurie J. Morrison i, Tom Aufderheide j , for the ROC Investigators** a

Ottawa, Ontario, Canada Pittsburgh, PA, USA c San Diego, CA, USA d Birmingham, AL, USA e Seattle, WA, USA f Portland, OR, USA g Iowa City, IA, USA h Dallas, TX, USA i Toronto, Ontario, Canada j Milwaukee, WI, USA b

Received 5 December 2007; received in revised form 18 January 2008; accepted 29 January 2008

KEYWORDS Prehospital; Cardiac arrest; CPR; Defibrillation

Summary Objective: The primary objective of the trial is to compare survival to hospital discharge with modified Rankin score (MRS) ≤3 between a strategy that prioritizes a specified period of CPR before rhythm analysis (Analyze Later) versus a strategy of minimal CPR followed by early rhythm analysis (Analyze Early) in patients with out-of-hospital cardiac arrest. Methods: Design—–Cluster randomized trial with cluster units defined by geographic region, or monitor/defibrillator machine. Population—–Adults treated by emergency medical service (EMS) providers for non-traumatic out-of-hospital cardiac arrest not witnessed by EMS.

夽 A Spanish translated version of the abstract of this article appears as Appendix in the final online version at doi:10.1016/j.resuscitation.2008.01.027. ∗ Corresponding author. Tel.: +1 613 798 5555x18683; fax: +1 613 761 5351. E-mail address: [email protected] (I.G. Stiell). ∗∗ See ROC investigators at www.uwctc.org/ROC.

0300-9572/$ — see front matter © 2008 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.resuscitation.2008.01.027

Resuscitation Outcomes Consortium (ROC) PRIMED cardiac arrest trial methods

187

Setting—–EMS systems participating in the Resuscitation Outcomes Consortium and agreeing to cluster randomization to the Analyze Later versus Analyze Early intervention in a crossover fashion. Sample size—–Based on a two-sided significance level of 0.05, a maximum of 13,239 evaluable patients will allow statistical power of 0.996 to detect a hypothesized improvement in the probability of survival to discharge with MRS ≤3 rate from 5.41% after Analyze Early to 7.45% after Analyze Later (2.04% absolute increase in primary outcome). Conclusion: If this trial demonstrates a significant improvement in survival with a strategy of Analyze Later, it is estimated that 4000 premature deaths from cardiac arrest would be averted annually in North America alone. © 2008 Elsevier Ireland Ltd. All rights reserved.

Introduction A presenting rhythm of ventricular fibrillation (VF) or pulseless ventricular tachycardia (PVT) provides the best chance for survival after a resuscitation attempt for victims of out-of-hospital cardiac arrest. The traditional approach to these patients has been to prioritize the analysis of cardiac rhythm and delivery of defibrillatory shocks, if indicated, as quickly as possible. As a result, administration of external chest compressions with ventilation (CPR) is often delayed until after the initial rhythm evaluation. Some now advocate the opposite strategy of delaying rhythm analysis and electrical shocks until after the provision of a period of CPR. Three clinical studies have each attempted to evaluate the strategy of sustained CPR with deferred rhythm analysis versus early rhythm analysis (CPR only until the defibrillator electrodes can be placed).1—3 While the findings from two studies supported the strategy of CPR before initial rhythm analysis,1,2 one did not.3 Furthermore, none were definitive and all had important limitations. The Resuscitation Outcomes Consortium (ROC) investigators are conducting a large clinical trial, using a partial factorial design, entitled ROC PRIMED (Prehospital Resuscitation using an IMpedance valve and Early versus Delayed analysis) that will test two strategies. One strategy (Part 1) involves the impedance threshold device (ITD), which enhances venous return and cardiac output by increasing the degree of negative intrathoracic pressure during decompression. The second strategy (Part 2) involves initiating resuscitation with a sustained period of manual compressions and ventilations (Analyze Later), rather than attempting analysis and, if indicated, defibrillation immediately (Analyze Early). The purpose of this paper is to describe the rationale and methodology for Part 2, a comparison of a strategy of Analyze Later versus a strategy of Analyze Early in patients with out-of-hospital cardiac arrest. The rationale and methodology for the use of the ITD (Part 1) is described in a companion paper.

Background and significance Conceptual framework for Analyze Later Our current paradigm of cardiac arrest defines VF as ‘‘shockable’’ with the optimal therapeutic approach being immediate direct countershock.4 Integral to this approach is the concept that defibrillation attempts should occur

without delay upon recognition of VF, either by prehospital personnel or the analysis software contained within automated external defibrillators (AED), which can be applied by first responders with limited training or even laypersons.5 This approach has defined current emergency cardiac care (ECC) algorithms, shaped the development of emergency medical service (EMS) systems and has resulted in improved survival for cardiac arrest victims with an initial rhythm of VF in some EMS systems.6—12 Others have questioned whether this current standard of care has measurably improved outcome from out-of-hospital cardiac arrests on a community level.1 One of the major limitations to this cardiac arrest paradigm is its consideration of VF as homogenous, without regard for variability in VF morphology or elapsed time since the arrest. In contrast, experimental models of VF arrest support three distinct phases, each with a different optimal therapeutic approach.13 The early moments following arrest define an ‘‘electrical phase’’ during which little ischemic injury has occurred and rapid defibrillation attempts appear to be most effective. After some time period, probably around 3—4 min, the optimal therapeutic approach no longer appears to be immediate countershock but instead includes a period of chest compressions prior to defibrillation attempts. Effective chest compressions, traditionally held to provide approximately 30% of normal cardiac output during the first several minutes, may be sufficient to modify favorably the status of the myocardium during ventricular fibrillation. Immediate defibrillation attempts during the circulatory phase may be unsuccessful due to persistent or recurrent VF or may result in terminal pulseless electrical activity (PEA) or asystole. Interestingly, outcomes in patients ‘‘shocked’’ into PEA or asystole are significantly worse than when these (rather than VF) are the presenting rhythms.14 After some additional elapsed time, even chest compressions prior to defibrillation attempts do not appear to change outcome. This may be due to the initiation of irreversible ischemic changes that ultimately lead to substantial myocyte and neuronal cell death. This ‘‘metabolic phase’’ is thought to start after about 10 min of total arrest duration, with no currently available therapies demonstrating efficacy once this phase is reached. Another consideration relates to the significant proportion of patients who present with asystole or PEA as their initial cardiac arrest rhythm. Early provision of CPR in such patients may be of benefit in promoting the return of spontaneous circulation, or perhaps, fostering the development of a subsequent shockable rhythm.

188 Preliminary studies The effect of early rhythm analysis versus later rhythm analysis has been evaluated in animal and human studies. Animal models demonstrate improved ROSC and neurological outcomes with delayed countershock following a period of chest compressions in VF of moderate duration.15—18 Other animal studies have identified various VF morphologic features as potentially useful in predicting successful defibrillation in animal models of VF.19—21 Limited human data exist to support morphological analysis of VF/PVT as a predictor of successful ROSC.21—23 In addition, animal and human data suggest that chest compressions alone can modulate these morphological features to a more favorable configuration for successful ROSC.20—22 None of these morphological features have been implemented prospectively into devices in a manner that is adequate to justify their clinical use; however, these data further support the therapeutic value of chest compressions prior to defibrillation in VF of moderate duration. Finally, the duration of time between cessation of chest compressions and direct countershock appears to influence success of ROSC and ultimate survival.24,25 This suggests that prehospital providers should attempt to minimize delays after pausing chest compressions for rhythm analysis or ventilation prior to defibrillation attempts. These recommendations were incorporated into the 2005 AHA ECC guidelines.26 Clinical studies Three clinical studies have compared outcomes from outof-hospital cardiac arrest due to VF when a period of CPR has or has not been prescribed prior to the first attempts at defibrillation.1—3 Cobb et al.’s prospective observational, populationbased study revealed that survival to hospital discharge significantly improved during the intervention period (n = 478) when out-of-hospital cardiac arrest patients were treated with 90 s of CPR prior to shock compared to the preintervention period (n = 639) when out-of-hospital cardiac arrest patients were treated immediately with an AED (30% versus 21%; p = 0.04).1 There was a non-significant trend (71% versus 79%, p = 0.11) toward a more favorable neurological outcome observed during the intervention period. A significant interaction also described a relatively greater survival benefit for CPR before defibrillation as the response interval of the first arriving unit increased, particularly in cases in which the response interval of the first arriving unit was 4 min or longer (p = 0.04). These findings, although encouraging, could not be considered definitive and confirmative randomized clinical trials were recommended. Wik et al. conducted a prospective randomized trial of 200 patients with out-of-hospital cardiac arrest due to VF to compare standard care with immediate defibrillation (n = 96) or 3 min of CPR prior to defibrillation (n = 104).2 The primary outcome of survival to hospital discharge did not differ significantly between the two treatment arms nor were there significant differences in ROSC, 1-year survival, and ‘‘good neurological recovery’’ at hospital discharge or 1 year after cardiac arrest. Yet, among the 119 patients with EMS response times longer than 5 min, post hoc subgroup analysis revealed that more patients in the CPR first strategy compared to the standard group achieved ROSC (58% versus 38%; p = 0.04), survived to hospital discharge (22% versus 4%;

I.G. Stiell et al. p = 0.006) and survived to 1 year (20% versus 4%; p = 0.01) but did not adjust for multiple comparisons. Jacobs et al. conducted a prospective prehospital randomized trial in Western Australia which randomized 256 patients to a strategy of 90 s of CPR before defibrillation versus immediate defibrillation.3 Results revealed no significant difference in survival to hospital discharge between the two groups or between patients with a response interval of ≤5 min versus >5 min. Summary of rationale While it has been recognized for many years that out-ofhospital cardiac arrest patients who received ‘‘bystander CPR’’ have positive outcomes,27 this impact has been relegated to a secondary or even tertiary role in resuscitation sequencing. Small randomized or observational studies suggest that CPR before defibrillation may increase survival but the results to date are inconclusive. Although there is some evidence favoring immediate defibrillation in cases where the response time is <2 min, such response times are rare and the frequent delay in recognition of the out-of-hospital cardiac arrest and calling 911, as well as the complexity of the resuscitation protocol, convince us that response time should not be used as an intervention modifier. Thus, for those who sustain a cardiac arrest before EMS arrival, there is clinical equipoise with regard to the competing strategies of Analyze Later versus Analyze Early. In contrast, patients who sustain a cardiac arrest after EMS arrival will be ineligible for inclusion in this study, in recognition that such a known brief period of arrest is best treated with early defibrillation.

Aim The primary aim of the trial is to compare survival to hospital discharge with modified Rankin score (MRS) ≤3 between a strategy of Analyze Later consisting of a sustained period of CPR first followed by rhythm analysis versus a strategy of Analyze Early consisting of minimal CPR while defibrillator electrodes are attached with early rhythm analysis in patients with out-of-hospital cardiac arrest. The secondary aims of the trial are to compare survival to discharge, functional status at discharge and at 1, 3 and 6 months follow-up as well as depression scores at 3 and 6 months.

Methods Study design The ROC PRIMED Part 2 study will be a single-blinded cluster randomized crossover controlled trial with two intervention groups: (a) an Analyze Early group, and (b) Analyze Later group. Subjects in the Analyze Early group will be assigned to receive 30—60 s of chest compressions (that is, a brief period of standard CPR sufficient to allow time for placement of defibrillation electrodes and assure readiness of the defibrillator for rhythm analysis) prior to ECG analysis and defibrillation shocks if indicated. The Analyze Later group will receive approximately 3 min of standard compression—ventilation CPR prior to ECG analysis and rescue defibrillation. The intervention will be imple-

Resuscitation Outcomes Consortium (ROC) PRIMED cardiac arrest trial methods

189

Figure 1 Study population inclusion and exclusion criteria. **Persons are designated as having ‘‘opted out’’ by wearing a marker (e.g. bracelet) indicating their wish to be excluded from the trial.

mented by the ROC first qualified provider to arrive at the scene of cardiac arrest. Qualified providers are defined as defibrillation-capable first responders, emergency medical technicians (EMTs), and paramedics. We will include all out-of-hospital locations within the participating ROC study communities. The ITD strategy (Part 1) and the Analyze Later versus Analyze Early strategy (Part 2) will be implemented simultaneously, capitalizing on the common infrastructure necessary to accomplish the study. We do not anticipate a substantial interactive effect between the two strategies. This is a partial factorial design study since the eligibility criteria for the two interventions are not identical.

Study population and primary comparison populations Efficacy population Analysis of primary and secondary efficacy outcomes will be conducted on a modified intent-to-treat basis. In order to be included in the efficacy analyses, patients must meet the inclusion/exclusion criteria for the Analyze Later versus Analyze Early intervention, as described in Figure 1. Furthermore, in order to be evaluable, they must also not have experienced cardiac arrest secondary to drowning, electrocution, or strangulation. All eligible patients are considered enrolled into the Analyze Later versus Analyze Early protocol by intention to treat regardless of how much CPR they actually received and will be included in the primary efficacy analysis. Safety population Evaluation of the safety of the Analyze Later versus Analyze Early strategies will be made using all data from patients who were treated, regardless of whether they are member of the efficacy population.

for carry-over from event to event, and because it would add unacceptable complexity for EMS providers. Rather, each ROC site will be subdivided into a goal of at least 20 clusters by the following means: (a) according to EMS agency or geographical boundaries, or (b) according to individual defibrillator device, rig, or station. All clusters will crossover between intervention assignments at least once (i.e. have at least two distinct treatment periods).

Intervention For those clusters allocated to Analyze Later, defibrillator analysis will not be initiated until after delivery of compressions equivalent to approximately 3 min of CPR, after which a rescue shock will be administered, if indicated. For those clusters allocated to Analyze Early, defibrillator analysis will not be initiated until after the delivery of 30—60 s of chest compressions while defibrillator electrodes are attached after which a rescue shock will be administered if indicated (Figure 2). Chest compressions Initiation of chest compressions will not be delayed. Recognition that a patient is in cardiac arrest will immediately prompt activation (‘‘power on’’) of the defibrillator and the start of CPR and ventilation with the ITD. This power-on event will initiate the time recording by the device, and serve as a surrogate marker for ‘‘time zero’’ of initiating CPR. Minimum interruptions Training will emphasize uninterrupted chest compressions, except for required ventilations. If endotracheal intubation or other advanced airway procedures are deemed medically necessary, the providers will be encouraged to proceed while ensuring that chest compressions are continued with minimal interruption.

Random allocation Adherence to protocol The intervention will be randomly allocated according to the cluster assignment. Randomization by event or by individual patient was deemed unfeasible because of the potential risk

As intention-to-treat principles apply, any breach of protocol will not alter the study group to which a patient has

190 Table 1

I.G. Stiell et al. Intervention compliance time interval targets

Analyze Early Analyze Later

Power-on to analysis interval

Power-on to shock interval

CPR to 1st rhythm analysis interval

30—60 s 180—200 s

<90 s 180—220 s

<60 s 150—210 s

been assigned. The time interval from power-on of the defibrillator at the first recognition of a cardiac arrest to the first ECG analysis (i.e. the power-on to analysis interval) and to first rescue shock (i.e. power-on to shock interval) will be calculated from the time annotated from the defibrillator clock on the electronic record. A study-monitoring committee will evaluate protocol compliance during the run-in (section below) and active phases of the trial. Feedback from this committee will be provided to sites in order to encourage continuous quality improvement. Explicit criteria will define the successful delivery of the intended therapy and this information will be provided back to the EMS providers (Table 1).

Outcome measures The primary outcome is survival to hospital discharge with a MRS ≤3.28,29 Patients who are transferred to another acute care facility will be considered to be still hospitalized. Patients transferred to a non-acute ward or facility will be considered discharged. The secondary30—33 and exploratory34,35 outcomes are listed in Table 2. In addition, the number of hospital days and time interval from 911 call to patient death will be described for all hospitalized patients as measures of in-hospital morbidity after resuscitation. Additional background information, rationale for selection, and details about specific functional status measures are given in Appendix A.

Sample size and analysis The sample size calculation was based on having 90% power for the ITD (Part 1) portion of the trial yielding 14,154 evaluable patients. Due to the entry criteria differences between the Part 1 and Part 2 (Analyze Later versus Analyze Early) studies, we expect to enrol 13,239 evaluable patients for Part 2. Given 13,239 maximum evaluable patients, we have 99.6% statistical power to detect an improvement in the probability of survival to discharge with MRS ≤3 rate from 5.41% after Analyze Early to 7.45% after Analyze Later. This assumes a two-sided p < 0.05 group sequential stopping rule with up to three analyses (two interim and the final analysis) after accruing approximately one-third, two-thirds, and all of the maximum sample size (O’Brien—Fleming Boundaries, Pd = 1.0, Pa = 1.0).36 Further, we did not assume the statistical information at a given analysis was proportionate to the sample size,

Table 2 Timing and content of secondary and exploratory outcome measures Discharge MRS CPC ALFI-MMSE HUI GDS

Figure 2 Intervention strategies. Analyze Early strategy. Analyze Later strategy.

1 month

3 months

6 months

X

X X X X X

X X X X X

Xa Xa

Abbreviations: MRS (modified Rankin score), CPC (cerebral performance category), ALFI-MMSE (adult lifestyle and function (ALFI) version of the mini-mental status exam (MMSE), HUI (health utilities index), and GDS (geriatric depression score). a By chart review for hospital discharge and by phone after hospital discharge.

Resuscitation Outcomes Consortium (ROC) PRIMED cardiac arrest trial methods Table 3

191

CPR performance standards

Parameter

Target

Minimum acceptable (per minute)

Maximum acceptable (per minute)

Criterion for remediation/retraining

Chest compression

100/mina

80

120

CPR fractionb

0.85

Above maximum or below minimum parameters in >20% of resuscitations Below minimum parameter in >20% of resuscitations

0.5

—

a

Refers to speed of compressions rather than actual number of compressions per minute. CPR fraction will be defined as = (total seconds with chest compressions)/(total seconds with interpretable signal and no evidence of spontaneous circulation). b

but instead we assumed that only one-sixth of the statistical information will be available at the first interim look, one-half will be available at the second interim look, and all (assuming a 5% loss of efficiency due to cluster randomization) at the final analysis. The lower amount of statistical information corresponds to the estimated proportion of time a cluster has spent in both crossover periods at the given analysis. A cluster with equal time in each intervention arm corresponds to maximum statistical information. Data analysis will be conducted in the framework of general linear mixed models, which includes a fixed effect for each treatment arm and random effects for each randomization cluster.37 Adjustments for site and other baseline confounders will be incorporated, if necessary.

Monitoring of CPR process Several recent studies have evaluated the quality of CPR38,39 and the importance of monitoring and improving CPR performance40,41 in out-of-hospital and in-hospital settings. For this trial, all ROC clinical trial sites will have implemented a high-quality system for monitoring individual components of CPR. The CPR process will be monitored for a minimum of the first analyzable 5 min in all resuscitations. In addition, CPR process will be monitored for a minimum of 5 min after placement of an advanced airway. Determination of whether a resuscitation effort meets minimally acceptable CPR performance standards will be based on chest compression rate and CPR fraction criteria as defined in Table 3. Further rationale for monitoring CPR and additional information regarding the method of monitoring the CPR process is provided in Appendix B.

Data collection and data entry Data will be abstracted from collated source documents; the EMS patient care report(s), EMS dispatch times, EMS/fire/first responder electronic ECGs, emergency and hospital records and entered using customized web entry forms with standard data encryption and authentication methods.

Recruitment and informed consent This study qualifies for exception from informed consent for emergency research as outlined in US FDA regulation 21CFR50.24 and the Canadian Tri-Council Agreement for research in emergency health situations (Article 2.8).

Training The training objectives for this study include: review of optimal CPR performance, scientific basis for and review of study protocols, practical (hands-on) session, and knowledge assessment test. Some retraining will occur at least every 6 months, including the use of written reminders, web-based training modules, etc. Initial and retraining performance criteria will emphasize: optimal chest compression rate and depth, complete chest wall recoil with each compression, minimizing ‘‘hands-off’’ intervals, 3 min of compressions in Analyze Later arm, 30—60 s of compressions in Analyze Early arm, rapid placement of defibrillator pads and monitor/defibrillator activation (‘‘power-on’’) immediately upon recognition of pulseless arrest.

Run-in phase Compliance with the protocol and timely submission of the data will be required during the run-in phase before the Study Monitoring Committee determines the agency is now in the active phase of the trial. Compliance monitoring includes: correct inclusion/exclusion criteria, adherence to intervention arm, CPR process measures reported, and correct completion of data elements including reporting of adverse events.

Conclusion A large, randomized clinical trial is underway to examine the impact of delayed defibrillation on survival to hospital discharge in patients who are presumed to be without circulation for several minutes. If this trial demonstrates a significant improvement in survival with a strategy of Analyze Later, we estimate the premature demise of 4000 victims of out-of-hospital cardiac arrest would be averted annually in North America alone.

192

Conflict of interest statement None.

Acknowledgments This study was supported by a cooperative agreement (5U01 HL077863) with the National Heart, Lung and Blood Institute in partnership with the National Institute of Neurological Disorders and Stroke, The Canadian Institutes of Health Research (CIHR)—–Institute of Circulatory and Respiratory Health, Defence Research and Development Canada, and the Heart and Stroke Foundation of Canada. We would like to thank Julie Cummins for editing and preparing the manuscript for submission.

Appendix A. Functional status measures A.1. Background Assessment of neurological recovery is an important component of evaluating the effects of resuscitation interventions. Neurological injury after cardiac arrest contributes to inhospital death often because of withdrawal of care.42 Some surviving patients have decrements in cognitive function,29,43,44 although most surviving patients enjoy good quality of life.45,46 Neurological status varies over time with large improvements during the first few days after resuscitation.47,48 Detailed neuropsychiatric testing reveals that most improvement occurs during the first 3 months after resuscitation,44,47 with only small clinical changes during rehabilitation in those with severe anoxic brain injury.49,50 These studies suggest that it is possible to reliably measure neurological status 3 months after resuscitation. Criteria for choosing an instrument to measure neurological status include prior data about reliability and validity, availability of instruments suitable for a multicenter trial, and application to prior cardiac arrest survivors. The modified Rankin scale (MRS) will be assessed from the clinical record prior to discharge and by telephone interview after discharge. It has face validity and can be determined in person or over the telephone. MRS has concurrent validity with other measures of neurological recovery after stroke and brain injury.51,52 Use of a structured interview in a recent study of stroke patients improved the weighted Ä from 0.71 to 0.91.53 The only previous published applications of MRS to survivors of cardiac arrest evaluated a cohort of neurosurgical patients with in-hospital cardiac arrest54 and a cohort of survivors of out-of-hospital cardiac arrest.29 The cerebral performance category (CPC) will be assessed from the clinical record prior to discharge and by telephone interview after discharge. Consensus statements recommend use of the CPC to assess functional outcomes after resuscitation from cardiac arrest.55,56 CPC is a fivepoint scale that was adapted from the Glasgow Outcome Scale.34,35 CPC has limited discrimination between mild and moderate brain injury. A small study with incomplete followup of survivors demonstrated only moderate correlation with other measures of health-related quality of life.57 However, CPC at discharge predicts long-term survival.58

I.G. Stiell et al. An interesting question is whether simple MRS or CPC scores at discharge correlate with more detailed assessments post-discharge. We primarily intend to apply the MRS and CPC at discharge and will apply more detailed measures once the patient has been contacted post-discharge. A limitation to the post-discharge assessment of MRS or CPC scores is that there are no well-validated instruments for either measure. Also, proxy respondents tend to overestimate patient burden,59 and may or may not be available for interview. Therefore, we shall assess MRS or CPC at discharge based on review of the clinical record, and after discharge based on interview at 3 and 6 months. MRS will be assessed after discharge by using an instrument validated for telephone administration.53 CPC will be assessed by using a structured questionnaire via telephone administration. These latter questions were developed based on experience assessing outcomes after discharge in the OPALS study, PAD trial and ASPIRE trial as well as previous work by others to develop a structured phone interview to assess the related Glasgow Outcome Score.60 However, it should be recognized that these questions have not been validated in their current format. Generic health-related quality of life61,62 will be measured 3 and 6 months after discharge by administering the Health Utilities Index Mark 3 HUI system.33 The interviewadministered version of HUI3 requires completion of a maximum of 39 questions. (http://www.healthutilities.com accessed on 14 October 2005). The HUI3 consists of eight attributes of health (vision, hearing, speech, mobility, dexterity, emotion, cognition, and pain) with five or six levels per attribute. For each attribute, no, or mild, impairment in health has been defined as better than level 3 function (or level 4 function in the cognitive attributes). For each respondent, health status is described as a vector that combines the levels of each attribute. The health status information is then converted into a utility score of healthrelated quality of life on a scale from perfect health (1.0) to death (0).33,63 The HUI3 has been used in several ongoing studies of interventions for individuals with sudden cardiac arrest45,46 and has been shown to be reliable and valid in other populations.64—67 This health-related quality of life index is consistent with current standards for economic evaluation of health technologies.68 Neurological status will be measured at one, 3 and 6 months after discharge by using the adult lifestyle and function (ALFI) version of the mini-mental status exam (MMSE).30,31 The ALFI-MMSE has 23 items. Our experience using it in the PAD trial suggests that it requires approximately 5 min to administer. It is scored from 0 to 22, with a cutoff of 16/17 used to determine cognitive impairment. ALFI-MMSE correlates strongly (Pearson’s r = 0.73—0.85) with face-to-face MMSE across all patients or when grouped by clinical dementia rating scale class.30 Five items were discrepant (p < 0.05) between phone and face-to-face versions: day of the week, county, name of place, phrase repetition, and registration of the term ‘‘penny’’. Respondents were more likely to give incorrect answers face-to-face to the first three of these items, while repetition and registration of the item ‘‘penny’’ were significantly less accurate over the phone. Compared to the brief neuropsychiatric screening test, which is a weighted score of the Trailmaking A,69 Word Fluency,70 Weschler Memory Scale-Mental Control and

Resuscitation Outcomes Consortium (ROC) PRIMED cardiac arrest trial methods Logical Memory,71 ALFI-MMSE had a sensitivity of 68% and specificity of 100% for mild cognitive impairment. The corresponding MMSE values were 67% and 100%. Finally, depression will be assessed 3 and 6 months after discharge by using the Geriatric Depression Scale administered by telephone (T-GDS)32 since depression is observed in those with cardiovascular disease72 or those who have survived cardiac arrest,44 and can be difficult to differentiate from cognitive impairment. This instrument has 30 items that are answered either ‘yes’ or ‘no’. It is scored from 0 to 30, with a cutoff of 10/11 used to determine cognitive impairment.73 T-GDS has moderate test—retest reliability during a 1-week period (Ä 0.35—0.7, mean 0.52). Using a cutoff of 10/11, T-GDS has a sensitivity of 86% and specificity of 70% for detecting depression compared to a comprehensive assessment by a geriatric psychiatrist. Although initially designed to measure depression in older adults, it is also internally consistent and valid in younger adults.74

193

addition, ongoing monitoring and review of the CPR process will be used throughout the conduct of the trial. At the completion of every resuscitation attempt, the electronic record from the BLS and ALS devices used during the call will be obtained. All electronic records will be reviewed manually by proprietary automated analysis software. Determination of whether a resuscitation effort meets minimally acceptable CPR performance standards for the ROC will be based on whether it meets acceptable chest compression rate, and CPR fraction criteria as defined in Table 3.

Appendix C See Appendix C online at doi:10.1016/j.resuscitation.2008.01.027.

References Appendix B. Monitoring of CPR process B.1. Rationale for monitoring CPR process Recent studies have demonstrated that CPR is frequently not performed according to evidence-based guidelines in both the out-of-hospital and in-hospital settings.38,39 Although these studies lacked power to detect a significant relationship between CPR performance and patient outcome, a related study demonstrated that a greater rate of chest compressions was associated with a greater likelihood of achieving ROSC.40 The importance of monitoring and improving CPR performance was confirmed by the observation of potentially deleterious hyperventilation in the Milwaukee pilot study of ITD.41

B.2. Method of monitoring CPR process All ROC clinical trial sites will implement a high-quality system for monitoring individual components of CPR; specifically, the rate of chest compressions, and the proportion of pulseless resuscitation time during which chest compressions are provided (i.e. CPR fraction), collectively referred to as CPR process. Recent studies show no significant differences in these parameters during the first 5 min of resuscitation as compared with the entire resuscitation episode.38,39 It is anticipated that during the initial period, interruption of CPR due to rhythm analysis or other procedures is more likely than throughout the resuscitation episode. After insertion of an advanced airway and initiation of ventilation that is asynchronous with chest compressions, hyperventilation is more likely than during the early resuscitation period. Therefore, CPR process will be monitored for a minimum of the first analyzable 5 min in all resuscitations. In addition, CPR process will be monitored for a minimum of 5 min after receipt of an advanced airway. Where feasible, sites will be encouraged to monitor CPR process for the entire resuscitation. Sites will be required to demonstrate an ability to adequately acquire and analyze these CPR process data, identify and attempt to correct any observed deficiencies, and meet minimum performance standards before being eligible to enroll patients in the present trial. In

1. Cobb LA, Fahrenbruch CE, Walsh TR, et al. Influence of cardiopulmonary resuscitation prior to defibrillation in patients with out-of-hospital ventricular fibrillation. JAMA 1999;281:1182—8. 2. Wik L, Hansen TB, Fylling F, et al. Delaying defibrillation to give basic cardiopulmonary resuscitation to patients with outof-hospital ventricular fibrillation: a randomized trial. JAMA 2003;289(11):1389—95. 3. Jacobs IG, Finn JC, Oxer HF, Jelinek GA. CPR before defibrillation in out-of-hospital cardiac arrest: a randomized trial. Emerg Med Australas 2005;17(1):39—45. 4. Guidelines 2000 for cardiopulmonary resuscitation and emergency cardiovascular care. International consensus on science. Circulation 2000;102(8 Suppl. I):1—291. 5. Weisfeldt ML, Kerber RE, McGoldrick RP, Moss AJ, et al. Public access defibrillation. A statement for health care professionals from the American Heart Association Task Force on Automatic External Defibrillation. Circulation 1995;92:2763. 6. Caffrey SL, Willoughby PJ, Pepe PE, Becker LB. Public use of automated external defibrillators. N Engl J Med 2002;347(16):1242—7. 7. Valenzuela TD, Roe DJ, Nichol G, Clark LL, Spaite DW, Hardman RG. Outcomes of rapid defibrillation by security officers after cardiac arrest in casinos. N Engl J Med 2000;343(17):1206—9. 8. Hallstrom AP, Ornato JP, Weisfeldt M, et al. Public-access defibrillation and survival after out-of-hospital cardiac arrest. N Engl J Med 2004;351(7):637—46. 9. Myerburg RJ, Fenster J, Velez M, et al. Impact of communitywide police car deployment of automated external defibrillators on survival from out-of-hospital cardiac arrest. Circulation 2002;106(9):1058—64. 10. Mosesso VN, Davis EA, Auble TE, Paris PM, Yealy DM. Use of automated external defibrillators by police officers for treatment of out of hospital cardiac arrest. Ann Emerg Med 1998;32:200—7. 11. Stiell IG, Wells GA, Field BJ, et al. Improved out-of-hospital cardiac arrest survival through the inexpensive optimization of an existing defibrillation program: OPALS study phase II. Ontario prehospital advanced life support. JAMA 1999;281(13):1175—81. 12. Nichol G, Stiell IG, Laupacis A, Pham B, De Maio V, Wells GA. A cumulative metaanalysis of the effectiveness of defibrillator-capable emergency medical services for victims of out-of-hospital cardiac arrest. Ann Emerg Med 1999;34(4):517—25. 13. Weisfeldt ML, Becker LB. Resuscitation after cardiac arrest: a 3-phase time-sensitive model. JAMA 2002;288(23):3035—8.

194 14. Niemann JT, Stratton SJ, Cruz B, Lewis RJ. Outcome of outof-hospital postcountershock asystole and pulseless electrical activity versus primary asystole and pulseless electrical activity. Crit Care Med 2001;29(12):2366—70. 15. Yakaitis RW, Ewy GA, Otto CW, Taren DL, Moon TE. Influence of time and therapy on ventricular defibrillation in dogs. Crit Care Med 1980;8(3):157—63. 16. Niemann JT, Cruz B, Garner D, Lewis RJ. Immediate countershock versus cardiopulmonary resuscitation before countershock in a 5-minute swine model of ventricular fibrillation arrest. Ann Emerg Med 2000;36(6):543—6. 17. Menegazzi JJ, Seaberg DC, Yealy DM, Davis EA, MacLeod BA. Combination pharmacotherapy with delayed countershock vs standard advanced cardiac life support after prolonged ventricular fibrillation. Prehosp Emerg Care 2000;4(1):31—7. 18. Menegazzi JJ, Davis EA, Yealy DM, et al. An experimental algorithm versus standard advanced cardiac life support in a swine model of out-of-hospital cardiac arrest. Ann Emerg Med 1993;22(2):235—9. 19. Lightfoot CB, Nremt P, Callaway CW, et al. Dynamic nature of electrocardiographic waveform predicts rescue shock outcome in porcine ventricular fibrillation. Ann Emerg Med 2003;42(2):230—41. 20. Berg RA, Hilwig RW, Kern KB, Ewy GA. Precountershock cardiopulmonary resuscitation improves ventricular fibrillation median frequency and myocardial readiness for successful defibrillation from prolonged ventricular fibrillation: a randomized, controlled swine study. Ann Emerg Med 2002;40(6):563—70. 21. Callaway CW, Sherman LD, Mosesso Jr VN, Dietrich TJ, Holt E, Clarkson MC. Scaling exponent predicts defibrillation success for out-of-hospital ventricular fibrillation cardiac arrest. Circulation 2001;103(12):1656—61. 22. Jekova I, Dushanova J, Popivanov D. Method for ventricular fibrillation detection in the external electrocardiogram using nonlinear prediction. Physiol Meas 2002;23(2):337—45. 23. Eftestol T, Sunde K, Steen PA. Effects of interrupting precordial compressions on the calculated probability of defibrillation success during out-of-hospital cardiac arrest. Circulation 2002;105(19):2270—3. 24. Berg RA, Hilwig RW, Kern KB, Sanders AB, Xavier LC, Ewy GA. Automated external defibrillation versus manual defibrillation for prolonged ventricular fibrillation: lethal delays of chest compressions before and after countershocks. Ann Emerg Med 2003;42(4):458—67. 25. Koster RW. Limiting ‘hands-off’ periods during resuscitation. Resuscitation 2003;58(3):275—6. 26. Hazinski MF, Nadkarni VM, Hickey RW, O’Connor R, Becker LB, Zaritsky A. Major changes in the 2005 AHA guidelines for CPR and ECC: reaching the tipping point for change. Circulation 2005;112(Suppl. 24). IV-206. 27. Stiell IG, Wells GA, Field B, et al. Advanced cardiac life support in out-of-hospital cardiac arrest. N Engl J Med 2004;351(7):647—56. 28. Rankin J. Cerebral vascular accidents in patients over the age of 60. II. Prognosis. Scott Med J 1957;2(5):200—15. 29. van Alem AP, de Vos R, Schmand B, Koster RW. Cognitive impairment in survivors of out-of-hospital cardiac arrest. Am Heart J 2004;148(3):416—21. 30. Roccaforte WH, Burke WJ, Bayer BL, Wengel SP. Validation of a telephone version of the mini-mental state examination. J Am Geriatr Soc 1992;40:697—702. 31. Folstein MF, Folstein SE, McHugh PR. Mini-mental state. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975;12(3):189—98. 32. Burke WJ, Roccaforte WH, Wengel SP, Conley DM, Potter JF. The reliability and validity of the geriatric depression rating scale administered by telephone. J Am Geriatr Soc 1995;43(6):674—9.

I.G. Stiell et al. 33. Feeny D, Furlong W, Torrance GW, et al. Multiattribute and single-attribute utility functions for the health utilities index mark 3 system. Med Care 2002;40(2):113—28. 34. Jennett B, Bond M. Assessment of outcome after severe brain damage. Lancet 1975;1(7905):480—4. 35. The Brain Resuscitation Clinical Trial II Study Group. A randomized clinical trial of calcium entry blocker administration to comatose survivors of cardiac arrest: design, methods, and patient characteristics. Control Clin Trials 1991;12:525—45. 36. O’Brien PC, Fleming TR. A multiple testing procedure for clinical trials. Biometrics 1979;35:549—56. 37. Liang K, Zeger SL. Longitudinal data analysis using generalized linear models. Biometrika 1986;73:13—22. 38. Abella BS, Alvarado JP, Myklebust H, et al. Quality of cardiopulmonary resuscitation during in-hospital cardiac arrest. JAMA 2005;293(3):305—10. 39. Wik L, Kramer-Johansen J, Myklebust H, et al. Quality of cardiopulmonary resuscitation during out-of-hospital cardiac arrest. JAMA 2005;293(3):299—304. 40. Abella BS, Sandbo N, Vassilatos P, et al. Chest compression rates during cardiopulmonary resuscitation are suboptimal: a prospective study during in-hospital cardiac arrest. Circulation 2005;111(4):428—34. 41. Aufderheide TP, Sigurdsson G, Pirrallo RG, et al. Hyperventilation-induced hypotension during cardiopulmonary resuscitation. Circulation 2004;109(16):1960—5. 42. Niemann JT, Stratton SJ. The Utstein template and the effect of in-hospital decisions: the impact of do-not-attempt resuscitation status on survival to discharge statistics. Resuscitation 2001;51(3):233—7. 43. Yarnell PR. Neurological outcome of prolonged coma survivors of out-of-hospital cardiac arrest. Stroke 1976;7(3):279—82. 44. Roine RO, Kajaste S, Kaste M. Neuropsychological sequelae of cardiac arrest. JAMA 1993;269:237—42. 45. Nichol G, Stiell IG, Hebert P, Wells G, Vandemheen K, Laupacis A. What is the quality of life of survivors of cardiac arrest? A prospective study. Acad Emerg Med 1999;6:95—102. 46. van Alem AP, Waalewijn RA, Koster RW, de Vos R. Assessment of quality of life and cognitive function after out-of-hospital cardiac arrest with successful resuscitation. Am J Cardiol 2004;93(2):131—5. 47. Edgren E, Hedstrand U, Kelsey S, Sutton-Tyrrell K, Safar P, BRCTI Study Group. Assessment of neurological prognosis in comatose survivors of cardiac arrest. Lancet 1994;343(8905):1055—9. 48. Booth CM, Boone RH, Tomlinson G, Detsky AS. Is this patient dead, vegetative, or severely neurologically impaired? Assessing outcome for comatose survivors of cardiac arrest. JAMA 2004;291(7):870—9. 49. Fertl E, Vass K, Sterz F, Gabriel H, Auff E. Neurological rehabilitation of severely disabled cardiac arrest survivors. Part I. Course of post-acute inpatient treatment. Resuscitation 2000;47(3):231—9. 50. Pusswald G, Fertl E, Faltl M, Auff E. Neurological rehabilitation of severely disabled cardiac arrest survivors. Part II. Life situation of patients and families after treatment. Resuscitation 2000;47(3):241—8. 51. Hop JW, Rinkel GJ, Algra A, van Gijn J. Changes in functional outcome and quality of life in patients and caregivers after aneurysmal subarachnoid hemorrhage. J Neurosurg 2001;95(6):957—63. 52. Weimar C, Kurth T, Kraywinkel K, et al. Assessment of functioning and disability after ischemic stroke. Stroke 2002;33(8):2053—9. 53. Wilson JTL, Hareendran A, Hendry A, Potter J, Bone I, Muir KW. Reliability of the modified Rankin scale across multiple raters: benefits of a structured interview. Stroke 2005;36(4):777—81. 54. Rabinstein AA, McClelland RL, Wijdicks EF, Manno EM, Atkinson JL. Cardiopulmonary resuscitation in critically ill

Resuscitation Outcomes Consortium (ROC) PRIMED cardiac arrest trial methods

55.

56.

57.

58.

59.

60.

61.

62.

63.

neurologic-neurosurgical patients. Mayo Clin Proc 2004;79(11): 1391—5. Cummins RO, Chamberlain DA, Abramson NS, et al. Recommended guidelines for uniform reporting of data from out-of-hospital cardiac arrest: the Utstein style. Circulation 1991;84(2):960—75. Jacobs I, Nadkarni V, Bahr J, Berg RA, Billi JE, Bossaert L, et al. Cardiac arrest and cardiopulmonary resuscitation outcome reports: update and simplification of the Utstein templates for resuscitation registries: a statement for healthcare professionals from a task force of the International Liaison Committee on Resuscitation (American Heart Association, European Resuscitation Council, Australian Resuscitation Council, New Zealand Resuscitation Council, Heart and Stroke Foundation of Canada, InterAmerican Heart Foundation, Resuscitation Councils of Southern Africa). Circulation 2004;110(21):3385—97. Hsu JWY, Callaham M, Madsen CD. Quality-of-life and formal functional testing of survivors of out-of-hospital cardiac arrest correlates poorly with traditional neurological outcome scales. Ann Emerg Med 1996;28:597—605. Herlitz J, Ekstrom L, Wennerblom B, Axelsson A, Bang A, Holmberg S. Prognosis among survivors of prehospital cardiac arrest. Ann Emerg Med 1995;25:58—63. Sneeuw KCA, Sprangers MAG, Aaronson NK. The role of health care providers and significant others in evaluating the quality of life of patients with chronic disease. J Clin Epidemiol 2002;55(11):1130—43. Wilson JT, Pettigrew LE, Teasdale GM. Structured interviews for the Glasgow Outcome Scale and the extended Glasgow Outcome Scale: guidelines for their use. J Neurotrauma 1998;15(8):573—85. Feeny D, Furlong W, Boyle W, Torrance G. Multi-attribute health status classification systems: health utilities index. PharmacoEconomics 1995;7(6):490—502. Torrance GW, Furlong W, Feeny D, Boyle M. Multi-attribute preference functions: health utilities index. PharmacoEconomics 1995;7(6):503—20. Torrance GW, Zhang Y, Feeny D, Furlong W, Barr R. Multiattribute preference function for a comprehensive health status

64.

65.

66.

67.

68.

69. 70.

71. 72.

73.

74.

195

classification system: Centre for Health Economics and Policy Analysis, McMaster University. 92-18. 1992. McMaster University. Ref Type: Report. Boyle M, Furlong W, Torrance G, Feeny D, Hatcher J. Reliability of the Health Utilities Index-Mark III used in the 1991 cycle 6 Canadian General Social Survey Health Questionnaire. Qual Life Res 1995;4:249—57. Gemke RJ, Bonsel GJ. Reliability and validity of a comprehensive health status measure in a heterogeneous population of children admitted to intensive care. J Clin Epidemiol 1996;49(3):327—33. Gold M, Franks P, Erickson P. Assessing the health of the nation. The predictive validity of a preference-based measure and selfrated health. Med Care 1996;34(2):163—77. Stiell IG, Spaite DW, Wells GA, et al. The Ontario Prehospital Advanced Life Support (OPALS) Study: rationale and methodology for cardiac arrest patients. Ann Emerg Med 1998;32(2):180—90. Gold MR, Siegel JE, Russell LB, Weinstein MC. In: Gold MR, Siegel JE, Russell LB, Weinstein MC, editors. Cost-effectiveness in health and medicine. New York: Oxford University Press; 1996. Armitage SG. An analysis of certain psychological tests used for the evaluation of brain injury. Psychol Monogr 1946;60:1—48. Thurston LL, Thurston TG. Examiner manual for the SRA primary mental abilities test. Chicago, IL: Science Research Associates; 1949. Weschler D, Stone CP. Manual: Weschler memory scale. New York: Psychological Corporation; 1983. Powell LH, Catellier D, Freedland KE, et al. Depression and heart failure in patients with a new myocardial infarction. Am Heart J 2005;149(5):851—5. McDowell I, Kristjansson E. Measuring health. A guide to rating scales and questionnaires. 2nd ed. New York: Oxford University Press; 1996. Rule BG, Harvey HZ, Dobbs AR. Reliability of the geriatric depression scale for younger adults. Clin Gerontol 1990;9(2):37—43.