• Periodic scientific journal

Neurosonology and Cerebral Hemodynamics

sp 2016

View or download the journal in .PDF format.

Volume 16, number 2, 2020; View on .pdf file
Volume 16, number 1, 2020; View on .pdf file
Volume 15, number 2, 2019; View on .pdf file
Volume 15, number 1, 2019; View on .pdf file
Volume 14, number 2, 2018; View on .pdf file
Volume 14, number 1, 2018
; View on .pdf file
Volume 13, number 2, 2017; View on .pdf file
Volume 13, number 2 supplement, 2017; View on .pdf file
Volume 13, number 1, 2017; View on .pdf file
Volume 12, number 2, 2016View on .pdf file
Volume 12, number 1, 2016; View on .pdf file
Volume 11, number 2, 2015; View on .pdf file
Volume 11, number 1, 2015; View on .pdf file
Volume 10, number 2, 2014; View on .pdf file
Volume 10, number 1, 2014; View on .pdf file
Volume 9, number 2, 2013; View on .pdf file
Volume 9, number 1, 2013; View on .pdf file
Volume 8, number 2, 2012; View on .pdf file
Volume 8, number 1, 2012; View on .pdf file

Archive

The Journal “Neurosonology and Cerebral Hemodynamics” is included in Web of Science and Emerging Sources Citation Index. It is a part of the Internet Platform Scholar One – a system for international distribution, indexing and editing of articles.
It is accredited with 5 credits, category “D” by the Bulgarian Medical Association.

Editors

Instructions for authors

Contacts

Usefull links:

Central Medical Libtary MU - Sofia

Faculty of Medicine - Sofia University

Hospital Lozenetz

Medicine University - Sofia

N. Ramshev

Neurosonology and Cerebral Hemodynamics 10, 2014:43–52

 

TITLE Cardiopulmonary Resuscitation after Cardiac Arrest – Modern Concepts and Approaches
AUTHOR/S N. Ramshev
PUBL. DATE May 2014
SOURCE Neurosonology and Cerebral Hemodynamics 10, 2014:43–52
TYPE Periodic scientific journal
ABSTRACT

Тhe present review article provides contemporary perspectives on the sudden cardiac arrest and presents the modern concepts and approaches for successful outcome after cardiopulmonary resuscitation. It is focused on prognostic factors of survival and determination of an adequate multidisciplinary therapeutic approach.

KEY WORDS

 cardiopulmonary resuscitation, cardiac arrest, predictors, survival

 

Cardiovascular diseases cause about 40% of all deaths in Europe below 75 years of age [22]. Based on epidemiological data the Sudden Cardiac Arrest (SCA) is responsible for 60% of all deaths in elderly group (above 75 years) as a result of coronary artery disease complication [26, 31]. Three basic types of cardiovascular complications are assumed to lead to Sudden Cardiac Death (SCD): acute myocardial infarction, ischemia without infarction and structural changes (myocardial sclerosis or ventricular dilatation as a secondary transformation of old myocardial infarction or chronic ischemia). Over 7 billion events of SCD worldwide are reported on annual basis. 300 thousands of them occur in USA and 700 thousands in Europe [22]. In 19-27% of autopsies pathoanatomical evidences of myocardial necrosis are identified; predominantly stable plaques and chronic myocardial changes in 50% of all cases [7].

Different other non-ischemic cardiac (hyper­trophic cardiomyopathy, coronary artery abnorma­lities, myocarditis, heart rhythm abnormalities, hypertensive heart disease, congestive heart failure, etc.) and non-cardiac causes (trauma, non-traumatic bleeding, drawing, intoxication, etc.) increase the risk of SCA [14, 18, 31].

The wide medical community started recognizing and promoting the technique combining the artificial respiration and chest compressions being a key part of cardiopulmonary resuscitation (CPR) after cardiac arrest in the middle of 20th century. The modern treatment related to development of CPR technique is a result of increasing the knowledge about the cardiac arrest (CA). Different terms are used in the modern medicine which comprehension is enriched constantly and their meaning should be known for the clinical practice.

Cardiac arrest. This term defines the stop of cardiac pumping function, confirmed by lack of pulse and lack of breathing or agonal breathing. It has wider meaning including all cases of cardiac arrest independently of the cause (including terminal stages of refractory heart failure). The cardiac arrest is a result of different causes which are combined in three electrocardiographic clinical groups:
• Tachy-form: ventricular fibrillation (VF), ventri­cular flatter (VFl) and pulseless ventricular tachycardia (PVT);
• Bradycardia-asystole form: asystole and extreme bradycardia (high grade of sinoatrial block, sinoatrial arrest, high grade of atrioventricular block and lack of effective replacement rhythm);
• Pulseless Electrical Activity (PЕА) is a condition with electrical activity (registered on the monitor) and lack of mechanical cardiac contraction and circulation (no pulse). PEA is observed when severe mechanical cardiac function disturbances are present (contractility failure in massive infarction, cardiac tampo­nade after ruptur of the free wall of the left ventricle after acute myocardial infarction, myocardial compression due to tension pneumothorax, obturation of pulmonary artery trunk or its main branches, etc.) and it is reversible after correction of the conditions (hypovolemia, hypoxia, hypothermia etc.).

Arrhythmias after СА are shockable (VF or PVT) and non-schockable (PEA or asystole) [22].

Sudden Cardiac Death (SCD). It is an unexpected death occurring very fast (in an hour after the begining of acute symptoms). It has cardiac origin with or without prior cardiac disease. Cardiac arrest is the clinical presentation due to cardiac causes only (Ischemic Cardiac Disease, Heart Failure or systolic left ventricular dysfunction, cardiomyopathy, arrhythmias, cardiac vices etc.). The most typical clinical manifestation of SCD is: sudden loss of consciousness, missing perceptible pulse in auscultation, no pulsations of large arteries and no breathing. The condition is irreversible if CPR is not administered and results in biological death. Since SCD is a sudden condition the cardiac diseases in terminal stage of heart failure and traumatic events are excluded.

Cardiopulmonary Resuscitation is a complex of diagnostic and therapeutic procedures in emergency setting after CA for restoring of circulation and providing of oxygenated blood to brain. Weisfeldt et al. defined three consecutive phases after СА according to ischemia progression: electrical phase (0-4 min) following by circulatory phase (4-10 min) and metabolic phase (more than 10 min) [29].

The electrical phase is manifested by arrhythmias; the process is reversible and imme­diate defibrillation is significantly successful especially if it starts in 3 minutes. Ischemic meta­­­bolites are accumulated in the myocardium during the circulatory phase leading to VF. The defibrillation might not be successful without prior CPR. Ventricular fibrillation features with lack of out-of-refractory period („free of excitation“) thus the defibrillation could not cease the re-entry mechanism. There is no possibility for recovering of cardiac rhythm supporting the active circulation during the metabolic phase. Biologic death occurs in 5 minutes after the CA. The timeframes could be extended considerably when applying an adequate CPR.

The International Guidelines for CPR lastly updated in 2010 remained almost unchanged compared to those from 2005 due to lack of considerably significant new evidence data. During the last years a multidisciplinary app­roach for CPR has been established. The American Heart Association (AHA) and American Stroke Association (ASA) Guidelines are unified in 2013.

Ischemia-Reperfusion Syndrome after СА (IRS, post-arrest syndrome). It is a result of the Return of Spontaneous Circulation (ROSC) during development of ischemia of tissues [5]. Clinically the condition manifests with [22]:
• Myocardial Dysfunction (MD) with hypotension, low cardiac index and arrhythmia. Controlling of mean blood pressure is critical for providing of an adequate urine output and normalizing of lactate plasma levels;
• CNS symptoms – coma, seizures, myoclonus, cognitive dysfunction and brain death. Seizures after ROSC are observed in 5-15% of adults and in up to 40% of patients in coma (generalized, non-generalized, myoclonuses, etc.). They increase the brain metabolism three times leading to additional brain damage.
• Polyorgan insufficiency as a result of decreased intravascular volume, vasodilation and increased risk of infections (result of acti­vation of immunologic and coagulogic path­ways);
• Precipitating pathology (acute coronary syndrome, cardiomyopathies, COPD, toxicological/thromboembolic/infectious diseases and hypovolemia) [22].

The severity of IRS varies depending on the continuation and the cause of the cardiac arrest. In case of short term CA, the IRS might not be displayed. Symptoms of brain damage might deepen by impaired microcirculation, impaired autoregulation, hypercarbia, hyperoxia, pyrexia, hyperglycemia and seizures. Significant MD is often observed after СА and usually recovers in 2-3 days. Neurologic complications are a cause of death in 2/3 of patients after СА before being admitted to an Intensive Care Unit and in 1/4 of hospitalized patients [22].

Terminal Conditions are classified based on the time after the begining of CA and unsuccessful CPR, as follows: 0-5 min – clinical death (CD); 5-10 min – CD without neurological deficiency; 10-15 min – CD with neurological deficiency; 15-20 min – cortical death; and over 20 min – brain and biologic death.

Clinical Death is a result of discontinuation of the effective brain circulation demonstrated with reversible and temporary failure of brain functions. СА is identified with the clinical death. After 2-3 minutes the alveolar anoxia leads to clinical death due to oxygen depletion of body reserves [30]. It might be a result of removal of the technical support of circulation and respiration.

Cortical Death is a result of severe destruction of the brain cortex and a loss of its functional connections with the other brain structures (apallic syndrome, chronic vegetative status). The patient is in bed and all living systems might operate for a long period of time [30].

Biologic Death is the death of all cells of the human body (cell and organ death).

Brain Death is a result of stopping of brain circulation and irreversible changes of brain functions and brain stem presenting with three main events: coma, absence of stem reflexes and apnea [2, 13, 30]. As of the current moment in medical regulation (Regulation No.14/ 15.04.2004) and in term of transplantation, the locally accepted definition of this condition is: “irreversible and permanent cessation of all functions of the brain in the presence of heartbeat”.

Chain of Survival – this terminology is used for the activities after the СА until patient’s survival with the aim of:
• Early recognition of the event and taking measures for preventing CA;
• Early CPR in term to prolong the time of reversible phase of CA;
• Early defibrillation for recovering of cardiac rhythm;
• Post-resuscitation care for restoring the quality of life of the patient.

Modern concepts for survival evaluation after cardiac arrest

The cardiac arrest is a global, socially significant medical condition associated with high mortality rate and severe disability. It is characterized with a high variety of etiological factors and pathogenesis which define the possibility for recovering after CPR. The efforts for achieving better therapeutic results have been focused in two directions:
• Predictors of CPR outcomes;
• Defining of risk factors causing СА.

For that purpose retrospective analyses, meta-analyses and prospective studies have been performed and a lot of data had been collected, grouped by different parameters. The survival rate after CA is assessed in analyses considering: time; demographic data; types of CA; risk factors and assessment of CPR outcome (quantitative and qualitative parameters).

A similar approach is observed in most survival analyses – grouping of results by absolute value and by percent. The largest observational study is published in 1977, 1063 patients have been observed for 10 years analyzing the survival after CA in a Multiprofile Hospital excluding Coronary and Intensive Care Units. The reported results after CPR were, as follows: died after unsuccessful CPR (65%); died later in hospital (23%) and discharged alive (7%)[25, 28]. In an observational study published in 2001, 557 patients are analyzed for a 5-year period in Cardiology Hospital and the survival by time after CA has been assessed in details. The patients after CPR have been divided in the following groups (in absolute value and in %): immediate death (52.4%); death in 24 hours (17%); survival more than 24 hours (30,6%) – death in 1 month (14.4%) and survival after 1 month (16.2%) incl. death up to and above 1 year [28]. In England, 2002 the largest up till now single site study has been completed in 1633 patients after CPR performed out-of-hospital by an emergency team. The survival has been assessed in three groups: until hospitalization (41%); up to 24 hours (28%) and until discharge (19%) [6]. Completed studies demonstrate a high level of immediate death after CPR. The rest of the parameters could not be analyzed due to the following reasons: the analyzed data are from CPR which is performed in different circumstances; the included patients have different characteristics; statistically significant differences are observed in results from similar parameters as well as the existence of heterogeneous, incomparable parameters.

During a survival analysis by demographic data the patients have been divided by gender and by age (0-10 y/a, 11-60 y/a, and above 70 y/a) and the mean age of patients after CA (47.2 years) has been reported. The highest level of mortality has been observed in the extreme age groups (0-10 y/a – no survival; 11-60 y/a – 19.7% and above 70 – 5.7%). No statistically significant difference in mortality rate by gender has been observed [28]. In a retrospective analysis of the influence of demographic factors on survival it was identified that the age, the gender and the race were not predictors of survival and those factors were not critical for the efficacy of reanimation [17].

In survival analysis within a month by CA forms it is identified that the elderly patients (over 75 years) the better prognosis is for patients with asystole followed by those with VF. An inverse correlation has been observed for patients survived more than a month and better prognosis is for VF compared to asystole. No patient with PEA survived more than a month [28]. In 2006 an analysis was published based on observation of 36 902 adults and 882 children experienced pulseless СА, using data from the National Registry of CPR (NRCPR), USA. First documented arrhythmias in both age groups have been asystole or PEA whereas the children had better survival than adults. Respiratory failure with progressing tissue hypoxia and acidosis leading to circulatory shock is the typical cause of PCA in children rather than a coronary disease. The rhythm of pulseless CA in children usually progresses from bradyarrhythmia to asystole or PEA and rarely to VF [21]. The first documented rhythm of CA in the age group above 75 y/a is VF or PVT in 36% of hospitalized and in 24% of ambulatory patients. VF or PVT is registered in 25% during the reanimation stages or after the first documented rhythm of asystole or PEA [22]. Relation is observed between VF and PVT as well as between PЕА and asystole in case identical risk factors are present. For example, acute myocardial infarction is associated with firstly documented VF or PVT; the acute respiratory failure and hypotension correlate with PЕА and asystole [20]. Mild therapeutic hypothermia improves survival after CA providing brain neuroprotection. The up-to-date guidelines of AHA and the International Liaison Committee on Resuscitation (ILCOR) refer to the cases with initially registered rhythm of VF, some studies even demonstrate worse results for another types of rhythm [1].

Quantitative assessment of CPR outcome is formed based on independent risk factors. Morbidity factors before CA having negative impact on CPR outcome are: obesity; cardiovascular diseases more frequently than those with surgical or respiratory etiology. Concomitant neurologic disease, trauma or electrolyte disbalance are associated with better survival rate [17]. According to the opinion of some authors the factors during the reanimation have greater impact on the CPR outcome than the comorbidity factors before CA [6]. Risk factors during reanimation predicting the poor outcome of CPR are: initiation of CPR later than 3 min after СА; CPR duration over 15 min; СА with asystole or PEA in adults over 75 y/a and unwitnessed СА [28]. A correlation is observed between the high levels of blood glucose after reanimation due to cardiac arrest and the severity of the neurological deficiency. In a large randomized trial in patients admitted to a general intensive care unit the conventional glucose control (≤10mmol/L) is compared with the intensive one (4,5–6.0 mmol/L). The 90-day mortality is higher in the group of patients with intensive glucose control [10]. According to the recent guidelines for adult patients after ROSC the level of blood glucose should be kept ≤10mmol/L and the intensive glucose control should be avoided due to the risk of hypoglycemia. Hyperthermia is common during the first 48 hours of IRS and associates with poor outcomes. Up to now no randomized, controlled trials have assessed the effectiveness of pyrexia control. [22].

For the purpose of a Quantitative assessment of CPR outcome, different scales and indexes have been developed for predicting of survival after CPR. All of them include a complex of risk factors resulting in CA as well as influencing factors during the reanimation. Most of the attempts for developing a scale or index for predicting of survival after CA have failed. Examples are: Resuscitation Predictor Scoring Scale (RPS); Karnofsky Performance Status (KPS) and Kentsch index [4, 6, 15, 16, 19]. RPS scale has not been widely implemented due to the fact that even a small present possibility for survival could not justify stopping of CPR. KPS has not demonstrated an accurate assessment of the physical performance thus it has not been implemented, including in patients with oncology diseases [4, 15]. The low sensitivity and specificity of the Kentsch index have been proved in a retrospective analysis for both – in-hospital and out-hospital patients thus it has not been implemented in the clinical practice [16, 19]. Pre Arrest Morbidity (РАМ) index was developed in 1989 including the following parameters which are associated with higher in-hospital mortality rate: pneumonia, hypotension, renal insufficiency, cancer and sedentary lifestyle; it has a negative relation with survival until discharge. During the hospitalization and 0 score of the index, the survival rate is up to 50.6% and the patients die with a score above 8 [3, 11]. Prognostics After Resuscitation (PAR index) is developed in 1992 by modification of PAM index based on a meta-analysis of 14 studies (2643 patients) assessing the in-hospital survival after CPR. The modified index is a better predictor of ineffective CPR [8, 23]. The reliability of PAR index as a predictor of CPR failure has not been sufficiently clinically proven and some controversial results have been reported [23, 24, 27]. No index or scale is available in order to be used in the practice in countries where a decision for “do” or “do not” resuscitate is defined by the locally acting regulation which requires searching for more reliable predictors of CPR outcome.

Qualitative assessment of CPR outcome after СА is a complex of diagnostic methods for prognosis of neurologycal status, cardiologycal status and respiration. Prognostic factors for poor neurological outcome are based on: electroencephalographic data; somatosensory evoked potentials; bispectral index (BIS); increased pressure of the cerebrospinal fluid; biochemical markers: neuron-specific enolase (NSE) and S-100β; sonography for cerebral circulatory arrest; Computed Tomography; CT angiography; digital subtraction angiography; isotope angiography and isotope scanning of brain [2, 22, 27]. A reliable prognosis could not be done based on only clinical signs, electrophysiological tests and biomarkers. Cerebral Performance Categories and Glasgow Outcome Scoring System are scales developed for assessment of neurologycal functions after CA, however they are not applicable in the first hours/days after SCA [22, 27].

Summary

Over the past years patients’ treatment after CA has considerably progressed due to improved organization and development of the CPR procedure. Independently of the registered progress, the mortality rate remains high, reaching 75% in developed countries [12]. The focus is directed on identifying the reliable and clear predictors of survival after CA and even before the CPR. For this purpose, it is necessary to grant conditions for retrospective and meta-analysis on a large-scale database since such has been established only in USA (NRCPR). Information obtained by new, prospective and controlled clinical trials and observations is expected as well.

REFERENCES

1. Рамшев Н, Енева В, Андонова С, Титянова Е. Терапевтична хипотермия в медицината – защо и кога. Невросонология и мозъчна хемодинамика 8, 2012:87–94.
2. Титянова Е, Каракънева С, Петров Н, Кирвиков 3, Петков А, Петров Л, Димова Р. Ултразвукова характеристика при мозъчна смърт. Невросонография и мозъчна хемодинамика 4, 2008:69–73.
3. Bedell S, Delbanco T, Cook E, Epstein F. Survival after cardiopulmonary resuscitation in the hospital. N Engl J Med 309, 1983:569–576.
4. Brezinski D, Stone P, Muller J, Tofler G, Davis V, Parker C, Hartley L, Braunwald E. Prognostic significance of the Karnofsky performance status score in patients with acute myocardial infarction: comparison with the left ventricular ejection fraction and the exercise treadmill test performance. Am Heart J 121, 1991:1374–1381.
5. Carden D, Granger D. Pathophysiology of ischaemia-reperfusion injury. J Pathol 190, 2000:255–266.
6. Cooper C, Evans C. Resuscitation Predictor Scoring Scale for inhospital cardiac arrests. Emerg Med 20, 2003:6–9.
7. Deo R, Albert C, Epidemiology and Genetics of Sudden Cardiac Death. Circulation 125, 2012:620–637.
8. Ebell M. Prearrest predictors of survival following in-hospital cardiopulmonary resuscitation: a meta-analysis. J Fam Pract 34, 1992:551–558.
9. Edward C, Jauch M, Edward C, Jeffrey L, Harold P, Bruno A, Connors J, Demaerschalk B, Khatri P, McMullan P, Qureshi A, Rosenfield K, Scott P, Summers D, Wang D, Wintermark M, Yonas H on behalf of the American Heart Association Stroke Council, Council on Cardiovascular Nursing, Council on Peripheral Vascular Disease, and Council on Clinical Cardiology,Guidelines for the Early Management of Patients With Acute Ischemic Stroke, A Guideline for Healthcare Professionals From the American Heart Association/American Stroke Association. Stroke 44, 2013:870–947.
10. Finfer S, Chittock D, Su S, Foster D, Dhingra V, Bellomo R, Cook D, Dodek P, Henderson W, Hebert P, Heritier S, Heyland D, McArthur C, McDonald E, Mitchell I, Myburgh J, Norton R, Potter J, Robinson B, Ronco J. Intensive versus conventional glucose control in critically ill patients. N Engl J Med 360, 2009:1283–1297.
11. George J, Folk B, Crecelius P, Campbell W. Pre-arrest morbidity and other correlates of survival after in-hospital cardiopulmonary arrest. Am J Med 87, 1989:28–34.
12. Girotra S, Nallamothu B, Spertus J, Li Y, Krumholz H, Chan P. Get with the Guidelines–Resuscitation Investigators Trends in Survival after In-Hospital Cardiac Arrest. N Engl J Med 367, 2012:1912–1920.
13. Goila А, Pawar М. The diagnosis of brain death. Indian J Crit Care Med 13, 2009:7–11.
14. Kannel W, Wilson P, D'Agostino R, Cobb J. Sudden coronary death in women. Am. Heart J 136, 1998:205–212.
15. Karnofsky D, Ablemann W, Craver L, Burchenal J. The use of nitrogen mustards in the palliative treatment of cancer. Cancer 1, 1948:634–641.
16. Kentsch M, Stendel M, Berkel H, Mueller-Esch G. Early prediction of prognosis in out-of-hospital cardiac arrest. Intensive Care Med 16, 1990:378–383.
17. Koldobskiy D, Groves S, Scharf S, Cowan M. Validation of Factors Affecting the Outcome of Cardiopulmonary Arrest in a Large, Urban, Academic Medical Center. Journal of Critical Care Medicine 1, 2014:1-7.
18. Kuisma M, Alaspaa A (July 1997). Out-of-hospital cardiac arrests of non-cardiac origin. Eur Hear J 18, 1997:1122–1128.
19. Martens P, Mullie A, Buylaert W, Calle P, Van Hoeyweghen R. Early prediction of non-survival for patients suffering cardiac arrest: a word of caution. Intens Care Med 18, 1992:11–14.
20. Meaney P, Nadkarni V; Kern K, Indik J, Halperin H, Berg R. Rhythms and outcomes of adult in-hospital cardiac arrest. Crit Care Med 38, 2009:101–108.
21. Nadkarni V, Larkin G, Peberdy M, Carey S, Kaye W, Mancini M, Nichol G, Lane-Truitt T, Potts J, Ornato J, Berg R. For the National Registry of Cardiopulmonary Resuscitation Investigators. JAMA 295, 2006:50–57.
22. Nolan J, Soar J, Zideman D, Biarent D, Bossaert L, Deakin C, Koster R, Wyllie J, Bottiger B, on behalf of the ERC Guidelines Writing Group. European Resuscitation Council Guidelines for Resuscitation 2010 Section 1. Executive summary. Resuscitation 81, 2010:1219–1276.
23. O’Keefe S, Ebell M. Prediction of failure to survive following in-hospital cardiopulmonary resuscitation: comparison of two predictive instruments. Resuscitation 28, 1994:21–25.
24. O’Keefe S, Redahan C, Keane P, Daly K. Age and other determinants of survival after in-hospital cardiopulmonary resuscitation. Quar. J Med 81, 1991:1005–1010.
25. Peatfield R, Taylor D, Sillet A, McNicol M. Survival after cardiac arrest in hospital. Lancet 1, 1977:1223–1225.
26. Sans S, Kesteloot H, Kromhout D. The burden of cardiovascular diseases mortality in Europe. Task force of the European Society of Cardiology on cardiovascular mortality and morbidity statistics in Europe. Eur Heart J 18, 1997:1231–1248.
27. Temple A, Porter R. Predicting neurological outcome and survival after cardiac arrest. Critical Care & Pain 12, 2012:283–287.
28. Timerman A, Sauaia N, Piegas L, Ramos R, Gun C, Santos E, Bianco A, Sousa J. Prognostic Factors of the Results of Cardiopulmonary Resuscitation in a Cardiology Hospital. Arq Bras Cardiol 77, 2001:142–160.
29. Weisfeldt M, Becker L. Resuscitation after cardiac arrest: a 3-phase time-sensitive model. JAMA 288, 2002:3035–3038.
30. Wijdicks E, Varelas P, Gary S, Gronseth D, Greer D. Evidence-based guideline update: Determining brain death in adults: Report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology 74, 2010:1911–1918. 31. Zheng Z, Croft J, Giles W, Mensah G. Sudden cardiac death in the United States, 1989 to 1998. Circulation 104, 2001:2158–2163.