Cardiac arrest

From Wikipedia, the free encyclopedia

Cardiac arrest
Classification and external resources
ICD-10 I46.
ICD-9 427.5
MeSH D006323

A cardiac arrest, also known as cardiorespiratory arrest, cardiopulmonary arrest or circulatory arrest, is the abrupt cessation of normal circulation of the blood due to failure of the heart to contract effectively during systole.[1]

A cardiac arrest is different from (but may be caused by) a heart attack or myocardial infarction, where blood flow to the still-beating heart is interrupted.

"Arrested" blood circulation prevents delivery of oxygen to all parts of the body. Cerebral hypoxia, or lack of oxygen supply to the brain, causes victims to lose consciousness and to stop normal breathing, although agonal breathing may still occur. Brain injury is likely if cardiac arrest is untreated for more than 5 minutes,[2] although new treatments such as induced hypothermia have begun to extend this time.[3][4] To improve survival and neurological recovery immediate response is paramount.[5]

Cardiac arrest is a medical emergency that, in certain groups of patients, is potentially reversible if treated early enough (See "Reversible causes" below). When unexpected cardiac arrest leads to death this is called sudden cardiac death (SCD).[1] The primary first-aid treatment for cardiac arrest is cardiopulmonary resuscitation (commonly known as CPR) which provides circulatory support until availability of definitive medical treatment, which will vary dependant on the rhythm the heart is exhibiting, but often requires defibrillation.

Contents

[edit] Characteristics and diagnosis

Cardiac arrest is an abrupt cessation of pump function (evidenced by absence of a palpable pulse) of the heart that with prompt intervention could be reversed, but without it will lead to death.[1]

However, due to inadequate cerebral perfusion, the patient will be unconscious and will have stopped breathing. The main diagnostic criterion to diagnose a cardiac arrest (as opposed to respiratory arrest, which shares many of the same features) is lack of circulation, however there are a number of ways of determining this.

In many cases, lack of carotid pulse is the gold standard for diagnosing cardiac arrest, but lack of a pulse (particularly in the peripheral pulses) may be a result of other conditions (e.g. shock), or simply an error on the part of the rescuer. Studies have shown that rescuers often make a mistake when checking the carotid pulse in an emergency, whether they are healthcare professionals[6][7] or lay persons.[8]

Owing to the inaccuracy in this method of diagnosis, some bodies such as the European Resuscitation Council (ERC) have de-emphasised its importance. The Resuscitation Council (UK), in line with the ERC's recommendations and those of the American Heart Association,[9] have suggested that the technique should be used only by healthcare professionals with specific training and expertise, and even then that it should be viewed in conjunction with other indicators such as agonal respiration.[10]

Various other methods for detecting circulation have been proposed. Guidelines following the 2000 International Liaison Committee on Resusciation (ILCOR) recommendations were for rescuers to look for "signs of circulation", but not specifically the pulse [9]. These signs included coughing, gasping, colour, twitching and movement.[11] However, in face of evidence that these guidelines were ineffective, the current recommendation of ILCOR is that cardiac arrest should be diagnosed in all casualties who are unconscious and not breathing normally.[9]

Following initial diagnosis of cardiac arrest, healthcare professionals further categorise the diagnosis based on the ECG/EKG rhythm. There are 4 rhythms which result in a cardiac arrest. Ventricular fibrillation (VF/VFib) and pulseless ventricular tachycardia (VT) are both responsive to a defibrillator and so are colloquially referred to as "shockable" rhythms, whereas asystole and pulseless electrical activity (PEA) are non-shockable. The nature of the presenting hearth rhythm suggests different causes and treatment, and is used to guide the rescuer as to what treatment may be appropriate[10] (see Advanced life support and Advanced cardiac life support, as well as the causes of arrest below).

[edit] Causes of cardiac arrest

Cardiac arrest is synonymous with Clinical death. All disease processes leading to death have a period of (potentially) reversible cardiac arrest: the causes of arrest are, therefore, numerous. However, many of these conditions, rather than causing an arrest themselves, promote one of the "Reversible causes" (see below), which then triggers the arrest (e.g. choking leads to hypoxia which in turn leads to an arrest). In some cases, the underlying mechanism cannot be overcome, leading to an unsuccessful resuscitation.

Among adults, ischemic heart disease is the predominant cause of arrest.[12] At autopsy 30% of victims show signs of recent myocardial infarction[citation needed]. Other cardiac conditions potentially leading to arrest include structural abnormalities, arrhythmias and cardiomyopathies. Non-cardiac causes include infections, overdoses, trauma and cancer, in addition to many others.

[edit] Reversible causes

Cardiopulmonary resuscitation (CPR), including adjunctive measures such as defibrillation, intubation and drug administration, is the standard of care for initial treatment of cardiac arrest. However, most cardiac arrests occur for a reason, and unless that reason can be found and overcome, CPR is often ineffective, or if it does result in a return of spontaneous circulation, this is short lived. [10]. As highlighted above, a variety of disease processes can lead to a cardiac arrest, however they usually boil down to one or more of the "Hs and Ts".[13][14][15]

[edit] H's

  • Hypovolemia - A lack of circulating body fluids, principally blood volume. This is usually (though not exclusively) caused by some form of bleeding, anaphylaxis, or pregnancy with gravid uterus. Peri-arrest treatment includes giving IV fluids and blood transfusions, and controlling the source of any bleeding - by direct pressure for external bleeding, or emergency surgical techniques such as esophageal banding, gastroesophageal balloon tamponade (for treatment of massive GI bleeding such as in esophageal varices), thoracotomy in cases of penetrating trauma or significant shear forces applied to the chest, or exploratory laparotomy in cases of penetrating trauma, spontaneous rupture of major blood vessels, or rupture of a hollow viscus in the abdomen.
  • Hypoxia - A lack of oxygen delivery to the heart, brain and other vital organs. Rapid assessment of airway patency and respiratory effort must be performed. If the patient is mechanically ventilated, the presence of breath sounds and the proper placement of the endotracheal tube should be verified. Treatment may include providing oxygen, proper ventilation, and good CPR technique. In cases of carbon monoxide poisoning or cyanide poisoning, hyperbaric oxygen may be employed after the patient is stabilized.
  • Hydrogen ions (Acidosis) - An abnormal pH in the body as a result of lactic acidosis which occurs in prolonged hypoxia and in severe infection, diabetic ketoacidosis, renal failure causing uremia, or ingestion of toxic agents or overdose of pharmacological agents, such as aspirin and other salicylates, ethanol, ethylene glycol and other alcohols, tricyclic antidepressants, isoniazid, or iron sulfate. This can be treated with proper ventilation, good CPR technique, buffers like sodium bicarbonate, and in select cases may require emergent hemodialysis.
  • Hyperkalemia or Hypokalemia - Both excess and inadequate potassium can be life-threatening. A common presentation of hyperkalemia is in the patient with end-stage renal disease who has missed a dialysis appointment and presents with weakness, nausea, and broad QRS complexes on the electrocardiogram. (Note however that patients with chronic kidney disease are often more tolerant of high potassium levels as their body often adapts to it.) The electrocardiogram will show tall, peaked T waves (often larger than the R wave) or can degenerate into a sine wave as the QRS complex widens. Immediate initial therapy is the administration of calcium, either as calcium gluconate or calcium chloride. This stabilizes the electrochemical potential of cardiac myocytes, thereby preventing the development of fatal arrhythmias. This is, however, only a temporizing measure. Other temporizing measures may include nebulized albuterol, intravenous insulin (usually given in combination with glucose, and sodium bicarbonate, which all temporarily drive potassium into the interior of cells. Definitive treatment of hyperkalemia requires actual excretion of potassium, either through urine (which can be facilitated by administration of loop diuretics such as furosemide) or in the stool (which is accomplished by giving sodium polystyrene sulfonate enterally, where it will bind potassium in the GI tract.) Severe cases will require emergent hemodialysis. The diagnosis of hypokalemia (not enough potassium) can be suspected when there is a history of diarrhoea or malnutrition. Loop diuretics may also contribute. The electrocardiogram may show flattening of T waves and prominent U waves. Hypokalemia is an important cause of acquired long QT syndrome, and may predispose the patient to torsades de pointes. Digitalis use may increase the risk that hypokalemia will produce life threatening arrhythmias. Hypokalemia is especially dangerous in patients with ischemic heart disease.
  • Hypothermia - A low core body temperature, defined clinically as a temperature of less than 35 degrees Celsius (95 degrees Fahrenheit). The patient is re-warmed either by using a cardiac bypass or by irrigation of the body cavities (such as thorax, peritoneum, bladder) with warm fluids; or warmed IV fluids. CPR only is given until the core body temperature reached 30 degrees Celsius, as defibrillation is ineffective at lower temperatures. Patients have been known to be successfully resuscitated after periods of hours in hypothermia and cardiac arrest, and this has given rise to the often-quoted medical truism, "You're not dead until you're warm and dead."
  • Hypoglycemia or Hyperglycemia - Low blood glucose from overdose of oral hypoglycemics such as sulfonylureas, or overdose of insulin. Rare endocrine disorders can also cause unexpected hypoglycemia. Generally, hyperglycemia is itself not fatal, however DKA will cause pH to drop, and nonketotic hyperosmolar coma leads to a severely hypovolemic state. Hypoglycemia is corrected rapidly by intravenous administration of concentrated glucose (typically 25 ml of 50% glucose in adults, but in children 25% glucose is used, and in neonates 10% glucose is used.) However, the patient will often require a continuous intravenous drip until the causative agent is completely metabolized. In DKA, the goal is correction of acidosis. In NKH, the goal is adequate fluid resuscitation.

[edit] T's

Checking respiration.
Checking respiration.
Checking carotid pulse.
Checking carotid pulse.
Insulfation mouth-to-mouth.
Insulfation mouth-to-mouth.

[edit] Treatment

[edit] Out of hospital arrest

Most out-of-hospital cardiac arrests occur following a myocardial infarction (heart attack), and present initially with a heart rhythm of ventricular fibrillation. The patient is therefore likely to be responsive to defibrillation, and this has become the focus of pre-hospital interventions. Several organisations promote the idea of a "chain of survival", of which defibrillation is a key step. The links are:

  • Early recognition - If possible, recognition of illness before the patient develops a cardiac arrest will allow the rescuer to prevent its occurrence. Early recognition that a cardiac arrest has occurred is key to survival - for every minute a patient is in cardiac arrest, their chances of survival drop by roughly 10% [10]
  • Early CPR - This buys time by keeping vital organs perfused with oxygen whilst waiting for equipment and trained personnel to reverse the arrest. In particular, by keeping the brain supplied with oxygenated blood, chances of neurological damage are decreased.
  • Early defibrillation - This is the only effective treatment for ventricular fibrillation, and also has benefit in ventricular tachycardia[10] and should be employed in such cases if the patient has signs of hemodynamic compromise, or if the patient has pulseless ventricular tachycardia. If defibrillation is delayed, then the rhythm is likely to degenerate into asystole, for which outcomes are markedly worse.
  • Early advanced care - Early Advanced Cardiac Life Support is the final link in the chain of survival.

If one or more links in the chain are missing or delayed, then the chances of survival drop significantly. In particular, bystander CPR is an important indicator of survival: if it has not been carried out, then resuscitation is associated with very poor results. Paramedics in some jurisdictions are authorised to abandon resuscitation altogether if the early stages of the chain have not been carried out in a timely fashion prior to their arrival.

Because of this, considerable effort has been put into educating the public on the need for CPR. In addition, there is increasing use of public access defibrillation. This involves placing automated external defibrillators in public places, and training key staff in these areas how to use them. This allows defibrillation to take place prior to the arrival of emergency services, and has been shown to lead to increased chances of survival. In addition, it has been shown that those who suffer arrests in remote locations have worse outcomes following cardiac arrest [16]: these areas often have first responder schemes, whereby members of the community receive training in resuscitation and are given a defibrillator, and called by the emergency medical services in the case of a collapse in their local area.

[edit] Hospital treatment

Treatment within a hospital usually follows advanced life support protocols. In the US, non-traumatic adult resuscitation is described by ACLS(advanced cardiac life support), pediatric resuscitation is described by PALS (pediatric advanced life support), and neonatal resusciation is described by NALS (neonatal advanced life support.) Depending on the diagnosis, various treatments are offered, ranging from defibrillation (for ventricular fibrillation or ventricular tachycardia) to surgery (for cardiac arrest which can be reversed by surgery - see causes of arrest, above) to medication (for asystole and PEA). All will include CPR.

While specific details may vary, all hospitals have protocols as to how resuscitations should be performed in patients, visitors, or employees who have arrested unexpectedly in the hospital. These protocols are often initiated by a Code Blue, which usually denotes impending or acute onset of cardiac arrest or respiratory failure, although in practice, Code Blue is often called in less life-threatening situations that require immediate attention from a physician.

If not already done, a definitive airway will be establish by the placement of an endotracheal tube which is then attached to a mechanical ventilator.

Cardiac arrest is generally divided into two cases: presence of disorganized mechanical cardiac activity, or complete absence of mechanical cardiac activity.

Disorganized mechanical cardiac activity includes ventricular fibrillation and hemodynamically unstable or pulseless ventricular tachycardia. This also includes torsade de pointes. These must all be treated primarily with defibrillation. Advanced cardiac life support algorithms also detail the stepwise administration of epinephrine, vasopressin, the antiarrhythmic agent amiodarone, as well as attempts to correct possible underlying causes.

Complete absence of mechanical cardiac activity includes asystole and pulseless electrical activity. This is treated entirely with pharmacologic agents, specifically epinephrine and atropine. However, resuscitation is rarely successful without effective treatment of the underlying cause.

Patients that survive cardiac arrest are said to have return of spontaneous circulation (ROSC). When these patients remain comatose, survival and neurologic damage can be improved by cooling patients. Mild Therapeutic Hypothermia can be induced by a variety of methods. The most beneficial protocols are still being determined.

[edit] Peri-arrest period

The period (either before or after) surrounding a cardiac arrest is known as the peri-arrest period. During this period the patient is in a highly unstable condition and must be constantly monitored in order to halt the progression or repeat of a full cardiac arrest. The preventative treatment used during the peri-arrest period depends on the causes of the impending arrest and the likelihood such an event occurring.

[edit] Prognosis

The out-of-hospital cardiac arrest (OHCA) has a worse survival rate (2-8% at discharge and 8-22% on admission), than an in-hospital cardiac arrest (15% at discharge). The principal determining factor is the initially documented rhythm. Patients with VF/VT have 10-15 times more chance of surviving than those suffering from pulseless electrical activity or asystole (as they are sensitive to defibrillation, whereas asystole and PEA are not).[citation needed]

Since mortality in case of OHCA is high, programs were developed to improve survival rate. A study by Bunch et al. showed that, although mortality in case of ventricular fibrillation is high, rapid intervention with a defibrillator increases survival rate to that of patients that did not have a cardiac arrest.[12][17]

Survival is mostly related to the cause of the arrest (see above). In particular, patients who have suffered hypothermia have an increased survival rate, possibly because the cold protects the vital organs from the effects of tissue hypoxia. Survival rates following an arrest induced by toxins is very much dependent on identifying the toxin and administering an appropriate antidote. A patient who has suffered a myocardial infarction due to a blood clot in the left coronary artery has a lower chance of survival as it cuts of the blood supply to most of the left ventricle (the chamber which must pump blood to the whole of the systemic circulation).

Cobbe et al (1996) conducted a study into survival rates from out of hospital cardiac arrest. 14.6% of those who had received resuscitation by ambulance staff survived as far as admission to an acute hospital ward. Of these, 59.3% died during that admission, half of these within the first 24 hours. 46.1% survived to hospital discharge (this is 6.75% of those who had been resuscitated by ambulance staff), however 97.5% suffered a mild to moderate neurological disability, and 2% suffered a major neurological disability. Of those who were successfully discharged from hospital, 70% were still alive 4 years after their discharge.[18]

Ballew (1997) performed a review of 68 earlier studies into prognosis following in-hospital cardiac arrest. They found a survival to discharge rate of 14% (this roughly double the rate for out of hospital arrest found by Cobbe et al (see above)), although there was a wide range (0-28%).[19]

[edit] Prevention

With positive outcomes following cardiac arrest so unlikely, a great deal of effort has been spent in finding effective strategies to prevent cardiac arrest.

As noted above, one of the prime causes of cardiac arrest outside of hospital is ischemic heart disease. Vast resources have been put into trying to reduce cardiovascular risks across much of the developed world. In particular schemes have been put in place to promote a healthy diet and exercise. For people considered to be particularly at risk of heart disease, measures such as blood pressure control, prescription of cholesterol lowering medications, and other medico-therapeutic interventions, have been widely used. A magnesium deficiency, or lower levels of magnesium, can contribute to heart disease and a healthy diet that contains adequte magnesium may help prevent heart disease.[20] Magnesium can be used to enhance long term treatment, so it may be effective in long term prevention.

Patients in hospital are far less likely to have a cardiac arrest caused of primary cardiac origin, and hence present in asystole or PEA, and have bleak outcomes.[citation needed] Extensive research has shown that patients in general wards often deteriorate for several hours or even days before a cardiac arrest occurs[10][21]. This has been attributed to a lack of knowledge and skill amongst ward based staff, in particular a failure to carry out measurement of the respiratory rate, which is often the major predictor of a deterioration[10] and can often change up to 48 hours prior to a cardiac arrest. In response to this, many hospitals now have increased training for ward based staff. A number of "early warning" systems also exist which aim to quantify the risk which patients are at of deterioration based on their vital signs and thus provide a guide to staff. In addition, specialist staff are being utilised more effectively in order to augment the work already being done at ward level. These include:

  • Crash teams (also known as code teams) - These are designated staff members who have particular expertise in resuscitation, who are called to the scene of all arrests within the hospital.
  • Medical emergency teams - These teams respond to all emergencies, with the aim of treating the patient in the acute phase of their illness in order to prevent a cardiac arrest.
  • Critical care outreach - As well as providing the services of the other two types of team, these teams are also responsible for educating non-specialist staff. In addition, they help to facilitate transfers between intensive care/high dependency units and the general hospital wards. This is particularly important, as many studies have shown that a significant percentage of patients discharged from critical care environments quickly deteriorate and are re-admitted - the outreach team offers support to ward staff to prevent this from happening.

[edit] Implantable cardioverter defibrillators

A technologically based intervention to prevent further cardiac arrest episodes is the use of an implantable cardioverter-defibrillator (ICD). This device is implanted in to the patient. They act as an instant defibrillator in the event of arrhythmia. Note that standalone ICDs do not have any pacemaker functions, but they can be combined with a pacemaker, and modern versions also have advanced features such as anti-tachycardic pacing as well as synchronized cardioversion. A recent study by Birnie et al. at the University of Ottawa Heart Institute has demonstrated that ICDs are underused in both the United States and Canada.[22] An accompanying editorial by Simpson explores some of the economic, geographic, social and political reasons for this.[23] Patients who are most likely to benefit from the placement of an ICD are those with severe ischemic cardiomyopathy (with systolic ejection fractions less than 30%) as demonstrated by the MADIT-II trial.[24]

[edit] Ethical issues

Cardiopulmonary resuscitation and advanced cardiac life support are not always in a person's best interest. This is particularly true in the case of terminal illnesses when resuscitation will not alter the outcome of the disease. Properly performed CPR often fractures the rib cage, especially in older patients or those suffering from osteoporosis. Defibrillation, especially repeated several times as called for by ACLS protocols, may also cause electrical burns.

Some people with a terminal illness choose to avoid such measures and die peacefully. People with views on the treatment they wish to receive in the event of a cardiac arrest should discuss these views with both their doctor and with their family. A patient may ask their doctor to place a do not resuscitate (DNR) order in the medical record. Alternatively, in many jurisdictions, a person may formally state their wishes in an advance directive or advance health directive.

[edit] See also

[edit] References

  1. ^ a b c Harrison's Principles of Internal Medicine 16th Edition, The McGraw-Hill Companies, ISBN 0-07-140235-7
  2. ^ Safar P (1986). "Cerebral resuscitation after cardiac arrest: a review". Circulation 74: IV138-153. Lippincott Williams & Wilkins. 
  3. ^ Holzer M, Behringer W (2005). "Therapeutic hypothermia after cardiac arrest". Current Opinion in Anaestesiology 18: 163-168. Lippincott Williams & Wilkins. 
  4. ^ Safar P et al (1996). "Improved cerebral resuscitation from cardiac arrest in dogs with mild hypothermia plus blood flow promotion". Stroke 27: 105-113. Lippincott Williams & Wilkins. 
  5. ^ Irwin and Rippe's Intensive Care Medicine by Irwin and Rippe, Fifth Edition (2003), Lippincott Williams & Wilkins, ISBN 0-7817-3548-3
  6. ^ Flesche CW, Breuer S, Mandel LP, Breivik H, Tarnow J. (1994) The ability of health professionals to check the carotid pulse. Circulation Vol. 90: I–288.
  7. ^ F. Javier Ochoa, E. Ramalle-Gomara, J.M. Carpintero et al. (1998) Competence of health professionals to check the carotid pulse. Resuscitation Vol. 37 pp. 173–175
  8. ^ Bahr, J., Klingler, H., Panzer, W., Rode, H., Kettler, D. (1997). Skills of lay people in checking the carotid pulse. Resuscitation. Vol. 35(1) pp. 23-26
  9. ^ a b c American Heart Association (2005) 2005 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation Vol. 112 pp. 19-34
  10. ^ a b c d e f g Resuscitation Council UK (2005). Resuscitation Guidelines 2005 London: Resuscitation Council UK.
  11. ^ St John Ambulance, St Andrew's Ambulance Association, British Red Cross (2002) (8th Ed.) First Aid Manual. London: Dorling Kindersley
  12. ^ a b Cardiac Resuscitation Mickey S. Eisenberg, M.D., Ph. D., and Terry J. Mengert, M.D. New England Journal of Medicine, Volume 344:1304-1313, April 26, 2001
  13. ^ ACLS: Principles and Practice. p. 71-87. Dallas: American Heart Association, 2003. ISBN 0-87493-341-2.
  14. ^ ACLS for Experienced Providers. p. 3-5. Dallas: American Heart Association, 2003. ISBN 0-87493-424-9.
  15. ^ "2005 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care - Part 7.2: Management of Cardiac Arrest." Circulation 2005; 112: IV-58 - IV-66.
  16. ^ Lyon, R.M, Cobbe, S.M., Bradley, J.M., Grubb, N.R. (2004)Surviving out of hospital cardiac arrest at home: a postcode lottery? Emergency Medical Journal Vol. 21 pp. 619-624
  17. ^ Long-Term Outcomes of Out-of-Hospital Cardiac Arrest after Successful Early Defibrillation T. Jared Bunch, M.D., Roger D. White, M.D., Bernard J. Gersh, M.B., Ch. B., Ryan A. Meverden, B.S., David O. Hodge, M.S., Karla V. Ballman, Ph. D., Stephen C. Hammill, M.D., Win-Kuang Shen, M.D., and Douglas L. Packer, M.D., New England Journal of Medicine, Volume 348:2626-2633, June 26, 2003
  18. ^ Survival of 1476 patients initially resuscitated from out of hospital cardiac arrest Stuart M Cobbe, Kirsty Dalziel, Ian Ford, Andrew K Marsden, British Medical Journal 1996;312:1633-1637 (29 June)
  19. ^ Recent advances: Cardiopulmonary resuscitation Kenneth A Ballew, British Medical Journal 1997;314:1462 (17 May)
  20. ^ Rosanoff, Andrea (PhD); Seelig, Mildred S (MD) (2004). "Comparison of Mechanism and Functional Effects of Magnesium and Statin Pharmaceuticals". Journal of the American College of Nutrition 23 (5): 501S–505S. 
  21. ^ Kause J, Smith G, Prytherch D, et al. (2004) A comparison of antecedents to cardiac arrests, deaths and emergency intensive care admissions in Australia and New Zealand, and the United Kingdom--the ACADEMIA study. Resuscitation Vol 62 pp. 275-82
  22. ^ Birnie, David H; Sambell, Christie; Johansen, Helen; Williams, Katherine; Lemery, Robert; Green, Martin S; Gollob, Michael H; Lee, Douglas S; Tang, Anthony SL (July 2007). "Use of implantable cardioverter defibrillators in Canadian and IS survivors of out-of-hospital cardiac arrest". Canadian Medical Association Journal 177 (1). 
  23. ^ Simpson, Christopher S (July 2007). "Implantable cardioverter defibrillators work - so why aren't we using them?". Canadian Medical Association Journal 177 (1). 
  24. ^ Moss, AJ; Cannom, DS; Daubert, JP; Hall, WJ; Higgins, SL; Klein, H; Wilber, D; Zareba, W; Brown, MW (1999). "Multicenter automatic defibrillator implantation Trial II (MADIT II) : Design and clinical protocol". Annals of non-invasive electrocardiology 4 (1): 83-91. 

[edit] External links