By MIKE McEVOY
Scenario One: Rescue 10 and Ambulance 15 are dispatched to a traffic crash with entrapment. Police on scene report a single vehicle into a concrete bridge abutment with a lone occupant who appears deceased. On arrival, EMT-firefighters find an approximately 16-year-old male driver lying across the front floor of a compact car with obvious open bilateral femur fractures, a rigid and distended abdomen, blood mixed with what appears to be cerebrospinal fluid draining from both ears, and no observable respirations or palpable pulses. The rescue truck officer cancels the responding ambulance, advises police that the driver is dead, and requests the medical examiner to the scene. Arriving one hour later, the medical examiner summons units back to the crash when he discovers the patient breathing with a barely palpable pulse. Resuscitation efforts begin; the patient is stabilized and transported to the trauma center across town, where he dies from massive head injuries two days later. The family contacts the news media and files a complaint with the State EMS Office.
Scenario Two: Engine 59 is dispatched to a reported obvious death in a low-income housing project. On arrival, fire paramedics encounter an elderly woman laying supine on her kitchen floor in an apparent advanced stage of decomposition. Large areas of skin have grotesquely peeled from the patient’s arms and torso and an overwhelmingly foul odor permeates the apartment. Medics contact the coroner to remove the body. Later that evening, the hospital morgue attendant summons a resuscitation team after the supposedly deceased patient begins moaning for help. The patient is admitted to the ICU with a massive Streptococcus pyrogenes (“flesh eating”) bacterial skin infection; she dies three days later. The national news media prominently carries the story.
Each year, 2.4 million Americans die, most of them in hospitals (61 percent) or in nursing homes (17 percent).1 Despite the small percentage of deaths (22 percent) that occur in the community, news stories of people mistakenly presumed dead by EMS appear regularly. These occurrences are embarrassing at the least and, more often, catastrophic for patients, families, firefighters, and department public relations. Although physicians in training receive formal instruction in death pronouncement, EMS training programs have no comparable component to prepare EMTs and paramedics for what many would consider a critical decision-making skill. The two cases described above are not fictional; both actually happened and could have been prevented. This article will give you the tools you need to be certain the next dead patient you encounter is really dead.
Live burial following mistaken pronouncement of death is believed to be one of the most widespread human fears,2 which explains why the media immediately seizes on mistaken pronouncements. Coffins in the 1800s were often outfitted with a variety of rescue devices designed to trigger an external signal if breathing or movement was detected inside. Concerned about live burials, New York State enacted legislation in 1899 requiring death pronouncement by a physician.3 Concerns increased with federal passage of the Uniform Anatomic Gift Act in 1968, allowing individuals to donate organs on their death.4 Worries about premature pronouncements to facilitate organ donation led a Harvard Medical School Committee to publish a report defining the characteristics of “irreversible coma,” now known as “brain death.” This report is still the basis for criteria used in patient care decisions involving organ transplantation.5
Despite published criteria defining brain death, consistent standards for death pronouncements were not assembled until the July 1981 report, “Defining Death: a report on the medical, legal, and ethical issues in the determination of death,” was published by the President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research.6 This report remains the basis for death pronouncement in the United States and describes criteria in existence at the time in other countries. Death pronouncement laws and regulations in all 50 states have been written using templates developed from this report outlined by the National Conference of Commissioners on Uniform State Laws in its 1980 document known as the Uniform Determination of Death Act.7
The 170-plus page “Defining Death” report suggested that the medical community in 1981 had many unanswered questions and concerns about death pronouncement. Most likely, the advances in life support equipment and technology that first appeared in the 1970s brought great uncertainties. Not only could technology keep apneic individuals breathing indefinitely, but artificial circulation support loomed right around the corner. It appeared that traditional criteria of pulselessness and apnea no longer recognized many deaths of ICU patients, whose ability to breathe spontaneously and circulate their own blood would never return. The issue here is the brain. Developing standardized criteria describing irreversible loss of all brain functions consumed most of the Commission’s energy and a substantial part of its report.
EMS providers don’t pronounce brain death, nor does a lone physician or nurse practitioner in the middle of the night. Such decisions take time, require specialized testing, and should be made by clinicians—such as neurologists—experienced in assessing the brain. Patients declared dead by EMS are those they find that way or those they cease resuscitating. For these individuals, the criteria are not incredibly complicated. “An individual with irreversible cessation of circulatory and respiratory function is dead. Cessation is recognized by an appropriate clinical exam,” whereas, “Irreversibility is recognized by persistent cessation of functions for an appropriate period of observation and/or trial of therapy.” (6, 133) The clinical exam suggested using the required confirmatory tests such as an electrocardiogram (ECG) in absence of responsiveness, heartbeat, and respiratory effort.
Irreversibility can be quite simple. If a patient is found decapitated, in an advanced state of decomposition, or with rigor mortis and significant dependent lividity, the observation period need not be prolonged. But be cautious: Not everything appearing nonsurvivable really is. Decapitation is an obvious fatal diagnosis, but massive internal injuries are often less so. Without the advantage of invasive imaging or ultrasound, a field provider cannot reliably diagnose a torn aorta or massive internal bleeding. Equal caution is warranted with what might appear to be massive head trauma. Experience offers the greatest advantage in difficult situations; seasoned providers have many past encounters on which to base their decisions. In any situation requiring field determination of death, the most senior medical providers on scene should be directly involved.
One additional consideration surrounding irreversibility is organ donation. For example, a single male operator ejected from a high-speed motor vehicle crash found pulseless and apneic in the roadway with a grossly deformed cervical spine and no other significant injuries will not walk, breathe, talk, or ever regain consciousness again. His neck is so badly broken that cord transection is obvious and would be clearly incompatible with life. Given the resources available, however, it may be reasonable to initiate resuscitation as a test for cardiovascular responsiveness with the potential that the patient may be an organ donor. Isolated massive head or spinal injuries, although clearly fatal, may provide opportunity to others through organ donation. Such considerations are mandatory in the hospital setting and ethically compelling for prehospital situations.
For patients who have failed resuscitation as a test for cardiovascular responsiveness, it may be necessary to observe them for the total duration of action of the medications administered. Virtually any experienced medic can recall a patient who regained vital signs after termination of resuscitation efforts. This is called the “Lazarus Phenomenon,” or autoresuscitation (AR), in medical literature and, although not commonly reported, is suspected to result from delayed resuscitation medication effects or intrathoracic pressure changes leading to spontaneous return of circulation once positive pressure ventilation is discontinued.8,9 Some of these patients survive to be transported; others live only transiently, but none live for very long.
AR also occurs in hospitals. To determine an appropriate observation period for AR following death, researchers recently reviewed the medical literature and found ARs occurring from seconds up to 33 minutes following cardio pulmonary resuscitation (CPR) termination. In the absence of CPR being provided, AR has never been reported when a patient was properly pronounced. The authors also could not locate any published reports of AR in children. Current recommendations call for observation ranging from two to 10 minutes following CPR cessation and, although the authors could not find a single instance where AR occurred more than seven minutes after CPR ended (when exact times and appropriate monitoring were reported), the authors could not support or refute the current 10-minute observation time without further study. (8) Following failed resuscitation attempts, including CPR and advanced cardiac life support, a 10-minute observation period is the current standard of practice to ascertain that AR has not occurred. Always incorporate this into prehospital field termination of resuscitation procedures.
When a patient is found dead and the death is not observed or expected or is sudden, and resuscitation is not provided, a much more detailed examination is needed, and longer observation is clearly indicated. Typically, physicians are taught to pronounce death using an examination that includes, at minimum, the general appearance of the body, no response to verbal or tactile stimulation, no pupillary light reflex (pupils fixed and dilated), absence of breath sounds, and absence of heart sounds. Deep, painful stimuli such as a sternal rubbing and nipple twisting are absolutely inappropriate. Some authors suggest testing for a corneal reflex (blinking when the cornea is touched with a gauze pad or cotton swab), but this is not included in these guidelines and is duplicative of pupillary reaction to light because both reflexes require some intact brainstem function to occur. When more sophisticated monitoring equipment is available such as an ECG, an oximeter able to measure low flow states, or portable ultrasound, use that equipment to confirm the physical exam findings.
A warning: Complicating conditions may resemble death. Drug intoxication can produce complete brain function cessation yet can be completely reversible. Total paralysis can also closely simulate death. Certain critical illnesses such as end-stage liver disease (hepatic coma) can make a live patient appear deceased. Drowning victims underwater for less than two hours have been known to survive. Shock states and profound hypothermia (body temperatures below 90°F) often require far more careful clinical exam because of reduced cerebral circulation. Children under the age of five bounce back remarkably better than adults after being unresponsive even for prolonged periods. These are red flag cases—situations where a “mistakenly declared dead” headline could include your name. However, looks can be deceiving. Some nursing homes and hospitals have patients whose life status appears questionable on initial observation. The second case at the beginning of this article presented all the distractions necessary to mislead both EMS and the local coroner to believe the patient had died without doing a more thorough exam.
As though there were not enough pitfalls regarding death announcements, enter the newest technology in artificial hearts: left ventricular assist systems (LVAS). Next generation centrifugal LVAS have been implanted in outpatients living in multiple communities around the world. Unlike their predecessors, which used air-driven pulsatile pumps, centrifugal devices use a magnetically levitated impeller to move blood forward through the body, resulting in nonpulsatile flow. Imagine the difficulties assessing a patient for signs of life when he has no pulse.
Also confounding the issue is a tendency for chest compressions to dislodge LVAS tubing from the heart, resulting in fatal internal hemorrhage. Fortunately, LVAS patients should be readily identifiable, exhibiting a sternotomy scar (healed incision over their sternum) and an additional healed incision with an often palpable implanted device under their diaphragm accompanied by conspicuous medical identification. When found unconscious, detecting signs of life in a centrifugal LVAS patient can be very challenging. Assess carefully for breathing, consider using an electronic noninvasive blood pressure cuff to determine mean arterial blood pressure (with no pulse, the calculated systolic and diastolic pressures are meaningless), or use portable ultrasound to assess blood flow. Always search for potential causes of unconsciousness such as hypoglycemia, and consider consulting the heart failure program that implanted the LVAS. All heart failure programs have 24/7 on-call clinicians primarily for troubleshooting LVAS problems.
Field pronouncement of death should follow local protocols. At minimum, an EMS exam must include apnea and pulselessness, lack of response to verbal or tactile stimulation, lack of pupillary light reflex, absence of breath sounds, and absence of heart sounds (in addition to no pulse). Documentation should describe the exam you conducted, the location of where the body was found, the body’s physical condition, any significant medical history or trauma, the conditions that precluded resuscitative efforts, any medical control contact, and the name of the person who was left in custody of the body. Note the time you completed your exam—this will be the official time of death in most cases. As in the hospital setting, double check your impression with every available resource. If you have an ECG monitor, attach it to the patient and make a 15-second recording in two leads. Leave the leads on the body as confirmation of your assessment. If you don’t have a monitor but carry an automated external defibrillator (AED), confirm a “No Shock Advised” reading when attached to the body. Some AEDs can be programmed to read “Asystole” when they encounter flatline for several minutes. There is no more potentially significant decision where portable ultrasound can contribute to prehospital care. Watch for this technology as it makes its way into EMS.
Familiarity with physical changes that occur following death can also help your exam and improve documentation. Dependent lividity (livor mortis) resulting from blood pooling in gravity-dependent parts of the body progresses fairly rapidly following cessation of circulation. This purple-red discoloration appears between 20 minutes and three hours following death and usually peaks at six to 12 hours. Dependent lividity does not occur in areas of the body that are in contact with the ground or another surface because those capillaries are compressed. Since blood usually congeals in the capillaries within four to five hours, dependent lividity can be helpful in determining if a body was moved. Within two to six hours following death (depending on ambient air temperature and physical health of the deceased), rigor mortis sets in. This muscle stiffening begins in the eyelids and progresses to the jaw and then the neck. The body also cools (algor mortis), largely in relation to ambient air temperature.
You will never find what you don’t look for. In every death, exhaust all available means looking for signs of life, no matter how obvious the situation may seem. Thoroughly document your actions and make them clearly evident to others. Virtually every patient mistakenly pronounced dead by EMS happened as a result of jumping to conclusions without a thorough and detailed search for all possible signs of life. In this article’s first case, the fire officer canceled response for a higher level of care (ALS); did not attach an AED to the patient; and, in all likelihood, missed observable vital signs that were present. Being humiliated in the media for being wrong is far more embarrassing than having your colleagues chide you for using every available means to make dead certain that you’re right.
1. Levine, C. Taking Sides. 11th ed. Dubuque, Iowa: McGraw-Hill, 2006: 93.
2. Bondeson, J. Buried Alive: The Terrifying History of Our Most Primal Fear. W.W. Norton & Co., 2002.
3. Anonymous. To Prevent Premature Burials. JAMA. 1899; 32:329.
5. A definition of irreversible coma. Report of the Ad Hoc Committee of the Harvard Medical School to Examine the Definition of Brain Death. JAMA. 1968; 6:337-40.
6. President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research. Defining Death: Medical, Legal, and Ethical Issues in the Determination of Death. U.S. Government Printing Office. Washington, DC. 1981. Retrieved February 7, 2007, from: www.bioethics.gov/reports/past_commissions/defining_death.pdf.
7. Uniform Determination of Death Act. National Conference of Commissioners on Uniform State Laws. Annual Conference, Kauai, Hawaii: July 26–August 1, 1980.
8. Hornby, K.; Hornby, L; Shemie, S.D. A systematic review of autoresuscitation after cardiac arrest. Critical Care Medicine. 2010; 38:1246-1253.
9. Wiese, C.H.; Bartels, U.E.; Orso, S; et al. Lazarus phenomenon. Spontaneous return of circulation after cardiac arrest and cessation of resuscitation attempts [article in German]. Anaesthesist. 2010; 4:333-41.
● MIKE McEVOY is the EMS coordinator for Saratoga County, New York, and the EMS technical editor for Fire Engineering. He is a nurse clinician in the cardiac surgical ICU at Albany Medical Center and teaches critical care medicine at Albany Medical College in New York. He is also a firefighter/paramedic and chief medical officer for West Crescent Fire Department in Halfmoon, New York.