How can I confirm that someone is dead with 100% certainty? I have read many articles about this issue but all of them have only what I consider superficial information. All of them focus about some essential points to confirm death:

  1. To listen for lung sounds for 1 min.

  2. To listen for heart sounds for 1 min.

  3. To check the carotid pulse.

  4. To assess the pupil reaction to light.

  5. To assess the response to a painful stimuli.

My argument is that for points 1 and 2 above, the person might be simply obese or having a pneumothorax/pleural effusion or pericardial effusion for example and thus the heart/lung sounds can be very low and difficult to hear. For point 3, the person might have simply a hypotension, and even the carotid artery pulse can't be palpated. For point 4, the person might have an old CVA or blind for other reasons and the pupil reaction to light is absent. For point 5, if the person is in a coma, he might not respond to a painful stimuli. Other signs of death like cold extremities can't be relied on.

The other related question is when to start CPR. Of course in a hospital if the person has a cardiac arrest (recognized as absent pulse + unresponsive + difficult/absent breathing), we start CPR right away, but in the ED department or in a clinic (where we might not have an ECG machine), how can I tell whether the person is dead and there is no need for CPR or he/she is not and I should start CPR?


1 Answer 1



DO perform CPR as soon as one stops breathing normally and is irresponsive, and don't stop. PERIOD. To be on the safe side, even perform CPR on people you presume to be dead for a longer period of time. There are certain definitive signs of death that doctors will use to pronounce a patient dead.

There is a difference between dead and dead

As a rule of thumb, for every minute without oxygen, the chances of survival decrease by roughly 10% because of irreversible brain damage.

Holmberg M. et al, (2000), "Effect of bystander cardiopulmonary resuscitation in out-of-hospital cardiac arrest patients in Sweden", Resuscitation, 47:59-70

The picture below is completely unscientific and has been drawn by me, with no data whatsoever, because I couldn't find any free-to-access statistics to prove my point. (The full version of the linked article. and quite a few other studies performed on this issue have data supporting this general trend).

enter image description here

Source: completely unscientific trend lines mingled together by me.

At some point, time has advanced so far that the patient has zero chances of survival left because the brain damage was too severe.

A patient is clinically dead as soon as they enter cardiac arrest (i.e. no oxygen circulation), but this status is reversible. However, due to ongoing brain damage due to lack of oxygen, the patients chances of survival decrease until they reach 0%, at which point a doctor will pronounce them (actually) dead.

So when do I perform CPR?

Simple, the European Resuscitation Council has published guidelines, the latest version (2015, as of now) says the following:

enter image description here

Source: Perkins, Gavin D. et al. Official Guidelines of the European Resuscitation Council, latest edition: 2015. Section 2, p.85

By the way, the guidelines are free to access, and definitely worth a read!

So as soon as the patient is unresponsive and not breathing normally, you should call the emergency services and start CPR. Even if you don't know how long they have been in this state of cardiac arrest, there are chances of survival. Once a doctor arrives, they will continue with CPR and/or pronounce the patient dead.

What about really dead people?

Of course, there is no gain in trying to reanimate a corpse that has been dead for a few days, and in this case it comes in handy to know if the patient was actually dead for a longer period of time.

You will know. Don't judge by the coldness (cold actually improves the chances of survival), but there are so-called certain signs of death.

  • Lethal injuries:
    This is actually quite obvious: If the patient is decapitated or otherwise so severely injured that it is simply impossible to survive, they are dead with a very high certainty. CAVE: Be careful about your own safety!
  • Pallor mortis:
    This is a paleness due to no circulation in the capillaries of the body. Gravity pulls the blood lower.
  • Algor mortis:
    The reduction in body temperature. CAVE: Depend highly on environmental factors, drowning patients will be quite cold and still have chances of survival
  • Rigor mortis:
    A stiffness in the body. For muscles to relax, ATP needs to be present, and because in a corpse, no ATP is produced/circulated, the muscles become stiff.
  • Livor mortis:
    The blood has been pulled down by gravity so that it forms a red decolorisation of the skin.
  • Decomposition:
    The proteins get broken down, the muscles relax because the binding proteins also get degenerated. This is accompanied with a strong odor.
  • Thank you for your effort but you didn't answer my main question. You mentioned doctors will pronounce the patient dead 2 or 3 times without saying how, which is my question. I quote There are certain definitive signs of death that doctors will use to pronounce a patient dead. and Once a doctor arrives, they will continue with CPR and/or pronounce the patient dead. Lethal injuries is obvious, pallor mortis and allgor mortis are not reliable because they can occur in a living person. No doctor will wait for rigor mortis, liver mortis, or decomposition to pronounce the patient dead.
    – user14945
    Commented Dec 25, 2018 at 16:26
  • @User Actually, depending on the jurisdiction they will need to wait. I can only speak for Germany, but here you can't pronounce someone dead until you have rigor or liver mortis. pallor and algor are fairly reliable (living person < 35 Celsius is pretty impossible, unless a drowning/diving victim). They can call off CPR before that because after ~1h of CPR and Advanced Life Support / Advanced Trauma Management with no pulse regained, the brain damage is so large that there is nothing to resuscitate left.
    – Narusan
    Commented Dec 25, 2018 at 16:50
  • Thank you, I have read the yellow papers. Luckily I know a little bit German 😊. Correct me if I am wrong. I uderstand that (at least in Germany) doctors can't pronounce a patient dead unless he/she has either rigor or liver mortis or has been resuscitated for approx. 1h with no signs of life. And doctors (in Germany) will try to resuscitate patients who had cardiac arrest unless they developed rigor or liver mortis. I excluded algor for the rare special case you mentioned and pallor because it can occur in CHF or due to vasoconstriction in cold weather.
    – user14945
    Commented Dec 25, 2018 at 17:15
  • 1
    I'll try to find the time to write a follow-on answer from a US perspective. The guidelines are similar but not quite as rigid. For example, pronouncing death is a legal determination that only a doctor can make, but US paramedics routinely presume death without a doctor being present, which has the same practical effect; ie, resuscitation efforts are terminated.
    – Carey Gregory
    Commented Dec 25, 2018 at 22:16