Body temperature during the dying process does not follow a neat, predictable decline. A study of hospital patients in their final hours found that average core temperature hovered around 37.2 °C right up until death, with individual swings of several degrees occurring regularly and sometimes dramatically. The reality is messier and more biologically interesting than the gradual cooling most people imagine, and what happens after death is just as complex.
Temperature Swings in the Final Hours of Life
One of the largest studies to track core body temperature in dying hospital patients found that the median temperature fluctuation in the last 24 hours before death was more than 1 °C, and singular deviations of up to 6 °C occurred in individual patients.1PubMed Central. Core body temperatures during final stages of life—an evaluation of data from in-hospital decedents That is not a gentle fade. Some patients spiked high fevers, others dropped low, and many did both within a short window. The average core temperature across the group stayed near 37.2 °C, but that average masks wild variation from person to person and hour to hour.
The same study found that men tended to have higher body temperatures than women at the time of death, a difference that reached statistical significance when researchers looked at temperatures recorded in the final two hours.1PubMed Central. Core body temperatures during final stages of life—an evaluation of data from in-hospital decedents Perhaps most surprising, there was no clear pattern linking temperature to a particular cause of death. Even among people who died from cardiac causes, temperatures at the moment of death spanned a broad range. The body does not simply wind down like a cooling engine. Its thermostat malfunctions in erratic, individual ways.
Why the Dying Body Loses Temperature Control
Your body normally keeps its core temperature within a remarkably tight range. The hypothalamus, a small region deep in the brain, acts as the thermostat. It detects when you are too warm or too cold and triggers responses: sweating and opening blood vessels near the skin to dump heat, or shivering and constricting those vessels to hold heat in. This system depends on a functioning brain, adequate blood flow, and responsive muscles and blood vessels. As someone approaches death, all of these can fail in unpredictable sequence.
In the final stage of many illnesses, blood pressure drops, blood flow to the brain decreases, and the hypothalamus can no longer send reliable signals. Organs competing for dwindling blood supply generate inconsistent amounts of metabolic heat. Infections, common in dying patients, trigger inflammatory responses that push temperature up. Meanwhile, muscle wasting reduces the body’s ability to generate heat through shivering. The result is temperature readings that jump around, with no single trajectory you can count on. Nurses and family members at the bedside sometimes notice extremities turning cold and mottled while the patient’s core remains warm, or even hot, because the body is pulling blood away from the skin and concentrating it in the vital organs.
Fever at the End of Life
Fever is common in people who are actively dying, and its management raises genuinely difficult questions. Some fevers near the end of life are caused by infections that could theoretically be treated with antibiotics, but evidence shows that antibiotic therapy is inconsistently useful at alleviating fever symptoms in terminally ill patients.2PubMed. Fever Near the End-of-Life Other fevers are caused by the disease process itself. Cancers, for instance, can produce so-called neoplastic fevers that do not respond to antibiotics at all.
Standard antipyretics work by blocking the production of prostaglandin E2, the chemical messenger that tells the hypothalamus to raise the body’s temperature set point. Acetaminophen is typically used first because of its low side-effect profile, and naproxen is considered particularly effective for neoplastic fever.2PubMed. Fever Near the End-of-Life But the real debate in palliative care is whether treating the number on the thermometer matters at all. A dying patient with a fever of 39 °C who seems comfortable may not benefit from aggressive cooling. Palliative guidelines increasingly frame the goal as treating the symptoms the patient actually feels, like discomfort or restlessness, rather than chasing the temperature reading itself.
When Low Temperature Signals Danger in the Living
While fever gets more attention, abnormally low body temperature can be a grimmer sign. Among patients with sepsis, those admitted with a core temperature below 36 °C had dramatically worse outcomes. One large database analysis found that patients with hypothermia had a 28-day mortality rate of about 54%, compared to roughly 19% in patients with elevated temperature.3PubMed Central. Association between body temperature and all-cause mortality in patients with sepsis: analysis of the MIMIC-IV database The relationship between temperature and death risk in sepsis followed a U-shaped curve: both very low and very high temperatures were dangerous, but the cold end was worse.
Body composition adds another layer. A study of sepsis patients stratified by weight found that hypothermia carried the highest mortality penalty in people with a normal BMI, where in-hospital death rates were roughly double those of non-hypothermic patients in the same weight category. That increased risk was not seen as clearly in underweight or overweight patients.4PubMed Central. Association between low body temperature on admission and in-hospital mortality according to body mass index categories of patients with sepsis The reasons are not fully understood, but the clinical takeaway is that a dropping temperature in a seriously ill person is a red flag, not a sign that the body is calming down. It often means the body has lost its ability to mount a proper inflammatory response.
What Happens to Temperature Right After Death
Most people assume that body temperature begins falling immediately once the heart stops. The truth is stranger. Research has shown that there is often an initial rise in body temperature shortly after death, measured rectally, probably because tissues and bacteria continue generating heat even as the body’s normal heat-dispersal mechanisms shut down.5PubMed. Body temperature is elevated in the early postmortem period Without a beating heart to circulate blood to the skin, without sweating, and without breathing to carry away heat through exhaled air, the core briefly traps the metabolic heat still being produced by dying cells.
This initial rise is typically small and short-lived, giving way within the first few hours to the well-known process of post-mortem cooling, traditionally called algor mortis. The body gradually equilibrates with the surrounding environment. In a room-temperature setting, this process usually takes roughly 18 to 24 hours for a body to reach ambient temperature, though the exact rate depends on dozens of variables: body size, clothing, ambient temperature, air movement, whether the surface beneath the body is conductive, and many others.
How Forensic Scientists Use Cooling to Estimate Time of Death
The rate at which a body cools after death has been used by forensic investigators for over a century to estimate the post-mortem interval, which is the time between death and discovery. The most widely used approach in practice is the Henssge nomogram method, which uses a mathematical model of heat loss along with the body’s rectal temperature and estimated weight to calculate a window for time of death. An assessment of this method found that body mass index and body surface area both influenced its accuracy. For every unit increase in BMI, a person had about 1.11 times the odds of the actual time of death falling within the method’s predicted window.6PubMed Central. An assessment of the Henssge method for forensic death time estimation in the early post-mortem interval In plain terms, larger bodies cool more predictably, probably because their greater thermal mass smooths out the noise that makes small bodies harder to model.
Newer computational methods are trying to improve on these estimates. One approach uses numerical simulations of heat transfer combined with optimization algorithms to reconstruct the post-mortem interval from skin temperature measurements. Researchers using this technique estimated perimortem body temperatures averaging about 37.1 °C, with individual post-mortem interval errors averaging just 11 minutes.7PubMed Central. Next-generation time of death estimation: combining surrogate model-based parameter optimization and numerical thermodynamics That level of precision is impressive for forensic work, though it was achieved under controlled conditions. Real-world crime scenes are messier.
The environment around the body matters enormously. A body found in water, for example, cools very differently from one found in air. Research presented at a 2023 engineering conference showed that the medium surrounding the body and the level of immersion can significantly and nonuniformly change internal body temperature over time.8ASME 2023 Heat Transfer Summer Conference. The Effect of Immersion in Water on Internal Body Temperature Post-Mortem Water conducts heat away from the body far more efficiently than air, meaning immersed bodies cool faster and in ways that standard air-based models do not predict well. This is one reason forensic temperature-based methods are considered supplementary tools rather than definitive evidence. They give investigators a useful estimate, but courts and coroners treat them as one piece of a larger puzzle.
The Assumption That Trips Up Forensic Work
Almost every traditional forensic cooling model starts from the same assumption: the body was at 37.2 °C at the moment of death. The data from dying hospital patients, with its wild individual variation, suggest this assumption is shaky. Some people die with a core temperature well above 38 °C; others are already below 36 °C. A person who died during a fever spike would appear to have been dead for less time than they actually had been, because their body started cooling from a higher point. Someone who died hypothermic would look like they had been dead longer. The researchers who tracked temperatures in dying hospital patients explicitly noted that their data “highlight potential biases for death time estimations when generally assuming a core body temperature of 37.2 °C.”1PubMed Central. Core body temperatures during final stages of life—an evaluation of data from in-hospital decedents
This is not a theoretical problem. In practice, forensic pathologists try to account for it by gathering contextual clues about the deceased person’s health: was there evidence of infection, was the person on medications that affect temperature, were they found in conditions suggesting they were already hypothermic? But in many cases, particularly unwitnessed deaths, the starting temperature is simply unknown. The newer computational methods described above attempt to address this by treating the perimortem body temperature as an unknown variable that the model estimates, rather than fixing it at a default value.
When Freezing Cold Causes Death
In deaths caused by environmental hypothermia, the temperature story runs in reverse. Instead of a warm body cooling after death, the body cools while the person is still alive, and the dropping temperature itself becomes the mechanism of death. As core temperature falls below about 30 °C, the brain becomes increasingly disoriented, and two bizarre behaviors have been documented that initially confused investigators.
The first is paradoxical undressing, in which a deeply hypothermic person removes their clothing in the final minutes before losing consciousness. This has been attributed to changes in peripheral vasoconstriction: as the body’s blood-vessel-tightening response collapses, warm blood rushes to the skin, creating an overwhelming and deceptive sensation of heat. The person, confused and now feeling unbearably hot, strips off layers in what researchers described as their last voluntary effort before unconsciousness and death.9PubMed. “Paradoxical undressing” in fatal hypothermia
The second is terminal burrowing, in which a dying hypothermic person crawls into a tight, enclosed space, such as under a bed, behind furniture, or into a closet. This also appears to result from peripheral vasodilation creating a sensation of warmth, combined with a primitive burrowing instinct that emerges as higher brain functions shut down.10PubMed. “Terminal burrowing behaviour”–a phenomenon of lethal hypothermia Both behaviors have been mistaken for signs of foul play at crime scenes: finding an undressed body wedged into a confined space understandably raises suspicions. Forensic pathologists now recognize these as hallmarks of death by hypothermia, not evidence of assault.
Temperature After Cardiac Arrest and Resuscitation
When someone’s heart stops and is restarted, the body enters a precarious state where temperature becomes both a symptom and a target. Spontaneous temperature changes after return of circulation are common, and patients who experience large swings in temperature during the first 24 hours tend to have worse outcomes than those who stay near normal.11Resuscitation. Body temperature changes are associated with outcomes following in-hospital cardiac arrest and return of spontaneous circulation The brain, injured by the period without blood flow, is especially vulnerable to temperature extremes.
Managing temperature after cardiac arrest has become a standard part of intensive care. Fever in the first few days after a cardiac arrest is consistently associated with worse neurological outcomes. Higher temperatures increase seizure activity, brain swelling, and metabolic demand in tissue already struggling to recover. Current guidelines recommend actively controlling temperature to prevent hyperthermia, and some evidence suggests that cooling select patients below normal body temperature may offer additional protection for the brain.12PubMed Central. Temperature control after cardiac arrest The practical challenge is that the body fights cooling. Shivering is a powerful heat-generating response, and suppressing it requires sedation or other drugs, which adds complexity and risk. Clinical temperature management after cardiac arrest involves not just cooling devices but a whole pharmacological strategy to prevent the body from undoing the cooling by ramping up its own heat production.
Extreme Cooling as a Medical Strategy
The relationship between cold and death cuts both ways. While hypothermia kills people in the wilderness, deliberately induced extreme cooling is being explored as a way to buy time for patients who would otherwise die. Emergency preservation and resuscitation, or EPR, is a technique developed for trauma patients whose hearts have stopped due to blood loss. The approach involves flushing ice-cold saline through the aorta to rapidly cool the body. Animal studies have demonstrated that cooling to a core temperature of about 10 °C during cardiac arrest from bleeding can allow up to two hours of circulatory arrest and surgical repair of injuries, with subsequent resuscitation and normal neurological recovery.13PubMed. Emergency preservation and resuscitation for cardiac arrest from trauma
At 10 °C, metabolic activity slows to a fraction of its normal rate. Brain cells that would die within minutes at normal temperature can survive for far longer. The idea is not so different from what hibernating animals do naturally, where a regulated suppression of heat production allows entry into torpor, dramatically reducing the body’s energy needs.14PubMed Central. Thermoregulation in hibernating mammals: The role of the “thyroid hormones system” Of course, a ground squirrel eases into torpor over days, while EPR drops a human’s temperature by 27 degrees in minutes. The comparison is imperfect, but the underlying biology is related: cold slows the chemistry of death.
Human trials of EPR have been underway, though the patient population is narrow, limited to trauma victims in cardiac arrest who would otherwise have virtually no chance of survival. Even if EPR never becomes routine, the research underscores a broader truth about temperature and dying. The line between life and death is not a clean threshold. It is a gradient shaped heavily by how warm or cold the tissue is, how fast metabolism is running, and how long the cells can hang on before the damage becomes irreversible. Temperature does not just change when you die; it plays an active role in determining whether, and how quickly, you do.