Burnout, Exhaustion and the Stress System: What Happens in the Body

Sunday evening, you are sitting on the sofa. Nothing is happening. And yet your heart is pounding as if you had an exam first thing tomorrow.

You sleep eight hours and wake up as if you had not slept at all. An email that used to take two minutes has been sitting open for days. And at some point you hear the sentence: your blood tests are fine.

Many people know this pattern, and both can be true at the same time. Standard lab tests can be unremarkable while something in your stress system may well have shifted, just not in a place that a lab value picks up.

This article is about the interplay of stress axis, cortisol, autonomic nervous system, sleep, immune system and brain in chronic exhaustion. You will find the overview in Burnout: neurobiology, diagnostics and the way back, and the symptoms in Physical symptoms of burnout. Where the research is inconsistent, I will tell you. That is part of the answer.

Two arms: how the stress system switches on and off again

Think of the moment before you have to speak in front of others. Your heart races, your hands get damp. That takes seconds. And hours later you are lying in bed, still wide awake, even though it is long over.

These are two arms of the same system, a fast one and a slow one.

Imagine the fast arm as a smoke detector. It goes off immediately, loudly and without thinking. The slow arm is the fire brigade's incident command. It organises supplies for hours and reports back when the operation is over.

The two do not work separately, but through largely overlapping circuits in the limbic forebrain, hypothalamus and brainstem, linked to circuits for memory and reward, according to a review of animal and human data (Ulrich-Lai and Herman 2009). What you remember and expect can help set how strongly the system switches on.

The brain decides what is threatening. Stress hormones are protective in the short term, and the hippocampus, amygdala and prefrontal cortex can themselves be remodelled under stress, a finding that largely comes from animal studies (McEwen 2007).

Cortisol follows a daily curve with a peak shortly after waking. What it can look like in burnout is covered in Cortisol and the HPA axis in burnout. Here, cortisol remains one player among several.

In anthroposophic medicine, rhythm, the alternation of tension and release, is considered a basic motif of health. That is clinical tradition without a strong evidence base, but it fits as an image: a healthy stress system swings out and back.

And now you know why you can lie awake for hours after a hard day: incident command can still be on duty even though the smoke detector fell silent long ago.

Allostatic load: why adaptation comes at a price

Maybe you functioned for years. More projects, less sleep, worries in the family on top, and you managed. Until one ordinary day, suddenly nothing worked anymore.

That is neither chance nor weakness. Stress research has a name for it.

In 1993, McEwen and Stellar took up the term allostasis and introduced a second one. Allostasis means stability through change: your body stays stable by constantly readjusting. Allostatic load is the price of that. They defined it as the cost of chronic exposure to fluctuating or heightened neural and endocrine responses when challenges recur and are experienced as particularly stressful.

Imagine a credit card on which you only ever pay the minimum each month. That goes well for a long time. But the interest keeps running, and the bill arrives spread across many accounts.

Over the years, this can turn into a cascade of shifts in several systems. An index made up of hormonal, immune, metabolic and cardiovascular values predicted illness and mortality better than conventional methods (Juster 2010).

A systematic review of 267 original studies linked allostatic load and overload with worse health outcomes (Guidi 2021).

And now you know why a breakdown can come on an ordinary Tuesday and not only in the hardest week.

Cortisol in burnout: why the studies point in different directions

Some people read after a saliva test: morning cortisol too low. Others with very similar symptoms read: cortisol raised. And both findings are explained with the same word, burnout.

How does that fit together? The studies themselves point in different directions. That belongs openly on the table.

A meta-analysis of 147 studies from 62 articles partly supports this direction: the awakening response was positively associated with work stress and general life stress, and negatively with fatigue, burnout or exhaustion (Chida and Steptoe 2009).

Reviews support this mixed picture. In an analysis of 190 studies, raised cortisol emerged in depression, while for burnout the studies were not sufficient for an evidence base, partly because burnout was assessed in very different ways (Rothe 2020). Jonsdottir and Sjörs Dahlman from the Institute of Stress Medicine in Gothenburg see no homogeneous, reliable hormonal or immunological changes and consider it possible that consequences show up in the response to acute stress rather than in resting values.

This fits with a randomised laboratory study of 70 working men: more burnout symptoms went along with more strongly perceived stress and, similar to depressive symptoms, with a blunted cardiovascular response (Wekenborg 2019).

Many people end up at the term adrenal fatigue here. Why it does not hold up as a distinct illness is covered in Adrenal fatigue: myth and facts and in Cortisol too high, making sense of adrenal fatigue.

And now you know why two people with similar symptoms can have opposite cortisol findings.

The autonomic nervous system: heart rate variability as a measurable window

Your watch shows you a number in the morning. For weeks it has been lower than it used to be. What does it tell you?

Your heart does not beat like a metronome. The intervals between two beats constantly vary by milliseconds. This variation is called heart rate variability, or HRV for short. A large part of it comes from the vagus nerve, the brake of the nervous system, which is loosened and tightened a little with every breath.

Imagine a good driver who constantly makes tiny steering corrections. It looks restless, yet it is a sign of control. A high HRV can stand for a nervous system with room to manoeuvre. A low one can mean that the brake grips less.

For work stress, a systematic review found a clear pattern of higher heart rate and lower HRV, but no clear conclusion for burnout (de Looff 2018). And a meta-analysis of neuroimaging studies linked HRV to the amygdala and the ventromedial prefrontal cortex, regions we will meet again when we get to the brain (Thayer 2012).

The number on your watch is a trend, not a diagnosis, more on that in Sleep trackers: how accurately they measure. Whether regulation can be influenced through breathing and the vagus nerve is discussed in Vagus nerve and stress regulation in burnout.

And now you know why your body can be tense while you are not doing anything at all.

Sleep and inflammation: the night shift and the immune system that is listening in

You go to bed early because you are so tired. At three o'clock you are awake, your head full of lists. And at the weekend, when you could sleep in, it does not get better.

Sleep is not a break from life. It is a recovery phase in which the stress axis usually winds down. If this night shift is missed, more can be left undone the next day.

A review by Grossi and colleagues notes that the data support sleep disturbances as causal and maintaining factors in clinical burnout. And sleep appears to be part of the off switch of the stress axis: in the laboratory, evening cortisol in young, healthy men was 37 percent higher after partial sleep deprivation and 45 percent higher after total sleep deprivation than on the previous day, and the nightly quiescent period began at least one hour later (Leproult 1997).

A frequently cited finding comes from animal research: in mice, the space between brain cells increased by 60 percent during sleep, and beta-amyloid was cleared more quickly (Xie 2013, Animal study, mouse). In humans, this has not been measured directly in this way.

More on the vicious circle in Burnout and sleep disorders, on snoring and breathing pauses in Recognising sleep apnoea, and on waking at three in Awake at 3 a.m..

The immune system is listening in

Cortisol is also a brake on inflammation. It seems likely that the immune system also changes when the stress axis is disrupted. But seeming likely does not yet mean proven.

Well described mechanistically, based on animal and human data, is the opposite direction: inflammatory messengers of the innate immune system can produce sickness behaviour in the brain, with withdrawal and loss of drive, and with persistent activation in vulnerable people even depressive symptoms (Dantzer 2008).

In 167 teachers, more severe burnout symptoms went along with higher TNF-alpha, with small proportions of variance explained, while the classification burnout yes or no showed no association (von Känel 2008). A meta-analysis of 72 studies with more than 50,000 people found disturbed sleep linked to somewhat higher CRP and interleukin-6. Experimental sleep deprivation or sleep restriction, by contrast, showed no association with CRP, IL-6 or TNF-alpha (Irwin 2016).

In summary: indications, not proof. Jonsdottir and Sjörs Dahlman see no firm conclusion for the immune system. What role the gut and mitochondria might play is covered in Burnout, gut, inflammation and mitochondria.

In my clinical experience, sleep often tips early in exhausted people and still only gets attention late in everyday life. That is experience, not a study result.

And now you know why a long weekend often does not make up for the exhaustion.

The brain under constant strain: amygdala, frontal cortex and why thinking stalls

You read an email three times and afterwards you do not know what it said. A harmless sentence in a meeting, and you are almost in tears. Many people feel ashamed of this. Neuroimaging can help to see it differently.

Two regions are important. The amygdala is the alarm system, it assesses in a flash whether something is dangerous. The prefrontal cortex behind the forehead is the control centre: it plans, holds information and can dampen the alarm.

Thinking can also be measured. In 33 patients with stress-related exhaustion, the difference from 37 controls was most pronounced in executive functions, that is, planning and control, along with weaker results in attention span, working memory and episodic memory (Jonsdottir 2013). In another study, patients maintained their performance under noise distraction but reported markedly more mental fatigue, performance at a high price (Krabbe 2017). That could explain why many affected people do not look overloaded from the outside.

What mental exhaustion feels like is described in Emotional and mental exhaustion. And because concentration problems can also have other causes, it is worth taking a look at Iron deficiency and brain fog.

And now you know why a meeting can exhaust you more than the task itself would explain.

Why recovery is not linear, and what the follow-up data show

Three weeks of sick leave. At first you do almost nothing but sleep, then things get better. You think, now I have made it. And then comes a day when nothing goes right, and you ask yourself what you are doing wrong.

Probably nothing. A slow course with setbacks is not unusual after long overload.

Think of a deep cut. After a few days the skin has closed, but the tissue underneath keeps remodelling for months. If you put strain on it too early, it pulls. A stress system after long overload resembles this wound more than a battery that charges overnight.

Well-being and measurable performance often diverge. After ten weeks of psychological treatment, burnout symptoms and self-reported cognitive problems improved, but test performance did not (Oosterholt 2012). After a year and a half, the cortisol awakening response in clinical burnout was back in the normal range, but complaints and slight test deficits persisted: better, but not yet well, as the study title puts it (Oosterholt 2016). So a normal hormone value does not mean that recovery is complete.

After an average of three years, 30 patients still performed worse than 27 controls on speed, attention and memory (Jonsdottir 2017). And seven to twelve years after first seeking help, 98 former patients classified as recovered fell between 51 who were still exhausted and 50 healthy people in terms of cognitive problems and fatigue (Ellbin 2021).

A meta-analysis of eight studies on treatment programmes found a faster return to work, with considerable heterogeneity. The effect on full return to work was not significant, and there were no significant effects on exhaustion, depression or anxiety, with a small body of studies (Perski 2017). That does not speak against treatment. It shows that returning to work and recovery of the stress system do not move at the same pace. Medical and psychotherapeutic support remains the foundation.

More on phase models in Burnout phases, realistically assessed, on returning to work in Burnout sick leave and returning to work, and on exercise in Training despite burnout.

And now you know why patience here is not a platitude, but part of the way.

What should be checked, and why a cortisol value does not support a burnout diagnosis

You are sitting in the consulting room. The blood count is unremarkable. Inside you, relief mixes with the worry that people will no longer believe you. Both deserve a clear answer.

First, the term. The WHO describes burnout in ICD-11 under code QD85 as a syndrome resulting from chronic workplace stress that has not been successfully managed, with three dimensions: exhaustion, mental distance or cynicism towards one's job, and reduced professional efficacy. Burnout refers explicitly to the occupational context and is not classified as a medical condition. The WHO lists it in the chapter on factors influencing health status or contact with health services. Burnout is therefore not a medical diagnosis in its own right.

That does not make your suffering any smaller. It explains why there is no burnout lab value, and why the assessment mainly asks: is there something else behind it that needs its own treatment?

Why a single cortisol value is not enough

Cortisol measurements have their established place when Cushing's syndrome or adrenal insufficiency is suspected. Which guideline applies is named in Adrenal fatigue: myth and facts. For the question burnout yes or no, a single value is not suitable, for three reasons:

Assessment in general practice is not an obstacle here, but the foundation. Functional medicine takes a broader complementary look, for example at iron status, sleep quality or blood sugar patterns. The guideline, on the other hand, warns of false findings when very broad testing is done without indications. I consider both objections justified and therefore test in a targeted way rather than across the board.

And now you know why body and experience do not simply start up again after long overload: the stress system probably does not have a single defect, it may have learned in many places to stay in alarm mode. The way back deserves time, targeted assessment, support and an honest look at the strain at work.

If you want to do more than read and get started right away: below this article you will find the option to book an appointment.

Frequently asked questions about burnout and the stress system

Where this topic leads