1. Stress hormones in ADHD¶
Various stress hormones are altered in ADHD.
The stress hormones cortisol, ACTH and CRH discussed below are the stress hormones of the 3rd, 2nd and 1st stages of the HPA axis. Alpha-amylase is a biomarker for the activity of the autonomic nervous system.
In summary, it must be assumed that the cortisol response to acute stressors, which is often flattened in ADHD-HI and very often excessive in ADHD-I, is a biomarker for a pathological change in the HPA axis, but that the change in the cortisol response is not the cause of ADHD. Comparable experiences have been made in the treatment of depression.
However, too many studies suffer from the fact that in ADHD, as in depression, the externalizing subtypes (ADHD-HI, atypical depression) and the internalizing subtypes (ADHD-I, melancholic depression) are not recorded separately.
2. Cortisol responses to stress and personality traits¶
2.1. High cortisol response: internalizing, low cortisol response: externalizing¶
We have collected a number of sources showing that high or low cortisol stress responses in healthy people correlate with gender and various general personality traits.
More on this at ⇒ Cortisol stress responses and personality traits in healthy people In the section ⇒ Stress response patterns in healthy people of the article ⇒ Stress theories and stress phenotypes: a possible explanation of ADHD subtypes.
2.2. Cortisol reactions as an expression of the stress phenotype¶
The available research strongly suggests that externalizing or disruptive behaviors such as aggression with or without the presence of (comorbid) Oppositional Defiant Disorder (ODD) or symptomatically more severe Conduct Disorder (CD) are associated with lower basal cortisol levels and a flattened cortisol response to acute stress.
Elevated basal blood cortisol levels in 15-year-old ADHD sufferers (decrease in the late afternoon) correlated with aggression in boys, but not in girls. Another study found a correlation of aggression with flattened cortisol responses to acute stress, while basal cortisol levels remained unchanged.
Depression or depressive personality traits as well as anxiety correlate with increased morning (CAR) or serum cortisol levels.
⇒ Cortisol in other disorders
If - as we assume - not only ADHD, but also the other mental disorders mentioned here are understood as dimensional disorders and furthermore ADHD-HI is understood as an externalizing stress phenotype and ADHD-I as an internalizing stress phenotype, it is reasonable to assume that primarily the stress phenotype could determine the cortisol response to acute stress (or that the cortisol response to acute stress could predict the stress phenotype).
This view leads to the conclusion that dexamethasone testing of the HPA axis for cortisol reactivity could only determine whether an HPA axis disorder is present and whether this leads to an externalizing or an internalizing stress phenotype, but not which mental disorder is present within the respective spectrum.
It is argued that cortisol stress responses say more about comorbidities than about ADHD. According to this view, blunted cortisol responses to stress are associated with comorbid DBD (disruptive behavioral disorder), while high cortisol responses are associated with comorbid anxiety disorders. This is consistent with our view of the correlation of blunted cortisol stress responses with externalizing disorders and elevated cortisol stress responses with internalizing disorders.
A reduced exchange between the brain hemispheres (especially in PTSD) appears to be a cause of a reduced cortisol stress response, as well as alexithymia. Furthermore, a coincidence of reduced basal cortisol levels (despite a normal cortisol response to ACTH) and alexithymia has been reported in some chronic stress sufferers, from which the authors concluded that ACTH-independent mechanisms reduce cortisol levels.
2.3. Cortisol stress responses over time¶
The problem is that the individual stress cortisol response maximum differs by up to 20 minutes. In a group of 54 healthy adults, the cortisol maximum in the early group occurred directly at the end of the 15-minute TSST stressor (cortisol increase from 0.04 to 0.21 ng/ml), whereas in the late group it did not occur until 20 minutes later (from 0.05 to 0.22 ng/ml). If both groups had been examined directly after the end of the TSST, the late group would have been incorrectly categorized as reacting to stress with a flattened cortisol increase with an increase from 0.05 to 0.07. In most studies, a cortisol measurement 20 minutes after the end of the TSST is common. The question arises as to whether the rise in cortisol in the early group lasts long enough to be properly recorded in a measurement after 20 minutes.
In many studies, not enough cortisol samples are taken at the end of the stress test to take this into account.
According to one source, the cortisol levels of ADHD sufferers only deviate (upwards) from those of non-affected people, even under stress, if relationship-related issues were the subject of the stress test (increased sensitivity to relationship issues), but not in the case of purely cognitive stressors. This is not confirmed by most studies. To clarify this, a distinction is made below between cognitive and emotional stressors.
2.4. Cortisol stress responses by gender¶
Cortisol responses to stress also differ according to gender.
2.5. Cortisol responses to stress in non-affected people¶
- Among 102 five-year-old children not diagnosed with ADHD, HPA activity was significantly higher in the girls than in the boys. Emotional stress (retelling a story) caused a significant increase in cortisol in all children, which was even higher in girls than in boys. Increased HPA system activity (basal and stress-reactive) correlated significantly with hyperactivity/impulsivity and emotional problems in the boys and with positive emotions in the girls.
-
Chronic stress reduces the morning cortisol maximum (CAR, typically 30 to 60 minutes after waking up) in healthy young men (average age 22.5 years). Even low levels of stress increased the baseline value on waking and significantly reduced the CAR maximum (30 min after waking). Severe stress led to an even higher baseline value with a flattened CAR, which also occurred 16 minutes after waking.
- The cortisol response to physical stress depends on
- The individual resting cortisol concentration and previous exposure
- Of muscle mass. The cortisol response to physical exertion is relatively lower with larger muscle mass than with smaller muscle mass. We suspect that this could simply be due to the lower relative load when the same work is performed by more muscle mass.
- The current blood glucose concentration and insulin availability before the start of exercise
- The increase in cortisol during physical exertion is mainly triggered by an adrenergically mediated increase in ACTH
- In addition, glucose-sensitive receptors in the liver and brain modulate the exercise-induced
Cortisol response through activation of the pituitary-adrenocortical axis
- A meta-study of 29 studies on 2601 subjects found no correlation between externalizing behaviour and a (flattened) cortisol stress response, as should exist according to the hypothesis we advocate.
2.6. Interplay between cortisol and dopamine¶
Glucocorticoid receptors are found on numerous dopaminergic cells in the midbrain and hypothalamus. It is assumed that cortisol can influence the release of dopamine in the basal ganglia and in nigrostriatal and mesolimbic pathways.
Cortisol inhibits tyrosine hydroxylase, an enzyme that limits catecholamine synthesis by acting as a catalyst for the conversion of tyrosine to DOPA. Tyrosine hydroxylase is inhibited by cortisol (as well as by dopamine and noradrenaline themselves (negative feedback).
This indicates a reciprocal interplay between dopamine and cortisol release.
A retrospective analysis found a correlation between the use of inhaled corticosteroids in younger children with moderate to severe asthma. This correlation was not found in older children.
2.7. Interplay between CRH and noradrenaline¶
The hypothalamus, which releases CRH, is closely connected to the nucleus coeruleus, which produces noradrenaline. CRH activates the release of noradrenaline, while at the same time noradrenaline stimulates CRH production.
3. Changes in cortisol levels in ADHD¶
A large number of studies have addressed the question of whether and how the HPA axis reacts differently in ADHD. There is now massive evidence that there are lasting changes in the HPA axis in ADHD.
Epidemiologic and preclinical studies have shown that the disruption of the HPA axis in ADHD may result from excessive cortisol exposure in the fetal and early postnatal period (early childhood stress). Glucocorticoid administration at this stage of life may permanently alter glucocorticoid receptors in the brain, causing dysregulation of HPA axis activity, disturbances in the biosynthesis of neurotransmitters and their receptors, and alterations in intracellular pathways. Glucocorticoids (cortisol) increase the activity of the dopaminergic system. Reduced expression of glucocorticoids could thus cause hypofunction of the dopaminergic system.
A fundamental distinction must be made between basal cortisol levels and cortisol responses (reactions) to acute stress. While the subtypes of ADHD differ from non-affected persons by reduced basal cortisol values, but not by deviations from each other, the cortisol response to acute stress of ADHD-HI/ADHD-C on the one hand and ADHD-I subtype on the other hand differs considerably and systematically from each other and deviates in various directions from the values of non-affected persons.
Basal cortisol levels therefore differentiate ADHD sufferers from those who are not affected, and cortisol responses also differentiate the subtypes.
It is also striking that the stress-induced cortisol increase is significantly higher in ADHD-I (without hyperactivity/impulsivity) than in those not affected, while acute stress triggers a significantly lower cortisol response in ADHD-HI (with hyperactivity) and ADHD-C than in those not affected.
This could be relevant for treatment, as cortisol not only mediates the stress response of the HPA axis, but also acts as a feedback loop to calm and deactivate the stress systems (the HPA axis and the central nervous system).
We hypothesize that a one-time higher cortisol administration in ADHD-HI could shut down the stress systems again. If the cortisol response that signals the shutdown of the stress systems is missing, it would at least be worth considering compensating for this with a single cortisol substitution. Since a high cortisol stress response is a healthy reaction to shut down the HPA axis again, a one-off intermittent administration should not necessarily have any disadvantages. For ADHD-I sufferers, on the other hand, this should naturally not make sense.
Dexamethasone could be recommended for this, as it addresses GR 30 times more strongly than MR. If necessary, a combination with an MR antagonist such as eplerenone, as used in a study for the treatment of back pain, would be conceivable.
Cortisol administration could not only downregulate the HPA axis, but also support the downregulation of the PFC.
Stress-induced elevated noradrenaline inhibits PFC and working memory function via noradrenaline α1 receptors to shift action control from slow analytical control by the PFC to fast instinctive control by older brain regions, because this is advantageous during fight or flight.
A stress-induced high release of cortisol leads to a stimulation of the same noradrenaline α1 receptors in the PFC. The simultaneous addressing of these receptors intensifies the effect caused by high noradrenaline levels.
In healthy individuals, the endocrine stress responses of noradrenaline and cortisol correlate, so that a coincidence of high noradrenaline and high cortisol levels can trigger PFC deactivation.
Cortisol plays a role in ADHD, but not in relation to inhibition.
3.1. Cortisol responses to stress in ADHD altered in a subtype-specific manner¶
A number of studies have looked at the cortisol response of ADHD sufferers to stress by measuring cortisol levels before and after a stressful event.
However, a distinction must be made between ADHD-HI (with hyperactivity) and ADHD-I (without hyperactivity). If this differentiation is omitted, meta-studies also find no difference in the cortisol stress response between ADHD and non-affected persons.
Chronic stress exposure reduces the expression of the glucocorticoid receptor (GR), which switches the HPA axis off again after a stress response. A slower decrease in the cortisol level to the initial value after an acute stressor was observed, i.e. a longer time to recovery
3.1.1. Reactions to cognitive stress in ADHD¶
- 51 of 68 children with ADHD (average age 8.8 years) showed no increase in cortisol in an unspecified “psychological test as a stressor”. It is possible that the Hamburg-Wechsler Intelligence Test (WISC-III) was also used as a stressor. The 17 test subjects with an increase in cortisol under stress had a higher IQ, although the report does not clearly state whether this was the IQ value measured after stress or the IQ value without stress. The level of cortisol before the test correlated with IQ. The lower the cortisol level after the test, the lower the IQ of the test subjects.
- 20 children with ADHD and comorbid oppositional defiant disorder (ODD) and social behavior disorder (CD), thus with high levels of aggression, showed no increase in cortisol to a cognitive stressor at the age of 4 to 6 years or 2 years later at the age of 6 to 8 years. ADHD sufferers whose problem persisted showed a flattened cortisol response. This is consistent with the other studies in that ODD and social behavior disorder are typical comorbidities of the hyperactive/impulsive ADHD-HI subtype, for which a reduced/absent cortisol rise is often reported.
- Of 48 children who were exposed to the Korean Hamburg-Wechsler Intelligence Test (KEDI-WISC) as a stressor, 28 showed a drop in cortisol and 15 showed an increase in cortisol levels. Children with a drop in cortisol made more errors of omission and incorrect responses, which is thought to demonstrate impulsivity problems. This suggests the hyperactive-impulsive ADHD-HI subtype. The children with an elevated cortisol stress response made more incorrect responses in the attention test, indicating attention problems. Non-affected children were not compared.
- Of 49 children with ADHD without other comorbidities between the ages of 6 and 17, 33 children showed no increase in cortisol on the Korean Hamburg-Wechsler Intelligence Test (KEDI-WISC) and the TOVA test as (cognitive) stressors, while 16 showed an increase in cortisol. The 33 children with a decrease in cortisol had an IQ of over 110, the 16 children with an increase had an IQ of less than 110 (IQ average: 92.3). A decrease in cortisol also correlated with higher aggression scores and lower attention scores in the K-CBCL. Children with an increase in cortisol tended to have a more withdrawn character and fewer social problems. This also fits the picture of a reduced/absent cortisol response in the hyperactive-impulsive ADHD-HI subtype. Non-affected individuals were not compared.
- 68 of 90 children with ADHD did not show an increased cortisol response to stress using the Korean-Wechsler Intelligence Scale for Children-Third Edition (K-WISC-III). Elevated cortisol responses to a stressor correlated with increased variance in response time. Non-affected individuals were not compared.
An increased variance in response times could be explained by an impairment in the performance of the PFC, as is particularly pronounced in the ADHD-I subtype.
Prolonged reaction times correlated with ADHD-I.
- 38 children with ADHD showed a significantly higher increase in cortisol in response to the Continuous Performance Test (CPT) as a stressor than 38 unaffected children, regardless of subtype. A follow-up study of the same test subjects 4 years later showed that the cortisol stress responses of the ADHD sufferers had decreased and now corresponded to those of the non-affected children.
Rating:
We consider stress testing by means of an intelligence test to be of little use. The fact that people with a higher IQ are less stressed by an IQ test than people with a lower IQ is probably the inherent purpose of an IQ test. We also assume that children with reduced intelligence are more stressed by cognitive tests and intelligence tests because they expect a negative assessment, which means that ultimately only the expected social evaluation would trigger stress. For the question of which circumstances trigger cortisolergic stress in the first place, see above.
Furthermore, the ADHD-HI (hyperactive/impulsive) and ADHD-I (without hyperactivity) subtypes are likely to react differently to different stressors due to their different stress phenotypes. We suspect that the more extroverted ADHD-HI type will be less stressed and possibly even more stimulated by social exposure (finishing a story and doing mental arithmetic aloud while being observed/judged by others) than the more introverted ADHD-I type, whose stress response is an inward flight.
This is consistent with the findings of van West.
3.1.2. Reactions to emotional stress in ADHD¶
- A study of 9 ADHD-HI, 10 ADHD-I and 14 ADHD-C sufferers and 33 non-sufferers (6 to 8 years, mean 6.3 years) using TSST-C also found a significantly lower cortisol response in ADHD-HI sufferers and a comparatively higher cortisol response in ADHD-I sufferers. All subtypes showed a lower baseline cortisol level than non-affected individuals. Non-affected individuals did not show a strong cortisol response. The latter could indicate problems with the test set-up. In addition, the number of subjects per subtype was too small for reliable conclusions.
- A study of 18 adults with ADHD and 18 adults without ADHD did not come to any clear conclusion. Only a general tendency towards a reduced cortisol response to acute stress and an increased subjective perception of stress in ADHD sufferers was reported. This may have been due to the small number of test subjects and the lack of separation according to subtypes.
- A study of young adults without differentiation between ADHD-HI and ADHD-I found generally increased cortisol stress responses in ADHD.
- A study that used simple sensory items rather than the TSST as stressors to investigate high sensitivity in ADHD found no significant differences in cortisol stress response between non-affected, ADHD sufferers without and ADHD sufferers with high sensitivity. It would have been interesting to see whether the responses to the TSST would have differed.
3.1.2.1. ADHD-HI, ADHD-C and comorbid externalizing disorders¶
- A study of 96 adults with ADHD and 25 non-affected individuals found a flattened cortisol response in ADHD-C compared to non-affected individuals, and an increased stress-induced cortisol response in the ADHD-I subtype.
- Children (6 to 12 years, mean 8.6 years) with the ADHD-HI-ADHD-C were significantly less likely to have the cortisol rise to emotional stress (Public Speaking Task, PST and Trier Social Stress Test, TSST) (expected in healthy individuals) than non-affected individuals, while the predominantly inattentive ADHD-I subtype showed a significantly stronger cortisol response than non-affected individuals. Only 35% of ADHD-C sufferers showed elevated cortisol levels due to stress, compared to 92% of ADHD-I sufferers and 88% of non-affected individuals. This study also confirmed a connection between aggression, externalization problems and a reduced cortisol stress response.
-
ADHD sufferers with high hyperactivity/impulsivity symptoms showed reduced cortisol suppression after the dexamethasone test. This may result from a flattened cortisol stress response or from less addressable glucocorticoid receptors.
- With a low alpha-amylase release (marker of the sympathetic stress system) in response to acute stress by the TSST, a simultaneous low cortisol response correlates with high levels of aggression; with a high alpha-amylase release in response to acute stress, the cortisol response has no correlation with aggression.
- Of 202 male adolescents (10 to 17 years, mean 14 years), the 107 who had a comorbid social behavior disorder (CD) responded to a psychosocial stressor with significantly more reduced cortisol stress reactivity compared to the 97 ADHD-HI only sufferers.
The more severe the ADHD symptoms were, the lower the basal cortisol level was. In contrast, a social behavior disorder was associated with an increased basal cortisol level and a decreased cortisol stress response. Impaired cortisol reactivity may reflect fearlessness and be associated with poor emotion regulation and inhibition of aggressive and antisocial behavior. Unfortunately, no unaffected individuals were included as controls.
- 7-12 year old (mean: 10 years) children with oppositional defiant disorder (ODD) have reduced activity patterns of the autonomic nervous system or HPA axis. 15 children with ODD, 31 children with ODD and AD, and 23 ADHD-only children were compared with 26 unaffected children. Cortisol response to competition-induced stress (non-relationship stress) was significantly decreased in the ODD children with or without ADHD-HI and was equally normal in the ADHD-only and unaffected children. Pulse rate was significantly decreased in ODD-affected with and without stress. ODD, ODD/ADHD-HI and ADHD-HI sufferers all had equally significantly lower skin conductance values under stress. As MPH medication was not suspended for the study, the results are likely to be biased. That 17 of the 23 ADHD-HI-only children received MPH could explain the normalization of the stress-induced cortisol response.
- Oppositional defiant behavior in ADHD correlated with deviant cortisol responses to acute stress.
-
ADHD-affected children showed a flattened cortisol response to the TSST the more strongly they showed psychopathic characteristics (callous unemotional traits = CU traits) such as lack of empathy, coldness of feeling, etc.
- 22 children with DBD (Disruptive Behavior Disorder) showed decreasing and increasing cortisol responses to stress, with the children with decreasing cortisol responses showing more severe symptoms than the children with increasing cortisol responses.
- Social behavior disorder (CD) correlates with a flattened cortisol stress response.
Since externalizing disorders correlate with ADHD-HI (with hyperactivity), we consider a tendency toward a flattened cortisol stress response in ADHD-HI and ADHD-C a plausible possibility.
3.1.2.2. ADHD-I and comorbid internalizing disorders¶
- A study of 96 adults with ADHD and 25 non-affected individuals found a higher stress-induced cortisol response in the ADHD-I subtype than in non-affected individuals, and a flattened cortisol response in ADHD-C.
-
ADHD-I sufferers of the purely inattentive subtype showed an excessive cortisol response to acute stress.
- A study of 20 ADHD-I sufferers and 19 non-sufferers (8 to 13 years, mean 10.3 years) using TSST-C revealed that the 7 ADHD-I sufferers with severe inattention symptoms showed a significantly reduced cortisol response to the stressor, unlike the 13 ADHD-I sufferers with moderate symptoms and the 19 non-sufferers.
A similar study found a more reduced cortisol response in ADHD sufferers with severe inattention symptoms than in sufferers with less severe inattention symptoms.
These results contradict our analysis that ADHD-I is characterized by an excessive cortisol stress response.
- One study found a strongly flattened cortisol stress response to the TSST in boys with pronounced inattention. Girls with marked inattention also showed a flattened cortisol stress response during and after the TSST, with elevated levels before the TSST. The basal cortisol levels were unchanged. The study also mentions that flattened cortisol stress responses are common in chronic stress.
The children were categorized as ADHD-I based on inattention scores alone, not on the absence of hyperactivity/impulsivity. This could explain the results if the subjects were also showing strong hyperactivity/impulsivity scores. Comorbid externalizing or internalizing disorders did not influence the results.
- In healthy adults, lower novelty seeking (which is more consistent with the ADHD-I subtype) correlated with increased cortisol responses to the TSST (as typically exhibited by ADHD-I), but not with ACTH levels.
- This also applies to a higher risk aversion (which also tends to fit the ADHD-I subtype).
- Anxiety, depression, subjective perception of stress, impulsivity and a higher ADHD symptom total score correlated with increased cortisol responses to acute stress in a study of adults (which is surprising in relation to impulsivity).
3.1.3. Reactions to physical stress¶
- Of 170 ADHD-affected children of primary school age, venipuncture as a stressor led to an increase in cortisol in those with comorbid anxiety problems, while those with comorbid DBD and oppositional defiant disorder had a decreased cortisol response. ADHD-HI subtypes did not elicit different cortisol responses.
- Of 62 ADHD-I affected children, the ADHD-HI type showed identical cortisol levels to non-affected children before venipuncture, while the ADHD-I subtype showed elevated cortisol levels compared to non-affected children even before venipuncture. Since basal cortisol levels are lower in both ADHD-I and ADHD-HI sufferers than in non-affected individuals, this could be an anxiety response in anticipation of the stressor. The same pattern was seen in relation to alpha-amylase as a representative of the autonomic nervous system.
- In the same 62 ADHD-HI-affected children, the ADHD-HI subtype showed a reduction in cortisol levels 10 minutes after venipuncture, while it increased slightly in the 40 non-affected children and increased more in the ADHD-I-affected children. The same pattern was seen in relation to alpha-amylase as a representative of the autonomic nervous system.
- Children with ADHD tended to show increased cortisol stress responses to venipuncture during the first procedure in a home setting and significantly lower basal cortisol levels and significantly flattened cortisol stress responses during the second procedure in a clinic setting compared to non-affected children.
- Children with ADHD showed a lower increase in cortisol in response to a visit to the dentist as a stressor than children without ADHD. The increase in cortisol was particularly low in children with high levels of hyperactivity/impulsivity.
3.2. Reduced basal cortisol levels in ADHD¶
The results on basal cortisol levels in ADHD show reduced values in the majority of cases.
3.2.1. Reduced basal cortisol levels: 17 studies¶
Several studies found that basal (constant, basic) cortisol levels are generally reduced in ADHD, with the reduction in cortisol levels being even more pronounced in ADHD-HI and in ADHD-C than in the ADHD-I subtype.
Reduced basal cortisol levels in the hair of 4-year-old children over several months are a good predictor of an increase in ADHD symptoms in boys (but not in girls) in the following year. Similarly, Pauli-Pott et al. Another study found that hair cortisol levels in children correlate between parental psychopathology and severe depression and ADHD in the children. One study found a correlation between low hair cortisol levels and low working memory scores and low intelligence performance only in boys, but not in girls. In boys, hair cortisol fully explained the correlations of ADHD inattention symptoms with working memory and intelligence performance.
Several other studies found reduced morning blood cortisol levels in ADHD.
The salivary cortisol level of 5- to 9-year-old boys directly after getting up at around 06:30 was on average 0.4 ng/ml lower than that of non-affected persons. However, the individual range of cortisol levels in people is significantly greater than the statistical difference (unaffected 0.7 to 10.5 ng/ml; ADHD sufferers 1.3 to 8.1 ng/ml).
Among 109 adults with ADHD, 64% were found to have a normal morning cortisol awakening response (with no significant differences between subtypes), compared with 84% of unaffected controls.
248 adolescents with ADHD showed lower basal cortisol levels and higher perceived stress, although basal cortisol levels and perceived stress did not correlate.
Of 202 male adolescents (10 to 17 years, mean 14 years), the 107 who had a comorbid social behavior disorder (CD) responded to a psychosocial stressor with significantly more reduced cortisol stress reactivity compared to the 97 ADHD-HI only sufferers. The more severe the ADHD symptoms were, the lower the basal cortisol level was. In contrast, a social behavior disorder was associated with an increased basal cortisol level and a decreased cortisol stress response. Impaired cortisol reactivity may reflect fearlessness and be associated with poor emotion regulation and inhibition of aggressive and antisocial behavior. Unfortunately, no unaffected individuals were included as controls.
Low basal stress levels (basal = not stress-induced) are interpreted as an adaptive response of the HPA axis (stress axis) to chronic stress experience. Not only children with ADHD, but also their mothers showed a lower cortisol level at 9 a.m. compared to unaffected children.
Reduced blood cortisol levels in 15-year-old ADHD sufferers (decrease in the late afternoon) correlated significantly with aggressiveness and impulsivity in boys, but only tendentially in girls.
Several studies show that reduced basal cortisol levels are not an exclusive feature of ADHD. They also correlate with (comorbid) social behavior disorders.
However, the measure of aggression does not result in different basal cortisol levels.
3.2.2. Unchanged basal cortisol levels: 5 studies¶
In ADHD-I sufferers, a study found no changes in daily cortisol levels.
Three other studies also found no change in basal cortisol levels in ADHD patients compared to non-affected patients.
One study that considered ADHD subtypes/presentations found:
- reduced hair cortisol in boys with ADHD-I (no longer significant after controlling for early trauma)
- increased hair cortisol in girls with ADHD-C (no longer significant after controlling for early trauma)
- weakened reactivity of the sympathetic nervous system in boys with ADHD and comorbid ODD/CD
3.2.3. Elevated basal cortisol levels: 3 studies¶
ADHD sufferers with high inattention symptoms, on the other hand, showed an increased cortisol awakening response (CAR), whereby this was predominantly due to comorbid anxiety or depression disorders.
In healthy adults, lower novelty seeking (which is more consistent with the ADHD-I subtype) correlated with elevated baseline cortisol levels, but not with elevated ACTH levels.
One study found consistently elevated basal cortisol levels.
3.3. Decrease in cortisol levels without stress induction in ADHD?¶
A relatively small study came to the conclusion that a steeper drop in cortisol levels from 06:30 to 09:00 is typical for ADHD. In 19 ADHD sufferers, the 66% drop in cortisol levels was significantly steeper than the 56% drop in 40 non-affected people.
Another study also reported abnormal changes in cortisol levels throughout the day, which correlate with the degree of hyperactivity.
3.4. GR variant influences cortisol effect and ADHD risk¶
For more information, see ⇒ Gene variants of the GR In the article ⇒ Cortisol and other stress hormones in AD(H)S
4. ACTH for ADHD¶
-
No differences in basal ACTH levels were found between 128 children with ADHD and 30 unaffected children. There was also no difference between the ADHD subtypes.
-
In healthy adults, lower novelty seeking (which is more consistent with the ADHD-I subtype) correlated with increased cortisol responses to the TSST (as is very common in the ADHD-I subtype), but not with altered ACTH levels.
-
This also applies to a higher risk aversion (which also correlates with the ADHD-I subtype).
-
In one study, sexually abused girls were found to have reduced basal ACTH levels and reduced ACTH responses to CRH stimulation, while the cortisol response was unremarkable.
-
Early childhood stress causes permanent changes in the HPA axis, which are reflected in altered basal and stress-induced cortisol levels. Children with internalizing problems often showed increased cortisol stress responses, while adults who had experienced early childhood psychological stress often showed decreased basal cortisol levels and increased ACTH responses to acute stress.
-
Early stress experiences can lead to disturbances in the ACTH receptor systems, which prevent the anxiety experience from being extinguished and thus cause long-term stress. This could be improved by administering ACTH. In our opinion, the change in the ACTH receptor systems could possibly be the result of a down/upregulation reaction. ⇒ Downregulation / upregulation
-
Rats that were separated from their mother at an early age or received little care showed
- Elevated basal ACTH levels
- Increased ACTH levels due to acute stress
- More than doubled CRH levels on inflammation
- Reduced density of CRH receptors in the anterior pituitary gland
- Changes in extrahypothalamic CRH systems
- 59% increase in the number of CRH receptor binding sites in the raphe nuclei
- Increase in immunoreactive CRH concentrations in the parabrachial nucleus by 86
- Behavioral abnormalities such as
- Increased anxiety
- Anhedonia
- Increased alcohol preference
- Sleep disorders
- Cognitive impairments
- Increased sensitivity to pain
-
In borderline sufferers without comorbid PTSD, the DST showed an increased ACTH response, in borderline sufferers with comorbid PTSD a significantly weakened ACTH response.
5. CRH for ADHD¶
- Young monkeys that grew up under early attachment stress had increased CRH and decreased adrenaline levels at the age of 4 years.
- After its release, cortisol initially mediates stress symptoms. Furthermore, a high cortisol level has the effect of shutting down the HPA axis and noradrenaline.
However, as the GR cortisol receptors are less sensitive after prolonged stress due to downregulation, they do not perceive this signal. The stress systems are not switched off. As a result, CRH levels remain permanently elevated and permanently trigger the symptoms mediated by CRH. This mechanism has already been described for depression.
- Prolonged stress caused chronic demethylation of the CRH gene in adult mice.
6. α-Amylase for ADHD¶
The alpha-amylase level represents the activity of the primarily adrenergically controlled autonomic nervous system. Alpha-amylase reacts very strongly to a psychosocial stressor, but does not correlate with other so-called stress markers such as cortisol, noradrenaline or heart rate. Alpha-amylase is therefore suitable for recording physical changes caused by stress. In contrast, concentration tasks (Stroop test) as a stressor do not increase the alpha-amylase level.
- In ADHD, the basal alpha-amylase level is usually reduced, with the α-amylase level reduction being even more pronounced in ADHD-HI and ADHD-C than in the ADHD-I subtype. This pattern is consistent with the reduced basal cortisol level (a representative of the HPA axis).
*In 62 children with ADHD, venipuncture only significantly increased the α-amylase levels in the control group, but not the cortisol levels. The same pattern was seen with regard to cortisol as a representative of the HPA axis.
- In the same 62 ADHD-affected children, the ADHD-HI subtype showed a decrease in α-amylase levels 10 minutes after venipuncture, while it increased slightly in the 40 non-affected and more in the ADHD-I-affected. The same pattern was seen with regard to cortisol as a representative of the HPA axis.
Alpha-amylase is an enzyme in the intestine that is responsible for breaking down carbohydrates into sugar. There could be a link between stress and eating problems or obesity.
6.1. Correlation between alpha-amylase and cortisol¶
An investigation of the time course of the occurrence of alpha-amylase as a representative of the autonomic nervous system and cortisol as a representative of the HPA axis in response to a TSST as a stressor indicates mutual dependencies between the two stress hormones:
- An increase in alpha-amylase was moderately often followed by an increase in cortisol 14 minutes later.
- A rise in cortisol was often slightly followed by a drop in alpha-amylase 14 minutes later.
- An increase in alpha-amylase was moderately often followed by a decrease in cortisol 42 minutes later.
7. Diagnosis and treatment of ADHD with cortisol / dexamethasone?¶
Dexamethasone is an (artificial) glucocorticoid that binds to the glucocorticoid receptors of the pituitary gland in healthy volunteers, thereby inhibiting the ACTH precursor protein POMC. As a result, the release of ACTH from the anterior pituitary gland is reduced. As ACTH normally stimulates the release of cortisol, a healthy response to the administration of dexamethasone would be to reduce cortisol levels - known as suppression.
Nonsuppression (a lack of cortisol reduction by dexamethasone) indicates a misalignment of the HPA axis, in the form of a lack of response of the glucocorticoid receptors that mediate HPA axis deactivation. We hypothesize such HPA axis misalignments as a possible cause of ADHD, and we believe the lack of HPA axis deactivation is typical of ADHD-HI.
The dexamethasone/CRH test, a further development of the previously used pure dexamethasone test, can be used to determine whether there is a problem with the shutdown of the HPA axis in the form of cortisol nonsupression.
For the procedure and evaluation of the dexamethasone test, see ⇒ Dexamethasone/CRH test and ⇒ Dexamethasone test in the chapter ⇒ Pharmacological endocrine function tests.
In melancholic depression, which like ADHD-I is very often characterized by an excessive cortisol response, but probably also frequently in atypical depression, which is characterized by a flattened cortisol stress response, the dexamethasone test shows insufficient negative feedback of the HPA axis and therefore insufficient cortisol suppression in response to dexamethasone administration.
We hypothesize that in ADHD-I (unlike melancholic depression) the excessive cortisol response leads to a frequent shutdown of the HPA axis. This could be a misconception, or could indicate a difference between melancholic depression and ADHD-I.
In support of the hypothesized difference between depression and ADHD-I, a large study (n = 2307) found reduced cortisol suppression to dexamethasone in hyperactivity/impulsivity (ADHD-HI), while inattention (ADHD-I) showed no correlation to nonsuppression. This is consistent with the results of a smaller study that found nonsuppression in only a proportion of ADHD sufferers and found that nonsuppression correlated with higher levels of hyperactivity. Another small study, in which only 9 children with ADHD were involved, found suppression in ADHD. Another smaller study found nonsuppression in 22.7% of ADHD sufferers with hyperactivity, which is 4-fold higher than the 5.7% in healthy people.
However, the fact that dexamethasone shows a reduction in ADHD-HI symptoms in rats that serve as an animal model for ADHD-HI (with hyperactivity) (spontaneous hypertensive rat) speaks against a general nonsupression. SHR have a genetically determined excessive expression of mineralocorticoid receptors (MR) and a normal expression of glucocorticoid receptors (GR).
These results suggest that glucocorticoid receptors are not desensitized in ADHD-I.
The results further indicate that in ADHD-HI, in addition to the reduced cortisol response, which naturally leads to an insufficient shutdown of the HPA axis, there is also a (relative) overactivity of the MR or a desensitization of the GR. The nonsuppression of the HPA axis observed in hyperactivity when dexamethasone is administered proves that there is also nonsuppression in the case of a short-term high cortisol level, i.e. an inadequate shutdown of the HPA axis.
To summarize, we interpret this as follows:
- Melancholic / psychotic depression:
l High cortisol stress response
- Nevertheless no shutdown of the HPA axis
- → Desensitization of the GR or overactivity of the MR
-
ADHD-I
HSHigh cortisol stress response
- Shutting down the HPA axis
- → No desensitization of the GR or overactivity of the MR
- Atypical depression
typeLow cortisol stress response
- No shutdown of the HPA axis
- Not only due to a lack of sufficient cortisol stress response
- Also with artificial glucocorticoid administration (dexamethasone)
ch→ more frequent desensitization of GR (more likely) or overactivity of MR (less likely, as dexamethasone addresses GR 30 times more strongly than MR and yet nonsupression occurs)
-
ADHD-HI
HSLow cortisol stress response
- More often no shutdown of the HPA axis
- Not only due to a lack of sufficient cortisol stress response
- Also with artificial glucocorticoid administration (dexamethasone)
ch→ more frequent desensitization of GR (more likely) or overactivity of MR (less likely, as dexamethasone addresses GR 30 times more strongly than MR and yet nonsupression occurs)
8. Treatment of ADHD with cortisol / dexamethasone?¶
8.1. Dexamethasone as a medication for ADHD¶
Another strong indication that the HPA axis, and cortisol in particular, plays a central role in ADHD is provided by a study that found that dexamethasone (as a medication) in SHR rats (which are considered an animal model for ADHD-HI (with hyperactivity)) was able to significantly reduce their ADHD-HI symptoms.
However, it must be taken into account that ADHD with hyperactivity can be caused by a variety of different gene constellations and the SHR only represent a single gene constellation, so the result cannot be easily generalized for every ADHD-HI.
W
The dopamine D1 transporter in the PFC, which is damaged in schizophrenia, can be reactivated by cortisol in animal models, which normalizes the dopamine level.
Dexamethasone has so far only been studied for its basic efficacy in rats. With regard to ADHD, there are no studies on its use in humans or the effects of long-term use, which could potentially have side effects (e.g. on the immune system).
On the other hand, compensating for the low cortisol level in ADHD and especially in ADHD-HI - provided that only the existing deficit is compensated for - could also have a positive influence in other ways.
We have developed the idea of a dexamethasone shock treatment as a hypothesis: ⇒ Dexamethasone for ADHD
8.2. Cortisol as a medication for emotional stabilization in ADHD?¶
In one study, the administration of 20 mg to 40 mg cortisol significantly reduced feelings of sadness, activity, alertness to acute stress and improved relaxation. Fatigue was also reduced.
Cortisol (together with other mechanisms) influences attention, vigilance and memory. The hippocampus, which is responsible for memory storage and retrieval processes, has the highest cortisol receptor density in the brain.
An administration of 10 mg cortisol before learning vocabulary significantly worsened memory recall performance. Whether this only occurred in men or also in women, who could be protected against this by their sex hormones, remains to be seen. ADHD-I sufferers react to acute stress with an increased release of cortisol. Their memory recall performance is also impaired after stress - but only in men.
While the existing medications for ADHD (stimulants, atomoxetine, guanfacine) primarily improve attention and hyperactivity problems, we believe that they are not able to reduce inner tension just as effectively. According to our hypothesis, infrequent high-dose administration of cortisol should be able to calm the HPA axis again. Long-term or regular administration, on the other hand, is detrimental because it can cause effects similar to chronic severe stress.
Due to the considerable disadvantages of corticoids that act on the mineralocorticoid receptors, only corticoids that selectively address the glucocorticoid receptors (such as dexamethasone, which addresses GR to MR in a ratio of 30:1) should be considered.
8.3. Mode of action of dexamethasone¶
Dexamethasone is a selective glucocorticoid receptor (GR) agonist and therefore has the effect of increasing cortisol levels.
The cortisol receptor agonist dexamethasone promotes PACAP mRNA transcription, cell proliferation and DA synthesis, while a cortisol receptor antagonist inhibits this.
Cortisol acts on the β-adrenoceptor -cAMP/protein kinase A (PKA) signaling pathway to facilitate the synthesis and release of norepinephrine, which binds postsynaptically to ɑ1-adrenoceptor and β-adrenoceptor. A β-adrenoceptor directly coupled to adenylate cyclase (AC) can stimulate cAMP production and trigger certain responses that can be regulated by the ɑ1-adrenoceptor. Glucocorticoid receptors, together with the ɑ1-adrenoceptor, have an influence on the β-adrenoceptor-cAMP system. and can activate the NE system by activating the NE group of neurons in the brainstem. GR can also regulate transcription.
The activated GR bound to the glucocorticoid response element (GRE) influences gene expression as well as transcription factors, such as NF-kB, AP-1, CREB, etc., which in turn influence the combination of cAMP and PKA regulatory subtypes that model DA neuron production and survival and regulate DA activity and secretion.
8.4. Dexamethasone for ADHD-HI, GR antagonists for ADHD-I?¶
The studies that found dexamethasone to be effective as an ADHD medication used SHR rats, which are considered a typical model of ADHD-HI (with hyperactivity).
This raises the theoretical question of whether, just as the GR agonist dexamethasone could function as an active agent in ADHD-HI, a GR antagonist could possibly be effective in ADHD-I.
This is unlikely to be the case.
A GR antagonist would prevent the HPA axis from shutting down. However, firstly, the resilencing of the HPA axis after a stress response is usually a healthy response and secondly, it is probably not impaired in ADHD-I.
Studies on the dexamethasone test in ADHD show that problems with the deactivation of the HPA axis correlate with the degree of hyperactivity, but not with inattention.
More on this under Medication for ADHD, here *⇒ Dexamethasone for ADHD.*ß