The catecholamines adrenaline and noradrenaline have been regarded for a long time as important hormones in the regulation of arousal reactions. The Swedish physiologist Ulf von Euler received the Nobel Prize for his discoveries regarding the role of these catecholamines in the sympathetic nervous system (Von Euler 1972a, b). The fact that von Euler was active in Sweden had a great impact on applied research on stress mechanisms in our country. Lennart Levi (Von Euler et al. 1959) and Frankenhaeuser et al. (1961), who both became very important in Swedish stress research with their collaborators were involved in basic research with von Euler on the psychophysiological role of catecholamines in stress reactions in the late 1950s. For some time “stress” became almost equal to noradrenaline and adrenaline.

If we use Hans Selye’s terminology (Selye 1950) the stress reaction arises as an adaptation to a situation that requires energy. This could be a threat or a challenge, potentially either negative or positive in consequence. The initial phase (alarm or alert) is a short period during which the brain cortex is not aware of the energy requirement—it just happens in a communication between the emotional brain (which is faster than the cognitive brain) on one hand (mainly the amygdala) and the parasympathetic and the sympathoadrenomedullary system on the other hand. The first event in this chain of reactions results from inhibition of activity in the parasympathic system (the vagus nerve) with immediate elevation of heart rate. The second phase is the result of increased activity in sympathic nerves that use noradrenaline as transmitter substance at the synapses—resulting in further increase in heart rate. The third phase in the immediate stress reaction is the result of increased secretion to the blood of adrenaline from the adrenal medulla. During this phase the cognitive brain is aware of the situation and there are psychological reactions to it. The catecholamines are central in these reactions and several catecholamine-sensitive receptors have been described (Frankenhaeuser 1982). Alpha receptors are for instance important in vasoconstriction while beta receptors are important in dilatation of bronchioles (regulating breathing; of importance in asthma) and in the regulation of heart rate and force of heart muscle contraction. It has been pointed out that noradrenaline is chemically more primitive than adrenaline. Adrenaline is not produced in lower animals or in the new born baby.

Researchers should be aware that the assessment of catecholamines is problematic. There are good methodologies for bioassay, but the noradrenaline concentration in plasma does not reflect acute stress reactions only. There is substantial input of noradrenaline from skeletal muscles activity for and it may therefore be difficult to relate psychological stress reactions (in Selye’s sense) per se to changes in plasma concentration of noradrenaline. Changes in plasma adrenaline concentration do to a greater extent reflect “stress” reactions but the concentration is relatively low, and in addition there are rapid and marked fluctuations in concentration making it necessary to make a large number of assessments in the same individual during the period of interest, if the researcher wants to capture peaks and troughs in adrenaline excretion (Hjemdahl 1993; Goldstein 1995). Since a venous puncture may be painful one has to avoid repeated punctures—pain stimulates adrenaline excretion so this could be a major source of error. The solution during experiments is to insert a venous catheter remaining in place during the taking of multiple samples.

The excretion of catecholamines in urine is an interesting and valid parameter in the assessment of stress reactions. Collection of urine during a defined time period (for instance working hours or 24 h) allows a proxy estimation of the amount of stress reactions that the individual has experienced during that particular period. The problem created by rapid fluctuations of the excretion is avoided since the excretion is integrated during several hours. Since the excretion of both adrenaline and noradrenaline shows pronounced variations over the 24-h cycle it is necessary to control for time of day/night. The adrenaline excretion mostly reaches its peak during the mid-day when it is in the order of three times higher than during night. In work environment research (see below) it has therefore been common for instance to compare the adrenaline excretion during working hours with the excretion during the same hours of a work-free day. On the whole, longitudinal studies with many assessments in each one of the individual participants during the sequence to be studied are more interesting and easier to interpret than studies with only one assessment per individual. The reason for this is that individuals have different basal levels and these levels are determined to a great extent by factors that are not “stress relevant”. It has also been known for a long time that there are several confounders in this kind of assessment, for instance coffee drinking, certain medications and drugs, smoking and several kinds of food intake.

Ever since the classical studies performed by Lundberg et al. (1981) and by Lundberg and Frankenhaeuser (1999), researchers have kept asking themselves whether the different catecholamine reaction patterns that we observe in men and women are due to a “species difference” (difference in reactivity) or whether they are due to different patterns of stressors. In particular, it has been emphasized that the interplay between home and work stressors may be gender specific.

In the study of 21 female and 21 male managers and professionals in high-ranking positions by Lundberg and Frankenhaeuser published in 1999, the observation that received most of the attention from researchers was that the patterns of unwinding related to home-coming—as they were mirrored in the urinary excretion of catecholamines—differed between men and women. Jobs in these positions were perceived as challenging and stimulating by both men and women, but the situation was more favourable for men. Women (managers in Volvo) reported that they were more stressed by their unpaid workload and by their greater responsibilities for duties related to home and family. In line with this, women had higher urinary norepinephrine excretion than men both at work and outside work. In addition the urinary excretion of catecholamines decreased in men when they came home while this unwinding phenomenon was delayed in women.

Lack of unwinding, however, was also related to type of job. In this classical study there were observations on women with less prestigious jobs in the same organization. In those women, the unwinding related to home-coming seemed to take place in the same way as in men.

Lundberg et al. (1981) had also been studying the effects of two types of social stressors on urinary catecholamine output; the workday and going to a doctor with a sick child. Their data indicated that women had more sympathetic arousal during the doctor visit while men were more aroused by their working day. These observations pointed at gender socialization as an important factor in the triggering of arousal in the daily round of life.

During later years research projects on urinary catecholamine excretion and psychosocial stress has been less frequent. A study of 12-h overnight excretion of catecholamines has been presented by Masi et al. (2004). This group of researchers studied a diverse population-based sample of men and women aged 50–67 in the Chicago area and introduced a method adjusting for individual differences in muscle mass—which is of relevance to the excretion of catecholamines (see above), particularly noradrenaline. Their study showed that both epinephrine and norepinephrine excretion was higher in men than in women after adjustment for muscle mass.

The same group of researchers (Kalil et al. 2010) studied the effects of job insecurity on various indicators of stress in 190 older (born 1935–1952) men and women in Chicago. The findings indicated that the effect of job insecurity was significant on adrenaline (not noradrenaline) output in older men but not in older women. While there were no effects of job insecurity on catecholamines in women, job insecurity had stronger effects on psychological parameters (depressive symptoms, hostility and loneliness) in women.

Another study illuminating gender differences was performed on 315 male and female nurses in the ages 20–60 (Deane et al. 2002). This study showed that premenopausal female nurses had consistently lower levels of urinary catecholamines than male nurses in the corresponding ages. The authors speculated about a role of female sex hormones in this gender difference.

Another group of researchers, Janicki Deverts et al. (2007), used the CARDIA study, which is another American population study. They showed a very consistent pattern of decreasing urinary catecholamine output (particularly adrenaline but to some extent also noradrenaline) with rising socioeconomic status (regardless of whether income, education or occupation was used as indicator of socioeconomic status) in young adults. This association was stronger for men than for women. This type of finding is consistent with the early studies indicating that occupational activity in general may be more important for catecholamines in men than in women.

Studies of stress related variations in plasma catecholamines and their possible gender specific effects on the cardiovascular system have also been studied. In a small experimental study, Schouwenberg et al. (2006) showed that when adrenaline (dosage adapted to body weight) was infused continuously for 20 min, the plasma adrenaline levels increased in the same way in men and women. The cardiovascular effects were different, however. Women had more elevation of systolic blood pressure whereas men had more elevation of heart rate. The authors speculated that the cardiovascular response in premenopausal women might be more alpha-receptor dominated whereas in men there may be more dominance of beta receptors. While this is speculation only and based upon one small study, it points to the possibility that even when the effects on plasma levels are the same in men and women, the resulting effects on the cardiovascular system may still be gender specific due to differences in receptor sensitivity.

Thus studies published during later years speak in favour of the hypothesis that men have a higher adrenaline output than women and that part of this gender difference may be due to occupational activity. It has been speculated that part of the difference may also be due a “stress protective” role that the female sex hormones may play.

When a stressful experience does not fade away immediately, the next system, the HPA axis (Hypothalamo-Pituitary-Adrenocortical axis), reacts in response to the energy demands. This is observable after minutes of exposure. The stimulation of the HPA axis starts in the hypothalamus part of the brain from which the peptide CRH (corticotropic releasing hormone) stimulates the pituitary to release ACTH (adrenocorticotropic hormone), which in turn stimulates the adrenal cortex to release corticosteroids. In a number of ways, these corticosteroids help the organism sustain its fight in a stressful situation. In the acute situation, this is purposeful since the release of energy is facilitated by mobilization of fuel for energy requiring actions (carbohydrates and free fatty acids), and there is retention of salt and fluid which may otherwise get lost in an uncontrollable way in a physically demanding situation. There is also inhibition of acute inflammatory reactions. In the stress reaction cortisol is of particular importance. Cortisol is important for all of the functions mentioned. It is one of the corticosteroids produced in the adrenal cortex and could be regarded as the main agent in the HPA axis. Cortisol has a relatively high concentration in plasma. It exists both in protein-bound and in free form. Like catecholamines it can be assessed in plasma and urine. The free plasma cortisol, however, also communicates with saliva. After some minutes an increased concentration of cortisol in plasma will be mirrored in saliva. This offers interesting possibilities since the collection of saliva is painless and repeated specimens can easily be collected. During later years the assessment of saliva cortisol has had a dominant position among biological stress measures.

Salivary cortisol has been extensively studied since Kirschbaum and Hellhammer (1989) published their report on the usefulness of salivary cortisol in studies of stress. A recent review (Kristenson et al. 2011) of all published studies in this particular field has shown the limitations and possibilities of salivary cortisol as a stress marker. The question regarding possible gender differences in physiological reactivity to different kinds of situations has been discussed in this research field as well. Stroud et al. (2002) have summarized this by stating, that “women appear to be more physiologically reactive to social rejection challenges (such as being systematically excluded by associates during a conversation) but men react more to achievement challenges (where study participants were told that the investigator studied the relation between intelligence and performance)”. These observations remind us of those made by Frankenhaeuser et al. mentioned above.

Studies of the relationship between environmental stressors and saliva cortisol do not show any consistent gender difference, neither in general (Garvin et al. 2011) nor in relation to work stressors (Karlson et al. 2011). On the basis of previous observations on gender differences, however, such differences are likely to be context dependent and it is possible that the work environment studies have not taken this into account. In studies of the general population it should also be remembered that women’s and men’s labour markets are different. Whereas men are overrepresented in transportation and industry, women are overrepresented in care and lower education, for instance. This may be of great importance to the interpretation of findings in work stress studies.

In a study published from Stockholm—the PART study which is a population study of psychiatric morbidity in the Stockholm region—saliva cortisol profiles were collected from subsamples of working men and women (n = 181 and 348, respectively), the group of women with the best working conditions (according to the demand control model) namely those who did not report excessive demands or lack of control (low strain) had significantly lower mean saliva cortisol concentration than other working women half an hour after awakening in the morning (Alderling et al. 2006). The most likely interpretation of this is that those women who had favourable working conditions had less arousal after awakening than the other women. Among working men there were no significant differences between the demand/control categories. In the same study (Alderling et al. 2008) the diurnal saliva cortisol profiles were related to psychiatric conditions assessed by means of standardized psychiatric interviews. The most significant finding was that women with anxiety disorder had a more pronounced increase in saliva cortisol from awakening to half an hour later (awakening response) and that women with sleep disorder had a significantly lower saliva cortisol at awakening and then remained on a lower saliva cortisol level throughout the day than healthy women. Together these two studies suggest that women with favourable working conditions may be in a better position to cope with everyday stressors and that proneness to anxiety is associated with marked cortisol responses to the expectations associated with the ordinary workday. In exhausted women (sleep disorder) low saliva cortisol levels were observed. In men there were no such associations.

Another study of the effects of environmental stimuli was performed in six metropolitan areas in Europe (Selander et al. 2009). Part of the participants in all the cities were living close to a large airport where they were exposed to airport noise whereas others did not have such exposure. Saliva samples were collected at awakening, at lunch and in the evening in 230 women and 209 men. The results showed that there was a considerable and significantly higher saliva cortisol awakening concentration in women exposed to at least 60 dB of air traffic noise. There were no corresponding associations among men exposed to air traffic noise. An interpretation of this is that the total stressor exposure in the morning was higher in women than in men making the extra burden of air traffic noise stressful.

On the other hand, it has been repeatedly shown that laboratory stress (Kudielka et al. 2009) induces a more pronounced increase in saliva cortisol concentration in men than in women. In particular this seems to be the case with social threat (the Trier Social Stress Test). A recent experimental study (Edelman et al. 2012) on 46 men and 46 women below age 35 has furthermore shown that a specific glucocorticoid receptor (NR3C1) may be of importance to the cortisol reaction in women but not in men and that this effect was further amplified by methylation of the receptor. This means that the protection against a pronounced saliva cortisol elevation that has been observed in women may be regulated by both genetic and epigenetic factors. Methylation corresponds to the environment’s ability to “turn on” and “switch off” the sensitivity of genes that (in this case) regulate cortisol response. These findings have to be replicated in other studies, however.

In adult women and men there may accordingly be differences in cortisol response patterns as discussed by Stroud et al. (2002). This may however be a phenomenon that arises in the adult person. It could be due to epigenetic phenomena in the form of differential activation of steroid receptors. It is certainly also due to gender differences in stressor exposure patterns. The gender difference in cortisol excretion has not been observed in small children, however. In a study of contrasting cohorts, children growing up in anthroposophic families are compared with children growing up in families who are partly anthroposophic and in families who are normal families. The anthroposophic upbringing of children is focussed on a low level of stressors especially for the small children, both physically (noise, temperature, light) and psychologically (avoidance of upsetting situations). The studied families are living in the same neighbourhood and the groups are comparable with regard to socioeconomic conditions. Salivary cortisol has been assessed in the morning, mid-day and at bedtime in these children at the age of 6, 12 and 24 months (Stenius et al. 2008; Swartz et al. 2012). The parents’ salivary cortisol has been assessed on the same occasions. The saliva cortisol concentration has been shown to be consistently and significantly lower in the children growing up in the anthroposophic families than in the other children but there was no difference between girls and boys from this point of view. Nor did the parents in the three groups differ with regard to saliva cortisol levels, neither fathers nor mothers. The correlations between the saliva cortisol levels in father and child and mother and child were similar to those observed between cortisol levels in mother and father. Thus there was no indication of a strong genetic component.

There is also a “good” counterbalancing system—an anti-stress system that protects from adverse effects of long lasting stress. This HPG (Hypothalamo-Pituitary Gonadal) axis has the same levels as the HPA axis ranging from the hypothalamus to the gonadal glands. HPG axis represents the “regenerative” or “anabolic” part of metabolism. The male testes and the female ovaries are two of the end organs of this axis and they represent the extremes of this activity, namely reproduction. “Building a new human individual” is of course the most pronounced “anabolic/regenerative activity” that the body can be involved in. Building new cells and repairing worn-out tissues is closely related to this, however. However, some of the production of anabolic corticosteroids takes place in the adrenal cortex. These are specifically testosterone and its precursors dehydroandrosterone (DHEA) with its water soluble sulphate form (DHEA-s). Accordingly, some of the release of these regenerative hormones takes place in the adrenal cortex (notably testosterone in women) together with the corticosteroids that are important for energy mobilization. This means that cells producing hormones mainly active in regeneration are located very close to those cells in the adrenal cortex that are producing and releasing hormones mainly active in energy mobilization. The two forces “energy mobilization” and “regeneration” are balancing one another on all levels of the HPA and HPG axes. It has been shown that long-term stress reduces the basal secretion of DHEA-s and also the DHEA-s response to challenging situations (Lennartsson et al. 2013a, b). The synthesis of DHEA-s will be inhibited during long periods of adverse stress.