Background:

NE and EPI are specifically assessed to elucidate the role of the sympathetic nervous system (SNS) in the mediation of biological stress pathways. The main sites of NE and EPI release are the brain (locus coeruleus (NE)), sympathetic neurons (NE), and adrenal medulla (EPI and small amounts of NE). NE and EPI are synthesized from the amino acid tyrosine (Figure 1). NE and EPI exert their diverse physiological effects via binding to G-protein-coupled adrenergic receptors (AR) and subsequent activation of signaling pathways. Whereas both a1-AR and b1-AR have a higher affinity for NE than EPI, b2-AR has a higher affinity for EPI than NE; a2-AR has a high affinity for both NE and EPI. For example, pertinent to the acute stress response, catecholamine stimulation of a1-AR results in vasoconstriction, of a2-AR in platelet activation, of b1-AR in increased heart rate, and of b2-AR in normal bronchodilation and natural killer cell adhesion. NE (also referred to as noradrenaline) detectable in the circulation mainly originates from sympathetic nerves that envelope arterioles and infuse organs such as the heart and kidneys. Specifically, plasma NE levels are applied in assessing SNS activity and acute, stress-induced changes therein. Plasma EPI (or adrenaline), is mainly secreted by the adrenal medulla’s chromaffin cells directly into the circulation, reflecting the neural outflow to the adrenal medulla (Goldstein, 2010a-b) and mainly adrenal-medullary-axis activity. However, it has been reported that a small portion NE is also released by the chromaffin cells (Eiden & Jiang, 2018; Tank & Wong, 2015). Therefore, approximately 80% of EPI and 20% of NE are released into circulation by the chromaffin cells (80%EPI:20% NE release ratio).

​

​

​

​

​

​

​

​

​

​

​

​

​

​

Stress research and NE and EPI

During conditions of acute physical, psychological, or environmental stress, the SNS is rapidly stimulated. This stimulation leads to large concurrent increases in NE and EPI, necessary to produce the changes in peripheral organ functions required to adapt to the stressor (e.g. physical activity, mental strain, extreme temperature exposure) (Tank & Wong, 2015). However, certain conditions evoke far greater increases in EPI compared to NE, which is highly dependent on stress severity as well as receptor affinity and distribution in the heart and vasculature (Wortsman et al, 1984; Malan et al, 2016). Plasma EPI levels increase as much as 10-30 fold during insulin-induced hypoglycaemia and post-surgical stress (Mannelli et al, 1982), whereas NE only increases by 2-fold in both cases (Blandini et al, 1995). Contrarily, immersion in cold water evokes a marked NE response compared to EPI change (Hasselhund et al, 2010). Furthermore, EPI responses are closely related to corticotropin responses – more so than NE responses – signifying a relevant link between hypothalamic-pituitary-adrenocortical (HPA) and adrenomedullary activation. This suggests that a more “physical” stressor (e.g. hypoglycaemia), leads to adrenomedullary activation and EPI release, in close association with HPA responses (Goldstein 2010b), more so than an association between adrenomedullary and sympathetic NE responses. Alternatively, a more “psychological” stressor (e.g. acute mental, psychological, psychosocial stress), leads to greater NE responses when compared to EPI responses (Goldstein 2010b). Yet, EPI is involved in processes concerning memory consolidation and alertness (Jänig 2006). Schommer et al (2003) reported dissociation between the HPA and sympathetic-adrenal-medullary (SAM) axis responses to repeated acute psychosocial stress. They particularly showed that HPA responses quickly habituated to repeated psychosocial stress, whereas the SNS (specifically the NE response) showed a rather uniform activation pattern with repeated exposure to a psychosocial challenge (Schommer et al, 2003). This might propose an explanation as to why, during chronic stress, elevations of NE are noted, even though lowered EPI might be observed (Wurtman, 2002). However, lowered NE has been related to ischemic heart disease risk in those experiencing chronic stress (Malan et al, 2017). It is important to emphasize this difference, as it may influence the interpretation and compilation of measures (e.g. allostatic load scores) when evaluating NE and EPI levels in acute and chronic stress settings.  In many cases, the NE/EPI ratio is applied to take these variations into account (Mancini & Brown, 1992; Jost et al, 1985; Vollmer, 2009). Consequently, the variable release of NE and EPI in response to a variety of stressors might suggest specificity in the overall sympathetic response to different stressful stimuli – a specificity that during chronic stress does not necessarily infer that NE and EPI levels will change in a parallel manner. 

​

Catecholamines do not cross the blood-brain-barrier (Kostrzewa, 2007); however, NE and EPI in peripheral circulation have been shown to produce effects mediated by the central nervous system (Tank & Wong, 2015). A significant example of the effect of stress-induced elevation of peripheral NE and EPI on central brain function is that of learning and memory, where the link between stress and memory is well-established in human and animal studies, illustrated in a review by Tank & Wong (2015). Furthermore, NE measured in the periphery and central (brain) NE may present an inverse relationship. This is apparent in patients that suffer from major depression, psychotic depression, and seasonal affective disorder (Moret & Briley, 2011; Goekoop et al, 2012). These patients may present with high plasma (peripheral) NE levels accompanied by low NE levels in the brain.

Chronically low NE levels were related to increased cardiovascular risk in a prospective bi-ethnic cohort (Malan et al, 2017). Notably, a great deal of controversy exists regarding the contributions of the SNS to HRV parameters, specifically frequency domain parameters (Reyes del Paso et al, 2013). Although changes in HRV during EPI and NE infusions have been reported in mammals (Ohmura et al, 2006), high levels of plasma NE do not necessarily indicate high sympathetic nerve traffic (Grassi et al, 2015; Esler et al, 1991; Meredith et al, 1992), as indicated by microneurography, in humans. Nonetheless, the complex relationship between plasma NE and sympathetic nerve traffic does not prohibit the use of plasma NE for research or clinical diagnostic purposes, but should remind us that plasma NE levels are to be interpreted cautiously, while always considering the context and design of the study, conditions and time of sampling and patient/ participant characteristics and history.

Due to the short half-life of NE and EPI (1-2 min), their widespread physiological role and complex interplay between reuptake and release, infusions of NE and EPI can be used to determine the kinetic mechanisms involved in acute stress activation (Clayton & Williams, 2000; Wirtz et al 2006-2009; von Känel et al, 2018). Plasma NE concentration is highly dependent on both the rate of NE release into the plasma as well as the rate of NE removal from the plasma. Direct adrenergic activation of the solitary tract nucleus (induced by EPI administration) potentiated amygdala NE release and aided the retention performance of emotionally rousing and spatial memory tasks in rodents (Clayton & Williams, 2000). In human subjects, acute stress-induced increases in NE were observed and associated with poor emotional regulation and low social support (Wirtz et al, 2006). NE and EPI responses related to emotional, cognitive (Phillips 2017), endocrine (Schug et al, 2015), and cardiovascular outcomes (Huang et al, 2013). Indeed, various studies demonstrated that acute cardiovascular responses during mental stress predicted the development of hypertension more accurately than resting blood pressure monitoring (Matthews et al, 1993; Wood et al, 1984). Acute responses to mental stress may elicit increases in cardiac output, systemic vascular resistance, and blood pressure. NE and EPI elevations to acute stress are transient in nature, but frequent and prolonged elevations of NE and EPI may result in exaggerated vasoconstriction of systemic arteries, leading to allostatic changes in the overall cardiovascular response (Gidron et al, 2002). These changes create the ideal, adverse conditions for the development of endothelial dysfunction, arteriosclerosis, hypertension, and stroke susceptibility (Huang et al, 2013; Spieker et al, 2002; Gidron et al, 2002). This may support the hypothesis that increased stress susceptibility possibly contributes to all-cause morbidity and mortality. Increased NE responses to acute stress also related to increased proinflammatory responses. A prothrombotic response was also noted when NE (von Känel et al, 2018) and EPI (von Känel & Dimsdale, 2000) were directly infused, elucidating the possible catecholamine-driven mechanisms by which acute mental stress may trigger acute coronary syndromes (Kivimäki & Steptoe, 2018).

​

NE, EPI and cortisol

As the adrenal medulla releases EPI and NE, the adrenal cortex releases steroids, including cortisol. Unlike EPI and NE, cortisol can readily cross the blood-brain-barrier, where it may produce direct effects on cells that determine other steroid releases (Goldstein, 2010) – thereby propagating a direct negative feedback loop with direct-acting chemical messengers. Cortisol release is regulated by the HPA-axis and, as previously mentioned, there is a relevant association between HPA-axis and adrenomedullary activation. The HPA-axis may function to inhibit sympathetic and adrenomedullary systems, terminating the immediate defense response, whilst favoring behavioral adaptation (Malan et al, 2017). The HPA-axis and SNS are alternative effectors for various homeostatic systems, which follow the principle of compensatory activation. Administration of corticotropin releasing hormone (CRH) in the brain not only increases peripheral levels of adrenocorticotrophic hormone (ACTH), but also EPI and NE, consistent with the stimulation of both the adrenomedullary hormonal system and SNS (Eto et al, 2014). One of the most prominent effects of cortisol is to facilitate the synthesis of EPI in the adrenal medulla, by promoting the activity of phenylethanolamine N-methyltransferase (the enzyme that converts NE to EPI) (Eto et al, 2014). Indeed, pathological conditions involving decreased ACTH, failure of the adrenal cortex or decreased cortisol production usually accompany decreased EPI production (Goldstein, 2010).

Clinical conditions, observations and NE and EPI

Alterations in NE and EPI levels often accompany stress-related psychiatric disorders. As central (brain) NE is a key player in executive functioning, regulating cognition and motivation, low plasma levels, comparable to low central NE levels, are observed in patients suffering from depression (Moret & Briley, 2011; Lambert et al, 2000). However, increased plasma NE levels have also been reported psychotically depressed compared to non-psychotic depressed patients possibly due to a high-stress level in the former group (Goekoop et al, 2012). Contrarily, increased urinary excretion of NE and EPI were associated with depressive symptoms and increased anxiety (Paine et al, 2015; Otte et al 2005).

Depression, anxiety and post-traumatic stress disorder (PTSD) often occur co-morbid, each contributing their share of changes in plasma and/or urine NE and EPI levels. Patients suffering from PTSD presented with significantly higher NE levels than controls, however, no significant differences were observed in EPI levels (Pan et al, 2018). Indeed, recent evidence indicated that a right-sided stellate ganglion block significantly reduced PTSD symptoms in patients after 8 weeks (Rae Olmested et al, 2019). Adverse childhood experiences are often identified as a determining factor in the development of specific stress-response mechanisms, severity of depression, anxiety, and PTSD (Yehuda, 2001). This may be because developmental factors, early-life experiences, and exposure play a prominent role in establishing biological stress responses and acute stress activation patterns (Gillies et al, 2012). It is postulated that higher NE and lower EPI contributes to the modulation of aggressive behavioural responses in individuals who have experienced aggressive stressors early-on (Coccaro et al, 2003), although attenuation of HPA-axis responses may precede such changes, increasing stress susceptibility later in life (Bodengom et al, 2017). Children as well as adults with attention deficit hyperactivity disorder (ADHD), exhibit lower NE and EPI levels and are often treated with medications increasing plasma NE concentrations/ prohibiting NE dysregulation. Distinct NE responses and elevated NE levels are also observed in a range of substance abuse disorders (Fitzgerald 2013), personality disorders (Simeon et al, 2007) and sleep disorders. In obstructive sleep apnea, elevated urinary and plasma NE and EPI levels have been recorded. In general, sleep deprivation, loss, or disordered sleep, are associated with elevated nocturnal levels of these catecholamines (Irwin et al, 1999). Patients with chronic fatigue syndrome presented with lower baseline levels, attenuated stress responses, and recovery of NE and EPI following a physical stressor (Strahler et al, 2013). Insufficient NE and EPI responses might explain some of the characteristic symptoms of chronic fatigue syndrome.

Alterations in NE and EPI baseline levels and stress-induced responses are also evident in disorders of catecholamine biosynthesis, metabolism, familial dysautonomia, syndromes of chronic autonomic failure, functional dysautonomia, surgical sympathectomies and neoplastic disorders of catecholaminergic cells (Goldstein 2010a).

​

NE metabolites and associated markers

As merely a fraction of NE derived from sympathetic nerves enters the circulation unchanged (Goldstein, 2010a), metabolites of NE can also be evaluated. The principal metabolite of NE, 3,4-dihydroxyphenylglycol (DHPG), provides unique insight into NE dynamics and turnover (Denfield et al 2018). Another NE metabolite, 3-methoxy-4-hydroxyphenylglycol (MHPG), is used as a surrogate marker for neural NE activity. Salivary alpha amylase (sAA) has emerged as a valid marker of autonomic nervous system activity in behavioural research. sAA is a salivary enzyme involved in the digestion of complex carbohydrates, which is used, within the context of acute stress evaluation, as a proxy measure for sympathetic arousal (Ali & Nater, 2020). However, there is some concern about whether sAA measured during stress reflects purely SNS activity or a combination of sympathetic and parasympathetic activation. Some evidence suggests that sAA may reflect central NE activation (Nater & Rohleder, 2009; Nater et al, 2005; Warren et al, 2017). Recent studies also discovered the relationship between acute stress induced changes in sAA and NE responses (Kuebler et al, 2014; Thoma et al, 2012), indicating that NE may stimulate sAA release. This may signify the use of sAA to predict changes in plasma NE, and to investigate whether stress-induced increases in sAA may be triggered by NE. In a study where the NE transporter blocker atomoxetine led to increases in sAA, changes in salivary cortisol were also positively correlated with sAA, supporting the theory that changes in sAA may reflect central NE (Warren et al, 2017). Petrakova et al (2015) found that, following psychological stress, both sAA and plasma NE reflected SNS activity, yet independently from one another. Attenuated sAA levels in response to acute mental stress were found in individuals suffering from burnout, alcohol dependency, and those with a history of childhood maltreatment (Muehlhan et al, 2017). The research context (e.g. acute stress testing or pharmacological intervention etc.) should be considered when interpreting sAA results.

​

Collection and Measurement

Resting plasma levels of NE and Epi are low, but adequate to maintain the normal sympathetic tone and blood pressure. Plasma and urine samples can be used to assess NE and EPI levels. Serum samples are unsuitable since catecholamines are stored in platelets and may be released during the clotting process. Increases/ decreases in urinary NE and EPI are reported to be the most comparable to respective changes in plasma NE and EPI, rather than their metabolites, the latter also measurable in urine (Reuben et al, 2000).

​

Sample collection

General sample requirements and precautions noted by the Association for Clinical Biochemistry (2013) state that a strict protocol is to be maintained for the analyses of both plasma NE and EPI. Patients or participants should be in a supine position and an indwelling catheter inserted (antecubital vein) 30min prior to the collection of baseline samples. Samples should be collected in a chilled lithium-heparin tube. Both NE and EPI are labile in plasma, therefore samples should be centrifuged at 4°C no more than 20min after sample collection. The resulting plasma should then either be frozen immediately or kept on dry ice with long-term storage at -80°C. When assessing NE and EPI in urine, samples should be collected in containers with a pH of less than 3.5 to 4 (e.g. for 3-5 L containers, filled with12ml 6 N HCl (or 9ml 20% HCl (UriSet24, Sarstedt®, Nümbrecht Germany)) to preserve and prevent the breakdown of the catecholamines and either kept on ice or refrigerated until aliquoted. In adults, various studies report the preferred use of 24h collection, although it is comparable with 8h and 12h collected samples (Paine et al 2015; Reuben et al, 2000). In children, however, the random collection is preferred as 24h urine sampling is often incomplete in child-cohorts. However, it is not always feasible to collect 24h urine samples, so 12h overnight collections are an alternative. In this case, it is common practice to use urinary creatinine as volume correction and to present urinary NE or EPI as a NE:creatinine/ EPI:creatinine ratio. It is important to note that creatinine is affected by muscle mass, body surface area, age, gender and ethnicity, therefore these factors should always be considered when reporting creatinine ratios.

​

Measurement

NE and especially EPI are powerful hormones, whose plasma levels are very low under normal conditions. Therefore, highly sophisticated and sensitive methods, such as high-pressure liquid chromatography (HPLC), liquid-chromatography-mass-spectrometry (LC-MS-MS), gas chromatography-mass spectrometry (GC-MS) and more recently developed carbon dot assays are employed to analyse NE, EPI and their associated metabolites (Jung-Klawitter & Huebschmann 2019; Zhang et al, 2019; Peaston & Weinkove 2004). Manual immunoassays are valid, yet should be applied with great caution to guarantee accuracy. When assessing levels in urine, the volume of the sample should be considered, and/or the urinary creatinine value should be calculated.

​

Analysis of NE and EPI changes/ responses

Experimental stress protocols refer to cognitive stress testing (Stroop), public speaking tasks (Trier Social Stress test, which incorporates key psychological elements such as social-evaluative threat and uncontrollability), noise exposure, emotional provocation, pain induction (Allen et al, 2017) and physical stress (e.g. exercise, cold pressure test). The magnitude of EPI and NE responses evoked by either mental-, social- or physical stressors are highly dependent on the composition of the study population. For instance, the circulating levels of catecholamines increase in healthy cyclists during exercise (Messan et al, 2017), but may show a slower/ diminished increase in students during similar exercise testing. A review published by Sluiter et al (2000) compared responses in different occupations.

​

The general procedure for assessing EPI and NE responses during mental or physical stress testing entails that the study participant remains in a supine position for 30 min prior to baseline sample collection, followed by sample collection at intermittent periods/ during the stress task. Some studies report as much as 3 samples during mental stress testing and 2 samples during physical stress – followed by a return to baseline samples taken 15min after the last stress test (Hasselhund et al, 2010). It is important to note that special care should be taken regarding time-of-day testing, to ensure comparability between participants/ patients as well as the timing of sampling during stress testing, when performing more than one stress test – to prevent the residual stress of one affecting the other.   Long-term stability of NE and EPI stress responses after 18-years have been reported (Hasselhund et al, 2010). However, Jern et al reported that stress-induced changes in plasma EPI and NE showed low reproducibility within 1h of repeated testing (1991). No habituation was seen in the magnitude of both NE and EPI responses to acute mental stress in healthy subjects who were tested three times with 4-week intervals (Schommer et al, 2003).

​

When determining the change in response to one applied stressor at one point in time, the difference from baseline or the percentage change can be calculated. Repeated measures across several time points should be taken to compare individuals or groups in terms of responses during different time intervals and recovery periods from acute mental stress. The area under the curve (AUC) is often determined to integrate data from repeated measures over time and allows for simplification of the analysis and interpretation of the results without sacrificing data. Both the change and intensity of change can be determined for a set period. This method is employed in population-based studies, clinical trials, and pharmacological studies. There are variations on the formulas used to calculate the AUC, and its applicability is described in Pruessner et al 2003. Multi-level modeling is an alternative analysis tool for repeated measures data in case of missing data points.

​

Strengths:

• Although currently, no “gold-standard” marker of SNS activity exists, NE and EPI are considered as some of the most widely accepted markers in determining various stress-related pathologies, psychopathologies and research of stress-induced health risks.

• NE and EPI can both be assessed in plasma and urine – the latter providing a less invasive approach, advantageous, specifically in vulnerable patient and participant populations.

• Surrogate markers, such as sAA offer a highly non-invasive relation to NE responses, yet, due to contrasting findings, should be interpreted with caution.

• High-sensitivity methods are constantly being refined to determine NE and EPI values with even greater accuracy.

• Aside from acute stress pathway activation, NE and EPI responses may reflect changes evolving or contributing to chronic stress – evidenced by the increased allostatic load (Seeman et al, 2001).

​

Limitations and Specific considerations:

• Difficulties regarding interpretation due to site of release, potentially interacting compounds, confounding factors, including chronic stress experiences, gender (Bangasser et al 2018), genetic dispositions, psychological distress, panic disorders, chronic depression, and altered receptor sensitivities and densities in the brain and target organ structures that influence its release.

• Specific factors to consider (Goldstein 2010a), many of which also apply to EPI:

1. The majority of NE released is recycled back into the nerve terminals, only a small fraction truly reaches the circulation. This increases the probability of erroneous interpretation, as a patient with normal sympathetic nerve traffic may have high plasma NE levels if the cell membrane NE transporter is blocked (e.g. calcium blockers, sympathomimetics, anti-depressants, anti-hypertensive medication – please refer to Table 1 for an overview of the classes of medication that may influence NE and EPI levels and should be taken into consideration when interpreting NE and EPI levels). Additionally, plasma EPI HPLC analyses may be influenced by a metabolite of labetalol, whilst urine EPI may be influenced by paracetamol;

2. Plasma NE is determined by both the rate of entry of NE into plasma and the rate of clearance of NE from plasma;

3. NE is released by sympathetic terminals via the action of several co-factors, enzymes, and transport molecules. Therefore, disruption in one of these factors may influence plasma NE levels, regardless of normal sympathetic traffic;

4. In most cases, the antecubital vein is used to sample plasma NE in humans. As sympathetic activity in the forearm and hand influences NE levels in the antecubital venous plasma, venous levels of NE measured in the arm may not translate to whole-body sympathetic activity;

5. Plasma NE levels are greatly influenced by posture, time of day, fasting status, environmental and room temperature, dietary factors, and various prescription and non-prescription medications (A comprehensive description by Farzam & Lakhkar, 2019).

• Although there are rudimentary reference values for NE and EPI in plasma and urine samples, these values are derived from population-specific observations and no globally accepted reference values currently exist clearly signifying or categorizing acute/ chronic stress conditions.

• Different methods for analyses (automated apparatuses, manual immunoassays, HPLC etc) – should be standardized across laboratories for more comparable results.

• Most allostatic load scores include NE and/or EPI. Once universal reference values/ranges have been established, this might help to advance our knowledge on the role of the SNS in diseases related to chronic stress. This may further contribute to refining the outcomes of the allostatic load scores – both for basic and clinical research relevance.