Nature Reviews Cardiology nature reviews cardiology https://doi.org/10.1038/s41569-023-00962-3 Review artic e Check for updates Cardiovascular autonomic dysfunction in post-COVID-19 syndrome: a major health-care burden Artur Fedorowski 1,2,3 , Alessandra Fanciulli 4 , Satish R. Raj 5,6 , Robert Sheldon 5 , Cyndya A. Shibao 6 & Richard Sutton 3,7 Abstract Cardiovascular autonomic dysfunction (CVAD) is a malfunction of the cardiovascular system caused by deranged autonomic control of circulatory homeostasis. CVAD is an important component of post-COVID-19 syndrome, also termed long COVID, and might affect one-third of highly symptomatic patients with COVID-19. The effects of CVAD can be seen at both the whole-body level, with impairment of heart rate and blood pressure control, and in specific body regions, typically manifesting as microvascular dysfunction. Many severely affected patients with long COVID meet the diagnostic criteria for two common presentations of CVAD: postural orthostatic tachycardia syndrome and inappropriate sinus tachycardia. CVAD can also manifest as disorders associated with hypotension, such as orthostatic or postprandial hypotension, and recurrent reflex syncope. Advances in research, accelerated by the COVID-19 pandemic, have identified new potential pathophysiological mechanisms, diagnostic methods and therapeutic targets in CVAD. For clinicians who daily see patients with CVAD, knowledge of its symptomatology, detection and appropriate management is more important than ever. In this Review, we define CVAD and its major forms that are encountered in post-COVID-19 syndrome, describe possible CVAD aetiologies, and discuss how CVAD, as a component of post-COVID-19 syndrome, can be diagnosed and managed. Moreover, we outline directions for future research to discover more efficient ways to cope with this prevalent and long-lasting condition. Sections Introduction Cardiovascular autonomic dysfunction CVAD in long COVID Detection and diagnosis Management Knowledge gaps and future directions Conclusions A full list of affiliations appears at the end of the paper. e-mail: artur.fedorowski@ki.se Nature Reviews Cardiology Review article We are at the early stages of understanding long COVID and the associated cardiovascular autonomic dysfunction (CVAD), but it seems to be similar to other post-viral syndromes 9,10 . In particular, there is sub- stantial overlap with myalgic encephalitis/chronic fatigue syndrome (ME/CFS), a condition that is frequently encountered in both long COVID 1,11 and postural orthostatic tachycardia syndrome (POTS) 12 . As in many post-viral syndromes, CVAD in patients with long COVID often seems to be worse than the COVID-19 itself, and the severity of the CVAD might not correlate with that of the initial infection 9 . However, the incidence of long COVID was most prominent with the early strains of the virus and seems to be less common with later strains. Factors that might have contributed to this change include amelioration by vaccination before infection and by antiviral therapy during the acute infection 4,13,14 . In parallel, several reports have suggested a temporal association between the onset of CVAD and recent COVID-19 vaccina- tion among some individuals who might be susceptible to adverse post-vaccine reactions 15–17 , although the extent of this complication has not been fully explored 18 . The prognosis and potential long-term mortality with long COVID-associated CVAD are not yet known 19 No specific treatments for long COVID are available, and current strategies rely on supportive care only. Many cases of long COVID-associated CVAD are similar in pres- entation to POTS and inappropriate sinus tachycardia 20–23 , conditions that are often attributed to post-viral syndromes 24–26 . Other forms of autonomic dysfunction affecting the cardiovascular system can also be seen, including hypertension 27 , orthostatic hypotension 28–30 , initial orthostatic hypotension 31 , orthostatic hypertension 32 , abnormal blood pressure (BP) variability and disruption of circadian BP rhythm 33,34 , low BP and hypotensive tendency 35–38 , chronotropic incompetence 39 , abnormal sinus bradycardia 40 and recurrent reflex syncope 37,39,41–43 In some patients, mild symptoms suggestive of CVAD might have been present before the COVID-19 illness, and the infection might have exac- erbated pre-existing CVAD. In other cases, patients had no pre-existing CVAD symptoms and a clear temporal association between COVID-19 and the development of CVAD can be documented. A common feature in all these conditions is that they can present with a substantial symp- tom burden, but the abnormality might be difficult to detect, posing a great diagnostic and therapeutic challenge for many clinicians 44 The focus of this Review is on understanding the autonomic aspects of the cardiovascular disorders in post-COVID-19 syndrome. Cardiovascular autonomic dysfunction CVAD or dysautonomia is a general term for a broad spectrum of con- ditions (not a specific disorder), in which autonomic control of the circulation, at a global and a local level, is impaired 41,45 (Box 1). The cir- culatory system depends on the appropriate action of three main components: the blood pumping ability of the heart, the blood trans- ferring ability of the blood vessels, and blood volume control, mainly through water ingestion, reabsorption and elimination, governed mainly by the kidneys 46 . All these crucial elements are controlled by the autonomic nervous system, directly via neural connections and indirectly via specific neuroendocrine adaptations 46,47 . Under normal conditions, humans are unaware of the adaptive processes in the circu- latory system resulting from regulation by the autonomic nervous sys- tem. One example is the response of the autonomic nervous system to counteract the changing circumstance of assuming a standing position in order to maintain BP and blood flow to the brain, thereby maintaining cardiovascular homeostasis 48 . When we stand, a baroreceptor reflex leads to increased heart rate (HR) and peripheral vascular tone, and Key points • Cardiovascular autonomic dysfunction (CVAD), in particular postural orthostatic tachycardia syndrome and inappropriate sinus tachycardia, are among the most frequent and distinct phenotypes of post-COVID-19 syndrome; one-third of highly symptomatic patients can be affected. • CVAD arises from a malfunction of the autonomic control of the circulation, and can involve failure or inadequate or excessive activation of the sympathetic and parasympathetic components of the autonomic nervous system. • As well as global circulatory disturbances, CVAD in post-COVID-19 syndrome can manifest as microvascular and endothelial dysfunction, with local symptoms such as headache, brain fog, chest pain, dyspnoea and peripheral circulatory symptoms, including skin discolouration, oedema, Raynaud-like phenomena, and heat and cold intolerance. • A structured diagnostic work-up based on a detailed patient history, cardiovascular autonomic testing, long-term electrocardiogram and blood-pressure monitoring, and ancillary cardiac and peripheral vascular tests will lead to an appropriate diagnosis. • Management of CVAD in post-COVID-19 syndrome should involve a correct diagnosis, patient education, and both non-pharmacological and pharmacological methods; a tailored exercise training programme, blood volume expansion and compression garments are especially effective. • Pharmacological approaches target heart rate control, blood volume expansion, promotion of vasoconstriction and venoconstriction, and reduction of hyperadrenergic drive. Introduction In May 2023, the United Nations declared an end to the coronavirus disease 2019 (COVID-19) pandemic as a global health emergency; how- ever, neither COVID-19 nor its long-term effects have gone away. A great many of those who were infected with severe acute respiratory syn- drome coronavirus 2 (SARS-CoV-2) are still living with chronic COVID-19 sequelae. These are early days in our understanding the management, prognosis and mortality associated with these persistent symptoms. The caseload is massive; approximately 10% of patients convalesc- ing from COVID-19 have post-COVID-19 syndrome, also termed long COVID, which equates to around 65 million individuals worldwide 1 Long COVID has been defined by the World Health Organization as persistent symptoms developed and lasting for more than 3 months after the initial SARS-CoV-2 infection, typically presenting as fatigue, shortness of breath, cognitive dysfunction, palpitations and chest pain, without alternative explanation 2 . Cardiovascular sequelae are one of the major aspects of long COVID and can be considered in three main categories: thromboembolic complications, exacerbation of existing cardiovascular disease, and autonomic dysfunction that includes, but is not limited to, abnormal sinus tachycardia, hypotension disorders and microvascular disturbances 3–5 (Box 1). Similar considerations can be given to other body systems; for example, an increased incidence of diabetes mellitus 6,7 , which itself will inevitably be accompanied by cardiovascular complications and autonomic dysfunction 8 Nature Reviews Cardiology Review article triggers the secretion of neuroendocrine catecholamines, vasopressin and renin 49–51 . Disruption of one or more autonomic control circuits can lead to symptoms of impaired circulatory adaptation: orthostatic hypotension, orthostatic tachycardia, abnormal BP fluctuations and persistent systemic hypertension 30,41,45,52 (Fig. 1). Likewise, local perfusion demands are satisfied by adaptive changes in vascular tone and permeability that are also controlled by autonomic and neuroendocrine mechanisms 46 . Inappropriate behav- iour of small vessels is usually termed microvascular dysfunction 53–59 If maladaptation occurs, local and peripheral symptoms can appear, such as headache, chest pain, muscle fatigue, peripheral oedema, Raynaud-like phenomenon, heat and cold intolerance, and paraesthesia 59 (Fig. 1). In this context, coronary microvascular dysfunction is probably the best-studied and defined phenomenon, although it has not been fully explored in the context of post-COVID-19 syndrome 53,58,60 CVAD can also present with paroxysmal events such as recur- rent reflex syncope 61,62 , with cardioinhibitory and/or vasodepressor components 28 . Patients with a low BP phenotype and hypotensive sus- ceptibility might have a predisposition to these episodic events 37,38,63 CVAD can easily be overlooked when the wide spectrum of nonspecific symptoms is not linked with the underlying autonomic dysfunction, nor examined with the appropriate diagnostic modalities. The most prevalent forms of CVAD are listed in Table 1 and illustrated in Figs. 2,3. CVAD in long COVID Epidemiology Among individuals with long COVID, POTS has been reported to be both the most frequent ( ∼ 15–30%) newly diagnosed CVAD (Table 1) and the most frequent previously diagnosed CVAD with major worsening after COVID-19 (refs. 20,22,42,43,64–67) (Table 2). Before the COVID-19 pandemic, the estimated prevalence of POTS in the USA was 0.2–1.0%, and mostly affected young women, possibly with a higher frequency among adolescents with orthostatic complaints 12 . The prevalence is very likely to grow following the pandemic. In some patients, POTS can be associated with initial (immediate or transient) orthostatic hypotension (Table 1), which further worsens the orthostatic symp- tom burden of affected individuals. Initial orthostatic hypotension can also develop as an isolated phenomenon, and studies focusing on postural haemodynamic changes in individuals with long COVID found initial orthostatic hypotension in 61% of these patients, exceeding the frequency of POTS and affecting both sexes equally 65 Individuals with long COVID can also have tachycardia indepen- dently of positional changes 21 . Causes of tachycardia in these indi- viduals can include inappropriate sinus tachycardia (Table 1), which is another form of CVAD, as well as hypoxia, sinus node dysfunction, myocarditis or persistent fever 21 . Inappropriate sinus tachycardia has been estimated to affect 1% of the adult population 52 , but was found in 7% of individuals hospitalized with COVID-19 (ref. 68) and in 12–20% of individuals referred to specialized post-COVID-19 clinics for persistent symptoms after acute SARS-CoV-2 infection 69 (Table 2). Palpitations were among the most frequent complaints in this study population 69 In a Spanish study conducted in a tertiary health-care centre, ~20% of patients with long COVID were diagnosed with inappropriate sinus tachycardia 23 . This cohort consisted mainly of women (85%), with a mean age of 40 ± 10 years. Orthostatic hypotension is found less frequently than sinus tachycardia in patients with long COVID (Table 1). The prevalence of orthostatic hypotension in the general population ranges from 5% in middle-aged cohorts to 43% in octogenarians, and is likely to be influ- enced by the presence of polypharmacy 70 , cardiovascular comorbidi- ties and risk factors such as hypertension and diabetes 71 . Orthostatic hypotension was the third most common newly diagnosed CVAD in individuals with long COVID 64 (Table 2). Some cases of orthostatic hypo- tension might have a neurogenic origin, with autonomic nerve damage resulting in efferent baroreflex failure; however, additional tests are needed to confirm this mechanism 30 . Afferent baroreflex failure is, by contrast, only sporadically reported in individuals with long COVID 72 Furthermore, chronotropic incompetence during physical exercise has been found in up to 30% of individuals with persistent cardiopulmonary symptoms 1 year after acute SARS-CoV-2 infection 73,74 CVAD might be under-diagnosed in patients with long COVID. Many large studies in individuals with long COVID have found frequent non-specific symptoms, such as post-exertional malaise, brain fog, palpitations, fatigue and dizziness in 55–85% of patients, complaints that are characteristic of POTS 75 . Importantly, studies indicate that individuals vaccinated against COVID-19 are less likely to develop long COVID symptoms or post-COVID-19 dysautonomia 76 . Long COVID symptoms can improve over the first 12 months from diagnosis, with a tendency to become permanent if they persist for longer 76 Reflex syncope is the most common type of episodic CVAD (Table 1). This often-benign cause of syncope is the predominant cause of transient loss of consciousness in all age categories and is by far the most common aetiology for syncope in younger individuals 28 . More than one-third of the worldwide population will experience at least one episode of syncope during their lifetime 77,78 , with a biphasic age distribution. The first episode of syncope usually occurs either before Box 1 Autonomic dysfunction Dysautonomia or autonomic dysfunction is a group of disorders that arise from malfunction of the autonomic nervous system (ANS). Dysautonomia can involve either failure or inadequate or excessive activation of sympathetic and parasympathetic components of the ANS. Aetiologies of dysautonomia include neurodegenerative disorders (such as Parkinson disease and pure autonomic failure), chronic diseases (such as diabetes mellitus and renal failure), genetic diseases (such as familial dysautonomia), poisoning, autoimmune diseases (such as autonomic autoimmune gangliopathy, Sjögren disease and systemic lupus erythematosus) and viral infections (as a part of post-COVID-19 syndrome). A group of disorders termed cardiovascular autonomic dysfunction manifest when ANS malfunction primarily affects the circulatory system. Typical cardiovascular dysautonomic conditions observed in post-COVID-19 syndrome are postural orthostatic tachycardia syndrome, inappropriate sinus tachycardia, systemic hypertension, orthostatic hypertension and hypotension, chronotropic incompetence, recurrent reflex syncope, hypotensive susceptibility and circadian rhythm disruption. Importantly, cardiovascular autonomic dysfunction can coexist with structural heart disorders (such as heart failure and obstructive coronary artery disease) and cardiac arrhythmias (such as atrial fibrillation and other forms of supraventricular tachycardia). Nature Reviews Cardiology Review article the age of 30 years or after the age of 65 years 79 . Reflex syncope was the most commonly reported form of CVAD during the acute phase of COVID-19 (ref. 42), with syncope occurring in 5–12% of individuals hospitalized with COVID-19 (ref. 80) (Table 2). Recurrent vasovagal reflex syncope can also develop in individuals with long COVID 64 and was anecdotally associated with a low BP phenotype, hypotensive susceptibility and high BP variability over 24 h 37,38 (Table 1). Finally, case reports and small observational studies have indicated that microvas- cular dysfunction and peripheral circulatory disturbances can be an important part of post-COVID-19 dysautonomia, partly overlapping with POTS 81–83 (Table 1). Aetiology and proposed pathophysiology The aetiology of CVAD after SARS-CoV-2 infection and in patients with long COVID is under investigation. POTS, one of the main manifesta- tions of long COVID, has previously been related to various underlying pathophysiological mechanisms with varying penetrance in inves- tigated cohorts, probably as a consequence of selection bias (for example, cardiology versus neurology as the leading specialty) and variability in access to investigational methods. Commonly proposed aetiologies include abnormally increased sympathetic activity and excess circulating catecholamine levels, peripheral autonomic neuro- pathy with inadequate macrocirculatory and microcirculatory control, hypovolaemia, and impaired vagal activity 12,25,84 . Each of these factors individually or in combination can result in secondary sinus tachy- cardia. The presence of circulating autoantibodies that interfere with cardiovascular signalling via G protein-coupled membrane receptors, such as adrenergic, muscarinic or angiotensin II type 1 receptors has also been postulated 85–87 . However, insufficient reproducibility, limited access to functional cell-based methodology and poor performance of traditional antibody tests such as enzyme-linked immunosorbent assays have hampered development in this field 88 The advent of long COVID has reinvigorated interest in under- standing the mechanisms underlying POTS and other forms of CVAD. Hypothetically, many of the proposed mechanisms of long COVID, such as autoimmune reactive inflammation associated with the release of autoantigens, hyperactivation of mast cells, persistence of viral particles stimulating a continued immune response, micro- vascular and endothelial dysfunction and dysfunctional neurological signalling 1,3,89,90 , could trigger various CVAD-related symptoms. More data are needed to link these mechanisms to circulatory aberrations and their contribution to specific types of CVAD. Interestingly, HR variability has been reported to be either increased 91 or reduced 92 in patients with long COVID. This inconsistency probably reflects the heterogeneity of long COVID and the multiple pathophysiological mechanisms, resulting in the different autonomic phenotypes discussed in this Review. The severity of the acute infec- tion, multiple comorbidities, medication use and prolonged bed rest during the convalescent period could have a role in determining the autonomic profile. Nevertheless, autonomic imbalance with increased sympathetic and reduced parasympathetic activities, specifically POTS, has been observed in other post-viral autonomic disorders 93 The mechanisms by which SARS-CoV-2 infection might affect autonomic pathways is yet to be understood. Studies have shown a neurotropic behaviour of SARS-CoV-2, as the virus can replicate in neuronal cell cultures 94 and viral particles have been found in human Control of heart rate affected • Postural orthostatic tachycardia syndrome • Inappropriate sinus tachycardia • Abnormal exercise-induced or tachypnoea-induced tachycardia • Chronotropic incompetence • Inappropriate sinus bradycardia Control of blood pressure affected • Hypertension • Orthostatic or postprandial hypotension • Orthostatic hypertension • Low blood pressure phenotype or hypotensive susceptibility • Abnormal blood pressure variability including non-dipping or reverse dipping Recurrent re lex syncope • Cardioinhibition and vasodilatation • Headache • Brain fog • Coronary microvascular dysfunction • Chest pain and angina-like symptoms • Dyspnoea • Venous pooling (abdomen, pelvis and lower limbs) • Heat and cold intolerance • Raynaud-like symptoms • Reddish or blue skin discolouration • Fatigue and exercise intolerance • Hypovolaemia General symptoms Local circulatory alterations Global circulatory alterations Microvascular or endothelial dysfunction Fig. 1 | Global and local manifestations of cardiovascular autonomic dysfunction. Two major categories of cardiovascular autonomic dysfunction — global and local circulatory alterations — usually overlap. Global circulatory disorders primarily affect the control of heart rate or blood pressure. Peripheral circulatory disorders are believed to stem from microvascular and endothelial dysfunction and can lead to localized symptoms, such as headache, cognitive impairment (brain fog), (angina-like) chest pain, dyspnoea, heat and cold intolerance, Raynaud-like phenomena, and reddish or blue skin discolouration, and can be associated with venous pooling in the subdiaphragmatic area. Abnormal circulatory control, especially if associated with hypotensive episodes, can lead to recurrent reflex (vasovagal) syncope. A relative and absolute hypovolaemia may be present. Affected individuals characteristically report fatigue, orthostatic intolerance and reduced physical capacity. Nature Reviews Cardiology Review article Table 1 | Typical manifestations of CVAD Type of CVAD Definition Symptoms Diagnosis Postural orthostatic tachycardia syndrome Increase in HR of ≥30 bpm or to >120 bpm on standing, and symptoms of orthostatic intolerance; absence of orthostatic hypotension; symptom duration ≥3 months; no alternative explanation for postural tachycardia Light-headedness, fatigue, palpitations, muscle weakness, headache, brain fog Active stand test; tilt table testing; 24-h ECG monitoring (HR spikes with ‘Manhattan skyline’, morning HR surge) (Figs. 2a,b; 3) Inappropriate sinus tachycardia Resting HR ≥100 bpm or daytime average 24-h ECG ≥90 bpm and no other sinus tachycardia aetiology, accompanied by unexplained fatigue and/or exercise intolerance Palpitations, fatigue, light-headedness, exercise intolerance As above (to differentiate from postural orthostatic tachycardia syndrome); repeated resting ECG, 24-h ECG monitoring (Fig. 2c) Chronotropic incompetence Failure to attain ≥80% of the predicted age-adjusted maximum HR measured during a graded exercise test in the absence of structural heart disease (such as heart failure or coronary artery disease) Exercise intolerance Exercise test; 24-h ECG monitoring Inappropriate sinus bradycardia Resting HR <50 bpm and no other sinus bradycardia aetiology, accompanied by unexplained fatigue and/or exercise intolerance Exercise intolerance, fatigue, presyncope Repeated resting ECG; 24-h ECG monitoring Recurrent vasovagal syncope Recurrent vasovagal reflex episodes at least three times per year and/or unpredicted or associated with severe trauma Repeated syncopal episodes with characteristic triggers Tilt table testing; 24-h ABPM; active stand test; implantable loop recorder in inconclusive cases Hypertension SBP ≥140 mmHg or DBP ≥90 mmHg; or ≥130/80 mmHg on 24-h ABPM Often asymptomatic; headache and vertigo in severe hypertension Office BP; home 24-h ABPM (Fig. 3a) Orthostatic hypotension SBP decrease of ≥20 mmHg or DBP decrease of ≥10 mmHg within 3 min of standing from a supine position Light-headedness on standing, (pre)syncope, fatigue; many patients are asymptomatic under normal conditions Active stand test; tilt table testing; 24-h ABPM Postprandial hypotension SBP decrease of ≥20 mmHg or DBP decrease of ≥10 mmHg within 1 h of a meal Light-headedness after a large meal; worse when combined with standing Measure BP before and at several time-points after a large meal Initial orthostatic hypotension Transient SBP decrease of ≥40 mmHg and/or DBP decrease of ≥20 mmHg that resolves within 30 s immediately after standing and that is associated with symptoms (light-headedness, syncope) Light-headedness immediately on standing, which resolves quickly with ongoing stand Supine-to-stand test with beat-to-beat BP monitoring is obligatory Orthostatic hypertension SBP increase of ≥20 mmHg and to a level >140 mmHg upon standing Usually asymptomatic; sometimes headache, pulsation in head As above Low BP phenotype Office SBP <110 mmHg, ambulatory 24-h SBP <105 mmHg, daytime SBP <115 mmHg, night-time SBP <97 mmHg (men); office SBP <100 mmHg, ambulatory 24-h SBP <98 mmHg, daytime SBP <105 mmHg, night-time SBP <92 mmHg (women) Light-headedness, (pre)syncope; fatigue, exercise and orthostatic intolerance, postprandial symptoms Office BP; home 24-h ABPM (Fig. 3b) Hypotensive susceptibility One or more episodes of daytime SBP <90 mmHg or two or more episodes of daytime SBP <100 mmHg on 24-h ABPM As above 24-h ABPM (Fig. 3b) Excessive BP variability 24-h ABPM daytime SBP standard deviation of repeated measurements ≥15 mmHg As above; can be asymptomatic 24-h ABPM (Fig. 3a) Circadian BP rhythm disruption 24-h ABPM extreme dipping (daytime–night-time SBP >20%), non-dipping (daytime–night-time SBP <10%) or reverse dipping pattern (night-time SBP > daytime SBP) Often asymptomatic; symptomatic if associated with other BP alterations 24-h ABPM (Fig. 3c) Microvascular and endothelial dysfunction Microvascular dysfunction involves alterations to the microcirculation (pre-arterioles, arterioles, capillaries and venules); endothelial dysfunction is a spectrum of phenotypes associated with heterogeneous alterations in endothelial function leading to impaired endothelium-dependent vasodilatation and vasoconstriction and loss of structural integrity of the endothelium Angina-like chest pain, dyspnoea, exercise intolerance, fatigue, migraine-like headache, peripheral oedema, erythromelalgia, heat and cold intolerance, reddish or blue skin discolouration, Raynaud-like symptoms Coronary circulation: absence of obstructive coronary artery disease (<50% diameter reduction or FFR >0.80), and evidence of impaired coronary microvascular function (impaired CFR or increased microvascular resistance) Peripheral circulation: reduced RHI (<1.67), impaired FMD, reduced ACh-induced forearm blood flow (venous occlusion plethysmography); direct and indirect imaging of microvasculature and peripheral blood flow ABPM, ambulatory blood pressure monitoring; ACh, acetylcholine; BP, blood pressure; CFR, coronary flow reserve; CVAD, cardiovascular autonomic dysfunction; DBP, diastolic blood pressure; ECG, electrocardiogram; FFR, fractional flow reserve; FMD, flow-mediated dilatation; HR, heart rate; RHI, reactive hyperaemia index; SBP, systolic blood pressure. Nature Reviews Cardiology Review article brain autopsy samples 95,96 . In addition, earlier studies indicated that viral invasion of the central nervous system occurs via neurotropism, particularly involving the cranial nerves innervating the otorhinolaryn- gopharyngeal mucosa (cranial nerves II, VII, IX and X). These gateways can provide access to neural structures located in the frontal lobe, pons and medulla 96,97 . Human specimens of the pons and medulla oblongata have been found to express very high levels of angiotensin-converting enzyme 2 (ACE2) 98 , the receptor for the SARS-CoV-2 spike protein, and inhibition of ACE2 function decreases vagal tone in animal models 99 Therefore, viral infection and retrograde axonal transport to bulbar centres might theoretically affect important autonomic nuclei, thereby setting the conditions for autonomic imbalance. Autonomic imbalance in patients with long COVID could also alter systemic inflammatory responses through attenuation of the cholinergic inflammatory reflex arc 100 . Persistent inflamma- tion, with elevated plasma levels of IL-1β, IL-6 and tumour necrosis factor, has been described in individuals after SARS-CoV-2 infection 101 This a pro-inflammatory state could put patients at increased risk of microvascular endothelial dysfunction, which is a predictor of cardiovascular morbidity 3 Alternatively, endothelial dysfunction might result directly from autonomic imbalance. A direct interaction exists between sympathetic activity and the vasodilatory function of nitric oxide 58,102 . Patients with POTS diagnosed before the COVID-19 pandemic have been shown to have impaired endothelial function 103 . Removal of sympathetic activ- ity with sympatholytic agents improved the nitric oxide-mediated dilatation of conduit arteries in these patients 103 . Preliminary reports suggest that both coronary and peripheral microvascular dysfunction is present in patients with long COVID and POTS, but more studies involving independent cohorts and different CVAD phenotypes are needed 81,82 . Other possible mechanisms of long COVID-associated CVAD, both on the global circulatory level and in the microcirculation, Fig. 2 | Typical tilt table test and 24-h ECG manifesta- tions of cardiovascular autonomic dysfunction in post-COVID-19 syndrome. a , A woman aged 40 years with post-COVID-19 syndrome examined with tilt table testing. Note the increase in heart rate during the head-up tilt phase (which is characteristic of postural orthostatic tachycardia syndrome (POTS)), associated with light-headedness, fatigue and tingling in the legs. No significant fall in blood pressure was observed. Symptoms resolved after the tilt-down procedure. b , A woman aged 40 years with post-COVID-19 syndrome, palpitations and rapid heartbeats examined with 24-h electrocardiogram (ECG) monitoring. The mean 24-h heart rate is 82 bpm, and the peak heart rate is 136 bpm. Heart rate peaks occur during the daytime (red arrow) and a heart rate surge of >60 bpm occurs in the morning (blue arrow), which are characteristic of POTS. A diagnosis of POTS was later confirmed using tilt table testing. c , A woman aged 52 years with post-COVID-19 syndrome, palpitations and rapid heartbeats examined with 24-h ECG monitoring. The mean 24-h heart rate is 106 bpm, and the peak heart rate is 157 bpm, which is characteristic of inappropriate sinus tachycardia. There is no decrease in heart rate during the night-time (non-dipping pattern) (red arrow). The 24-h SDNN (standard deviation of NN intervals) is 51, which is lower than the normal value of >70, indicating impaired heart rate variability. 200 150 100 50 0 0 20 40 60 80 100 120 140 160 180 200 Tilt up a b c Tilt down Maximum: 150 bpm Maximum: 136 bpm Mean: 82 bpm Maximum: 157 bpm Mean: 106 bpm Baseline: 80 bpm 00:33:20 00:50:00 Heart rate (bpm) Heart rate (bpm) Heart rate (bpm) Day 1 12:00 h Day 1 18:00 h Day 2 00:00 h Day 2 06:00 h 0 20 40 60 80 100 120 140 160 180 200 Day 1 12:00 h Day 1 18:00 h Day 2 00:00 h Day 2 06:00 h Time Time (hh:mm:ss) Time Nature Reviews Cardiology Review article involve low-grade inflammation in the central nervous system paral- leled by impaired cardiovascular autonomic control: inappropriate baroreflex function and increased peripheral resistance 104 , small-fibre neuropathy potentially affecting peripheral autonomic responses 105 , and the presence of fibrin amyloid, microclots and hyperactivated platelets potentially compromising the microcirculation 106 Moreover, beyond the obvious sympathoparasympathetic imbal- ance as the main driver of cardiovascular dysautonomia in long COVID, other neuroendocrine systems involved in the aetiology of CVAD might include the hypothalamic–pituitary–adrenocortical system, the renin– angiotensin–aldosterone system and the arginine–vasopressin sys- tem, but these pathways have been less studied in this context 107 . The increased susceptibility to reflex syncope that is observed in some individuals with long COVID might be provoked by sympathoadrenal imbalance and hyperactivation of the arginine–vasopressin system 51,108 , reflecting an amplified neuroendocrine response to haemodynamic instability and imminent hypotension 37,38 . Finally, symptoms of brain fog, a frequent comorbidity of POTS in patients with long COVID 20 , could be caused by endothelial dysfunction in the cerebral circulation and abnormally reduced cerebral blood flow during orthostasis or exercise 109 Detection and diagnosis Detailed history and physical examination A focused medical history is the first step in the diagnostic work-up for individuals with long COVID and suspected CVAD, and should always take into account the educational and cultural background of the patient, as well as information that the patient might already have gathered from social media 110,111 . The main symptoms and their severity are recorded in the medical history, which should also include information on the temporal relationship between the symptoms and the SARS-CoV-2 infection, as well as whether any symptoms suggest- ing that CVAD was present before COVID-19 and might have worsened after it (Fig. 4). In particular, patients should be asked about symptoms and signs of organ hypoperfusion, such as light-headedness, cognitive impair- ment, blurred vision (indicating retinal hypoperfusion), coat-hanger pain (pain in the shoulders and neck due to muscular hypoperfusion), pallor, chest pain, palpitations, dyspnoea, weakness or fatigue, which might be present on standing and be ameliorated when sitting or lying down 110,111 . Common daily activities, such as meals, heat exposure, Fig. 3 | Typical 24-h ABPM manifestations of cardiovascular autonomic dysfunction in post-COVID-19 syndrome. a , Ambulatory 24-h blood pressure monitoring (ABPM) recording from a woman aged 49 years with post-COVID-19 syndrome, postural orthostatic tachycardia syndrome (POTS) and new-onset hypertension. The mean daytime blood pressure is 146/85 mmHg, and the mean heart rate is 78 bpm. The difference in mean systolic blood pressure between day and night is 12.3% (normal values 10–20%). Characteristic peaks in systolic blood pressure to >160 mmHg (dashed line) can be observed. One hypotensive episode (76/50 mmHg) is marked with an arrow, most likely after a lunchtime meal at 1300 h. b , 24-h ABPM recording from a woman aged 27 years with post- COVID-19 syndrome, POTS and recurrent syncope. The mean blood pressure is 103/70 mmHg, and the mean heart rate is 77 bpm. The difference in mean sys- tolic blood pressure between day and night is 13.7%. Characteristic drops in systolic blood pressure to <90 mmHg can be observed (arrows). c , 24-h ABPM recording from a man aged 51 years with post-COVID-19 syndrome and POTS. The difference in mean systolic blood pressure between day and night is 6.6%. Characteristic abnormal blood pressure and heart rate fluctuations can be observed. a b c 0 10 10 12 14 16 18 20 22 00 02 04 06 08 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 20 30 40 50 60 70 80 90 100 110 120 130 140 150 0 14 16 18 20 22 00 02 04 06 08 10 12 16 18 20 22 00 02 04 06 08 10 12 14 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 Day Night Day Day Night Day Day Night Blood pressure (mmHg) Heart rate (bpm) Blood pressure (mmHg) Heart rate (bpm) Blood pressure (mmHg) Heart rate (bpm) Time (h) Time (h) Time (h) Nature Reviews Cardiology Review article alcohol intake or light physical exercise, might worsen the symptoms of CVAD. These effects should be documented as specific triggers and guide further autonomic function testing. Individuals should also be questioned about syncope episodes. Documenting the conditions under which loss of consciousness occurred, any prodromal symptoms, the phenomena observed by witnesses during the loss of consciousness and the recovery time can help physicians distinguish between syncope and other possible causes of transient loss of consciousness 28 , such as epileptic seizures and psychogenic pseudosyncope 64 . The semiology of syncope also guides physicians in understanding the possible mechanisms underlying the syncope event (for example, vasovagal reflex, orthostatic hypoten- sion or cardiac causes of syncope), each of which requires a targeted diagnostic work-up 28 Impairment of other autonomic domains can frequently accom- pany CVAD. The medical history should include any changes in body sweat; bladder, bowel and sexual function; and heat or light toler- ance. The Composite Autonomic Symptom Score-31 (COMPASS-31) is a useful questionnaire for the accurate screening of multiple auto- nomic complaints, with validated translations in several languages 112 , although its utility in post-COVID-19-associated CVAD has not been fully studied. Other questionnaires, such as the Orthostatic Hypoten- sion Questionnaire 113 and the Malmö POTS symptom score 114 target symptoms of CVAD more specifically in individuals diagnosed with orthostatic hypotension or POTS, respectively, and can be used for monitoring the symptomatic course over time (Fig. 4). Individuals with long COVID commonly have accompanying dis- turbances such as headache, anosmia, ageusia, cognitive dysfunction, sleep–wake disturbances or sequelae of acute COVID-19 complications, such as encephalopathy, stroke and meningoencephalitis 80,115 . These disturbances can persist over months 116 and are major determinants of an individual’s perception of their illness 117 . They should, therefore, be documented and treated in a targeted way as part of a holistic approach, by taking into account not only the present illness but also past medi- cal conditions and carefully reviewing any prescribed medications for effects on cardiovascular autonomic function, especially in older individuals 70 Physical examination should focus on the heart and lungs and screen for additional signs that might indicate widespread forms of autonomic failure. These signs include pupillary abnormalities, venous pooling, cold violaceous hands or feet 118 , marbled skin 119 or joint hypermobility 119 (Fig. 4). If patients report tingling sensations in the hands or feet, a neurological examination should be performed to document any additional signs of peripheral neuropathy 120 . A referral to a neurologist can be considered at this stage. Initial testing First, 12-lead electrocardiography (ECG) should be performed to screen for resting tachycardia, heart rhythm, repolarization disturbances or indirect signs of cardiac hypertrophy 28,52 . If there is a suspicion of under- lying structural heart disease, cardiac imaging should be performed (such as echocardiography, cardiac MRI, coronary CT angiography or invasive coronary angiography with assessment of coronary micro- vascular function, if appropriate) 28 (Fig. 4). Basic laboratory tests are indicated at the initial diagnostic work-up. These tests should include indices of kidney and liver function, blood glucose levels, haemoglobin A1c levels, electrolyte levels, blood cell count and thyroid hormone levels to exclude conditions that might predispose to or aggravate CVAD. Further testing should be driven by clinical suspicion and could include measurement of autoimmune and neuroendocrine biomarker levels (Fig. 4). Orthostatic vital signs Office BP measurements are usually taken with the individual in the s