Original AnalysisDysautonomia Research Is Measuring the Wrong Thing Under the Wrong Conditions

The largest drug trial conducted in POTS ended by separating the measurement from the illness. RECOVER-AUTONOMIC randomized 181 adults with post-COVID POTS to ivabradine or placebo, using a drug that lowers heart rate by selective inhibition of the sinoatrial node without altering blood pressure or contractility, and the heart rate duly fell. In the trial's own summary, ivabradine "did produce a significant reduction in heart rate compared to placebo, but the lower heart rate did not improve POTS symptoms" (RECOVER, 2026).

A systematic review published the previous year explains why. Pooling 17 cohorts and 1,321 patients, it found stroke volume reduced in 70% of studies but cardiac output reduced in only 10%, concluding that cardiac output is generally maintained or increased in POTS and is therefore unlikely to cause symptoms (Hogwood et al., 2025). The heart is delivering, the tachycardia is the means of delivery, and removing it removes a compensation rather than a cause.

The trial itself wasn't designed badly. Its primary endpoint was a symptom questionnaire and heart rate was secondary, so it was built from the outset to be judged on how patients felt. It also carried a second randomization in a factorial design, assigning participants to coordinated nonpharmacologic care, meaning volume expansion, an abdominal binder, exercise rehabilitation, education and a care coordinator, or to usual care (Fudim et al., 2026). Among participants receiving ivabradine, those given coordinated care reported symptom improvement over those given usual care (RECOVER, 2026).

The second finding needs reporting without being built on. It was presented at a conference session with the full report still pending, the results for each of the 4 cells haven't been released, and 181 participants divided 4 ways leaves roughly 45 in each. Nothing in what is public establishes how much of that improvement belongs to the drug and how much to the program delivered alongside it. What survives without qualification is the main effect, which is that the drug on its own changed heart rate and changed nothing a patient noticed. What failed alongside it was the inference that had accumulated around the measurement, in which a threshold adopted to define the syndrome had been carried forward into a target worth treating. The fault therefore sits distal to the heart, in the vasculature that decides where the output goes.

A second and more consequential error runs alongside that one, because it doesn't merely waste evidence, it manufactures it. In 1998, investigators examined whether neck afferents contribute to cardiovascular control in humans by rotating subjects' heads and sustaining the position for 3 minutes. Sympathetic traffic didn't change (Ray and Hume, 1998). A review 2 years later carried the finding forward as the conclusion that human studies don't indicate that low threshold mechanoreceptors in the neck modulate cardiovascular responses (Bolton and Ray, 2000). In 2014, an author of that review tested the same pathway again, stretching the neck muscles sinusoidally rather than holding them in position, and recorded a 45.4% modulation of sympathetic outflow to the lower limbs (Bolton, Hammam and Macefield, 2014). The earlier experiment hadn't disproved the pathway, it had only immobilized the variable the pathway responds to.

The same structure recurs throughout this literature, wherever a supine baseline, a sustained hold, resting heart rate variability in asymptomatic volunteers, a single static reading: each is reproducible, each is standardizable, and none of them can see a control system fail, because a control system only fails under load. Conditions selected for stability can't detect a failure that only appears under provocation, and a null produced under those conditions records the limits of the method rather than the absence of the mechanism.

Both errors resolve into a single question about what is measured and under what conditions. This review examines what governs blood distribution during orthostasis, which sensory pathways feed it, what stimulus parameters determine their effect, and how much of what is currently taken as settled rests on measurements taken at rest. Much of the trial evidence below is drawn from cohorts recruited under the POTS label, because that is where trials have been funded. The argument concerns orthostatic intolerance and autonomic control generally, and one of the findings examined here is that the label itself is a poor boundary around the disorder.

Blood distribution is set by nerve traffic

Standing displaces roughly 500 to 800 ml of blood into the vessels below the diaphragm. Maintaining cerebral perfusion depends on constricting those vessels quickly enough to preserve central filling, and that constriction is applied by sympathetic vasoconstrictor fibers innervating specific vascular beds. The regulated quantity is the distribution of blood across compartments rather than its total.

The orthostatic intolerance literature has been describing this for 2 decades. Stewart's mechanistic framework separates orthostatic intolerance arising from regional blood volume redistribution from that arising from any deficit in total volume (Stewart, 2012). The same group characterized normal-flow POTS, in which patients have adequate blood volume and adequate cardiac output yet pool in the splanchnic bed through excessive nitric oxide synthase activity and splanchnic vasodilation (Stewart et al., 2011), and distinguished low-flow from normal-flow subgroups by measuring regional volumes and flows directly, finding uncoupling of baroreflex heart rate regulation in the low-flow group (Stewart et al., 2007). The problem is where blood goes rather than how much of it there is.

Regional vasoconstrictor tone isn't a quantity that can be topped up. It's nerve traffic, computed by brainstem nuclei from sensory information about posture, pressure and motion, and it has a direct unit of measurement. Muscle sympathetic nerve activity, recorded by microelectrode from a peripheral nerve, is the outgoing constrictor instruction to the vascular bed that nerve supplies. Any intervention that changes where blood sits during orthostasis must either substitute mechanically for that instruction or change the instruction itself.

Compression is usually described as a mechanical substitute, and what was actually measured is narrower than that. Applied across 4 conditions in a randomized crossover, it lowered heart rate in proportion to how much was used, from 109 beats per minute uncompressed to 92 with full abdominal and leg compression, with symptoms improving along the same gradient (Bourne et al., 2021). That trial recorded stroke volume rising without recording why, and passive reduction of venous capacity and a reflex venoconstriction driven by the garment would produce the same curve. Sustained pressure on skin is itself an autonomic input, and the point has been demonstrated with the relevant control: gentle mechanical stimulation of the skin inhibits the somatocardiac sympathetic C-reflex by up to 40%, an inhibition that outlasts the contact by 15 minutes and disappears when the cutaneous nerves supplying the stimulated area are cut (Hotta et al., 2010). Compression garments also improve measured proprioception and balance in hypermobile patients over 2 years of use (Benistan et al., 2023). A garment worn against the skin loads mechanoreceptors continuously, and no trial in this population has separated that from the pressure it applies to the vessels underneath. Sodium loading raises plasma volume and lowers standing norepinephrine (Garland et al., 2021), although a subset of patients can't retain sodium and intravenous fluid affected neither hemodynamics nor peak oxygen uptake in the pooled analysis (Hogwood et al., 2025). Exercise reconditioning carries the strongest outcome figures, with 71% of 103 completers no longer meeting POTS criteria (George et al., 2016), though the same pooled review found peak oxygen uptake reduced in 80% of studies and in only 30% once matched for deconditioning, locating most of that deficit in inactivity rather than in the syndrome. These are systemic interventions acting on load, and none of them targets the constrictor instruction itself.

One intervention does target it, since transcutaneous vagus nerve stimulation delivers current to the auricular branch of a predominantly afferent nerve projecting to the nucleus tractus solitarius, and 2 trials have applied it in POTS. They are worth separating, because they recruited on different principles and the difference determines what each can support.

The first enrolled 26 patients meeting the POTS label and delivered stimulation for 1 hour daily over 2 months against sham, reducing the postural heart rate increase to 17.6 beats per minute from 31.7, alongside lower antiadrenergic autoantibodies and inflammatory cytokines (Stavrakis et al., 2024). It's a positive randomized result and it establishes little about mechanism, because the cohort was defined by a heart rate threshold that groups together patients with unrelated drivers. A clinician reading it can't determine which patient in front of them it describes.

The second recruited on mechanism instead, enrolling 22 patients with hyperadrenergic POTS specifically, and recorded sympathetic traffic directly rather than inferring it. Muscle sympathetic nerve activity fell both supine and tilted, baroreflex sensitivity rose during tilt, and part of the effect outlasted the stimulation period (Shiffer et al., 2026). It was uncontrolled and small, and it's the only one of the 2 that can state what moved.

Peripheral sensory stimulation altered the constrictor instruction, and the alteration persisted. The pair also sets the constraint on everything downstream, since enrollment by diagnostic label produces results that can't be assigned to a patient, and only measurement of the mechanism can do that.

The afferent pathways that set constrictor tone

If sensory input can move sympathetic outflow, the relevant question is which inputs, and the anatomy is established.

The vestibulosympathetic reflex

Across 30 years, animal and human work has, in the words of the field's own review, conclusively demonstrated that input from the vestibular otolith organs contributes to blood pressure control during movement and changes in posture, running through the caudal vestibular nucleus complex, interneurons of the lateral medullary reticular formation, and bulbospinal neurons of the rostral ventrolateral medulla. The reflex is anatomically separate from the baroreflex and faster, because it can be elicited before blood has finished redistributing (Yates, Bolton and Macefield, 2014).

Timing makes it a feed-forward channel, one that signals a postural change while the change is occurring rather than reporting the hemodynamic consequence afterward. Degraded feed-forward input produces a compensation that arrives late and overshoots.

Cervical proprioceptive input

The otoliths report head motion in space and carry no information about head position relative to the trunk. Neck muscle spindles supply that coordinate, which is why both are required to determine whether a given movement demands vasoconstriction in the legs. Recording muscle sympathetic nerve activity from the peroneal nerve in 10 supine subjects, with the head fixed in space and the body displaced beneath it to isolate neck input from vestibular input, sinusoidal neck muscle stretch modulated sympathetic outflow to the lower limbs with a modulation index of 45.4% (Bolton, Hammam and Macefield, 2014).

The measured output there was sympathetic constrictor traffic to the vessels of the lower limb, the same variable that decides whether blood pools below the diaphragm on standing or returns to the chest. The cervical spine is wired to the quantity that is disordered in orthostatic intolerance.

Somato-autonomic reflexes

Beneath both sits the general case, in which mechanical stimulation of somatic tissue modulates visceral function through reflexes mapped in detail more than 50 years ago and catalogued since across several hundred studies, with effects that are general for some organs and strictly segmental for others (Sato and Schmidt, 1973) (Sato, 1997). The afferent limb of those reflexes is the muscle spindle, joint and cutaneous mechanoreceptor population of the somatic tissues themselves, which means that any mechanical contact applied to paraspinal tissue with sufficient force and speed is an input to this system whether or not it's administered with that intention.

This body of work is anesthetized-animal electrophysiology using electrical stimulation. It establishes the architecture rather than the magnitude of response to any clinical contact. What it does establish is that the somatic surface is a route into autonomic control, and that the receptors carrying that route are the same class of mechanoreceptor that manual and manipulative techniques are designed to load.

Stimulus parameters determine the autonomic response

Sensory input to these pathways doesn't behave as a switch, and across every preparation in which anyone has varied the delivery, the autonomic response has proved dependent on timing, frequency and repetition, to the point that the same stimulus at different settings produces opposite results.

The clearest demonstration is a randomized trial that varied nothing but schedule. In it, 96 susceptible subjects received optokinetic drum-chair training in 1 of 2 forms: short and intense, 1 session daily for 3 days each terminated at the first report of pronounced nausea, or stepwise, 1 session daily for 6 days at 60, then 120, then 180 seconds. The graded arm produced the larger symptom reduction, a median change of 9.0 points against 5.0, and the 2 arms produced different autonomic signatures. The graded protocol reduced both low and high frequency power and raised the ratio between them, the intensive protocol reduced only high frequency power, and skin temperature fell by 2 to 4 degrees under the intensive protocol against under 1 degree in the graded one (Zhang et al., 2026). Scheduling alone determined both the clinical and the autonomic result, and the arm driven to tolerance performed worse on both.

Vestibular stimulation shows the dependence as outright failure at the wrong setting. Brief galvanic stimulation at 2 mA for 1 second, sufficient to drive postural and oculomotor responses, produced no change in muscle sympathetic nerve activity. Continuous sinusoidal stimulation over 60 to 100 cycles raised total sympathetic traffic by 156% and generated bursts coupled to the input, an influence the authors described as potent and capable of operating independently of the baroreceptor system (Bent, Bolton and Macefield, 2006). Frequency proved to matter as much as continuity, with modulation inversely related to stimulation frequency across a 100-fold range and strongest at 0.05 Hz (Singh, Hammam and Macefield, 2019).

Neck input shows the dependence as a dissociation between movement and position. In the protocol that produced 45.4% modulation under sinusoidal stretch, a static ramp-and-hold displacement of 17.5 degrees sustained for 54 seconds produced no significant change (Bolton, Hammam and Macefield, 2014), and the same laboratory reported the identical split in sympathetic supply to skin (Bolton, Hammam and Macefield, 2018). The 1998 study contained the same signal in its own data, since head-down neck flexion in those subjects raised sympathetic traffic by 79% while sustained rotation raised nothing (Ray and Hume, 1998). The condition that moved and the condition that didn't were reported side by side, and the negative one became the citation.

Mechanical input to the spine follows the same arithmetic and has been quantified most precisely. Recording paraspinal muscle spindle discharge during controlled thrusts across durations from 12.5 to 400 milliseconds, discharge rose nonlinearly as duration fell, with the inflection near the duration used clinically (Pickar et al., 2007). Beyond that window the sign reverses, with mechanically assisted thrusts under 10 milliseconds decreasing spindle discharge and most afferents requiring over 6 seconds to return to baseline (Reed et al., 2017).

Optokinetic training, galvanic vestibular stimulation, neck muscle stretch, spinal thrust and vagal stimulation are separate literatures examining different tissue, and each independently identifies timing, frequency, repetition and progression as the governing variables. Any stimulus delivered once, held static, or driven to tolerance is being applied under the conditions in which its effect is smallest or reversed.

Why the cervical and vestibular channels are implicated in dysautonomia

An existing pathway isn't grounds for suspicion. What implicates these channels in dysautonomia is the antecedent history and the comorbidity pattern.

Concussion has been reviewed systematically against autonomic function, and across 36 qualifying studies only 3 failed to identify anomalies, supporting the graded conclusion that concussion likely causes autonomic disturbance (Pertab et al., 2018). Surgery, accident and head injury together form the largest non-infectious trigger cluster in population surveys of POTS (Shaw et al., 2019).

Whiplash supplies a mechanism for persistence, and it's self-reinforcing. Sympathetic activity modulates the sensitivity of muscle spindles to stretch, so elevated sympathetic tone degrades the quality of position signal from the muscles it supplies (Passatore and Roatta, 2006). Degraded cervical proprioception produces an inappropriate autonomic response, and the resulting sympathetic tone further degrades the signal. The hypermobile population arrives at the same position without injury, since tissue that deforms more under load reports position less reliably, and proprioceptive impairment accompanies autonomic dysfunction in hypermobility, characterized across 270 patients on cerebrovascular, autonomic and neuropathic measures simultaneously (Novak et al., 2025).

Available means of loading these pathways

Vestibular and oculomotor rehabilitation, gaze stabilization, habituation protocols and cervical manual therapy all deliver graded mechanical and positional input to these channels, and the bedside examinations used to decide which pathway to load are themselves long-established tests originating with physicians in mainstream medicine. Multidisciplinary concussion programs already pair oculomotor and vestibular training with autonomic regulation as named treatment domains (Hawash-Kuemmerle et al., 2026), and the association runs in both directions, with autonomic state indexed by pupil light reflex predicting performance on visual, vestibular and somatosensory testing in 90 adults (McGrath et al., 2026).

Spinal manipulation as somatosensory input

Whatever else a manipulation is understood to be, mechanically it's a brief high-velocity load applied to paraspinal tissue, which places it in the somato-autonomic afferent pathway described above and, when applied to the cervical spine, in the neck proprioceptive channel that reaches the vestibular nuclei.

The mechanism side hasn't been disputed for 20 years. Muscle spindle and Golgi tendon organ afferents are stimulated by spinal manipulation, and the biomechanical event has physiological consequences through its effect on sensory inflow to the central nervous system (Pickar, 2002). A single cervical session measurably alters cortical sensorimotor processing, reducing parietal N20 and frontal N30 evoked potential amplitudes for approximately 20 minutes in 12 subjects, with a separate passive head movement group showing no change (Haavik-Taylor and Murphy, 2007).

Autonomic outcomes are a separate matter and the evidence there is unfavorable. The best-powered sham-controlled test randomized 95 healthy young adults to seated cervical manipulation or sham and found no significant change in any frequency domain measure of heart rate variability in either group, with plasma norepinephrine declining in both (Budgell, Injeyan and Teodorczyk-Injeyan, 2023). An earlier crossover trial in 28 adults reporting that thoracic manipulation raised the ratio of low to high frequency power didn't survive that replication (Budgell and Polus, 2006). Both originate with the same investigator, who has separately documented the blinding failures that make heart rate variability an unreliable endpoint in manual therapy trials (Injeyan and Budgell, 2022).

The design of that null deserves the same scrutiny applied to the 1998 study. The endpoint was resting heart rate variability, recorded without orthostatic or positional challenge. A stimulus proposed to alter vasoconstrictor responses to postural load was assessed while no postural load was applied. The result stands as reported, and what it can be taken to have excluded is narrower than the way it's usually cited.

Where systematic review does identify a signal, it falls where the parameter data predict. Across 29 trials drawn from 2,267 screened, oscillatory mobilization probably produces a short-term increase in skin sympathetic nerve activity regardless of region treated, rated moderate certainty, while sustained apophyseal glides and single-thrust manipulation show no acute effect or conflicting results at very low to low certainty (Picchiottino et al., 2019) (Araujo et al., 2019). The single moderate-certainty autonomic finding in this literature attaches to the rhythmic repeated technique, which is the delivery form the parameter data would have predicted.

Static structural measures fare worse still, since correlations between cervical alignment, forward head posture and skin sympathetic response derive from cross-sectional and case-control designs concentrated in a single research network (Moustafa et al., 2020), and static position is precisely the variable microneurography found doesn't modulate sympathetic outflow.

The systematic review most often cited as settling this question assessed spinal manipulation as a treatment for infantile colic, childhood asthma, hypertension, dysmenorrhea and migraine, and excluded by design all non-clinical studies of physiological mechanism together with animal work, removing every study cited in this section (Côté et al., 2021). On safety, a population study of 818 vertebrobasilar strokes found patients under 45 approximately 3 times more likely to have consulted a chiropractor beforehand and equally more likely to have consulted a primary care physician, consistent with care-seeking for the neck pain of a dissection already in progress (Cassidy et al., 2008). That undercuts causation without establishing safety, and consent covering arterial dissection remains the defensible standard.

Cerebral perfusion is the endpoint that tracks symptoms

Distribution matters because of what reaches the brain, and Orthostatic intolerance tracks cerebral perfusion more closely than heart rate, and cerebral blood flow can fall before heart rate rises. Normal vitals don't establish normal brain blood flow.

The threshold that defines the syndrome can be examined against that endpoint directly. A retrospective series compared 127 patients with hypocapnic cerebral hypoperfusion, meaning orthostatic intolerance without orthostatic tachycardia, against 125 patients with POTS and 42 healthy controls, using tilt testing with transcranial Doppler, capnography, sudomotor testing, skin biopsy and invasive cardiopulmonary exercise testing. The 2 patient groups didn't differ in symptom duration, demographics, comorbidities, autonomic and sensory complaints, supine or standing norepinephrine, inflammatory markers, or treatment. Both showed reduced orthostatic cerebral blood flow velocity, reduced end-tidal CO2, preload failure, mild autonomic failure and small fiber neuropathy of similar degree and distribution. The authors reported the groups as indistinguishable "except for tachycardia in POTS," concluded that reduced orthostatic cerebral blood flow is the unifying feature of both, and proposed that the tachycardia itself reflects central nervous system overcompensation for the orthostatic challenge (Novak et al., 2024).

Which leaves 127 patients with the same physiology, the same perfusion deficit and the same peripheral pathology falling outside the diagnostic criterion because a single number stayed below a cutoff. Orthostatic intolerance occurs without tachycardia, which places the same measurement at the center of both errors described at the outset. It's not the fault, and the threshold drawn across it excludes patients who have the disorder it was written to identify.

This sets the measurement standard for any future work. Resting heart rate variability in asymptomatic subjects isn't an orthostatic endpoint. A stimulus aimed at regional vasoconstrictor control should be assessed against regional vasoconstriction, postural hemodynamics, or cerebral blood flow velocity under challenge.

What hasn't been tested

The disordered quantity in these syndromes is regional vasoconstrictor tone, which brainstem nuclei set from sensory information. Two of the pathways feeding those nuclei, the vestibular apparatus and the cervical proprioceptors, have been shown by direct nerve recording to modulate sympathetic traffic to the lower limbs in humans. The stimulus parameters governing that modulation have been characterized across 5 independent literatures. No clinical trial has been directed at any of it in this population.

The required study is specifiable in full: dynamic graded vestibular and cervical loading, progressed across weeks rather than applied once, in diagnosed POTS patients rather than asymptomatic volunteers, assessed against postural hemodynamics or cerebral blood flow velocity rather than resting variability, conducted by an unaffiliated group. Each specification derives from a null result already on record.

The objection that hands-on intervention can't be blinded has already been addressed by the field. The RECOVER-AUTONOMIC protocol states that participants and investigators are blinded to the drug but not to the non-pharmacologic intervention, because blinding the latter isn't possible, and the trial randomized it regardless and prespecified the analysis (Fudim et al., 2026). Rigor derived from randomization rather than masking, and the unmasked arm moved the symptom endpoint.

Sparse randomized evidence is also not specific to manual approaches. The 2 investigators behind the exercise protocol with the strongest outcome figures in POTS closed their own review of non-pharmacological treatment by noting that randomized trials are needed to evaluate the efficacies of those treatments (Fu and Levine, 2018).

The way this absence is usually reported compounds it. Summaries of what is known in this area are assembled from the trials that exist, and the trials that exist were conducted under the conditions described throughout this review. A systematic review that restricts itself to randomized clinical trials of named disease outcomes, and excludes physiological and mechanistic studies by protocol, will retrieve a body of nulls generated in asymptomatic subjects at rest and will correctly report that it found nothing (Côté et al., 2021). The finding is accurate as a statement about the retrieved literature. It's not a statement about the physiology, and the distinction between those 2 things is the whole of what this review has argued.

Implications for practice

The physiology supports a form of delivery rather than a specific therapy. It favors input targeted at an identified pathway, delivered dynamically, graded across a course of care, and assessed against an objective orthostatic measurement. It argues against static correction, single application, and provocation to tolerance, the last of which performed worse than graded loading on both clinical and autonomic outcomes in the only trial to compare them directly.

It also relocates the disagreement, because the division that determines whether a clinician is measuring the relevant variable doesn't run between specialties or credentials. It runs between assessment performed at rest and assessment performed under the load that provokes the failure. A clinician who examines vestibular, oculomotor and cervical function under positional and orthostatic challenge, and who tracks cerebral perfusion and ventilation while doing so, is measuring the quantity this review has described. A clinician who records a supine baseline, a static posture and a resting rhythm strip is measuring stability, and stability is the state in which this disorder doesn't appear.

The standard is available to anyone prepared to apply it, and it's also the standard by which the existing evidence should be read. Every null reviewed here was produced under conditions selected against the mechanism being tested. Every positive result reviewed here came from a protocol that loaded the system and recorded what it did. The literature isn't ambiguous about which sensory pathways reach vasoconstrictor control. It's ambiguous about what happens when a clinician loads them deliberately in a patient who is unwell, and it will remain ambiguous until somebody measures it under conditions capable of showing an answer.

References

1. Araujo FX, Ferreira GE, Angellos RF, Stieven FF, Plentz RDM, Silva MF. Autonomic effects of spinal manipulative therapy: systematic review of randomized controlled trials. Journal of Manipulative and Physiological Therapeutics. 2019;42(8):623-634. doi:10.1016/j.jmpt.2018.12.005
2. Benistan K, Pontier B, Leblond C, Flageul O, Le Guicher G, Enjalbert M, Gillas F. The effectiveness of compression garments for reducing pain in non-vascular Ehlers-Danlos syndromes: a prospective observational cohort study. Healthcare. 2023;11(13):1862. doi:10.3390/healthcare11131862
3. Bent LR, Bolton PS, Macefield VG. Modulation of muscle sympathetic bursts by sinusoidal galvanic vestibular stimulation in human subjects. Experimental Brain Research. 2006;174(4):701-711. doi:10.1007/s00221-006-0515-6
4. Bolton PS, Hammam E, Macefield VG. Neck proprioceptors contribute to the modulation of muscle sympathetic nerve activity to the lower limbs of humans. Experimental Brain Research. 2014;232(7):2263-2271. doi:10.1007/s00221-014-3917-x
5. Bolton PS, Hammam E, Macefield VG. Neck movement but not neck position modulates skin sympathetic nerve activity supplying the lower limbs of humans. Journal of Neurophysiology. 2018;119(4):1283-1290. doi:10.1152/jn.00043.2017
6. Bolton PS, Ray CA. Neck afferent involvement in cardiovascular control during movement. Brain Research Bulletin. 2000;53(1):45-49. doi:10.1016/s0361-9230(00)00307-5
7. Bourne KM, Sheldon RS, Hall J, Lloyd M, Kogut K, Sheikh N, Jorge J, Ng J, et al. Compression garment reduces orthostatic tachycardia and symptoms in patients with postural orthostatic tachycardia syndrome. Journal of the American College of Cardiology. 2021;77(3):285-296. doi:10.1016/j.jacc.2020.11.040
8. Budgell B, Polus B. The effects of thoracic manipulation on heart rate variability: a controlled crossover trial. Journal of Manipulative and Physiological Therapeutics. 2006;29(8):603-610. doi:10.1016/j.jmpt.2006.08.011
9. Budgell BS, Injeyan HS, Teodorczyk-Injeyan J. Effect of seated cervical spinal manipulation on autonomic nervous system activity as measured by heart rate variability and plasma norepinephrine levels: a randomized pre- and poststudy. Journal of Manipulative and Physiological Therapeutics. 2023;46(4):220-228. doi:10.1016/j.jmpt.2024.02.003
10. Cassidy JD, Boyle E, Côté P, He Y, Hogg-Johnson S, Silver FL, Bondy SJ. Risk of vertebrobasilar stroke and chiropractic care: results of a population-based case-control and case-crossover study. Spine. 2008;33(4 Suppl):S176-S183. doi:10.1097/BRS.0b013e3181644600
11. Côté P, Hartvigsen J, Axén I, Leboeuf-Yde C, Corso M, Shearer H, Wong J, Marchand AA, et al. The global summit on the efficacy and effectiveness of spinal manipulative therapy for the prevention and treatment of non-musculoskeletal disorders: a systematic review of the literature. Chiropractic and Manual Therapies. 2021;29:8. doi:10.1186/s12998-021-00362-9
12. Fu Q, Levine BD. Exercise and non-pharmacological treatment of POTS. Autonomic Neuroscience. 2018;215:20-27. doi:10.1016/j.autneu.2018.07.001
13. Fudim M, Novak P, Taub PR, Chung T, Zimmerman KO, Moy OV, Fissler HZ, Wen J, et al. Design and rationale of RECOVER-AUTONOMIC: a randomized platform trial evaluating interventions for Long COVID postural orthostatic tachycardia syndrome. American Heart Journal. 2026;296:107384. doi:10.1016/j.ahj.2026.107384
14. Garland EM, Gamboa A, Nwazue VC, Celedonio JE, Paranjape SY, Black BK, Okamoto LE, Shibao CA, et al. Effect of high dietary sodium intake in patients with postural tachycardia syndrome. Journal of the American College of Cardiology. 2021;77(17):2174-2184. doi:10.1016/j.jacc.2021.03.005
15. George SA, Bivens TB, Howden EJ, Saleem Y, Galbreath MM, Hendrickson D, Fu Q, Levine BD. The international POTS registry: evaluating the efficacy of an exercise training intervention in a community setting. Heart Rhythm. 2016;13(4):943-950. doi:10.1016/j.hrthm.2015.12.012
16. Haavik-Taylor H, Murphy B. Cervical spine manipulation alters sensorimotor integration: a somatosensory evoked potential study. Clinical Neurophysiology. 2007;118(2):391-402. doi:10.1016/j.clinph.2006.09.014
17. Hawash-Kuemmerle K, Brodsky JR, Molind H, Mysak K, O'Brien M, Pidgeon J, Raghuram A, Shah AS. A multidisciplinary, symptom-targeted approach to rehabilitation in pediatric concussion. Seminars in Pediatric Neurology. 2026;58:101292. doi:10.1016/j.spen.2026.101292
18. Hogwood AC, Abbate G, Thomas G, Markley R, Priday AG, Arena R, Abbate A, Canada JM. Determinants of exercise intolerance in postural orthostatic tachycardia syndrome: a systematic review. Exercise, Sport, and Movement. 2025;3(4). doi:10.1249/esm.0000000000000055
19. Hotta H, Schmidt RF, Uchida S, Watanabe N. Gentle mechanical skin stimulation inhibits the somatocardiac sympathetic C-reflex elicited by excitation of unmyelinated C-afferent fibers. European Journal of Pain. 2010;14(8):806-813. doi:10.1016/j.ejpain.2010.02.009
20. Injeyan HS, Budgell BS. Mitigating bias in the measurement of heart rate variability in physiological studies of spinal manipulation: a comparison between authentic and sham manipulation. Journal of Manipulative and Physiological Therapeutics. 2022;45(2):104-113. doi:10.1016/j.jmpt.2022.03.019
21. McGrath M, Bissett B, Fearon A, Witchalls J, Waddington G. Do measures of autonomic nervous system function correlate with sensorimotor performance? Perceptual and Motor Skills. 2026. doi:10.1177/00315125261459736
22. Moustafa IM, Youssef A, Ahbouch A, Tamim M, Harrison DE. Is forward head posture relevant to autonomic nervous system function and cervical sensorimotor control? Cross sectional study. Gait and Posture. 2020;77:29-35. doi:10.1016/j.gaitpost.2020.01.004
23. Novak P, Systrom DM, Witte A, Marciano SP. Orthostatic intolerance with tachycardia (postural tachycardia syndrome) and without (hypocapnic cerebral hypoperfusion) represent a spectrum of the same disorder. Frontiers in Neurology. 2024;15:1476918. doi:10.3389/fneur.2024.1476918
24. Novak P, Systrom DM, Marciano SP, Witte A, Warren A, Felsenstein D, Giannetti MP, Hamilton MJ, et al. Hypermobile Ehlers-Danlos syndrome: cerebrovascular, autonomic and neuropathic features. American Journal of Medicine Open. 2025;14:100111. doi:10.1016/j.ajmo.2025.100111
25. Passatore M, Roatta S. Influence of sympathetic nervous system on sensorimotor function: whiplash associated disorders (WAD) as a model. European Journal of Applied Physiology. 2006;98(5):423-449. doi:10.1007/s00421-006-0312-8
26. Pertab JL, Merkley TL, Cramond AJ, Cramond K, Paxton H, Wu T. Concussion and the autonomic nervous system: an introduction to the field and the results of a systematic review. NeuroRehabilitation. 2018;42(4):397-427. doi:10.3233/NRE-172298
27. Picchiottino M, Leboeuf-Yde C, Gagey O, Hallman DM. The acute effects of joint manipulative techniques on markers of autonomic nervous system activity: a systematic review and meta-analysis of randomized sham-controlled trials. Chiropractic and Manual Therapies. 2019;27:17. doi:10.1186/s12998-019-0235-1
28. Pickar JG. Neurophysiological effects of spinal manipulation. The Spine Journal. 2002;2(5):357-371. doi:10.1016/s1529-9430(02)00400-x
29. Pickar JG, Sung PS, Kang YM, Ge W. Response of lumbar paraspinal muscle spindles is greater to spinal manipulative loading compared with slower loading under length control. The Spine Journal. 2007;7(5):583-595. doi:10.1016/j.spinee.2006.10.006
30. Ray CA, Hume KM. Neck afferents and muscle sympathetic activity in humans: implications for the vestibulosympathetic reflex. Journal of Applied Physiology. 1998;84(2):450-453. doi:10.1152/jappl.1998.84.2.450
31. RECOVER Initiative. RECOVER-AUTONOMIC clinical trial results shared at 2026 ACC Conference. National Institutes of Health, March 28, 2026. recovercovid.org
32. Reed WR, Pickar JG, Sozio RS, Liebschner MAK, Little JW, Gudavalli MR. Characteristics of paraspinal muscle spindle response to mechanically assisted spinal manipulation: a preliminary report. Journal of Manipulative and Physiological Therapeutics. 2017;40(6):371-380. doi:10.1016/j.jmpt.2017.03.006
33. Sato A, Schmidt RF. Somatosympathetic reflexes: afferent fibers, central pathways, discharge characteristics. Physiological Reviews. 1973;53(4):916-947. doi:10.1152/physrev.1973.53.4.916
34. Sato A. Neural mechanisms of autonomic responses elicited by somatic sensory stimulation. Neuroscience and Behavioral Physiology. 1997;27(5):610-621. doi:10.1007/BF02463910
35. Shaw BH, Stiles LE, Bourne K, Green EA, Shibao CA, Okamoto LE, Garland EM, Gamboa A, et al. The face of postural tachycardia syndrome: insights from a large cross-sectional online community-based survey. Journal of Internal Medicine. 2019;286(4):438-448. doi:10.1111/joim.12895
36. Shiffer D, Rigo S, Minonzio M, Yarsuvat DT, Tobaldini E, Furlan L, Montano N, Cairo B, et al. Short and long term effects of a two-week transcutaneous vagus nerve stimulation in hyperadrenergic postural orthostatic tachycardia syndrome: a proof-of-concept trial. European Journal of Internal Medicine. 2026;143:106529. doi:10.1016/j.ejim.2025.106529
37. Singh N, Hammam E, Macefield VG. Vestibular modulation of muscle sympathetic nerve activity assessed over a 100-fold frequency range of sinusoidal galvanic vestibular stimulation. Journal of Neurophysiology. 2019;121(5):1644-1649. doi:10.1152/jn.00679.2018
38. Stavrakis S, Chakraborty P, Farhat K, Whyte S, Morris L, Abideen Asad ZU, Karfonta B, Anjum J, et al. Noninvasive vagus nerve stimulation in postural tachycardia syndrome: a randomized clinical trial. JACC: Clinical Electrophysiology. 2024;10(2):346-355. doi:10.1016/j.jacep.2023.10.015
39. Stewart JM, Taneja I, Medow MS. Reduced central blood volume and cardiac output and increased vascular resistance during static handgrip exercise in postural tachycardia syndrome. American Journal of Physiology. Heart and Circulatory Physiology. 2007;293(3):H1908-H1917. doi:10.1152/ajpheart.00439.2007
40. Stewart JM, Nafday A, Ocon AJ, Terilli C, Medow MS. Cutaneous constitutive nitric oxide synthase activation in postural tachycardia syndrome with splanchnic hyperemia. American Journal of Physiology. Heart and Circulatory Physiology. 2011;301(3):H704-H711. doi:10.1152/ajpheart.00171.2011
41. Stewart JM. Update on the theory and management of orthostatic intolerance and related syndromes in adolescents and children. Expert Review of Cardiovascular Therapy. 2012;10(11):1387-1399. doi:10.1586/erc.12.139
42. Yates BJ, Bolton PS, Macefield VG. Vestibulo-sympathetic responses. Comprehensive Physiology. 2014;4(2):851-887. doi:10.1002/cphy.c130041
43. Zhang Q, Chen X, Cao C, Zhang Y, Zhao J, Shi L, Yan S, Jin Z. Vestibular habituation is associated with distinct autonomic signatures and greater reduction in visually induced motion sickness: a randomized controlled trial. Hearing Research. 2026;472:109554. doi:10.1016/j.heares.2026.109554