Prodromal autonomic dysfunction in Lewy body diseases
Date: May 2026
Prepared by SIC Members: Sephira Ryman, PhD, and Michele Matarazzo, MD
Authors: David S. Goldstein, MD, PhD, Abhimanyu Mahajan MD, MHS; Alison Yarnall, PhD, MBBS
Editors: Lorraine Kalia, MD, PhD; Per Svenningsson, MD, PhD
Introduction
Prodromal autonomic dysfunction has emerged as an important area of research in understanding Lewy body diseases, including Parkinson's disease (PD) and Dementia with Lewy bodies (DLB). Increasing evidence suggests that disturbances in cardiovascular, gastrointestinal, urinary, thermoregulatory, and other autonomic domains may precede the onset of classic motor or cognitive symptoms by years.
This discussion aims to discuss the evidence supporting autonomic dysfunction in prodromal Lewy body diseases, which autonomic domains may be most informative for early detection, emerging diagnostic tools, and insights of these findings and their relevance to clinical care.
This blog posting was produced in collaboration with Michelle Matarrazo and the MDS Podcast.
Listen to the podcast episode:
1. What clinical or biomarker evidence supports the presence of autonomic dysfunction in the prodromal phase of Lewy body diseases?
Dr. David Goldstein
The strongest evidence for a specific autonomic abnormality in Lewy body diseases has come from cardiac sympathetic neuroimaging. Our initial report in 1997 in the New England Journal of Medicine described for the first time that in patients with Lewy body diseases, Parkinson disease and pure autonomic failure, exhibit loss of myocardial sympathetic nerve terminals using [18F] fluorodopamine PET scanning. We also demonstrated no differences in the non-Lewy body synucleinopathy, Shy-Drager syndrome, which came to be called multiple system atrophy.
Over the course of the last three decades we confirmed these findings postmortem by assaying norepinephrine content in the heart. By now there is no question that there is profound cardiac noradrenergic deficiency in Lewy body diseases and by and large none in multiple system atrophy.
When in the course of these diseases does this autonomic abnormality become manifest? At least in what's now called the body-first form of central synucleinopathies, the cardiac sympathetic lesion precedes the onset of either Parkinsonism or cognitive dysfunction as in dementia with Lewy bodies.
Dr. Abhimanyu Mahajan
One of the best early papers on this topic was published in 2013 by Professor Ron Postuma, where he described autonomic features in a group of 91 patients. They noted at least one symptom per autonomic subdomain, which included orthostatic symptoms, urinary dysfunction, constipation, and erectile dysfunction in the prodromal phase of the disease.
We recently used the Parkinson’s Progression Markers Initiative (PPMI) database to evaluate whether early autonomic burden predicts incident cognitive impairment in prodromal PD. We used the Scales for Outcomes in Parkinson’s Disease Autonomic Dysfunction (SCOPA-AUT), which is a validated scale by the MDS. If you take a positive answer on one of those questions in each subdomain, it is amazing how many of these patients in the prodrome have at least one autonomic symptom.
Dr. Alison Yarnall
In prodromal DLB, mild cognitive impairment with Lewy bodies, we have deeply phenotype cohorts here in Newcastle that we followed up over a number of years and compared that to mild cognitive impairment with Alzheimer's and with controls. Certainly, from our experience, we found that autonomic symptoms were commonly reported, and that was more common compared to normal aging. However, it wasn't more common compared to those with MCI with Alzheimer's type pathology.
1.1. Are specific autonomic domains (e.g. cardiovascular vs. gastrointestinal) more frequently abnormal in prodromal cohorts?
Dr. Abhimanyu Mahajan
There is merit to the suggestion that cardiovascular autonomic dysfunction has a strong influence on phenoconversion and cognitive outcomes. We demonstrated that early autonomic dysfunction based on the SCOPA-AUT was significantly associated with increased risk of mild cognitive impairment or dementia (P < 0.001). Cardiovascular autonomic symptom burden, specifically, was a strong independent predictor (HR = 5.21, P = 0.01).
Cardiovascular autonomic dysfunction has had robust biomarkers developed. However, the more we know, the more we know how little we know about autonomic dysfunction. If better biomarkers are developed in other subdomains, these may equally, if not more influential.
1.2. How do prodromal autonomic symptoms compare in sensitivity and specificity to other prodromal markers?
Dr. Alison Yarnall
We found that gastric and secretory psychomotor symptoms were commonly reported, but cardiovascular symptoms were more severe in our prodromal DLB cohorts. Many of these symptoms have high sensitivities, so they are commonly present. We see them in more than half of prodromal DLB cases, especially constipation or orthostatic hypotension. These symptoms have much lower specificity compared to other core markers that we see, such as the presence of REM sleep behavior disorder or abnormal dopamine transporter imaging.
It is an evolving landscape and we are recognizing autonomic dysfunction more and more. On the flip side, a lot of these symptoms are also commonly reported in older adults.
2. Which emerging methods for evaluating autonomic function hold the most promise for detecting prodromal disease?
Dr. David Goldstein
Orthostatic hypotension is under-diagnosed. Neurologists don't often think about it unless the patient has a symptom. You're going to underestimate the frequency of orthostatic hypotension if you depend on symptoms.
There are physiological, neurochemical, neuroimaging, and micro-neuroimaging biomarkers that overlap in a Venn diagram. When it comes to physiological, indices of baroreflex and sympathetic neural function may be helpful. When it comes to neurochemical, CSF levels of catecholamines and their metabolites may be helpful. When it comes to neuroimaging, we already talked about the cardiac PET scanning in body-first PD and DLB. In the real world, MIBG SPECT scanning may be helpful. When it comes to micro-neuroimaging, skin biopsies may be helpful, looking for alpha-synuclein deposition in sympathetic noradrenergic nerves supplying blood vessels, sweat glands, and arrector pili muscles. In addition to the neurochemistry about catecholamines, there's CSF synuclein seeding, which was predictive in our longitudinal prospective study.
In the future, I hope there will be a switch from purely clinical observations to an algorithmic approach that incorporates biomarkers. Start with things that are quick, cheap, safe, but adequately sensitive--from that point of view, heart rate variability analysis especially in the frequency domain. After that, move to synuclein markers followed by fluorodopa PET scanning of the brain. If it can be done, fluorodopamine PET scanning of the heart.
Dr. Abhimanyu Mahajan
Having very specific biomarkers is important for us to understand pathophysiology. Not just pathophysiology, physiology of how all of this works. At the same time, there is quite a bit of merit to not underestimate the benefit of what we can do at the bedside.
Orthostatic blood pressures are underused. We conducted a study at the University of Cincinnati and evaluated bedside objective measurements of orthostatic hypotension. About 30% of patients did not know that their blood pressures were orthostatic. Importantly, even if they were asymptomatic, the risk of worse activities of daily living and falls was still high, highlighting that even when the patient doesn't recognize orthostatic hypotension, it does not mean that it's not impactful.
Despite the relatively low sensitivity of symptoms, it is very important to capture them at the bedside. A lot of our information that we have about the impact of orthostatic symptom burden and its impact on other domains in PD is from observational studies and cohorts followed in clinic, which include measures such as SCOPA-AUT and others. It is important capture throughout the disease course. The ability to do these scales and these assessments at the bedside affords us the ability to capture information from large sample sizes which helps us to account for some of the disease heterogeneity that we see.
Both biomarker and clinical approaches to assessing autonomic dysfunction have to work in parallel. They're complimentary, of course.
2.1. How do you see continuous remote monitoring improving diagnosis and our real-world understanding of patients outside the clinic?
Dr. Alison Yarnall
Continuous monitoring is the way the world is going. Here in the UK we have a 10-year plan for our NHS. One of the three pillars is moving from hospital to community, from sickness to prevention, to managing care at home. This will help measure autonomic dysfunction, as well as a multitude of other measures, such as sleep, fatigue, mobility, gait, and cognition. Measuring this in someone's own environment over a seven day period, or similar, means we don't have to rely on a single snapshot that won't capture the inherent fluctuations that are part of PD and DLB.
It could potentially reduce burden on clinician time. If we have real time patient outcomes, we might be able to make changes remotely to prevent people having to come up to clinic. It allows for inclusion as well. Older, frailer patients, people in a rural location would have similar care to those living near university centers both in terms of clinical and a research opportunities.
Taking algorithmic approach similar to what David discussed would be the best approach. Capturing a plethora of measures and a score can be developed to differentiate those who progress more quickly, more slowly, and those who require intervention.
Dr. David Goldstein
I would add measures of continuous ambulatory blood pressure monitoring. It’s really important to find out when orthostatic hypotension is worse during the day. Often it's worse in the morning, and if it's worse in the morning, then you're not going to treat the person with a vasoconstrictor around the clock.
3. What clinical implications do these insights have for risk stratification and monitoring of individuals with prodromal autonomic dysfunction?
Dr. David Goldstein
I'll refer to the PD risk study because it's one of the few longitudinal prospective studies of people who have multiple risk factors to determine the predictive value of different types of biomarkers. The prediction, just based on risk factors, family history, dream enactment behavior, olfactory dysfunction, and orthostatic intolerance are very strong. In combination they substantially increase predictive and prognostic outcomes in terms of phenoconversion in the future. When objective laboratory biomarkers of catecholamine deficiency of alpha-synucleinopathy are added, then in essence it is not really a matter of risk anymore. You have the disease. It's a matter of time before clinical symptoms present. Those biomarkers will change a relative risk to a hundred percent. The future is in adding in biomarkers to the risk factors that we've already talked about.
4. Looking ahead, what are the potential therapeutic implications?
Dr. Alison Yarnall
Perhaps in the future, the patient would come into the clinic, you would identify autonomic physiology and symptoms that could be used to define their treatments and their treatments options. Autonomic prodromal features can tell us about risk and risk stratification. That can lend support to who you put forward into a clinical trial.
The ultimate clinical trial will be something that prevents someone converting from prodromal disease into manifest DLB or PD. Clinical and biomarker features of autonomic disease may help us decide who goes into which clinical trial.
Dr. Abhimanyu Mahajan
Enriching clinical trials absolutely is the right way to go. Now that we know early autonomic dysfunction is associated with a faster time to phenoconversion, greater morbidity, lesser survival, and perhaps greater cognitive risk, it's important to stratify the sample in clinical trials.
5. What are the current research priorities in this area?
Dr. David Goldstein
Identifying biomarkers of preclinical disease. Even in prodromal disease there already is extensive damage. The hope for the future is to identify the disease process before the person has symptoms at all. The goal is not to prolong the period of symptomatic disease, it's just the opposite. The goal is to delay the onset of symptomatic disease. If you could prolong healthspan, that would be the societal goal.
Dr. Alison Yarnall
Management of these difficult to treat autonomic symptoms is important. There is a relative dearth in the current evidence base at the present time, and it often significantly impacts quality of life. That would be my top research priority.
Dr. Abhimanyu Mahajan
Identifying ways capture of autonomic burden for every patient with a synucleinopathy who comes to the clinic, whether it is a prodromal patient, early PD, or late DLB. As long as you can appropriately capture it and treat those symptoms, not just for the effect of those symptoms, but also, and this is a pie in the sky picture, to prevent associated symptoms in the future. That would be the first priority for me.
We also don't have enough options in our arsenal to attack cognitive impairment. Autonomic dysfunction, including cardiovascular autonomic dysfunction, may be beneficial to target to improve cognitive impairment. We need to identify interventions that improve cognition.
References
-
Goldstein DS, Holmes C, Cannon RO, Eisenhofer G, Kopin IJ. Sympathetic cardioneuropathy in dysautonomias. N Engl J Med 1997;336(10):696–702. doi:10.1056/NEJM199703063361004
-
Goldstein DS, Holmes C, Lopez GJ, Wu T, Sharabi Y. Cardiac sympathetic denervation predicts Parkinson disease in at-risk individuals. Parkinsonism Relat Disord 2018;52:90–93. doi:10.1016/j.parkreldis.2017.10.003
-
Goldstein DS. Preclinical catecholaminergic biomarkers of central Lewy body diseases. J Clin Neurol 2026;22(1):6. doi:10.3988/jcn.2026.22.1.6
-
Hamilton CA, Frith J, Donaghy PC, Barker SA, Durcan R, Lawley S, Barnett N, Firbank M, Roberts G, Taylor JP, Allan LM, O’Brien JT, Yarnall AJ, Thomas AJ. Assessment of autonomic symptoms may assist with early identification of mild cognitive impairment with Lewy bodies. Int J Geriatr Psychiatry 2022;37(4). doi:10.1002/gps.5697
-
Hamilton CA, Frith J, Donaghy PC, Barker SA, Durcan R, Lawley S, Barnett N, Firbank M, Roberts G, Taylor JP, Allan LM, O’Brien JT, Yarnall AJ, Thomas AJ. Blood pressure and heart rate responses to orthostatic challenge and Valsalva manoeuvre in mild cognitive impairment with Lewy bodies. Int J Geriatr Psychiatry 2022;37(5). doi:10.1002/gps.5703
-
Martinez-Nunez AE, Mills KA, Seemiller J, Morrow CB, Pontone GM, Mahajan A. Early autonomic burden in prodromal Parkinson disease predicts cognitive impairment. Mov Disord 2026. doi:10.1002/mds.29972 (epub ahead of print)
-
Mahajan A, Morrow CB, Seemiller J, Mills KA, Pontone GM. The effect of dysautonomia on motor, behavioral, and cognitive fluctuations in Parkinson disease. Mov Disord 2025;40(1):157–162. doi:10.1002/mds.29614
-
Mahajan A, Dwivedi AK, Galvin JE, Maixner SM, Paulson HL, Manning CA, Fields JA, Boeve BF, Mills KA, Morrow C, Pontone GM, Armstrong MJ. Cognitive fluctuations, neuropsychiatric burden, and quality of life in moderate-advanced dementia with Lewy bodies: the role of dysautonomia. Neurology 2026;106(9):e214934. doi:10.1212/WNL.0000000000214934
-
Merola A, Romagnolo A, Rosso M, Lopez-Castellanos JR, Wissel BD, Larkin S, Bernardini A, Zibetti M, Maule S, Lopiano L, Espay AJ. Orthostatic hypotension in Parkinson disease: does it matter if asymptomatic? Parkinsonism Relat Disord 2016;33:65–71. doi:10.1016/j.parkreldis.2016.09.013
-
Patino J, Duque K, Abanto J, Bloemer A, Deraz H, Espay AJ, Mahajan A. Hypotensive episodes precede cognitive fluctuations in Parkinson disease dementia. Mov Disord Clin Pract 2026. doi:10.1002/mdc3.70590 (epub ahead of print)
-
Postuma RB, Gagnon JF, Pelletier A, Montplaisir J. Prodromal autonomic symptoms and signs in Parkinson disease and dementia with Lewy bodies. Mov Disord 2013;28(5):597–604. doi:10.1002/mds.25473