This is the first study to analyze the influence of autonomic dysfunction on survival in patients with DLB. We show that presence of persistent autonomic dysfunction is a possible predictor of a shorter survival in this population. The frequency of symptoms related to autonomic dysfunction was high in this DLB cohort and the most frequent manifestation was orthostatic hypotension.
In PD, non-motor symptoms like autonomic system dysfunction, depression, psychosis and sleep disturbances are considered by many patients to be more disabling than the motor symptoms 
. Despite this, they are often poorly recognized and inadequately treated, in contrast to the motor symptoms of the disease. 
. Patients with DLB and PDD report a more impaired quality of life compared to AD 
. Given that non-motor symptoms, including autonomic dysfunction, have a notable impact on quality of life in PD 
, this may be the case also in DLB.
There are different methods to evaluate the presence and severity of autonomic dysfunction. Many earlier studies on different diagnostic populations have a retrospective study design and they share the unreliable nature of such data. Patients may not report symptoms of autonomic dysfunction spontaneously, and the prevalence is most likely underestimated. Although orthostatic hypotension is known to give symptoms like light-headedness, visual blurring, dizziness, generalized weakness, fatigue, coat-hanger ache, nausea and abdominal discomfort it is shown that only 43% of non-demented patients with profound orthostatic hypotension have typical symptoms 
. Corresponding findings are reported for demented patients 
. The great majority of recent studies on PD use scales based on reports from patients and caregivers to detect non-motor symptoms. We have found only one scale where presence of orthostatic hypertension is based on objective blood pressure measurements; Composite Autonomic Severity Score (CASS) 
In this prospective study we focus on the three most common 
dysautonomic symptoms; constipation, urinary incontinence and orthostatic hypotension. A strength of the study is that orthostatic hypotension was detected by repeated and standardized blood pressure measurements. We do not specify whether patients with orthostatic hypotension are asymptomatic or symptomatic. In most studies, the time of standing during an orthostatic test is 3–5 minutes. Importantly, demented patients may not show significant falls in blood pressure until as late as after 10 minutes 
. Measurements of blood pressure should therefore be recorded for at least 10 minutes after standing up. It has also been shown that DLB patients react different to orthostatic challenge compared to AD and controls, with a more prolonged period of orthostasis after standing up 
. All patients in our study had three documented orthostatic tests during the follow up, although a few did not manage to accomplish the 10 min measure point. The 12 patients who did not complete the three assessments were excluded, which means that there is a possibility that those with the most severe orthostatic hypotension are not in the study. This may strengthen our results.
Our results suggest that orthostatic hypotension is the more important prognostic factor compared to constipation and incontinence as neither of these two manifestations alone was found to affect survival. We have not found any earlier studies addressing this issue. A possible explanation is that constipation and urinary incontinence are common in an elderly population due to a variety of causes and their specificity for autonomic dysfunction is probably low, especially when they exist without concurrent orthostatic hypotension. However, there are also several possible explanations to orthostatic hypotension. Common reasons in the elderly are dehydration and congestive heart failure, but also side effects from drugs, for example antihypertensive medication, antiparkinson medication or certain antidepressants. The negative effect of orthostatic hypotension on survival may be independent of the underlying cause of orthostatic hypotension. In this study, information on constipation and urinary incontinence is based on prospectively collected reports from patients and caregivers but no objective test. Furthermore, we did not rate the severity, as we did with orthostatic hypotension. The major limitation of our study is the small sample size. Due to this, and especially the low number of events (n
7) a multivariate analysis to adjust for possible confounders, was not feasible. In our population with DLB/PDD patients, the frequency of coexisting heart disease and the use of antihypertensive and antiparkinson medication were equal in both groups. There was no significant difference in the use of antidepressants between groups, but it tended to be more common in the subgroup with persistent orthostatic hypotension. It is possible that the more severe orthostatic hypotension is at least partly influenced by concomitant medication.
The pathophysiological mechanism of the association between autonomous dysfunction and mortality is not clear. Autonomic dysfunction is thought to be related to the α-synuclein pathology, since several neuropathological studies describe the presence of lewy bodies in strategic locations like locus coeruleus, sympathetic ganglia and parasympathetic plexus 
. In PD, the non-motor symptoms are related to advanced stages in patients with a fully developed motor phenotype 
, but are reported significantly more common in the disease across all stages than in controls 
. The latency from onset of disease to orthostatic hypotension seems to be delayed in DLB compared to MSA 
and PD 
Studies comparing survival in AD and DLB are inconclusive. Some studies report a similar rate of cognitive decline, but a shorter survival 
and shorter time to institutionalization in DLB 
. These findings suggest that progression of non-cognitive symptoms differ between AD and DLB, and our findings add to these findings by suggesting that autonomic dysfunction may be one of the contributing factors to this difference.
The extent and pattern of autonomic dysfunction in dementia with Lewy bodies are poorly documented, even though many authors mention it as a well-known clinical feature. To our knowledge, this is the first study to investigate its influence on prognosis and survival in DLB even though there are a few studies proposing that repeated hypotensive episodes may exaggerate cognitive decline on the basis of cerebral hypoperfusion and microvascular lesions 
. However, further research is needed to clarify the impact of autonomic dysfunction on factors such as than cognition, quality of life, activities of daily living and behavioral and psychiatric symptoms.
In summary we have found a high frequency of symptoms related to autonomic dysfunction in patients with DLB/PDD and importantly, patients with persistent orthostatic hypotension had a shorter survival. This needs to be verified in larger studies. Orthostatic hypotension seems to be the most important feature in autonomic dysfunction and we strongly recommend orthostatic blood pressure measurement on a routine basis in all patients with DLB/PDD.