It happens unexpectedly: a person long thought lost to the ravages of dementia, unable to recall the events of their lives or even recognize those closest to them, will suddenly wake up and exhibit surprisingly normal behavior, only to pass away shortly thereafter.
This phenomenon, which experts refer to as terminal or paradoxical lucidity, has been reported since antiquity, yet there have been very few scientific studies of it.
That may be about to change.
In an article published in the August issue of Alzheimer’s & Dementia , an interdisciplinary workgroup convened by the National Institutes of Health’s (NIH) National Institute on Aging and led by Michigan Medicine’s George A. Mashour, M.D., Ph.D., outlines what is known and unknown about paradoxical lucidity, considers its potential mechanisms, and details how a thorough scientific analysis could help shed light on the pathophysiology of dementia.
“We’ve assumed that advanced dementia is an irreversible neurodegenerative process with irreversible functional limitations,” says Mashour, professor in the department of anesthesiology, faculty in the neuroscience graduate program, and director of the Center for Consciousness Science.
“But if the brain is able to access some sort of functional network configuration during paradoxical lucidity, even in severe dementia, this suggests a reversible component of the disease.”
The paper describes earlier work documenting case studies of individuals with advanced dementia, including Alzheimer’s disease, appearing to be able to communicate and recall in a seemingly normal fashion at the end of life, to the astonishment of their caregivers.
“The accumulation of anecdotal reports about paradoxical lucidity in the scientific literature prompts several important research questions,” says NIA medical officer Basil Eldadah, M.D., Ph.D.
“We look forward to additional research in this area, such as better characterization of lucidity in its varying presentations, new instruments or methods to assess episodes of lucidity retrospectively or in real-time, tools to analyze speech patterns or other behavioral manifestations of lucidity, and evidence to inform decision-making challenges and opportunities prompted by unexpected lucidity.”
One precedent for investigating such events exists in the study of so-called near-death experiences.
In 2013, Mashour and his collaborators at Michigan Medicine published a basic science study showing evidence of electrical brain features indicative of a conscious state following cardiac arrest.
“We don’t know that the same thing is occurring with paradoxical lucidity, but the fact that this is usually happening around the time of death suggests there could be some common neural network mechanism,” he says.
Mashour admits that studying paradoxical lucidity will be a challenge, given the fleeting nature of the event.
Case studies report episodes lasting from mere seconds to at most several days for a small minority of cases.
The workgroup also outlines important ethical implications of this work, including the ability of vulnerable patients to participate in research and how the observation of paradoxical lucidity might change the way caregivers interact with people with dementia.
“Would research that might identify a systematically observable paradoxical lucidity provide comfort, for example, by offering loved ones a potential channel for closure, or might it induce worry if loved ones are left to wonder if a reversible cause of the dementia could have been found?
We do not know the answers but these could be important research questions in their own right,” says co-first author Lori Frank, Ph.D., of the RAND Corporation and former Health and Aging Congressional fellow with the National Institute on Aging.
The workgroup hopes their paper will help raise awareness within the scientific community to advance paradoxical lucidity research, and help validate the experiences of a multitude of caregivers.
Says Mashour, “Science is now trying to be thoughtful and attentive to something that has long been reported.”
- What is known about Paradoxical lucidity
Cognitive fluctuations in patients with dementia have been observed and documented, but usually in patients with early or moderate stages of the disease .
Nahm et al. [2–4] have collected literature reports, and Batthyany  collected reports from a 12-month retrospective survey among physicians, nurses, and care providers, usually based on the time shortly before the death of patients.
As these studies focused on the time around death, the phenomenon in this context is sometimes called terminal lucidity [2–4], but here, we will retain the designation of PL.
Anecdotal reports of dramatic fluctuations of cognitive abilities in severe neurodegenerative diseases that were seemingly not related to dying can be found in non academic sources; a population-based study of lucid episodes in patients with severe dementia also contained no mention of episodes that appeared to be specifically related to the death of patients .
Our focus here, however, is on cases involving severe dementia and includes the following considerations related to timing and duration.
In a sample of 49 cases, many with dementia, 43% of PL episodes occurred within the last day of life, 41% within 2-7 days before death, and 10% within 8–30 days before death .
In the enhanced case collection by Nahm , PL in patients with dementia seems to take place predominantly within 1–2 days before death.
This is consistent with Batthyany’s study  of patients with dementia. Of 38 case descriptions, 44% occurred within 1 day before death, 31% within 2–3 days, and 6% within 4–7 days before death.
Similarly, in a study of end-of-life experiences, seven out of ten caregivers in a nursing home reported that they had observed patients with dementia and confusion becoming lucid a few days before death during the past five years .
Of the 38 cases collected by Batthyany , 3% of the lucid episodes lasted less than 10 minutes, 16% lasted 10–30 minutes, 24% lasted 30–60 minutes, 29% lasted several hours, 11% lasted one day, and 5% lasted several days.
Nevertheless, episodes of PL may also be brief, lasting only a few seconds, and the patients may only speak a few words that express something meaningful of relevance to a given situation.
Hence, PL displays a considerable range of degree and variety. Lucid episodes that occur shortly before death may be accompanied by so-called deathbed visions, for example, visions of deceased loved ones .
There is a lack of systematic studies to assess the neurologic underpinnings as well as the epidemiology and phenomenological characteristics of lucid intervals among patients with severe dementia .
Similarly, although instruments to assess fluctuations of cognitive abilities of patients with dementia exist , a specific scale to describe the extent and qualitative aspects of lucid episodes occurring late in disease progression is not yet available.
- Related phenomena
There are numerous phenomena related to PL in dementia that have been more extensively investigated, might provide mechanistic insight, and might argue for biological plausibility.
The near-death experience (NDE) and unexpected arousal phenomena share the ostensibly paradoxical nature of PL but often have systems neuroscience explanations.
In this section, we will consider NDEs and other phenomena of unexpected behavioral recovery or experience.
NDEs have been reported across cultures since antiquity and are arguably the phenomena most closely aligned with PL in dementia, especially when the latter occurs just before death.
NDEs represent phenomenologically rich experiences in the setting of clinical death or a hypofunctioning brain [14–16].
Similar to PL in dementia, NDEs were primarily reported anecdotally, retrospectively, or in case studies until the early 2000s.
In 2001, two prospective epidemiological studies in cohorts of patients who had cardiac arrest revealed that the incidence of NDEs in this population could be as high as 18% [17,18], which is substantially more common than that might have been predicted from case reports alone.
It has been reported that dying patients in the critical care or operative setting can exhibit a surge of electrical activity in frontal montage electroencephalography [21–23].
However, the relevance of this electrical surge to the phenomenology of the NDE is unclear because such electrical surges could reflect nothing more than a nonspecific discharge attributable to the uncoordinated activity of neuronal firing in an excitotoxic cascade associated with hypoxia and loss of membrane integrity.
Experiments in a rodent model of both cardiac and respiratory arrest replicated the frontal surge of electrical activity but further demonstrated increased functional and directional connectivity between frontal and posterior cortices that were similar to neural correlates of consciousness identified in humans .
Follow-up studies in rodents have confirmed this surge of large-scale cortical communication after cardiac arrest and have detailed associted neurochemical surges .
The progression of NDE investigation from anecdote to case report to epidemiology to mechanistic investigation could be informative for future research programs of PL in dementia.
There are other forms of paradoxical improvements in cognitive function or arousal of possible relevance to PL.
For example, the sleep-promoting drug zolpidem has been shown to enhance arousal and behavioral recovery in patients in a vegetative state .
Zolpidem appears to cause what has been referred to as paradoxical metabolic and vascular changes [27,28].
However, such changes have been explained in the context of a mesocircuit model of disordered consciousness .
Similarly, children with autism have been found to show signs of recovery across a number of domains in the setting of fever .
These two sit- uations (zolpidem in vegetative states and fever in autism) share the common feature of a disordered neural network that recovers function after a perturbation that would typically have a depressive or disorganizing effect in the normal brain.
There has been recent work in anesthetic-induced uncon- sciousness reporting a phenomenon that has been referred to as paradoxical emergence.
In animals anesthetized with iso- flurane, the addition of the intravenous anesthetic ketamine induces a deeper state of unconsciousness (as indicated by the appearance of burst suppression) .
However, despite the deeper state of anesthesia, the animals treated with ketamine recover from unconsciousness 44% faster than controls.
This ostensible paradox, that is, an intervention that more profoundly disables the brain but allows it to recover function more quickly, was associated with enhanced cholinergic tone in the prefrontal cortex.
A follow-up study demonstrated that agonizing acetylcholine receptors in the prefrontal cortex could actually reverse the anesthetized state despite the ongoing administration of inhaled general anesthetics .
This return of function in the setting of clinical levels of general anesthetic in the brain has also been explained through a mesocircuit model, prominently involving the cholinergic system of the basal forebrain .
It is also important to note that there is evidence for preserved consciousness and selfhood in the setting of ostensibly devastating structural brain damage.
For example, patients with a diagnosis of vegetative state exhibit such covert consciousness, losing the ability to interact behaviorally with the world but able to demonstrate volition through brain activity, which is identified through functional magnetic imaging .
It is thus conceivable that patients with dementia could, in some cases, have a preserved self with the inability to engage behaviorally except in sporadic cases such as PL.
In summary, there are phenomena that, similar to PL in dementia, represent an unexpected or ostensibly paradoxical recovery of neural function with a systems neuroscience explanation.
More information: George A. Mashour et al, Paradoxical lucidity: A potential paradigm shift for the neurobiology and treatment of severe dementias, Alzheimer’s & Dementia (2019). DOI: 10.1016/j.jalz.2019.04.002
Provided by University of Michigan