Dementia with Lewy bodies, a common but obscure disease, was thrust into public consciousness in 2014 with the suicide of the actor Robin Williams. Now recognized as its own disorder, DLB is difficult to diagnose because it has both similarities with, and differences from, its better-known cousins Alzheimer’s and Parkinson’s. Now, new diagnostic and management guidelines aim to help clinicians and specialists detect and treat the disorder. In the July 4 issue of Neurology, a quorum of leading DLB clinicians across the world update their previous report on the state of the art in DLB research, which was published more than a decade ago, with knowledge shared and debated at the 2015 International Dementia with Lewy Bodies Conference (see Dec 2015 Conference Series). “We’ve tried to maintain strengths of the previous guidelines and add high-quality, evidence-based changes,” said lead author Ian McKeith, Newcastle University, U.K. “We want these criteria to be widely used and understandable by specialists and clinicians, as well as those who may diagnose DLB less frequently.” 

Heart Loses Its Nerve: In MIBG scintigraphy, researchers measure the signal from the heart (circle). It drops off in DLB compared with normal controls and people with AD. The rectangle encloses the reference chest space, the mediastinum. [McKeith et al., 2017; CC4.0.]

The 2005 consensus report on DLB has been highly cited and widely used, said McKeith (see McKeith et al., 2005). However, many practitioners found that the guidelines were difficult to apply and often missed cases, McKeith said. 

In the current update, the authors have reorganized the diagnostic information into clearly defined clinical features vs. biomarkers, whereas previous versions combined the two categories. This separation should make the guidelines clearer to understand and easier to apply more widely, McKeith told Alzforum. The authors specify defining “core” clinical features and “indicative” biomarkers, a combination of which can confirm diagnosis. These are separate from “supportive” clinical or biomarker features, which can strengthen a diagnosis but are not required.

New to the core clinical features category is REM sleep behavior disorder (RBD), which was included in the last report but considered merely suggestive. Since then, much data has accrued to show that RBD is common in DLB; almost three-quarters of patients act out their dreams while sleeping, often years before the onset of other symptoms. RBD now joins several other core clinical features of DLB that were previously on the list; they include fluctuating cognition, vivid and recurrent hallucinations; and some sign of parkinsonism—be it slowness of movement, tremor, or rigidity.

Another addition comes in the biomarkers category: the inclusion of iodine-123-metaiodobenzylguanidine (MIBG) myocardial scintigraphy (see Dec 2015 news). This type of imaging visualizes the amount of postganglionic sympathetic innervation to the heart, which plummets in DLB (see image above). Approved and reimbursed in Japan, this technique has a sensitivity and specificity of 69 and 87 percent for distinguishing DLB from AD, which rises to 77 and 94 percent if symptoms are still mild. Other indicative biomarkers include reduced dopamine transporter uptake in the basal ganglia in SPECT or PET, as well as polysomnography confirmation of REM sleep without the usual temporary muscle paralysis.

The authors recommend the minimum requirements for a diagnosis of probable DLB to be dementia with two or more of those core clinical features, or dementia with one clinical feature and one or more indicative biomarker. If signs of disease are limited only to biomarkers, or if only one clinical feature is present with no biomarkers, the diagnosis should be limited to possible DLB.

Supportive clinical features new to the list include excessive daytime sleepiness, and weakening sense of smell. These are included with sensitivity to antipsychotics, postural instability, repeated falls, severe autonomic dysfunction, delusions, and depression. Among the supportive biomarkers are relatively preserved medial temporal lobe structures on an MRI scan, unusual slow-wave electroencephalography in posterior brain regions, and hypometabolism in visual brain areas as measured by FDG-PET.

Although DLB has overlapping symptoms and pathologies with Alzheimer’s and Parkinson’s diseases, it is important to distinguish DLB as its own disorder, McKeith said. DLB patients respond differently to medications used in AD and PD, and it’s helpful to know which symptoms could arise later. In addition, DLB tends to progress and lead to death faster than AD or PD, so a correct diagnosis helps families plan.

The new guidelines feature few changes in disease management, as the published literature on DLB therapies remains scant. Based on clinical experience, cholinesterase inhibitors appear to improve DLB patients’ cognition, activities of daily living, and global function, the authors wrote. Doctors should avoid prescribing antipsychotics to these patients, as they can have severe adverse reactions. Likewise, dopaminergic drugs, used widely and effectively in PD, can cause psychosis in DLB patients, though some may tolerate minimal doses.

“The updated criteria represent a clear improvement, reflecting scientific advances in the field,” wrote Matthew Barrett, University of Virginia School of Medicine, Charlottesville, and Melissa Armstrong, University of Florida College of Medicine, Gainesville, in an accompanying editorial.

Michael Schlossmacher, Ottawa Hospital Research Institute, Canada, agreed. “The criteria are straightforward and make a lot of sense to me as a clinician,” he told Alzforum. They will help screen patients with clinically diagnosed dementia for upcoming DLB trials, as well as avoid the inclusion of people with DLB in separate trials for AD. In Schlossmacher’s experience, the core clinical features included in the guidelines are “right on,” though he would have liked to see reference made to GBA1 mutations as a genetic risk marker of disease. Studies have implicated such allelic variants in almost a quarter of DLB cases (Goker-Alpan et al., 2006). 

In the future, clinical trial scientists might consider how to subtype DLB patients and specify target outcomes, the authors wrote. They also called for assembling large, population-based cohorts, and sampling publicly available large genetic and exomic repositories to better understand the biology of the disease. McKeith and colleagues also suggested developing animal models. A subset of the authors are planning to develop a similar set of guidelines for prodromal DLB.—Gwyneth Dickey Zakaib


  1. The new modifications of the clinical criteria for diagnosis of DLB are a beautiful contribution, especially the inclusion of RBD as a core clinical feature together with visual hallucinations, parkinsonism, and cognitive fluctuation. No doubt this change will increase the sensitivity of the clinical diagnosis to predict the pathological findings in this disease.

    However, it would seem that there may be an error that should be amended in the criteria of probable DLB. In the case of a patient who has only one core clinical feature such as RBD, and the only supportive biomarker is REM without atonia in polysomnography, according to the criteria this patient should be considered probable DLB. In fact, the polysomnographic findings would only confirm the clinical report. Thus, a single criterion would ensure the diagnosis of probable DLB, and this would not be correct.

  2. The new guidelines have rightly put greater emphasis on the diagnostic role of REM behavior disorder, which is likely to be the most specific among the “core features.” In addition, a greater diagnostic weight has been given to MIBG myocardial scintigraphy (which now has the same value of DAT-scan), in light of the results of multiple studies during the last decade which have highlighted the high sensitivity and specificity of this procedure. These modifications can increase the identification of this (still underdiagnosed) condition during life and undoubtedly represent an advance in the field.

    Based on these new guidelines, at least two core features, or one core and one indicative biomarker, are required for a diagnosis of “probable DLB” while, as for prior formulations of Consortium Criteria, the presence of a single core feature only warrants the uncertain and unstable diagnosis of “possible DLB.” However, in terms of specificity, the core features have a different diagnostic weight (REM behavior disorder > visual hallucinations > fluctuations > parkinsonism), so that one can wonder whether the presence of RBD alone may be sufficient to warrant a diagnosis of probable rather than possible DLB. Perhaps the authors might have stressed that there is a hierarchy among the core features.

    The authors have differentiated between indicative (previously suggestive) and supportive biomarkers on the basis of the greater specificity of the former in supporting the clinical diagnosis, but have not clarified which procedure should be preferred in particular circumstances. For example, when parkinsonism is the single core feature exhibited by the patient with dementia, MIBG myocardial scintigraphy or FDG-PET imaging are likely to be more appropriate diagnostic options than DAT-scan. That is because, in general, one would expect the latter exam to be abnormal in the context of parkinsonism, which means that DAT-scan abnormalities add no or negligible diagnostic extra value when parkinsonism is already present on examination. 

Make a Comment

To make a comment you must login or register.


Conference Coverage Series Citations

  1. International Dementia with Lewy Bodies Conference 2015

News Citations

  1. Through the Heart? Cardiology Tracer to Nail DLB Diagnosis

Paper Citations

  1. . Diagnosis and management of dementia with Lewy bodies: third report of the DLB Consortium. Neurology. 2005 Dec 27;65(12):1863-72. PubMed.
  2. . Glucocerebrosidase mutations are an important risk factor for Lewy body disorders. Neurology. 2006 Sep 12;67(5):908-10. PubMed.

Further Reading


  1. . New tracers and new perspectives for molecular imaging in Lewy body diseases. Curr Med Chem. 2017 Jun 8; PubMed.
  2. . Lewy Body Disorders. Neurol Clin. 2017 May;35(2):325-338. PubMed.

Primary Papers

  1. . Diagnosis and management of dementia with Lewy bodies: Fourth consensus report of the DLB Consortium. Neurology. 2017 Jul 4;89(1):88-100. Epub 2017 Jun 7 PubMed.
  2. . Dementia with Lewy bodies advances: A new consensus report. Neurology. 2017 Jul 4;89(1):18-19. Epub 2017 Jun 7 PubMed.