Subscribe to eTOC

International Consensus on the First Diagnostic Guidelines for Posterior Cortical Atrophy



SINGLE PARTICIPANT axial images for one control participant and one patient with PCA showing cerebral blood flow (ASL), glucose metabolism (FDG-PET), atrophy (structural magnetic resonance imaging), and amyloid deposition (florbetapir-PET). For clinical purposes, 18F-florbetapir images should be read on a gray scale. Arterial spin labeling (ASF); cerebral blood flow (CBF); 18F-labeled fluorodeoxyglucose positron emission tomography (FDG-PET); SUVR, standard uptake value ratio.

An international panel of experts has developed an updated consensus document on the signs and symptoms of posterior cortical atrophy to help neurologists better understand and diagnose the condition.

Decades after the late D. Frank Benson, MD, of the University of California, Los Angeles (UCLA), first published a report on a series of patients with deficits in higher visual function and sensory aphasia, uncertainty remains about how many people develop posterior cortical atrophy (PCA), a form of dementia usually considered an atypical variant of Alzheimer's disease (AD).

But now a new consensus report by an international team of Alzheimer's disease experts aims to provide a roadmap to help neurologists and other health professionals better understand and diagnose the condition.

The report, published in the March 1 online edition of Alzheimer's & Dementia, provides the first diagnostic guidelines for the condition, which causes atrophy of the posterior part of the cerebral cortex, resulting in the progressive disruption of complex visual processing.

For most people the pathology is clear: The same amyloid plaques and tau tangles of AD show up in posterior cortical areas that govern vision and sensory perception, said the lead author of the guideline, Sebastian Crutch, PhD, a neuropsychologist and Alzheimer's senior research fellow at University College London (UCL) who has spent almost two decades studying PCA.

Optometrists and ophthalmologists generally see these patients first, and a referral to a neurologist may not occur until years later, Dr. Crutch said, adding that many may have neither seen nor diagnosed PCA.

“We needed a common language to talk about PCA,” Dr. Crutch said. While a few centers had independently published diagnostic criteria, no universal agreement existed on how best to identify the most salient features of PCA and arrive at a more uniform diagnosis.


DR. SEBASTIAN J. CRUTCH said that while a few centers had independently published diagnostic criteria no universal agreement existed on how best to identify the most salient features of PCA and arrive at a more uniform diagnosis.

“These are very important research guidelines and ultimately they can help clinicians nail down the clinical features we see in posterior cortical atrophy,” added Victoria Pelak, MD, another study author and professor of neurology and ophthalmology at the University of Colorado School of Medicine. “We also have to figure out a way to talk about PCA with our patients. While PCA is most commonly associated with Alzheimer's disease pathology, it is a very different presentation. Many patients have signs and symptoms for three to four years before they receive a correct diagnosis. Some of the delay can be blamed on a lack of awareness.”

The Alzheimer's Association International Society to Advance Alzheimer's Research and Treatment (ISTAART) funded the development of the new guidelines.


A 2012 review paper about PCA in The Lancet Neurology by Dr. Crutch and colleagues at the University of California, San Francisco (UCSF) and the University College London (UCL), highlighted problems with the criteria at the time, and the group set out to create a unified diagnostic language. They were particularly interested in the impact the diagnosis, or a misdiagnosis, was having on research.

For instance, PCA patients were often not considered in research studies on Alzheimer's disease that were heavily weighted to identify changes in memory. These patients, at least in early stages, did not have memory problems. When they were included, they had difficulty taking the tests being used to measure outcomes as they experience difficulty with processing visual information.

Gil D. Rabinovici, MD, associate professor of neurology at UCSF and one of the authors of the 2012 review, noted that the pathology in many PCA cases was not consistent with AD.

In 2012, about 25 experts came together to create a web-based questionnaire to gather data on the frequency of the signs and symptoms – including those that were never documented as well as those that were always present. They asked PCA experts around the world to rate their level of diagnostic agreement on various statements. About 36 neurologists, psychologists, pathologists, psychiatrists, gerontologists and neuroscientists completed the online assessment and these experts made up the working group for the diagnostic guidelines.

The working group met four times over several years, reviewing the diagnostic criteria defined at single centers, as well as papers in the field — all of which were considered in their revised and expanded definitions. A smaller group worked with the UCL and UCSF researchers and neurologists to draft the guidelines. The UCL group had seen about 120 PCA patients at their center. The UCSF group had worked up about 100 cases.


The consensus group found that PCA may present in different forms. Some people have greater problems seeing spatial relationships — where things are — and other people have more trouble with object perception — seeing what things are.

Some people can't see any visual details and still others complain of problems calculating or planning movements. Others have field-cut difficulties with lights and color and elemental aspects of visual perception. Many have difficulty reading. Some are virtually blind but maintain a sense of self, with daily function limited by visual impairments moreso than general cognitive deficits.

“We don't understand what makes people more vulnerable to specific visual or spatial problems,” said Dr. Rabinovici.

The consensus among the experts is that patients should have at least three of these symptoms — “basic visual, visuoperceptual, visuospatial, literacy, numeracy, praxis, and higher sensory deficits” – for a diagnosis of PCA. (For more details, see “Common PCA Symptoms and A Suggested Work-up.”)

The criteria in the current paper include the use of AD biomarkers, which were not available when Dr. Benson began describing these patients and coined the term posterior cortical atrophy. The working group also wrote a description of the different stages of the disease. Many, but not all, go on to a more global amnestic state, which also parallels the spread of the disease seen on brain scans and at autopsy.

Last year, the group published the first international multisite study on the genetics of PCA in the journal Alzheimer's & Dementia. In work led by Jonathan M. Schott, MD, a reader in clinical neurology at UCL, DNA from 300 PCA patients was tested.

So far, it appears that apolipoprotein E4 does not confer as great a risk as it does in amnestic forms of AD, said Dr. Crutch. Among the intriguing hits was one marker involved with the development of the visual cortex and another with cholesterol. The research team is also hunting for clues in early childhood to determine if patients with PCA had difficulties with reading or math.

The group continues to collect samples to replicate and extend the findings. They also want to understand why this rare form of AD starts in the posterior parts of the brain and what factors predict age of onset and its spread.

Dr. Rabinovici hopes that the guidelines help neurologists and ophthalmologists recognize PCA early enough for a diagnosis. In life, 90 percent of people with amnestic AD will have visual or spatial problems. He said it's important to ask about earlier signs of visuospatial deficits.

Dr. Crutch agreed. “Getting AD pathology in the occipital lobe is a matter of when, not if,” he said.

“Diagnosing PCA is a huge challenge,” added Dr. Rabinovici. “People with this condition are generally younger than those with amnestic AD and the atypical symptoms often lead to a long delay in diagnosis. They face a long road. Many start at the ophthalmologist's office and can undergo cataract surgery, and it takes a good ophthalmologist to figure out that this is a brain problem. Then, it takes a good neurologist to figure out that this is a neurodegenerative disease.”

Dr. Rabinovici said that neurologists should have a low threshold for ordering magnetic resonance imaging if patients show up with these visual and spatial problems. The scan would show changes in the posterior visual association areas and the integrative visual networks.

The scientists also hope that the guidelines help in research. PCA patients are often unable to do some of the tests used in clinical trials. Biomarkers may also be different because of the anatomy. For instance, hippocampal atrophy is not a good marker for PCA.

“These are the early days in studying PCA. These patients are young and healthy and we need to figure out a way to get them into clinical trials,” said Dr. Rabinovici. The guidelines might also help in determining PCA's incidence and prevalence.


“There has never been an international group to agree and harmonize over the criteria for PCA,” said David S. Knopman, MD, FAAN, professor of neurology at the Mayo Clinic in Rochester, MN. A Mayo Clinic group had published a series of PCA cases using its own criteria.

“PCA has an unmistakable presentation if you are acquainted with it. The problem is that it is rare,” Dr. Knopman said. “The guideline formalized a diagnostic approach that we were already using at the Mayo Clinic, but reference to a formal set of criteria enhances the stature of a diagnosis of PCA. An internationally created set of diagnostic criteria will not only aid in research but will also assist in improving recognition of PCA in the broader medical community.”

Dr. Knopman said that in his experience the patients are slow to progress and have relatively spared memory functions. Their visual problems invariably progress to the point where they are usually blind.

“Patients have insight into their problems and can benefit from visual aids and training to help them navigate their world,” he explained. “They have problems stitching together a visual scene. The brain is not able to distinguish the target object from the background.”

“Doing a clinical exam would generally tell us what parts of the brain are involved,” added Ronald C. Petersen, MD, PhD, FAAN, director of the Mayo Clinic Alzheimer's Disease Research Center and the Mayo Clinic Study of Aging. “Now, it is much more sophisticated and the clinical syndrome is not the only characteristic in making a diagnosis of PCA.”

“These days, we talk about the disease, the syndrome, and the pathology,” he continued. “It is a scholarly exercise at refining criteria for PCA and bringing in the pathology. The guidelines should help clinicians in making the diagnosis.”


  • Difficulty locating objects in space
  • Problems judging spatial relationships
  • Poor navigation and trouble driving
  • Visual agnosia
  • Prosopagnosia
  • Alexia or agraphia
  • Visual field deficit
  • Elements of Balint or Gerstmann syndrome (inability to perceive the visual field [simultanagnosia], difficulty in fixating the eyes [oculomotor apraxia], and inability to move the hand to a specific object by using vision)
  • Limb apraxia

The experts also recommended that clinicians do the following work-up of suspected PCA:

  • Detailed ophthalmology exam to rule out primary ocular disease
  • Cognitive testing: assess for visuospatial or visuoperceptual deficits with relative sparing of non-visual cognitive domains
  • Standard lab work-up for reversible metabolic causes of cognitive impairment
  • Structural brain imaging: CT or MRI to assess for occipito-parietal or occipito-temporal atrophy
  • Consider FDG-PET or SPECT to assess for posterior cortical hypometabolism or hypoperfusion
  • Consider amyloid PET to rule out underlying Alzheimer's disease


• Crutch SJ, Schott JM, Rabinovici GD, et al; for the Alzheimer's Association ISTAART Atypical Alzheimer's Disease and Associated Syndromes Professional Interest Area. Consensus classification of posterior cortical atrophy Alzheimers Dement 2017; Epub 2017 Mar 1.
    • Crutch SJ, Lehmann M, Schott JM, et al. Posterior cortical atrophy Lancet Neurology 2012; 11:170–178.
      • Schott JM, Crutch SJ, Carrasquillo MM, et al. Genetic risk factors for the posterior cortical atrophy variant of Alzheimer's disease Alzheimers Dement 2016; 12:862–871.
        • Crutch SJ, Schott JM, Rabinovici GD, et al. Shining a light on posterior cortical atrophy Alzheimers Dement 2013; 9:463–465.
          • Benson DF, Davis RJ, Snyder BD. Posterior cortical atrophy Arch Neurology 1988; 45(7): 789–793.