bs_bs_banner

doi:10.1111/jgh.13706

REVIEW ARTICLE

Gluten-induced cognitive impairment (“brain fog”) in coeliac

disease
Gregory W Yelland*,†
*Department of Gastroenterology, Central Clinical School, Monash University and †School of Health Sciences, RMIT University, Melbourne, Victoria,
Australia

Key words
coeliac disease, cognitive impairment, neural
inflammation, speed of processing, Subtle
Cognitive Impairment Test.
Accepted for publication 28 November 2016.
Correspondence
Dr Gregory W Yelland, Department of
Gastroenterology, Central Clinical School,
Monash University, The Alfred Hospital,
Melbourne, Vic, Australia..
Email: gwyelland@gmail.com
Disclosures: Dr Yelland is co-director of
NeuroTest Pty. Ltd.
Abstract
Much is known about the serious neurological effects of gluten ingestion in coeliac disease
patients, such as sporadic ataxia and peripheral neuropathy, although the causal links to
gluten are still under debate. However, such disorders are observed in only a small percent-
age of coeliac patients. Much less is known about the transient cognitive impairments to
memory, attention, executive function, and the speed of cognitive processing reported by
the majority of patients with coeliac disease. These mild degradations of cognitive func-
tions, referred to as “brain fog,” are yet to be formally recognized as a medical or psycho-
logical condition. However, subtle tests of cognitive function are measurable in untreated
patients with coeliac disease and improve over the first 12 months’ therapy with a
gluten-free diet. Such deficits also occur in patients with Crohn’s disease, particularly in as-
sociation with systemic inflammatory activity. Thus, cognitive impairments associated with
brain fog are psychologically and neurologically real and improve with adherence to a
gluten-free diet. There is not yet sufficient evidence to provide a definitive account of the
mechanism by which gluten ingestion causes the impairments to cognitive function associ-
ated with brain fog, but current evidence suggests that it is more likely that the causal factor
is not directly related to exposure to gluten.

Coeliac disease (CD) is an inflammatory autoimmune reaction to
gluten ingestion causing small bowel damage, for which the only
existing treatment is a gluten-free diet (GFD). The intestinal le-
sions are defined by varying degrees of villous atrophy and
intraepithelial lymphocytosis. However, the consequences of CD
are not restricted to the intestinal tract. It is a systemic inflamma-
tory disease that impacts the brain and neural system in at least
two ways: major (or gross) neurological disorders and “silent”
neurological complaints. While the prevalence of gluten-related
major neurological disorders is very low, it is where the bulk of
empirical research has been focused. However, little attention
has been paid to the silent neurological consequences of gluten in-
gestion in CD and yet it would appear that its behavioral conse-
quences, “brain fog,” are considerably more prevalent.
The major neurological complications of CD are identifiable
neurological disorders where there are clear patterns of neural
and behavioral dysfunctions that have serious impact on the life
of the patients. Gluten ataxia is a disorder of the central nervous
system, a specifically gluten-related cerebellar atrophy that results
in a lack of coordination of complex movements like walking,
speaking, and swallowing.1 Peripheral neuropathy is consequence
of inflammation in peripheral nerve fibers, which variously impairs
sensation (sensory nerves), movement (motor nerves), or gland or
organ function (autonomic nerves).2 There is increasing evidence
that untreated CD could result in severe and degenerative loss of
global cognitive function, in particular loss of memory function,
that is, dementia.3,4 Estimates of the prevalence of these neurolog-
ical disorders in CD patients vary from 2% to 10% of the CD
population and is highest amongst the late-diagnosed elderly and
individuals who are not compliant with the GFD.
The so-called silent neurological complications of CD refer to
changes in structure of the brain that are not focal in nature and
have no overt neurological symptoms.5 They are characterized
by small, diffuse reductions in size of neural cell bodies or white
matter hyperintensities, that is, inflammation of nerve fibers. This
inflammation results in a reduction in the speed of signal transmis-
sion, with the volume of white matter hyperintensities associated
with the magnitude of cognitive decline.4
The behavioral correlates of silent neurological damage are the
subtle impairments to memory, attention, decision-making, and
the speed of cognitive processing collectively referred to as “brain
fog.” Instances of “brain fog” are frequently reported by patients
with CD who are inadvertently exposed to gluten, and improve-
ment in these symptoms has been noted in those commencing a
GFD in newly diagnosed disease. Research into the subtle cogni-
tive deficits associated with brain fog is limited, existing as anec-
dotal reports from patients and physicians, and as such, brain fog
is not yet formally recognized as a medical or psychological con-
dition. Even the prevalence of brain fog is difficult to establish
although the weight of anecdotal evidence suggests that it is wide-
spread. A recent study of 120 individuals recruited through
Coeliac Australia found that all (100%) had experienced brain
fog Symptoms following the ingestion of gluten.6
Objective empirical evidence has been difficult to obtain but is
required in order to ascertain the true nature, prevalence, and mag-
nitude of brain fog in CD. There are few existing studies of

90 Journal of Gastroenterology and Hepatology 2017; 32 (Suppl. 1): 90–93

© 2017 Journal of Gastroenterology and Hepatology Foundation and John Wiley & Sons Australia, Ltd
GW Yelland Cognitive impairment in coeliac disease

cognitive impairments in CD patients, and most were not focused presentation of the target stimulus is followed by a visual mask.
on the subtle impairments characterized by brain fog. For example, The mean response time (RT; in milliseconds) and mean percent-
Hallert et al. examined 19 untreated CD patients using the Down’s age error rate are calculated for each exposure duration. For the
syndrome cognitive (dementia) test battery and found no evidence four stimulus exposure durations less than 64 ms, attention cannot
of cognitive impairment.7 Bürk et al. found evidence of signifi- be brought to bear and so the participant’s response relies entirely
cantly impaired recall memory and trends towards deficits in ver- on automatic processes. However, for the exposure durations
bal fluency and executive function, but all eight of their greater than 65 ms, the stimulus is available to conscious attention
participants presented with gluten ataxia.8 A retrospective study and so controlled processes are available to aid in the response.
by Hu and colleagues of 13 patients aged between 45 and 79 years, Thus, collectively, the SCIT provides measures of the speed of
all of whom presented with cognitive impairment varying in sever- processing (SCIT-RT) and the effectiveness (SCIT-Error) of pro-
ity from dementia to minor impairment (nine with gluten ataxia cessing for both automatic and controlled processes, factors that
also), reported that 92% had impaired memory.9 Casella and col- underlie most of cognitive functions.
leagues observed impairments to attention and semantic and The SCIT was used in a study of 11 newly diagnosed CD pa-
visual–spatial memory in 36 elderly CD patients who were signif- tients examined over a period of 12 months from the onset of
icantly impaired on the Mini-Mental State Examination, a screen- GFD.18 Participants were aged between 22 and 39 years, with no
ing measure for dementia.3 The problem for these studies is that overt neurological symptoms and high compliance with the
they either (i) include participants who have existing major neuro- GFD. A battery of cognitive tests (including SCIT) were adminis-
logical deficits, such as dementia or gluten ataxia, where signifi- tered at and 12 and 52 weeks after the onset of the GFD. A range
cant cognitive deficits would be expected and as such were not of biological markers were measured, at the same time points and
examining brain fog; or (ii) the tests used were not sufficiently sen- small bowel biopsies were collected via routine gastroscopy at di-
sitivity to assess minor impairments in cognitive function associ- agnosis and at 12 and 52 weeks post-GFD onset. Significant im-
ated with brain fog.5 provement from 0 to 52 weeks for the speed of automatic
Sufficiently sensitive tests were developed only relatively re- processing (SCIT-RT; Fig. 2), the speed and control of motor
cently in response to the growing need to detect the very early pre-
cursors of Alzheimer’s disease. Some have utility for the detection
of minor cognitive impairments generally. One such test is the Sub-
tle Cognitive Impairment Test (SCIT, Neurotest.com),10,11 a very
short (3–5 min), computer-based, perceptual judgement task that
has been found to be sensitive to very small differences in global
cognitive performance across a wide range of clinical and non-
clinical populations.12–16 It is a reliable and valid measure that has
no practice effects.17 It has been used successfully in the lab, home,
or at the bedside with participants aged between 7 and 96 years.
The SCIT requires participants to indicate which of two un-
equal, parallel vertical lines in the target stimulus (Fig. 1) is shorter
(the line on the left or that on the right). Participants indicate their
choice by pressing the corresponding mouse button. The target
stimuli are presented very briefly with eight different exposure du-
rations between 16 and 128 ms across 96 trials. On each trial, the

Figure 1 Examples of the Subtle Cognitive Impairment Test stimuli. A

visual fixation stimulus shown in the center of the computer screen for Figure 2 Subtle Cognitive Impairment Test (SCIT) response time (RT)
500 ms. This is followed by a brief presentation (16–128 ms) of one of (a) and SCIT error (b) outcomes for 0, 12, and 52 weeks post gluten-free
the four test stimuli, which is replaced on the screen by a visual mask diet ( ) onset for automatic and ( ) controlled processes. Re-
for a further 500 ms. sults are shown as mean  1 standard error of the mean.

Journal of Gastroenterology and Hepatology 2017; 32 (Suppl. 1): 90–93 91

© 2017 Journal of Gastroenterology and Hepatology Foundation and John Wiley & Sons Australia, Ltd
Cognitive impairment in coeliac disease
processing (Trail Making Test Part A), visual–spatial memory
(Rey–Osterrieth complex figure), and importantly, two markers
of CD severity—Marsh scores on duodenal histology and serum
levels of tissue transglutaminase. They also found significant cor-
relations between changes in both Marsh score and serum tissue
transglutaminase levels and changes in the speed of both automatic
and controlled processing (SCIT-RT), verbal fluency (Controlled
Oral Word Association Test, COWAT) and motor function (Trails
A) over the 12 months of the study (Fig. 3). The only cognitive
function that is common between these three cognitive tests is
the speed of processing and as such suggests that cognitive func-
tion, particularly the speed of processing, improves as intestinal
health improves. The authors concluded that the subjective reports
of brain fog symptoms are supported by empirical evidence of mi-
nor deficits to the processing speed and visual–spatial memory.18
These cognitive impairments, while subclinical in nature, do
have an impact on the daily life of CD patients. It should be noted
that the magnitude of the impairment on SCIT-RT at week 0 is
similar to that found in people with a blood alcohol level of
0.05 g/100 ml,14 which is the legal limit for driving in Australia.
It is also of equivalent magnitude to the impairment in SCIT-RT
observed by participants with moderate to severe jetlag (SR
Robinson & GW Yelland, unpublished data). More importantly,
these cognitive impairments are ameliorated by adopting a diet
that is free from gluten, although there is some evidence that this
may be age dependent. A study of 32 late-diagnosed CD patients
of 65 years and older found no improvement in cognitive function,
relative to healthy controls, as a function of a GFD.3
The most likely explanation for the existence of minor cognition
impairments (i.e., brain fog) in patients with untreated CD, which
can also account for white matter hyperintensities within the brains
of CD patients, is the existence of elevated levels of circulating cy-
tokines associated with the systemic inflammation associated with
CD. At elevated levels, circulating pro-inflammatory cytokines
bind to blood–brain barrier epithelial cells and facilitate the migra-
tion of leukocytes into the brain.19 Leukocytes promote inflamma-
tion in the brain, particularly inflammation of nerve fibers (i.e.,
Figure 3 Participants’ performance on four cognitive tests showed sig-
nificant correlations with each other and with histological scores and tis-
sue transglutaminase levels (*P < 0.05; **P < 0.001; Spearman’s rho).
Modified from Litchwark et al. (2014).18 COWAT, Controlled Oral Word
Association Test; RT, response time; SCIT, Subtle Cognitive Impairment
Test.
92
© 2017 Journal of Gastroenterology and Hepatology Foundation and John Wiley & Sons Australia, Ltd
GW Yelland
white matter hyperintensities),20 which, in turn, reduces the speed
of neural transmission, that is, reduces the speed of processing.4,20
Other mechanisms have been suggested to account for the minor
cognitive impairments associated with the brain fog experienced
by CD patients that are specific to the ingestion of gluten. How-
ever, the observation of brain fog in patients with multiple
sclersois,19 patients with fibromyalgia,21 and patients undergoing
chemotherapy20,22,23 suggests that the mechanism is not related
specifically to gluten ingestion. All of these disorders have one
thing in common with CD—they are associated with systemic in-
flammation. The case for systemic inflammation involvement has
been further strengthened by a recent study of patients with
Crohn’s disease (another inflammatory bowel disease) in which
impairments to the speed of processing (SCIT-RT) were observed
and the magnitude of the impairment was found to be associated
with markers of systemic inflammation.24
In summary, subjective reports of brain fog from CD patients
are both psychologically and neurologically real and can be quan-
tified when sufficiently sensitive cognitive tests are used. The cog-
nitive impairments associated with brain fog improve on a GFD.
The primary cognitive domain affected appears to be the speed
of processing, and its likely cause is the increase in the circulating
cytokines associated with systemic inflammation, resulting in
damage to or inflammation of neural fibers in the brain.
Acknowledgements
The author wishes to acknowledge the contributions of his collab-
orators over 8 years of research on the impact of gastrointestinal
disorders on cognitive function: Prof Stephen Robinson (School
of Health Sciences, RMIT University), Prof Peter Gibson (Depart-
ment of Gastroenterology, Monash University), and Dr Evan
Newnham (Department of Gastroenterology, Eastern Health and
Monash University). He would also like to extend special thanks
to Prof Peter Gibson, Dr Jane Muir, and the students of the Depart-
ment of Gastroenterology at the Alfred Centre, Monash Univer-
sity, for accepting him, a stray psychologist, within their team.
References
1 Currie S, Hadjivassiliou M, Clark MJ et al. Should we be ‘nervous’
about coeliac disease? Brain abnormalities in patients with coeliac
disease referred for neurological opinion. J. Neurol. Neurosurg.
Psychiatry 2012; 83: 1216–21.
2 Hadjivassiliou M, Sanders DS, Grunewald RA, Woodroofe N, Boscolo
S, Aeschlimann D. Gluten sensitivity: from gut to brain. Lancet Neurol.
2010; 9: 318–30.
3 Casella S, Zanini B, Lanzarotto F et al. Cognitive performance is
impaired in coeliac patients on gluten free diet: a case–control study in
patients older than 65 years of age. Dig. Liver Dis. 2012; 44: 729–35.
4 Debette S, Markus HS. The clinical importance of white matter
hyperintensities on brain magnetic resonance imaging: systematic
review and meta-analysis. BMJ 2010; 341: c3666.
5 Bilgic B, Aygun D, Arslan AB et al. Silent neurological involvement in
biopsy-defined coeliac patients. Neurol. Sci. 2013; 34: 2199–204.
6 McPhedran M. Identifying “brain fog”: Features, triggers, and
development of a psychometric tool for detection. [Research]: RMIT
University; 2014.
7 Hallert C, Astrom J. Intellectual ability of adults after lifelong intestinal
malabsorption due to coeliac disease. J. Neurol. Neurosurg. Psychiatry
1983; 46: 87–9.
Journal of Gastroenterology and Hepatology 2017; 32 (Suppl. 1): 90–93
GW Yelland
8 Burk K, Bosch S, Muller CA et al. Sporadic cerebellar ataxia
associated with gluten sensitivity. Brain 2001; 124: 1013–9.
9 Hu WT, Murray JA, Greenaway MC, Parisi JE, Josephs KA. Cognitive
impairment and celiac disease. Arch. Neurol. 2006; 63: 1440–6.
10 Yelland GW, Robinson SR, Friedman TW, Hutchison CW, inventors;
Yelland, G.W & Robinson, S.R, assignee. An automated method for
measuring cognitive impairment. Australia patent 2004203679. 2004.
11 Yelland GW, Robinson SR, Friedman TW, Hutchison CW,
inventors; Yelland, G.W. & Robinson, S.R., assignee. An
automated method for measuring cognitive impairment. Europe
patent 04700416.3 2004.
12 Bruce KM, Yelland GW, Smith JA, Robinson SR. Recovery of
cognitive function after coronary artery bypass graft operations. Ann.
Thorac. Surg. 2013; 95: 1306–13.
13 Bruce KM, Yelland GW, Almeida AA, Smith JA, Robinson SR.
Effects on cognition of conventional and robotically assisted cardiac
valve operation. Ann. Thorac. Surg. 2014; 97: 48–55.
14 Friedman TW, Robinson SR, Yelland GW. Impaired perceptual
judgment at low blood alcohol concentrations. Alcohol 2011; 45:
711–18.
15 Friedman TW, Yelland GW, Robinson SR. Subtle cognitive
impairment in elders with Mini-Mental State Examination scores
within the ‘normal’ range. Int. J. Geriatr. Psychiatry 2012; 27: 463–71.
16 Speirs SJ, Rinehart NJ, Robinson SR, Tonge BJ, Yelland GW. Efficacy
of cognitive processes in young people with high-functioning autism
spectrum disorder using a novel visual information-processing task.
J. Autism Dev. Disord. 2014; 44: 2809–19.
Journal of Gastroenterology and Hepatology 2017; 32 (Suppl. 1): 90–93
© 2017 Journal of Gastroenterology and Hepatology Foundation and John Wiley & Sons Australia, Ltd
Cognitive impairment in coeliac disease
17 Bruce KM, Robinson SR, Smith JA, Yelland GW. Validity of a
screening tool for detecting subtle cognitive impairment in the middle-
aged and elderly. Clin. Interv. Aging 2014; 9: 2165–76.
18 Lichtwark IT, Newnham ED, Robinson SR et al. Cognitive impairment
in coeliac disease improves on a gluten-free diet and correlates with
histological and serological indices of disease severity. Aliment.
Pharmacol. Ther. 2014; 40: 160–70.
19 Larochelle C, Alvarez JI, Prat A. How do immune cells overcome the
blood–brain barrier in multiple sclerosis? FEBS Lett. 2011; 585: 3770–
80.
20 Seigers R, Fardell JE. Neurobiological basis of chemotherapy-induced
cognitive impairment: a review of rodent research. Neurosci. Biobehav.
Rev. 2011; 35: 729–41.
21 Glass JM, Williams DA, Fernandez-Sanchez ML et al. Executive
function in chronic pain patients and healthy controls: different cortical
activation during response inhibition in fibromyalgia. J. Pain 2011; 12:
1219–29.
22 Argyriou AA, Assimakopoulos K, Iconomou G, Giannakopoulou F,
Kalofonos HP. Either called “chemobrain” or “chemofog,” the long-
term chemotherapy-induced cognitive decline in cancer survivors is
real. J. Pain Symptom Manag. 2011; 41: 126–39.
23 Raffa RB. A proposed mechanism for chemotherapy-related cognitive
impairment (‘chemo-fog’). J. Clin. Pharm. Ther. 2011; 36: 257–9.
24 van Langenberg DR, Yelland GW, Robinson SR, Gibson PR.
Cognitive dysfunction in Crohn’s disease is associated with indices
of inflammation. United European Gastroenterology Journal. 2016;
DOI: 10.1177/2050640616663397
93