Journal of Diabetes and Its Complications 32 (2018) 1040–1045

Contents lists available at ScienceDirect

Journal of Diabetes and Its Complications

journal homepage: WWW.JDCJOURNAL.COM

Impaired cognitive processing speed in type 1 diabetic patients who had

severe/recurrent hypoglycaemia
Stefano Bortolotti a,1, Lisa Zarantonello a,1, Ambra Uliana b, Nicola Vitturi b, Sami Schiff a, Patrizia Bisiacchi c,
Angelo Avogaro b, Piero Amodio a,⁎,2, Alberto Maran b,2
a
Unit of Internal Medicine 5, Department of Medicine, University of Padova, Padova 35128, Italy
b
Unit of Metabolic Disease, Department of Medicine, University of Padova, Padova 35128, Italy
c
Department General Psychology and CIRMANMEC, University of Padova, Padova 35128, Italy

a r t i c l e i n f o a b s t r a c t

Article history: Aims: To detect whether adults with type 1 diabetes mellitus (T1DM) have lower cognitive performance than
Received 10 April 2018 healthy individuals and to detect risk factors for low cognitive performance.
Received in revised form 16 July 2018 Methods: Twenty-six adults with T1DM and twenty-six healthy subjects matched for age, gender and educational
Accepted 6 August 2018 level were compared for cognitive performance by a chronometric computerized test measuring visuo-spatial
Available online 10 August 2018
working memory (N-Back) and by two validated neuropsychological tests (Mini Mental State Examination, An-
imal Naming Test). Clinical data about diabetes duration, average daily insulin dosage, glycated haemoglobin, ret-
Keywords:
Diabetes type 1
inopathy, urine albumin-creatinine ratio, previous hypoglycaemic coma and awareness of hypoglycaemia were
Awareness of hypoglycaemia obtained from medical records. Basal pre-test glycemia and blood pressure were measured for each patient.
Unawareness of hypoglycaemia Results: No differences were found between patients (n = 26) and healthy controls (n = 26) in neuropsycholog-
Cognition ical tests. Within diabetic patients, those with impaired awareness of hypoglycaemia (n = 7) or history of coma
Cognitive processing speed in the recent 1–3 years (n = 5) had psychomotor slowing at the N-Back test (592 ± 35 vs. 452 ± 21 ms and
619 ± 40 vs. 462 ± 19 ms, respectively; both p b 0.01). The variables related to diabetic severity did not show
a relationship with reaction times of the N-Back test.
Conclusion: Psychomotor speed slowing is detectable in patients with T1DM who have a history of previous
hypoglycaemic episodes or coma.
© 2018 Elsevier Inc. All rights reserved.

1. Introduction sustained attention, executive functions, general intelligence and

T1DM is characterized by acute and chronic complications. The
main long-term complication is diabetic angiopathy, which is
classified as microvascular (retinopathy, nephropathy, neuropathy)
and macrovascular (atherosclerosis). Neurological sequelae can be
neuropathic (distal symmetric polyneuropathy, autonomic neuropathy,
mononeuropathy) or cognitive (depression, cognitive decline).
Evidence about type 1 diabetes-associated cognitive decline (DACD) is
conflicting, both on magnitude of impairment and on which cognitive
domains are involved. The most common deficits consist in the slowing
of cognitive processing speed (CPS) and the worsening of mental
flexibility. 1,2 The specific nature of the deterioration of memory,
Declaration of interest: The authors declare that they have no conflicts of interest rel-
evant to this article.
⁎ Corresponding author at: Dipartimento di Medicina – DIMED – and CIRMANMEC, Via
Giustiniani 2, 35128 Padova, Italy.
E-mail address: piero.amodio@unipd.it (P. Amodio).
1
Shared first authorship.
2
Shared senior authorship.
visuoconstruction is uncertain, and the pathogenesis of DACD is still
unknown.
The main variables affecting the risk appear to be the age of onset of
diabetes, the development of microvascular complications and the level
of glycaemic control (hypoglycaemia or hyperglycaemia). Early onset of
diabetes (4–7 years of age) is linked to greater neurocognitive dysfunc-
tions and structural alterations than late onset. This event is probably
due both to the duration of diabetes and to its effects on neuro-
development.3,4 In adults with T1DM, the occurrence of microvascular
complications is related to increased and early cognitive decline. 5,6
Moreover, in some neuroimaging studies, type 1 diabetic patients with
retinopathy have shown an augmented focal cortical atrophy and a re-
duced white matter volume. 7,8 Longitudinal studies have demonstrated
that patients with elevated glycated haemoglobin (N8%) present worse
cognitive efficiency and velocity than subjects with good glycaemic
control.9,10 These deficits seem to be reversed by improved metabolic
control. The intensive use of insulin to prevent hyperglycaemia and its
complications has incremented the incidence of hypoglycaemic crises.
Patients treated with an intensive therapeutic regimen, especially

https://doi.org/10.1016/j.jdiacomp.2018.08.005
1056-8727/© 2018 Elsevier Inc. All rights reserved.
 S. Bortolotti et al. / Journal of Diabetes and Its Complications 32 (2018) 1040–1045 1041

those with glycated haemoglobin near the non-diabetic range, have a 3 Table 1
times greater risk of severe hypoglycaemia than subjects with a less in- Demographic characteristics, neuropsychological and behavioural performance of the
study participants.
tensive insulin therapy. Subjects affected by T1DM have on average 2
episodes of mild hypoglycaemia a week; about one third of them pres- Patients with Control p
ent a severe episode annually. 11 DM1 subjects

The severe detrimental effects of acute hypoglycaemia on cognitive N 26 26 –

performance, especially on memory and executive functions, are widely Age (years) 45.8 ± 12.6 45.9 ± 12.8 0.97
Gender (M/F) 17/9 17/9 –
known. However, the role of hypoglycaemia and its complications on
Education (years) 13.4 ± 3.8 13.9 ± 3.7 0.63
long-term cognitive decline are still uncertain. Some studies support Humeral arterial Pressure Diastolic 75 ± 5 – –
the hypothesis that the brain may adapt to hypoglycaemic insult.5,12,13 (mm Hg) Systolic 129 ± 9 – –
Other studies, in contrast, argue that hypoglycaemia can sensitize neu- Mean 93 ± 5 – –
rons, increasing the negative effects of glycopenia.14 Recurrent episodes Glycaemia (mg/dl) 130 ± 34 – –
DM1 duration (years) 18 ± 10 – –
of hypoglycaemia cause both a down-regulation of counter-regulatory
Insulin (U/die) 49 ± 21 – –
response (hypoglycaemia-associated autonomic failure–HAAS) and a HbA1c (%) 7.3 ± 0.6 – –
reduction of autonomic hypoglycaemic symptoms (resulting in un- HbA1c (mmol/mol) 56 ± 6
awareness of hypoglycaemia), which are associated with a sixfold in- Retinopathy Absent 46% – –
Simple 39% – –
crease in the risk of severe hypoglycaemia and a prolonged encephalic
Proliferative 15% – –
exposure to glycopenic insult.15,16 Perros et al.17 underlined that alter- Urine albumin/creatinine (mg/mMol) 6 ± 3.5 – –
ation of EEG, induced by hypoglycaemia can become permanent after se- Previous hypoglycaemic coma 19% – –
vere or recurrent crises. Increased brain susceptibility to hypoglycaemia Unawareness of hypoglycaemia 27% – –
seems to occur especially from birth to 20 years of age. 4,18 However, CriQ Education* 115 ± 21 101 ± 13 0.03
Work 109 ± 15 111 ± 16 0.66
at a later age also, the brain may become unable to cope with
Free-time 118 ± 23 110 ± 16 0.2
hypoglycaemia. activity
Thus, since i) the occurrence of cognitive alterations in T1DM and Total 119 ± 24 108 ± 17 0.13
the role that hypoglycaemia may have in their pathophysiology is de- MMSE 29.4 ± 0.9 29.5 ± 0.9 0.97
ANT 24 ± 6 23 ± 5 0.6
bated and ii) there are few data about the cognitive sequelae of individ-
N-Back condition (ms) 0-Back 425 ± 93 385 ± 58 0.07
uals with or without hypoglycaemia awareness, we designed a study to 1-Back 450 ± 109 409 ± 94 0.16
detect whether adults with T1DM have lower cognitive performance 2-Back 593 ± 131 557 ± 148 0.37
than healthy individuals, as well as to investigate the risk factors for Mean 490 ± 31 451 ± 31 0.08
low cognitive performance in type 1 diabetic subjects. Data are mean ± SD; *, statistical significance; Significance level was p b 0.05.

2. Subjects, materials and methods 2.2. Neuropsychological evaluation

2.1. Subjects and data
This study involved 26 participants with T1DM and 26 healthy con-
trol subjects matched for age, gender and educational level. The demo-
graphic characteristics of the two groups are shown in Table 1. Diabetic
patients were recruited from the Diabetes Clinic at the University Hospi-
tal of Padua where they were seen regularly each month. All partici-
pants attended two sessions: in the first, they provided information
about their medical history, and then they submitted to preliminary
neuropsychological tests (Mini Mental State Examination, Back Depres-
sion Inventory-II, Cognitive Reserve Index Questionnaire, Animal Nam-
ing Test) to verify the presence of exclusion criteria. Exclusion criteria
were: serious systemic disease (heart failure, chronic kidney disease,
cirrhosis, severe lungs and airways diseases, addiction), cerebrovascular
disease, coronary artery disease, severe neuropathy, major psychiatric
disorders and use of tranquillizer, antidepressant or anti-psychotic
drugs. Based on tests administered during preliminary evaluation, par-
ticipants were also excluded if they had cognitive decline (Mini Mental
State Examination b26 and Animal Naming Test ≤14) or depression
(Back Depression Inventory N13). Clinical data about diabetes duration
(years from onset of the disease), average daily insulin dosage (U/die),
glycated haemoglobin (HbA1c – %), presence of retinopathy (absent,
proliferative, non-proliferative), urine albumin-creatinine ratio (mg/
mMol), previous hypoglycaemic coma and awareness of hypoglycaemia
were recorded for each patient. In the second session, participants per-
formed the computerized N-Back test. Pre-test glycemia (mg/dl) and
pre-test blood pressure (mm Hg) were measured for each diabetic pa-
tient by glucometer and sphygmomanometer, respectively. The proto-
col was reviewed and approved by the Institutional Review Board of
University of Padova. The purpose and the nature of the study were ex-
plained to all subjects, and informed written consent was obtained be-
fore their participation.
• Mini Mental State Examination (MMSE): it assesses cognitive de-
cline. The MMSE explores: spatial and temporal orientation, learn-
ing, calculation, comprehension (written and oral), writing and
constructional apraxia.19
• Beck Depression Inventory – II (BDI-II): a self-administered ques-
tionnaire that identifies depressive symptoms and investigates the
depression level.20
• Cognitive Reserve Index Questionnaire (Cri-q): a test that evaluates
cognitive reserve by educational level (school years and extra-
scholastic courses), work and free-time activities. 21
• Animal Naming Test (ANT): a verbal fluency test in which partici-
pants have to name as many animals as possible in 1 min. The
total sum of animals correctly reported (excluding errors of intru-
sion or perseveration) represents the final score.22
2.3. Cognitive test protocol
• N-Back Test: a computerized chronometric test providing insight
into both CPS in a go/no-go condition and working memory
(WM). The task was an adapted version of the visuospatial N-Back
test from Cui et al.23 and Haberecht et al.24 The stimuli were admin-
istered using a personal computer (17-in. display; resolution: 1024
× 768 pixels) by E-Prime software (Psychology Software Tools, Pitts-
burgh, PA). The blank screen was divided into 9 equal parts by a
black grid during the test. The stimulus was the letter O (66
× 156 pixels), which appeared in one of the grid's sections for
1500 ms. After this period, there was a fixed inter-stimulus interval
of 500 ms in which only the grid was displayed. The position of the
letter changed in each trial. The test was block design, and it was
composed of three conditions with increasing working memory
load. During the control condition (0-Back), participants had to
press the response button if the letter O appeared in the central
1042 S. Bortolotti et al. / Journal of Diabetes and Its Complications 32 (2018) 1040–1045

part of the grid. This is a simple RT; however, it requires the inhibi-
tion of the response if the letter O is not in the centre of the screen,
and thus it reflects both a go/no-go task and a control condition for
WM, which is not required for this task. In the 1-Back condition
(lower working memory load), subjects were required to respond
when the stimulus appeared in the same position as in the previous
trial. Finally, in the 2-Back condition (higher working memory load),
the instruction was to press the response button every time the let-
ter appeared in the same position as two trials before. One block was
composed of 6 sub-blocks for the 1-Back alternate, with 6 sub-
blocks for 0-Back. The other block was composed of 6 sub-blocks
for the 2-Back condition as well as 0-Back. Before each block, there
was a practice block in which participants performed the task until
they understood it completely. The order in which 1-Back and 2-
Back blocks were administered was counterbalanced among partici-
pants to prevent confounding effects (learning and mental fatigue).
Each condition was formed by 23 stimuli, 7 of which were targets
and 16 were non-targets, for a total of 552 stimuli, 168 of which
were targets and 384 were non-targets. The percentage of targets
(33% of total stimuli) was decided based on the study by Jaeggi et al.25
2.4. Statistical analysis
Results are expressed as mean ± SD, unless otherwise indicated. The
comparison between groups (patients vs. controls, awareness vs. un-
awareness of hypoglycaemia, previous vs. no previous coma) was per-
formed by general linear model, using age and education years as
covariates. Between factors were the reaction times for different condi-
tions of the N-Back test (0-Back, 1-Back, 2-Back). The reaction time of
the N-Back were adjusted for the accuracy and expressed as weighted
RT (RTW) in accordance with the following formula: RTW = RT + [RT
(1 − Accuracy)]. This was done to overcome the problem posed by
the speed-accuracy trade-off. Post hoc comparison of more than two
variables was performed using the Newman-Keuls correction for multi-
ple comparison. Statistical analysis was performed by TIBCO Software
Inc. (2017), Statistica (data analysis software system), version 13.
http://statistica.io.
3. Results
The demographic and neuropsychological variables were compara-
ble in participants with and without diabetes, except for Cri-Education
(Table 1). The RTW of the N-Back task increased with memory load in
both groups (ANOVA: N-Back condition (0,1,2) - F2,94 = 6.1, p b 0.01).
The education level improved the efficacy of WM, since the higher the
education level, the lower the increment of RTW from the 0-Back condi-
tion to the 2-Back condition in both groups (ANOVA: interaction N-Back
condition (0,1,2) × Education - F2,94 = 6.1, p b 0.01). Age influenced the
RTW homogeneously in all conditions (F1,47 = 11,02 p b 0.01; interac-
tion age × N-Back condition (0,1,2) - F2,94 = 2.31 p = 0.10).
Thus, all the analyses were adjusted by age and education.
The performance of the N-Back task was comparable in diabetic pa-
tients and controls (F1,47 = 3.14, p = 0.08); however, a trend for cogni-
tive processing slowing in patients with T1DM in the 0-Back condition
was detectable (Table 1).
Patients with T1DM and unawareness of hypoglycaemia had
prolonged RTW in each condition compared both to patients aware of
hypoglycaemia and control subjects (F2,45 = 5.8, p b 0.01; Table 2, Fig.
1a). The effect of WM load on RTW was similar for the three subgroups
(control subjects, aware patients, unaware patients): interaction
subgroups × N-Back conditions (0,1,2) - F4,92 = 0.11, p = 0.98.
Likewise, hypoglycaemic coma was associated with cognitive pro-
cessing slowing, both compared to patients who had not had previous
hypoglycaemic coma and controls (F2,46 = 5.9, p b 0.01), with an effect
of WM load similar in the three groups (interaction with the N-Back
condition (0,1,2) - F4,92 = 0.28, p = 0.9) (Table 2, Fig. 1b).

Table 2
Demographic and clinical characteristics, neurophysiological and behavioural evaluation of patients.

Patients with DM1

Hypoglycaemia History of hypoglycaemic coma

Unawareness Awareness Yes No

N 7 19 5 21
Age (years)*, † 59 ± 8 41 ± 11 58 ± 10 43 ± 12
Gender (M/F) 4/3 13/6 2/3 15/6
Education (years) 13 ± 5.2 13 ± 3.2 11 ± 4,5 14 ± 3,4
Humeral arterial pressure (mm Hg) Diastolic 74 ± 5 75 ± 5 74 ± 5 76 ± 5
Systolic 131 ± 8 129 ± 9 132 ± 10 129 ± 9
Mean 93 ± 5 93 ± 4 93 ± 6 93 ± 4
Glycaemia (mg/dl) 121 ± 32 133 ± 35 128 ± 32 130 ± 35
DM1 duration (years) 21 ± 16 18 ± 8 24 ± 18 17 ± 8
Insulin (U/die) 45 ± 20 52 ± 21 41 ± 23 50 ± 21
HbA1c (%) 7.2 ± 0.6 7.3 ± 0.6 7.2 ± 0.7 7.3 ± 0.6
HbA1c (mmol/mol) 55 ± 6 56 ± 6 55 ± 7 56 ± 6
Retinopathy Absent 14% 58% 20% 52%
Simple 86% 21% 80% 29%
Proliferative 0% 21% 0% 19%
Urine albumin/creatinine (mg/mMol) 6.8 ± 4.5 5.6 ± 3.4 8.7 ± 3.8 5.2 ± 3.3
Previous hypoglycaemic coma 72% 0% – –
Unawareness of hypoglycaemia – – 100% 10%
CriQ Education 110 ± 25 117 ± 20 107 ± 28 117 ± 20
Work 109 ± 17 108 ± 15 102 ± 8 111 ± 17
Free-Time 118 ± 39 118 ± 13 109 ± 38 121 ± 16
Total 117 ± 36 119 ± 16 106 ± 32 123 ± 20
MMSE 29 ± 0.9 29 ± 0.9 29 ± 1.1 29 ± 0.8
ANT 27 ± 5 23 ± 6 26 ± 5 24 ± 6
N-Back condition (ms) 0-Back*, † 512 ± 99 393 ± 69 496 ± 46 408 ± 94
1-Back*, † 546 ± 119 412 ± 80 556 ± 104 423 ± 95
2-Back*, † 702 ± 149 552 ± 100 767 ± 113 551 ± 98
Mean*, † 592 ± 35 452 ± 21 619 ± 40 462 ± 19

Data are mean ± SD; ⁎, p b 0.05: Awareness vs. Unawareness of hypoglycaemia; †, p b 0.05 history of hypoglycaemic coma vs. no history of hypoglycaemic coma. Free-Time = Free-Time
Activities. The interval between previous hypoglycaemic coma and psychometric testing was 1–4 years.
 S. Bortolotti et al. / Journal of Diabetes and Its Complications 32 (2018) 1040–1045 1043

Fig. 1. Left panel: patients with unawareness of hypoglycaemia displayed slower RTW than patients with awareness of hypoglycaemia and controls; right panel: patients who had
hypoglycaemic coma displayed slower RTW than patients who did not have hypoglycaemic coma and controls. The bars represent confidence intervals.

The mean arterial pressure (MAP) of diabetic patients showed an in-
verse correlation with RTW during the 2-Back condition; thus, the lower
the MAP, the longer the RTW (r = 0.435, p = 0.03).
The variables related to diabetic severity (pre-test glycemia, T1DM
duration, average daily insulin dosage, HbA1c, retinopathy, urine
albumin-creatinine ratio) did not show a relationship with the RTW of
the N-Back task (Table 3).
Of note, no relationship was found between the N-Back and the ANT
(p = 0.9), or between the N-Back and the MMSE (p = 0.3).
4. Discussion
Our study has shown that patients with T1DM who suffer from un-
awareness of hypoglycaemia as well as patients who have had previous
episodes of hypoglycaemic coma, present cognitive processing slowing
on accurate chronometric neuropsychological investigation, despite the
fact that the MMSE and a semantic memory task did not reveal cognitive
dysfunction. In contrast, no relationship was found between psycho-
metric tests and indexes of poor glycaemic control. Thus, our data sup-
port the view that overly strict control of diabetes, exposing patients to
hypoglycaemic episodes, selectively impairs cognition in patients with
T1DM.
The group of patients in the study was well representative of the
adult type 1 diabetic population. The average age of the sample was
46 years (maximum: 69; minimum: 25 years of age). Some 46% of the
patients had a high-school degree, whereas 31% of them had an aca-
demic degree. There was a slight prevalence of male gender (65%) in
the sample. This is not surprising as T1DM, despite being an immune-
mediated pathology, is more common in men and boys.26 Some recent
longitudinal studies have shown that, since the spreading of the
Table 3
Summary results of repeated measures ANOVA for variables related to diabetic severity for
the three conditions of the N-Back.
F p
DM1 duration 0.3
Average daily insulin dosage 0.2
A/C ratio 0.3
Retinopathy 2.9
Basal glycaemia 0.07
HbA1c 0.47
intensive insulin regimen in diabetic population, the incidence of
long-term complications is lower than in the past. Even if it is difficult
to quantify correctly the prevalence and incidence of T1DM complica-
tions, the cumulative incidence of proliferative retinopathy, nephropa-
thy and cardiovascular disease after 30 years from the diagnosis is
supposed to be 21%, 9% and 9%, respectively.27 Fifteen percent of the pa-
tients enrolled have a proliferative retinopathy, whereas the study has
ruled out patients with overt nephropathy and cardiovascular disease
to prevent confounding effects of these complications on cognitive
performance.
Both in controls and in diabetic subjects the RTW of the N-Back task
had the expected shape: the RTW in the lower cognitive load condition
(0-Back) were significantly shorter than those in the 1-Back condition
which, in turn, were shorter than those in the 2-Back condition. This is
in accordance with previous studies.28,29
In the patients group, as well as in control subjects, high educational
level (N16 years) facilitates the mnestic task. It appears that education is
a protective factor for cognitive efficiency in diabetic patients and in
healthy people. This hypothesis is credible, given the numerous studies
demonstrating the role of education in cognitive performance. 30,31
The finding that patients with hypoglycaemia unawareness showed
a reduction in CPS, compared to those with hypoglycaemia awareness,
supports the theory of a detrimental effect of recurrent hypoglycaemia
on cognitive functions. 32 The fact that cognitive performances on
MMSE and ANT were comparable between healthy subjects and dia-
betic patients, as well as between patients with or without awareness
of hypoglycaemia, highlights that only sensitive cognitive tools oriented
to studying specific domains vulnerable to hypoglycaemia are able to
detect the phenomenon.
This finding is not surprising, as paper and pencil tests are adequate
only for detecting gross cognitive deficits, while the highly accurate
chronometric N-Back test, used in this study, is oriented to measuring
a potentially vulnerable area with more sensitive results. Thus, our in-
vestigation has pointed out the importance of using the most precisely
chosen methods in the study of cognitive impairment in T1DM to pre-
vent misleading conclusions.33,34
Cognitive processing slowing affects actions that require mental alert-
ness, like driving or practicing sports, while intellectual activities, which
0.57
0.63
can be performed more slowly, might not be involved at all. However, it
0.59 cannot be overlooked that the cognitive slowing showed by patients
0.07 with unaware hypoglycaemia, even if moderate, impairs everyday life.
0.86 As regards patients with previous hypoglycaemic coma, they have
0.50
shown longer RTW compared to subjects without a history of coma.
1044 S. Bortolotti et al. / Journal of Diabetes and Its Complications 32 (2018) 1040–1045
This result is in agreement with a previous study conducted by our
centre35 and other data in the literature 36 that showed permanent cere-
bral sequelae caused by severe hypoglycaemia.
When considering all of these issues, our data do not support the
hypothesis that recurrent severe hypoglycaemia episodes do not
have long-term adverse effects on the cognitive capacity of patients
with T1DM. 9 An explanation to this discrepancy can be given by
our data showing that MMSE and ANT were not influenced by a his-
tory of hypoglycaemia because of their probable lack of sensitivity
and the selectivity of the cognitive impairment produced by
hypoglycaemia.
The main strengths of this study consist of: (1) the use of a highly
sensitive chronometric test (N-Back), which investigates cognitive pro-
cessing slowing and WM, in addition to basic cognitive tests to study
DACD; (2) a rigorous definition of hypoglycaemia unawareness, deter-
mined using validated methods (Clarke and Gold Scores37,38), and of se-
vere hypoglycaemia (which only includes subjects with previous
hypoglycaemic coma); and (3) the usage of strict exclusion criteria
that have ruled out patients with nephropathies, cardiovascular disease,
depression and other comorbidities, to ensure that healthy volunteers
are only compared with patients with the best cognitive performance.
The limitation in our study is to be found in the relatively small number
of cases observed, which partly restricts analysis of subgroups. How-
ever, the effect of recurrent hypoglycaemia seems to be sufficiently rel-
evant to allow for the identification of cognitive changes even with a
small sample size. It would be important, for the future, to extend this
study to a less select patient population and to validate the findings
with an independent group of diabetic patients.
5. Conclusion
In conclusion, our study showed that a subset of patients with T1DM
were found to have neuropsychological defects mainly regarding CPS.
This alteration was related to a history of previous hypoglycaemic epi-
sodes and coma; in contrast, complications related to poor glycaemic
control did not have any relationship to cognitive dysfunction, at least
in the observed individuals.
Author contributions
S.B. and L.Z. collected/analysed data, wrote the manuscript and con-
tributed to the discussion. A.U. and N.V. collected data. S.S. collected data
and contributed to the discussion. P.B. contributed to the discussion and
edited the manuscript. P.A. and A.M. designed the study, analysed the
data, contributed to the discussion and reviewed/edited the manuscript.
A.A. contributed to discussion.
Additional information
P.A. is the corresponding author and guarantor of this work and, as
such, had full access to all the data in the study and takes responsibility
for the integrity of the data and the accuracy of the data analysis.
Funding
This research did not receive any specific grant from funding agen-
cies in the public, commercial, or not-for-profit sectors. It was funded
by Padova University funding to P.A.
References
1. Mijnhout GS, Scheltens P, Diamant M, et al. Diabetic encephalopathy: a concept in
need of a definition. [1]Diabetologia 2006;49:1447-8, https://doi.org/10.1007/
s00125-006-0221-8.
2. Brands AMA, Biessels GJ, DeHaan EH, Kappelle LJ, Kessels RP. The effects of type 1 di-
abetes on cognitive performance. Diabetes Care 2005;28:726-35.
3. Gaudieri PA, Chen R, Greer TF, Holmes CS. Cognitive function in children with type 1
diabetes. Diabetes Care 2008;31:1892-7, https://doi.org/10.2337/dc07-2132.
4. Biessels GJ, Deary IJ, Ryan CM. Cognition and diabetes: a lifespan perspective. Lancet
Neurol 2008;7:184-90, https://doi.org/10.1016/S1474-4422(08)70021-8.
5. Ferguson SC, Blane A, Perros P, et al. Cognitive ability and brain structure in type 1
diabetes: relation to microangiopathy and preceding severe hypoglycemia. Diabetes
2003;52:149-56, https://doi.org/10.2337/diabetes.52.1.149.
6. Ryan CM, Geckle MO, Orchard TJ. Cognitive efficiency declines over time in adults
with type 1 diabetes: effects of micro- and macrovascular complications.
Diabetologia 2003;46:940-8, https://doi.org/10.1007/s00125-003-1128-2.
7. Wessels AM, Simsek S, Remijnse PL, et al. Voxel-based morphometry demonstrates
reduced grey matter density on brain MRI in patients with diabetic retinopathy.
Diabetologia 2006;49:2474-80, https://doi.org/10.1007/s00125-006-0283-7.
8. Musen G, Lyoo IK, Sparks CR, et al. Effects of type 1 diabetes on gray matter density as
measured by voxel-based morphometry. Diabetes 2006;55:326-33.
9. DCCT study research group D. Long-term effect of diabetes and its treatment on cognitive
function. N Engl J Med 2007;356(18):1842-52. [http://www.nejm.org/doi/pdf/10.1056/
NEJMoa066397].
10. Reichard P, Pihl M, Rosenqvist U, Sule J. Complications in IDDM are caused by ele-
vated blood glucose level: the Stockholm diabetes intervention study (SDIS) at 10-
year follow up. Diabetologia 1996;39:1483-8.
11. Frier BM. Hypoglycaemia in diabetes mellitus: epidemiology and clinical implica-
tions. Nat Rev Endocrinol 2014;10:711-22, https://doi.org/10.1038/nrendo.2014.170.
12. Boyle PJ, Nagy RJ, O'Connor AM, Kempers SF, Yeo RA, Qualls C. Adaptation in brain glu-
cose uptake following recurrent hypoglycemia. Proc Natl Acad Sci U S A 1994;91:9352-6.
13. McCrimmon R, Öz G. Cerebral adaptation to recurrent hypoglycemia. 2012;3:89-114.
14. Brismar T, Hyllienmark L, Ekberg K, Johansson B-L. Loss of temporal lobe beta power
in young adults with type 1 diabetes mellitus. Neuroreport 2002;13:2469-73, https://
doi.org/10.1097/01.wnr.0000047688.08940.ab.
15. Seaquist ER, Anderson J, Childs B, et al. Hypoglycemia and diabetes: a report of a
workgroup of the American Diabetes Association and the Endocrine Society. Diabetes
Care 2013;36:1384-95, https://doi.org/10.2337/dc12-2480.
16. Graveling AJ, Frier BM. Altération de la perception des hypoglycémies. Revue générale. Di-
abetes Metab 2010;36:S64-74, https://doi.org/10.1016/S1262-3636(10)70470-5.
17. Perros P, Bjorgaas M. In: Frier BM, Heller SR, McCrimmon RJ, eds. Hypoglycaemia in
Clinical Diabetes. 3rd edn. Wiley Blackwell; 2014. p. 305-22.
18. Blasetti A, Chiuri RM, Tocco AM, et al. The effect of recurrent severe hypoglycemia on
cognitive performance in children with type 1 diabetes: a meta-analysis. J Child
Neurol 2011;26:1383-91, https://doi.org/10.1177/0883073811406730.
19. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”: a practical method for grading
the cognitive state of patients for the clinician. J Psychiatr Res 1975;12:189-98.
20. Beck AT. Depression: Clinical, Experimental and Theoretical Aspects. University of Penn-
sylvania Press. 1967.
21. Nucci M, Mapelli D, Mondini S. Cognitive reserve index questionnaire (CRIq): a new
instrument for measuring cognitive reserve. Aging Clin Exp Res 2012;24:218-26,
https://doi.org/10.3275/7800.
22. Sager MA, Hermann BP, La Rue A, Woodard JL. Screening for dementia in community-
based memory clinics. Wis Med J 2006;105:25-9.
23. Cui X, Bray S, Bryant DM, Glover GH, Reiss AL. A quantitative comparison of NIRS and
fMRI across multiple cognitive tasks. Neuroimage 2011;54:2808-21, https://doi.
org/10.1016/j.neuroimage.2010.10.069.
24. Haberecht MF, Menon V, Warsofsky IS, et al. Functional neuroanatomy of visuo-
spatial working memory in turner syndrome. Hum Brain Mapp 2001;14:96-107.
25. Jaeggi SM, Buschkuehl M, Perrig WJ, Meier B. The concurrent validity of the N-back
task as a working memory measure. Memory 2010;18:394-412, https://doi.
org/10.1080/09658211003702171.
26. Atkinson MA, Eisenbarth GS, Michels AW. Type 1 diabetes. Lancet 2014;383:69-82,
https://doi.org/10.1016/S0140-6736(13)60591-7.
27. Nathan DM, Zimman B, Cleary PA, Backlund JYC, Genuth S, Miller R, et al. Modern-day
clinical course of type 1 diabetes mellitus after 30 years' duration. Arch Intern Med
2009;169:1307-16.
28. Luu P, Caggiano DM, Geyer A, Lewis J, Cohn J, Tucker DM. Time-course of cortical net-
works involved in working memory. Front Hum Neurosci 2014;8:1-15, https://doi.
org/10.3389/fnhum.2014.00004.
29. Chen Y-N, Mitra S, Schlaghecken F. Sub-processes of working memory in the N-back
task: an investigation using ERPs. Clin Neurophysiol 2008;119:1546-59, https://doi.
org/10.1016/j.clinph.2008.03.003.
30. Kaufman AS, Reynolds CR, McLean JE. Age and WAIS-R intelligence in a national sample of
adults in the 20- to 74-year age range: a cross-sectional analysis with educational level con-
trolled. Dermatol Int 1989;13:235-53, https://doi.org/10.1016/0160-2896(89)90020-2.
31. Berryhill ME, Jones KT. tDCS selectively improves working memory in older adults
with more education. Neurosci Lett 2012;521:148-51, https://doi.org/10.1016/J.
NEULET.2012.05.074.
32. Hansen TI, Olsen SE, Haferstrom ECD, et al. Cognitive deficits associated with im-
paired awareness of hypoglycaemia in type 1 diabetes. Diabetologia 2017;60:
971-9, https://doi.org/10.1007/s00125-017-4233-3. [Internet].
33. Koski L, Brouillette MJ, Lalonde R, et al. Computerized testing augments pencil-and-
paper tasks in measuring HIV-associated mild cognitive impairment. HIV Med
2011;12:472-80, https://doi.org/10.1111/j.1468-1293.2010.00910.x.
34. Silbert BS, Maruff P, Evered LA, et al. Detection of cognitive decline after coronary sur-
gery: a comparison of computerized and conventional tests. Br J Anaesth 2004;92:
814-20, https://doi.org/10.1093/bja/aeh157.
35. Maran A, Crepaldi C, Del Piccolo F, et al. Cognitive, neurophysiologic and metabolic
sequelae of previous hypoglycemic coma revealed by hyperinsulinemic-
hypoglycemic clamp in type 1 diabetic patients. Metab Brain Dis 2017;32:1543-51,
https://doi.org/10.1007/s11011-017-0041-1.
 S. Bortolotti et al. / Journal of Diabetes and Its Complications 32 (2018) 1040–1045 1045

36. Tonoli C, Heyman E, Roelands B, et al. Type 1 diabetes-associated cognitive decline: a
meta-analysis and update of the current literature. J Diabetes 2014;6:499-513, https:
//doi.org/10.1111/1753-0407.12193.
37. Clarke WL, Cox DJ, Gonder-Frederick LA, Julian D, Schlundt D, Polonsky W. Reduced
awareness of hypoglycemia in adults with IDDM: a prospective study of
hypoglycemic frequency and associated symptoms. Diabetes Care 1995;18:517-22,
https://doi.org/10.2337/diacare.18.4.517.
38. Gold AE, Macleod KM, Frier BM. Frequency of severe hypoglycemia in patients with
type I diabetes with impaired awareness of hypoglycemia. Diabetes Care 1994;17:
697-703, https://doi.org/10.2337/diacare.17.7.697.