REVIEW
SA JOURNAL OF DIABETES & VASCULAR DISEASE
102
VOLUME 7 NUMBER 3 • SEPTEMBER 2010
Metabolic memory in type 1 diabetes
AlEx D WrIghT
Abstract
M
etabolic memory and its possible mechanisms are
reviewed. In clinical practice in type 1 diabetes the
concept of metabolic memory has developed largely
from the observations of the Epidemiology of Diabetes
Interventions and Complications (EDIC) study, which followed
the Diabetes Control and Complications Trial (DCCT). In the
former intensive treatment group, after 10 years follow
up, when glycated haemoglobin levels had converged
completely, there was less progression of retinopathy
and lower rates of proliferative retinopathy. Diabetic
nephropathy and neuropathy were similarly reduced. The
combined DCCT and EDIC studies showed a reduction in
the risk of any cardiovascular disease. Good early metabolic
control affects outcome for at least 10 years and it is hoped
this information can be translated into clinical practice to
reduce significantly the burden of long-term complications.
Keywords:
DCCT, EDIC, glycaemic memory, legacy effect, long-
term complications, metabolic memory
Introduction
Memory is commonly perceived as a neurological phenomenon
allowing mental or sensory information to be retained and
reproduced. The word anamnesis is a good way to describe such
recall and, for example, may be used to describe the account a
patient gives of their medical history. A more general effect of
neurological memory has been shown to occur in the delayed
benefits of early education in under-privileged children.
1
In other
biological systems immunological memory has a different, but
equally specific, meaning of a more rapid and enhanced immune
response to the repeated exposure to a particular antigen.
Immunological memory is thought to be based on long-lived
memory in B and T lymphocytes. Does metabolic memory have an
equally good definition and scientific basis? This review will explore
the current use of this term as applied to patients with type 1
diabetes.
One of the earlier uses of the term metabolic memory
2
was
not in diabetes but was concerned with control mechanisms and
futile cycles. Examples given included the effect of diet altering
metabolic patterns for several days and the effects of exercise on
increasing high-density lipoprotein cholesterol and insulin receptor
kinetics. More recently the broad category of metabolic memory
has been used in diabetes to describe the delayed development
of a long-term complication associated with previous exposure to
glucotoxicity, lipotoxicity and other metabolic disturbances. Other
terms covering the same general area include metabolic imprint,
legacy effects, glycaemic memory, hyperglycaemic memory and
latent hyperglycaemic damage. Metabolic memory is an attractive,
memorable alliteration, but the idea of enduring beneficial effects
of good metabolic control early in the course of diabetes is difficult
to define and for which it is difficult to establish basic mechanisms.
It is better described as early or nascent pathology. Indeed the idea
that poor foetal nutrition and nutrition in early life are associated
with subsequent insulin resistance, metabolic syndrome and type 2
diabetes is a well established example of more long-term effects of
nutrition/metabolism on development. Another example of long-
term effects of asymptomatic tissue damage comes from studies
of former smokers whose morbidity from pulmonary and vascular
disease continues for many years after smoking has stopped.
Mechanisms of metabolic memory
Memory probably results from tissue damage that is very slowly
repaired or cannot be fully repaired. Two current hypotheses
suggest that poor control of diabetes results in some irreversible
mitochondrial or vascular change which then predisposes or
progresses to overt long-term complications. To what extent the
cause of the damage is glucotoxicity or lipotoxicity or a combination
of these factors is unknown and no doubt these toxic effects are
modified by other inherited or acquired metabolic processes. The
well established time relationship of duration of diabetes and
incidence of long-term complications is therefore shifted to the left
by poor metabolic control. For most patients with type 1 diabetes
HbA
1c
measurements are the only evidence available of their overall
Alex D Wright
Correspondence to: Dr Alex D Wright
The Diabetes Centre, University Hospital, Selly Oak, Raddlebarn Road, Selly
Oak, Birmingham, B29 6JD, UK.
Tel: +44 (0)121 627 1627
Fax: 01922 656625
E-mail:
S Afr J Diabetes Vasc Dis
2010;
7
: 102–104
Abbreviations and acronyms
AGEs
advanced glycation end-products
DCCT
Diabetes Control and Complications Trial
EDIC
Epidemiology of Diabetes Interventions and Complications
HbA
1C
glycated haemoglobin A
1C
UKPDS
UK Prospective Diabetes Study
