Diabetes Care Volume 38, August 2015 1583

Minimizing Hypoglycemia in International Hypoglycaemia Study Group*

Diabetes
Diabetes Care 2015;38:1583–1591 | DOI: 10.2337/dc15-0279

Hypoglycemia caused by treatment with a sulfonylurea, a glinide, or insulin

coupled with compromised defenses against the resulting falling plasma glucose
concentrations is a problem for many people with diabetes. It is often recurrent,
causes significant morbidity and occasional mortality, limits maintenance of
euglycemia, and impairs physiological and behavioral defenses against sub-
sequent hypoglycemia. Minimizing hypoglycemia includes acknowledging the
problem; considering each risk factor; and applying the principles of intensive
glycemic therapy, including drug selection and selective application of diabetes
treatment technologies. For diabetes health-care providers treating most people
with diabetes who are at risk for or are suffering from iatrogenic hypoglycemia,
these principles include selecting appropriate individualized glycemic goals and
providing structured patient education to reduce the incidence of hypoglycemia.
This is typically combined with short-term scrupulous avoidance of hypoglycemia,
which often will reverse impaired awareness of hypoglycemia. Clearly, the risk of

REVIEW
hypoglycemia is modifiable.

Hypoglycemia is the major limiting factor in the glycemic management of diabetes

with a sulfonylurea, a glinide, or insulin (1,2). It is often recurrent, causes significant
morbidity in most people with type 1 diabetes and in many with advanced type 2
diabetes (i.e., those with absolute endogenous insulin deficiency), and is sometimes
fatal. Hypoglycemia limits maintenance of euglycemia over a lifetime of diabetes
and, thus, generally prevents full realization of the benefits of glycemic control. It
impairs defenses against subsequent falling plasma glucose concentrations and can
cause impaired awareness of hypoglycemia, therefore resulting in a vicious cycle of
recurrent hypoglycemia.
The problem of hypoglycemia in diabetes has been recently reviewed in detail
(1,2). The intent of this article is not to reiterate that information but, rather, to
summarize the relevant background and then focus on pragmatic approaches to
minimizing hypoglycemia. Efforts to minimize hypoglycemia include acknowledging
the problem, considering each risk factor, and applying the relevant principles of
intensive glycemic therapy (3–6). The principles of intensive glycemic therapy in-
clude avoiding sulfonylureas and glinides; using more physiological insulin regi- Corresponding author: Philip E. Cryer, pcryer@
mens, such as insulin analogs, when insulin is indicated; ensuring users are wustl.edu.
confident in their self-management; considering insulin treatment technologies Received 7 February 2015 and accepted 24 May
such as continuous subcutaneous insulin infusion (CSII), continuous glucose moni- 2015.
toring (CGM), and CSII with CGM (ideally with suspension of insulin infusion when *A complete list of the members of the Interna-
tional Hypoglycaemia Study Group can be found
glucose levels fall to a selected low value) for selected patients; and closed-loop
in the APPENDIX.
insulin or insulin and glucagon replacement or pancreas or pancreatic islet trans-
© 2015 by the American Diabetes Association.
plantation for the few patients in whom hypoglycemia persists. However, for the Readers may use this article as long as the work
majority of people with diabetes who are at risk for or are suffering from iatrogenic is properly cited, the use is educational and not
hypoglycemia, the principles include selecting appropriate individualized glycemic for profit, and the work is not altered.
1584 Hypoglycemia Diabetes Care Volume 38, August 2015

goals (7,8) and providing structured pa- hypoglycemia in both types of diabe- raise circulating insulin levels at normal
tient education to reduce the incidence tes. As a function of treatment with a or low plasma glucose concentrations.
of hypoglycemia (9–17). This is typically sulfonylurea, a glinide, or insulin as However, to the extent they lower
coupled with short-term scrupulous well as compromised physiological plasma glucose concentrations, they in-
avoidance of hypoglycemia and often and behavioral defenses against fall- crease the risk of hypoglycemia when
will reverse impaired awareness of hy- ing plasma glucose concentrations (as given with insulin, a sulfonylurea, or a
poglycemia (18–21). discussed shortly), the frequency of glinide.
hypoglycemia increases with the dura- Findings of differences in clinical asso-
BACKGROUND ON tion of diabetes (26). ciations with symptomatic and severe
HYPOGLYCEMIA IN DIABETES Clinical hypoglycemia is a plasma glu- hypoglycemia (33) suggest that hypogly-
Classification and Frequency of cose concentration low enough to cause cemic events that do or do not require
Hypoglycemia in Diabetes symptoms and/or signs, including im- the assistance of another person might
Hypoglycemia in diabetes has been de- paired brain functioning (27). The glyce- be considered separately. They are re-
fined as “all episodes of abnormally low mic thresholds for symptoms and other lated because an increase in the fre-
plasma glucose concentration that ex- manifestations of hypoglycemia shift to quency of the former predicts the
pose the individual to potential harm” lower plasma glucose concentrations in occurrence of the latter (34), but an ep-
(22,23). It has been classified as the people with well-controlled diabetes isode of severe hypoglycemia is a clinical
following: (28) and to higher plasma glucose con- red flag that demands action.
centrations in those with poorly con-
1. Severe hypoglycemia. An event re- trolled diabetes (28,29). Of note, both Impact of Hypoglycemia
quiring assistance of another person childhood and poor glycemic control The morbidity of hypoglycemia includes
to actively administer carbohydrate, shift the glycemic thresholds to higher an array of symptoms. Autonomic (or
glucagon, or other resuscitative plasma glucose concentrations in chil- neurogenic) symptoms include palpita-
actions. dren than in adults (30). Therefore, the tions, tremor and anxiety/arousal
2. Documented symptomatic hypogly- plasma glucose concentration at which (which are adrenergic or catecholamine
cemia. An event during which typical responses occur is variable between mediated), and sweating; hunger; and
symptoms of hypoglycemia are ac- and even within individuals, making it paresthesias (which are cholinergic or
companied by a low measured plasma difficult to define a specific low plasma acetylcholine mediated) (35). Auto-
glucose concentration. glucose concentration as indicative nomic symptoms are largely the result
3. Asymptomatic hypoglycemia. An of clinical iatrogenic hypoglycemia in of sympathetic neural rather than adre-
event not accompanied by typical diabetes. nomedullary activation (36). Neurogly-
symptoms of hypoglycemia but with We recommend that people with copenic symptoms include weakness,
a measured low plasma glucose diabetes treated with a sulfonylurea, drowsiness, impaired cognition ranging
concentration. a glinide, or insulinddrugs that raise from difficulty concentrating to frank
4. Probable symptomatic hypoglyce- circulating insulin levels even at normal confusion, incoordination, and behav-
mia. An event during which symp- or low plasma glucose concentrationsd ioral changes. It can result in seizure,
toms typical of hypoglycemia are become aware of the possibility of de- coma, and ultimately death (1,2).
not accompanied by a plasma glu- veloping hypoglycemia and take action Clearly, hypoglycemia is associated
cose determination but that are pre- to prevent severe hypoglycemia at with a lower health-related quality of
sumed to be caused by a low plasma a self-monitoring of plasma glucose life (37). People with diabetes treated
glucose concentration. (SMPG) concentration #70 mg/dL with a sulfonylurea, a glinide, or insulin
5. Relative (or pseudo-) hypoglycemia. (#3.9 mmol/L). Within the errors of glu- are at risk for hypoglycemia and must
An event during which the person cose monitoring, that glucose alert level learn to detect such symptoms, inter-
with diabetes reports any of the approximates the lower limit of the non- pret them as possibly indicative of hypo-
typical symptoms of hypoglycemia diabetic postabsorptive plasma glucose glycemia, document them with a plasma
and interprets those as indicative concentration range and the glycemic glucose measurement, and ingest ap-
of hypoglycemia with a measured thresholds for activation of physiological propriate carbohydrates to reverse the
plasma glucose concentration that glucose counterregulatory systems episode.
is not low. (31) and is low enough to reduce some The symptoms of hypoglycemia are
of the hormonal defenses against sub- not specific. In addition, the patterns
Hypoglycemia is common in diabetes. sequent hypoglycemia in individuals appear to be somewhat different in chil-
Population-based data indicate that 30– without diabetes (32). dren and in elderly adults (2,38,39) who
40% of people with type 1 diabetes ex- Metformin, thiazolidinediones, may rely more on neuroglycopenic
perience an average of one to three a-glucosidase inhibitors, dipeptidyl symptoms. Evidence suggests that
episodes of severe hypoglycemia peptidase-4 inhibitors, glucagon-like both children with type 1 diabetes and
each year; those with insulin-treated peptide 1 receptor agonists, and sodium- their parents fail to recognize 40–50% of
type 2 diabetes experience about glucose cotransporter 2 inhibitors should hypoglycemic episodes (40). The very
one-third that number (24–26). The not, and presumably do not, cause hypo- young rely on observations by their
rates of any type of hypoglycemia are glycemia when given alone or in combina- caregivers who often observe behav-
;50-fold higher than those of severe tion with one another because they do not ioral changes as a clue to hypoglycemia,
care.diabetesjournals.org International Hypoglycaemia Study Group 1585

although many parents note pallor, an judged as possibly caused by hypoglyce- The compromised defenses in estab-
autonomic sign. mia (57). Hypoglycemia is associated lished type 1 diabetes and advanced
Although treatment of diabetes with an increased risk of cardiovascular type 2 diabetes (those with absolute
with a sulfonylurea, a glinide, or insulin events and all-cause mortality in insulin- deficiency of endogenous insulin) in-
is a common cause of hypoglycemia, treated patients with type 1 and type 2 clude loss of a decrease in insulin and
there are other causes of low plasma diabetes (58). Although prolonged, pro- loss of an increase in glucagon, both of
glucose concentrations, including renal, found hypoglycemia can result in brain which are probably the result of b-cell
hepatic, and cardiac failure; sepsis; and death, and some hypoglycemic deaths failure (64), and attenuation of an in-
inanition (27). Thus, a low plasma glucose are accidental or suicidal, most fatal ep- crease in epinephrine as plasma glu-
concentration in a patient not treated isodes are believed to be the result of cose concentrations fall. These cause
with such drugs, and even in some pa- other mechanisms, such as cardiac ar- the clinical syndrome of defective glu-
tients treated with these drugs, may rhythmias (59,60). cose counterregulation, which increases
well be a marker of another disorder In addition to the morbidity and mor- the risk of severe hypoglycemia by a fac-
rather than an iatrogenic cause of mor- tality associated with hypoglycemia, tor of $25 (1). Attenuation of the sym-
bidity and mortality (41). Nonetheless, there can also be negative consequen- pathoadrenal response is believed to
there is substantial evidence that iatro- ces for emotional well-being and quality cause the clinical syndrome of impaired
genic hypoglycemia is sometimes fatal of life. Patients can develop a fear of awareness of hypoglycemia, which in-
[reviewed in Cryer (8)]. hypoglycemia that not only decreases creases the risk of severe hypoglycemia
It has been known since the discovery their quality of life but also that of their by a factor of $6 (1,2). Defective glu-
of insulin that hypoglycemia can cause family members (61,62). Fear of hypo- cose counterregulation and impaired
death. There is substantial evidence of glycemia should be assessed by diabetes awareness of hypoglycemia are compo-
associations between hypoglycemia and health-care providers by either survey nents of the syndrome of hypoglycemia-
mortality in diabetes. For example, se- or inquiry. An extreme level of fear associated autonomic failure (HAAF) in
vere hypoglycemia was associated with may have a negative impact on diabetes diabetes. HAAF can be caused by recent
increased mortality in six randomized management, leading, for example, to antecedent hypoglycemia, sleep, or
controlled trials of intensive glycemic maintenance of higher-than-desirable prior exercise, but its precise mecha-
therapy, two in intensive care unit plasma glucose levels to avoid hypogly- nisms are not known (64,65). HAAF is a
(ICU) patients (42,43), and four in pa- cemia. However, awareness of the po- dynamic syndrome distinct from dia-
tients with type 2 diabetes (44–47). Al- tential dangers of hypoglycemia and betic autonomic neuropathy.
though these associations do not some level of concern are appropriate The clinical diagnosis of impaired
establish a causal connection, the con- for individuals at risk. Data support awareness of hypoglycemia is based on
sistent pattern in all six trials increases that high-fear/high-risk patients would the patient’s subjective assessment (or
the probability that hypoglycemia was benefit most from efforts to lower the evidence from close associates) because
the cause of some of the deaths. Fur- likelihood of hypoglycemia. A minority documentation of substantially reduced
thermore, there was increased mortal- of patients have inappropriately high symptoms during experimental hypo-
ity in the intensive glycemic therapy levels of fear when there is no history glycemia, which is critical for research
arms of randomized controlled trials in of severe hypoglycemia or an inappro- studies (64), is not practical clinically.
ICU patients (42) and in patients with priate lack of concern when there is a Clinical scoring of symptoms has been
type 2 diabetes (44). Although excess high risk of recurrent episodes. developed (66–68). Because impaired
mortality could have been the result of awareness is inducible and reversible,
some nonglycemic aspect of the inten- Pathophysiology of Hypoglycemia such a classification cannot be assumed
sive therapy regimen in the patients As plasma glucose concentrations fall, to be stable. Therefore, people with di-
with type 2 diabetes (44), in the ICU pa- the prevention or rapid correction of hy- abetes at risk for hypoglycemia by virtue
tients, only the glycemic goals differed poglycemia normally involves physio- of their diabetes treatment regimen
(42). Finally, whereas older series indi- logical defenses (a decrease in insulin should have their awareness status re-
cated that 2–4% of patients with type 1 and an increase in glucagon and, in the viewed regularly. The fact that impaired
diabetes died of hypoglycemia (48–50), absence of the latter, an increase in epi- awareness of hypoglycemia, and at least
more recent series have indicated nephrine) and a behavioral defense (car- in part the attenuated sympathoadrenal
higher hypoglycemic mortality rates of bohydrate ingestion prompted by the response, can be reversed by as little as
4% (51), 6% (52), 7% (53), and 10% (54). perception of symptoms of hypoglyce- 2–3 weeks of scrupulous avoidance of
Indeed, hypoglycemia at the time of mia) (31,63). Hypoglycemia in diabetes hypoglycemia in most affected patients
death of a patient with type 1 diabetes is typically the result of the interplay of (18–21) provides compelling additional
has been documented with CGM (55). therapeutic hyperinsulinemia caused by support for the clinical relevance of
Hypoglycemic mortality rates have also treatment with a sulfonylurea, a glinide, HAAF and for a portion of the therapeu-
been reported in series of patients with or insulin (all of which raise circulating tic approach to the problem of hypogly-
type 2 diabetes (56,57). Of the evaluable insulin levels regardless of the plasma cemia in diabetes, as discussed shortly.
deaths in the ACCORD (Action to Control glucose concentration) and compro-
Cardiovascular Risk in Diabetes) trial, 1% mised physiological and behavioral de- Risk Factors for Hypoglycemia
were judged as definitely or probably fenses against the resulting falling The risk factors for iatrogenic hypogly-
caused by hypoglycemia, and 9% were plasma glucose concentrations (64). cemia in diabetes are those relevant to
1586 Hypoglycemia Diabetes Care Volume 38, August 2015

therapeutic hyperinsulinemia and those and the relationship between be expected to improve glycemic con-
relevant to compromised defenses lower A1C levels and higher rates trol more safely.
against the resulting falling plasma of severe hypoglycemia (69) is
glucose concentrations (1,2). The con- now less marked with less hypogly- MINIMIZING HYPOGLYCEMIA IN
ventional risk factors for hypoglyce- cemia at a given A1C level (70,71), DIABETES
mia in diabetes are based on the undoubtedly due to progressive General Approach to the Problem
premise that absolute or relative ther- improvements in the glycemic The problem of iatrogenic hypoglycemia
apeutic hyperinsulinemia is the sole management of diabetes, a low can be solved only if it is recognized by
determinant of risk and include the A1C is a risk factor for hypoglyce- the diabetes health-care provider, the
following: mia during intensive therapy of di- patient, and the caregivers. In a patient
abetes [reviewed in Cryer (8)] treated with a sulfonylurea, a glinide, or
1. Insulin (or sulfonylurea or glinide) (33,72–74). Nonetheless, as dis- insulin, it is fundamental that the provider
doses are excessive, ill-timed, or of cussed shortly, structured patient acknowledges the possibility of real, or
the wrong type. education is intended to reduce feared, hypoglycemia and gives the patient
2. Exogenous glucose delivery is de- both the frequency of severe hypo- and those close to the patient the oppor-
creased as it is after a missed or glycemia and the A1C level (9–17). tunity to express their observations and
low-carbohydrate meal and during thoughts. The provider must discuss with
the overnight fast. Glycemic Goals in Diabetes patients the frequency and timing of hypo-
3. Endogenous glucose production is The selection of a glycemic goal for glycemic episodes of any severity, seek ev-
decreased as it is after alcohol a person with diabetes treated with a idence for asymptomatic episodes by
ingestion. sulfonylurea, a glinide, or insulin is a asking whether others sometimes tell the
4. Glucose utilization is increased as it is trade-off between the benefits of glyce- patient that he or she is hypoglycemic and
during and shortly after exercise. mic control (partial prevention or delay by asking close acquaintances directly, and
5. Sensitivity to insulin is increased as it of microvascular complications and per- review SMPG (and CGM) data with the
is in the middle of the night, late after haps that of macrovascular compli- patient to search for evidence of hypo-
exercise, and after weight loss or im- cations) and the risk of recurrent glycemia.
proved fitness. morbidity, and potential mortality, of When hypoglycemia is recognized to
6. Insulin clearance is decreased as it is hypoglycemia (8). A reasonable individ- be a problem, the diabetes health-care
with renal failure, hepatic failure, hy- ualized glycemic goal is the lowest A1C provider and patient should first con-
pothyroidism, or, rarely, high levels that does not cause severe hypoglyce- sider each conventional risk factor and
of insulin-binding antibodies. mia and preserves awareness of hypo- those indicative of compromised de-
glycemia, preferably with little or no fenses against falling plasma glucose
Risk factors for hypoglycemia indica- symptomatic or even asymptomatic hy- concentrations. Sometimes, a cause of
tive of compromised defenses against poglycemia, at a given stage in the evo- recurrent hypoglycemia can be identi-
falling plasma glucose concentrations lution of the individual’s diabetes (8). fied and corrected.
include the following situations: Thus, the glycemic goal should be linked In the absence of a simple solution,
not only to the level of glycemic control the provider must review the patient’s
1. There is absolute endogenous insulin (i.e., the A1C) but also to the risk of management strategies (3–6). A de-
deficiency. The frequency of hypo- hypoglycemia, specifically to the drugs tailed discussion of these strategies
glycemia increases with the duration used and the degree of endogenous in- and their application in special popula-
of diabetes (26) probably due to pro- sulin deficiency and the anticipated tions, such as children, the elderly, and
gressive endogenous insulin defi- benefit of the targeted level of glycemic pregnant women, to particular activi-
ciency that develops rapidly in type control. For example, a nondiabetic A1C ties, such as exercise and driving, and
1 diabetes and more slowly in type 2 would be reasonable in a patient with to specific drugs and therapeutic regi-
diabetes. Absolute endogenous insu- early type 2 diabetes who is treated ef- mens is provided elsewhere (1,2).
lin deficiency predicts loss of the glu- fectively with lifestyle changes and/or In general, diabetes treatment regi-
cagon response to hypoglycemia drugs that do not cause hypoglycemia, mens should be constructed with a
(1,64). whereas a higher A1C that is just suffi- view to minimizing hypoglycemia as
2. There is a history of severe hypogly- cient to prevent symptoms of hypergly- well as hyperglycemia. Those principles
cemia, impaired awareness of hy- cemia would be reasonable in a patient relevant to minimizing hypoglycemia in-
poglycemia, or both as well as a who has a limited life expectancy. Ac- clude drug selection, the use of insulin
relationship of hypoglycemia to re- tion is required when a sulfonylurea, analogs, and the use of insulin treat-
cent antecedent hypoglycemia, glinide, or insulin regimen causes severe ment technologies as well as patient ed-
sleep, or exercise. These are features hypoglycemia, impaired awareness of ucation. Discontinuing a sulfonylurea
of HAAF in diabetes (64). hypoglycemia, or an unacceptable num- or a glinide would obviously reduce hy-
3. There is intensive glycemic therapy ber of symptomatic or asymptomatic poglycemia and generally would require
per se evidenced by lower glycemic hypoglycemic episodes. Actions to the addition of an alternative glucose-
goals. Although hypoglycemia can minimize the risk of hypoglycemia lowering medication. But sulfonylureas
occur in patients with relatively taken by the diabetes health-care pro- are both inexpensive and widely avail-
high hemoglobin A 1c (A1C) levels vider and the person with diabetes can able and sometimes might be the most
care.diabetesjournals.org International Hypoglycaemia Study Group 1587

practical medication. When therapy hypoglycemia outweighed any benefit impaired awareness of hypoglycemia
with insulin is necessary, as it is in of CSII and CGM. (18–21).
type 1 diabetes and ultimately in most Work continues on closed-loop insu- Although generally confined to ob-
individuals with type 2 diabetes, the use lin or insulin and glucagon replacement servational studies, longstanding and
of more physiological insulin regimens (93,94) and on pancreas and pancreatic substantial evidence shows that struc-
and genetically engineered insulin ana- islet transplantation (95). When these tured patient education on implement-
logs reduces the frequency of at least become consistently successful, they ing flexible intensive glycemic therapy,
nocturnal hypoglycemia (75,76), includ- will likely eliminate hypoglycemia. particularly with insulin, reduces the in-
ing severe nocturnal hypoglycemia (77). cidence of hypoglycemia without com-
CSII offers potential advantages over Structured Education Coupled With promising glycemic control (9–17). We
multiple daily injections (MDIs) of insu- Scrupulous Avoidance of would reason that it is unsafe to en-
lin in a basal-bolus insulin regimen, es- Hypoglycemia courage such intensive therapy without
pecially in type 1 diabetes, because one Given documented evidence that glyce- accompanying it with high-quality edu-
can vary the rate of basal insulin infusion mic control partially prevents or delays cation that provides sufficient knowl-
throughout a 24-h period. CSII may be microvascular complications (retinopa- edge of insulin actions and empowers
better than MDI in selected, capable, and thy, nephropathy, and neuropathy) in patients with the tools to prevent and
motivated patients with hypoglycemia- type 1 diabetes (69) and type 2 diabetes treat hypoglycemia.
prone diabetes (70,78) largely because of (44,45,96,97) and might partially pre- In addition to basic diabetes educa-
the premise that CSII may reduce A1C to vent or delay some macrovascular com- tion that includes the principles of nutri-
some extent without increasing the risk of plications in type 1 diabetes (98) and tion, all people with diabetes treated
hypoglycemia. One meta-analysis con- type 2 diabetes (99), it follows that a with a sulfonylurea, a glinide, or insulin
cluded that compared with MDI, CSII re- lower A1C is in the best interest of peo- and not only those treated with insulin
duces hypoglycemia, but this conclusion ple with diabetes if that can be achieved should receive structured education
was based on three randomized con- and maintained safely. However, when about hypoglycemia and how to avoid
trolled trials using an NPH or lente insu- hypoglycemia becomes a problem (the it. The therapeutic objective is to mini-
lin–based MDI regimen (79); trials using regimen causes severe hypoglycemia, mize the number of episodes of hypo-
insulin analog MDI comparison groups impaired awareness of hypoglycemia, glycemia and their severity and duration
have not reported lower rates of hypogly- an unacceptable number of episodes without promoting hyperglycemia and
cemia with CSII (80–82). A subsequent of symptomatic or asymptomatic hypo- raising A1C levels. Indeed, the goal is
meta-analysis of randomized controlled glycemia, or a combination of these), to reduce both hypoglycemia and A1C
trials of CSII versus MDI disclosed only a the core action for essentially all af- levels. This patient education must
small lowering of A1C (20.2%) with no fected patients is consideration of indi- cover a broad range of information and
significant difference in severe or noctur- vidualized glycemic goals (7,8) and skills training as well as include a moti-
nal hypoglycemia (83). There is evidence delivery of structured patient education vational element.
that insulin pump bolus calculators are (often reeducation) to reduce the inci- Patients need to understand the risk
effective (84). Despite its conceptual at- dence of hypoglycemia without com- factors for hypoglycemia and how their
tractiveness, subcutaneous CGM alone promising glycemic control (9–17). particular sulfonylurea, glinide, or insu-
has had rather minimal effects in reducing This is typically coupled with short- lin regimen can cause hypoglycemia and
the frequency of hypoglycemia (85–88), term scrupulous avoidance of hypogly- when that is most likely to occur. Thus,
although studies focused on more patients cemia, which, when successful, reverses they should be familiar with the effects
with problematic hypoglycemia are
needed. However, the combination of
real-time CGM and CSII (sensor-augmented
pump therapy) including an insulin pump Table 1—Recommendations of the International Hypoglycaemia Study Group
programmed to suspend insulin infusion People with diabetes treated with a sulfonylurea, a glinide, or insulin should
when glucose levels fall to a selected low c Be educated about hypoglycemia.

value has been reported to reduce the c Treat SMPG levels #70 mg/dL (3.9 mmol/L) to avoid progression to clinical iatrogenic
hypoglycemia.
frequency of severe hypoglycemia in
c Regularly be queried about hypoglycemia, including the glucose level at which symptoms
type 1 diabetes (89–91). Based on ran- develop. Those developing symptoms at a glucose level ,55 mg/dL (3.0 mmol/L) should
domized controlled trials published up to be considered at risk.
2012, Yeh et al. (92) concluded that CSII When hypoglycemia becomes a problem, the diabetes health-care provider should
(compared with MDI), real-time CGM c Consider each conventional risk factor and those indicative of compromised glucose
(compared with SMPG), and CSII plus counterregulation.
CGM (compared with MDI and SMPG) c Avoid sulfonylureas (and glinides) if possible, using insulin analogs when insulin is

had not been shown to reduce the inci- required, and consider using CSII, CGM, and CSII + CGM in selected patients.
c Provide structured education and, in patients with impaired awareness of hypoglycemia,
dence of severe hypoglycemia in either
prescribe short-term scrupulous avoidance of hypoglycemia.
type 1 or insulin-treated type 2 diabetes. c Seek to achieve the lowest A1C level that does not cause severe hypoglycemia and
Of note, one recent study (16) reported preserves awareness of hypoglycemia with an acceptable number of less-than-severe
that the impact of ongoing personal sup- episodes of hypoglycemia, provided that benefit from glycemic control can be anticipated.
port for patients with problematic
1588 Hypoglycemia Diabetes Care Volume 38, August 2015

of the dose and timing of their individual psychosocial interventions to facilitate (Division of Endocrinology and Metabolism, Li Ka
glucose-lowering drugs and the risks of behavioral changes (14). Those beliefs Shing Knowledge Institute and Keenan Research
Centre for Biomedical Science, St. Michael’s Hos-
missed or low-carbohydrate meals, the range from an obsessive attachment pital, and Professor of Medicine and Nutritional
overnight fast, and alcohol ingestion. to a very low A1C level to an overriding Sciences, University of Toronto, Toronto, ON,
Patients must learn strategies to defend fear of hypoglycemia. Canada), Yingying Luo (Endocrinology and
against the glycemic effects of planned Metabolism Department, Peking University
and unanticipated exercise and the oc- People’s Hospital, Beijing, China), Robert Vigersky
CONCLUSIONS
(Colonel, Army Medical Core, and Director, Di-
currence of delayed hypoglycemia after Clinical iatrogenic hypoglycemia is a abetes Institute, Walter Reed National Military
intense physical activity. In short, they problem for many people with diabetes Center, Bethesda, MD), and Sophia Zoungas
need to learn to adjust their medica- treated with a sulfonylurea, a glinide, or (School of Public Health and Preventive
tions, meal plans, and exercise to opti- Medicine, Monash University, Melbourne,
insulin. Minimizing hypoglycemia in- VIC, Australia).
mize glycemic control and minimize cludes acknowledging the problem, con- Acknowledgments. The manuscript was pre-
hypoglycemia. The patient as well as sidering each risk factor, and applying pared by Janet Dedeke, an administrative
the provider and caregivers must under- the principles of intensive glycemic ther- assistant at Washington University School of
stand that episodes of hypoglycemia sig- apy, including selecting appropriate Medicine in St. Louis, St. Louis, MO.
nal an increased likelihood of future, Duality of Interest. The International Hypogly-
individualized glycemic goals and pro- caemia Study Group (IHSG) is supported through
often more severe, hypoglycemia (34). viding structured patient education typ- an unrestricted educational grant from Novo
This insight is fundamental to under- ically combined with short-term Nordisk awarded to Six Degrees Academy (SDA)
standing that short-term scrupulous scrupulous avoidance of hypoglycemia. of Toronto, Ontario, Canada. Along with the
avoidance of hypoglycemia is key The recommendations of the Interna- IHSG chair, SDA has been solely responsible for
to long-term reversal of impaired membership recruitment/selection and con-
tional Hypoglycaemia Study Group are tent/outcomes for the meetings. The rationale
awareness of hypoglycemia (18–21). In summarized in Table 1. The risk of hypo- for the formation of IHSG is that hypoglycemia is
addition, a patient using CGM needs glycemia is modifiable and can be mini- an underrecognized problem that deserves in-
to learn how to use monitoring data mized with the methods discussed in creased awareness and focus across the health-
to minimize hypoglycemia as well as this review. care community. The group’s ultimate goal is to
hyperglycemia. improve the lives of patients with diabetes.
Patients at risk for hypoglycemia need P.A. has served on scientific advisory
boards and/or as a lecturer for AstraZeneca,
to know the common symptoms of hy- Appendix Boehringer Ingelheim/Lilly, Bristol-Myers
poglycemia, particularly those most Squibb, GlaxoSmithKline, Janssen, Merck Sharp &
International Hypoglycaemia Study Group.
meaningful to them. Being able to rec- Stephanie A. Amiel (RD Lawrence Professor of Dohme, Novartis, and Sanofi. B.C. had research
ognize even subtle symptoms and Diabetic Medicine, Division of Diabetes and grant support from Halozyme and Lilly to the
correctly interpret them as possibly in- Nutritional Sciences, King’s College London, former MidAmerica Diabetes Associates. P.E.C.
London, U.K.), Pablo Aschner (Associate Profes- has served on scientific advisory boards for
dicative of hypoglycemia can empower Boehringer Ingelheim/Lilly, Calibrium, Merck
sor of Endocrinology, Javeriana University
the patient to take immediate action, School of Medicine; Director of Research, San Sharp & Dohme, Novo Nordisk, and Pfizer.
including SMPG and ingesting carbo- Ignacio University Hospital; and Scientific Dire- B.d.G. has served on scientific advisory boards
hydrates if the value is #70 mg/dL ctor, Colombian Diabetes Association, Bogotá, for Merck Sharp & Dohme, Novo Nordisk, and
(3.9 mmol/L). Consistently treating Colombia), Belinda Childs (Clinical Nurse Spe- Sanofi and received research grant support
cialist, Great Plains Diabetes, and Adjunct Pro- from Europharma Group. S.R.H. has served on
SMPG (or CGM) values below the alert scientific advisory boards and provided consul-
fessor, Wichita State University, Wichita, KS),
value and not delaying treatment in the Philip E. Cryer (Professor of Medicine Emeritus, tations for which his institution has received
absence of symptoms are important. If Washington University in St. Louis, St. Louis, MO), remuneration from Lilly, Novo Nordisk, Takeda,
plasma glucose testing is not practical, it Bastiaan de Galan (Department of Internal Merck Sharp & Dohme, and Becton Dickinson
is wise to treat symptoms of hypoglyce- Medicine, Radboud University Nijmegen Med- and has served as a speaker for which he re-
ical Centre, Nijmegen, the Netherlands), Simon ceived remuneration from Lilly, Novo Nordisk,
mia whenever they arise and to perform Boehringer Ingelheim, and Takeda. B.M.F. has
R. Heller (Professor of Clinical Diabetes, Univer-
SMPG as soon as possible. The patient sity of Sheffield, and Director of Research and served on scientific advisory boards and as a
needs to know how to treat (and not Development and Honorary Consultant Physi- speaker for Boehringer Ingelheim, Janssen,
overtreat) hypoglycemia with appropri- cian, Sheffield Teaching Hospitals NHS Founda- Lilly, Merck Sharp & Dohme, Novo Nordisk, and
ate carbohydrates and how to follow an tion Trust, Sheffield, U.K.), Brian M. Frier Lilly. L.G.-F. has served as a consultant or speaker
(Honorary Professor of Diabetes, The Queen’s and/or has received research grant support from
episode with monitoring and, if indi- Abbott Diabetes Care, AstraZeneca, Dexcom,
Medical Research Institute, University of Edin-
cated, regimen adjustments. Close asso- burgh, Edinburgh, Scotland, U.K.), Linda Johnson & Johnson, and Merck Sharp & Dohme.
ciates, such as a spouse or partner, also Gonder-Frederick (Associate Professor, De- T.J. has served as a speaker for Novo Nordisk,
need to recognize severe hypoglycemia partment of Psychiatry and Neurobehavioral Lilly, Medtronic, and Sanofi. K.K. has served
and how to prepare and administer Sciences, and Clinical Director, Behavioral as a consultant or speaker for AstraZeneca,
Medicine Center, University of Virginia Health Boehringer Ingelheim, Janssen, Lilly, Merck
glucagon. System, Charlottesville, VA), Timothy Jones (Clinical Sharp & Dohme, Novartis, Novo Nordisk, and
Diabetes health-care providers also Professor, School of Paediatrics and Child Health, Sanofi and has received research grant support
should be aware that some patients, Telethon Institute for Child Health Research, Uni- from AstraZeneca, Boehringer Ingelheim, Lilly,
particularly those with longstanding versity of Western Australia, and Head, Depart- Novartis, Novo Nordisk, Roche, and Sanofi.
impaired awareness of hypoglycemia, ment of Endocrinology and Diabetes, Princess L.A.L. has served as a consultant or speaker
Margaret Hospital for Children, Perth, WA, Aus- for Abbott, AstraZeneca, Bristol-Myers Squibb,
may have health beliefs that interfere tralia), Kamlesh Khunti (Professor of Primary Boehringer Ingelheim, Lilly, GlaxoSmithKline,
with their ability to benefit from educa- Care Diabetes and Vascular Medicine, University Janssen, Merck Sharp & Dohme, Novo Nordisk,
tion alone and thus may benefit from of Leicester, Leicester, U.K.), Lawrence A. Leiter Sanofi, Servier, and Takeda. R.V. has served as a
care.diabetesjournals.org International Hypoglycaemia Study Group 1589

consultant for Medtronic and has received re- (MEMO) study. Diabetes Res Clin Pract 2011;93: predictors of hypoglycaemia in type 1 and
search grant support from Dexcom. S.Z. has 328–336 insulin-treated type 2 diabetes: a population-
served on scientific advisory boards for Amgen, 13. Hopkins D, Lawrence I, Mansell P, et al. Im- based study. Diabet Med 2005;22:749–755
Bristol-Myers Squibb, AstraZeneca, Janssen, proved biomedical and psychological outcomes 26. UK Hypoglycaemia Study Group. Risk of hy-
Merck Sharp & Dohme, Novo Nordisk, Sanofi, 1 year after structured education in flexible in- poglycaemia in types 1 and 2 diabetes: effects of
and Takeda; has served as a speaker for Bristol- sulin therapy for people with type 1 diabetes: treatment modalities and their duration. Diabe-
Myers Squibb, AstraZeneca, Janssen, Merck the U.K. DAFNE experience. Diabetes Care 2012; tologia 2007;50:1140–1147
Sharp & Dohme, Servier, and Takeda; and has re- 35:1638–1642 27. Cryer PE, Axelrod L, Grossman AB, et al.;
ceived research grant support from Bristol-Myers 14. Leelarathna L, Little SA, Walkinshaw E, et al. Endocrine Society. Evaluation and management
Squibb and AstraZeneca. No other potential Restoration of self-awareness of hypoglycemia of adult hypoglycemic disorders: an Endocrine
conflicts of interest relevant to this article were in adults with long-standing type 1 diabetes: Society clinical practice guideline. J Clin Endo-
reported. hyperinsulinemic-hypoglycemic clamp sub- crinol Metab 2009;94:709–728
Author Contributions. The content of this study results from the HypoCOMPaSS trial. Di- 28. Amiel SA, Sherwin RS, Simonson DC,
review was first discussed at a meeting of the abetes Care 2013;36:4063–4070 Tamborlane WV. Effect of intensive insulin ther-
IHSG. The manuscript was then drafted by P.E.C. 15. de Zoysa N, Rogers H, Stadler M, et al. A apy on glycemic thresholds for counterregula-
It was revised and revised again by the members psychoeducational program to restore hypogly- tory hormone release. Diabetes 1988;37:901–
of the IHSG. cemia awareness: the DAFNE-HART pilot study. 907
Diabetes Care 2014;37:863–866 29. Boyle PJ, Schwartz NS, Shah SD, Clutter
16. Little SA, Leelarathna L, Walkinshaw E, et al. WE, Cryer PE. Plasma glucose concentrations
References
Recovery of hypoglycemia awareness in long- at the onset of hypoglycemic symptoms in pa-
1. Cryer PE. Hypoglycemia in Diabetes: Patho- standing type 1 diabetes: a multicenter 2 3 2 tients with poorly controlled diabetes and in
physiology, Prevalence, and Prevention. 2nd ed. factorial randomized controlled trial comparing nondiabetics. N Engl J Med 1988;318:1487–
Alexandria, VA, American Diabetes Association, insulin pump with multiple daily injections 1492
2013 and continuous with conventional glucose self- 30. Jones TW, Boulware SD, Kraemer DT, Caprio
2. Frier BM, Heller SR, McCrimmon RJ. Hypogly- monitoring (HypoCOMPaSS). Diabetes Care S, Sherwin RS, Tamborlane WV. Independent
caemia in Clinical Diabetes. 3rd ed. Chichester, 2014;37:2114–2122 effects of youth and poor diabetes control on
West Sussex, Wiley Blackwell, 2014 17. Elliott J, Jacques RM, Kruger J, et al. Sub- responses to hypoglycemia in children. Diabetes
3. Cryer PE, Davis SN, Shamoon H. Hypoglyce- stantial reductions in the number of diabetic 1991;40:358–363
mia in diabetes. Diabetes Care 2003;26:1902– ketoacidosis and severe hypoglycaemia epi- 31. Cryer PE. The prevention and correction of
1912 sodes requiring emergency treatment lead to hypoglycemia. In Handbook of Physiology. Vol.
4. Rossetti P, Porcellati F, Bolli GB, Fanelli CG. reduced costs after structured education in II, The Endocrine Pancreas and Regulation of
Prevention of hypoglycemia while achieving adults with type 1 diabetes. Diabet Med 2014; Metabolism. Jefferson LS, Cherrington AD, Eds.
good glycemic control in type 1 diabetes: the 31:847–853 New York, Oxford University Press, 2001,
role of insulin analogs. Diabetes Care 2008;31 18. Fanelli CG, Epifano L, Rambotti AM, et al. p. 1057–1092
(Suppl. 2):S113–S120 Meticulous prevention of hypoglycemia normal- 32. Davis SN, Shavers C, Mosqueda-Garcia R,
5. Amiel SA. Hypoglycemia: from the laboratory izes the glycemic thresholds and magnitude of Costa F. Effects of differing antecedent hypogly-
to the clinic. Diabetes Care 2009;32:1364–1371 most of neuroendocrine responses to, symptoms cemia on subsequent counterregulation in nor-
6. Graveling AJ, Frier BM. Hypoglycaemia: an of, and cognitive function during hypoglycemia in mal humans. Diabetes 1997;46:1328–1335
overview. Prim Care Diabetes 2009;3:131–139 intensively treated patients with short-term 33. ORIGIN Trial Investigators. Predictors of
7. Inzucchi SE, Bergenstal RM, Buse JB, et al.; IDDM. Diabetes 1993;42:1683–1689 nonsevere and severe hypoglycemia during
American Diabetes Association; European As- 19. Fanelli C, Pampanelli S, Epifano L, et al. glucose-lowering treatment with insulin glargine
sociation for the Study of Diabetes. Manage- Long-term recovery from unawareness, defi- or standard drugs in the ORIGIN trial. Diabetes
ment of hyperglycemia in type 2 diabetes, cient counterregulation and lack of cognitive Care 2015;38:22–28
2015: a patient-centered approach: update dysfunction during hypoglycaemia, following in- 34. Cox DJ, Gonder-Frederick L, Ritterband L,
to a position statement of the American Diabe- stitution of rational, intensive insulin therapy in Clarke W, Kovatchev BP. Prediction of severe
tes Association and the European Association IDDM. Diabetologia 1994;37:1265–1276 hypoglycemia. Diabetes Care 2007;30:1370–
for the Study of Diabetes. Diabetes Care 2015; 20. Cranston I, Lomas J, Maran A, Macdonald I, 1373
38:140–149 Amiel SA. Restoration of hypoglycaemia aware- 35. Towler DA, Havlin CE, Craft S, Cryer P.
8. Cryer PE. Glycemic goals in diabetes: trade- ness in patients with long-duration insulin- Mechanism of awareness of hypoglycemia. Per-
off between glycemic control and iatrogenic hy- dependent diabetes. Lancet 1994;344:283–287 ception of neurogenic (predominantly choliner-
poglycemia. Diabetes 2014;63:2188–2195 21. Dagogo-Jack S, Rattarasarn C, Cryer PE. Re- gic) rather than neuroglycopenic symptoms.
9. Mühlhauser I, Jörgens V, Berger M, et al. Bi- versal of hypoglycemia unawareness, but not Diabetes 1993;42:1791–1798
centric evaluation of a teaching and treatment defective glucose counterregulation, in IDDM. 36. DeRosa MA, Cryer PE. Hypoglycemia and
programme for type 1 (insulin-dependent) dia- Diabetes 1994;43:1426–1434 the sympathoadrenal system: neurogenic
betic patients: improvement of metabolic con- 22. Workgroup on Hypoglycemia, American Di- symptoms are largely the result of sympathetic
trol and other measures of diabetes care for up abetes Association. Defining and reporting hy- neural, rather than adrenomedullary, activa-
to 22 months. Diabetologia 1983;25:470–476 poglycemia in diabetes: a report from the tion. Am J Physiol Endocrinol Metab 2004;287:
10. Sämann A, Mühlhauser I, Bender R, Kloos Ch, American Diabetes Association Workgroup on E32–E41
Müller UA. Glycaemic control and severe hypogly- Hypoglycemia. Diabetes Care 2005;28:1245– 37. Marrett E, Radican L, Davies MJ, Zhang Q.
caemia following training in flexible, intensive in- 1249 Assessment of severity and frequency of self-
sulin therapy to enable dietary freedom in people 23. Seaquist ER, Anderson J, Childs B, et al. Hy- reported hypoglycemia on quality of life in pa-
with type 1 diabetes: a prospective implementation poglycemia and diabetes: a report of a work- tients with type 2 diabetes treated with oral
study. Diabetologia 2005;48:1965–1970 group of the American Diabetes Association antihyperglycemic agents: a survey study.
11. Plank J, Köhler G, Rakovac I, et al. Long-term and the Endocrine Society. Diabetes Care BMC Res Notes 2011;4:251–257
evaluation of a structured outpatient education 2013;36:1384–1395 38. Amin A, Lau L, Crawford S, Edwards A,
programme for intensified insulin therapy in pa- 24. Pedersen-Bjergaard U, Pramming S, Heller Pacaud D. Prospective assessment of hypogly-
tients with type 1 diabetes: a 12-year follow-up. SR, et al. Severe hypoglycaemia in 1076 adult cemia symptoms in children and adults with
Diabetologia 2004;47:1370–1375 patients with type 1 diabetes: influence of risk type 1 diabetes. Can J Diabetes 2014;38:263–
12. Crasto W, Jarvis J, Khunti K, et al. Multi- markers and selection. Diabetes Metab Res Rev 268
factorial intervention in individuals with type 2 2004;20:479–486 39. Graveling AJ, Noyes KJ, Allerhand MH, et al.
diabetes and microalbuminuria: the Microalbumin- 25. Donnelly LA, Morris AD, Frier BM, et al.; Prevalence of impaired awareness of hypogly-
uria Education and Medication Optimisation DARTS/MEMO Collaboration. Frequency and cemia and identification of predictive symptoms
1590 Hypoglycemia Diabetes Care Volume 38, August 2015

in children and adolescents with type 1 diabe- 55. Tanenberg RJ, Newton CA, Drake AJ. Con- 70. Cooper MN, O’Connell SM, Davis EA, Jones
tes. Pediatr Diabetes 2014;15:206–213 firmation of hypoglycemia in the “dead-in-bed” TW. A population-based study of risk factors for
40. Gonder-Frederick L, Zrebiec J, Bauchowitz syndrome, as captured by a retrospective con- severe hypoglycaemia in a contemporary cohort
A, et al. Detection of hypoglycemia by children tinuous glucose monitoring system. Endocr of childhood-onset type 1 diabetes. Diabetolo-
with type 1 diabetes 6 to 11 years of age and Pract 2010;16:244–248 gia 2013;56:2164–2170
their parents: a field study. Pediatrics 2008;121: 56. Holstein A, Egberts EH. Risk of hypoglycae- 71. Karges B, Rosenbauer J, Kapellen T, et al.
e489–e495 mia with oral antidiabetic agents in patients Hemoglobin A1c levels and risk of severe hypo-
41. Zoungas S, Patel A, Chalmers J, et al.; with type 2 diabetes. Exp Clin Endocrinol Diabe- glycemia in children and young adults with type
ADVANCE Collaborative Group. Severe hypo- tes 2003;111:405–414 1 diabetes from Germany and Austria: a trend
glycemia and risks of vascular events and 57. Bonds DE, Miller ME, Bergenstal RM, et al. analysis in a cohort of 37,539 patients between
death. N Engl J Med 2010;363:1410–1418 The association between symptomatic, severe 1995 and 2012. PLoS Med 2014;11:e1001742
42. Finfer S, Liu B, Chittock DR, et al.; NICE- hypoglycaemia and mortality in type 2 diabetes: 72. Lipska KJ, Warton EM, Huang ES, et al.
SUGAR Study Investigators. Hypoglycemia and retrospective epidemiological analysis of the HbA1c and risk of severe hypoglycemia in type
risk of death in critically ill patients. N Engl J Med ACCORD study. BMJ 2010;340:b4909 2 diabetes: the Diabetes and Aging Study. Di-
2012;367:1108–1118 58. Khunti K, Davies M, Majeed A, Thorsted BL, abetes Care 2013;36:3535–3542
43. Macrae D, Grieve R, Allen E, et al.; CHiP Wolden ML, Paul SK. Hypoglycemia and risk of 73. Monami M, Dicembrini I, Kundisova L,
Investigators. A randomized trial of hyperglyce- cardiovascular disease and all-cause mortality in Zannoni S, Nreu B, Mannucci E. A meta-analysis
mic control in pediatric intensive care. N Engl J insulin-treated people with type 1 and type 2 of the hypoglycaemic risk in randomized con-
Med 2014;370:107–118 diabetes: a cohort study. Diabetes Care 2015; trolled trials with sulphonylureas in patients
44. Gerstein HC, Miller ME, Byington RP, et al.; 38:316–322 with type 2 diabetes. Diabetes Obes Metab
Action to Control Cardiovascular Risk in Diabe- 59. Chow E, Bernjak A, Williams S, et al. Risk of 2014;16:833–840
tes Study Group. Effects of intensive glucose cardiac arrhythmias during hypoglycemia in pa- 74. Weinstock RS, Xing D, Maahs DM, et al.;
lowering in type 2 diabetes. N Engl J Med tients with type 2 diabetes and cardiovascular T1D Exchange Clinic Network. Severe hypogly-
2008;358:2545–2559 risk. Diabetes 2014;63:1738–1747 cemia and diabetic ketoacidosis in adults with
45. Patel A, MacMahon S, Chalmers J, et al.; 60. Reno CM, Daphna-Iken D, Chen YS, type 1 diabetes: results from the T1D Exchange
ADVANCE Collaborative Group. Intensive blood VanderWeele J, Jethi K, Fisher SJ. Severe clinic registry. J Clin Endocrinol Metab 2013;98:
glucose control and vascular outcomes in pa- hypoglycemia-induced lethal cardiac arrhyth- 3411–3419
tients with type 2 diabetes. N Engl J Med mias are mediated by sympathoadrenal activa- 75. Siebenhofer A, Plank J, Berghold A, et al.
2008;358:2560–2572 tion. Diabetes 2013;62:3570–3581 Short acting insulin analogues versus regular
46. Duckworth W, Abraira C, Moritz T, et al.; 61. Wild D, von Maltzahn R, Brohan E, human insulin in patients with diabetes melli-
VADT Investigators. Glucose control and vascu- Christensen T, Clauson P, Gonder-Frederick L. tus. Cochrane Database Syst Rev 2006;2:
lar complications in veterans with type 2 diabe- A critical review of the literature on fear of hy- CD003287
tes. N Engl J Med 2009;360:129–139 poglycemia in diabetes: Implications for diabe- 76. Horvath K, Jeitler K, Berghold A, et al. Long-
47. Mellbin LG, Ryd én L, Riddle MC, et al.; tes management and patient education. Patient acting insulin analogues versus NPH insulin
ORIGIN Trial Investigators. Does hypoglycaemia Educ Couns 2007;68:10–15 (human isophane insulin) for type 2 diabetes
increase the risk of cardiovascular events? A re- 62. Johnson SR, Cooper MN, Davis EA, Jones mellitus. Cochrane Database Syst Rev 2007;2:
port from the ORIGIN trial. Eur Heart J 2013;34: TW. Hypoglycaemia, fear of hypoglycaemia CD005613
3137–3144 and quality of life in children with Type 1 diabe- 77. Pedersen-Bjergaard U, Kristensen PL, Beck-
48. Deckert T, Poulsen JE, Larsen M. Prognosis tes and their parents. Diabet Med 2013;30: Nielsen H, et al. Effect of insulin analogues on
of diabetics with diabetes onset before the age 1126–1131 risk of severe hypoglycaemia in patients with
of thirty-one. I. Survival, causes of death, and 63. Gerich J, Davis J, Lorenzi M, et al. Hormonal type 1 diabetes prone to recurrent severe hypo-
complications. Diabetologia 1978;14:363–370 mechanisms of recovery from insulin-induced glycaemia (HypoAna trial): a prospective, ran-
49. Tunbridge WM; Medical Services Study hypoglycemia in man. Am J Physiol 1979;236: domised, open-label, blinded-endpoint crossover
Group and British Diabetic Association. Factors E380–E385 trial. Lancet Diabetes Endocrinol 2014;2:553–
contributing to deaths of diabetics under fifty 64. Cryer PE. Mechanisms of hypoglycemia- 561
years of age. Lancet 1981;2:569–572 associated autonomic failure in diabetes. N 78. Johnson SR, Cooper MN, Jones TW, Davis
50. Laing SP, Swerdlow AJ, Slater SD, et al. The Engl J Med 2013;369:362–372 EA. Long-term outcome of insulin pump therapy
British Diabetic Association Cohort Study, I: all-cause 65. Chan O, Sherwin R. Influence of VMH fuel in children with type 1 diabetes assessed in a
mortality in patients with insulin-treated diabetes sensing on hypoglycemic responses. Trends En- large population-based case-control study. Dia-
mellitus. Diabet Med 1999;16:459–465 docrinol Metab 2013;24:616–624 betologia 2013;56:2392–2400
51. Patterson CC, Dahlquist G, Harjutsalo V, 66. Gold AE, MacLeod KM, Frier BM. Fre- 79. Pickup JC, Sutton AJ. Severe hypoglycaemia
et al. Early mortality in EURODIAB population- quency of severe hypoglycemia in patients and glycaemic control in type 1 diabetes: meta-
based cohorts of type 1 diabetes diagnosed in with type I diabetes with impaired awareness analysis of multiple daily insulin injections
childhood since 1989. Diabetologia 2007;50: of hypoglycemia. Diabetes Care 1994;17:697– compared with continuous subcutaneous insu-
2439–2442 703 lin infusion. Diabet Med 2008;25:765–774
52. Jacobson AM, Musen G, Ryan CM, et al.; 67. Clarke WL, Cox DJ, Gonder-Frederick LA, 80. Doyle EA, Weinzimer SA, Steffen AT, Ahern
Diabetes Control and Complications Trial/ Julian D, Schlundt D, Polonsky W. Reduced JA, Vincent M, Tamborlane WV. A randomized,
Epidemiology of Diabetes Interventions and awareness of hypoglycemia in adults with prospective trial comparing the efficacy of con-
Complications Study Research Group. Long-term IDDM. A prospective study of hypoglycemic fre- tinuous subcutaneous insulin infusion with mul-
effect of diabetes and its treatment on cognitive quency and associated symptoms. Diabetes tiple daily injections using insulin glargine.
function. N Engl J Med 2007;356:1842–1852 Care 1995;18:517–522 Diabetes Care 2004;27:1554–1558
53. Feltbower RG, Bodansky HJ, Patterson CC, 68. Høi-Hansen T, Pedersen-Bjergaard U, 81. Hirsch IB, Bode BW, Garg S, et al.; Insulin
et al. Acute complications and drug misuse are im- Thorsteinsson B. Classification of hypoglycemia Aspart CSII/MDI Comparison Study Group. Con-
portant causes of death for children and young awareness in people with type 1 diabetes in tinuous subcutaneous insulin infusion (CSII) of
adults with type 1 diabetes: results from the York- clinical practice. J Diabetes Complications insulin aspart versus multiple daily injection of
shire Register of Diabetes in Children and Young 2010;24:392–397 insulin aspart/insulin glargine in type 1 diabetic
Adults. Diabetes Care 2008;31:922–926 69. The Diabetes Control and Complications patients previously treated with CSII. Diabetes
54. Skrivarhaug T, Bangstad HJ, Stene LC, Trial Research Group. The effect of intensive Care 2005;28:533–538
Sandvik L, Hanssen KF, Joner G. Long-term mor- treatment of diabetes on the development 82. Bolli GB, Kerr D, Thomas R, et al. Compari-
tality in a nationwide cohort of childhood-onset and progression of long-term complications in son of a multiple daily insulin injection regimen
type 1 diabetic patients in Norway. Diabetologia insulin-dependent diabetes mellitus. N Engl J (basal once-daily glargine plus mealtime lispro)
2006;49:298–305 Med 1993;329:977–986 and continuous subcutaneous insulin infusion
care.diabetesjournals.org International Hypoglycaemia Study Group 1591

(lispro) in type 1 diabetes: a randomized open glucose monitoring: accuracy and reliability of hy- 94. Russell SJ, El-Khatib FH, Sinha M, et al. Out-
parallel multicenter study. Diabetes Care 2009; poglycemia detection in patients with type 1 di- patient glycemic control with a bionic pancreas in
32:1170–1176 abetes at high risk of severe hypoglycemia. type 1 diabetes. N Engl J Med 2014;371:313–325
83. Fatourechi MM, Kudva YC, Murad MH, Diabetes Technol Ther 2013;15:371–377 95. Ang M, Meyer C, Brendel MD, Bretzel RG,
Elamin MB, Tabini CC, Montori VM. Clinical re- 89. Bergenstal RM, Klonoff DC, Garg SK, et al.; Linn T. Magnitude and mechanisms of glucose
view: Hypoglycemia with intensive insulin ther- ASPIRE In-Home Study Group. Threshold- counterregulation following islet transplanta-
apy: a systematic review and meta-analyses of based insulin-pump interruption for reduc- tion in patients with type 1 diabetes suffering
randomized trials of continuous subcutaneous tion of hypoglycemia. N Engl J Med 2013; from severe hypoglycaemic episodes. Diabeto-
insulin infusion versus multiple daily injections. 369:224–232 logia 2014;57:623–632
J Clin Endocrinol Metab 2009;94:729–740 90. Ly TT, Nicholas JA, Retterath A, Lim EM, 96. UK Prospective Diabetes Study (UKPDS)
84. Ramotowska A, Golicki D, D żygało K, Davis EA, Jones TW. Effect of sensor-augmented Group. Intensive blood-glucose control with sul-
Szypowska A. The effect of using the insulin insulin pump therapy and automated insulin phonylureas or insulin compared with conven-
pump bolus calculator compared to standard suspension vs standard insulin pump therapy tional treatment and risk of complications in
insulin dosage calculations in patients with on hypoglycemia in patients with type 1 diabe- patients with type 2 diabetes (UKPDS 33). Lan-
type 1 diabetes mellitus - systematic review. tes: a randomized clinical trial. JAMA 2013;310: cet 1998;352:837–853
Exp Clin Endocrinol Diabetes 2013;121:248–254 1240–1247 97. Ohkubo Y, Kishikawa H, Araki E, et al. Inten-
85. Battelino T, Phillip M, Bratina N, Nimri R, 91. Maahs DM, Calhoun P, Buckingham BA, sive insulin therapy prevents the progression of
Oskarsson P, Bolinder J. Effect of continuous et al.; In Home Closed Loop Study Group. A ran- diabetic microvascular complications in Japa-
glucose monitoring on hypoglycemia in type 1 domized trial of a home system to reduce noc- nese patients with non-insulin-dependent dia-
diabetes. Diabetes Care 2011;34:795–800 turnal hypoglycemia in type 1 diabetes. betes mellitus: a randomized prospective 6-year
86. Pickup JC, Freeman SC, Sutton AJ. Glycae- Diabetes Care 2014;37:1885–1891 study. Diabetes Res Clin Pract 1995;28:103–117
mic control in type 1 diabetes during real time 92. Yeh HC, Brown TT, Maruthur N, et al. Com- 98. Nathan DM, Cleary PA, Backlund JY, et al.;
continuous glucose monitoring compared with parative effectiveness and safety of methods Diabetes Control and Complications Trial/Epi-
self monitoring of blood glucose: meta-analysis of insulin delivery and glucose monitoring for demiology of Diabetes Interventions and Com-
of randomised controlled trials using individual diabetes mellitus: a systematic review and plications (DCCT/EDIC) Study Research Group.
patient data. BMJ 2011;343:d3805 meta-analysis. Ann Intern Med 2012;157: Intensive diabetes treatment and cardiovascu-
87. Langendam M, Luijf YM, Hooft L, Devries JH, 336–347 lar disease in patients with type 1 diabetes. N
Mudde AH, Scholten RJ. Continuous glucose 93. Leelarathna L, Dellweg S, Mader JK, et al.; Engl J Med 2005;353:2643–2653
monitoring systems for type 1 diabetes mellitus. AP@home Consortium. Day and night home 99. Holman RR, Paul SK, Bethel MA, Matthews
Cochrane Database Syst Rev 2012;1:CD008101 closed-loop insulin delivery in adults with type DR, Neil HA. 10-year follow-up of intensive glu-
88. Bay C, Kristensen PL, Pedersen-Bjergaard U, 1 diabetes: three-center randomized crossover cose control in type 2 diabetes. N Engl J Med
Tarnow L, Thorsteinsson B. Nocturnal continuous study. Diabetes Care 2014;37:1931–1937 2008;359:1577–1589