• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information

Q 28What specific advice can be given to adults with Type 1 diabetes for the management and prevention of hypoglycaemia?

Author / Title / Reference / YrYale JF, Begg I, Gerstein H, Houlden R, Jones H, Maheux P, Pacaud D 2001 Canadian diabetes association clinical practice guidelines for the prevention and management of hypoglycaemia in diabetes. Canadian Journal of Diabetes 26:22–35
N=47 studies of various design in people with type 1 diabetes
Research DesignSystematic review - Guidelines
AimTo determine practical management options to reduce the effect of hypoglycaemia
PopulationType 1 diabetes
InterventionManagement and prevention of hypoglycaemia by various interventions
ComparisonVaried between studies
OutcomeCorrection or reduction in frequency of hypoglycaemia
CharacteristicsVaried between studies
ResultsHypoglycaemia occurs on average in people with type 1 diabetes at a frequency of approximately 2 episodes per week.
Hypoglycaemia and insulin therapy
Intensive vs. conventional insulin therapy
Hypoglycaemia is the most common effect of intensive insulin therapy in type 1 diabetes. Some studies have reported an increase in severe hypoglycaemia, although this is not universal
Rapid-acting insulin analogues vs. human regular insulin
Studies have found no differences in onset, magnitude and temporal pattern of the physiological, symptomatic and counter- regulatory hormonal responses to acute hypoglycaemia induced by regular human insulin compared with the rapid acting insulin analogues
Animal vs. human insulin
A number of studies have suggested no significant clinical difference in the symptomatic response to or in the frequency of hypoglycaemia between animal and human insulin.
Lifestyle factors
Studies suggest self-management behaviours, less food, more insulin and more activity are associated with 85% of hypoglycaemic episodes.
Food and snacks
Patients on fixed-dose insulin regimens should have an individualised meal and activity plan developed, that the person can and will follow.
Patients should be taught how to make adjustments to insulin dosage, diet and physical activity in response to blood glucose levels
Bedtime snack may be needed to avoid nocturnal hypoglycaemia. One study showed that the addition of protein to carbohydrate at bedtime has not demonstrated a reduction in hypoglycaemia.
Prepared snack bars with cornstarch have demonstrated some effectiveness in reducing overnight hypoglycaemia
High protein content (vs. high fat content) of an evening meal has also demonstrated some protection against nocturnal hypoglycaemia
Low-to moderate-intensity exercise lowers glucose levels during and after activity, increasing the risk of a hypoglycaemic episode.
This effect can be altered by modifying diet, insulin and type and timing of the exercise.
Self monitoring of glucose level before during and after exercise is important for establishing the patient’ s response to exercise and guideline the appropriate management of exercise.
Hypoglycaemia unawareness and glucose counter regulation
Major risk factors for severe hypoglycaemia are: prior episode of severe hypoglycaemia, HbA1c less than 6%, hypoglycaemia unawareness, long duration of diabetes and autonomic neuropathy
Severe hypoglycaemic episodes occur mainly at night, or in the absence of awareness that makes patients alert to correct their glucose levels
Following the first few years of diagnosis when glucagon responses to hypoglycaemia are lost, patients depend on sympathodrenal responses for glucose counter regulation and hypoglycaemia awareness.
Autonomic neuropathy has been shown to be an independent risk factor for severe hypoglycaemia, and further reduces epinephrine and norepinephrine responses to hypoglycaemia.
The incidence of prior hypoglycaemic episodes is a crucial factor leading to hypoglycaemia unawareness due to a worsening in the defect of the hormonal responses to hypoglycaemia, leading to a reduction in the self-detection of hypoglycaemia.
Strict avoidance of hypoglycaemia from 2 days to 3 months has been associated with an improvement in the recognition of severe hypoglycaemia in the counter regulatory hormone responses, or both
Blood glucose awareness training (BGAT) may have a positive effect on increasing accurate detection and treatment of hypoglycaemia. The BGAT programme involves instruction in interpretation of physical symptoms, performance cues and moods and feelings as internal cues to blood glucose awareness; it also involves instruction on food, exercise, insulin dosage and action, time of day, and last blood glucose reading as external cues to estimate blood glucose level. BGAT allowed reduced-awareness subjects (with some, but not many, hypoglycaemic symptoms) to detect a greater percentage of blood glucose levels less than 3.9 mmol/l (35–45%, p=0.006)
Ingestion of caffeine increases the sympathoadrenal and symptomatic responses during moderate hypoglycaemia. However, there is no data on the impact of caffeine consumption on the frequency of severe hypoglycaemic episodes.
Long-term complications of severe hypoglycaemia
The potential long-term complications of severe hypoglycaemia are mild intellectual impairment and permanent neurologic sequelae such as haemiparesis and pontine dysfunction. The latter are rare and reported only in case studies. available prospective studies did not find an association between intensive diabetes management and cognitive function. Since intensive management is linked with more frequent severe hypoglycaemia, it is extrapolated that, similarly, severe hypoglycaemia is not linked with a decrease in cognitive function. However, the time interval studied was relatively short (5 years)
However, retrospective studies suggested a link with frequent severe hypoglycaemia (5 or more episodes) and a decrease in intellectual performance. People with diabetes with a history of severe hypoglycaemia, when compared to matched subjects with diabetes without severe hypoglycaemia and patients without diabetes, were found to perform more poorly in a number of intellectual tests: immediate memory and finger tapping, word recall test and verbal fluency, performance IQ, Weschler
Performance test and Trail Making Test. In some studies, a correlation was found between the frequency of severe hypoglycaemic reactions and the performance IQ.
Treatment for hypoglycaemia
Little evidence is available to support the widely recommended treatment for acute hypoglycaemia as 10g of a variety of carbohydrates.
More recent evidence suggests that 15g of glucose (monosaccharide) is required to produce a rise in blood glucose of approximately 2.1 mmol/l within 20 minutes, with adequate symptom relief for most people.
A 20g oral glucose dose will produce a blood glucose increment of approximately 3.6 mmol/L at 45 minutes.
Other treatments such as milk and orange juice are slower to raise blood glucose levels and provide symptom relief
Glucose gel is a slower treatment (less than 1mmol/l rise at 20 mins) and therefore needs swallowing to have a significant effect
No evidence is available for the buccal administration of glucose gel as absorption through the mucosa is minimal, if any Patients taking alpha-glucosidase inhibitors should use glucose (dextrose) tablets, or if unavailable, milk or honey to treat hypoglycaemia.

Physical Training and Exercise
In anyone treated with insulin, recommendations regarding alterations of diet, insulin regimen, injection sites and self-monitoring should be appropriate for the general level of physical activity or specific types of exercise undertaken. Oral agent doses may need to be decreased. [Grade D, consensus]
Self-monitoring of glucose level before, during and especially for many hours after exercise, is important for establishing the patient’ s response to exercise and guiding the appropriate management of exercise. [Grade D, consensus]
General advice regarding physical activity include:
For those on insulin or insulin secretagogues, ingest rapidly absorbed carbohydrate if pre-exercise glucose level is under 5 mmol/L.
For those on insulin injections, administer insulin into a site away from the most actively exercising extremities. [Grade D, consensus]

Insulin Use in Type 1 Diabetes
All patients currently on or starting intensive insulin programs should be counselled about the risk and prevention of hypoglycaemia. They should be advised to perform frequent blood glucose monitoring and receive appropriate instruction on how to make adjustments in insulin dosage, diet and physical activity in response to blood glucose levels. The diabetes health care team should review the patient’ s experience with hypoglycaemia at each visit. This should include an estimate of cause, frequency, symptoms, and recognition, severity and treatment. [Grade D, consensus]
To reduce the risk of symptomatic nocturnal hypoglycaemia, patients should periodically monitor overnight blood glucose levels at a time that corresponds with the peak action time of their overnight insulin and consume a bedtime snack with at least 15 g carbohydrate and protein if the bedtime blood glucose level is under 7 mmol/L. [Grade D, consensus]
Regular or a rapid-acting insulin analogue, or both, can be used before meals in intensified therapy (multiple daily injections and CSII). Lispro has been associated with lower post-prandial glucose levels and lower rates of hypoglycaemia than regular insulin. [Grade A, Level 1]
Aspart insulin has been associated with lower rates of hypoglycaemia compared to human regular insulin. [Grade B, Level 2]

Patients experiencing frequent hypoglycaemic episodes on regular insulin should be tried on a fast-acting insulin analogue. [Grade D, consensus]
Lispro is the preferred insulin for use in CSII. [Grade B, Level 2]
Substituting a rapid-acting insulin analogue for human regular insulin at suppertime may prevent the delayed night time effect of regular insulin and reduce the risk of nocturnal hypoglycaemia (45–48).Administering basal insulin at bedtime rather than at suppertime, or instituting CSII, may also reduce the risk of nocturnal hypoglycaemia. [Grade D, consensus]
Risk factors for severe hypoglycaemia should be identified in people with type 1 diabetes so that appropriate strategies can be used to prevent hypoglycaemia. (Grade A). Established risk factors include a history of previous severe hypoglycaemic event [Level 1], a greater reduction in HbA1c [Level 1] and recurrent previous hypoglycaemic reactions [Level 1]
The patients at high risk should be informed of their risk, counselled along with their significant others on avoidance and treatment (including glucagon), and if necessary have their insulin regimen adjusted appropriately to avoid these events. [Grade D, consensus]
During insulin therapy of type 1 diabetes, the frequency of mild hypoglycaemic episodes should be minimized, particularly in those at high risk, in an attempt to reduce the development of hypoglycaemia unawareness. [Grade D, consensus]
In individuals with hypoglycaemia unawareness, the following strategies should be implemented to reduce the risk of hypoglycaemia, the risk of hypoglycaemia unawareness, and to increase physiologic counter-regulatory responses to hypoglycaemia:
increased frequency of glucose monitoring, increase in the glucose targets, and multiple insulin injections with increased glucose targets [Grade D, level 4]
Patients switching from animal to human insulin do not require counselling about any change in frequency or perception of hypoglycaemia. [Grade A, Level 1]
In hospitalised patients, efforts must be made to ensure that patients on insulin have ready access to an appropriate form of glucose at all times, particularly when NPO or during diagnostic procedures. [Grade D]

Treatment of hypoglycaemia
Mild to moderate hypoglycaemia should be treated by the oral ingestion of 15 g of carbohydrate, preferably as glucose or sucrose tablets or solution or hydrolysed polysaccharide. These are preferable to orange juice and glucose gels [Grade B, level 2]. Patients should be encouraged to wait 15 minutes, retest blood glucose and retreat with another 15 g of glucose if the blood glucose remains less than 4.0 mmol/L. [Grade D, consensus]
Severe hypoglycaemia in a conscious person should be treated by the oral ingestion of 20 g of carbohydrate, preferably as glucose tablets or equivalent. Patients should be encouraged to wait 15 minutes, retest blood glucose and retreat with another 15 g glucose if it remains less than 4.0 mmol/L. [Grade D, consensus]
To prevent repeated hypoglycaemia, the person should have in addition to the fast-acting treatment above, once the hypoglycaemia has been reversed, their usual meal or snack. A snack (including 15 g of carbohydrate and a protein source) is recommended if a meal is more than 1 hour away and in the absence of complicating factors. [Grade D, consensus]
All patients currently on or starting therapy with insulin or insulin secretagogues should be counselled about the recognition and prevention of drug-induced hypoglycaemia. [Grade D, consensus]
Hierarchy of Evidence GradingIa (various)
CommentsAn Expert committee was established including healthcare professionals from diabetologists to dietitians, and methodologists from Canada.
The methodologists then evaluated the literature and developed recommendations on the key areas, assigning levels of evidence to the relevant citations and making grading recommendations based on the validated methods. This evidence was then summarised in a series of documents for the Expert committee.
All recommendations reviewed by 3 methodologists not directly involved in initial assessment of evidence and grading of recommendations.
Recommendations are an update of the Canadian medical association 1998 clinical guidelines for the management of diabetes.
Hypoglycaemic symptoms (% frequency):
Neurogenic (Autonomic)Neuroglycopenic
Trembling (32–78%)Difficulty concentrating (31–75%)
Palpitations (8–62%)Confusion (13–53%)
Sweating (47–84%)Weakness (28–71%)
Anxiety (10–44%)Drowsiness (16–33%)
Hunger (39–49%)Vision changes (24–60%)
Nausea (5–20%)Difficulty speaking (7–41%)
Tingling (10–39%)Headache (24–36%)
Dizziness (11–41%)
Tiredness (38–46%)
No details given of search strategy employed by stakeholders, or even if a systematic search was employed.
No indication given as to whether papers were rejected due to poor quality
Details not provided for magnitude of effect in each study, or population size.
Trials includedSee original study
Reference / Citation98
Author / Title / Reference / YrBroers, S., Van Vliet, K. P., Everaerd, W., Le Cessie, S., & Radder, J. K. 2002, “Modest contribution of psychosocial variables to hypoglycaemic awareness in Type 1 diabetes”, Journal of Psychosomatic Research, vol. 52, no. 2, pp. 97–106.
Research DesignCase control study
AimA study to assess relationships between hypoglycaemic awareness and diabetes-related, psychosocial and demographic characteristics
PopulationType 1 diabetes
InterventionThe study assesses the contribution of a range of demographic, diabetes related, and psychological characteristics to hypoglycaemic awareness. Over a 4 to 6 week period. The parameters include Negative affectivity, Somatic awareness, Bustle and variety of the daily life, and degree each of 20 symptoms; as assessed by questionnaire. Sensitivity and specificity of the symptom beliefs in terms of the degree of agreement between ‘actual’ and ‘believed’ hypoglycaemic symptoms were made for every patient by the use of measured blood glucose values from the handheld computer with data dichotomised to Hypoglycaemia (less than 3.9 mmol/l) versus no hypoglycaemia (greater than 3.9 mmol/l),
ComparisonComparison of different scores in the same parameters
OutcomeHypoglycaemic awareness. The percentage of recognised hypoglycaemic episodes (% RH) was defined as the percentage of measured blood glucose values less than 3.9 mmol/l that was accompanied by an estimation less than 3.9 mmol/l, or by an estimation within 20% of the measured value, for each of the test occurrences
CharacteristicsAge =42yrs, Male =55%, Duration of diabetes =21.4 yrs, HbA1c= 7.5%
ResultsUnivariate analysis
A moderate but significant positive correlation (r=.52, p=.000) between hypoglycaemic awareness and the number of actual hypoglycaemic symptoms (see was observed.
The four symptoms that were actually related to hypoglycaemia for the most patients were: trembling (41.1%), problems with coordination (34.7%), difficulty concentrating (33.7%) and dizziness (32.6%). The four symptoms that were believed to be hypoglycaemic symptoms by the most patients were: difficulty concentrating (69.5%), sweating (54.7%), confusion (51.6%) and irritability (43.2%).
Of the diabetes-related variables, disease duration (r= −.25, p=0.01) and antecedent hypoglycaemia (r=−.20, p=0.05) were both negatively associated to the % RH.
Multivariate regression
Hypoglycaemic awareness was predicted by the duration of the diabetes B=−0.79 (SE= 0.37)(p=0.04), the use of CSII versus insulin injections B= 24.7 (SE= 11.9) (p=0.04), and the sensitivity of the symptom beliefs (B=0.21 (SE =0.11) (p=0.07).
Only 17% of the variance in hypoglycaemic awareness was explained by the model
Hierarchy of Evidence GradingIIa
CommentsThe indication that those who withdrew from the study had a higher HbA1c level than those participating suggests that people in the study already had good glycaemic control and may be less prone to hypoglycaemia with the outcomes thus being diminished.
The measure of assessment of psychological characteristics was not validated and could have had unknown effects on the results.
Only 65 people were included in final multivariate model with data available for all parameters
People who used an insulin pump recognised hypoglycaemia more often than those who used insulin injections.
Possibly, the single questions used were too crude to assess the underlying concepts, and a validated questionnaire would have led to other results.
People may become more sensitive to physical sensations when they regularly take time to ‘listen’ to their body, and may learn which symptoms represent hypoglycaemia by the keeping of a ‘symptom diary.’
Only a small proportion of the variance in hypoglycaemic awareness was explained in the present study. It seems likely that other variables, which have not been measured, play a role in hypoglycaemic awareness.
Reference / Citation
Author / Title / Reference / YrKingsley BT, Weinger K, Bajaj M, Levy CJ, Simonson DC, Quigley M, Cox DJ, Jacobson AM 1999 Blood glucose awareness training and epinephrine responses to hypoglycaemia during intensive treatment in Type 1 diabetes. Diabetes Care 7:1022–1028
Research DesignRandomised controlled trial
AimA study to assess the effectiveness of blood glucose awareness training to reduce incidence of Hypoglycaemia
PopulationType 1 diabetes
Intervention8 session group education programme in Blood glucose awareness training (BGAT) (using the revised BGAT-3 version involving instruction in interpretation of physical symptoms, performance cues and moods, and feelings as internal cues to blood glucose awareness, also including instruction on food, exercise, insulin dosage and action, time of day and last blood glucose reading as external cues to estimate blood glucose level.)
ComparisonAn equivalent number and duration of sessions in cholesterol education (control)
OutcomeGlycaemic control measured by HbA1c and hypoglycaemia frequency
CharacteristicsSex (M/F): 23/24; Age: 34±8 years; Duration of diabetes: 9.0±3 (range 3–15) years; HbA1c 9.0±1.2%
ResultsGlycaemic control and hypoglycaemia frequency
During 4 months IDT, glycaemic control measured by HbA1c improved in BGAT and control groups. HbA1c fell from 9.1±1.4 to 7.9±1.1% (p less than 0.001) in the BGAT group 9.0±1.1 to 7.8±8.0% (p less than 0.001) in the controls. No significant difference was seen between the two groups
Hypoglycaemia frequency measured by the daily number of readings less than 3.9 mmol/L increased in both groups from 0.45±0.06 to 0.69±0.07 episodes per day in the BGAT group (p less than 0.001) and 0.05±0.08 to 0.68±0.06 episodes in controls (p less than 0.05). Differences between the groups were not significant.
No differences were recorded in the severity of hypoglycaemia between the two intervention groups.
Counter-regulatory hormones
Epinephrine levels increased from baseline following IDT in both groups. Following IDT epinephrine levels were significantly lower in the control group compared with the BGAT group at blood glucose levels of 2.8 mmol/L (1,162±165 vs. 1,850±255 pmol/l) and 2.2 mmol/l (2,217±263 vs. 3,220±382 pmol/l) p less than 0.05 between groups.
Levels of norepinephrine, ACTH, cortisol, and hGH did not differ between control and BGAT groups before or after IDT and education.
Symptom scores
Neurogenic and neuroglycopenic symptom scores did not differ between control and BGAT groups before or after IDT.
Self-reported neurogenic symptoms were significantly reduced in subjects undergoing BGAT during IDT compared to the control group (p less than 0.004), whereas neuroglycopenic symptoms did not differ in BGAT and control groups.
Blood glucose estimation accuracy was similar in control and BGAT groups before IDT. Following IDT blood glucose estimations errors did not differ significantly between groups.
BGAT subjects had a greater improvement in detection of low blood glucose levels and fewer undetected low blood glucose readings.

Subgroup analysis of patients at high risk of hypoglycaemia (n=26)
HbA1c improved from 9.0±0.4 to 7.4±0.2% in the control group (p less than 0.01) and from 8.8±0.5 to 7.5±0.2% in the BGAT group (p less than 0.01), but no significance difference was seen between the two groups.
No difference was seen between the groups in hypoglycaemia frequency following BGAT compared to controls
An increase in the proportion of hypoglycaemic episodes ≤2.8 mmol/l was seen in the control but not BGAT group; however, the difference between the two groups was not significant.
Counter regulatory hormones
Epinephrine levels did not differ between control and BGAT groups prior to IDT.
After 4 months of IDT the change in epinephrine response to hypoglycaemia with IDT differed in the control and BGAT groups (repeated measure ANOVA, F=4.4, p less than 0.05)
Levels of norepinephrine, ACTH, cortisol and hGH did not differ in control and BGAT groups before or after BGAT.
Symptom scores
These did not differ between the two groups; no significant decrease in neurogenic or neuoglycopenic symptoms was seen in the BGAT group compared to the controls (p=0.11 and 0.93 respectively)
Blood glucose estimation
Accuracy did not differ in control and BGAT groups before or after 4 months intensive diabetes treatment.
Subjects undergoing BGAT had fewer undetected low blood glucose readings than those in the control group (p less than 0.04)
Hierarchy of Evidence GradingIb
Comments60 subjects enrolled in the study. 8 patients dropped out because of moving from the area (n=3), or having a change in job and time involvement (n=4) or non-study-related injury (n=1). 5 subjects experienced technical difficulties either during one of the clamp procedures (n=3) or with laboratory assays (n=2). Final study sample therefore consisted of 47 subjects.
None of the remaining patients had any evidence of complications.
Subjects were followed in an outpatient clinic over a 4 to 5 month period.
Subjects were seen monthly by study physicians, nurse educators and a nutritionist, and had weekly telephone contact with their nurse educator to optimise glycaemic control. Patients took 3–5 insulin injections/day and performed ~ 5 home blood glucose measurements/day.
Before and after 4 months of intensive diabetes treatment, subjects underwent paired identical hypoglycaemic insulin clamp procedures.
At baseline at each glucose level, subjects completed a 35-item self-administered mood and symptom questionnaire (MSQ), rating each item on a 7 point likert scale, with 0 standing for feeling the symptom “ not at all” and 6 “ a lot”. Mean score of trembling, sweating, pounding heart and fast pulse was used to represent neurogenic symptoms, and the mean of being light headed, difficulty concentrating, uncoordinated, confused and feeling weak to represent neuroglycopenic symptoms.
On completion of questionnaire, patients estimated and recorded their blood glucose level.
HbA1c was measured at baseline and at each monthly visit. Home blood glucose meter readings were downloaded to a p.c., providing data on glucose levels for 4 weeks, and were used to analyse frequency of hypoglycaemia.
An estimate of blood glucose within 20% of the actual meter reading was taken as correct
Validated BGAT protocol applied.
Validated symptom and mood questionnaire used.
No details of randomisation procedure or concealment of allocation
Blinding of patients and investigators not discussed.
Findings on hypoglycaemia frequency based on home blood glucose testing, leaving potential for subjects to not have treated or recorded all of their low blood glucose readings and no data is available on nocturnal hypoglycaemic episodes.
Reference / Citation170
Author / Title / Reference / YrFritsche S, Stumvoll M, Renn W, Schmulling RM 1998 Diabetes teaching program improves glycemic control and preserves perception of hypoglycaemia. Diabetes Research and Clinical Practice 40:129–135
Research DesignCase series
AimTo track the progression of hypoglycaemia in people undergoing a teaching programme
PopulationType 1 diabetes
InterventionStructured inpatient diabetes education programme
ComparisonNot applicable
OutcomeHbA1c, blood glucose accuracy index, sensitivity and prevalence of low blood glucose
CharacteristicsAge: 33.7±11.7 years; Sex (M/F): 26/28; duration of diabetes: 11.7±9.3 years
Results33 patients (61%) kept a constant diary and estimated blood glucose regularly during the study. The remaining 21 patients were excluded from the study.
Of the excluded patients 38% had a history of severe hypoglycaemia. These patients did not differ from the patients included in the study.
Of the remaining patients, those with repeated severe hypoglycaemia were older and had longer duration of diabetes, compared to patents without repeated severe hypoglycaemia, although levels of HbA1c did not differ at baseline.
Blood glucose levels
Accuracy of blood glucose perception in all the patients did not change following introduction of DTTP and IT (p=0.25).
Taken according to history of hypoglycaemia, patients with a history of repeated hypoglycaemia were able to improve the accuracy of blood glucose perception following DTTP, unlike those with no history.
No significant improvement was seen in blood glucose sensitivity following a teaching programme either when considering the entire cohort, or when patients were divided up according to hypoglycaemia history.
Prevalence of low blood glucose
Prevalence of blood glucose levels less than 3.9 mmol/l did not change following the teaching programme, however there was a tendency towards a lower frequency of low blood glucose readings in group H.
Correlation of sensitivity for blood glucose levels and glycaemic control
There was no linear correlation between the change of sensitivity for low blood glucose levels and HbA1c (r=0.1; p=0.4).
No linear correlation was seen between sensitivity for and prevalence of low blood glucose levels (r=0.3; p=0.1)
Hierarchy of Evidence GradingIII
Comments54 Patients consecutively referred to teaching and treatment programme for intensification of insulin regimen. 33 completed the study. These patients were grouped according to presence (n=11) or absence (n=22) of history of repeated hypoglycaemia.
No participants had clinical evidence of autonomic neuropathy or advanced secondary complications e.g. advanced retinopathy, macroproteinuria or symptomatic macrovascular disease
Patient’ s blood glucose diaries examined at end of study.
Patients underwent a structured interview about the frequency and severity of severe hypoglycaemia (requiring i.v. glucose or s.c. glucagon), results were double checked with relatives and GPs.
The 5 day inpatient program consisted of 25 60-min lessons emphasising insulin dose adaptation for food intake and exercise as well as correction or prevention of hypoglycaemia and hyperglycaemia. Patients learned to estimate carbohydrate content of foods and how to treat and recognise hypoglycaemia.
Patients also specifically leaned how to prevent unrecognised night-time hypoglycaemia. When blood glucose readings were below 8 mmol/l at bedtime patients were instructed to eat a snack with a variable carbohydrate content aiming for a blood glucose level of 8–10 mmol/l before going to bed.
Patients were tested and trained by nurses in their ability to accurately measure blood glucose.
Patients also estimated blood glucose before each measurement ≥ 4 times/day. No specific training was given in blood glucose perception, but received written instructions detailing mechanisms of blood glucose perception.
Instructions also listed actions influencing blood glucose such as amount, type and timing of insulin, food, exercise, alcohol and illness.
Study looking at the ability to alleviate the increase in hypoglycaemia and reduction in blood glucose awareness following improved metabolic control, through the use of structured inpatient diabetes treatment and teaching programme.
Small sample size (n=11 in severe hypoglycaemia group)
Study has high drop out rate, and only compliant patients were included in the final results, therefore there was a positive selection of estimating patients in this study.
Some patients may have already been familiar with estimating blood glucose since two thirds reported to occasionally have relied on blood glucose estimates before the study when adjusting their insulin dose.
Patients with and without a history of repeated severe hypoglycaemia differed in age and diabetes duration, and also in certain aspect of blood glucose awareness.
Reference / Citation171

From: Appendix D, Evidence tables

Cover of Type 1 Diabetes in Adults
Type 1 Diabetes in Adults: National Clinical Guideline for Diagnosis and Management in Primary and Secondary Care.
NICE Clinical Guidelines, No. 15.1.
National Collaborating Centre for Chronic Conditions (UK).
Copyright © 2004, Royal College of Physicians of London.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.