Tuesday , July 17 2018
Home / Resources / Clinical Gems / Managing Clinical Problems in Diabetes, Case Studies, Part 1

Managing Clinical Problems in Diabetes, Case Studies, Part 1

Apr 25, 2011

Edited by Trisha Dunning AM, RN, MEd, PhD, CDE, FRCNA and Glenn Ward MBBS, BSc, DPhil (Oxon), FRACP, FRCPath

ThDCMS30_CG_Imagee aims of the book are to: (1) address commonly encountered diabetes management problems; (2) develop comprehensive responses from a range of relevant health professionals who suggest management approaches relevant to their area of practice. The specific health professionals who provide comments about each case depend on the specific clinical issue; and (3) stimulate thought and discussion.

The target readership is health professionals from a range of professional backgrounds and general as well as specialist professionals such as general practitioners, nurses, dietitians, and podiatrists. The book will be particularly useful for beginner practitioners specializing in diabetes. In addition, it will provide suggestions or food for thought for more experienced practitioners. The cases will be excerpts from the book are all real and are presented exactly as the information was received from the person making the referral. General practitioners, diabetes educators and people with diabetes referred most of the cases; some were self-referrals by people with diabetes. They represent referrals to various diabetic health professionals and concern commonly encountered clinical issues.


Mrs. TP was referred to a tertiary diabetes clinic by a rural GP Please advise about this 50-year-old female with supposed Type 1 diabetes, which she has had for some years. She is on Protaphane (intermediate-acting insulin) 12 units at night. Her blood glucose ranges between 54-108mg/dL (3 and 6 mmol/L) fasting, rising to about 148mg/dL.(8.2 mmol/L) post-food. HbA1c usually 7.4%. She is having hypos (hypo unawareness) at lunchtime, but increased her own morning NovoLog (NovoRapid) because she thought her HbA1c was high. Basically she cannot understand the difference between her blood glucose tests and HbA1c . Her old notes show no GAD antibodies but she did initially present in a comatose state following a 42 lbs (three-stone weight) loss.

What else should I do for this woman?

Diabetes educator: Three key issues need to be addressed:

(1) What type of diabetes does Mrs. TP have?

(2) How can the frequency of hypoglycemic episodes be reduced to enable her to recognize symptoms again?

(3) Mrs. TP should stop driving until her hypoglycemia unawareness resolves.

Mrs. TP’s history is relatively common in tertiary diabetes centers, but may be uncommon in general practice. The type of diabetes Mrs. TP has is unclear. She may have Type 1 diabetes. Up to 80% of people with Type 1 diabetes exhibit either GAD or IA 2 antibodies, which means that 20% do not have such antibodies but may still have Type 1 diabetes. Measuring her C-peptide would demonstrate how much insulin her pancreas produces. A normal or raised C-peptide level suggests Type 2 diabetes, while a low level suggests Type 1 diabetes. Measuring her IA 2 antibodies would also be useful because people with Type 1 diabetes can be positive for IA 2 and not GAD antibodies. Mrs. TP was diagnosed with a 42 pound (three-stone) weight loss, which suggests Type 1 diabetes. Often people with Type 2 diabetes present with weight gain rather than weight loss.

People with Type 1 diabetes are at risk of developing diabetic ketoacidosis (DKA) and this may account for her comatose state on presentation. Alternatively, the coma might have been due to hyperosmolar non-ketonic coma (HONK). There is no mention of ketones or indication of her pH on presentation, which again could suggest the diabetes type. The initial treatment for both DKA and HONK is insulin, which is required permanently in Type 1 and often for several weeks or months following recovery from HONK, after which oral hypoglycemic agents can be slowly introduced once the blood glucose levels are stable. During the recovery period the beta cells start to produce more insulin. If a definitive diagnosis of Type 1 diabetes cannot be made, insulin should be slowly withdrawn over a period of weeks and oral agents introduced.

However, if insulin therapy needs to be maintained, Mrs. TP’s current insulin regimen needs to be revised. Bedtime intermediate-acting insulin has variable effects on the blood glucose. Its peak action can be anywhere from 4-12 hours after administration and last up to 24 hours. The site of administration affects the onset of action and therefore its peak effect. The abdomen is the preferred injection site because the action profile is more predictable.

These are the possible management options:

  • Her prescription could be changed from Protaphane (intermediate acting insulin) to a long acting basal analogue such as Lantus or Levemir to improve the fasting blood glucose levels, which will in turn contribute to a reduction in HbA1c and reduce the need to increase the breakfast dose of NovoLog (NovoRapid). Hypoglycemia before lunch is more likely to be due to her bedtime Protaphane (intermediate acting insulin) than her breakfast NovoLog (NovoRapid) dose, which peaks approximately 90 minutes after administration and is almost completely gone by mid-morning.
  • If Mrs. TP does have Type 2 diabetes, combining the basal analogue insulin with an oral agent such as the sustained release metformin may be the first step towards eventually ceasing insulin therapy and using oral agents to maintain glycemic control.
  • A three-day period of continuous blood glucose monitoring could help identify when Mrs. TP becomes hypoglycemic — which may occur at times other than when she tests her blood glucose — and identify hypoglycemic unawareness. Significant hypoglycemia may mean that Mrs. TP should not drive her car until she can recognize hypoglycemia symptoms and her blood glucose levels are stable, which may impact on her employment and will need to be managed sensitively.

Mrs. TP is not the only person to find it difficult to understand the difference between the HbA1c and capillary blood glucose tests. Many health professionals do not know the difference. HbA1c measures the red blood cell exposure to glucose over a 3-month period. However, a number of factors can affect the accuracy of the HbA1c result. For example, an HbA1c of 7.4% could occur due to the daily blood glucose fluctuations between 3 and 8.2 mmol/L that Mrs. TP reports. Her fasting blood glucose range is a little too low (54-148mg/dL or 4-6 mmol/L would be preferable) and may mean she consumes extra fast-acting carbohydrate to correct hypoglycemia, which in turn could cause a postprandial glucose excursion. Although the fasting blood glucose levels contribute to an HbA1c under 7%, high postprandial readings account for most of the rise in HbA1c. In addition, a number of other medical conditions directly affect the lifespan of red blood cells and consequently the accuracy of the HbA1c . Schneider (2006) described 700 different variables that could possibly affect the HbA1c . Despite these limitations, the HbA1c is a valuable indicator of metabolic control. However, it should be interpreted considering the clinical picture and the day-to-day capillary blood glucose results. The accuracy of Mrs. TP’s blood glucose testing technique should be checked and it is important to ensure that:

  • her meter is calibrated correctly;
  • the strips/electrodes are within the expiry date;
  • her technique is correct; and
  • when she tests — in relation to food — is recorded.

Generally, the accepted time to measure blood glucose after meals is 2 hours after eating. Testing earlier measures the post absorptive state rather than postprandial. If Mrs. TP tests 30 minutes after eating, her results will be falsely raised and will not be a true indication of her postprandial glycemic control. In summary, I suggest the following.

  • Measure her IA 2 antibodies and C-peptide to determine whether she has Type 1 or Type 2 diabetes.
  • Review her medical notes at the time of diagnosis to ascertain:
    • whether ketones were present at diagnosis;
    • and the pH at the time.
  • Check her blood glucose meter and testing technique.
  • Elicit more information about when she tests her blood glucose, administers her insulin, and how she manages hypoglycemia.
  • Check her injection sites.
  • Explore the feasibility of 72 hours of continuous blood glucose monitoring to identify when and how often her blood glucose levels drop below 63mg/dL(3.5 mmol/L) without accompanying symptoms.
  • If the review of her medical record at the time of diagnosis and the results of the recommended blood test indicate she has Type 2 diabetes, start to gradually wean her off insulin and commence oral hypoglycemic agents, in particular, metformin.


Mrs. TP’s story raises several issues: does she really need insulin? I would like more information about how she makes decisions about her NovoLog (NovoRapid) doses. She is on a relatively small basal insulin dose and has an exceptionally good blood glucose range, possibly too low. I would aim to increase the range to 72-126mg/dL (4–7 mmol/L). Is she using appropriate NovoLog (NovoRapid) doses and an algorithm to adjust the doses?

I would determine her lifestyle, weight and activity level to determine whether her insulin sensitivity could be improved and insulin resistance reduced. The consistency of her carbohydrate intake needs to be checked and she may need to be educated about carbohydrate counting, glycemic index (GI) and mixed meals.

Despite her tight control (HbA1c 7.4%), Mrs.TP may need more education about her meals and blood glucose testing, especially the difference between pre- and postprandial test results. I would take the opportunity to review her dietary intake by taking a detailed diet history, which could be compared with accepted nutritional recommendations (50–55% carbohydrate, 30–35% saturated fat, 10–15% protein) (Canadian Diabetes Association 2003; American Diabetes Association 2004; Diabetes Nutrition Study Group 2004). I would use a plate model or food pyramid to demonstrate a balanced diet pictorially. In addition, I would promote physical activity and aim to achieve a body mass index (BMI)/waist circumference within normal limits.


As well as the comments above, several further points could be addressed. First, the disparity of the HbA1c with capillary glucoses might be clarified by measuring serum fructosamine, which represents a glycosylated component of serum proteins, rejecting the average blood glucose over the previous 2-3 weeks while not being influenced by red blood cell turnover or the level of hemoglobin variants. The HbA1c could also be measured by an affinity method, which is less affected by hemoglobin variants, and combined with hemoglobin electrophoresis to explore the presence of such variants. Renal function should be measured, because carbamylated hemoglobin, which is raised according to the severity of renal impairment, can cross-react in some HbA1c assays. It is now clear that some patients glycosylate their own tissues at a higher-than-normal rate for the same average glucose levels.

Second, it needs to be kept in mind that we now know that hypoglycemic unawareness in a percentage of patients can be reversed, and is secondary to the hypoglycemia itself ‘stunning’ the autonomic nervous system. The stunning impairs the development of the typical ‘adrenergic’ symptoms characteristic of hypoglycemia. The way to confirm whether this process is occurring is to concentrate on eliminating hypoglycemia, even if metabolic control is temporarily poor, until the warning signs return. The doses of NovoLog (NovoRapid) insulin were not specified, but reducing the breakfast dose would be critical in this patient to eliminate the lunchtime hypoglycemia. Her target glucose should be 90-144mg/dL (5–8 mmol/L) before meals and any glucose levels less than 4 more than occasionally would trigger a dose readjustment.

Diabetes Educator 2

While accepting the previous comments, I suggest that Mrs. TP’s insulin sensitivity is not in question since her control is very good. Nor does insulin resistance appear to be a problem given her HbA1c and capillary blood glucose range. In addition to the issues already raised, her menopausal status needs to be evaluated. Menopause occurs at around 51 years of age and Mrs. TP is 51, so she is likely to be peri- or postmenopausal and her changing hormone status could affect her blood glucose levels.

Oestrogen production has individual effects on blood glucose, which can often be more variable and contribute to symptoms such as mood changes, fatigue and hot flashes that can be mistaken for hypoglycemia. Menopausal symptoms often occur at night and compromise sleep. Blood glucose testing, including at the time she experiences the symptoms, will help determine the cause. If the symptoms are due to the menopause, Mrs. TP may be unnecessarily consuming extra calories that can compromise her control. 

Her medicines may need to be reviewed and possibly adjusted. The need for other medicines — such as lipid-lowering agents, antihypertensive agents and low-dose aspirin — should be assessed and her weight monitored, because it is common for women to gain weight after menopause.

Women who are obese at the time of menopause are also at greater risk of developing breast cancer due to higher levels of oestrogen as a result of excess abdominal fat (Bernstein 2002). If she smokes she should be counseled to stop. She should discuss menopausal symptom management with her doctor. Short-term hormone replacement therapy (HRT) may be indicated, but the benefits need to be weighed against the risk of breast cancer. If HRT is indicted, the lowest effective dose should be used for the shortest possible time (Khoo and Mahesh 2005). Mrs. TP might like to consider complementary therapies such as Dong Quai, red clover and black cohosh, which help relieve symptoms for some but not all women. Mrs. TP may be at increased risk of osteoporosis after menopause and her diet and activity may need to be revised.

I would also take the opportunity to check when she last had a pap smear, mammogram and general health check, and refer her for the relevant investigations if necessary. Bone density studies might be useful, particularly if she smokes.

Sexual dysfunction in women is a prevalent problem and needs to be considered with Mrs. TP. In most cases, neither the woman nor her health professionals discuss the issue (Ohl 2007). Female dysfunction is more often psychogenic in origin than male sexual dysfunction. However, it has been linked to medical conditions such as hypertension and antihypertensive medications (Burchardt et al. 2002), diabetes (Enzlin et al. 2002), and coronary artery diseases (Salonia et al. 2002). It appears the vascular complications that cause male erectile dysfunction are associated with sexual dysfunction in women, especially at the arousal stage.

In addition, the hypoglycemia could be associated with sexual activity. Taking a careful sexual history could help identify whether Mrs. TP is experiencing any sexual difficulties or has any other stressors, or psychological issues that could be affecting her blood glucose levels and quality of life. Specific management would depend on the findings. The PLISST model (permission, limited information, specific suggestions, and intensive therapy) (Annon 1976; Dunning 1993) is a useful framework for providing sexual counseling and can help health professionals identify their level of competence to do so.

Next Week: Another Case Discussion

For more information on the book, just follow this link to Amazon.com, Managing Clinical Problems in Diabetesalt

Copyright © 2008 by Blackwell Publishing Ltd, UK

Edited by Trisha Dunning AM, RN, MEd, PhD, CDE, FRCNA and Glenn Ward MBBS, BSc, DPhil (Oxon), FRACP, FRCPath