This site is intended for health professionals only

At the heart of general practice since 1960

Read the latest issue online

A faulty production line

What I learned from my diabetes near miss

Polyuria and an elevated blood glucose and HbA1c in a 63-year-old might make you think of type 2 diabetes. In the second of our new series on cases that could have ended disastrously, Dr David Morris describes a situation that might make you think again

Polyuria and an elevated blood glucose and HbA1c in a 63-year-old might make you think of type 2 diabetes. In the second of our new series on cases that could have ended disastrously, Dr David Morris describes a situation that might make you think again

The case

The patient was a 63-year-old retired teacher who came to see me with a 4kg weight loss over four months, accompanied by polyuria, thirst and fatigue. But she looked well, and cardiovascular, respiratory and abdominal findings were normal. Her BMI was also in the normal range.

She had a past history of thyrotoxicosis that had been treated with radioactive iodine and now received a maintenance dose of levothyroxine. She was using salbutamol and tiotropium inhalers for mild-to-moderate COPD, acquired as a result of smoking since teenage years.

She had had two children without evidence of gestational diabetes; one of her daughters had type 1 diabetes mellitus.

The clinical diagnosis of diabetes mellitus was straightforward. Urine dipstick results were glucose +++ and ketones +++. A fingerprick BM reading of 27 was obtained. A fasting blood glucose came back later as 19.0 mmol/l with an HbA1C of 13.2%; other blood tests, including thyroid function testing, were satisfactory.

I provisionally diagnosed her as having type 2 diabetes and – because of the high glucose readings – started her immediately on gliclazide 40mg bd in an attempt to gain glycaemic control rapidly.

I thought about admitting her but in view of her good clinical condition, I opted for an early review in the practice.

At follow-up a few days later, the osmotic symptoms persisted and again a urine dipstick revealed heavy glycosuria and ketonuria, with no obvious response to gliclazide. By now I was seriously concerned that the patient was at risk of diabetic ketoacidosis and arranged for her to be admitted for urgent insulin treatment.

She was discharged after a blood test had shown her not to be acidotic, with advice to start insulin – which seemed to me to be a wasted opportunity to commence insulin within the safe confines of a hospital.

She was slotted in as an extra to the practice diabetes clinic the following day.

The outcome

The diagnosis then was consistent with late-onset type 1 diabetes, often called LADA (latent autoimmune diabetes in adults). Gliclazide was stopped and she started taking insulin Novomix 30 – eight units with breakfast and six with evening meal.

The patient was taught to monitor her blood glucose by our diabetic nurse and her daughter was happy to supervise injections until she was seen by the community diabetic nurse specialist.

Within one week of starting insulin, her osmotic symptoms had cleared, BM readings had fallen and ketonuria had resolved. The Novomix doses were subsequently titrated up to achieve single figure BM readings and her weight loss reversed. Islet cell and GAD (glutamic acid decarboxylase) antibodies were positive, supporting the diagnosis of type1 diabetes. An ultrasound of the abdomen was unremarkable, with no suggestion of pancreatic pathology.

Why it nearly went wrong

The problem here was that the patient was thought to have type 2 diabetes because of her age. In fact she had LADA, which is best regarded as late-onset type 1 diabetes in terms of management. These patients are not insulin-resistant but rather develop an absolute need for insulin as progressive pancreatic ß-cell destruction occurs.

Response to oral hypoglycaemics such as the sulphonylureas will be poor and there is a real risk of diabetic ketoacidosis unless insulin is commenced. In addition, such patients face the same long-term complications as ‘standard' type 1 diabetes.

Pointers towards LADA are marked osmotic symptoms and relatively rapid weight loss in non-obese patients.

Ketonuria is a crucial finding, indicative of insulin deficiency (other causes being starvation and alcohol excess) and must not be ignored. Measurement of blood ketones is a more accurate gauge of insulin deficiency and the risk of diabetic ketoacidosis, now easily measured using fingerprick blood with standard meters.

A personal or family history of autoimmune disease strengthens the possibility of LADA and this patient had both – thyrotoxicosis herself and a daughter with type 1 diabetes.

The findings of islet cell and GAD antibodies (or of low C-peptide levels) support an insulin-deficient diabetic state.

The moral of the story

Although the immediate reaction on diagnosing diabetes in an elderly person may be to assume it is type 2, it's crucial to keep an open mind to the possibility of late-onset type 1. Ketosis is a key finding and can indicate an absolute need for insulin treatment. Staying alert to the possibility of late onset type 1 diabetes means the risk of diabetic ketoacidosis may be anticipated and warning signs dealt with early.

The lessons learned will stay with me and hopefully prevent any catastrophes.

Dr David Morris is a GP in Shrewsbury

Was it really type 2 diabetes? Was it really type 2 diabetes? Near misses Near misses

Have you had a near miss you think other GPs could learn from? Send a brief description to clinical editor Adam Legge at

Rate this article 

Click to rate

  • 1 star out of 5
  • 2 stars out of 5
  • 3 stars out of 5
  • 4 stars out of 5
  • 5 stars out of 5

0 out of 5 stars

Have your say