A 45 year old right handed Marathi speaking Hindu male working as a cobbler educated upto std 2 residing at Lalbaug was brought to Nair Hospital in October 2000 with 4th episode of altered behaviour in the form of

• Suspiciousness
• Hearing voices
• Gesticulating and muttering of self
• Decreased sleep and appetite

Patients family, personal and medical history was not contributory and patient had cluster C pre morbid personality traits.
Mental state examination at that time showed that the patient had delusions of persecution, auditory hallucination, impaired judgement and absence of insight.

Patient was diagnosed as a case of paranoid schizophrenia. All routine investigations in the wards including blood sugars were within normal limits, his fasting plasma glucose being 92 mg/dl.

Patient was then started T. Risperidone 3 mg/day with which patient improved gradually. After discharge the patient followed regularly in the OPD. Patient was compliant on medication and was working well.

Three years later i.e. Jan. 2004, patient was taken up for a study of hyperglycaemic effects of antipsychotics. Patient was incidentally found to have impaired oral glucose tolerance test (75 gms glucose). The glucose levels were as follows. Fasting plasma glucose 98 mg/dl, at 30 minutes 135 mg/dl, at 90 minutes 148 mg/dl, while at the end of 2 hours it was 180 mg/dl. It is to be noted that values of more than 140 mg/dl at the end of 2 hours implies impaired glucose tolerance. (Patient had no personal, family history or risk factors associated with Diabetes Mellitus).

When the patient's sugars were repeated after 6 weeks, he had frank diabetes with fasting plasma glucose of 169 mg/dl and post prandial of 216 mg/dl. Patient was shifted to 10 mg of Trifluoperazine, 2 mg of trihexyphenidyl with 5 mg of diazepam. An endocrinology reference was made and a diabetic diet was advised. Blood sugars repeated later have shown a downward trend. Four weeks later his fasting plasma glucose was 124 mg/dl while the post prandial sugars were 192 mg/dl. After 6 weeks on continuous diet and exercise alone the plasma glucose reduced further to fasting 108 mg/dl and post prandial 148 mg/dl.

A total of 17 patients on risperidone from our department underwent oral Glucose tolerance test (75 gms). One case of frank diabetes (F > 140 mg/dl, 2 hrs PG > 200 mg/dl) and this case of impaired glucose tolerance was found with fasting hyperinsulinaemia of 67 mIU/ml who later became diabetic. Ten out of 17 patients had fasting hyperinsulinaemia inspite of normal fasting plasma glucose levels. (Normal range of insulin as established by TNMC Endocrine research lab is 7-25 mIU/ml on using the BARC (BRIT) - RIA-K-1 kit).

The available literature suggest that patients with schizophrenia are at a higher risk for diabetes. The prevalence of Diabetes Mellitus in patients with schizophrenia is 2-4 times greater than that in the general population. Taking antipsychotic medication further increases the chance of developing non insulin dependant hyperglycaemia.2

The mechanisms by which atypical antipsychotics cause hyperglycaemia are varied.

Case reports, chart reviews and results from clinical drug trials implicate a relationship between hyperglycaemia and clozapine or olanzapine.3

A study published by Frank and Amy et al studied the differential effects of Risperidone, Olanzapine and Clozapine on type 2 diabetes based on 12 months of exposure. The odds of type 2 diabetes for risperidone treated patients was not significantly different from that for untreated patients, whereas patients receiving other atypical antipsychotics had a significantly greater risk of diabetes than untreated patients.4

Although such studies and numerous other articles and references show nonsignificant association of risperidone with hyperglycaemia, we came across a few studies and case reports of hyperglycaemia associated with risperidone. We cite a few references.

A study done by Doraiswamy et al reviewed the FDA surveillance data from 1993 to 2002 of patients on risperidone. One hundred and thirty one cases of risperidone associated hyperglycaemia were noted. Findings suggested that

Another study done by Allison et al compared the hyperglycaemic effects of Risperidone with olanzapine. The findings showed that the increase in blood glucose levels of olanzapine treated patients was not significantly higher than those of Risperidone treated patients.6 (Note : Olanzapine is known for its hyperglycaemic effects).

An article published by Michael et al in the American Journal of Psychiatry 2002 studies the association of diabetes mellitus with use of antipsychotics in treatment of schizophrenia. The study showed that for patients less than 40 years age all of the atypical neuroleptics including Risperidone were associated with a significantly increased prevalence of diabetes. However Risperidone failed to show the same association in age groups above 40 years of age.7

Another study published in the journal of clinical endocrinology and Metabolism found that the magnitude of decrease in insulin sensitivity observed with both, olanzapine and Risperidone was identical (18%).8

In addition 3 cases of Risperidone associated diabetic ketoacidosis have been reported separately by Gordon et al, Coarkin et al and Newcomer et al.9

We conclude that atypical antipsychotics may unmask or precipitate hyperglycaemia. Although such cases attributed to Clozapine and Olanzapine are numerous than those associated with Risperidone, hyperglycaemia caused by Risperidone may be higher than that observed with conventional neuroleptics.

An important population to remember in the context of treatment emergent hyperglycaemia are the people with impaired glucose tolerance i.e. the patients whose fasting plasma glucose more than 140 mg/dl and less than 200 mg/dl. This group has a 5-10% annual risk of converting to diabetes.6

The relationship between fasting plasma glucose concentration and fasting insulin level is complex and resembles an inverted U or horse shoe in normal weight type 2 diabetes patients. Because this curve closely resembles starlings curve of the heart Defronzo and colleagues have referred to it as starling curve of the pancreas. As the fasting glucose level rises from 80 to 140 mg/dl, there is a progressive rise in fasting plasma insulin concentration which peaks at a value 2-2.5 times greater than that in normal age weight matched control subjects. This progressive rise in basal insulin secretion can be viewed as an adaptive response by the pancreas to offset the progressive deterioration in glucose homoeostasis, as demonstrated in this patient. However when the fasting glucose concentration exceeds 140 mg/dl the insulin secretion drops precipitously.10

Should we prefer high potency typical antipsychotics over the atypicals for treating patients with high risk for developing hyperglycaemia or those with pre-existing diabetes or should we routinely check baseline blood sugars before starting a patients on atypical antipsychotics, even for Risperidone?

John Buse in his article in the Journal of Clinical Psychiatry 2002 puts it this way.

“With the availability of several highly effective oral hypoglycaemic agents, the benefits of controlling psychotic disorders dramatically outweigh the potential risk associated with elevation of blood glucose”.

Or

Do you think otherwise?