VOL. 18 NO. 3 Summer 2000


Case Study: A 50-Year-Old Man With Schizoaffective Disorder and a History of Decompensated Type 2 Diabetes

Ginny Lewis, ARNP, FNP, CDE

J.G. is a 50-year-old African-American man with a chronic schizoaffective disorder. He was recently diagnosed with type 2 diabetes and has a history of hypertension and hypercholesterolemia. His psychiatric medications included gabapentin (Neurontin), valproic acid (Depakote), benztropine (Cogentin), and haloperidol (Haldol). His height is 68 inches and his weight is 250 lb. There is no evidence of retinopathy or microalbuminurea and his hypertension and hypercholesterolemia are well controlled with verapamil SR (Calan SR) and simvastatin (Zocor).

At the time of diagnosis of diabetes, he was in a psychiatric hospital and was successfully managed with metformin (Glucophage). After discharge from the psychiatric hospital, he stopped taking metformin and subsequently presented to his primary care provider with complaints of urinary frequency, urgency, and incontinence. Laboratory testing revealed a glucose of 1,572 mg/dl, sodium of 113 mEq/l, potassium of 4.8 mEq/l, blood urea nitrogen of 47 mg/dl, creatinine of 2.5 mg/dl, and HbA1c of >14%.

He was admitted to the hospital, treated for his hyperosmolar state, and stabilized on insulin before discharge to the psychiatric hospital. One month later, he was discharged to an independent living facility and had daily contact with an outpatient psychiatric clinic.

J.G. received all of his medications through the outpatient psychiatric clinic. The morning medications were given in the clinic, and the evening medications were provided in a unit dose package. He was also given his weekend medications in a unit dose package every Friday. He was responsible for taking the evening and weekend medications on his own, and thus there was no method to control the timing or whether he took the medications with food.

The staff at the facility was concerned with J.G.'s ability to manage his insulin appropriately. His thought patterns were very disorganized and tangential, making it difficult to obtain an accurate history regarding his daily activities. He was unable to measure his insulin dose correctly or to take appropriate problem-solving steps in response to his blood glucose values. He was able to learn blood glucose monitoring using a simple meter with date, time, and a memory function, which allowed for review of blood glucose values at clinic visits. It was not possible to determine whether these values were pre- or postprandial. However, the random values provided an opportunity to track blood glucose trends.

It was finally decided that the risks of outpatient insulin management were too great for this patient. The decision was made to discontinue insulin and seek alternative treatment options.


  1. What treatment options are available for patients with type 2 diabetes?
  2. What monitoring guidelines and treatment goals would be appropriate for this patient?

Three defects characterize type 2 diabetes: impaired insulin secretion, increased hepatic glucose output, and insulin resistance.1 These defects can affect individuals to varying degrees.1 The goal in diabetes management is to prevent acute and long-term complications by maintaining glucose levels as close to normal as possible. This is generally accomplished with meal planning, exercise, and medications, if necessary. For J.G., the goal became preventing acute complications, such as hyperosmolar coma and hypoglycemia, while minimizing risk factors for long-term complications. J.G. was at increased risk for macrovascular and microvascular complications because of his history of hypertension and hypercholesterolemia.

Maintenance of a consistent meal plan was difficult because of J.G.'s schizoaffective disorder. His case manager was able to control food choices by helping with grocery shopping, although the timing and quantity of meals could not be regulated. J.G. does not participate in a formal exercise program but walks as a part of his daily routine.

When medications become necessary in diabetes, the choice is usually based on several criteria including degree of hyperglycemia, weight, presence of concurrent diseases, and the patient's ability to manage self-care routines.2 In J.G.'s case, his chronic mental illness played a major role in the choice of the most appropriate medication. Insulin was not a treatment option given his severe thought disorder.

Oral antidiabetic medications have been developed to address each of the metabolic defects that define type 2 diabetes.3 Sulfonylureas stimulate insulin release from the pancreas, but differ in duration of action and side effect profile.3 Sulfonylureas have a significant risk of hypoglycemia because of the prolonged duration of action, which make them an unattractive option for J.G. because of his erratic meal intake. While glipizide has a slightly shorter duration of action, hypoglycemia continues to be a concern. In addition, first-generation sulfonylureas can interact with other highly protein-bound drugs, as well as drugs that inhibit their elimination, resulting in hypoglycemia. Although there appears to be less interaction with the second-generation sulfonylureas, the potential still exists. Two of J.G.'s medications, valproic acid and simvastatin, are also highly protein-bound and have the potential to interact with sulfonylureas.

Repaglinide (Prandin) is not a sulfonylurea but binds to the same ion channel as sulfonylureas and thereby stimulates insulin release.3 It has a rapid onset and short duration of action (<4 h), which reduces the risk for hypoglycemia compared to sulfonylureas. Repaglinide is most effective at decreasing postprandial blood glucose levels.

Metformin, a biguanide, does not cause hypoglycemia when used as monotherapy. It also has a positive effect on high-density lipoprotein cholesterol and triglyceride levels. A major concern with metformin is development of lactic acidosis in patients with renal or hepatic insufficiency or during periods of acute dehydration.

J.G. had initially been treated with metformin while in the psychiatric facility. J.G.'s serum creatinine had returned to normal following his hyperosmolar state. However, concern was raised regarding his ability to recognize the need for medical care should he become dehydrated.

Thiazolidinediones enhance insulin sensitivity but do not stimulate insulin secretion. Hypoglycemia is not a side effect of these medications when used as monotherapy. However, weight gain can be profound, and fluid retention occurs in some patients.4 Changes in liver function are a potential problem with this class of medications, and alanine aminotransferase (ALT) monitoring must be obtained every other month for the first year of therapy. There was concern that J.G. would not be able to follow through with the appropriate clinic visits and ALT monitoring.

Several of J.G.'s medications (valproic acid, haloperidol, and simvastatin) are metabolized via the cytochrome P450 pathways. Although there is no documentation of drug interactions between these medications and the thiazolidinediones, there is competition for enzyme binding sites as well as protein binding sites, which suggests there is potential for drug interaction.

Repaglinide was chosen as the best alternative because of the short duration of action and reduced risk for hypoglycemia with J.G.'s current lifestyle. In case of an overdose, repaglinide's elimination does not appear to be prolonged; therefore, the duration of hypoglycemia is not prolonged. J.G. was initially given repaglinide, 0.5 mg twice a day. His HbA1c has decreased from >14% at time of initial presentation to 8.1% currently. He has had no episodes of hypoglycemia and is testing his blood glucose several times a week, which has provided a satisfactory method of tracking glucose values.

Clinical Pearls

  1. Concurrent medications and preexisting medical conditions are important considerations when selecting medication options.
  2. With the multiple medication options available for diabetes, health care providers have the opportunity to choose an option that best fits their patients' medical treatment goals.
  3. Treatment and monitoring goals need to be adapted to fit patients' self-care abilities.


1Haas L: Pathophysiology of diabetes mellitus. Nurse Pract Forum 9:42-45, 1998.

2Lewis G: Drug therapy in diabetes management. Nurse Pract Forum 9:58-65, 1998.

3Feinglos MN, Bethel MA: Oral agent therapy in the treatment of type 2 diabetes. Diabetes Care 22 (Suppl 3):C61-64, 1999.

4Valve R, Sievenius K, Miettinen R, Pihlajamaki J, Rissanen A, Deeb SS, Auwerx J, Uusitupa M, Laakso M: Two polymorphisms in the peroxisome proliferator-activated receptor-g gene are associated with severe weight among obese women. J Clin Endocr Metab 84:3708-12, 1999.

Ginny Lewis, ARNP, FNP, CDE, is a nurse practitioner at the Diabetes Care Center at the University of Washington School of Medicine in Seattle.

Copyright 2000American Diabetes Association
Updated 7/00
For ADA Related Issues contact

For Technical Issues contact