A Weighty Problem—Diagnosis and Treatment of Type 2 Diabetes in Adolescents

Orit Pinhas-Hamiel, MD, and Philip Zeitler, MD, PhD

The prevalence of overweight in the United States has increased dramatically during the past decade¹ and affects not only adults, but also adolescents and children. One-fifth of adolescents aged 12–19 years are overweight,² while the prevalence of obesity among younger children is estimated to be between 5 and 25%.^3,4

It is well known that obesity leads to increased morbidity and mortality in adults.^5,6 However, the association between adolescent obesity and morbidity has been less well studied. In the following case, we will illustrate and discuss one potential morbidity associated with adolescent obesity.

CASE STUDY

The Obese Adolescent Girl With Ketoacidosis: Things Are Not Always What They Seem

A 12-year-old African-American girl presented with polyuria, polydipsia, and a 12-kg weight loss over approximately 3 weeks. Symptoms were prominent for the 3–4 days immediately preceding the admission.

She had no significant past medical or surgical history. Her family history was significant for type 2 diabetes in her father, with onset in his 30s, and in both her maternal grandmother and grandfather, with onset at 60 years and 70 years, respectively.

Initial assessment revealed an alert, obese girl in minimal distress. Blood pressure was 122/60 mmHg, pulse 95 beats per minute, respiratory rate 24 breaths per minute, temperature 97.6 F. The patient’s weight was 74 kg, and her height was 151 cm, with a body mass index (BMI) of 32 kg/m2. Her weight before this illness was 86 kg (BMI = 36 kg/m2).

The patient had prominent acanthosis nigricans (AN) around her neck. She also had subtle AN over the flexural creases of her antecubital fossae. Her abdomen was obese and soft with mild epigastric tenderness. She had Tanner III breast development and pubic hair. The rest of her examination was normal.

Initial laboratory evaluation revealed a serum glucose of 52 mmol/L (950 mg/dl), pH 7.23, and CO2 10 mmol/L. The rest of her electrolytes were normal. Urinalysis showed glucose >55 mmol/L (>1,000 mg/dl) and ketones >80 mg/dl.

The patient was diagnosed with diabetic ketoacidosis (DKA) and dehydration. She was rehydrated with IV fluids, and her hyperglycemia and acidosis were treated with a continuous insulin infusion. The acidosis gradually resolved. However, during the remainder of her hospitalization, good glycemic control was extremely difficult to obtain. Her eventual insulin requirement was ~2 U/kg/day.

Because of her obesity, acanthosis nigricans, family history of type 2 diabetes, and apparent insulin resistance during therapy, a fasting glucose and 1-hour and 2-hour postprandial C-peptide readings were determined 2 weeks after resolution of the DKA. The results revealed a fasting C-peptide of 390 pmol/L (1.2 ng/ml) and fasting glucose 5.3 mmol/l (96 mg/dl). There was an adequate postprandial C-peptide increase. The 1-hour C-peptide was 630 pmol/L (1.9 ng/ml) and glucose was 19.1 mmol/l (343 mg/dl). The 2-hour C-peptide was 830 pmol/L (2.5 ng/ml), and glucose was 21.3 mmol/l (383 mg/dl). Islet cell antibodies (anti-IA-2, anti-GAD, and insulin autoantibodies) obtained at presentation were negative.

The patient was given the presumptive diagnosis of type 2 diabetes. Although the health-care team remained alert to the possibility that she might have type 1 diabetes with atypical presentation, her chronic course confirmed the initial diagnosis.

Initially, the patient was followed bimonthly, and, although she gained back all of her weight, her glycemic control was fair. HbA1c levels ranged between 9–11% (normal range 4–8%). Her insulin doses were gradually decreased, and she was started on an oral hypoglycemic agent.

Subsequently, the patient discontinued monitoring blood glucose and missed several clinic visits. Eighteen months after her initial diagnosis, she returned to clinic. She reported that she had discontinued all of her medications, did not follow a specific diet, and had nocturia. She continued to gain weight. Her HbA_1c was 11%, and her urine ketones were negative. Otherwise, she was doing well.

Four years later, the patient was still coming to the clinic sporadically. Insulin was restarted for a short period, but HbA_1c levels did not improve.

DISCUSSION

This case points out a number of features characteristic of the adolescent type 2 diabetic population. The patient is a mid-pubertal, African-American girl, who is markedly obese, sedentary, and has signs of insulin resistance (acanthosis nigricans). She has a family history of type 2 diabetes. Although she had DKA at presentation and required high doses of insulin to attain glycemic control, she was subsequently able to discontinue insulin for a long period without becoming ketotic. She had negative islet cell antibodies at the time of presentation, and she also had intact C-peptide secretion.

These clinical features are classic for type 2 diabetes. Unlike patients with type 1 diabetes, who suffer rapid deterioration when not taking insulin, patients with type 2 diabetes can forget about their disease, the need for diet control, and self-monitoring of blood glucose for long periods without becoming critically ill.

Thus, ongoing education about the importance of good metabolic control is essential, as are efforts to bolster adolescents’ motivation to maintain difficult and unpleasant lifestyle changes. Unfortunately, adolescents typically have difficulty planning for the future and making major lifestyle changes.

Epidemiology of Type 2 Diabetes in Adolescents
Type 2 diabetes has been considered rare in the pediatric population, with estimates of ~2% of newly diagnosed cases among the pediatric population.⁷ However, this situation is rapidly changing. We reported a 10-fold increase in the incidence of type 2 diabetes among adolescents in Cincinnati over the past decade (Figure 1).⁸ Type 2 diabetes accounted for nearly 40% of new-onset adolescent diabetes cases. Similar observations have been made in other parts of the United States^9,10 and throughout the world.¹¹

This finding implies that a significant number of new adolescent patients with diabetes will have type 2 diabetes, a situation very different from the past. Most concerning is the fact that in the adult population, undiagnosed type 2 diabetes represents half of all cases, irrespective of gender, age, and race.¹² If the same is true for adolescents, type 2 diabetes is even more common in this age-group than indicated by current data.

About one-third of the adolescents with type 2 diabetes in our study group8 and in the group from San-Diego10 were non-Hispanic white, whereas the rest (68%) were either African-American or Hispanic. The overall female-to-male ratio was 1.7:1, with a predominance of females irrespective of race.

The common dominator in our patients, as well as in those reported by others, was their morbid obesity. Known environmental risk factors for type 2 diabetes include overweight, high caloric intake, and low physical activity—the hallmarks of a modern Western lifestyle.^13,14 Indeed, in a subgroup of our patients subsequently evaluated for their habits (observations not yet published), we found that, despite education sessions and our recommendations, the vast majority (85%) did not practice any physical activity. Moreover, all patients were found to consume a high-fat, low-fiber diet. Therefore, the increasing incidence of type 2 diabetes among adolescents appears to be a consequence of the well-documented increase in adolescent obesity combined with a marked decrease in physical activity leading to earlier presentation of type 2 diabetes in genetically susceptible patients.

Characterization of Patients
Patients with type 2 diabetes presenting in adolescence generally are obese or extremely obese. The mean BMI for all patients seen in Cincinnati was 37.7 ± 9.6 kg/m2 and the mean age was 13.8 ± 1.9 year. All patients were in puberty (Tanner stage III or greater).

Interestingly, females were diagnosed on average 1 year earlier than males. Because puberty starts on average 1 year earlier in females, the earlier presentation of type 2 diabetes in the females may be related to timing of sexual development. The temporal relationship between onset of type 2 diabetes and pubertal development suggests that the relative insulin resistance characteristic of puberty¹⁵ may be an etiological factor or contributing factor in the appearance of overt type 2 diabetes in these adolescents. Furthermore, since the age of onset of puberty among obese children is earlier than that of their peers,¹⁶ the relationship between body mass and puberty may further contribute to the early presentation of type 2 diabetes in these patients.

A strong family history of type 2 diabetes is characteristic of this group of patients. However, there is generally no autosomal dominant inheritance pattern suggesting the diagnosis of mature-onset diabetes of the young (MODY).

Presentation of Disease
In the present case, the patient’s main complaints were polyuria, polydipsia, and weight loss. However, one-third of our patients were diagnosed by urinalysis during a routine physical examination. This observation points out that primary health care providers must be alert to young patients with extreme obesity, acanthosis nigricans (present in 60% of patients), hypertension, other signs of insulin resistance, and/or a family history of type 2 diabetes. Also, vaginal monilial infection was present in 25% of females.

In many instances, patients are aware of their increased polyuria and polyphagia but enjoy the accompanying weight loss and do not seek medical help. The importance of early recognition of type 2 diabetes in adolescents cannot be over-emphasized, as micro- and macrovascular complications have been diagnosed in adults with unrecognized diabetes.

Adolescents diagnosed with type 2 diabetes may present in DKA,¹⁷ as in the present case. The presentation of adolescents with DKA or some degree of ketosis may confuse the initial diagnosis. In the DCCT, a basal C-peptide level <200 pmol/L was consistent with the diagnosis of type 1 diabetes,¹⁸ and the VA Cooperative Study of type 2 diabetes used an exclusion criterion of basal C-peptide <210 pmol/L.¹⁹ A concordance rate of ~90% was found between C-peptide levels and clinical diagnosis of type 2 diabetes.²⁰ In our patient, the basal C-peptide level of 390 pmol/L is nearly double the above criteria and indicative of type 2 diabetes despite ketoacidosis at presentation. The observation that ketosis occurs among adolescent patients with type 2 diabetes agrees with recent reports of DKA in obese African-American adult patients subsequently diagnosed with type 2 diabetes.²¹ Thus, in young, obese adolescents who develop DKA, the presence of insulin-dependent diabetes cannot be assumed and the diagnosis of type 2 diabetes should be considered.

Treatment Strategies
Early-onset type 2 diabetes among adolescents may have particularly devastating effects on long-term health because these patients will have many years with the disease. Therefore, it is even more critical than in adult diabetes that efforts be focused on treating the underlying causes of the disorder, namely obesity, poor diet, and sedentary lifestyle.

Poor dietary adherence, emotional lability, and rebellion are characteristic of adolescents in general. Thus, adolescents with diabetes present a number of difficulties for the design of treatment strategies. In addition, adolescent type 2 diabetes patients are overwhelmingly sedentary and are unrestrained eaters. These facts, as well as the family constellation, must be taken into account when planning treatment strategies.

Education. Continuous education, involving the physician, diabetes nurse educator, and registered dietitian, has a critical role in the management of adolescents with diabetes mellitus. This is particularly true for patients with type 2 diabetes. In our experience, initial education sessions patterned after the education sessions given to type 1 diabetes patients are not sufficient for type 2 diabetes patients. Unfortunately, these patients quickly resume their old eating habits with foods rich in fat. Education must therefore be specific and repeated. Since dietary control is of fundamental importance in the treatment of type 2 diabetes, the development of a strong relationship between the patient and a dietitian with a special interest in this area should be a focus of the initial education experience.

Education should include all family members. This is particularly important in this population, because other family members are often overweight and many have diagnosed or undiagnosed diabetes or are at high risk for developing type 2 diabetes. Further-more, maintenance of a high-fat, high-calorie, low-fiber diet by other family members makes it almost impossible for adolescents with type 2 diabetes to comply with dietary recommendations.

These patients are often initially diagnosed as having type 1 diabetes. It is important that they be assigned and understand their correct diagnosis as soon as possible. Furthermore, it is wise to be judicious in assigning an initial diagnosis in patients who fit the characteristics of this population, in order to avoid confusion and inappropriate initial education.

Once the diagnosis is made, the patients themselves, as well as family members, must be educated regarding the difference in pathogenesis and treatment between type 1 diabetes and type 2 diabetes. Education should focus on the difference in natural history of the two diseases, the treatment options, the distinction between insulin use and insulin dependence, the possible complications, and approaches to avoidance of complications. Often, patients assume that their disease is of a less serious nature if they do not require insulin. Thus, an emphasis on good treatment adherence to avoid early complications is important for all family members.

Patients must be followed closely after the appropriate initial education to monitor their compliance, provide support for their successes, and work with their families to change multiple aspects of behavior.

Nutrition. The aim of nutrition management should be gradual, sustained weight loss. Realistic goals should be set, e.g., moderate weight reduction and not normalization of body weight, as the latter will quickly discourage patients. Even modest weight loss can result in marked improvement in glycemic control and plasma lipoprotein patterns in adult patients with type 2 diabetes.²²

Moreover, reduction in total body adiposity and improvement in exercise tolerance are the most effective means to improve insulin resistance in patients with type 2 diabetes.²³ Reduction in adiposity also lowers blood pressure and improves hypertriglyceridemia.²⁴

Besides weight loss, the nutrition goals, as in adult type 2 diabetes, are aimed at achieving optimum metabolic control of blood glucose levels and decreasing lipids and lipoproteins.

According to recent recommendations of the American Diabetes Asso-ciation,²⁵ patients should have an individualized approach to medical nutrition therapy based on their lifestyle and individual diabetes management goals. As mentioned above, this is specifically important in this population, as patients often are initially diagnosed as having type 1 diabetes and given a meal plan that includes mandatory snacks between meals. Attention also must be paid to the development or presence of binge-eating disorders in this high-risk population.

Daily exercise. Exercise is of primary importance in the prevention and management of type 2 diabetes. Yet these adolescent type 2 diabetes patients are very obese with extremely low exercise performance. Therefore, compliance with an exercise regimen may be very difficult physically, as well as emotionally.

Another problem that may arise is the expense and logistics associated with arranging for routine organized exercise for these children. They may be alone at home after school without supervision and may live in neighborhoods where outdoor activity is not attractive. Access to institutional exercise facilities may be beyond the financial means of the family. Therefore, exercise programs need to be tailored to patients’ physical abilities, as well as to their family circumstances. Structuring an appropriate regimen will require some creativity on the part of the diabetes management team.

Above all, it should be emphasized that even moderate exercise improves control.²⁶ Patients should be encouraged to choose physical activities they can perform easily and frequently. For example, daily walking is recommended in obese, adult type 2 diabetes patients not only for weight reduction but also for improvement of insulin sensitivity.²⁷ An exercise plan that includes other family members may be helpful to promote family lifestyle changes.

Emotional support. As in any chronic disease in this age-group that influences daily activity and has implications for long-term morbidity, emotional support must play a significant role in therapy. In our experience, many patients have been treated for depression or eating disorders before presentation, while additional patients have signs or symptoms of emotional disorders but have not sought evaluation. Furthermore, the development of depression and adolescent adjustment disorder can significantly complicate the ability of the patient to comply with the lifestyle changes required. For these reasons, the team approach to treatment of adolescents with type 2 diabetes ideally includes a mental health professional with expertise in treating adolescents.

Integral involvement with the team would permit initial evaluation of patients’ emotional needs and ongoing re-evaluation as adolescents tackle the demands of coping with type 2 diabetes. In the absence of routine mental health participation on the diabetes team, clinicians should remain sensitive to the emotional status of these patients and maintain a low threshold for referral to mental health resources in the community.

Medication. An extensive body of literature exists regarding the effect of oral hypoglycemic agents, alone and in combination, among adults. However, their effects on adolescents have not been studied. We have generally started treatment with 5 mg of glyburide and increased the dose up to 20 mg/day. However, this regimen has not proven particularly successful.

Metformin, which is not a sulfonylurea, decreases glucose production in the liver, increases skeletal muscle glucose uptake, and does not cause clinical hypoglycemia.²⁸ It has no effect on pancreatic insulin secretion and requires the presence of insulin to be effective. When endogenous secretion of insulin is adequate, it can be used alone to overcome insulin resistance. However, its use in adolescents has not been studied and therefore must be undertaken with caution in these patients.

Another potential drug for use in this population is troglitazone, which has also recently received approval for use in adults with type 2 diabetes in the United States. Its administration in adult type 2 diabetes patients improves hyperglycemia and hyperinsulinemia.²⁹ However, again, there is no reported experience with this drug in adolescent patients with type 2 diabetes. Therefore, its use must be based on the adult literature at present.

Weight loss medications. Besides treating hyperglycemia and improving insulin resistance, treatment should be aimed toward solving the underlying problem of obesity. However, the pharmaceutical options currently available to assist in this endeavor are limited. Dexfenfluramine and fenfluramine, two widely prescribed agents thought to have a role in treatment of morbidly obese patients,^30,31 were removed from the North American market in September 1997 secondary to safety concerns regarding primary pulmonary hypertension and development of valvular insufficiency. Whether these or related compounds will reappear in the market is unclear.

Some studies have addressed the beneficial effects of fluoxetine in adults with type 2 diabetes.³² Patients treated with fluoxetine had a greater weight loss, a significantly lower HbA_1c, and decreased total daily insulin dose compared with those taking placebo. Since we found a high prevalence of depression among obese adolescents with type 2 diabetes in our population (unpublished observations), adolescents with type 2 diabetes may benefit from this or similar treatment. However, this has not been systematically studied.

Finally, there are a number of stimulant medications, including phentermine, which have been approved by the Food and Drug Administration as short-term adjuncts in weight loss. However, none of these agents has proven to be particularly effective, and, most importantly, there is currently no experience with these drugs in adolescents. Therefore, the use of pharmaceutical agents to assist with weight loss in this population is not recommended and should be undertaken only in the most severely obese patients and with the utmost caution.

Insulin. Unlike the adult population, a substantial number of these adolescent patients are started on insulin treatment because the initial diagnosis is thought to be type 1 diabetes. In most cases, patients can be weaned off of insulin, but this may be a prolonged process secondary to physician anxiety regarding the diagnosis and the appropriate treatment for adolescent type 2 diabetes.

We believe that all efforts consistent with adequate glycemic control should be taken to prevent chronic insulin therapy, since it heightens the risk of hypoglycemia, entails frequent injections and blood measurements, and most importantly, is associated with hyperinsulinemia, which by itself may promote hypertension and hyperlipidemia.³³ However, in many patients, insulin will eventually be required.

Numerous studies in the adult population have examined combined insulin and oral hypoglycemic treatment.³⁴ While experience with adolescent type 2 diabetes is still limited, the approach in patients who fail oral therapy should consider that adult patients do not benefit from multiple daily insulin injections. Currently, nocturnal NPH insulin administration with oral hypoglycemic agents is the preferred initial approach³⁵ to insulin therapy.

Summary

Type 2 diabetes is an increasingly important adolescent health problem with potentially significant long-term consequences. An increasing percentage of adolescents presenting with new-onset diabetes, with or without ketosis, will have type 2 diabetes. Because the distinction between type 2 diabetes and type 1 diabetes has implications for optimal long-term therapy, it is important for clinicians to take into account multiple features of the presentation in making a provisional diagnosis of either type 2 diabetes or type 1 diabetes in these patients. Furthermore, the initial diagnosis should be re-evaluated based on the subsequent course.

The primary pathology in these patients is severe obesity, and the primary goal of treatment should be weight control. In addition, emotional support and exercise programs are mandatory.

Insulin therapy may lead to increased weight gain, increased burdens on the family, and an increase in the incidence of diabetic complications. Therefore, the possibility of controlling diabetes without insulin therapy in these obese patients must be considered.

Above all, it must be kept in mind that obesity and type 2 diabetes are chronic conditions requiring continuous support and follow-up.

1Kuczmarksi RJ, Flegal KM, Campbell SM, Johnson CL: Increasing prevalence of overweight among U.S. adults. JAMA 272:238-39, 1994.

2Centers for Disease Control and Prevention: Prevalence of overweight among adolescents — United States. 1988-91. Morbid Mortal Weekly Report 43:818-21, 1994.

3Dietz WH: Childhood obesity: susceptibility, cause, and management. J Pediatr 103:676-86, 1983.

4Javier NF, Szklo M, Comstock GW: Childhood weight and growth rate as predictors of adult mortality. Am J Epidemiol 136:201-13, 1992.

5Van Itallie TB: Health implications of overweight and obesity in the united states. Ann Intern Med 103:983-88, 1985.

6Bray GA: Complications of obesity. Ann Intern Med 103:1052-61, 1985.

7Drash AL: The classification of diabetes mellitus in children and adolescents. Acta Paediatr Jpn 29:325-34, 1987.

8Pinhas-Hamiel O, Dolan LM, Daniels SR, Standiford D, Khoury PR, Zeitler P: Increased incidence of non-insulin dependent diabetes mellitus among adolescents. J Pediatr 128:608-15, 1996.

9Umpierrez GE, Woo W, Stewart P, Hagopiam W, Kitabchi AE: Insulin secretion and antibodies to glutamic decarboxylase in obese African-Americans with diabetic ketoacidosis. 10th International Congress of Endocrinology Abstract Book, 1996, p. 90.

10Glaser N, Jones KL: Non insulin dependent diabetes mellitus in childhood. Pediatr Res 37:89A, 1995.

11Owada M, Hanaoka Y, Tanimoto Y, Kitagawa T: Descriptive epidemiology of non-insulin dependent diabetes mellitus detected by urine glucose screening in school children in Japan. Acta Pediatr Jpn 32:716-24, 1990.

12Harris M: Undiagnosed type 2 diabetes: clinical and public health issues. Diabetes Care 16:643-52, 1993.

13Bray GA: Obesity increases risk for diabetes. Int J Obes 16:13-17, 1992.

14Dietz WH, Gortmaker SL: Do we fatten our children at the television set? Obesity and television viewing in children and adolescents. Pediatrics 75:807-12, 1985.

15Savage MO, Smith CP, Dunger DB, Gale EA, Holly JM, Preece MA: Insulin and growth factors adaptation to normal puberty. Horm Res 37:70-73, 1992.

16Morrison JA, Barton B, Biro FM, Sprecher DL, Falkner F, Obarzanek E: Sexual maturation and obesity in 9- and 10-year-old black and white girls: the National Heart, Lung, and Blood Institute Growth and Health Study. J Pediatr 124:889-95, 1994.

17Pinhas-Hamiel O, Dolan LM, Zeitler P: Diabetic ketoacidosis among obese African-American adolescents with NIDDM. Diabetes Care 20:484-86, 1997.

18The DCCT Research Group: The Diabetes Control and Complications Trial (DCCT) design and methodological considerations for the feasibility phase. Diabetes 35:530-45, 1986.

19Arbraira C, Emanuele N, Colwell J, Henderson W, Comstock J, Levis S, Nuttall F, Sawin C: VA Cooperative Study Group: glycemic control and complications in type II diabetes: design of a feasibility trial. Diabetes Care 15:1560-71, 1992.

20Service EJ, Rizza RA, Zimmmerman BR, Dyck PJ, O’Brien PC, Melton LJ: The classification of diabetes by clinical and C-peptide criteria. Diabetes Care 20:198-201, 1997.

21Umpierrez GE, Casals MMC, Gebhart SSP, Mixon PS, Clark WS, Phillips LS: Diabetic ketoacidosis in obese African-Americans. Diabetes 44:790-95, 1995.

22Colditz GA, Willett WC, Rotnitzkii A, Manson JE: Weight gain as a risk factor for clinical diabetes mellitus in women. Ann Intern Med 122:481-86, 1995.

23Olefsky J, Reaven GM, Farquhar JW: Effects of weight reduction on obesity studies of lipid and carbohydrate metabolism in normal and hyperlipoproteinemic subjects. J Clin Invest 53:64-76, 1974.

24Reisen E, Abel R, Modan M, Silverberg DS, Eliahou HE, Modan B: Effect of weight loss without salt restriction on the reduction of blood pressure in overweight hypertensive patients. N Engl J Med 298:1-6, 1978.

25Bantle JP: Current recommendations regarding the dietary treatment of diabetes mellitus. The Endocrinologist 4:189-96, 1994.

26Ras I, Hauser E, Bursztyn M: Moderate exercise improves glucose metabolism in uncontrolled elderly patients with non-insulin-dependent diabetes mellitus. Isr J Med Sci 30:766-70, 1994.

27Yamanouchi K, Shinozaki T, Chikada K, Nishikawa T, Ito K, Shimizu S, Ozawa N, Suzuki Y, Maeno H, Kato K: Daily walking combined with diet is a useful means for obese type 2 diabetes patients not only to reduce body weight but also to improve insulin sensitivity. Diabetes Care 18:775-78, 1995.

28DeFronzo R, Goodman A: Efficacy of metformin in patients with noninsulin dependent diabetes mellitus: The Multicenter Metformin Study Group. N Engl J Med 333:541-49, 1995.

29Nolan JJ, Ludvik B, Beerdsen P, Joyce M, Olefsky J: Improvement in glucose tolerance and insulin resistance in obese subjects treated with troglitazone. N Engl J Med 331:1183-93, 1994.

30Abenhaim L, Moride Y, Brenot F, Rich S, Benichou J, Kurz X, Higenbottam T, Oakley C, Wouters E, Aubier M, Simonneau G, Begaud B: Appetite suppressant drugs and the risk of primary pulmonary hypertension. N Engl J Med 335:609-16, 1996.

31Manson JE, Faich GA: Pharmacotherapy for obesity: do the benefits outweigh the risks? N Engl J Med 335:659-60, 1996.

32O’Kane M, Wiles PG, Wales JK: Fluoxetine in the treatment of obese type 2 diabetic patients. Diabetic Med 11:105-10, 1994.

33Despres JP, Lamarche B, Mauriege P, Cantin B, Dagenais GR, Moorjani S, Lupien PJ: Hyperinsulinemia as an independent risk factor for ischemic heart disease. New Engl J Med 334:952-57, 1996.

34Raskin P: Combination therapy in NIDDM. N Engl J Med 327:1453-54, 1992.

35Landsted-Hallin L, Adamson U, Arner P, Bolinder J, Line PE: Comparison of bedtime NPH or preprandial regular insulin combined with glibenclamide in secondary sulfonylurea failure. Diabetes Care 18:1183-86, 1995.