VOL. 17 NO. 4 1999
Case Study: A 55-Year-Old Man With Obesity, Hypertriglyceridemia, and Newly Diagnosed Type 2 Diabetes Who Collapsed and Died
Deborah Thomas-Dobersen, RD, MS, CDE, and Michael J. Dobersen, MD, PhD
A 55-year-old Caucasian man presented with polyuria, polydipsia, and "feeling dry" during the past 2 months. His medical history was remarkable for a 3-year history of poorly controlled hypertriglyceridemia. His initial fasting serum cholesterol was 299 mg/dl, triglycerides were 928 mg/dl, and high-density lipoprotein (HDL) cholesterol was 30 mg/dl before treatment.
He was treated with gemfibrozil (Lopid) 600 mg twice daily and told to watch his diet and exercise. No referral was made to a registered dietitian.
Two years later, a fasting triglyceride level of 570 mg/dl prompted further increase of gemfibrozil to 600 mg three times daily (this exceeds usual recommended dosing). The patient took no other medications and denied drinking alcohol and smoking.
Physical examination revealed a height of 5'11", weight of 240 lb (body mass index [BMI] of 34.4 kg/m2), blood pressure of 150/88 mm Hg, and pulse of 80/min. There was no abdominal tenderness or organomegally. Laboratory evaluation showed a serum glucose of 397 mg/dl. Urinalysis revealed 3+ glucose and negative ketones.
The patient was started on 5 mg glyburide [Micronase] daily. He was also given a referral to a dietitian. That evening, the patient complained of abdominal pain, nausea, vomiting and flu-like symptoms. He collapsed at home and died a short time later. At autopsy, it was found that he died of acute hemorrhagic pancreatitis. The patient was also found to have severe arteriosclerotic cardiovascular disease with severe two-vessel coronary artery atherosclerosis.
1. What is a normal level of serum triglyerides?
2. What is the medical nutrition therapy for hypertriglyeridemia?
3. What are current recommendations for screening for diabetes?
4. What effect did the onset of diabetes have on this patient's hypertriglyc- eridemia?
The National Cholesterol Education Program (NCEP) Adult Treatment Panel II1 gives the following classification for triglyerides:
normal 200 mg/dl
borderline high 200400 mg/dl
high 4001,000 mg/dl
very high >1,000 mg/dl
Hypertriglyceridemia can be primary (associated with familial hypertriglyceridemia) or secondary (due to diabetes mellitus, hypothyroidism, kidney disease, or medication). Important exacerbating factors are obesity and excess alcohol intake.
Hypertriglyceridemia and low HDL cholesterol (<35 mg/dl) are commonly seen in the insulin resistance syndrome, or Syndrome X. In fact, an increase in plasma triglyceride is the most common metabolic characteristic of Syndrome X. Although all insulin-resistant patients do not develop type 2 diabetes, many do. Insulin resistance is involved in the pathogenesis and clinical course of type 2 diabetes as well as hypertension and coronary heart disease.2
In type 2 diabetes, a common abnormal lipid pattern is an elevation of very-low-density lipoprotein (VLDL) cholesterol, a reduction in HDL, and a low-density lipoprotein (LDL) cholesterol that contains a greater proportion of small, dense atherogenic LDL particles.3 Diabetes, as a possible cause of the hypertriglyceridemia, should be evaluated and treated if found, as several studies have shown that this pattern of dyslipidemia precedes the onset of type 2 diabetes mellitus.4
The goals for medical nutrition therapy, the first line of treatment1 for borderline to high triglyceride values, are:
1. weight loss if indicated
2. restriction of alcohol intake
3. increased physical activity
4. the American Heart Association (AHA) Step 1 progressing to Step 2 diets, individualized for the patient
A weight loss of only 5% of total body weight effectively lowers triglycerides. Exercise can lower triglycerides by approximately 10%. If triglycerides increase on the AHA Step 2 meal plan, the amount of carbohydrate should be decreased and the amount of mono-unsaturated fats increased. At present, it is still controversial whether this has a long-term or a short-term benefit.
The NCEP Adult Treatment Panel noted that the expertise of a registered dietitian is very helpful in achieving adherence with these protocols.1 When referring a patient to a dietitian, include laboratory data on hyperlipidemia.
People with triglycerides >500 mg/dl are at risk of pancreatitis. This risk increases as triglycerides increase, becoming very high when serum triglycerides approach 2,000 mg/dl.5 Special immediate attention to lower triglycerides to <400 mg/dl is recommended. Severe dietary fat restriction (<10% of calories) in addition to pharmacological therapy is needed to reduce the risk of pancreatitis,3 as gemfibrozil will not be able to decrease serum triglycerides when they are extremely high (>1,500 mg/dl). This severe diet can decrease serum triglycerides by 2025%. Further reduction to Adult Treatment Panel II goals of <200 mg/dl may be beneficial.
This patient's triglyceride level was inadequately treated. The patient did not make some follow-up appointments, and the dyslipidemia may have been refractory to treatment.
The onset of type 2 diabetes in this patient may have deleteriously raised the serum triglyceride levels in two ways: by directly increasing VLDL production and by decreasing catabolism due to decreased lipoprotein lipase activity. Several assumptions must be made at this point to understand the onset of acute hemorrhagic pancreatitis. It is reasonable to assume that the triglyceride level soared to >1,000 mg/dl sometime during the onset of the type 2 diabetes. It is also reasonable to assume that the actual onset of diabetes predated the onset of symptoms by several months, possibly a year. If diabetes had been detected, improved glycemic control would have been very effective in reducing serum triglycerides.
The exact mechanism whereby hypertriglyceridemia causes pancreatitis is unknown. Presumably, the extremely high level of serum triglycerides causes "sludging" in the pancreatic vasculature, resulting in ischemia and necrosis. Alternatively, pancreatic lipase breaks down triglycerides to free fatty acids, which at high levels can also cause pancreatitis by direct toxicity (personal communication, Robert H. Eckel, MD, Lipid Research Clinic, University of Colorado Health Sciences Center, Denver, Colo.).
Diabetes occurs more frequently in individuals with hypertriglyceridemia (>250 mg/dl) and/or an HDL level of <35 mg/dl, in patients who are obese (BMI >27 kg/m), and in people over the age of 45.6 The American Diabetes Association has recommended that screening for diabetes may be appropriate for individuals with one or more risk factors.6 This patient had three risk factors. The recommended screening interval is 3 years.
More aggressive treatment of hypertriglyceridemia and earlier detection of diabetes may have lessened the impact of the onset of diabetes on triglyceridemia, thereby preventing the premature death of this patient. A referral to a lipid specialist may have helped in treating this severe, complex, or refractory disorder. An inadequate understanding of the importance of weight loss, exercise, and diet changes may have prevented a satisfactory lowering of serum triglycerides.
1. Hypertriglyceridemia should be carefully monitored and aggressively treated by weight loss, diet, exercise, alcohol restriction, and pharmacological means to keep serum levels <400 mg/dl to prevent possible pancreatitis and <200 mg/dl to prevent coronary heart disease.
2. Patients presenting with lipid patterns similar to those found in type 2 diabetes (high triglycerides and low HDL) should be screened for diabetes.
3. Medical nutrition therapy is the cornerstone of treatment for hyperlipidemia. Referral to a registered dietitian and a lipid specialist is recommended to help with such severe dyslipidemia.
1National Cholesterol Education Program (NCEP) Expert Panel: Summary of the second report of the NCEP expert panel on detection, evaluation, and treatment of high blood cholesterol (Adult Treatment Panel II). JAMA 269:3015-23, 1993.
2Reaven GM: Role of insulin resistance in human disease. Diabetes 37:1595-1607, 1988.
3American Diabetes Association: Position statement: Management of dyslipidemia in adults with diabetes. Diabetes Care 22:S56-59, 1999.
4Haffner SM: Management of dyslipidemia in adults with diabetes: Technical review. Diabetes Care 22:160-78, 1998.
5Grundy SM, Vega GL: Two different views of the relationship of hypertriglyeridemia to coronary heart disease. Arch Intern Med 152:28-34, 1992.
6American Diabetes Association: Position statement: Screening for type 2 diabetes. Diabetes Care 22:S20-23, 1999.
Deborah Thomas-Dobersen, RD, MS, CDE, is in private practice, and Michael J. Dobersen, MD, PhD, is a forensic pathologist at the Arapahoe County Coroner's Office, in Littleton, Colo.
Copyright � 1999 American Diabetes Association
For ADA Related Issues contactCustomerService@diabetes.org
For Technical Issues email@example.com
Diabetes is a chronic, progressive disease with potentially serious sequelae. Treatment for type 2 diabetes often begins with oral agents and eventually requires insulin therapy. As the disease progresses, drug therapies are often intensified and rarely reduced to control glycemia. Conversely, in type 1 diabetes, some patients experience a “honeymoon period” shortly after diagnosis, wherein insulin needs decrease significantly before intensification is needed (1). No comparable honeymoon period has been widely described for type 2 diabetes. However, a few studies have demonstrated that drug-free glycemic control can be achieved in type 2 diabetes for 12 months on average after a 2-week continuous insulin infusion (2–4). Here, we describe an unusual case of a 26-month drug holiday induced with outpatient basal insulin in a patient newly diagnosed with type 2 diabetes.
A 69-year-old white woman (weight 72.7 kg, height 59 inches, BMI 32.3 kg/m2) was diagnosed with type 2 diabetes in June 2011. She presented with an A1C of 17.6% (target <7%) and a fasting blood glucose (FBG) of 452 mg/dL (target 70–130 mg/dL). Before diagnosis, the patient had not used any oral or parenteral steroids nor had she experienced any traumatic physical or emotional event or illness that could have abruptly increased her blood glucose. Metformin 500 mg twice daily was initiated at diagnosis, but was discontinued 9 days later to avoid risk of lactic acidosis, as her serum creatinine was 1.5 mg/dL. At that time, her fasting self-monitoring of blood glucose (SMBG) values ranged from 185 to 337 mg/dL. Treatment with 25 units of insulin detemir daily (0.34 units/kg/day) was initiated in place of metformin. The patient was counseled on diet modifications and encouraged to exercise.
One month later (July 2011), the patient’s fasting SMBG values had improved to a range of 71–212 mg/dL with a single hypoglycemic episode (58 mg/dL); her weight and BMI increased slightly to 74.1 kg and 32.9 kg/m2, respectively. Hypoglycemia education was reinforced, and insulin therapy was switched from 25 units of detemir delivered with the Levemir FlexPen to 28 units (0.38 units/kg/day) of insulin glargine delivered with the Lantus SoloStar due to the patient’s preference for this device. Two weeks later, the patient reported continued improvements in fasting SMBG (70–175 mg/dL) with one hypoglycemic episode (67 mg/dL). In response to the hypoglycemic episode, her insulin glargine dose was decreased to 25 units daily.
In September, the patient reported fasting SMBG values ranging between 71 and 149 mg/dL, and her A1C was 7.9%. On days when the patient skipped lunch, her midday blood glucose level would drop to <70 mg/dL (54–60 mg/dL). She was counseled not to skip meals, and her insulin glargine dose was maintained.
In October, the patient’s weight was 71.4 kg, and her BMI was 31.7 kg/m2. She reported recently initiating a cinnamon supplement and switching her beverage intake from sugar-sweetened products to water and diet soda. Although the majority of her fasting SMBG values were controlled (80–110 mg/dL), she had experienced six hypoglycemic episodes (FBG 13–64 mg/dL). All values were objectively confirmed in the patient’s glucose meter, and the meter was replaced in case of device error. Her daily insulin glargine dose was decreased to 20 units (0.28 units/kg/day).
In December, her SMBG values ranged between 70 and 106 mg/dL preprandially and 111 and 207 mg/dL postprandially, and she had had six additional hypoglycemic episodes (42–66 mg/dL). The patient’s weight remained stable at 71.4 kg (BMI 31.7 kg/m2). At this follow-up visit, her daily insulin glargine dose was decreased further to 15 units (0.21 units/kg/day).
The patient self-discontinued daily insulin glargine in March 2012 but continued using the cinnamon supplements. She continued to perform SMBG 1–3 times/day, anticipating loss of glycemic control. During the next 2 years, her A1C remained stable (from 6.3% in January 2012 to 6.9% in May 2014) (Figure 1).
Daily basal insulin dose and A1C over time. Black triangle = insulin units; black square = A1C.
At a follow-up visit in May 2014, the patient’s SMBG indicated a need for resumed drug therapy (FBG 107–169 mg/dL, postprandial blood glucose 108–328 mg/dL). Her weight at this time was 65.5 kg (BMI 29.1 kg/m2). Insulin glargine was reinitiated at 5 units daily (0.08 units/kg/day).
During the drug-free period of March 2012 to May 2014, the patient maintained her lack of sugar-sweetened beverage consumption. However, she reported having difficulties purchasing healthy food options because of financial constraints. In August 2013, she was specifically encouraged to incorporate physical activity (walking) into her daily routine. The patient’s weight during the drug-free interval declined from 70 kg in March 2012 to 65.5 kg in May 2014.
Hyperglycemia causes pancreatic β-cell toxicity, leading to decreased insulin release (3). In type 1 diabetes, the honeymoon period occurs when residual pancreatic β-cell function is partially restored for an average of 7.2 months, as hyperglycemic stress is removed before the β-cells are ultimately destroyed (1,3).
Past studies demonstrated induction of a drug-free period when patients newly diagnosed with type 2 diabetes were treated with 2–3 weeks of intensive insulin therapy (2–5). Ilkova et al. (2) induced a 12-month drug-free period in 46.2% (n = 6) of patients using an insulin infusion averaging 0.61 units/kg/day. Three patients maintained glycemic control for 37–59 months. Li et al. (3) also induced a 12-month drug-free period in 47.1% (n = 32) of patients with an insulin infusion of 0.7 units/kg. Additional studies indicate that basal-bolus insulin therapy (0.37–0.74 units/kg/day) using NPH and regular insulin can also induce a 12-month drug-free period in a similar percentage of patients (43.8–44.9%) (4,5).
The mechanism of remission appears to be related to resumption of endogenous insulin production after glucotoxicity is resolved. Glucotoxicity has been shown to inhibit first-phase insulin secretion from the pancreatic β-cells (3). Li et al. (3) theorized that an insulin infusion corrects hyperglycemia and removes stress from the β-cells, allowing them to produce insulin, resulting in euglycemia. Their study quantified an increase in secretion of endogenous insulin (44%) and C-peptide (26%) after 2 weeks of continuous insulin infusion. The mechanism through which insulin induces a period of drug-free glycemic control in type 2 diabetes appears to be similar to that causing the honeymoon period in type 1 diabetes.
To our knowledge, this is the first report of basal insulin monotherapy–induced remission of type 2 diabetes. Previous studies required multiple daily injections in a basal-bolus therapy regimen using NPH and regular insulin or hospitalization of patients administered a continuous insulin infusion (2–5).
Basal-only insulin therapy may be a slower method of achieving remission compared to more intensive insulin regimens. In this case, basal insulin was maintained for 9 months. However, according to the FBG trend, discontinuation could have occurred sooner. This report suggests that a trial of basal insulin dosed at 0.2–0.3 units/kg/day, with follow-up every 2–4 weeks in severely hyperglycemic patients with newly diagnosed type 2 diabetes, may be an alternative method to achieving temporary remission. Although this insulin regimen requires a longer timeframe compared to remission induced by basal-bolus therapy or continuous insulin infusion, it provides a more convenient outpatient therapeutic option at a lower cost.
Limitations of this case study include the patient’s use of cinnamon supplementation, which was continued throughout the drug-free period. Although reports are conflicting regarding its efficacy in type 2 diabetes, it is possible that cinnamon may have exerted a mild antidiabetic effect. Positive cinnamon studies have demonstrated a 0.36% A1C reduction after 3 months of use (6). Additionally, the patient’s weight declined by 3.75% during the 9 months of basal insulin therapy, which was likely in response to introducing dietary modifications related to beverage consumption. Most studies suggest that an A1C reduction of 0.36% (7) to 0.66% (8) can be achieved with intensive lifestyle interventions. Therefore, it is unlikely that cinnamon in combination with the mild lifestyle modifications accounted for a nearly 11% A1C reduction from baseline.
Eliminating the consumption of sugar-rich beverages alters the postprandial glycemic curve. In clinical practice, suppressing postprandial blood glucose excursions by adopting significant dietary improvements may postpone or obviate the need for bolus insulin therapy. Likewise, the remission of diabetes potentially may be achieved, as seen in this case, with monotherapy basal insulin when dietary modifications significantly alter the postprandial glycemic curve. However, it is unknown whether remission can be achieved using basal insulin administration alone in patients who choose not to incorporate lifestyle modifications or in patients with baseline healthy eating and exercise habits.
Although weight changes did not appear to contribute to disease remission, the moderate weight loss (6.5%) achieved during the drug-free interval and continued SMBG both may have contributed to maintaining and extending the remission period. The Diabetes Prevention Program (9) showed that lifestyle modifications aimed at achieving a 7% reduction of weight significantly delay the onset of diabetes compared to placebo and metformin. Finally, performing SMBG through the drug-free period may have empowered the patient by providing objective criteria necessary to validate the benefits of lifestyle modifications.
Based on this case, it is possible that initial type 2 diabetes management with basal insulin can temporarily restore β-cell function to a degree to which blood glucose control can be maintained without drug therapy. Although previous studies conducted with intensive insulin regimens have reported response rates nearing 50% for ∼12 months (2–5), future studies should investigate the ideal basal dose, percentage of patient responders, duration of drug-free glycemic control, and mechanism through which this phenomenon occurs. This case further highlights the need to educate every newly diagnosed patient about the treatment of hypoglycemic events.
The purposeful remission of diabetes is not widely attempted or generally considered possible. Although literature exists regarding the temporary honeymoon period experienced after insulin initiation in some people with type 1 diabetes (1), comparatively little research is available regarding the influence of insulin on the remission of type 2 diabetes. Current literature suggests benefit in nearly 50% of patients newly diagnosed with type 2 diabetes using one of the following strategies: a 2-week inpatient insulin infusion or multiple daily injections of basal-bolus therapy (2–5). However, there are disadvantages to these methods. A continuous insulin infusion requires inpatient admission, whereas a basal-bolus insulin regimen requires purchase of two products and administration of multiple subcutaneous injections daily. Unfortunately, both methods may be impractical, costly, and inconvenient for many patients newly diagnosed with type 2 diabetes.
This case outlines a third potential option for inducing remission of type 2 diabetes: basal insulin monotherapy. Using this approach avoids the costly and inconvenient hospital admission required for the continuous insulin infusion strategy. Furthermore, the cost of drug therapy is reduced with the purchase of one rather than two insulin products, as needed in a basal-bolus insulin regimen. Additionally, using basal insulin alone reduces the risk of hypoglycemic events that may occur with stacking of multiple insulin products. Finally, requiring only one injection of insulin each day offers a more manageable alternative for newly diagnosed patients compared to the multiple daily injections required with a basal-bolus insulin regimen.
By using this basal insulin strategy, the patient in this case was able to achieve drug-free glycemic control for 26 months. Early initiation of basal insulin monotherapy in patients newly diagnosed with type 2 diabetes is a more convenient and cost-effective approach than methods previously described and could potentially induce remission of type 2 diabetes in other patients.
Duality of Interest
No potential conflicts of interest relevant to this article were reported.
- © 2016 by the American Diabetes Association.
Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. See http://creativecommons.org/licenses/by-nc-nd/3.0 for details.