When Oral Agents Are Failing: The Transition to Insulin Therapy Volume 1 Number 1 August 2003

Review each section of this article using the links below.

• Overview
• Case 1
• Case 2
• Conclusion
• References

Overview

Type 2 diabetes mellitus is a chronic progressive disease that affects approximately 17 million people in the United States. This disease is characterized by insulin resistance and progressive failure of the pancreas to produce sufficient insulin. Whereas the disease can be controlled with oral medications (particularly early in its course), pancreatic dysfunction often progresses to the point at which oral agents are no longer effective at controlling blood sugars. In the United Kingdom Prospective Diabetes Study (UKPDS),¹ which followed nearly 4000 patients for 10 years, almost half the patients required insulin by study end, supporting clinical evidence that most patients with type 2 diabetes will need insulin during their lifetime. Other studies, such as the VA Cooperative Study of Type 2 Diabetes Mellitus (VA CSDM) and the Kumamoto study, demonstrated that intensive insulin treatment has efficacy in controlling A1C levels and preventing or reducing some complications associated with diabetes.

Case 1

A 57-year-old woman diagnosed with type 2 diabetes mellitus 9 years previously was referred for management of uncontrolled blood glucose. Diabetes-related complications included retinopathy and nephropathy. She described several weeks of increased polyuria and polydipsia. Blood glucose levels at home were in the 347- to 514-mg/dL range. Her physical activity was limited by comorbid osteoarthritis. She had recently seen a nutritionist and was following a reasonable dietary regimen. At presentation, this patient's medications included lisinopril 5 mg/QD and glipizide 10 mg BID. Previous experience of gastrointestinal side effects of metformin precluded its use. Pertinent laboratory studies included an A1C of 11.5% (normal 4.3%–5.8%). Six months ago the A1C was 7.8%. Serum chemistries revealed normal renal and liver function.

Comment

This patient presents with marked, uncontrolled hyperglycemia despite lifestyle interventions and pharmacotherapy with maximal doses of a sulfonylurea. Because laboratory testing 6 months previously revealed a suboptimal A1C, reflecting poor glycemic control, the addition of other oral agents was contemplated. Although she was previously metformin-intolerant, the extended-release form of the agent, which may be better tolerated, had never been prescribed. The patient has normal liver function tests, so a thiazolidinedione such as rosiglitazone or pioglitazone may also be considered. However, with the addition of one more oral agent, fasting blood glucose is expected to drop by only 20 to 60 mg/dL, and A1C may go down by an additional 1% to 1.5%.^2,3 Therefore, her blood sugars are too high to expect additional oral agents to provide a significant impact on glucose control. The most effective option remaining to enable attainment of targeted glucose control, to relieve the symptoms of hyperglycemia, and to prevent the development or worsening of diabetic complications is initiation of insulin therapy. The physician considered the use of one of the analog mixes including, insulin aspart protamine/insulin aspart (NovoLog^® Mix 70/30) or insulin lispro protamine/insulin lispro (Humalog^® Mix 75/25).

The patient was started on insulin aspart protamine/insulin aspart (NovoLog® Mix 70/30) beginning with 10 U, 15 minutes before breakfast, and 10 U, 15 minutes before dinner. This premixed insulin contains 70% intermediate-acting insulin aspart protamine and 30% rapid-acting aspart insulin. The NPH provides a basal effect, and the aspart has prandial insulin action. Her glipizide was discontinued, and the insulin dose was titrated over the telephone to 20 U in the morning and 22 U in the evening. She returned to the office a few weeks later with the following blood glucose levels:   Breakfast Lunch Dinner Bedtime Sun 213 147 166 193 Mon 146 108 115 84 Tues 260 129 -- 98 Wed 130 70 -- 129 Two months later, a repeat A1C was 7.4%, which is much closer to the ADA recommended A1C goal of <7% ADA and AACE recommended goal of <6.5%.

In this patient, insulin therapy was successfully initiated for long-standing, poor glycemic control, which raises several questions:

• What factors should be considered in choosing the type of insulin or insulin analog?
• How should the initial dose be determined and subsequently titrated?
• Should the patient's oral agents be discontinued?

Available therapies

	Time to peak insulin (min)	Onset of action (min)	Time to peak action (hr)	Duration of action (hr)
Aspart	40-50	10-20	1-3	3-5
Lispro	30-90	10-15	.8-4.3	3-5
Regular	80-120	30	2.5-5	8
NPH	--	1.5	4–12	24
Lente	--	2.5	7-15	22
Ultralente	--	6-10	16-20	28
Glargine	--	1-1.5	Peakless	24

Human insulins and insulin analogs are available in preparations that provide basal or prandial and/or supplemental insulin action. Basal insulin is used to control fasting and between-meal glucose levels. Prandial insulin controls postprandial glucose excursions. Supplemental insulin doses are used for the correction of hyperglycemia.

Basal insulin needs are met by NPH, lente, ultralente and/or glargine preparations. To provide a continuous basal insulin effect over a 24-hour period, NPH and lente insulin are generally given twice daily, before breakfast and at dinner or bedtime. Ultralente and glargine are administered once daily, generally at bedtime. Even though these are basal insulins, they do have the potential to cause hypoglycemia if patients do not time mealtimes or snacktimes with the insulin's peak effect. In a study by Yki-Jarvinen and reported in Diabetes Care, insulin glargine produced less nocturnal hypoglycemia and lower post-dinner glucose concentrations compared with NPH.⁴ It also has more consistent absorption than human insulins. However, basal therapy with insulin glargine alone will not allow control of postprandial hyperglycemia in patients who do not have a residual ability to secrete endogenous insulin with meals.

Prandial and supplemental/correction insulins are short-acting regular insulin and rapid-acting aspart or lispro insulin preparations. Regular insulin is administered 30 to 45 minutes before meals and aspart or lispro 0 to 15 minutes before meals, to allow time to onset of action of each to match the postprandial rise in blood glucose level.

Basal insulin may be used in combination with prandial insulin or oral agents to attain targeted blood glucose control. Multiple daily insulin injections produce the most physiologic insulin regimen: Long-acting analog glargine is given once daily and prandial insulin, (eg, aspart or lispro), is given with each meal. Although this regimen requires 4 injections per day, these insulins give patients the greatest flexibility with regard to timing and carbohydrate content of meals. A sulfonylurea or meglitinide (insulin secretagogue) agent can be used for stimulation of insulin secretion with meals. For initiation of insulin therapy in reluctant patients, once-daily basal insulin in combination with a secretagogue drug, may be more acceptable than multiple injections.

Premix insulins provide ease of administration: no mixing or BID dosing, and availability in pen devices. Although these features may be important for elderly or visually impaired patients who cannot draw up insulin into a syringe, one cannot finely titrate the premix dose for tight control without increased risk of hypoglycemia. The premixed combinations also require that patients have consistent meals. Analog mixes may be a little better than human mixes at controlling blood glucose and minimizing hypoglycemia.⁵

Guidelines for starting insulin doses
There is no recommended starting dose; however, a person's body weight, degree of insulin resistance, and lifestyle (including diet and exercise) should be factors in determining the initial dose. In addition, preventing hypoglycemia during initiation of insulin therapy is desirable. In type 2 diabetes, the usual total daily insulin requirement is 0.4-1.0 units/kg/day. Half (50%) of this amount is generally given as basal insulin. The rest is divided into prandial doses. Individual doses vary widely. Frequent adjustments based on resultant blood glucose testing results should be made.

In this patient, it was thought that both the long duration of diabetes and high blood glucose levels were indicative of advanced beta cell dysfunction, which would require both background as well as prandial insulin. Therefore, a premixed analog regimen was prescribed as the initial insulin therapy. A low dose was begun empirically but was titrated quickly during frequent phone calls and visits with the nurse practitioner, diabetes educator, and physician.

As to whether the patient should remain on an oral agent, some data address the issue. In a study by Riddle,⁶ patients who were taking glimepiride were randomized to either (1) insulin 70/30 once in the evening in addition to the glimepiride, or (2) insulin 70/30 plus placebo. The A1C in both groups was reduced from 9.9% to 7.7% in 24 weeks. Patients who remained on glimepiride improved more quickly and required significantly less insulin (49 U vs 78 U). It is not known how long the glimepiride would have continued to benefit patients or whether the lower insulin dose would have had any impact on long-term outcomes.

Case 2

A 60-year-old woman with type 2 diabetes mellitus, diagnosed 6 years ago, was referred for diabetes management. She reported that her blood glucose levels, tested at home, ranged from 200 to mid-300 mg/dL. She had a regular exercise routine that included walking 4 to 5 days a week and was reasonably adherent to her diet. This patient is 5'5" and weighs 159 pounds with a body mass index (BMI) of 26.5. She described burning pains in both feet at night. Medications at presentation included glyburide 10 mg BID and metformin 500 mg BID; she reported that higher doses of metformin had caused diarrhea. Pioglitazone had been prescribed in the past, but its use was discontinued because of pedal edema. Laboratory results are significant for A1C of 9.3%. She was reluctant to start insulin but agreed to one injection of insulin glargine, 10 U, once at bedtime. The dose was then titrated over the telephone. Three weeks later, she came for a follow-up visit, at which time she was administering insulin glargine 25 U at bedtime and was continuing her oral agents. Fasting blood glucose (FPG) levels were between 70 and 100 mg/dL, but postprandial glucose levels were all in the high 200s. Her physician suggested supplementing with prandial insulin, but the patient did not want to take 4 injections per day. However, she was willing to take 2 injections per day, so her physician prescribed insulin aspart protamine/insulin aspart (NovoLog® Mix 70/30, containing 70% basal insulin aspart protamine and 30% prandial insulin aspart) with breakfast and dinner. The dose was titrated to 40 U BID. Blood sugars were erratic, either high in the 200s or low in the 50s, with symptoms of hypoglycemia. At this point, she accepted her physician's recommendations for multiple daily injections. She was changed back to glargine 30 U once a day with insulin aspart (NovoLog®) 10 U with meals using the FlexPen™. She is very pleased with her ability to control her blood sugars and does not mind administering insulin 4 times per day.

Comment

This case illustrates several important principles involved in selecting insulin regimens. Patients are often reluctant to initiate insulin therapy, and when they do, they prefer regimens that are the least complex. Once-daily injection of glargine, alone or combined with oral agents, can be effective. It clearly improved glycemic control in this patient, but postprandial blood sugars were well outside the target range; it was apparent that she needed a prandial insulin as well in order to achieve optimal glycemic control. Adding a fast-acting analog such as insulin aspart or lispro to the glargine would have been effective. The patient did not want to take 4 injections, so an analog mix was prescribed. Unlike the patient in Case 1 who took the premix insulin, this patient's blood glucose control was erratic, with an excessive number of hypoglycemic episodes. In order for patients to be successful on a premixed insulin, it is necessary for them to adhere to a mealtime schedule that enables the postprandial rise in blood glucose to coincide with the peak effect of the insulin mix. Insulin mixes have become the most popular of the prescribed insulin products due to their ease of use and reasonable effectiveness in improving hyperglycemia; however, as this case illustrates, they can have their limitations. The four injection per day regimen with a short-acting insulin and a background insulin (see Table above) provides more physiologic insulin dynamics and allows patients greater flexibility with regard to the timing of meals. Once this patient found she could control her blood glucose well with glargine, as a background insulin, and prandial insulin aspart injections, she was willing to continue with 4 injections per day. She administers the glargine at bedtime and carries her disposable NovoLog FlexPen wherever she goes. She is working with a dietitian to learn carbohydrate counting so that she may adjust the aspart dose to the carbohydrate content of each meal. As a result, both she and her physician are more satisfied with her glycemic control.
Conclusion
As demonstrated in these case studies, insulin should be initiated without delay in patients with uncontrolled diabetes with markedly elevated blood sugars. It may be beneficial to maintain patients on their oral medications. Depending on the patient, a physician may choose from once- or twice-daily injections of a background insulin or insulin mix or multiple daily injections. The starting dose of insulin will depend on patient factors and should be titrated weekly, by telephone or brief office visits, until acceptable glycemic control is achieved. The selection of the type of insulin can be made by considering the advantages and limitations of the various regimens as illustrated in the above cases. A stepwise approach of starting insulin glargine and titrating it, then adding a short-acting analog when needed, is a reasonable strategy.

References

1. UK Prospective Diabetes Study (UKPDS) Group. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet. 1998;352:837-853.
   
   
2. DeFronzo RA, Goodman AM, for the Multicenter Metformin Study Group. Efficacy of metformin in patients with non-insulin-dependent diabetes mellitus. N Engl J Med. 1995;333:541-549.
   
   
3. Yale JF, Valiquett TR, Ghazzi MN, et al. The effect of a thiazolidinedione drug, troglitazone, on glycemia in patients with type 2 diabetes mellitus poorly controlled with sulfonylurea and metformin. A multicenter, randomized, double-blind, placebo-controlled trial. Ann Intern Med. 2001;134(9 pt 1):737-745.
   
   
4. Yki-Jarvinen H, Dressler A, Ziemen M; HOE 901/300s Study Group. Less nocturnal hypoglycemia and better post-dinner glucose control with bedtime insulin glargine compared with bedtime NPH insulin during insulin combination therapy in type 2 diabetes. Diabetes Care. 2000;23:1130-1136.
   
   
5. Boehm BO, Home PD, Behrendt C, Kampt NM, Lindholmt A. Premixed insulin aspart 30 vs. premixed human insulin 30/70 twice daily: a randomized trial in Type 1 and Type 2 diabetic patients. Diabetic Medicine. 2002; 19: 393-399.
   
   
6. Riddle MC, Schneider J, for the Glimepiride Combination Group. Beginning insulin treatment of obese patients with evening 70/30 insulin plus glimepiride versus insulin alone. Diabetes Care. 1998;21:1052-1057.