Insulin is often prescribed for the management of diabetes during pregnancy. Severe insulin resistance, defined as requiring more than 100 units of insulin per injection or more than 200 units in a day,1 makes achieving glycemic control challenging. Higher doses of insulin require an increased volume of medication, often necessitating multiple injections because generally available syringes hold only 1 mL (100 units). Higher injection volumes increase the risk of medication leakage at withdrawal, extravasation during administration, decrease medication absorption, and increase patient discomfort potentially leading to noncompliance.
In nonpregnant patients, 500 units/mL concentrated insulin (five times more concentrated than conventional insulin) has been used to reduce the volume of delivery in patients who require large insulin doses.2 There is a paucity of literature on both severe insulin resistance in pregnancy as well as the use of 500 units/mL concentrated insulin to treat it.3–7 Moreover, there are no trials available to guide treatment of these patients. A recent editorial in this journal decried this lack of data and pointed out the potential risks of hypoglycemia attendant to 500 units/mL concentrated insulin and noted that hypoglycemic deaths have been increasing in (nonpregnant) individuals with diabetes as a result of tight glycemic control.8 Other approaches to treatment were suggested, including the use of rapid-acting insulin analogs, insulin pumps, and continuous glucose monitoring. The editorial concluded, “…there is still not enough clinical evidence to recommend its use in any patient population, especially pregnant women.” However, in our center we have been using 500 units/mL concentrated insulin for severely insulin-resistant pregnant women with diabetes for over 7 years. The goal of this study is to describe and compare outcomes in pregnant women with severe insulin resistance treated with 500 units/mL concentrated insulin and similarly insulin-resistant patients treated with 100 units/mL concentrated insulin.
PATIENTS AND METHODS
This retrospective cohort study was institutional review board-approved and was designed to compare characteristics and outcomes between women with severe insulin resistance (daily insulin dose requirement of greater than 100 units per injection or greater than 200 units total insulin/d) treated with either 500 units/mL concentrated insulin during their pregnancy or managed with short-acting and intermediate-acting or long-acting 100 units/mL concentrated insulin regimens. All women were cared for in our Diabetes in Pregnancy Program at Women & Infants Hospital in Providence, Rhode Island. In this program, trained nurses provide counseling to patients with both pregestational and gestational diabetes, and tailored nutritional counseling by a certified dietitian is offered. Patients in this program are instructed to monitor their blood glucose at least four times throughout the day, fasting and 2 hours after each meal. Treatment is initiated and modified with the goal of maintaining fasting blood glucose at 95 mg/dL or less and 2-hour postprandial levels at 120 mg/dL or less. At each clinical visit, stored blood glucose values are downloaded by our diabetes educator to an electronic spreadsheet. Patients requiring more than 100 combined units of insulin during their morning or nighttime dose or requiring more than 100 units short-acting or rapid-acting insulin at any dose during the day are considered candidates for 500 units/mL concentrated insulin at the same initial dose. The decision to offer and initiate 500 units/mL concentrated insulin was at the discretion of the health care provider and included consideration of individual patient characteristics such as gestational age (patients at advanced gestational ages were less often offered 500 units/mL concentrated insulin), level of comprehension, likelihood of compliance, and insurance status (some carriers would not pay for 500 units/mL concentrated insulin). All patients with severe insulin resistance seen and treated in our clinic between January 2006 and December 2012 are included in this analysis.
After constructing a list of all eligible study participants, we reviewed the electronic charts and nurse-maintained blood glucose spreadsheets and extracted several variables of interest including patient demographics (age at enrollment, race, parity, prepregnancy weight [by patient recall], weight and height at initial visit, prepregnancy body mass index [calculated as weight (kg)/[height (m)]2], insurance status, and tobacco use), and diabetes characteristics (diabetes type [based on clinical characteristics and history], types and doses of insulin used, gestational age at first visit to the program, recorded blood glucose values, hemoglobin A1C values, gestational age at which 500 units/mL concentrated insulin was initiated, number of hypoglycemia episodes, and need for and reason for antenatal hospitalization). We also collected information regarding obstetric and neonatal complications including birth weight, admission to the neonatal intensive care unit, diagnosis of respiratory complications, or development of neonatal hypoglycemia. The effect of insulin use on glycemic control was estimated by dividing the total number of recorded values for each specific testing time period by the total number of elevated values for that time period. To characterize glycemic control at the end of gestation, we calculated the average fasting and 2-hour postprandial blood glucose values during the last week of treatment. Information regarding compliance with blood glucose monitoring was calculated by dividing the number of actual recorded values over the number of recommended tests for a specific time period. Data on hypoglycemia were also obtained based on two thresholds: less than 60 mg/dL and less than 50 mg/dL.
A two-part analysis was conducted. We first compared baseline patient characteristics, pregnancy outcomes and complications, and glucose control between patients with severe insulin resistance treated with 500 units/mL concentrated insulin at some point during their pregnancy and patients treated only with conventional insulin therapy (a combination of short-acting and intermediate-acting or long-acting 100 units/mL concentrated insulin). Second, we evaluated glucose control and treatment characteristics within the group treated with 500 units/mL concentrated insulin both before and after initiation of this therapy. We included all eligible patients treated in our clinic during the study period and did not calculate a sample size in advance for this observational study. Continuous variables were compared using Student’s t test, whereas categorical variables were compared using χ2 or Fisher’s exact test as appropriate. All tests were two-tailed with P<.05 considered significant. We made no adjustments for multiple comparisons. All statistical analyses were completed using SAS 9.2.
From January 2006 to December 2012, 1,565 pregnant women with preexisting and gestational diabetes were counseled and treated in our clinic. Of these, 166 (10.6%) were identified as having severe insulin resistance. In 78 patients, 500 units/mL concentrated insulin was used during their pregnancies, two of which ended in miscarriage at 16 and 17 weeks of gestation and were excluded from analysis. An additional three women with twin gestation were also excluded, leaving 73 individuals in this analysis. The remaining 88 patients with severe insulin resistance were not started on 500 units/mL concentrated insulin. Two women in this group experienced two pregnancies, each requiring high doses of insulin; only their first pregnancy was included in this analysis. Four pregnancies ended in miscarriage and there were four sets of twins, and all were excluded, leaving 78 women in the conventionally treated group.
Women treated with 500 units/mL concentrated insulin were on average older, more likely to have a diagnosis of type 2 diabetes mellitus, and were seen at an earlier gestational age in our clinic when compared with other patients with severe insulin resistance (Table 1). Patients treated with 500 units/mL concentrated insulin were started on it at a median gestational age of 26.6 weeks and were treated for a median duration of 8.6 weeks during pregnancy (range 1.5–26.3 weeks). Mean blood glucose values in women treated with 500 units/mL concentrated insulin were similar in the week of initiation to the gestational-aged matched mean values of women treated with conventional insulin (Table 1).
The median insulin requirement at the end of the pregnancy for the women not started on 500 units/mL concentrated insulin was 2.5 units of insulin/kg/d (median total dose 240 units, range 140–820 units) compared with 3.0 units of insulin/kg/d (median total dose 300 units, range 82–742 units) in patients treated with 500 units/mL concentrated insulin (P<.01 comparing units/kg/d). A high rate of preeclampsia was noted in both groups (26.8% compared with 17.3%, P=.17). Rates of admission to the neonatal intensive care unit as well as other neonatal complications were similar (Table 2).
The average blood glucose values for the last week of pregnancy were comparable between groups (Table 3). Glycemic control was compared between the conventionally treated group using all values throughout pregnancy and the 500 units/mL concentrated insulin cohort using only values after this medication had been initiated. The rate of overall elevated glucose values was similar between groups, 29.1% compared with 30.4% (P=.62). Compliance with testing was similar between both groups. The rate of glucose values less than 60 mg/dL but not less than 50 mg/dL was significantly higher in patients treated with 500 units/mL concentrated insulin when compared with conventionally treated patients (Table 3). A median of one hospital admission for glycemic control was required within each group (range zero to four in the 500 units/mL concentrated insulin group and zero to six in the conventionally treated group).
Among women treated with 500 units/mL concentrated insulin, the mean percentage of elevated values decreased after initiation of the therapy (Fig. 1). Among women treated with 500 units/mL concentrated insulin, there were 68 admissions for glycemic control; some women were admitted more than once. The frequency of admissions decreased after the initiation of 500 units/mL concentrated insulin; before initiation, admission occurred once every 154 patient-days in our program compared with once every 256 patient-days after initiation. Compliance with blood glucose testing was 69.3% before the initiation of 500 units/mL concentrated insulin compared with 66.5% afterward (P=.38). The percentage of recorded blood glucose values less than 60 mg/dL and less than 50 mg/dL increased after the initiation of 500 units/mL concentrated insulin, from 2.0% to 4.8% (P<.01) and from 0.7% to 2.0% (P<.01), respectively. A single patient required hospital admission for a hypoglycemic episode while using 500 units/mL concentrated insulin. She was symptomatic with a blood glucose value of 37 mg/dL 2 hours after she missed her scheduled meal. With oral alimentation, her blood glucose quickly normalized.
In our program, the use of 500 units/mL concentrated insulin during pregnancy appeared to improve glycemic control in patients with severe insulin resistance without having a measurable effect on compliance. When compared with a group of conventionally treated patients with severe insulin resistance, 500 units/mL concentrated insulin use conferred the same level of glycemic control as the conventional insulin regimen combinations. Patients on 500 units/mL concentrated insulin experienced slightly but significantly more low blood glucose values after medication initiation, but no serious hypoglycemia-attributable adverse outcomes were reported. Obstetric and neonatal complications were similar in patients with severe insulin resistance irrespective of their insulin regimen.
Five hundred units/mL concentrated insulin has been available since 1952; a recombinant formulation of the drug was introduced to the market in 1997. Only a few case reports have described the use of 500 units/mL concentrated insulin during pregnancy.4–6 Multiple trials in nonpregnant patient populations have demonstrated that this therapeutic option has led to improved glycemic control in patients with severe insulin resistance. A recent clinical review evaluated all published trials on 500 units/mL concentrated insulin and determined that its use was associated with an overall decrease in the mean HbA1C by 1.6% from a 10.0% baseline value (P<.001).9
Different algorithms have been proposed for converting 100 units/mL concentrated insulin doses to 500 units/mL concentrated insulin. Pharmacokinetic studies done in nonpregnant populations have demonstrated that 500 units/mL concentrated insulin has an onset of action within the first 30 minutes of administration; peak blood levels are reached after 1.75–4 hours and peak action is seen at 3.5–4.5 hours.10 Five hundred units/mL concentrated insulin appears to have the same onset and peak level time compared with 100 units/mL concentrated insulin but appears to have a longer effect, perhaps because of delayed absorption. Despite this fact, some authors have recommended a 1:1 conversion rate when HbA1C levels are above 8%.11 In our experience, using 1:1 conversion rates and maintaining the dose of intermediate-acting or long-acting insulin led to adequate glycemic control with only a handful of patients requiring decreased dosing of their intermediate-acting insulin. Pharmacokinetic studies of 500 units/mL concentrated insulin have not been conducted in pregnant women and are needed.
One of the major concerns with the use of 500 units/mL concentrated insulin is the risk of hypoglycemia and increased weight gain with escalating insulin doses. Because pregnancy is confined to a finite time period, long-term effects such as weight gain may not be evident. We lacked consistent data on weight gain over the course of the pregnancy to assess this outcome. Trials reporting the use of 500 units/mL concentrated insulin in nonpregnant cohorts have described cases of hypoglycemia associated with its use; in one trial, this rate was calculated to have been one to two episodes per month. However, most trials fail to report any complications with hypoglycemia.12,13
Patient satisfaction has been evaluated in a number of trials. Dailey et al14 reported that patients started on 500 units/mL concentrated insulin were not only more significantly satisfied with this treatment compared with prior treatment, but also had increased satisfaction with the time devoted to their management of diabetes and with their diabetes control. Higher satisfaction rates have also been reported with the use of 500 units/mL concentrated insulin in insulin pumps.13 The increase in satisfaction has been attributed to a decrease in the number of injections, a decrease in the volume of administration, and decreased pain at the site of injection. We did not formally evaluate patient satisfaction in this study.
Certain limitations of our study warrant comment. It was retrospective and patients were not randomly allocated to the two treatment groups. Neither treatment nor outcome assessment was blinded, and the selection of patients for 500 units/mL concentrated insulin therapy was fairly subjective. Our study design did not allow for the evaluation of symptomatic hypoglycemia, because this parameter was not uniformly reported throughout the medical records. Compliance with blood glucose testing was only 60%, introducing the possibility that we either overestimated or underestimated the rates of hyperglycemia or hypoglycemia in either group. Because a minority of included patients had type 1 diabetes mellitus, our findings are more applicable to patients with type 2 diabetes mellitus or gestational diabetes. Although we did not perform formal power analyses, almost certainly we had limited statistical power to detect meaningful differences in outcomes that occurred at a low rate.
Strengths of the study include the fact that, to our knowledge (based on a PubMed search of the English language using the terms “500 units/mL concentrated insulin,” “U-500,” “pregnancy,” and “insulin” from 1970 to March 2013), this is the largest reported cohort (n=73) of pregnant patients treated with 500 units/mL concentrated insulin. We found only five other reported cases.3–7 All patients included in this analysis were enrolled and treated at a single specialized clinic with a standardized approach to care. The availability of the patient's entire blood glucose log allowed the evaluation of patterns before and after administration of 500 units/mL concentrated insulin.
In conclusion, in carefully selected patients, the use of 500 units/mL concentrated insulin appears to be a reasonable option for the management of severe insulin resistance during pregnancy. Its use is associated with a small increase in the risk of hypoglycemia. Prospective and randomized studies are needed to better estimate safety, appropriate dosing and pharmacokinetics, pregnancy outcomes, patient satisfaction, and compliance when 500 units/mL concentrated insulin is used in pregnancy.
1. Quinn SL, Lansang MC, Mina D. Safety and effectiveness of U-500 insulin therapy in patients with insulin-resistant type 2 diabetes mellitus. Pharmacotherapy 2011;31:695–702.
2. Cochran E, Musso C, Gorden P. The use of U-500 in patients with extreme insulin resistance. Diabetes Care 2005;28:1240–4.
3. Dolberg BK, Lenhard MJ. Successful outcome of pregnancy in a patient with generalized lipoatrophic diabetes mellitus. Endocr Pract 2000;6:34–6.
4. Zuckerwise LC, Werner EF, Pettker CM, McMahon-Brown EK, Thung SF, Han CS. Pregestational diabetes with extreme insulin resistance: use of U-500 insulin in pregnancy. Obstet Gynecol 2012;120:439–42.
5. Hatipoglu B, Soni S, Espinosa V. Glycemic control with continuous subcutaneous insulin infusion with use of U-500 insulin in a pregnant patient. Endocr Pract 2006;12:542–4.
6. Wong VW, Pape AV. Extreme insulin resistance in a patient with pre-existing diabetes during pregnancy: role of U-500 insulin. Med J Aust 2009;190:650–1.
7. Okeigwe I, Yeaton-Massey A, Kim S, Vargas JE, Murphy EJ. U-500R and aspart insulin for the treatment of severe insulin resistance in pregnancy associated with pregestational diabetes. J Perinatol 2013;33:235–8.
8. Reece EA. Use of U-500 insulin in pregnancy is intriguing, but controlled trials are needed. Obstet Gynecol 2012;120:435–6.
9. Lane WS, Cochran EK, Jackson JA, Scism-Bacon JL, Corey IB, Hirsch IB, et al.. High-dose insulin therapy: is it time for U-500 insulin? Endocr Pract 2009;15:71–9.
10. Wafa WS, Khan MI. Use of U-500 regular insulin in type 2 diabetes. Diabetes Care 2006;29:2175–6.
11. Segal AR, Brunner JE, Burch FT, Jackson JA. Use of concentrated insulin human regular (U-500) for patients with diabetes. Am J Health Syst Pharm 2010;67:1526–35.
12. Ballani P, Tran MT, Navar MD, Davidson MB. Clinical experience with U-500 regular insulin in obese, markedly insulin-resistant type 2 diabetic patients. Diabetes Care 2006;29:2504–5.
13. Bulchandani DG, Konrady T, Hamburg MS. Clinical efficacy and patient satisfaction with U-500 insulin pump therapy in patients with type 2 diabetes. Endocr Pract 2007;13:721–5.
14. Dailey AM, Williams S, Taneja D, Tannock LR. Clinical efficacy and patient satisfaction with U-500 insulin use. Diabetes Res Clin Pract 2010;88:259–64.