- Email: [email protected]
A comparison of insulin lispro and buffered regular human insulin administered via continuous subcutaneous insulin infusion pump
Journal of Diabetes and Its Complications 15 (2001) 295 – 300
A comparison of insulin lispro and buffered regular human insulin administered via continuous subcutaneous insulin infusion pump$ Philip Raskina,*, John H. Holcombeb, William V. Tamborlanec, John I. Maloned, Suzanne Strowiga, Jo Ann Ahernc, Francine Laventd a
Department of Internal Medicine, University of Texas Southwestern Medical Center at Dallas, 5323 Harry Hines Boulevard, G5-238, Dallas, TX 75390-8858, USA b Lilly Research Laboratories, Eli Lilly and Company, Lilly Corporate Center, Indianapolis, IN 46285, USA c Section of Pediatric Endocrinology, Yale University School of Medicine, PO Box 208064, 33 Cedar St., New Haven, CT 06520-8064, USA d Department of Pediatrics, College of Medicine, University of South Florida, 4202 E. Fowler Ave., Tampa, FL, USA Received 24 May 2001; accepted 20 June 2001
Abstract This study compared glycemic control achieved with insulin lispro or buffered regular human insulin in patients with Type 1 diabetes treated with continuous subcutaneous insulin infusion (CSII) using an external insulin pump. In this 24-week multicenter, randomized, twoway crossover, open-label trial, 58 patients on CSII with adequate glycemic control received either insulin lispro or buffered regular human insulin for 12 weeks, followed by the alternate treatment for another 12 weeks. Efficacy and safety measures included hemoglobin A1c (HbA1c) at baseline and endpoint, home blood glucose monitoring, hypoglycemia, and frequency of pump catheter occlusion. Patients consumed a standard test meal on three occasions, with determinations of fasting, 1- and 2-h postprandial glucose values. Insulin lispro use was associated with a significantly lower HbA1c than was buffered regular human insulin (7.41 ± 0.97 vs. 7.65 ± 0.85 mmol/l; P = .004). Fasting serum glucose values before the test meal were similar between the two therapies. The 1-h (11.16 ± 4.29 vs. 13.20 ± 4.68 mmol/l; P = .012) and 2-h (9.64 ± 4.10 vs. 12.53 ± 4.64 mmol/l; P = .001) postprandial glucose concentrations were significantly lower during treatment with insulin lispro. No differences between treatments were observed in basal or bolus insulin doses, weight gain, or the incidence and rate of hypoglycemia, hyperglycemia, or pump occlusions. When used in external pumps, insulin lispro provides better glycemic control than buffered regular human insulin with a similar adverse event profile. D 2001 Elsevier Science Inc. All rights reserved. Keywords: Insulin; Lispro; Insulin pump; HbA1c
1. Introduction The goal of treatment for individuals with diabetes should be a normalization of blood glucose levels, such that the hemoglobin A1c (HbA1c) is in the range of individuals without diabetes. The results of the Diabetes Control and Complications Trial (DCCT) clearly show that the microvascular complications, and perhaps even macrovascular complications of diabetes, are related to hyperglycemia (The Diabetes Control and Complications Trial Research Group, 1993). Intensive therapy resulting in
$
Eli Lilly and Company sponsored this research. * Corresponding author. Tel.: +1-214-648-2017; fax: +1-214-6484854.
near-normal blood glucose levels lead to an approximately 50% reduction in the development and progression of microvascular complications. However, this therapy was associated with a threefold increase in the incidences of hypoglycemia compared to standard treatment. Thus, it is imperative to develop treatment strategies that can achieve euglycemia without excessive hypoglycemia. To date, that has been difficult; for example, less than 5% of the intensively treated subjects in the DCCT study were able to achieve HbA1c values in the range of nondiabetic individuals (The Diabetes Control and Complications Trial Research Group, 1993). This goal has been difficult to reach, in part, because we still lack the treatment tools needed to completely normalize blood glucose levels. However, new therapeutics and tools are becoming available that show promise in improving blood glucose control.
1056-8727/01/$ – see front matter D 2001 Elsevier Science Inc. All rights reserved. PII: S 1 0 5 6 - 8 7 2 7 ( 0 1 ) 0 0 1 6 8 - 4
296
P. Raskin et al. / Journal of Diabetes and Its Complications 15 (2001) 295–300
Portable insulin infusion devices are effective insulin delivery systems for managing Type 1 diabetes (Mecklenburg et al., 1982; Tamborlane, Sherwin, Genel, & Felig, 1979), lower HbA1c levels have been achieved when treatment with insulin pumps is compared to multiple daily insulin injections (The Diabetes Control and Complications Trial Research Group, 1995). However, severe hypoglycemia remains common with both methods of insulin administration (The Diabetes Control and Complications Trial Research Group, 1995). Rapidly absorbed insulin analogs, such as insulin lispro, may offer an advantage over regular human insulin for insulin pumps by more effectively lowering postprandial hyperglycemia without increasing the prevalence of hypoglycemia (Zinman, Tildesley, Chiasson, Tsui, & Strack, 1997). Insulin lispro is a rapid-acting analog of human insulin with a quicker onset and a shorter duration of action than regular human insulin (Howey, Bowsher, Brunelle, & Woodworth, 1994; Koivisto, 1998). After subcutaneous injection of insulin lispro, peak activity is observed between 30 and 90 min, compared with regular human insulin, which has a peak effect between 2 and 4 h (Howey et al., 1994). The peak effect of regular human insulin is usually achieved 2 to 6 h after injection and its effects may last as long as 16 h. However, since the peak glycemic response to a mixed meal has been found to be between 26 and 70 min after ingestion, regular human insulin may peak too late to control postprandial blood glucose levels. This pharmacodynamic difference is important because an early peak of insulin secretion is very effective in suppressing hepatic glucose production, the main contributor to the postprandial hyperglycemia observed in patients with Type 1 diabetes (Tuominen, Karonen, Melamies, Bolli, & Koivisto, 1995). In addition, the use of insulin lispro is associated with a lower frequency of hypoglycemia (Anderson et al., 1997) or severe hypoglycemia (Brunelle, Llewelyn, Anderson, Gale, & Koivisto, 1998) compared with regular human insulin. Various unmodified insulin preparations have been used in continuous subcutaneous insulin infusion (CSII) pumps since they became available in the late 1970s. These insulins have included mixed beef – pork, purified pork, and biosynthetic human insulins. Insulins containing phosphate buffer may be better suited for external pump usage, because the incidence of catheter occlusion is reportedly less than with use of neutral insulin (Ratner & Steiner, 1987). Newer infusion catheter tubing, such as Polyfin (MiniMed, Sylmar, CA) may decrease the occlusion rate without regard to the type of insulin used. Velosulin BR (buffered regular human insulin injection, Novo Nordisk, Princeton, NJ) is the only phosphate-buffered insulin approved in the United States for use in external insulin pumps. However, insulin lispro (Humalog, Eli Lilly and Company, Indianapolis, IN) is also phosphate-buffered. Therefore, this study was designed to compare the effects of the two phosphate-buffered insulins, insulin lispro and buffered regular human insulin, on
glycemic control in patients with Type 1 diabetes who were receiving insulin by CSII.
2. Patients and methods 2.1. Subjects Patient demographics are shown in Table 1. Patients were invited to participate if they had Type 1 diabetes according to the World Health Organization classification (Pickup & Williams, 1991), were between the ages of 13 and 60 years, had achieved acceptable compliance with insulin pump therapy and a nutritional regimen, and had been treated with CSII for at least 6 months before entering this study. Patients were excluded from the study if they had total glycated hemoglobin values or an HbA1c more than 2.0 times the upper limit of the normal range, clinically significant renal, hepatic, or cardiac disease, cancer, drug or alcohol abuse, insulin allergy, recurrent severe hypoglycemia, anemia; life expectancy of less than 3 years, lactating, pregnant, or intended to become pregnant, or require dilution of insulin in their pump. All patients gave informed consent according to Good Clinical Practice Guidelines and the Declaration of Helsinki. 2.2. Study design This was a 6-month open-label randomized crossover clinical trial conducted at three sites in the United States. During this study, only MiniMed models 504, 504-S, 506, or 507 were used, along with that manufacturer’s associated tubing catheters (Sof-set, Sof set QR, Polyfin, or Polyfin QR). Each patient continued to use the same model of MiniMed pump throughout this study. At the initial visit, all patients began treatment with buffered regular human insulin (Velosulin BR). During a 2- to 4-week lead-in period, diabetes control was to be maintained or improved according to individual goals by adjusting the basal or bolus insulin doses, or both. Because the stability of diluted Table 1 Patient characteristics at baseline (mean ± S.D.) Treatment sequence Variable N Female Male Age Weight (kg) Duration of diabetes (years) Fasting serum glucose (mmol/l) HbA1c (%)
Insulin lispro; regular insulin
Regular insulin; insulin lispro
28 12 16 40.5 ± 8.7 78.3 ± 17.9 18.8 ± 7.6
30 16 14 37.8 ± 9.7 77.3 ± 16.7 17.4 ± 8.5
8.7 ± 3.2
7.45 ± 2.7
.169
7.9 ± 1.1
7.6 ± 0.8
.234
P value
P. Raskin et al. / Journal of Diabetes and Its Complications 15 (2001) 295–300
insulin lispro had not been investigated at the time of this study, insulin dilution was not allowed. After the lead-in period, the patients were randomized either to continue buffered regular human insulin or to begin insulin lispro. Randomization was such that there would be approximately equal numbers of patients in each treatment sequence at each investigative site. Premeal insulin doses were given immediately before each meal. Both treatments were continued for 12 weeks after which time the patients received the alternate insulin for an additional 12 weeks. The patients were seen every 6 weeks throughout the study. Glycemic goals for this study were preprandial blood glucose values below 6.7 mmol/l ( < 120 mg/dl) and maintenance of 2-h postprandial glucose values below 10 mmol/l ( < 180 mg/dl) without causing hypoglycemia. The dose of insulin was increased to achieve these goals according to the results of home blood glucose monitoring, unless further increase would result in an increased risk of hypoglycemia. The basal and bolus doses were adjusted according to each patient’s metabolic needs. At the end of the lead-in period and at the end of each treatment period, HbA1c and fasting plasma glucose were measured. After these blood samples were obtained, each patient consumed a test meal (Sustacal, Mead Johnson, Evansville, IN) consisting of 360 cal, with 55% of calories from carbohydrate, 21% from fat, and 24% from protein. The patient was given 1– 2 min to consume the test meal. One- and two-hour postprandial serum glucose measurements were obtained with the patient at rest. The appropriate study insulin (insulin lispro or buffered regular human insulin) was administered immediately before the test meals. The HbA1c measurements were performed by a central laboratory (Covance, Indianapolis, IN) using high-performance cation exchange chromatography (reference range: 4.3 –6.1%) at baseline and at the end point of each study period. Serum glucose levels were also measured by Covance, using a Hitachi 747 analyzer. Patients recorded in a diary the insulin doses (bolus and basal doses) used on the day before each clinic visit. In addition to the test meal data, patients were asked to perform home blood glucose monitoring using a meter with memory capability. At the clinic visits, the results of the self-blood glucose monitoring were downloaded and the summary statistics of glucose values obtained during the 2 weeks before each visit were recorded. At each visit, information on hypoglycemic and hyperglycemic events, in addition to other questions related to safety, were recorded. Hypoglycemia was defined as any time a patient believed (or another person observed) that he or she was experiencing at least one sign or symptom that could be associated with hypoglycemia or the patient experienced a measured blood glucose less than 3.0 mmol/l (54 mg/dl). These episodes were reported spontaneously and were recorded by the patients in their study diaries. In addition to information on hypoglycemia, unexplained episodes of hyperglycemia were also recorded. An episode of hyper-
297
glycemia was defined as 2 or more consecutive blood glucose measurements greater than 13.8 mmol/l (250 mg/ dl) occurring within a 2- to 4-h period and that did not respond to insulin boluses or otherwise remained unexplained. For each episode of hyperglycemia, the patient was asked to answer three questions intended to characterize the cause of the hyperglycemia. If the hyperglycemia was thought to be due to occlusion, the patient was asked to identify the source of the interrupted insulin flow through the catheter, such as blood in the tubing, kink in the tubing, visible occlusion, or other causes. 2.3. Statistical methods An intent-to-treat analysis was performed using all the data from all the patients. All comparisons for efficacy variables were performed using two-tailed tests with a nominal significance level of .05. A separate analysis of variance (ANOVA) model was used to examine for carryover, investigator or treatment effects.
3. Results Table 1 contains the patient characteristics. Fifty-eight patients completed the study. Three patients discontinued before randomization (two due to personal reasons; one due to failure to meet protocol entry criteria) and one patient discontinued the study after randomization, for personal reasons. Because no carryover or investigator effects were discovered by ANOVA for any of the efficacy measures, the data from the crossover treatment periods were combined. Mean insulin doses were similar between the two treatments. The mean basal insulin doses were 0.34 and 0.35 U/ kg and the mean bolus insulin doses were 0.28 and 0.30 U/ kg for the insulin lispro and regular human insulin treatment groups, respectively. Fig. 1 shows the results of the test meal. The fasting glucose values on the morning of the test meal did not differ between the treatment groups. The 1-h glucose value was lower during treatment with insulin lispro (insulin lispro: 11.16 ± 4.29 mmol/l vs. regular human insulin: 13.20 ± 4.68 mmol/l; P = .012, two-sample t test). The 2-h glucose value was also significantly lower during therapy with insulin lispro (insulin lispro: 9.64 ± 4.10 mmol/l vs. regular human insulin: 12.53 ± 4.64 mmol/l; P = .001, two-sample t test). Fig. 2 shows the HbA1c values at baseline and at the end of each treatment period. No difference was observed between the therapies at baseline. For the combined treatment groups, use of insulin lispro was associated with significantly lower HbA1c compared with regular human insulin (insulin lispro: 7.41 ± 0.97% vs. regular human insulin: 7.65 ± 0.85%; P = .004, two-tailed t test). At the end of the first treatment period, the change from baseline was greater for the insulin lispro group compared with regular human insulin (insulin lispro: 0.34 ± 0.60
298
P. Raskin et al. / Journal of Diabetes and Its Complications 15 (2001) 295–300
Fig. 1. Mean serum glucose before and after a test meal following a preprandial injection of either insulin lispro or buffered regular human insulin. Closed circles: buffered regular human insulin; open triangles: insulin lispro. Error bars represent S.D.
vs. regular human insulin: 0.06 ± 0.58; P = .028, two-tailed t test) (Fig. 3). Similarly, for the combined treatment groups, the overall change from baseline was greater during therapy with lispro compared with regular human insulin (insulin lispro: 0.34 ± 0.59% vs. regular human insulin: 0.09 ± 0.63%; P = .004). The number of home blood glucose measurements carried out during the 2 weeks before each visit did not differ between the two therapy groups (insulin lispro: 56 ± 23 vs. regular human insulin: 56 ± 19; P = .943). The mean blood glucose level during therapy with insulin lispro and regular human insulin was also similar (8.1 ± 2.0 vs. 8.1 ± 1.6 mmol/l, respectively; P = .792). The overall mean of the maximum and the mean of the minimum glucose values did not differ between the therapies. Patients had 41% of their glucose measurements within the target range, irrespective of therapy. Forty-one percent of the patients’
Fig. 2. Mean HbA1c values at baseline and at the end of each 3-month period for both treatment sequence groups.
Fig. 3. Change in HbA1c from baseline to the end of the first 3-month period in patients treated with insulin lispro or buffered regular human insulin. Error bars represent S.D.
glucose values were above target and 18% of readings were below target. During the entire study, 10 patients experienced a total of 26 episodes of hypoglycemia. For the lead-in period, during which all patients received buffered regular human insulin, seven episodes of hypoglycemia were observed, two episodes required treatment with intravenous glucose. After randomization, hypoglycemic events were infrequent and similar between therapies. With insulin lispro treatment, seven patients reported a total of eight hypoglycemic episodes, and during therapy with regular human insulin, seven patients reported a total of 11 episodes. Hypoglycemia requiring intravenous glucose occurred in two patients (three episodes) and in three patients (three episodes) during treatment with regular human insulin and insulin lispro, respectively. Thirty-eight patients recorded 109 episodes of hyperglycemia, 39 (35.7%) were caused by occlusion, 47 (43.1%) were caused by reasons other than occlusion, and 23 (21.1%) had no identifiable cause. The 39 episodes of hyperglycemia due to occlusion were reported by 20 patients (eight patients using insulin lispro, 16 episodes; 12 patients using buffered regular human insulin, 23 episodes). There were no significant associations between therapies and a specific cause of occlusion, such as kinked tubing, blood in tube, or visible occlusion. None of the episodes of occlusion resulted in an adverse event. There were no discontinuations during the study due to adverse events. One patient receiving buffered regular human insulin was hospitalized for treatment of fever, vomiting, and dehydration, and one patient receiving insulin lispro was hospitalized for treatment of ketosis. There was no significant difference in mean ( ± S.D.) weight at endpoint between patients receiving insulin lispro therapy and patients receiving buffered regular human insulin therapy (79.2 ± 17.1 kg vs. 78.8 ± 17.3 kg, respect-
P. Raskin et al. / Journal of Diabetes and Its Complications 15 (2001) 295–300
ively; P = .780). Although patients in both therapy groups gained approximately 1 kg by the end of the first treatment period, this change in weight from baseline was not statistically significant.
4. Discussion The introduction of analogs of human insulin such as insulin lispro has greatly improved the therapeutic options for the management of Type 1 diabetes. In particular, insulin lispro seems well suited for use in portable insulin infusion devices. The quick onset and short duration of action of insulin lispro allow individuals with Type 1 diabetes to take a premeal insulin bolus immediately before or after eating and helps to reduce postprandial hyperglycemia (Anderson et al., 1997; Ebeling et al., 1997; Heinemann et al., 1996; Koivisto, 1998; Pampanelli et al., 1995). Consistent with the absorption kinetics of the insulins used in this study, the use of insulin lispro resulted in significantly lower 1- and 2-h postprandial glucose levels following the test meal than that achieved with regular human insulin. Although Sustacal is frequently used for test meals and is nutritionally balanced, it does not represent an actual meal as it is in liquid form. However, the postprandial findings remain valid because two insulins were compared in the present study. In this study, therapy with insulin lispro lowered HbA1c compared with regular human insulin (7.4% vs. 7.6%; P = .004). This difference in HbA1c is consistent with the findings of Zinman et al. (1997), who compared insulin lispro with regular human insulin in a 3-month, masked, crossover study involving 30 patients with Type 1 diabetes. In that study, use of insulin lispro was associated with significantly lower HbA1c values (insulin lispro: 7.7%, regular human insulin 8.0%; P = .004). In another crossover trial similar in design to the current study, Melki et al. (1998) also reported a significantly greater reduction in HbA1c from baseline during use of insulin lispro compared with regular human insulin (insulin lispro: 0.62% vs. regular human insulin: 0.09%; P = .01). Improved HbA1c with the use of insulin lispro in external insulin pumps has been observed in other studies as well (Bode, Steed, & Davidson, 1997; Campbell et al., 1998; Lins, Johansson, Wredling, & Adamson, 1998; Renner et al., 1999). While the magnitude of the change in HbA1c varies across the studies, the differences may be related to number of subjects, duration of the study, intensity of glucose control, or the timing of the regular human insulin bolus does relative to meals. The rapid disappearance of insulin lispro also reduces the likelihood of hypoglycemia 3 to 4 h after a meal. In several long-term studies using multiple daily insulin injections with either NPH or ultralente as the basal insulin, insulin lispro reduced the number of hypoglycemic episodes compared to regular human insulin (Anderson et al., 1997). The improved HbA1c observed with insulin lispro in the present
299
study was not associated with an increase in the frequency of hypoglycemia. In the present study, no apparent difference was seen in the type or severity of adverse events between therapy with insulin lispro or buffered regular human insulin. The occurrences of hyperglycemia did not differ between the therapies, including hyperglycemia due to tubing occlusion from any cause. Although diluted insulin was not allowed in the present study, decreasing the insulin concentration through dilution has been suggested to decrease the occurrence of occlusion by insulin aggregation (Hirsch, FarkasHirsch, & McGill, 1992). Another aspect of using insulin lispro in external pumps is a potential risk of hyperglycemia or ketoacidosis during needle loss or tubing occlusion. In a small crossover study, Reichel et al. (1998) observed in seven patients that metabolic decompensation following insulin cessation appeared to occur about 1.5 to 2 h earlier with insulin lispro compared with regular human insulin (after 3 h). A crossover study carried out in 10 patients by Guerci et al. (1999) gave similar results. In contrast, Attia, Jones, Holcombe, and Tamborlane (1998) studied 18 patients in a parallel study design and observed no differences after infusion cessation between insulin lispro and regular human insulin with respect to glucose or beta-hydroxybutyrate concentration. These discrepancies may be due to differences in study design, particularly with regard to the timing between the last bolus of insulin or the time of day the insulin infusions were stopped. In the study by Reichel et al., for example, the insulin infusion was stopped at 10 p.m., 4 h after the last bolus dose of insulin. Four hours after a bolus injection, sufficient insulin depot may remain at the site because the duration of effect of regular human insulin is prolonged with larger doses. Therefore, discontinuing an insulin infusion at times other than during basal infusion may result in different rates of hyperglycemia onset because of the residual action of regular human insulin. Attia et al. (1998) discontinued the insulin infusions at 3 a.m., when the insulin infusion would be more likely to represent a true basal rate. In summary, insulin lispro delivered through an external insulin pump provides a safe, convenient, and a biologically significant advantage over regular human insulin. This advantage is easily seen in reduced HbA1c and postmeal glucose levels. Importantly, these benefits are seen without an increase in the frequency of insulininduced hypoglycemia.
Acknowledgments The authors thank Hunter Heath III for carefully reading the manuscript and for the suggestions for improvement, Vipin Arora for the statistical analysis, Kenneth E. Robertson for the revision of the manuscript, and Michele Skinner for editorial assistance.
300
P. Raskin et al. / Journal of Diabetes and Its Complications 15 (2001) 295–300
References Anderson, J. H. Jr., Brunelle, R. L., Koivisto, V. A., Pfu¨tzner, A., Trautmann, M. E., Vignati, L., & DiMarchi, R. (1997). Reduction of postprandial hyperglycemia and frequency of hypoglycemia in IDDM patients on insulin-analog treatment. Diabetes, 46, 265 – 270. Attia, N., Jones, T. W., Holcombe, J. H., & Tamborlane, W. V. (1998). Comparison of human regular and lispro insulins after interruption of continuous subcutaneous insulin infusion and in the treatment of acutely decompensated IDDM. Diabetes Care, 21, 817 – 821. Bode, B. W., Steed, R. D., & Davidson, P. C. (1997). Continuous subcutaneous insulin infusion (CSII): a study comparing lispro to regular human insulin. Diabetes, 46 (Suppl. 1), 167A. Brunelle, R. L., Llewelyn, J., Anderson, J. H. Jr., Gale, E. A. M., & Koivisto, V. A. (1998). Metaanalysis of the effect of insulin lispro on severe hypoglycemia in patients with type 1 diabetes. Diabetes Care, 21, 1726 – 1731. Campbell, R. K., Sclar, D. A., Robison, L. M., Stowers, J. K., Ferguson, J. C., Trippe, B. S., & Kayne, D. M. (1998). Impact on clinical status and quality of life of switching from regular human insulin to insulin lispro among patients using insulin pumps. Diabetes Educator, 24, 95 – 99. The Diabetes Control and Complications Trial Research Group. (1993). The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin dependent diabetes mellitus. New England Journal of Medicine, 329, 977 – 986. The Diabetes Control and Complications Trial Research Group. (1995). The relationship of glycemic exposure (HbA1c) to the risk of development and progression of retinopathy in the diabetes control and complications trial. Diabetes, 44, 968 – 983. Ebeling, P., Jansson, P.-A., Smith, U., Lalli, C., Bolli, G. B., & Koivisto, V. A. (1997). Strategies toward improved control during insulin lispro therapy in IDDM: importance of basal insulin. Diabetes Care, 20, 1287 – 1289. Guerci, B., Meyer, L., Salle´, A., Charrie´, A., Dousset, B., Ziegler, O., & Drouin, P. (1999). Comparison of metabolic deterioration between insulin analog and regular human insulin after a 5-hour interruption of a continuous subcutaneous insulin infusion in type 1 diabetic patients. Journal of Clinical Endocrinology and Metabolism, 84, 2673 – 2678. Heinemann, L., Heise, T., Wahl, L. C., Trautmann, M. E., Ampudia, J., Starke, A. A., & Berger, M. (1996). Prandial glycaemia after a carbohydrate-rich meal in Type I diabetic patients: using the rapid acting insulin analogue [Lys (28), Pro (B29)] human insulin. Diabetic Medicine, 13, 625 – 629. Hirsch, I. B., Farkas-Hirsch, R., & McGill, J. B. (1992). Catheter obstruction with continuous subcutaneous insulin infusion: effect of insulin concentration. Diabetes Care, 15, 593 – 594.
Howey, D. C., Bowsher, R. R., Brunelle, R. L., & Woodworth, J. R. (1994). [Lys (B28), Pro (B29)] Human insulin: a rapidly absorbed analogue of human insulin. Diabetes, 43, 396 – 402. Koivisto, V. A. (1998). The human insulin analogue insulin lispro. Annals of Medicine, 30, 260 – 266. Lins, P. E., Johansson, U. B., Wredling, R., & Adamson, U. (1998). Blood glucose variability and well-being during CSII using insulin lispro and regular human insulin. Diabetes, 47 (Suppl. 1), 39A (Abstract). Mecklenburg, R. S., Benson, J. W. Jr., Becker, N. M., Brazel, P. L., Fredlund, P. N., Metz, R. J., Nielsen, R. L., & Sannar, C. A. (1982). Clinical use of the insulin infusion pump in 100 patients with Type 1 diabetes. New England Journal of Medicine, 307, 513 – 518. Melki, V., Renard, E., Lassmann-Vague, V., Boivin, S., Guerci, B., Hanaire-Broutin, H., Bringer, J., Belicar, P., Jeandidier, N., Meyer, L., Blin, P., Augendre-Ferrante, B., & Tauber, J.-P. (1998). Improvement of HbA1c and blood glucose stability in IDDM patients treated with lispro insulin analog in external pumps. Diabetes Care, 21, 977 – 982. Pampanelli, S., Torlone, E., Lalli, E., Del Sindaco, P., Ciofetta, M., Lepore M., Bartocci, L., Brunetti, P., & Bolli, G. B. (1995). Improved postprandial metabolic control after subcutaneous injection of a short-acting insulin analog in IDDM of short duration with residual pancreatic-cell function. Diabetes Care, 18, 1452 – 1459. The WHO classification of diabetes. (1991). In: J. C., Pickup, & G., Williams (Eds.) Textbook of diabetes. Oxford, England: Blackwell Scientific Publications, p.39. Ratner, R. E., & Steiner, M. L. (1987). Insulin-pump therapy: effect of phosphate-buffered human insulin. Diabetes Care, 10, 787 – 788. Reichel, A., Rietzsch, H., Kohler, H. J., Pfu¨tzner, A., Gudat, U., & Schulze, J. (1998). Cessation of insulin infusion at night-time during CSII-therapy: comparison of regular human insulin and insulin lispro. Experimental and Clinical Endocrinology & Diabetes, 106, 168 – 172. Renner, R., Pfu¨tzner, A., Trautmann, M., Harzer, O., Sauter, K., & Landgraf, R. (1999). Use of insulin lispro in continuous subcutaneous insulin infusion treatment. Diabetes Care, 22, 784 – 788. Tamborlane, W. V., Sherwin, R. S., Genel, M., & Felig, P. (1979). Reduction to normal of plasma glucose in juvenile diabetes by subcutaneous administration of insulin with a portable infusion pump. New England Journal of Medicine, 300, 573 – 578. Tuominen, J. A., Karonen, S. L., Melamies, L., Bolli, G., & Koivisto, V. A. (1995). Exercise induced hypoglycaemia in IDDM patients treated with short-acting insulin analogue. Diabetologia, 38, 106 – 111. Zinman, B., Tildesley, H., Chiasson, J.-L., Tsui, E., & Strack, T. (1997). Insulin lispro in CSII: results of a double-blind crossover study. Diabetes, 46, 440 – 443.
A comparison of insulin lispro and buffered regular human insulin administered via continuous subcutaneous insulin infusion pump$ Philip Raskina,*, John H. Holcombeb, William V. Tamborlanec, John I. Maloned, Suzanne Strowiga, Jo Ann Ahernc, Francine Laventd a
Department of Internal Medicine, University of Texas Southwestern Medical Center at Dallas, 5323 Harry Hines Boulevard, G5-238, Dallas, TX 75390-8858, USA b Lilly Research Laboratories, Eli Lilly and Company, Lilly Corporate Center, Indianapolis, IN 46285, USA c Section of Pediatric Endocrinology, Yale University School of Medicine, PO Box 208064, 33 Cedar St., New Haven, CT 06520-8064, USA d Department of Pediatrics, College of Medicine, University of South Florida, 4202 E. Fowler Ave., Tampa, FL, USA Received 24 May 2001; accepted 20 June 2001
Abstract This study compared glycemic control achieved with insulin lispro or buffered regular human insulin in patients with Type 1 diabetes treated with continuous subcutaneous insulin infusion (CSII) using an external insulin pump. In this 24-week multicenter, randomized, twoway crossover, open-label trial, 58 patients on CSII with adequate glycemic control received either insulin lispro or buffered regular human insulin for 12 weeks, followed by the alternate treatment for another 12 weeks. Efficacy and safety measures included hemoglobin A1c (HbA1c) at baseline and endpoint, home blood glucose monitoring, hypoglycemia, and frequency of pump catheter occlusion. Patients consumed a standard test meal on three occasions, with determinations of fasting, 1- and 2-h postprandial glucose values. Insulin lispro use was associated with a significantly lower HbA1c than was buffered regular human insulin (7.41 ± 0.97 vs. 7.65 ± 0.85 mmol/l; P = .004). Fasting serum glucose values before the test meal were similar between the two therapies. The 1-h (11.16 ± 4.29 vs. 13.20 ± 4.68 mmol/l; P = .012) and 2-h (9.64 ± 4.10 vs. 12.53 ± 4.64 mmol/l; P = .001) postprandial glucose concentrations were significantly lower during treatment with insulin lispro. No differences between treatments were observed in basal or bolus insulin doses, weight gain, or the incidence and rate of hypoglycemia, hyperglycemia, or pump occlusions. When used in external pumps, insulin lispro provides better glycemic control than buffered regular human insulin with a similar adverse event profile. D 2001 Elsevier Science Inc. All rights reserved. Keywords: Insulin; Lispro; Insulin pump; HbA1c
1. Introduction The goal of treatment for individuals with diabetes should be a normalization of blood glucose levels, such that the hemoglobin A1c (HbA1c) is in the range of individuals without diabetes. The results of the Diabetes Control and Complications Trial (DCCT) clearly show that the microvascular complications, and perhaps even macrovascular complications of diabetes, are related to hyperglycemia (The Diabetes Control and Complications Trial Research Group, 1993). Intensive therapy resulting in
$
Eli Lilly and Company sponsored this research. * Corresponding author. Tel.: +1-214-648-2017; fax: +1-214-6484854.
near-normal blood glucose levels lead to an approximately 50% reduction in the development and progression of microvascular complications. However, this therapy was associated with a threefold increase in the incidences of hypoglycemia compared to standard treatment. Thus, it is imperative to develop treatment strategies that can achieve euglycemia without excessive hypoglycemia. To date, that has been difficult; for example, less than 5% of the intensively treated subjects in the DCCT study were able to achieve HbA1c values in the range of nondiabetic individuals (The Diabetes Control and Complications Trial Research Group, 1993). This goal has been difficult to reach, in part, because we still lack the treatment tools needed to completely normalize blood glucose levels. However, new therapeutics and tools are becoming available that show promise in improving blood glucose control.
1056-8727/01/$ – see front matter D 2001 Elsevier Science Inc. All rights reserved. PII: S 1 0 5 6 - 8 7 2 7 ( 0 1 ) 0 0 1 6 8 - 4
296
P. Raskin et al. / Journal of Diabetes and Its Complications 15 (2001) 295–300
Portable insulin infusion devices are effective insulin delivery systems for managing Type 1 diabetes (Mecklenburg et al., 1982; Tamborlane, Sherwin, Genel, & Felig, 1979), lower HbA1c levels have been achieved when treatment with insulin pumps is compared to multiple daily insulin injections (The Diabetes Control and Complications Trial Research Group, 1995). However, severe hypoglycemia remains common with both methods of insulin administration (The Diabetes Control and Complications Trial Research Group, 1995). Rapidly absorbed insulin analogs, such as insulin lispro, may offer an advantage over regular human insulin for insulin pumps by more effectively lowering postprandial hyperglycemia without increasing the prevalence of hypoglycemia (Zinman, Tildesley, Chiasson, Tsui, & Strack, 1997). Insulin lispro is a rapid-acting analog of human insulin with a quicker onset and a shorter duration of action than regular human insulin (Howey, Bowsher, Brunelle, & Woodworth, 1994; Koivisto, 1998). After subcutaneous injection of insulin lispro, peak activity is observed between 30 and 90 min, compared with regular human insulin, which has a peak effect between 2 and 4 h (Howey et al., 1994). The peak effect of regular human insulin is usually achieved 2 to 6 h after injection and its effects may last as long as 16 h. However, since the peak glycemic response to a mixed meal has been found to be between 26 and 70 min after ingestion, regular human insulin may peak too late to control postprandial blood glucose levels. This pharmacodynamic difference is important because an early peak of insulin secretion is very effective in suppressing hepatic glucose production, the main contributor to the postprandial hyperglycemia observed in patients with Type 1 diabetes (Tuominen, Karonen, Melamies, Bolli, & Koivisto, 1995). In addition, the use of insulin lispro is associated with a lower frequency of hypoglycemia (Anderson et al., 1997) or severe hypoglycemia (Brunelle, Llewelyn, Anderson, Gale, & Koivisto, 1998) compared with regular human insulin. Various unmodified insulin preparations have been used in continuous subcutaneous insulin infusion (CSII) pumps since they became available in the late 1970s. These insulins have included mixed beef – pork, purified pork, and biosynthetic human insulins. Insulins containing phosphate buffer may be better suited for external pump usage, because the incidence of catheter occlusion is reportedly less than with use of neutral insulin (Ratner & Steiner, 1987). Newer infusion catheter tubing, such as Polyfin (MiniMed, Sylmar, CA) may decrease the occlusion rate without regard to the type of insulin used. Velosulin BR (buffered regular human insulin injection, Novo Nordisk, Princeton, NJ) is the only phosphate-buffered insulin approved in the United States for use in external insulin pumps. However, insulin lispro (Humalog, Eli Lilly and Company, Indianapolis, IN) is also phosphate-buffered. Therefore, this study was designed to compare the effects of the two phosphate-buffered insulins, insulin lispro and buffered regular human insulin, on
glycemic control in patients with Type 1 diabetes who were receiving insulin by CSII.
2. Patients and methods 2.1. Subjects Patient demographics are shown in Table 1. Patients were invited to participate if they had Type 1 diabetes according to the World Health Organization classification (Pickup & Williams, 1991), were between the ages of 13 and 60 years, had achieved acceptable compliance with insulin pump therapy and a nutritional regimen, and had been treated with CSII for at least 6 months before entering this study. Patients were excluded from the study if they had total glycated hemoglobin values or an HbA1c more than 2.0 times the upper limit of the normal range, clinically significant renal, hepatic, or cardiac disease, cancer, drug or alcohol abuse, insulin allergy, recurrent severe hypoglycemia, anemia; life expectancy of less than 3 years, lactating, pregnant, or intended to become pregnant, or require dilution of insulin in their pump. All patients gave informed consent according to Good Clinical Practice Guidelines and the Declaration of Helsinki. 2.2. Study design This was a 6-month open-label randomized crossover clinical trial conducted at three sites in the United States. During this study, only MiniMed models 504, 504-S, 506, or 507 were used, along with that manufacturer’s associated tubing catheters (Sof-set, Sof set QR, Polyfin, or Polyfin QR). Each patient continued to use the same model of MiniMed pump throughout this study. At the initial visit, all patients began treatment with buffered regular human insulin (Velosulin BR). During a 2- to 4-week lead-in period, diabetes control was to be maintained or improved according to individual goals by adjusting the basal or bolus insulin doses, or both. Because the stability of diluted Table 1 Patient characteristics at baseline (mean ± S.D.) Treatment sequence Variable N Female Male Age Weight (kg) Duration of diabetes (years) Fasting serum glucose (mmol/l) HbA1c (%)
Insulin lispro; regular insulin
Regular insulin; insulin lispro
28 12 16 40.5 ± 8.7 78.3 ± 17.9 18.8 ± 7.6
30 16 14 37.8 ± 9.7 77.3 ± 16.7 17.4 ± 8.5
8.7 ± 3.2
7.45 ± 2.7
.169
7.9 ± 1.1
7.6 ± 0.8
.234
P value
P. Raskin et al. / Journal of Diabetes and Its Complications 15 (2001) 295–300
insulin lispro had not been investigated at the time of this study, insulin dilution was not allowed. After the lead-in period, the patients were randomized either to continue buffered regular human insulin or to begin insulin lispro. Randomization was such that there would be approximately equal numbers of patients in each treatment sequence at each investigative site. Premeal insulin doses were given immediately before each meal. Both treatments were continued for 12 weeks after which time the patients received the alternate insulin for an additional 12 weeks. The patients were seen every 6 weeks throughout the study. Glycemic goals for this study were preprandial blood glucose values below 6.7 mmol/l ( < 120 mg/dl) and maintenance of 2-h postprandial glucose values below 10 mmol/l ( < 180 mg/dl) without causing hypoglycemia. The dose of insulin was increased to achieve these goals according to the results of home blood glucose monitoring, unless further increase would result in an increased risk of hypoglycemia. The basal and bolus doses were adjusted according to each patient’s metabolic needs. At the end of the lead-in period and at the end of each treatment period, HbA1c and fasting plasma glucose were measured. After these blood samples were obtained, each patient consumed a test meal (Sustacal, Mead Johnson, Evansville, IN) consisting of 360 cal, with 55% of calories from carbohydrate, 21% from fat, and 24% from protein. The patient was given 1– 2 min to consume the test meal. One- and two-hour postprandial serum glucose measurements were obtained with the patient at rest. The appropriate study insulin (insulin lispro or buffered regular human insulin) was administered immediately before the test meals. The HbA1c measurements were performed by a central laboratory (Covance, Indianapolis, IN) using high-performance cation exchange chromatography (reference range: 4.3 –6.1%) at baseline and at the end point of each study period. Serum glucose levels were also measured by Covance, using a Hitachi 747 analyzer. Patients recorded in a diary the insulin doses (bolus and basal doses) used on the day before each clinic visit. In addition to the test meal data, patients were asked to perform home blood glucose monitoring using a meter with memory capability. At the clinic visits, the results of the self-blood glucose monitoring were downloaded and the summary statistics of glucose values obtained during the 2 weeks before each visit were recorded. At each visit, information on hypoglycemic and hyperglycemic events, in addition to other questions related to safety, were recorded. Hypoglycemia was defined as any time a patient believed (or another person observed) that he or she was experiencing at least one sign or symptom that could be associated with hypoglycemia or the patient experienced a measured blood glucose less than 3.0 mmol/l (54 mg/dl). These episodes were reported spontaneously and were recorded by the patients in their study diaries. In addition to information on hypoglycemia, unexplained episodes of hyperglycemia were also recorded. An episode of hyper-
297
glycemia was defined as 2 or more consecutive blood glucose measurements greater than 13.8 mmol/l (250 mg/ dl) occurring within a 2- to 4-h period and that did not respond to insulin boluses or otherwise remained unexplained. For each episode of hyperglycemia, the patient was asked to answer three questions intended to characterize the cause of the hyperglycemia. If the hyperglycemia was thought to be due to occlusion, the patient was asked to identify the source of the interrupted insulin flow through the catheter, such as blood in the tubing, kink in the tubing, visible occlusion, or other causes. 2.3. Statistical methods An intent-to-treat analysis was performed using all the data from all the patients. All comparisons for efficacy variables were performed using two-tailed tests with a nominal significance level of .05. A separate analysis of variance (ANOVA) model was used to examine for carryover, investigator or treatment effects.
3. Results Table 1 contains the patient characteristics. Fifty-eight patients completed the study. Three patients discontinued before randomization (two due to personal reasons; one due to failure to meet protocol entry criteria) and one patient discontinued the study after randomization, for personal reasons. Because no carryover or investigator effects were discovered by ANOVA for any of the efficacy measures, the data from the crossover treatment periods were combined. Mean insulin doses were similar between the two treatments. The mean basal insulin doses were 0.34 and 0.35 U/ kg and the mean bolus insulin doses were 0.28 and 0.30 U/ kg for the insulin lispro and regular human insulin treatment groups, respectively. Fig. 1 shows the results of the test meal. The fasting glucose values on the morning of the test meal did not differ between the treatment groups. The 1-h glucose value was lower during treatment with insulin lispro (insulin lispro: 11.16 ± 4.29 mmol/l vs. regular human insulin: 13.20 ± 4.68 mmol/l; P = .012, two-sample t test). The 2-h glucose value was also significantly lower during therapy with insulin lispro (insulin lispro: 9.64 ± 4.10 mmol/l vs. regular human insulin: 12.53 ± 4.64 mmol/l; P = .001, two-sample t test). Fig. 2 shows the HbA1c values at baseline and at the end of each treatment period. No difference was observed between the therapies at baseline. For the combined treatment groups, use of insulin lispro was associated with significantly lower HbA1c compared with regular human insulin (insulin lispro: 7.41 ± 0.97% vs. regular human insulin: 7.65 ± 0.85%; P = .004, two-tailed t test). At the end of the first treatment period, the change from baseline was greater for the insulin lispro group compared with regular human insulin (insulin lispro: 0.34 ± 0.60
298
P. Raskin et al. / Journal of Diabetes and Its Complications 15 (2001) 295–300
Fig. 1. Mean serum glucose before and after a test meal following a preprandial injection of either insulin lispro or buffered regular human insulin. Closed circles: buffered regular human insulin; open triangles: insulin lispro. Error bars represent S.D.
vs. regular human insulin: 0.06 ± 0.58; P = .028, two-tailed t test) (Fig. 3). Similarly, for the combined treatment groups, the overall change from baseline was greater during therapy with lispro compared with regular human insulin (insulin lispro: 0.34 ± 0.59% vs. regular human insulin: 0.09 ± 0.63%; P = .004). The number of home blood glucose measurements carried out during the 2 weeks before each visit did not differ between the two therapy groups (insulin lispro: 56 ± 23 vs. regular human insulin: 56 ± 19; P = .943). The mean blood glucose level during therapy with insulin lispro and regular human insulin was also similar (8.1 ± 2.0 vs. 8.1 ± 1.6 mmol/l, respectively; P = .792). The overall mean of the maximum and the mean of the minimum glucose values did not differ between the therapies. Patients had 41% of their glucose measurements within the target range, irrespective of therapy. Forty-one percent of the patients’
Fig. 2. Mean HbA1c values at baseline and at the end of each 3-month period for both treatment sequence groups.
Fig. 3. Change in HbA1c from baseline to the end of the first 3-month period in patients treated with insulin lispro or buffered regular human insulin. Error bars represent S.D.
glucose values were above target and 18% of readings were below target. During the entire study, 10 patients experienced a total of 26 episodes of hypoglycemia. For the lead-in period, during which all patients received buffered regular human insulin, seven episodes of hypoglycemia were observed, two episodes required treatment with intravenous glucose. After randomization, hypoglycemic events were infrequent and similar between therapies. With insulin lispro treatment, seven patients reported a total of eight hypoglycemic episodes, and during therapy with regular human insulin, seven patients reported a total of 11 episodes. Hypoglycemia requiring intravenous glucose occurred in two patients (three episodes) and in three patients (three episodes) during treatment with regular human insulin and insulin lispro, respectively. Thirty-eight patients recorded 109 episodes of hyperglycemia, 39 (35.7%) were caused by occlusion, 47 (43.1%) were caused by reasons other than occlusion, and 23 (21.1%) had no identifiable cause. The 39 episodes of hyperglycemia due to occlusion were reported by 20 patients (eight patients using insulin lispro, 16 episodes; 12 patients using buffered regular human insulin, 23 episodes). There were no significant associations between therapies and a specific cause of occlusion, such as kinked tubing, blood in tube, or visible occlusion. None of the episodes of occlusion resulted in an adverse event. There were no discontinuations during the study due to adverse events. One patient receiving buffered regular human insulin was hospitalized for treatment of fever, vomiting, and dehydration, and one patient receiving insulin lispro was hospitalized for treatment of ketosis. There was no significant difference in mean ( ± S.D.) weight at endpoint between patients receiving insulin lispro therapy and patients receiving buffered regular human insulin therapy (79.2 ± 17.1 kg vs. 78.8 ± 17.3 kg, respect-
P. Raskin et al. / Journal of Diabetes and Its Complications 15 (2001) 295–300
ively; P = .780). Although patients in both therapy groups gained approximately 1 kg by the end of the first treatment period, this change in weight from baseline was not statistically significant.
4. Discussion The introduction of analogs of human insulin such as insulin lispro has greatly improved the therapeutic options for the management of Type 1 diabetes. In particular, insulin lispro seems well suited for use in portable insulin infusion devices. The quick onset and short duration of action of insulin lispro allow individuals with Type 1 diabetes to take a premeal insulin bolus immediately before or after eating and helps to reduce postprandial hyperglycemia (Anderson et al., 1997; Ebeling et al., 1997; Heinemann et al., 1996; Koivisto, 1998; Pampanelli et al., 1995). Consistent with the absorption kinetics of the insulins used in this study, the use of insulin lispro resulted in significantly lower 1- and 2-h postprandial glucose levels following the test meal than that achieved with regular human insulin. Although Sustacal is frequently used for test meals and is nutritionally balanced, it does not represent an actual meal as it is in liquid form. However, the postprandial findings remain valid because two insulins were compared in the present study. In this study, therapy with insulin lispro lowered HbA1c compared with regular human insulin (7.4% vs. 7.6%; P = .004). This difference in HbA1c is consistent with the findings of Zinman et al. (1997), who compared insulin lispro with regular human insulin in a 3-month, masked, crossover study involving 30 patients with Type 1 diabetes. In that study, use of insulin lispro was associated with significantly lower HbA1c values (insulin lispro: 7.7%, regular human insulin 8.0%; P = .004). In another crossover trial similar in design to the current study, Melki et al. (1998) also reported a significantly greater reduction in HbA1c from baseline during use of insulin lispro compared with regular human insulin (insulin lispro: 0.62% vs. regular human insulin: 0.09%; P = .01). Improved HbA1c with the use of insulin lispro in external insulin pumps has been observed in other studies as well (Bode, Steed, & Davidson, 1997; Campbell et al., 1998; Lins, Johansson, Wredling, & Adamson, 1998; Renner et al., 1999). While the magnitude of the change in HbA1c varies across the studies, the differences may be related to number of subjects, duration of the study, intensity of glucose control, or the timing of the regular human insulin bolus does relative to meals. The rapid disappearance of insulin lispro also reduces the likelihood of hypoglycemia 3 to 4 h after a meal. In several long-term studies using multiple daily insulin injections with either NPH or ultralente as the basal insulin, insulin lispro reduced the number of hypoglycemic episodes compared to regular human insulin (Anderson et al., 1997). The improved HbA1c observed with insulin lispro in the present
299
study was not associated with an increase in the frequency of hypoglycemia. In the present study, no apparent difference was seen in the type or severity of adverse events between therapy with insulin lispro or buffered regular human insulin. The occurrences of hyperglycemia did not differ between the therapies, including hyperglycemia due to tubing occlusion from any cause. Although diluted insulin was not allowed in the present study, decreasing the insulin concentration through dilution has been suggested to decrease the occurrence of occlusion by insulin aggregation (Hirsch, FarkasHirsch, & McGill, 1992). Another aspect of using insulin lispro in external pumps is a potential risk of hyperglycemia or ketoacidosis during needle loss or tubing occlusion. In a small crossover study, Reichel et al. (1998) observed in seven patients that metabolic decompensation following insulin cessation appeared to occur about 1.5 to 2 h earlier with insulin lispro compared with regular human insulin (after 3 h). A crossover study carried out in 10 patients by Guerci et al. (1999) gave similar results. In contrast, Attia, Jones, Holcombe, and Tamborlane (1998) studied 18 patients in a parallel study design and observed no differences after infusion cessation between insulin lispro and regular human insulin with respect to glucose or beta-hydroxybutyrate concentration. These discrepancies may be due to differences in study design, particularly with regard to the timing between the last bolus of insulin or the time of day the insulin infusions were stopped. In the study by Reichel et al., for example, the insulin infusion was stopped at 10 p.m., 4 h after the last bolus dose of insulin. Four hours after a bolus injection, sufficient insulin depot may remain at the site because the duration of effect of regular human insulin is prolonged with larger doses. Therefore, discontinuing an insulin infusion at times other than during basal infusion may result in different rates of hyperglycemia onset because of the residual action of regular human insulin. Attia et al. (1998) discontinued the insulin infusions at 3 a.m., when the insulin infusion would be more likely to represent a true basal rate. In summary, insulin lispro delivered through an external insulin pump provides a safe, convenient, and a biologically significant advantage over regular human insulin. This advantage is easily seen in reduced HbA1c and postmeal glucose levels. Importantly, these benefits are seen without an increase in the frequency of insulininduced hypoglycemia.
Acknowledgments The authors thank Hunter Heath III for carefully reading the manuscript and for the suggestions for improvement, Vipin Arora for the statistical analysis, Kenneth E. Robertson for the revision of the manuscript, and Michele Skinner for editorial assistance.
300
P. Raskin et al. / Journal of Diabetes and Its Complications 15 (2001) 295–300
References Anderson, J. H. Jr., Brunelle, R. L., Koivisto, V. A., Pfu¨tzner, A., Trautmann, M. E., Vignati, L., & DiMarchi, R. (1997). Reduction of postprandial hyperglycemia and frequency of hypoglycemia in IDDM patients on insulin-analog treatment. Diabetes, 46, 265 – 270. Attia, N., Jones, T. W., Holcombe, J. H., & Tamborlane, W. V. (1998). Comparison of human regular and lispro insulins after interruption of continuous subcutaneous insulin infusion and in the treatment of acutely decompensated IDDM. Diabetes Care, 21, 817 – 821. Bode, B. W., Steed, R. D., & Davidson, P. C. (1997). Continuous subcutaneous insulin infusion (CSII): a study comparing lispro to regular human insulin. Diabetes, 46 (Suppl. 1), 167A. Brunelle, R. L., Llewelyn, J., Anderson, J. H. Jr., Gale, E. A. M., & Koivisto, V. A. (1998). Metaanalysis of the effect of insulin lispro on severe hypoglycemia in patients with type 1 diabetes. Diabetes Care, 21, 1726 – 1731. Campbell, R. K., Sclar, D. A., Robison, L. M., Stowers, J. K., Ferguson, J. C., Trippe, B. S., & Kayne, D. M. (1998). Impact on clinical status and quality of life of switching from regular human insulin to insulin lispro among patients using insulin pumps. Diabetes Educator, 24, 95 – 99. The Diabetes Control and Complications Trial Research Group. (1993). The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin dependent diabetes mellitus. New England Journal of Medicine, 329, 977 – 986. The Diabetes Control and Complications Trial Research Group. (1995). The relationship of glycemic exposure (HbA1c) to the risk of development and progression of retinopathy in the diabetes control and complications trial. Diabetes, 44, 968 – 983. Ebeling, P., Jansson, P.-A., Smith, U., Lalli, C., Bolli, G. B., & Koivisto, V. A. (1997). Strategies toward improved control during insulin lispro therapy in IDDM: importance of basal insulin. Diabetes Care, 20, 1287 – 1289. Guerci, B., Meyer, L., Salle´, A., Charrie´, A., Dousset, B., Ziegler, O., & Drouin, P. (1999). Comparison of metabolic deterioration between insulin analog and regular human insulin after a 5-hour interruption of a continuous subcutaneous insulin infusion in type 1 diabetic patients. Journal of Clinical Endocrinology and Metabolism, 84, 2673 – 2678. Heinemann, L., Heise, T., Wahl, L. C., Trautmann, M. E., Ampudia, J., Starke, A. A., & Berger, M. (1996). Prandial glycaemia after a carbohydrate-rich meal in Type I diabetic patients: using the rapid acting insulin analogue [Lys (28), Pro (B29)] human insulin. Diabetic Medicine, 13, 625 – 629. Hirsch, I. B., Farkas-Hirsch, R., & McGill, J. B. (1992). Catheter obstruction with continuous subcutaneous insulin infusion: effect of insulin concentration. Diabetes Care, 15, 593 – 594.
Howey, D. C., Bowsher, R. R., Brunelle, R. L., & Woodworth, J. R. (1994). [Lys (B28), Pro (B29)] Human insulin: a rapidly absorbed analogue of human insulin. Diabetes, 43, 396 – 402. Koivisto, V. A. (1998). The human insulin analogue insulin lispro. Annals of Medicine, 30, 260 – 266. Lins, P. E., Johansson, U. B., Wredling, R., & Adamson, U. (1998). Blood glucose variability and well-being during CSII using insulin lispro and regular human insulin. Diabetes, 47 (Suppl. 1), 39A (Abstract). Mecklenburg, R. S., Benson, J. W. Jr., Becker, N. M., Brazel, P. L., Fredlund, P. N., Metz, R. J., Nielsen, R. L., & Sannar, C. A. (1982). Clinical use of the insulin infusion pump in 100 patients with Type 1 diabetes. New England Journal of Medicine, 307, 513 – 518. Melki, V., Renard, E., Lassmann-Vague, V., Boivin, S., Guerci, B., Hanaire-Broutin, H., Bringer, J., Belicar, P., Jeandidier, N., Meyer, L., Blin, P., Augendre-Ferrante, B., & Tauber, J.-P. (1998). Improvement of HbA1c and blood glucose stability in IDDM patients treated with lispro insulin analog in external pumps. Diabetes Care, 21, 977 – 982. Pampanelli, S., Torlone, E., Lalli, E., Del Sindaco, P., Ciofetta, M., Lepore M., Bartocci, L., Brunetti, P., & Bolli, G. B. (1995). Improved postprandial metabolic control after subcutaneous injection of a short-acting insulin analog in IDDM of short duration with residual pancreatic-cell function. Diabetes Care, 18, 1452 – 1459. The WHO classification of diabetes. (1991). In: J. C., Pickup, & G., Williams (Eds.) Textbook of diabetes. Oxford, England: Blackwell Scientific Publications, p.39. Ratner, R. E., & Steiner, M. L. (1987). Insulin-pump therapy: effect of phosphate-buffered human insulin. Diabetes Care, 10, 787 – 788. Reichel, A., Rietzsch, H., Kohler, H. J., Pfu¨tzner, A., Gudat, U., & Schulze, J. (1998). Cessation of insulin infusion at night-time during CSII-therapy: comparison of regular human insulin and insulin lispro. Experimental and Clinical Endocrinology & Diabetes, 106, 168 – 172. Renner, R., Pfu¨tzner, A., Trautmann, M., Harzer, O., Sauter, K., & Landgraf, R. (1999). Use of insulin lispro in continuous subcutaneous insulin infusion treatment. Diabetes Care, 22, 784 – 788. Tamborlane, W. V., Sherwin, R. S., Genel, M., & Felig, P. (1979). Reduction to normal of plasma glucose in juvenile diabetes by subcutaneous administration of insulin with a portable infusion pump. New England Journal of Medicine, 300, 573 – 578. Tuominen, J. A., Karonen, S. L., Melamies, L., Bolli, G., & Koivisto, V. A. (1995). Exercise induced hypoglycaemia in IDDM patients treated with short-acting insulin analogue. Diabetologia, 38, 106 – 111. Zinman, B., Tildesley, H., Chiasson, J.-L., Tsui, E., & Strack, T. (1997). Insulin lispro in CSII: results of a double-blind crossover study. Diabetes, 46, 440 – 443.