[Adapted from Canadian Agency for Drugs and Technologies in Health. Rapid-Acting Insulin Analogues for the Treatment of Diabetes Mellitus: Meta-analyses of Clinical Outcomes. Update of CADTH Technology Report No. 87. (Optimal therapy report; vol.2 no.2). Ottawa: The Agency; 2008.]
For more information on this project, visit https://www.cadth.ca/rapid-acting-insulin-analogues-treatment-diabetes-mellitus-meta-analyses-clinical-outcomes
Diabetes comprises a group of common metabolic disorders characterized by hyperglycemia.1 It is a chronic condition in which the body is unable to produce sufficient insulin and/or unable to properly use insulin.1 Insulin, a hormone secreted by pancreatic islet cells in response to increased blood glucose levels, promotes the uptake of glucose into cells where it can be used as a source of energy.1 Diabetes is classified as:2
- Type 1 diabetes — little or no insulin produced by the body (previously classified as insulin-dependent diabetes or juvenile-onset diabetes)
- Type 2 diabetes — the body produces insulin but is unable to use it properly (previously classified as non–insulin-dependent diabetes)
- Gestational diabetes — glucose intolerance with first onset during pregnancy. It is usually a temporary condition
- Other — mainly specific, genetically defined forms of diabetes, or diabetes associated with other disease or drug use.
Without adequate control of blood glucose, vascular and non-vascular complications may ensue. In combination with diet modifications, weight control, and adequate exercise, medications can assist patients in controlling blood glucose levels to reduce their risk of developing long-term diabetic complications.3 Maintaining glycemic levels near normal has been shown to lower the risk of microvascular complications3,4 and macrovascular complications.5-8
Insulin is indicated for all patients with type 1 diabetes, as well as for patients with type 2 diabetes who are unable to achieve adequate glycemic control by other measures (exercise, diet, and/or other antidiabetes drugs). Insulin products can be classified as human insulin, insulin analogues, or animal-sourced insulin, according to the source of insulin.
This report describes the results of a systematic review and meta-analysis conducted as part of the clinical evaluation of the rapid-acting insulin analogues.
The objective of this study was to conduct a systematic review and meta-analysis of the clinical efficacy and safety of the rapid-acting insulin analogues compared with intermediate-acting, unmodified human insulin (HI) and oral antidiabetes drugs (OADs) for the treatment of type 1, type 2, and gestational diabetes.
The research questions addressed were:
- What are the patient-relevant and clinical benefits and harms associated with rapid-acting insulin analogues (i.e., insulin aspart, insulin lispro) compared with short-acting HI or OADs in the treatment of diabetes (type 1, type 2, or gestational)?
- Are there subpopulations of diabetic patients (e.g., pregnant patients, children, elderly people, aboriginal peoples and other ethnic minorities, patients using continuous subcutaneous insulin infusion [CSII]) who may particularly benefit from treatment with rapid-acting insulin analogues in comparison with short-acting HI or OADs?
- What are the benefits and harms of combining rapid-acting insulin analogues with OADs compared with combining short-acting HI with OADs in the treatment of type 2 diabetes from a clinical and patient perspective?
- Compared with short-acting HI, are there differences in the clinical effects of rapid-acting insulin analogues when used at the onset of the disease versus later on for patients with type 2 diabetes?
- Are there clinically significant differences between insulin aspart and insulin lispro in the treatment of diabetes (type 1, type 2, or gestational)?
The literature search strategy and methodology for CADTH’s Technology Report No. 87: Short-acting Insulin Analogues for Diabetes Mellitus: Meta-analysis of Clinical Outcomes and Assessment of Cost-effectiveness9 formed the basis for this current research.
The following bibliographic databases were searched through the Ovid interface: MEDLINE (MEDLINE In-Process & Other Non-Indexed Citations, MEDLINE Daily Update; 1966 to present), Embase (1980 to present), and BIOSIS Previews (1989 to present). The Cochrane Library was searched using the Wiley interface. The search strategy was comprised of both controlled vocabulary, such as the National Library of Medicine’s MeSH (Medical Subject Headings), and keywords. The main search concepts were “diabetes” and “long-acting insulin analogues” (glargine and detemir) and “short-acting insulin analogues” (glulisine, lispro, and aspart). A literature filter was applied to limit retrieval to randomized controlled trials (RCTs).
The search was restricted only by date, from 1990 onward, and by human population. Monthly update searches were established following the initial search in December 2006. Alert results were reviewed from January 2006 until April 2007. To reduce bias, oversight, and inconsistency, two reviewers independently determined whether studies met inclusion criteria. Discrepancies were resolved by consensus or by a third reviewer in the event that consensus was not reached.
A modified Jadad scale was used to evaluate the methodological quality of the RCTs. Outcomes of interest included glycemic control (glycosylated hemoglobin [A1C], fasting plasma glucose), hypoglycemia, body weight, quality of life, patient satisfaction, health care utilization, and long-term complications of diabetes. Two reviewers independently extracted data from each article, and appraised methodological quality. Differences were discussed and resolved by consensus. Where appropriate, quantitative pooling of results through random effects meta-analysis was conducted using Review Manager 4.2 to generate estimates of treatment effect. In the absence of reported carry-over effects, data from crossover trials were combined with those from parallel trials in a single meta-analysis. When a carryover effect was reported in a crossover RCT for a particular outcome, these data were excluded from meta-analysis. A number of subgroup and sensitivity analyses were conducted.
Fifty RCTs were included in the meta-analyses for patients with type 1 diabetes: eight for children (age range from five to 15 years) and 42 for adults (age range from 23 to 48 years). Sample sizes ranged from 10 to 1,008 patients. For patients with type 2 diabetes, 30 RCTs were included (age range from 42 to 68 years), and the number of patients ranged from seven to 876. Three RCTs were available for gestational diabetes (age range from 30 to 35 years), and the number of patients in each trial ranged from 41 to 49. The majority of RCTs were of low methodological quality (Jadad score ≤ 2). Because of incomplete reporting of data, not all outcomes reported in RCTs could be pooled in meta-analyses.
For adult patients with type 1 diabetes, A1C was significantly lower with insulin lispro compared with HI in the combined analysis of multiple daily injection (MDI) and continuous subcutaneous insulin infusion (CSII) users (weighted mean difference [WMD] [95% confidence interval (CI)] = –0.09% [–0.16 to –0.02]). This was also the case in the CSII subgroup (WMD [95% CI] = –0.18% [–0.32 to –0.05]), but not in the MDI subgroup (WMD [95% CI] = –0.06 [–0.14 to 0.02]). Insulin aspart also significantly decreased A1C compared with HI (WMD [95% CI] = –0.13% [–0.20 to –0.07]). A1C was not significantly different between insulin lispro and insulin aspart.
Insulin lispro significantly decreased the relative risk (RR) of severe hypoglycemia compared with HI (RR [95% CI] = 0.80 [0.67 to 0.96]) when MDI and CSII trials were pooled. In subgroup analysis, the RR remained statistically significant in favour of insulin lispro in the MDI group but not in the CSII group. There was no significant difference in the RR of severe hypoglycemia between insulin aspart and HI. The frequency of nocturnal hypoglycemia was significantly decreased with insulin lispro or insulin aspart compared with HI (rate ratio [95% CI] = 0.60 [0.40 to 0.90] and 0.55 [0.43 to 0.70], respectively). There was no difference in the rate ratio of nocturnal hypoglycemia between insulin lispro and insulin aspart. The rate ratio of overall hypoglycemia was not significantly different between insulin aspart or insulin lispro and HI, except for insulin aspart used as CSII (rate ratio [95% CI] = 0.58 [0.40 to 0.85]). A high degree of heterogeneity was observed in meta-analyses of overall hypoglycemia, possibly because of variation in the definitions for this outcome across studies. Insulin lispro demonstrated a significantly higher rate ratio of overall hypoglycemia compared with insulin aspart (rate ratio [95% CI] = 1.49 [1.37 to 1.63]).
Mean two-hour, post-prandial plasma glucose was significantly decreased with insulin lispro compared with HI (WMD [95% CI] = –1.31 mmol/L [–2.35 to –0.27]). However, estimates were available only from three10-12 studies. There was no difference in body weight or diabetic ketoacidosis between the rapid-acting insulin analogues and HI.
Only a few RCTs provided mortality data; no differences between treatments were apparent. Regarding quality of life, limited evidence indicated that insulin lispro was better than HI. Overall, patients seemed to prefer insulin lispro over HI because of its convenience of use.13-17 However, caution is advised in interpreting these results because of the open-label design of most studies and variation in the instruments used.
No RCT data were available for long-term complications of diabetes.
Of five RCTs18-22 that compared insulin lispro with HI in preadolescent patients with type 1 diabetes, four18-21 used MDI and one22 used CSII. There were no significant differences for A1C and severe or nocturnal hypoglycemia. The only significant difference for overall hypoglycemia was observed in a single small RCT22 that evaluated the use of insulin lispro versus HI in preadolescent patients using CSII (18% rate reduction with insulin lispro versus HI).
Regarding quality of life and patient satisfaction, limited evidence21,22 suggests that parents of preadolescent children using either MDI or CSII preferred insulin lispro over HI because of convenience. At this time, there are no data available on complications or mortality associated with long-term use of insulin lispro or insulin aspart in this population.
No studies were identified that compared insulin aspart with HI in preadolescents.
The only RCT23 to investigate the use of insulin lispro compared with HI in adolescents reported no significant differences between treatments for A1C and severe hypoglycemia. In contrast, there were statistically significant rate reductions for nocturnal hypoglycemia and overall hypoglycemia (rate ratios 0.61 [95% CI 0.57 to 0.64] and 0.90 [95% CI 0.99 to 0.93], respectively). No data were available for quality of life, patient satisfaction, diabetes-related complications, or mortality.
In adult patients with type 2 diabetes, there was no significant difference in A1C levels between either of the rapid-acting insulin analogues and HI. There was also no difference in A1C between insulin lispro and OADs; however, patients who had failed previous OAD therapy demonstrated a greater decrease in A1C with biphasic insulin lispro compared with OADs (WMD [95% CI] = –0.85% [–1.18 to –0.53]) and versus metformin (WMD [95% CI] = –0.60% [–1.09 to –0.11]). Insulin apart significantly decreased A1C compared with sulfonylurea (WMD [95% CI] = –0.63% [–1.04 to –0.22]).
The risks of severe and nocturnal hypoglycemia with rapid-acting insulin analogues were not significantly different from HI, although insulin lispro was found to have a lower risk of nocturnal hypoglycemia than sulfonylurea therapy (RR [95% CI] = 0.20 [0.06 to 0.70]). The rates of nocturnal hypoglycemia were significantly decreased with insulin lispro compared with HI (rate ratio [95% CI] = 0.58 [0.48 to 0.70])24-26 or sulfonylureas (RR [95% CI] = 0.20 [0.06 to 0.70]). For overall hypoglycemia, there was no significant difference in risk between insulin lispro or insulin aspart and HI. The rate of overall hypoglycemia was also not significantly different between insulin lispro and HI or insulin lispro and insulin aspart, but significantly favoured insulin aspart compared with HI (rate ratio [95% CI] = 0.72 [0.64 to 0.80]).27,28 Insulin lispro and insulin aspart both increased the rate of overall hypoglycemia compared with sulfonylureas and metformin.
There were no differences in fasting plasma glucose between insulin aspart and HI or insulin lispro and sulfonylureas. There were also no significant differences in body weight or body mass index, cholesterol levels, or all-cause mortality between rapid-acting insulin analogues and HI or sulfonylureas. There was no significant improvement in patient satisfaction with insulin lispro compared with HI, except on the “worry related to diabetes” scale. However, insulin lispro demonstrated a significant improvement in quality of life compared with sulfonylureas. Data for long-term outcomes and mortality were not reported.
Based on the limited evidence comparing rapid-acting insulin analogues and HI for the treatment of pregnant women who had either type 1 diabetes or gestational diabetes, there was no significant difference in A1C level or rates of overall hypoglycemia. There was also limited evidence that showed no significant difference in severe hypoglycemia rates between insulin lispro or insulin aspart compared with HI in pregnant women with type 1 diabetes.
This systematic review and meta-analysis has limitations. Not all studies reported data on all the outcomes of interest. Also, not all RCTs identified could be included in the meta-analyses of all outcomes, thereby reducing power. There was heterogeneity in some meta-analyses, as indicated by high I2 values. Heterogeneity may have resulted from variations in patient population or in study methods. The largest degree of heterogeneity was observed for hypoglycemia outcomes, possibly due to variation in the manner that hypoglycemia was defined by investigators.
The methodological quality of the available evidence was generally poor. For example, allocation concealment was not mentioned in most trials, which may have introduced bias. Lower-quality trials can overestimate benefit; hence, even the relatively modest benefits of the rapid-acting insulin analogues reported here may not be representative of the effects of these agents in clinical practice.
Although quality of life is important in the treatment of diabetes patients, not all trials addressed this issue. Those that did address this issue did not always use the same quality-of-life scale, making comparisons difficult. Most trials excluded patients with diabetes complications, and none of them addressed long-term complications associated with diabetes. Also, most trials lasted for six months or less.
The bulk of the available evidence on rapid-acting insulin analogues for both type 1 and type 2 diabetes consists of short- to medium-term comparisons with HI for A1C and hypoglycemia. Most studies were of poor methodological quality. Based on the available evidence, the benefits of short-acting insulin analogues over HI appears to be marginal at best.
In adult patients with type 1 diabetes, treatment with insulin lispro or insulin aspart significantly reduced A1C levels compared with HI when used by CSII. Insulin aspart but not insulin lispro also reduced A1C levels when used by MDI. The rate of overall or severe hypoglycemia was not significantly different between the two rapid-acting insulin analogues and HI, but nocturnal hypoglycemia was less frequent with insulin lispro or insulin aspart compared with HI.
In children with type 1 diabetes, A1C levels and the rate of hypoglycemia were similar between insulin lispro and HI. A small benefit in reducing the rate of overall and nocturnal hypoglycemia in adolescent patients was shown by a single study.23
In adult patients with type 2 diabetes, there were no differences in A1C levels, risk of hypoglycemia, and quality of life with rapid-acting insulin analogues compared with HI. A slight reduction in the rates of nocturnal and overall hypoglycemia was observed with rapid-acting insulin analogues compared with HI. A marginal improvement of A1C levels and well-being, but no reduction of hypoglycemia, was observed with insulin analogues compared with sulfonylureas.
The limited evidence regarding pregnant women with type 1 diabetes or gestational diabetes showed no difference between rapid-acting insulin analogues and HI for A1C level, overall hypoglycemia, and severe hypoglycemia.
High-quality and long-term studies are required to measure the impact of rapid-acting insulin analogues on quality of life, health care resource utilization, long-term diabetes-related complications, and mortality.
- Diabetes in Canada: facts & figures [Internet]. Ottawa: Public Health Agency of Canada; 2008. [cited 2010 Sep 22]. Available from: http://www.phac-aspc.gc.ca/publicat/2008/ndfs-fnrd-08/ndfs_ff-fnrd_fc-eng.php
- Canadian Diabetes Association 2003 clinical practice guidelines for the prevention and management of diabetes in Canada. Can J Diabetes. 2003;27(Suppl 2):i-S140.
- Diabetes Control and Complications Trial Research Group. Effect of intensive therapy on the microvascular complications of type 1 diabetes mellitus. JAMA. 2002 May 15;287(19):2563-9.
- Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med. 1993;329(14):977-86.
- Selvin E, Marinopoulos S, Berkenblit G, Rami T, Brancati FL, Powe NR, et al. Meta-analysis: glycosylated hemoglobin and cardiovascular disease in diabetes mellitus. Ann Intern Med [Internet]. 2004 Sep 21;141(6):421-31.
- Khaw KT, Wareham N, Bingham S, Luben R, Welch A, Day N. Association of hemoglobin A1c with cardiovascular disease and mortality in adults: the European prospective investigation into cancer in Norfolk. Ann Intern Med [Internet]. 2004 Sep 21;141(6):413-20.
- Nathan DM, Cleary PA, Backlund JY, Genuth SM, Lachin JM, Orchard TJ, et al. Intensive diabetes treatment and cardiovascular disease in patients with type 1 diabetes. N Engl J Med. 2005 Dec 22;353(25):2643-53.
- Nathan DM, Lachin J, Cleary P, Orchard T, Brillon DJ, Backlund JY, et al. Intensive diabetes therapy and carotid intima-media thickness in type 1 diabetes mellitus. N Engl J Med. 2003 Jun 5;348(23):2294-303.
- Banerjee S, Tran K, Li H, Cimon K, Daneman D, Simpson S, et al. Short-acting insulin analogues for diabetes mellitus: meta-analysis of clinical outcomes and assessment of cost-effectiveness [Internet]. Ottawa: Canadian Agency for Drugs and Technologies in Health; 2007. [cited 2010 Sep 22]. (Technology Report; no.87). Available from: http://www.cadth.ca/media/pdf/341A_Insulin_tr_e.pdf
- Anderson JH, Brunelle RL, Koivisto VA, Pfützner A, Trautmann ME, Vignati L, et al. Reduction of postprandial hyperglycemia and frequency of hypoglycemia in IDDM patients on insulin-analog treatment. Diabetes. 1997 Feb;46(2):265-70.
- Caixàs A, Pérez A, Payés A, Otal C, Carreras G, Ordóñez-Llanos J, et al. Effects of a short-acting insulin analog (insulin lispro) versus regular insulin on lipid metabolism in insulin-dependent diabetes mellitus. Metab Clin Exper. 1998;47(4):371-6.
- Raskin P, Holcombe JH, Tamborlane WV, Malone JI, Strowig S, Ahern JA, et al. A comparison of insulin lispro and buffered regular human insulin administered via continuous subcutaneous insulin infusion pump. J Diabetes Complicat. 2001 Nov;15(6):295-300.
- Annuzzi G, Del Prato S, Arcari R, Bellomo DA, Benzi L, Bruttomesso D, et al. Preprandial combination of lispro and NPH insulin improves overall blood glucose control in type 1 diabetic patients: a multicenter randomized crossover trial. Nutr Metab Cardiovasc Dis. 2001 Jun;11(3):168-75.
- Holleman F, Schmitt H, Rottiers R, Rees A, Symanowski S, Anderson JH, et al. Reduced frequency of severe hypoglycemia and coma in well-controlled IDDM patients treated with insulin lispro. Diabetes Care. 1997 Dec;20(12):1827-32.
- Janes JM, Bradley C, Rees A. Preferences for, and improvements in aspects of quality of life (QoL) with, insulin lispro in a multiple injection regimen [abstract]. Diabetologia. 1997;40(Suppl 1):A353.
- Bott U, Ebrahim S, Hirschberger S, Skovlund SE. Effect of the rapid-acting insulin analogue insulin aspart on quality of life and treatment satisfaction in patients with Type 1 diabetes. Diabet Med. 2003 Aug;20(8):626-34.
- Home PD, Lindholm A, Riis A, the European Insulin Aspart Study Group. Insulin aspart vs. human insulin in the management of long-term blood glucose control in type 1 diabetes mellitus: a randomized controlled trial. Diabet Med. 2000 Nov;17(11):762-70.
- Deeb LC, Holcombe JH, Brunelle R, Zalani S, Brink S, Jenner M, et al. Insulin lispro lowers postprandial glucose in prepubertal children with diabetes. Pediatrics [Internet]. 2001 Nov [cited 2010 Sep 22];108(5):1175-9. Available from: http://pediatrics.aappublications.org/cgi/reprint/108/5/1175
- Heller SR, Amiel SA, Mansell P, U.K. Lispro Study Group. Effect of the fast-acting insulin analog lispro on the risk of nocturnal hypoglycemia during intensified insulin therapy. Diabetes Care [Internet]. 1999 Oct [cited 2010 Sep 22];22(10):1607-11. Available from: http://care.diabetesjournals.org/cgi/reprint/22/10/1607
- Ford-Adams ME, Murphy NP, Moore EJ, Edge JA, Ong KL, Watts AP, et al. Insulin lispro: a potential role in preventing nocturnal hypoglycaemia in young children with diabetes mellitus. Diabet Med. 2003 Aug;20(8):656-60.
- Tupola S, Komulainen J, Jääskeläinen J, Sipilä I. Post-prandial insulin lispro vs. human regular insulin in prepubertal children with type 1 diabetes mellitus. Diabet Med. 2001 Aug;18(8):654-8.
- Tubiana-Rufi N, Coutant R, Bloch J, Munz-Licha G, Delcroix C, Montaud-Raguideau N, et al. Special management of insulin lispro in continuous subcutaneous insulin infusion in young diabetic children: a randomized cross-over study. Horm Res. 2004;62(6):265-71.
- Holcombe JH, Zalani S, Arora VK, Mast CJ, Lispro in Adolescents Study Group. Comparison of insulin lispro with regular human insulin for the treatment of type 1 diabetes in adolescents. Clin Ther. 2002 Apr;24(4):629-38.
- Roach P, Trautmann M, Arora V, Sun B, Anderson JH, Jr. Improved postprandial blood glucose control and reduced nocturnal hypoglycemia during treatment with two novel insulin lispro-protamine formulations, insulin lispro mix25 and insulin lispro mix50. Mix50 Study Group. Clin Ther. 1999 Mar;21(3):523-34.
- Ross SA, Zinman B, Campos RV, Strack T, Canadian Lispro Study Group. A comparative study of insulin lispro and human regular insulin in patients with type 2 diabetes mellitus and secondary failure of oral hypoglycemic agents. Clin Invest Med. 2001 Dec;24(6):292-8.
- Anderson JH, Brunelle RL, Keohane P, Koivisto VA, Trautmann ME, Vignati L, et al. Mealtime treatment with insulin analog improves postprandial hyperglycemia and hypoglycemia in patients with non-insulin-dependent diabetes mellitus. Arch Intern Med. 1997 Jun 9;157(11):1249-55.
- Boehm BO, Vaz JA, Brøndsted L Home PD. Long-term efficacy and safety of biphasic insulin aspart in patients with type 2 diabetes. Eur J Intern Med. 2004;15(8):496-502.
- Bretzel RG, Arnolds S, Medding J, Linn T. A direct efficacy and safety comparison of insulin aspart, human soluble insulin, and human premix insulin (70/30) in patients with type 2 diabetes. Diabetes Care [Internet]. 2004 May;27(5):1023-7. Available from: http://care.diabetesjournals.org/cgi/reprint/27/5/1023?maxtoshow=&HITS=10&hits=10&RESULTFORMAT=&author1=bretzel&andorexactfulltext=and&searchid=1&FIRSTINDEX=0&sortspec=relevance&volume=27&firstpage=1023&resourcetype=HWCIT