DIABETES TREATMENTS

Pivotal Role of Timely Basal Replacement After Failure in Sustaining Long-Term Blood Glucose Control at a Target in

GEREMIA B. BOLLI, MD FRANCESCA PORCELLATI, MD, PHD metformin at diagnosis of type 2 diabetes PAOLA LUCIDI, MD, PHD CARMINE G. FANELLI, MD, PHD to rapidly reduce A1C levels to ,7.0%. This “step 1” is generally well accepted, and, in fact, it is common clinical experi- ence that combining medical nutrition ype 2 diabetes over time associates not yet developed vascular complications treatment and physical exercise, along , with the development of vascular and who presumably have a long life- with metformin, reduces A1C to 7.0% T in a significant number of subjects. How- complications (1). The causative role expectancy (6). of long-term elevation of blood glucose is At present, the question is not whether ever, in subjects presenting with severely . – well established, at least for microvascular to intensively treat people with type 2 di- elevated blood glucose (A1C 9 10%), complications, since intervention strate- abetes at onset of the disease to prevent initiation of insulin controls hyperglycemia gies directed at reducing hyperglycemia long-term complications. The question more rapidly than oral hypoglycemic lower onset and/or progression of micro- rather is how to intensively treat type 2 agents (10). angiopathy (1,2). The role of hyperglyce- diabetes over the many years and decades Some patients never satisfactorily fi mia and its treatment in the development of the progression of the disease to con- respond to lifestyle modi cations and of macrovascular complications is less sistently keep A1C levels ,7.0% over the metformin and do not reduce A1C to , well established. In fact, it takes a longer entire cycle of type 2 diabetes. At present, 7.0% within a few months of treat- time to observe a positive effect of better this question is difficult to answer, pri- ment initiation. Other patients, after an blood glucose control (in addition to re- marily because of the lack of evidence of initial successful response to lifestyle fi ducing the multiple risk factors often as- long-term effects of one specificinterven- modi cation + metformin, may lose the sociated with type 2 diabetes such as tion, compared with several other possi- response within months or a few years; hypertension, visceral obesity, and hyper- ble intervention strategies (7) in type 2 therefore, their A1C progressively in- dyslipidemia) on macroangiopathy com- diabetes. In addition, one should always creases beyond the value of 7.0%. This is “ ” pared with microangiopathy (1,3,4). keep in mind that type 2 diabetes is a the time at which an aggressive add on Today’s understanding of the complex re- complex disease characterized by large treatment has to be initiated to rapidly , lationship between hyperglycemia and heterogeneity among individuals and var- control A1C to 7.0% and maintain it “ ” complications in type 2 diabetes predi- iable progression over time that may over a long period of time ( step 2 of cates that only an early and aggressive eventually result in a nearly total loss of the ADA-EASD consensus) (6). To suc- blood glucose–lowering intervention (in pancreatic b-cell function in just a few cessfully implement step 2, several prelim- addition to reduction of the above men- years (8,9). inary considerations should be made. tioned risk factors), successfully sus- First, the initiation of step 2 should be tained over time, will translate into THE AMERICAN DIABETES timely, i.e., the decision should be made benefits on macrovascular complications ASSOCIATION—EUROPEAN as early as possible after failure of step 1 to , several years later (likely 10–15 years) ASSOCIATION FOR THE STUDY maintain A1C 7.0%. This scenario re- (1,3–5). Thus, the present recommenda- OF DIABETES CONSENSUS—The quires consistent monitoring of A1C – tion is to intensively treat people with 2008 consensus of the American Diabetes every 2 3 months, which makes it possi- type 2 diabetes from the clinical onset of Association (ADA)–European Association ble to identify the trend for progressive – the disease, particularly subjects with for the Study of Diabetes (EASD) (7) pro- increases of A1C from values of 6 6.5 to short diabetes duration who likely have poses lifestyle intervention along with 7.0%. Perhaps it would be better to move to step 2 immediately before, not after, ccccccccccccccccccccccccccccccccccccccccccccccccc A1C has already trespassed the threshold From the Department of Internal Medicine, Section of Internal Medicine, Endocrinology and Metabolism, of 7.0%. University of Perugia, Perugia, Italy. Corresponding author: Geremia B. Bolli, [email protected]. OPTIONS IN STEP 2 OF THE This publication is based on the presentations at the 3rd World Congress on Controversies to Consensus in — Diabetes, Obesity and Hypertension (CODHy). The Congress and the publication of this supplement were ADA-EASD CONSENSUS In step made possible in part by unrestricted educational grants from AstraZeneca, Boehringer Ingelheim, Bristol- 2, the simplest option is addition of a Myers Squibb, Daiichi Sankyo, Eli Lilly, Ethicon Endo-Surgery, Generex Biotechnology, F. Hoffmann- second oral agent, either a La Roche, Janssen-Cilag, Johnson & Johnson, Novo Nordisk, Medtronic, and Pfizer. (SU) or a (7). This ad- DOI: 10.2337/dc11-s224 , © 2011 by the American Diabetes Association. Readers may use this article as long as the work is properly dition will likely reduce A1C to 7.0% in cited, the use is educational and not for profit, and the work is not altered. See http://creativecommons.org/ some patients, but the durability is ques- licenses/by-nc-nd/3.0/ for details. tionable. Although combinations of two

S220 DIABETES CARE, VOLUME 34, SUPPLEMENT 2, MAY 2011 care.diabetesjournals.org Bolli and Associates drugs have not been examined, the dura- given in monotherapy, in contrast to basal lowers in a predictable dose-dependent bility of single oral agents is quite limited insulin. However, these benefits of GLP-1 manner both fasting blood glucose and to over time (11). Perhaps the most rational analogs have to be weighed against their some extent postprandial blood glucose, combination would be adding a dipep- relevant side effects, primarily the limited thus reducing the 24-h mean blood glu- tidyl peptidase (DPP)-4 inhibitor, not an tolerability in a significant number of pa- cose concentration and A1C (20). The ad- SU, on top of metformin. This step would tients (gastrointestinal side effects). In ad- ministration of insulin is painless and the reduce the SU-related hypoglycemia risk dition, the elevated cost restricts the titration of the basal preparation is simple, for similar glycemic control (12) and also treatment to a minority of wealthy type with low risk for hypoglycemia. Insulin the risk for apoptosis of pancreatic b-cells 2 diabetic patients (and/or countries). Fi- dose can be tailored to individual needs with SU demonstrated at least in vitro nally, experience with GLP-1 agonists is on a unit-to-unit basis (21). In contrast to (13) and invoked to explain the short du- limited, and durability of such a treatment other drugs, there is no intolerance or ad- rability of SU in vivo (11). Based on this is unknown. In contrast, (basal) insulin verse effects to insulin. Insulin has been in kind of observation, one might anticipate has no limitations in terms of tolerability use for nearly 90 years, which is the lon- that in type 2 diabetes, b-cell function (except for hypoglycemia) and durability gest experience than with any other drug, would benefitlongtermsimplybynot (except possibly requiring the addition of with the notable exceptions of digitalis using an SU and considering alternative prandial insulin) and is a rather cheap and aspirin. options that resulted in similar glycemic treatment. Insulin has relevant extra-glucose control. However, the interesting hypoth- beneficial effects that make insulin unique, esis of superiority of DPP-4 inhibitors BENEFITS OF INITIATION although GLP-1 agonists may have some over SU in terms of durability of b-cell WITH INSULIN (BASAL) OF effects beyond glucose (Table 1). Insulin is function over time in type 2 diabetes is STEP 2 OF THE ADA-EASD the only drug that directly reduces lipo- still to be demonstrated. At present, use CONSENSUS—Insulin (basal) can lysis and free fatty acid concentrations in of DPP-4 inhibitors is limited because of easily be initiated immediately after fail- blood (22), thus reducing lipotoxicity (23) cost. ure of metformin (and lifestyle inter- and improving insulin action and secre- Thesecondoptioninstep2isinject- vention), thus bypassing the add-on tion (24) and the liver fat accumulation able, i.e., either insulin (as basal prepara- treatment of SU, although it is reasonable in non-alcoholic steato-hepatitis (25). In tion) or a -like (GLP)-1 to consider the more traditional use of addition, it improves lipoprotein metabo- agonist (7). Several studies have com- insulin after failure of the combination lism (26), decreases LDL cholesterol and pared head-to-head treatments with basal metformin + SU (or DPP-4 inhibitor). In triglycerides, and increases HDL choles- insulin versus GLP-1 agonists in type 2 fact, although SUs are effective, very terol (27). Insulin promotes nitric oxide diabetes (14–17).However,atpresent, popular, and cheap, and there have been synthesis by endothelium, thus reversing there are not clear recommendations decades of experience with their use, they endothelial dysfunction of type 2 diabetes when it would be more convenient to ini- have negative aspects. SUs stimulate (28), a well-known cardiovascular risk fac- tiate patients with basal insulin compared insulin secretion in a rather glucose- tor (29). No other hypoglycemic drug has with GLP-1 agonists. independent manner, in contrast to in- such a large number of pleiotropic effects cretins (19); thus, they do not specifically in the treatment of type 2 diabetes. OUTCOMES OF CLINICAL target postprandial hyperglycemia, but Insulin, including basal insulin, also STUDIES COMPARING BASAL rather fasting blood glucose. Most impor- has drawbacks—namely the risk for hy- INSULIN AND GLP-1 tantly, they increasehypoglycemicrisk poglycemia and weight gain. However, AGONISTS IN TYPE 2 (7,19), and last, but not least, there is ev- the hypoglycemia risk is associated pri- DIABETES—The demonstration of the idence at least in vitro that they increase marily with prandial and premixed, not elegant physiology of the system apoptosis (13). basal, insulin treatment (30,31). Al- in the regulation of blood glucose ho- Insulin (basal) has a number of bene- though the risk for hypoglycemia is meostasis (18) has prompted research to fits when initiated at step 2 of type 2 higher with basal insulin than with GLP-1 develop GLP-1 agonists for treatment of diabetes treatment. Insulin (basal) easily analogs (14,15), the absolute events are type 2 diabetes, with successful improve- ment of blood glucose primarily in the Table 1—Comparison of positive and negative effects of basal insulin compared postprandial situation, but to some extent with GLP-1 treatment in type 2 diabetes also in the fasting state (14–17). Although GLP-1 agonists and basal Positive Negative insulin treatments both reduce A1C by a similar extent (14–17), there are relevant Insulin Lowers blood glucose Increases body weight and differences in terms of other outcomes. Lowers free fatty acids risk for hypoglycemia Individual GLP-1 agonists have different Increases HDL/decreases LDL cholesterol effects, but as a class, all GLP-1 analogs Decreases hepatic fat content in NASH result in loss of body weight, in contrast to Increases hepatic and muscle insulin sensitivity the increase in body weight with basal in- Improves pancreatic islet b-cell function sulin; they all reduce the risk of hypogly- Reverses endothelial dysfunction cemia compared with basal insulin; and GLP-1 agonists Lower blood glucose similarly to basal insulin Limited tolerability they do not require dose titration, and Reduce body weight Expensive therefore blood glucose monitoring is Virtually no hypoglycemia Short-term experience theoretically not necessary, at least when NASH, non-alcoholic steato-hepatitis. care.diabetesjournals.org DIABETES CARE, VOLUME 34, SUPPLEMENT 2, MAY 2011 S221 Early insulin treatment in type 2 diabetes infrequent, especially with long-acting in- account the frequency of hypoglycemia The recommendation of early use of sulin analogs compared with NPH (32). observed for each level of A1C achieved. insulin in type 2 diabetes emphasized in Weight gain increases between 2 and Recently, the and this article still requires large prospective 4 kg on insulin treatment (30,31,33), pharmacodynamics of therapeutic doses trials in Caucasians to confirm the bene- whereas with GLP-1 agonists, there is a of the basal NPH, glargine, and fits observed in the Chinese population decrease of similar magnitude (14–17). detemir have been examined in the only (10). Nevertheless, today is not too early However, the weight gain on insulin is as- study comparing head-to-head basal in- for an “early initiation” of insulin in type 2 sociated with improved insulin sensitivity, sulins in type 2 diabetes (40). The phar- diabetes. The traditional view of insulin better plasma lipids, and improved glyce- macodynamic effect of glargine on as a late or “last resort” in type 2 diabetes mic control, as mentioned above. Alto- glucose metabolism is superior to that of should be opposed in the year 2011, i.e., gether, these observations account for detemir and NPH. Likewise, glargine bet- in an era where we have learned about the the overall decrease in cardiovascular ter than detemir and NPH suppresses possibility of prevention of complications risk on long-term insulin treatment de- lipolysis and pancreatic a-andb-cell islet by early institution of strict glycemic con- spite some weight gain in interventional function. This latter effect of basal insulin trol and b-cell protection. studies (1,6). likely is the mechanism of the protective The alternative choice is treatment effect of insulin on b-cell function in type with GLP-1 agonists. This choice is ap- 2 diabetes (10). Although all basal insu- pealing because of its efficacy, possible SUPERIORITY OF BASAL lins have a similarly long median duration weight loss, and low risk for hypoglyce- INSULIN COMPARED WITH of action (.31 h), the intersubject vari- mia. However, the limitations of GLP-1 PRANDIAL AND PREMIXED ability of glargine is lower than that of agonists (Table 1) make this a second —Although similar A1C val- and NPH. Interestingly, none of choice of step 2 compared with basal in- ues can be achieved after initiation of the basal insulins stimulate glucose utili- sulin, with the notable exception of obese basal or prandial insulin, basal is more zation, their pharmacodynamic effect subjects with type 2 diabetes. convenient than prandial (30,31) and being explained solely by suppression The present dualism, i.e., choice be- premixed (31,34) insulin because of the of hepatic glucose production. Finally, tween basal insulin versus GLP-1 agonists, lower increased body weight, the lower the nocturnal activity of NPH insulin will likely be replaced in the near future hypoglycemia risk, and the lower need indicates a peak of between 0100 and by combination strategies where GLP-1 for blood glucose monitoring. Those pa- 0400 h, whereas glargine is peakless and agonists would be added to basal insulin. tients who are not at target, or who can no detemir increases its activity overnight In physiology, cooperate with longer control A1C on basal insulin only, slowly after an initial lag phase (40), insulin in glucose homeostasis in the should add prandial insulin to basal (35), thus explaining the protection against fasting and, to a larger extent, postpran- not switch to premixes (36). In fact, al- risk of nocturnal hypoglycemia of the dial situation (18). In fact, in type 2 di- though aggressive titration of premixed two analogs compared with NPH in clin- abetes, addition of a DPP-4 inhibitor or insulin can lower A1C to target (33) or ical studies (32). Thus, the three basal in- GLP-1 agonists improves short-term post- near-to-target (31), this result is associ- sulins exhibit relevant differences not prandial blood glucose (43). Addition of ated with elevated frequency of hypogly- only in type 1 diabetes (32), but in type GLP-1 agonists to basal insulin improves cemia (31,34) due to inappropriate 2 diabetes as well (40). A recent study long-term A1C in subjects who cannot pharmacokinetics and pharmacodynam- (41) has compared head-to-head NPH, reach the target despite optimization of ics, primarily in the interprandial state glargine, and detemir with results at var- basal insulin (44). Although additional of the late-morning and early-night hours iance compared with those described studies are needed to learn more about (37). above (40). However, this study (41) has this potentially rational and synergistic investigated normal nondiabetic subjects combination in subjects with type 2 dia- DIFFERENCES BETWEEN who do not need basal insulin treatment betes with different characteristics and of BASAL INSULINS—The old NPH in their daily life, and therefore the results long duration, it is likely that in the near and the new long-acting insulin analogs remain applicable to normal subjects only future, GLP-1 agonists will be combined glargine and detemir similarly decrease and the data cannot be extrapolated to the early with basal insulin in step 2 of the A1C to target in type 2 diabetes (32). wide spectrum of insulin deficiency and consensus (7). The goals are reducing or However, both analogs are superior to insulin resistance that characterizes sub- even neutralizing the increase in body NPH, because for the same A1C achieved, jects with type 1 diabetes (32) or type 2 weight observed with basal insulin and they reduce the risk of hypoglycemia diabetes (40). lowering more blood glucose, especially compared with NPH (32). On the other in the postprandial state, thus reducing hand, the curvilinear relationship be- CONCLUSIONS—After lifestyle mod- A1C further compared with basal insulin tween A1C and frequency for hypoglyce- ification and metformin fail to sustain A1C only. Notably, these additive effects of mia with basal insulins (38,39) also to ,7.0%, insulin (basal) should be initi- GLP-1 occur in the absence of the greater indicates that for the same frequency of ated in a timely manner and preferred to hypoglycemia risk (45) observed when SU hypoglycemia, A1C is consistently lower the other options in step 2. Metformin or prandial insulin are added to basal in- with both analogs—glargine and detemir should be continued (whenever tolerated sulin (20,30,31). If this were the case, then versus NPH. All of these observations in- and/or not contraindicated) when insulin is incretins, namely GLP-1 agonists, would dicate that any level of A1C achieved initiated, because of the well-demonstrated become the add-on treatment of choice with a given insulin preparation or treat- multiple benefits on reduction of body when basal insulin alone is no longer ment should be analyzed not in absolute, weight and A1C and lower risk for macro- able to reduce A1C to ,7.0%. Addition but rather in relative terms, taking into vascular events (42). of GLP-1 under these circumstances

S222 DIABETES CARE, VOLUME 34, SUPPLEMENT 2, MAY 2011 care.diabetesjournals.org Bolli and Associates would prevent use of SU or rapid-acting a consensus statement of the American glucagon-like peptide 1 (7–36 amide) insulin at mealtime. The main benefitof Diabetes Association and the European in type 2 (non-insulin-dependent) di- such a combination would be reducing Association for the Study of Diabetes. abetic patients. Diabetologia 1993;36: – – hypoglycemic risk whenever targeting an Diabetes Care 2008;31:173 175 741 744 A1C level of ,7.0%. At present, such a 8. Holman RR. Assessing the potential for 19. Szoke E, Gosmanov NR, Sinkin JC, et al. alpha-glucosidase inhibitors in prediabetic Effects of and glyburide on risk increases considerably because intensi- fi states. Diabetes Res Clin Pract 1998;40 glucose counterregulation and recovery cation of treatment is based on SU and/or (Suppl.):S21–S25 from hypoglycemia. Metabolism 2006;55: basal-bolus or premixed insulin regimens 9. U.K. Prospective Diabetes Study Group. 78–83 (45). Thus, a future treatment paradigm of U.K. Prospective Diabetes Study 16. Over- 20. Riddle MC, Rosenstock J, Gerich J; Insulin type 2 diabetes after basal insulin might view of 6 years’ therapy of type II diabetes: Glargine 4002 Study Investigators. The foresee no add-on of SU and use of pran- a progressive disease. Diabetes 1995;44: treat-to-target trial: randomized addition dial insulin on top of basal at a later stage 1249–1258 of glargine or human NPH insulin to oral whentheresponsetoGLP-1islost. 10. Weng J, Li Y, Xu W, et al. Effect of in- therapy of type 2 diabetic patients. Di- tensive insulin therapy on beta-cell func- abetes Care 2003;26:3080–3086 tion and glycaemic control in patients 21. Yki-Järvinen H, Kauppinen-Mäkelin R, Acknowledgments—G.B.B. has received hon- with newly diagnosed type 2 diabetes: Tiikkainen M, et al. or oraria for lecturing and scientificadvisingfrom a multicentre randomised parallel-group NPH combined with metformin in type 2 sanofi-aventis, Eli Lilly, Novo Nordisk, and trial. Lancet 2008;371:1753–1760 diabetes: the LANMET study. Diabetologia MannKind. No other potential conflicts of in- 11. Kahn SE, Haffner SM, Heise MA, et al. 2006;49:442–451 terest relevant to this article were reported. Glycemic durability of , met- 22. Taskinen MR, Sane T, Helve E, Karonen The critical comments of H. Gerstein (De- formin, or glyburide monotherapy. N Engl SL, Nikkilä EA, Yki-Järvinen H. Bedtime partment of Medicine and Population Health J Med 2006;355:2427–2443 insulin for suppression of overnight free- Research Institute, McMaster University and 12. Ferrannini E, Fonseca V, Zinman B, et al. fatty acid, blood glucose, and glucose Hamilton Health Sciences) and J. Rosenstock Fifty-two-week efficacy and safety of production in NIDDM. Diabetes 1989;38: (Dallas Diabetes and Endocrine Center at Med- vs. glimepiride in patients 580–588 ical City) are greatly appreciated. with type 2 diabetes mellitus inadequately 23. McGarry JD, Dobbins RL. Fatty acids, controlled on metformin monotherapy. lipotoxicity and insulin secretion. Dia- Diabetes Obes Metab 2009;11:157–166 betologia 1999;42:128–138 References 13. Maedler K, Carr RD, Bosco D, Zuellig RA, 24.Yki-JärvinenH,EskoN,EeroH,Marja- 1. UK Prospective Diabetes Study (UKPDS) Berney T, Donath MY. Sulfonylurea in- Riitta T. Clinical benefits and mechanisms Group. Intensive blood-glucose control duced beta-cell apoptosis in cultured hu- of a sustained response to intermittent with sulphonylureas or insulin compared man islets. J Clin Endocrinol Metab 2005; insulin therapy in type 2 diabetic patients with conventional treatment and risk of 90:501–506 with secondary drug failure. Am J Med complications in patients with type 2 di- 14. Heine RJ, Van Gaal LF, Johns D, Mihm MJ, 1988;84:185–192 abetes (UKPDS 33). Lancet 1998;352: Widel MH, Brodows RG; GWAA Study 25. Juurinen L, Tiikkainen M, Häkkinen AM, 837–853 Group. versus insulin glargine Hakkarainen A, Yki-Järvinen H. Effects of 2. Patel A, MacMahon S, Chalmers J, et al. in patients with suboptimally controlled insulin therapy on liver fat content and Intensive blood glucose control and vas- type 2 diabetes: a randomized trial. Ann hepatic insulin sensitivity in patients with cular outcomes in patients with type 2 Intern Med 2005;143:559–569 type 2 diabetes. Am J Physiol Endocrinol diabetes. N Engl J Med 2008;358:2560– 15. Bunck MC, Diamant M, Cornér A, et al. Metab 2007;292:E829–E835 2572 One-year treatment with exenatide im- 26. Taskinen MR, Kuusi T, Helve E, Nikkilä 3. Gaede P, Lund-Andersen H, Parving HH, proves beta-cell function, compared with EA, Yki-Järvinen H. Insulin therapy in- Pedersen O. Effect of a multifactorial in- insulin glargine, in metformin-treated duces antiatherogenic changes of serum tervention on mortality in type 2 diabetes. type 2 diabetic patients: a randomized, lipoproteins in noninsulin-dependent di- N Engl J Med 2008;358:580–591 controlled trial. Diabetes Care 2009;32: abetes. Arteriosclerosis 1988;8:168–177 4. Holman RR, Paul SK, Bethel MA, Matthews 762–768 27. Kotronen A, Westerbacka J, Bergholm R, DR, Neil HAW. 10-year follow-up of in- 16. Jendle J, Nauck MA, Matthews DR, et al. Pietiläinen KH, Yki-Järvinen H. Liver fat in tensive glucose control in type 2 diabetes. Weight loss with , a once-daily the metabolic syndrome. J Clin Endo- N Engl J Med 2008;359:1577–1589 human glucagon-like peptide-1 analogue crinol Metab 2007;92:3490–3497 5. Turnbull FM, Abraira C, Anderson RJ, for type 2 diabetes treatment as mono- 28. Vehkavaara S, Yki-Järvinen H. 3.5 years of et al. Intensive glucose control and mac- therapy or added to metformin, is pri- insulin therapy with insulin glargine im- rovascular outcomes in type 2 diabetes. marily as a result of a reduction in fat proves in vivo endothelial function in type Diabetologia 2009;52:2288–2298 tissue. Diabetes Obes Metab 2009;11: 2 diabetes. Arterioscler Thromb Vasc Biol 6. Skyler JS, Bergenstal R, Bonow RO, et al. 1163–1172 2004;24:325–330 Intensive glycemic control and the pre- 17. Nauck MA, Ratner RE, Kapitza C, Berria 29. Perticone F, Ceravolo R, Pujia A, et al. vention of cardiovascular events: im- R, Boldrin M, Balena R. Treatment with Prognostic significance of endothelial dys- plications of the ACCORD, ADVANCE, the human once-weekly glucagon-like function in hypertensive patients. Circula- and VA diabetes trials: a position state- peptide-1 analog in combi- tion 2001;104:191–196 ment of the American Diabetes Associa- nation with metformin improves glycemic 30. Bretzel RG, Nuber U, Landgraf W, Owens tion and a scientific statement of the control and lowers body weight in pa- DR, Bradley C, Linn T. Once-daily basal American College of Cardiology Founda- tients with type 2 diabetes inadequately insulin glargine versus thrice-daily pran- tion and the American Heart Association. controlled with metformin alone: a double- dial in people with type 2 Circulation 2009;119:351–357 blind placebo-controlled study. Diabetes diabetes on oral hypoglycaemic agents 7. Nathan DM, Buse JB, Davidson MB, et al. Care 2009;32:1237–1243 (APOLLO): an open randomised con- Management of hyperglycemia in type 2 18. Nauck MA, Kleine N, Orskov C, Holst JJ, trolled trial. Lancet 2008;371:1073–1084 diabetes: a consensus algorithm for the Willms B, Creutzfeldt W. Normalization 31. Holman RR, Farmer AJ, Davies MJ, et al. initiation and adjustment of therapy: of fasting hyperglycaemia by exogenous Three-year efficacy of complex insulin care.diabetesjournals.org DIABETES CARE, VOLUME 34, SUPPLEMENT 2, MAY 2011 S223 Early insulin treatment in type 2 diabetes

regimens in type 2 diabetes. N Engl J Med insulin regimen in individuals with type 2 diabetes: a randomized, cross-over study. 2009;361:1736–1747 diabetes: results of the GINGER study. Diabetes Care. In press 32. Owens DR, Bolli GB. Beyond the era of Diabetes Obes Metab 2010;12:115–123 41. Sørensen LP, Brock B, Mengel A, et al. NPH insulin: long-acting insulin analogs: 37. Luzio S, Dunseath G, Peter R, Pauvaday V, Similarity of pharmacodynamic effects chemistry, comparative pharmacology, Owens DR. Comparison of the pharmaco- of a single injection of insulin glargine, and clinical application. Diabetes Technol kinetics and pharmacodynamics of biphasic insulin detemir and NPH insulin on glu- Ther 2008;10:333–349 and insulin glargine in people cose metabolism assessed by 24-h eugly- 33. Rosenstock J, Davies M, Home PD, Larsen with type 2 diabetes. Diabetologia 2006;49: caemic clamp studies in healthy humans. J, Koenen C, Schernthaner G. A rando- 1163–1168 Diabet Med 2010;27:830–837 mised, 52-week, treat-to-target trial com- 38. Hermansen K, Davies M, Derezinski T, 42. Kooy A, de Jager J, Lehert P, et al. Long- paring insulin detemir with insulin glargine Martinez Ravn G, Clauson P, Home P. A term effects of metformin on metabolism when administered as add-on to glucose- 26-week, randomized, parallel, treat-to- and microvascular and macrovascular lowering drugs in insulin-naive people target trial comparing insulin detemir disease in patients with type 2 diabetes with type 2 diabetes. Diabetologia 2008;51: with NPH insulin as add-on therapy to mellitus. Arch Intern Med 2009;169:616– 408–416 oral glucose-lowering drugs in insulin- 625 34. Raskin P, Allen E, Hollander P, et al. Ini- naive people with type 2 diabetes. Di- 43. Arnolds S, Dellweg S, Clair J, et al. Further tiating insulin therapy in type 2 diabetes: abetes Care 2006;29:1269–1274 improvement in postprandial glucose a comparison of biphasic and basal in- 39. Mullins P, Sharplin P, Yki-Jarvinen H, control with addition of exenatide or sulin analogs. Diabetes Care 2005;28: Riddle MC, Haring HU. Negative binomial to combination therapy with 260–265 meta-regression analysis of combined gly- insulin glargine and metformin: a proof- 35. Lankisch MR, Ferlinz KC, Leahy JL, cosylated hemoglobin and hypoglycemia of-concept study. Diabetes Care 2010;33: Scherbaum WA; Orals Plus Apidra and outcomes across eleven phase III and IV 1509–1515 LANTUS (OPAL) study group. Introducing studies of insulin glargine compared with 44. Buse JB, Bergenstal RM, Glass LC, asimplified approach to insulin therapy in neutral protamine Hagedorn insulin in et al. Use of twice-daily exenatide in type 2 diabetes: a comparison of two single- type 1 and type 2 diabetes mellitus. Clin basal insulin-treated patients with type dose regimens of plus in- Ther 2007;29:1607–1619 2 diabetes: a randomized, controlled trial. sulin glargine and oral antidiabetic drugs. 40.LucidiP,PorcellatiF,RossettiP,etal. Ann Intern Med 2011;154:103–112 Diabetes Obes Metab 2008;10:1178–1185 Pharmacokinetics and pharmacodynamics 45. Gerstein HC, Miller ME, Byington RP, 36. Fritsche A, Larbig M, Owens D, Häring of therapeutic dose of basal insulins NPH, et al. Effects of intensive glucose lowering H-U; GINGER study group. Comparison glargine and detemir after one week of in type 2 diabetes. N Engl J Med 2008; between a basal-bolus and a premixed daily administration at bedtime in type 2 358:2545–2559

S224 DIABETES CARE, VOLUME 34, SUPPLEMENT 2, MAY 2011 care.diabetesjournals.org