RESEARCH
Review
Management of Type 2 Diabetes: Oral Agents,
Insulin, and Injectables
VIVIAN A. FONSECA, MD; KARMEEN D. KULKARNI, MS, RD
ABSTRACT
This article highlights the use of antidiabetic agents,
including insulin. Medical nutrition therapy (MNT) and
physical activity are the cornerstones of management of
type 2 diabetes. The progressive nature of type 2 diabetes
requires use of one or more oral agents and eventually
insulin, along with MNT and physical activity. The Amer-
ican Association of Clinical Endocrinologists and the Eu-
ropean Association for the Study of Diabetes have recom-
mended a lower hemoglobin A1c target of ⬍6.5%, and
many health care providers strive to achieve normaliza-
tion of blood glucose. Achievement of these goals often
prompts providers to consider combination therapy to
target different pathogenic mechanisms and manage
both fasting and postprandial blood glucose levels. Main-
tenance of glycemic control over the lifespan of a patient
with diabetes is overwhelmingly likely to require combi-
nation therapy with oral diabetes medications. The UK
Prospective Diabetes Study reported that ⬍50% of pa-
tients maintained glycemic control on MNT or mono-
therapy with oral agents at 3 years, and that number
dropped to ⬍25% at 9 years. Newer agents and insulins
have become available since the UK Prospective Diabetes
Study was completed and have enhanced our armamen-
tarium of therapeutics for treatment of diabetes.
J Am Diet Assoc. 2008;108:S29-S33.
T
his review describes how antidiabetic agents and
insulin are used with medical nutrition therapy
(MNT) for management of type 2 diabetes. The cor-
nerstones of therapy for type 2 diabetes are MNT and
V. A. Fonseca is professor of Medicine and Pharmacol-
ogy, Tullis Tulane Alumni Chair in Diabetes, and chief
of Section of Endocrinology, Tulane University Medical
Center, New Orleans, LA. K. D. Kulkarni is a certified
diabetes educator and director of Scientific Affairs, In-
tensive Diabetes Management, Abbott Diabetes Care,
Salt Lake City, UT.
STATEMENT OF CONFLICT OF INTEREST: See
page S33.
Address correspondence to: Vivian A. Fonseca, MD,
Tulane University Medical Center, SL-53, 1430 Tulane
Avenue, New Orleans, LA 70112-2699. E-mail: vfonseca@
tulane.edu
Manuscript accepted: January 27, 2008.
Copyright © 2008 by the American Dietetic
Association.
0002-8223/08/10804-1013$34.00/0
doi: 10.1016/j.jada.2008.01.047
© 2008 by the American Dietetic Association
physical activity, as well as weight management for over-
weight patients (1). The American Diabetes Association
recommends that people with diabetes receive individu-
alized MNT provided by a knowledgeable registered die-
titian (RD) as needed to achieve treatment goals (2).
When fasting plasma glucose remains ⬎126 mg/dL
(⬎6.993 mmol/L) or hemoglobin A1c (HbA1c) is ⬎7% after
lifestyle interventions have been implemented for at least
6 weeks, pharmacological intervention should be initi-
ated (2). The American Diabetes Association glycemic
targets were established to minimize risk of microvascu-
lar complications, specifically retinopathy. The American
Diabetes Association target for HbA1c is ⬍7% in general
and as near to normal (6%) as possible in selected indi-
viduals, provided they do not have substantial hypogly-
cemia. It is not clear whether these targets will also
provide protection from macrovascular disease. The
American Association of Clinical Endocrinologists and
the European Association for the Study of Diabetes have
recommended a lower HbA1c target of ⬍6.5%, and many
health care professionals strive to achieve normalization
of blood glucose. This effort often prompts consideration
of combination therapies that target different pathogenic
mechanisms and manage both fasting and postprandial
blood glucose levels (2).
In the UK Prospective Diabetes Study, patients with
type 2 diabetes received intensive therapy with a sulfo-
nylurea or insulin during 10 years and were compared
with patients in a control group receiving dietary therapy
alone. Patients who received intensive therapy had a 25%
risk reduction in microvascular end points compared with
those who received dietary therapy (3). In addition, obese
patients were randomized to these therapies and met-
formin. The latter was associated with a considerable
reduction in myocardial infarction and death.
Recently, several new antidiabetic agents and insulin
preparations have expanded therapeutic choices for gly-
cemic control, more effectively enabling patients to reach
blood glucose targets (1). These products also allow a
choice of effective combinations of therapeutic agents.
Insulin remains an important option among these
choices.
In 2006, the American Diabetes Association and the
European Association for the Study of Diabetes published
a consensus algorithm for managing type 2 diabetes (Fig-
ure 1) (1). This algorithm recommends initiation of met-
formin therapy at diagnosis, along with lifestyle interven-
tion that includes MNT. The algorithm calls for the
addition of another oral agent or insulin if the HbA1c goal
of ⬍7% is not met or maintained (1). Thus, combination
therapy is likely to be needed in most cases. Insulin is
Supplement to the Journal of the AMERICAN DIETETIC ASSOCIATION S29
Figure 1. American Diabetes Association/European Association for the Study of Diabetes consensus algorithm for the metabolic management of
type 2 diabetes. Reinforce lifestyle intervention at every visit. *Check HbA1c every 3 months until ⬍7% and then at least every 6 months. ⫹Although
three oral agents can be used, initiation and intensification of insulin therapy is preferred based on effectiveness and expense. From Nathan and
colleagues (1). Copyright © 2006 American Diabetes Association. From Diabetes Care®, Vol. 29, 2006; 1963-1972. Modified with permission from
The American Diabetes Association.
included as an option at this stage of the algorithm be-
cause it is most likely to result in attainment of the
HbA1c goal. However, because of continuing clinician and
patient resistance to early insulin use, most patients are
treated with combinations of two or three oral agents
before starting insulin (1).
MNT and physical activity enhance the actions of dia-
betes medications in the treatment regimen. We will ex-
plain when and why available medications are recom-
mended (4).
ORAL AGENTS FOR TYPE 2 DIABETES
Insulin Secretagogues and Sensitizers
Secretagogues. Secretagogues stimulate the pancreas to
secrete insulin, which then reduces hepatic glucose pro-
duction and improves glucose uptake by muscles. There
are two classes of secretagogues: sulfonylureas and
nonsulfonylureas.
Sulfonylureas These drugs reduce fasting and postprandial
glucose levels and have been shown to lower HbA1c levels
by 1 to 2 percentage points. They are metabolized by the
liver and cleared by the kidney, except for glimepiride,
which is used cautiously in patients with impaired renal
function. In the elderly, sulfonylureas are used at low
doses because some patients can have a decreased glo-
merular filtration rate, even with a normal serum creat-
inine concentration (3,5). RDs should inform patients
who use a sulfonylurea that missed meals or snacks could
cause hypoglycemia. RDs should also inform patients
that sulfonylureas can cause an increase in appetite and
possible weight gain, emphasize the importance of MNT
and its effectiveness for weight management, and discuss
appropriate weight-management techniques (4).
S30 April 2008 Suppl 1 Volume 108 Number 4
Nonsulfonylurea secretagogues These agents stimulate rapid
insulin production by the pancreas and have been shown to
reduce postprandial blood glucose and to reduce HbA1c
levels by 0.5 to 2 percentage points. Compared with sulfo-
nylureas, they have a quicker onset and a shorter duration
of action (4 to 6 hours). They also carry a reduced risk of
hypoglycemia and cause less weight gain.
The nonsulfonylurea agents are potential options for
patients with erratic meal schedules and those concerned
about weight gain. RDs should caution patients who use
these agents that they should be taken only before meals
or large snacks that contain substantial amounts of car-
bohydrate; taking them before low-carbohydrate meals or
when a meal is missed could result in hypoglycemia.
The two available nonsulfonylurea secretagogues are
repaglinide and nateglinide. They are cleared hepatically
and may be used in patients with renal impairment (4).
They are metabolized by the liver and so should be used
with caution in people who have impaired hepatic func-
tion (5).
Sensitizers. These agents, which enhance insulin action,
work through a variety of mechanisms. They can inhibit
gluconeogenesis and glycogenolysis, inhibit hepatic glu-
cose absorption, or increase glucose uptake in fat and
muscle. There are three categories in this class: bigua-
nides (metformin), thiazolidinediones, and ␣-glucosidase
inhibitors (4,5).
Biguanides Biguanides inhibit hepatic gluconeogenesis
and, to a lesser extent, glycogenolysis. Insulin sensitivity
is also enhanced. Metformin, the only available bigua-
nide, is typically given in two equal doses daily, before
breakfast and supper. It can also be given three times
daily with meals, and the slow-release formulation can be
given once daily. Metformin has been shown to lower
fasting and postprandial blood glucose levels and to lower
HbA1c by 1 to 2 percentage points.
RDs should inform patients that metformin does not
cause hypoglycemia when it is used as monotherapy and
that it can cause weight loss. Other side effects can in-
clude abdominal cramping, nausea, diarrhea, and a me-
tallic taste. These can be minimized by starting with a
small dose and titrating gradually. RDs also should ad-
vise patients to limit foods such as cauliflower, cabbage,
broccoli, lentils, and legumes, which could aggravate
these gastrointestinal side effects (4).
Lactic acidosis is a rare but potentially fatal adverse
effect of metformin. The risk of lactic acidosis is increased
if baseline renal function is abnormal or if there is any
other condition affecting the kidneys, such as dehydra-
tion, major surgery, or chronic heart failure (5).
Thiazolidinediones These agents enhance insulin sensitiv-
ity by increasing the efficiency of the glucose transport-
ers. They are usually taken once or twice a day. They
have been shown to lower HbA1c values by 1 to 2 per-
centage points and to reduce fasting and postprandial
blood glucose levels. They do not cause hypoglycemia
when used as monotherapy (4,5). The two available thia-
zolidinediones (TZDs) are rosiglitazone and pioglitazone.
TZDs can have a positive effect on lipids by decreasing
triglycerides, increasing high-density lipoprotein choles-
terol, and increasing the particle size of low-density li-
poprotein cholesterol (4,5).
Rosiglitazone’s effect on cardiovascular morbidity and
mortality has not been determined. A meta-analysis by
Nissen and Wolski of trials comparing rosiglitazone with
placebo or active comparators assessed the effect of this
agent on cardiovascular outcomes (6). Rosiglitazone was
associated with a substantial increase in risk of myocar-
dial infarction and a borderline substantial increased risk
of death from cardiovascular causes (6). The US Food and
Drug Administration now requires rosiglitazone’s pre-
scribing information to include a boxed warning provid-
ing information about myocardial infarction and isch-
emia. The prescribing information also includes warnings
about the use of rosiglitazone with nitrates, coadminis-
tration of rosiglitazone and insulin, and use of rosiglita-
zone in patients experiencing an acute coronary event.
Pioglitazone does not appear to carry the same cardio-
vascular risk and may have a protective effect of reducing
triglycerides and slightly increasing high-density lipopro-
tein cholesterol.
The main side effects of TZDs are weight gain and mild
edema. RDs should provide patients who use TZDs with
information about reducing caloric consumption and re-
ducing sodium to reduce fluid retention (4).
␣-Glucosidase inhibitors ␣-Glucosidase inhibitors block the
enzyme ␣-glucosidase in the brush borders of the small
intestines, which delays absorption of carbohydrates.
They lower HbA1c values by 0.5 to 1 percentage point.
The three ␣-glucosidase inhibitors are acarbose, miglitol,
and voglibose (not approved for use in the United States).
␣-Glucosidase inhibitors are taken before carbohy-
drate-containing meals; they should not be taken when
meals are missed. ␣-Glucosidase inhibitors should not be
used by people who have severe renal or hepatic impair-
ment or any gastrointestinal disease. Their side effects
April 2008 ● Supplement to the Journal of the AMERICAN DIETETIC ASSOCIATION
are bloating, abdominal cramps, flatulence, and diarrhea
(7). RDs counseling patients who use ␣-glucosidase inhib-
itors should discuss the importance of taking these med-
ications before carbohydrate-containing meals only. They
should also review the gastrointestinal side effects and
how to minimize them by reducing intake of foods that
could cause abdominal bloating and flatulence, such as
cauliflower, broccoli, cabbage, legumes, and lentils (4).
Dipeptidyl Peptidase IV Inhibitors
These agents inhibit dipeptidyl peptidase IV, the enzyme
that degrades endogenously secreted incretins, including
glucagon-like peptide 1 and glucose-dependent insulino-
tropic polypeptide. Increased levels of the incretin hor-
mones result in increased insulin secretion and suppres-
sion of glucagon secretion. Sitagliptin is the first
dipeptidyl peptidase IV inhibitor approved for use as
monotherapy or in combination with a TZD or metformin.
The daily dose of sitagliptin is 100 mg taken with or
without a meal. Sitagliptin’s main effect is on postpran-
dial blood glucose levels, and it has been shown to reduce
HbA1c values by 0.7 to 1.4 percentage points.
The most common side effects are headaches and na-
sopharyngitis. It is metabolized in the liver but excreted
in the urine. In patients with renal insufficiency, the dose
should be decreased by 50% to 75%.
Sitagliptin appears to be weight-neutral. RDs can focus
on helping patients make healthful food choices without
focusing on its effects on appetite. Sitagliptin does not
cause hypoglycemia when used with an insulin sensitizer
or as monotherapy. Therefore, increased caloric intake
from potential overtreatment of hypoglycemia is not an
MNT concern (5).
Combination Medications
Several products that combine two classes of drugs are
available and can benefit patients by reducing the num-
ber of pills that must be purchased and taken each day.
Available combinations include:
● pioglitazone and metformin in doses of 15/500 mg and
15/850 mg;
● rosiglitazone and glimepiride in doses of 4/1 mg, 4/2 mg,
and 4/4 mg;
● rosiglitazone and metformin in doses of 1/500 mg, 2/500
mg, 4/500 mg, 2/1,000 mg, and 4/1,000 mg;
● glyburide and metformin in doses of 1.25/250 mg, 2.5/
500 mg, and 5/500 mg; and
● glipizide and metformin in doses of 2.5/250 mg, 2.5/500
mg, and 5/500 mg (8).
Progressing with Therapy
Too often, patients are started on a low dose of an insulin
sensitizer or secretagogue but are not followed diligently
for dose titration or additional therapies as needed to
improve glycemic control. This clinical inertia can result
in patients maintaining HbA1c values higher than target
for months or years. Health care professionals and pa-
tients should view the initial medication prescription as a
first step in an ongoing process of active management and
follow-up to achieve glycemic goals (1). If these measures
S31
do not achieve or maintain glycemic control, insulin is the
next step, usually as an adjunct to oral agents (9).
INSULIN
A physiological approach to insulin therapy consists of
long-acting basal insulin given once daily and rapid-act-
ing bolus insulin given preprandially to cover mealtime
glucose fluctuations. However, some patients may find
starting such an intensive regimen daunting. Therefore,
short- and long-acting insulins are sometimes mixed (in-
cluding premixed commercial formulations) and used
twice daily. Another option is to add a single injection of
basal insulin to the oral agent regimen and adjust slowly
as needed to include prandial insulin.
Very few long-term data are available to help deter-
mine the best initial insulin regimen. A few short-term
trials have compared basal insulin regimens with either
prandial or mixed insulins, with varying results (10,11).
Recently, Holman and colleagues reported on the interim
1-year results of a trial comparing the addition of bipha-
sic, prandial, or basal insulin to oral agent therapy (12).
Unfortunately, the titration scheme used was possibly
suboptimal because most patients did not achieve the
stated glycemic goal. HbA1c improvements were best
with prandial insulin aspart given three times daily and
worst with basal insulin detemir given usually once or
twice daily, with statistically significant differences
(P⬍0.001). However, rates of hypoglycemia and degree of
weight gain were greatest with prandial insulin and least
with basal insulin. These side effects of insulin therapy
are important for RDs to explain when counseling indi-
viduals with type 2 diabetes. The regimen with biphasic
aspart 70/30 insulin twice daily achieved glycemic con-
trol, hypoglycemia rates, and weight gain that were in-
termediate between the other two regimens. Thus, it is
impossible to identify the single-best regimen to achieve
glycemic control while minimizing hypoglycemia and
weight gain.
This is particularly true given that patient preferences
were not considered. We know from clinical experience
that patients are most likely to choose a regimen with
fewer daily injections and possibly the least risk of hypo-
glycemia and weight gain. For people using a premixed
insulin, RDs should advise eating three meals daily and
keeping the carbohydrate content of meals as consistent
as possible to avoid hypoglycemia.
Because of the progressive nature of type 2 diabetes,
most patients eventually will require insulin (3). To min-
imize the number of injections, insulin is often initiated
with a single injection of basal insulin added to the oral
agent regimen. Many studies have demonstrated the ad-
vantages of administering this injection at night, which is
convenient and can lead to improved fasting blood glucose
levels in the morning, possibly allowing oral agents to
work more effectively throughout the day. For patients
with elevated fasting glucose values, this is an appropri-
ate initial approach.
In the Treat-to-Target study, (13) long-acting insulin
glargine was compared with neutral protamine Hagedorn
(NPH). Both insulins were given at night starting with 10
U, with an aggressive dose titration to achieve a fasting
glucose level ⬍100 mg/dL (⬍5.55 mmol/L). Both insulins
resulted in a substantial decline in fasting glucose and
S32 April 2008 Suppl 1 Volume 108 Number 4
Treat to Target and Hypoglycemia
Combination Oral Agents+Glargine vs NPH at hs
250
196 Mean FPGa During Study (both treatment groups)
200
177
*Mean FPG (mg/dL)
158
150
141 136
150 133 127
123 121 119 118 117 117 118
100
50
750 686 718 698 737 659 676 667 713 661 694 563 570 464 266 =n
0
0 1 2 3 4 5 6 7 8 10 12 15 18 21 24
Weeks in Study
Figure 2. Fasting blood glucose results in the Treat-to-Target study with
both insulin glargine and neutral protamine Hagedorn (NPH). aFPG⫽fasting
plasma glucose. From Riddle and colleagues (13). Copyright © 2003 Ameri-
can Diabetes Association. From Diabetes Care®, Vol. 26, 2003; 3080-
3086. Modified with permission from The American Diabetes Association.
HbA1c, with no difference between groups (Figure 2).
Although the fasting glucose target was not achieved in
many patients, ⬃60% of patients in both groups achieved
target HbA1c concentrations of ⬍7% within the 24-week
trial period. However, use of glargine was associated with
substantially fewer episodes of hypoglycemia, including
nocturnal hypoglycemia, than NPH. Similar results have
been obtained with detemir insulin added as basal insu-
lin for patients with type 2 diabetes suboptimally con-
trolled on oral agents (14). The Treat-to-Target algorithm
has been successfully applied in clinical practice to help
patients reach their glycemic goals (15,16). It is possible
to teach patients to make adjustments in insulin dose
based on their own fasting blood glucose values.
When patients are unable to reach HbA1c goals despite
adequate titration of basal insulin, either postprandial
hyperglycemia or a lack of regimen adherence is usually
to blame. There are no clinical trials to determine the best
approach at this stage. Postprandial hyperglycemia may
be addressed by moving patients to the next step of ther-
apy, namely a physiological (basal-bolus) regimen. It is
usually appropriate at this stage to discontinue sulfonyl-
ureas, but to continue any insulin sensitizers. Continuing
metformin may be particularly beneficial; the combina-
tion of insulin and metformin has been associated with
less weight gain than the use of insulin alone (17). Even
this step may be gradually introduced by adding rapid-
acting insulin initially only with the main meal of the day
and then adding the other two premeal injections as
needed to achieve glucose targets.
NONINSULIN INJECTABLES
Exenatide
Exenatide is a synthetic form of a protein found in the
saliva of the Gila monster that mimics the action of in-
cretins such as glucagon-like peptide 1. It improves gly-
cemic control through several mechanisms, including in-
creased insulin synthesis and secretion in the presence of
elevated glucose concentrations, improvement of first-
phase insulin response, slowed gastric emptying, reduced
glucagon concentrations during glycemic excursions, and
reduced food intake.
Exenatide is injected in the thigh, abdomen, or upper
arm within 60 minutes before the morning or evening
meals. It should not be given after a meal. The initial dose
should be 5 ␮g twice daily, titrated after 1 month to 10 ␮g
twice daily. RDs working with patients who take ex-
enatide should consider that it slows gastric emptying.
Thus, exenatide will cause a feeling of fullness halfway
through meals and can also cause some nausea or feelings
of satiety early in meals (8).
Pramlintide
Pramlintide is a synthetic version of the natural-occur-
ring hormone amylin, and is used in people with type 1 or
type 2 diabetes who also use mealtime insulin. Pramlin-
tide works with insulin to reduce blood glucose excursions
after a meal by slowing down absorption of glucose into
the bloodstream after eating. Pramlintide may allow pa-
tients to reduce their insulin requirements by 50%, lead
to weight loss, and improve HbA1c levels. The starting
dose for patients with type 2 diabetes is 60 ␮g, titrated to
120 ␮g as tolerated. Nausea is a common side effect of
pramlintide, but is usually transient and can be ad-
dressed through slow titration (8). Patients should be
advised that pramlintide cannot be mixed with insulin
and must be administered as a separate injection.
SUMMARY
For type 2 diabetes, the first line of intervention should
always include MNT and physical activity. Diabetes med-
ications support these lifestyle interventions, and their
dosage and time of use should be based on patients’
eating and activity patterns. RDs knowledgeable about
diabetes medications can use the information provided in
this review, along with their patients’ blood glucose re-
sults and eating and activity habits, to promote optimal
glycemic control even as diabetes progresses over time.
STATEMENT OF CONFLICT OF INTEREST: The au-
thors report the following conflict of interest with the
sponsor of this supplement article or products discussed
in this article. Vivian A. Fonseca, MD: Research Support
(to Tulane): Grants from Glaxo Smith Kline, Novartis,
Novo-Nordisk, Takeda, Astra-Zeneca, Pfizer, Sanofi-
Aventis, Eli Lilly, Daiichi-Sankyo, NIH, ADA. Honoraria
for Consulting and Lectures: Glaxo Smith Kline, Novar-
tis, Takeda, Pfizer, Sanofi-Aventis, Eli Lilly, Daiichi
Sankyo. Karmeen D. Kulkarni, MS, RD: Currently em-
ployed by Abbott Diabetes Care as Director of Scientific
Affairs Intensive Diabetes Management.
April 2008 ● Supplement to the Journal of the AMERICAN DIETETIC ASSOCIATION
References
1. Nathan DM, Buse JB, Davidson MB, Heine RJ, Holman RR, Sherwin
R, Zinman B. Management of hyperglycemia in type 2 diabetes: A
consensus algorithm for the initiation and adjustment of therapy: A
consensus statement from the American Diabetes Association and the
European Association for the Study of Diabetes. Diabetes Care. 2006;
29:1963-1972.
2. American Diabetes Association. Standards of medical care in diabe-
tes—2007. Diabetes Care. 2007;30(suppl 1):S4-S41.
3. Turner RC, Cull CA, Frighi V, Holman RR. Glycemic control with diet,
sulfonylurea, metformin, or insulin in patients with type 2 diabetes
mellitus: Progressive requirement for multiple therapies (UKPDS 49).
JAMA. 1999;281:2005-2012.
4. Austin MA, Kulkarni K, Powers M. Medications to support blood
glucose control. In: Austin MA, Kulkarni K, Powers M. Blood Glucose
Monitoring: Essential Skills for Health Care Professionals. St. Paul,
MN: Peekytoe Productions Inc.; 2003:33-48.
5. American College of Physicians. ACP Diabetes Care Guide: A Team-
Based Practice Manual and Self-Assessment Program. Philadelphia,
PA: American College of Physicians; 2007.
6. Nissen SE, Wolfski K. Effect of rosiglitazone on the risk of myocardial
infarction and death from cardiovascular causes. N Engl J Med.
2007;356:2457-2471.
7. Chiasson JL, Josse RG, Gomis R, Hanefeld M, Karasik A, Laakso M,
STOP-NIDDM Trial Research Group. Acarbose treatment and the
risk of cardiovascular disease and hypertension in patients with im-
paired glucose tolerance: The STOP-NIDDM trial. JAMA. 2003;290:
486-494.
8. Steil C. Pharmacologic therapies for glucose management. In:
Mensing C, Cypress M, Halstenson C, McLaughlin S, Walker E, eds.
The Art and Science of Diabetes Self-Management Education. Chicago,
IL: American Association of Diabetes Educators; 2006:321-355.
9. UK Prospective Diabetes Study Group. Intensive blood glucose control
with sulphonylureas or insulin compared with conventional treat-
ment and risk of complications with type 2 diabetes (UKPDS 33).
Lancet. 1998;352:837-853.
10. Bastyr EJ 3rd, Stuart CA, Brodows RG, Schwartz S, Graf CJ, Zagar
A, Robertson KE. Therapy focused on lowering postprandial glucose,
not fasting glucose, may be superior for lowering HbA1c. Diabetes
Care. 2000;23:1236-1241.
11. Raskin P, Allen E, Hollander P, Lewin A, Gabbay RA, Hu P, Bode B,
Garber A, INITIATE Study Group. Initiating insulin therapy in type
2 diabetes: A comparison of biphasic and basal insulin analogs. Dia-
betes Care. 2005;28:260-265.
12. Holman RR, Thorne KI, Farmer AJ, Davies MJ, Keenan JF, Paul S,
Levy JC, 4-T Study Group. Addition of biphasic, prandial, or basal
insulin to oral therapy in type 2 diabetes. N Engl J Med. 2007;357:
1716-1730.
13. Riddle MC, Rosenstock J, Gerich J. The Treat-to-Target trial: Ran-
domized addition of glargine or human NPH insulin to oral therapy of
type 2 diabetic patients. Diabetes Care. 2003;26:3080-3086.
14. Meneghini LF, Rosenberg KH, Koenen C, Merilainen MJ, Luddeke
HJ. Insulin detemir improves glycaemic control with less hypoglycae-
mia and no weight gain in patients with type 2 diabetes who were
insulin naive or treated with NPH or insulin glargine: Clinical prac-
tice experience from a German subgroup of the PREDICTIVE study.
Diabetes Obes Metab. 2007;9:418-427.
15. Davies M, Storms F, Shutler S, Bianchi-Biscay M, Gomis R. Improve-
ment of glycemic control in subjects with poorly controlled type 2
diabetes: Comparison of two treatment algorithms using insulin
glargine. Diabetes Care. 2005;28:1282-1288.
16. Rosenstock J. Basal insulin supplementation in type 2 diabetes: Re-
fining the tactics. Am J Med. 2004;116(suppl 3A):10S-16S.
17. Yki-Jarvinen H, Nikkila K, Makimattila S. Metformin prevents
weight gain by reducing dietary intake during insulin therapy in
patients with type 2 diabetes mellitus. Drugs. 1999;58(suppl 1):
53-54.
S33