1. Limitations and opportunities of insulin therapy
Luigi Meneghini
June 8th, 2012

2. Outline Insulin need versus implementation
Options for initiating insulin in T2DM
Limitations & opportunities for more stable basal insulins
Degludec pharmacodynamics and clinical studies
Adding an incretin to basal insulin replacement

3. Metabolic Status at Diagnosis of Type 2 Diabetes
Insulin resistance
40%
100 75
Beta-cell function
50%
Beta Cell Function (%) 50
IGT 25
Postprandial Hyperglycemia
Diabetes
-12
-10 -6 -2 2 6 10 14
Years From Diagnosis
Adapted from Lebovitz HE. Diabetes Reviews. 1999;7(3):139–153.

4. Glycemic Control with Monotherapy in the UKPDS Over 9 Years
Short-acting insulin added in 44% by 9 years
KEY POINT: While insulin is clearly most effective in maintaining FPG over time, overall glycemic control (as per A1C) still limited, underscoring need to address both fasting and prandial treatments
Insulin treatment details: The initial insulinregimenconsistedof
aonce-dailydose
of long-acting or isophane insulin. If the
daily dosewas above 14 U or if premeal
or prebedtime FPG concentration was
higher than 7 mmol/L (126 mg/dL),
regular insulin was added to the regimen.
Larger doses of insulin were used
when FPG concentration was higher
than6mmol/L(108mg/dL).The median
insulin doses at 6 and 9 years from diagnosis
of type 2 diabetes were 28 U and
34U, respectively. At 9 years, the median
dose was 24 U in non obese and 53 U in
obese subjects (BMI, #25 and "35
kg/m2, respectively). If hypoglycemia
occurred, the doses were reduced.
Although patients initially were treated
with a basal insulin supply from long acting
insulin, if home blood glucose
monitoring or HbA1c levels were unsatisfactory,
patients (44%) were transferred
to mixtures of long- and short acting
insulin or a twice-daily mixture
at 9 years.
Context Treatment with diet alone, insulin, sulfonylurea, or metformin is known to
improve glycemia in patients with type 2 diabetes mellitus, but which treatment most
frequently attains target fasting plasma glucose (FPG) concentration of less than 7.8
mmol/L (140 mg/dL) or glycosylated hemoglobin A1c (HbA1c) below 7% is unknown.
Objective To assess how often each therapy can achieve the glycemic control target
levels set by the American Diabetes Association.
Design Randomized controlled trial conducted between 1977 and Patientswere
recruited between 1977 and 1991 and were followed up every 3 months for 3, 6, and
9 years after enrollment.
Setting Outpatient diabetes clinics in 15 UK hospitals.
Patients A total of 4075 patients newly diagnosed as having type 2 diabetes ranged
in age between 25 and 65 years and had a median (interquartile range) FPG concentration
of 11.5 ( )mmol/L [207 ( )mg/dL], HbA1c levels of 9.1%(7.5%-
10.7%), and a mean (SD) body mass index of 29 (6) kg/m2.
Interventions After 3 months on a low-fat, high-carbohydrate, high-fiber diet, patients
were randomized to therapy with diet alone, insulin, sulfonylurea, or metformin.
Main Outcome Measures Fasting plasma glucose and HbA1c levels, and the proportion
of patientswho achieved target levels below7%HbA1c or less than 7.8mmol/L
(140 mg/dL) FPG at 3, 6, or 9 years following diagnosis.
Results The proportion of patientswhomaintained target glycemic levels declinedmarkedly
over 9 years of follow-up. After 9 years of monotherapy with diet, insulin, or sulfonylurea,
8%, 42%, and 24%, respectively, achieved FPG levels of less than 7.8 mmol/L
(140 mg/dL) and 9%, 28%, and 24% achieved HbA1c levels below 7%. In obese patients
randomized to metformin, 18%attained FPG levels of less than 7.8 mmol/L (140
mg/dL) and 13% attained HbA1c levels below 7%. Patients less likely to achieve target
levels were younger, more obese, or more hyperglycemic than other patients.
Conclusions Each therapeutic agent, as monotherapy, increased 2- to 3-fold the
proportion of patients who attained HbA1c below 7%compared with diet alone. However,
the progressive deterioration of diabetes control was such that after 3 years approximately
50% of patients could attain this goal with monotherapy, and by 9 years
this declined to approximately 25%. The majority of patients need multiple therapies
to attain these glycemic target levels in the longer term.
Turner RC et al. JAMA 1999; 281:

5. Physicians delay intensifying therapy for months, especially initiating insulin
9.5%
N=2319
N=3394
N=513
N=982
A1C>8% (mos) 4 17 12 26
A1C>7% (mos) 16 37 51
Brown et al. Diabetes Care 2004; 27: 1535

6. Options for Initiating & Intensifying Insulin Therapy in Type 2 Diabetes

7. Less hypoglycemia with basal initiation
Insulin Initiation & Intensification Outcomes in T2DM at Baseline, 1 & 3 Years
-1.3%*
-1.4%*
-1.2%*
3.0
5.5
1.7
Less hypoglycemia with basal initiation
(events/pt/yr) *
2009
Background
Evidence supporting the addition of specific insulin regimens to oral therapy in
patients with type 2 diabetes mellitus is limited.
Methods
In this 3-year open-label, multicenter trial, we evaluated 708 patients who had suboptimal
glycated hemoglobin levels while taking metformin and sulfonylurea therapy.
Patients were randomly assigned to receive biphasic insulin aspart twice daily,
prandial insulin aspart three times daily, or basal insulin detemir once daily (twice
if required). Sulfonylurea therapy was replaced by a second type of insulin if hyperglycemia
became unacceptable during the first year of the study or subsequently if
glycated hemoglobin levels were more than 6.5%. Outcome measures were glycated
hemoglobin levels, the proportion of patients with a glycated hemoglobin level of
6.5% or less, the rate of hypoglycemia, and weight gain.
Results
Median glycated hemoglobin levels were similar for patients receiving biphasic (7.1%),
prandial (6.8%), and basal (6.9%) insulin-based regimens (P = 0.28). However, fewer
patients had a level of 6.5% or less in the biphasic group (31.9%) than in the prandial
group (44.7%, P = 0.006) or in the basal group (43.2%, P = 0.03), with 67.7%,
73.6%, and 81.6%, respectively, taking a second type of insulin (P = 0.002). Median
rates of hypoglycemia per patient per year were lowest in the basal group (1.7),
higher in the biphasic group (3.0), and highest in the prandial group (5.7) (P<0.001
for the overall comparison). The mean weight gain was higher in the prandial
group than in either the biphasic group or the basal group. Other adverse event
rates were similar in the three groups.
Conclusions
Patients who added a basal or prandial insulin-based regimen to oral therapy had
better glycated hemoglobin control than patients who added a biphasic insulin-based
regimen. Fewer hypoglycemic episodes and less weight gain occurred in patients adding
basal insulin.
2007
Adding insulin to oral therapy in type 2 diabetes mellitus is customary when glycemic
control is suboptimal, though evidence supporting specific insulin regimens is
limited.
In an open-label, controlled, multicenter trial, we randomly assigned 708 patients with
a suboptimal glycated hemoglobin level (7.0 to 10.0%) who were receiving maximally
tolerated doses of metformin and sulfonylurea to receive biphasic insulin
aspart twice daily, prandial insulin aspart three times daily, or basal insulin detemir
once daily (twice if required). Outcome measures at 1 year were the mean
glycated hemoglobin level, the proportion of patients with a glycated hemoglobin
level of 6.5% or less, the rate of hypoglycemia, and weight gain.
At 1 year, mean glycated hemoglobin levels were similar in the biphasic group
(7.3%) and the prandial group (7.2%) (P = 0.08) but higher in the basal group (7.6%,
P<0.001 for both comparisons). The respective proportions of patients with a glycated
hemoglobin level of 6.5% or less were 17.0%, 23.9%, and 8.1%; respective
mean numbers of hypoglycemic events per patient per year were 5.7, 12.0, and 2.3;
and respective mean weight gains were 4.7 kg, 5.7 kg, and 1.9 kg. Rates of adverse
events were similar among the three groups.
A single analogue-insulin formulation added to metformin and sulfonylurea resulted
in a glycated hemoglobin level of 6.5% or less in a minority of patients at 1 year.
The addition of biphasic or prandial insulin aspart reduced levels more than the
addition of basal insulin detemir but were associated with greater risks of hypoglycemia
and weight gain.
235
222
201
239
222
188
234
224
189
* P<0.05
Holman, et al. NEJM 2009;361: Holman, et al. NEJM 2007;357:

8. Hypoglycaemia limits further reduction of FPG with basal insulin
Mean HbA1c [%]
Mean annual fasting blood glucose [mmol/l] 12
Frequency of Hypoglycaemic Episodes [%] 10 8 6 4 40 30 20 3 5 7 9 11
n = 13,072
Yki-Jarvinen et al. Ann Int Med 1999

9. How do Pharmacodynamics of Basal Insulin Preparations Affect Outcomes

10. Pharmacodynamics of NPH versus Glargine Insulin
Plasma glucose
Pharmacodynamics of NPH versus Glargine Insulin
Glucose infusion rate
Lepore, et al. Diabetes 1999; 48 (suppl 1): A97
Bolli et al. The Lancet • Vol 356 • August

11. Biologic activity over 24-hours more consistent for basal insulin analogs
Insulin detemir
Levemir® Offers a Consistent Blood Glucose Response
This randomized, double-blind, parallel-group study (enrolled n=54, completed n=51) compared the within-subject variability of the glucose-lowering effect of Levemir® with insulin glargine and NPH in patients with type 1 diabetes
Average A1C at baseline was 7.5%
Each patient was randomly assigned to receive 4 subcutaneous injections of 0.4 unit/kg under euglycemic glucose clamp conditions on 4 identical study days
Levemir®: n=18
Insulin glargine: n=16
NPH: n=17
The properties of each of the insulin preparations were recorded for 24 hours post-dose
The graph reflects the individual time-action profiles over 24 hours
GIR = Glucose Infusion Rate
Heise et al. Diabetes 2004; 53 (6):
Reference
Heise T, Nosek L, Rønn BB, et al. Lower within-subject variability of insulin detemir in comparison to NPH insulin and insulin glargine in people with type 1 diabetes. Diabetes. 2004;53:

12. Less hypoglycemia with basal analogues vs. NPH* * * *
*P<0.05
Riddle et al. Diabetes Care 2003; 26: 3080–3086. Philis-Tsimikas et al. Clin Ther 2006; 28 (10).

13. Modeled risk of hypoglycemia based on achieved A1C levels
Little S, et al. Diab Tech Ther 2011; 13 (S1)

14. Improving on current basal insulin analogs
Extend duration of action
Flat pharmacodynamic profile
Reduced day-to-day variability

15. Molecular size determines rate of subcutaneous absorption
Subcutaneous tissue
Molecular size
6 kDa
36 kDa
72 kDa
>5000 kDa
Insulin association state
Insulin
High molecular weight forms
Zn2+
Zn2+
Zn2+
Absorption
Absorption rate
Capillary membrane
Rapid absorption
Slow absorption
Brange et al. Diabetes Care 1990;13:923–54 15

16. Insulin degludec from solution to subcutaneous depot
Phenol Zn2+
Insulin degludec
injected
As phenol from the vehicle diffuses degludec hexamers link up via single side-chain contacts
Long multi-hexamers assemble 16

17. Insulin degludec multi-hexamers visible with transmission electron microscopy
SOLUTION
SC DEPOT
Main picture shows elongated IDeg structures in absence of phenol; inset (white box) shows absence of elongated IDeg structures in presence of phenol.
Main picture shows elongated insulin degludec structures in absence of phenol; inset shows absence of elongated insulin degludec structures in presence of phenol
Kurtzhals et al. Diabetes 2011;60(Suppl . 1):LB12 (Abstract 42-LB) (NN MOA) 17

18. Insulin degludec: slow release following injection
Subcutaneous depot
Zn2+
Insulin degludec
multi-hexamers
Zinc diffuses slowly causing individual hexamers to disassemble, releasing
monomers
Monomers are absorbed from the depot into the circulation 18

19. Insulin degludec PD profile at steady state in T1D2 4 6 8 10 12 14 16 18 20 22 24
Time (hours) 1 3 5
GIR (mg/kg/min)
Mean profile, n=66
IDeg = 0.4 U/kg
PD, pharmacodynamic
Heise et al. Diabetologia 2011;54(Suppl. 1):S425

20. Terminal half-life & coefficient of variation at steady state
Harmonic mean
(h) CV
(%)
Terminal half-life (steady state)
Degludec
24.5 23
Glargine
12.2 56
Degludec has a double as long half-life as insulin glargine. The long half-life is what makes the pharmacokinetic profiles flat and thereby gives the stable insulin 454 exposure during the 24 hour period between two doses over 24 hours.

21. Basal insulin initiation in T2DM
IDeg OD + metformin ± DPP-4 (n=773)
Insulin-naïve patients with type 2 diabetes
(n=1030)
IGlar OD + metformin ± DPP-4 (n=257)
Inclusion criteria
Type 2 diabetes ≥6 months
Insulin naïve treated with metformin ± SU, DPP-4 or acarbose for ≥3 months
HbA1c 7.0–10.0%
BMI ≤40 kg/m2
Age ≥18 years
52 weeks
Randomised 3:1 (IDeg OD:IGlar OD)
Open label
Insulin degludec administered with main evening meal
Insulin glargine administered according to label
DPP-4, dipeptidyl peptidase-4 inhibitor SU, sulphonylurea
OD, once daily Data on file: NN ; Accepted for presentation at ADA 2012

22. Pre-breakfast plasma glucosea
Weekly titration algorithm for insulin degludec and insulin glargine in T2DM
Pre-breakfast plasma glucosea
Adjustment
mmol/L
mg/dL U
<3.1b
<56b –4
3.1–3.9b
56–70b –2
4.0–4.9
71–89
5.0–6.9
90–125 +2
7.0–7.9
126–143 +4
8.0–8.9
144–161 +6
≥9.0
≥162 +8
a Mean of 3 consecutive days’ measurements for up titration.
b Unless there is obvious explanation for the low value, such as a missed meal

23. Serum IDeg concentration Proportion of Day 10 level (%)
Insulin degludec steady state is reached within 2–3 days of once-daily dosing 1 2 3 4 5 6 7 8 9 10 20 30 40 50 60 70 80 90
100
110
120
Days since first dose
Serum IDeg concentration Proportion of Day 10 level (%)
Relative serum IDeg trough concentrations during initiation of once-daily (0.4 U/kg) dosing in patients with T1DM
Values are estimated ratios and 95% CI relative to day 10
Heise T et al. IDF st World Congress Abstract Book. IDF: Dubai, 2011; Poster 1453

24. Pharmacokinetics of insulin steady state
Receptor activation &
insulin clearance
Absorption from the SC depot

25. No difference in HbA1c decrease over time between degludec & glargine
Degludec (n=773)
Glargine (n=257)
Treatment differences are derived from an LSMeans-based model
0.0
Time (weeks)
Mean±SEM; FAS; LOCF
Comparisons: Estimates adjusted for multiple covariates Data on file: NN ; Accepted for presentation at ADA 2012

26. No difference in overall confirmed hypoglycaemia
HYPOGLYCEMIA
BG < 56 mg/dl or severe
18% (ns)
Degludec (n=773)
Glargine (n=257)
Please note that after 26 weeks of the 52-week trial, visit (and therefore, reporting) frequency decreased.
Statistics based on FAS
Time (weeks)
SAS
Comparisons: Estimates adjusted for multiple covariates Data on file: NN ; Accepted for presentation at ADA 2012

27. Lower nocturnal confirmed hypoglycaemia with insulin degludec
Degludec (n=773)
Glargine (n=257)
36% p<0.05
Please note that after 26 weeks of the 52-week trial, visit (and therefore, reporting) frequency decreased.
Statistics based on FAS
Time (weeks)
SAS
Comparisons: Estimates adjusted for multiple covariates Data on file: NN ; Accepted for presentation at ADA 2012 27

28. Forced flexible insulin degludec study design
Degludec OD Flexible ±OADs (n=229)
(metformin/SU/pioglitazone)
Patients with type 2 diabetes
(n=687)
Degludec OD Fixed ±OADs (n=228)
(metformin/SU/pioglitazone)
Glargine OD ±OADs (n=230)
(metformin/SU/pioglitazone)
Inclusion criteria
Type 2 diabetes ≥6 months
Previously treated with OADs and/or basal insulin
HbA1c: OADs only 7–11% Basal insulin ± OADs 7–10%
BMI ≤40 kg/m2
Age ≥18 years
26 weeks
Open label
Birkeland et al. IDF 2011:P-1443; Bain et al. IDF 2011:O-0508; Birkeland et al. Diabetologia 2011;54(suppl. 1):S423;
Atkin et al. Diabetologia 2011;54(suppl. 1):S53; Meneghini et al. Diabetes 2011;60(suppl. 1A):LB10 (NN )

29. Timing of flexible insulin degludec administration29
Timing of flexible insulin degludec administration
Mon
Tue
Wed
Thu
Fri
Sat
Sun
morning
morning
morning 8h 8h
40h
40h
40h
8-12 AND hours between insulin administration
evening
evening
evening
evening
24h

30. No difference in A1C between flexible degludec and fixed dosing
Degludec Flexible OD
Degludec OD
Glargine OD
Treatment differences are derived from an LSMeans-based model
0.0
Time (weeks)
Birkeland et al. IDF 2011:P-1443; Bain et al. IDF 2011:O-0508; Birkeland et al. Diabetologia 2011;54(suppl. 1):S423;
Atkin et al. Diabetologia 2011;54(suppl. 1):S53; Meneghini et al. Diabetes 2011;60(suppl. 1A):LB10 (NN ) 30

31. Nocturnal hypoglycemia
No difference in hypoglycemia between flexible degludec and fixed dosing
Degludec Flexible OD Degludec OD Glargine OD
Overall hypoglycemia
Nocturnal hypoglycemia
23%(ns)
18%(ns)
cumulative events/patient/yr
cumulative events/patient/yr
Time (weeks)
Birkeland et al. IDF 2011:P-1443; Bain et al. IDF 2011:O-0508; Birkeland et al. Diabetologia 2011;54(suppl. 1):S423;
Atkin et al. Diabetologia 2011;54(suppl. 1):S53; Meneghini et al. Diabetes 2011;60(suppl. 1A):LB10 (NN )

32. Insulin Lispro Pegylation
= kDa

33. Pegylated Lispro Insulin PD

34. Fasting vs. post-prandial contribution to A1C: baseline & after basal insulin
Fasting hyperglycemia
Post-prandial hyperglycemia
OBJECTIVE To determine the relative contributions of basal hyperglycemia (BHG) versus postprandial hyperglycemia (PPHG) before and after treatment intensification in patients with glycated hemoglobin A1c (A1C) >7.0% while on prior oral therapy.
RESEARCH DESIGN AND METHODS Self-measured, plasma-referenced glucose profiles and A1C values were evaluated from participants in six studies comparing systematically titrated insulin glargine with an alternative regimen (adding basal, premixed, or prandial insulin, or increasing oral agents). Hyperglycemic exposure (>100 mg/dL [5.6 mmol/L]) as a result of BHG versus PPHG was calculated.
RESULTS On prior oral therapy, 1,699 participants (mean age 59 years, diabetes duration 9 years) had mean fasting plasma glucose (FPG) of 194 mg/dL (10.8 mmol/L), and mean A1C was 8.7%. BHG contributed an average of 76–80% to hyperglycemia over the observed range of baseline A1C levels. Adding basal insulin for 24 or 28 weeks lowered mean FPG to 117 mg/dL (6.5 mmol/L), A1C to 7.0%, and BHG contribution to 32–41%. Alternative regimens reduced FPG to 146 mg/dL (8.1 mmol/L), A1C to 7.1%, and the contribution of BHG to 64–71%. BHG contributions for patients with A1C averaging 7.6–7.7% were 76% at baseline and 34 and 68% after adding basal insulin or other therapies, respectively.
CONCLUSIONS When A1C is >7.0% despite oral therapy, BHG routinely dominates exposure. Intensified therapy reduces A1C and changes this relationship, but BHG amenable to further intervention still accounts for one-third of total hyperglycemia after basal insulin treatment and two-thirds after alternative methods.
Basal insulin
Riddle, et al. Diabetes Care 2011; 34 (12):

35. Exenatide added to basal insulin glargine improves control in T2DM
A1C %
Insulin
0.5 u/kg
BMI
33-34
-1.0%
+20u
+1.0kg
-1.7%
+13u
-1.8kg
BACKGROUND:
Insulin replacement in diabetes often requires prandial intervention to reach hemoglobin A₁(c) (HbA₁(c)) targets.
OBJECTIVE:
To test whether twice-daily exenatide injections reduce HbA₁(c) levels more than placebo in people receiving insulin glargine.
DESIGN:
Parallel, randomized, placebo-controlled trial, blocked and stratified by HbA₁(c) level at site, performed from October 2008 to January Participants, investigators, and personnel conducting the study were masked to treatment assignments. (ClinicalTrials.gov registration number: NCT )
SETTING:
59 centers in 5 countries.
PATIENTS:
Adults with type 2 diabetes and an HbA₁(c) level of 7.1% to 10.5% who were receiving insulin glargine alone or in combination with metformin or pioglitazone (or both agents).
INTERVENTION:
Assignment by a centralized, computer-generated, random-sequence interactive voice-response system to exenatide, 10 µg twice daily, or placebo for 30 weeks.
MEASUREMENTS:
The primary outcome was change in HbA₁(c) level. Secondary outcomes included the percentage of participants with HbA₁(c) values of 7.0% or less and 6.5% or less, 7-point self-monitored glucose profiles, body weight, waist circumference, insulin dose, hypoglycemia, and adverse events.
RESULTS:
112 of 138 exenatide recipients and 101 of 123 placebo recipients completed the study. The HbA₁(c) level decreased by 1.74% with exenatide and 1.04% with placebo (between-group difference, -0.69% [95% CI, -0.93% to -0.46%]; P < 0.001). Weight decreased by 1.8 kg with exenatide and increased by 1.0 kg with placebo (between-group difference, -2.7 kg [CI, -3.7 to -1.7]). Average increases in insulin dosage with exenatide and placebo were 13 U/d and 20 U/d. The estimated rate of minor hypoglycemia was similar between groups. Thirteen exenatide recipients and 1 placebo recipient discontinued the study because of adverse events (P < 0.010); rates of nausea (41% vs. 8%), diarrhea (18% vs. 8%), vomiting (18% vs. 4%), headache (14% vs. 4%), and constipation (10% vs. 2%) were higher with exenatide than with placebo.
LIMITATIONS:
The study was of short duration. There were slight imbalances between groups at baseline in terms of sex, use of concomitant glucose-lowering medications, and HbA₁(c) levels, and more exenatide recipients than placebo recipients withdrew because of adverse events.
CONCLUSION:
Adding twice-daily exenatide injections improved glycemic control without increased hypoglycemia or weight gain in participants with uncontrolled type 2 diabetes who were receiving insulin glargine treatment. Adverse events of exenatide included nausea, diarrhea, vomiting, headache, and constipation
OBJECTIVE To determine variables associated with glycemic and body weight responses when adding exenatide to basal insulin-treated type 2 diabetes. RESEARCH DESIGN AND METHODS Exploratory subgroup analyses based on baseline A1C, disease duration, and BMI of a 30-week study comparing exenatide twice daily to placebo, added to optimized insulin glargine (intent-to-treat analysis: 137 exenatide; 122 placebo). RESULTS Exenatide participants had greater A1C reductions compared with optimized insulin glargine alone, irrespective of baseline A1C (P < 0.001). Exenatide participants with longer diabetes duration and those with lower BMI had greater A1C reductions (P < 0.01). Exenatide participants lost more weight, regardless of baseline A1C or BMI (P < 0.05). Exenatide participants with longer diabetes duration lost the most weight (P < 0.001). CONCLUSIONS Exenatide added to optimized basal insulin was associated with improved glycemic control and weight loss, irrespective of baseline A1C, diabetes duration, and BMI. Changes were evident in modestly obese patients and in those with longer diabetes duration.
Longer diabetes duration and lower BMI had greater A1C reductions. Longer diabetes duration also lost the most weight.
Minor hypoglycemia
25% (EXE) vs 29% (PLB)
Buse, et al. Ann Intern Med. 2011;154: Rosenstock, et al. Diabetes Care 2012; 35(5): Epub 2012 Mar 19.

36. Effective combinations of basal replacement and GLP-1 Ras
Conclusions
Ultra-long acting basal insulin with improved consistency & less hypoglycemia
Effective combinations of basal replacement and GLP-1 Ras
Smarter & simpler approaches to treatment
Ultimately, more intensive insulin regimens may be required (see Figure 3.)
Dashed arrow line on the left-hand side of the figure denotes the option of a more rapid progression from a 2-drug combination directly to multiple daily insulin doses, in those patients with severe hyperglycaemia (e.g. HbA1c ≥ %).
Consider beginning with insulin if patient presents with severe hyperglycemia (≥ mg/dl [≥ mmol/l]; HbA1c ≥ %) with or without catabolic features (weight loss, ketosis, etc).
Diabetes Care, Diabetologia.
19 April 2012 [Epub ahead of print]