Insulin management in overweight or obese type 2 diabetes patients: the role of insulin glargine

Type 2 diabetes mellitus (T2DM) and obesity commonly co‐exist. Improved clinical management of T2DM and improved glycaemic control with traditional therapies including insulin usually result in some weight gain – a frequently perceived barrier to the introduction of insulin by both patient and healthcare professionals. Weight gain of 2.5 kg per 1% change in haemoglobin A1c (HbA1c) is common in many studies. Strategies to minimize weight gain, particularly in obese patients, are essential to help patients better manage their diabetes and improve quality of life. Insulin analogues with lower risk of hypoglycaemia and better within‐patient variability compared with human insulin may help facilitate reaching treatment goals. Moreover, weight gain can be minimized by earlier insulinization and the use of basal insulin, such as insulin glargine, instead of premixed insulin. Data specific to the obese patient with T2DM are presented; they are currently limited but do indicate that insulin glargine therapy is associated with improved glycaemic control as well as less weight gain than other insulins, such as premixed insulin and prandial insulin regimens. Retrospective subanalyses of earlier trials and ongoing studies would shed further light on the impact of insulin therapy in obese people with T2DM in addition to determination of optimal therapeutic strategies.


Introduction
Type 2 diabetes mellitus (T2DM) is a significant health problem, the prevalence of which is steadily increasing [1][2][3], and is associated with increased risk of vascular disease. However, it is well established that T2DM is also closely associated with obesity. It is predicted that up to 60% of cases of T2DM could potentially be avoi-ded if the body mass index (BMI) of the population was maintained within the normal healthy range [4]. In addition, with greater understanding of the pathophysiology of diabetes, it appears that there are some differences in insulin secretion profiles between people with normal weight and obese people with T2DM, which partly relates to the degree of peripheral insulin resistance [5].
For many patients, insulin therapy used in conjunction with oral glucose-lowering drugs, most commonly sulphonylurea and metformin, is essential to maintain good glycaemic control. However, such a therapeutic strategy is associated with weight gain, which is a common barrier among patients and providers alike [6]. Understandably, therefore, there is some reluctance to use insulin in obese patients with T2DM as there is a perception that they may be more likely to gain relatively more weight, potentially worsening insulin resistance; thus, weight gain remains a significant barrier to insulin initiation [7][8][9]. Newer analogues, such as insulin glargine and insulin detemir, offer a simple approach for the introduction of insulin, with oncedaily dosing for insulin glargine and some insulin detemir patients, improved within-patient variability and lower risk of hypoglycaemia vs. neutral protamine Hagedorn (NPH) insulin [10,11]. The aim of this review is to discuss the role of insulin therapy in obese patients with T2DM and strategies to minimize insulin-induced weight gain.

Obesity in T2DM
The global prevalence of obesity is steadily increasing, particularly in highly industrialized regions such as Europe [2] and North America [1,12]. Developing nations, and some ethnic groups, are also showing a trend towards increasing prevalence of obesity and T2DM [13,14]. Obese individuals, typically defined as having a BMI > 30 kg/m 2 (although this threshold is lower in some ethnic groups; e.g. the threshold for obesity is 27.5 kg/m 2 for South Asian populations [15]), are at greater risk of developing impaired fasting glucose and impaired glucose tolerance, which often progress to overt T2DM as a result of progressive dysfunction in b-cell function [16]. In addition, obese individuals also show abnormal liver, muscle and adipose tissue lipid handling, which has also been implicated in the development of insulin resistance and T2DM [17]. This is of significant concern, given the trend towards increasing prevalence of obesity and T2DM [18], particularly over the next 25 years [3], in addition to the strong association between these risk factors and microvascular [19][20][21] and macrovascular diseases [22][23][24].
Indeed, in people with T2DM and who are obese, the level of insulin resistance is typically greater than that in lean people with T2DM, as was demonstrated in a study by Røder et al. [5] 5.5 AE 1.2 years, respectively)}, there were clinically relevant differences in b-cell peptides in the fasting state and after glucagon stimulation. In particular, levels of intact proinsulin (6.6 AE 1.0 vs. 7.7 AE 2.0 pmol/l, respectively; p < 0.01) and C-peptide (598 AE 32 vs. 893 AE 112 pmol/l, respectively; p < 0.05) were significantly different in both groups, while insulin tended (although not statistically significant) to be higher in the obese group compared with the lean group (32.5 AE 4.9 vs. 63.9 AE 20.4 pmol/l, respectively), which likely reflects increased insulin resistance in obese patients [25,26].
Treatment of T2DM and improved metabolic control are associated with weight gain, particularly when insulin, thiazolidinediones (TZDs) or sulphonylureas are used [27]. In ADOPT (A Diabetes Outcome Progression Trial), weight gain over 5 years was 4.8 kg with rosiglitazone and 1.6 kg with glyburide [27]. This compares with a weight loss of 1.6 kg in the metformin group in that study [27]. However, for many patients, weight gain is unacceptable and commonly results in poor adherence to therapy or refusal to accept more intensive therapy, particularly the addition of insulin, despite its proven efficacy [9]. Therefore, it is important to understand why weight gain is common with insulin therapy and then consider potential strategies to minimize weight gain.

T2DM, Weight Gain and Therapeutic Strategies to Minimize Weight Gain
Management of T2DM typically involves the intensive use of a number of therapeutic agents that should be optimally titrated according to the patient's needs and glycaemic status. However, a common feature is that the most commonly used therapeutic agents are associated with some weight gain, except metformin, which is largely weight neutral [28]. Weight gain is a barrier to the introduction of insulin therapy. However, the natural progression of T2DM means that insulin therapy is often essential [29]. Therefore, it is important to consider how best to use insulin therapy to minimize weight gain, which is particularly true for obese patients. Indeed, it has been suggested that for each 1% absolute reduction in haemoglobin A 1c (HbA 1c ), an increase in body weight of 2.5 kg can be expected [30], which is supported by data showing a strong, inverse association between improvements in fasting glucose and changes in body weight (figure 1). However, before strategies to minimize weight gain can be discussed, it is important to understand why insulin therapy is associated with weight gain.

Mechanisms of Weight Gain in Insulin-treated T2DM -Pharmacodynamic Studies of Insulin
A number of pharmacodynamic studies have determined potential mechanisms that would conceptually lead to weight gain with long-term insulin therapy. Key factors included hypoglycaemia-associated snacking, reduced glucose excretion (glycosuria) and reduced metabolic rates.

Hypoglycaemia and Snacking
A common unwanted effect of insulin therapy is that of hypoglycaemia, and patients may respond to symptomatic hypoglycaemia, in particular, by increased snacking to maintain stable blood glucose levels [31], although patients on insulin therapy should be aware of the risks of hypoglycaemia. The modern insulin analogues are associated with lower risk of hypoglycaemia [10,11], lower within-patient variability [32] and lower fluctuation [33] in action compared with the human equivalent. Thus, the need for snacking to avoid hypoglycaemia should be reduced. One might anticipate that reduced hypoglycaemia-related snacking with newer insulin analogues would allow for less weight gain. In studies with insulin glargine, the reduced weight gain vs. NPH insulin was evident in the studies by Rosenstock et al. [34] and by Yki-Jarvinen et al. [35], while reduced weight gain is consistent in studies with insulin detemir vs. NPH insulin [36], although the underlying mechanism for this is still not fully understood.

Glycosuria
In people with poorly controlled T2DM, excess glucose is usually excreted in urine (glycosuria). In one study of six obese patients with T2DM, baseline urinary glucose excretion was 48 AE 19 g/day. After switching to glyburide treatment, glucose excretion fell to 20 AE 9 g/day and further declined to 2 AE 1 g/day with insulin therapy [37]. This reversal of glycosuria will inevitably lead to weight gain unless diet or exercise levels are changed, as demonstrated in figure 2.

Basal Metabolic Rate
The basal metabolic rate is also influenced by insulin therapy and is largely decreased by insulin with reduced resting energy expenditure, which may reflect increased efficiency in fuel selection as a result of better glycaemic control [30]. Therefore, weight gain will ensue unless diet or activity levels are adjusted accordingly.

Strategies to Reduce Weight Gain
As described above, weight gain is largely an expected effect of insulin therapy but is, nevertheless, unwanted by the majority of people treated with insulin. However, there are strategies to minimize the extent of weight gain associated with improved glycaemic control in patients on insulin.

Insulin vs. Alternative Oral Agents
Typically, insulin is added as a third agent after metformin and sulphonylurea doses have reached the maximum tolerated, although the joint American Diabetes Association and the European Association for the Study of Diabetes consensus statement advocates the earlier use of basal insulin [29]. For the third agent, there is a choice between adding a TZD or adding insulin. This was evaluated in a study comparing triple therapy of insulin glargine vs. rosiglitazone (both added to sulphonylurea plus metformin) [38]. In that study, insulin glargine was associated with significantly less weight gain (1.6 vs. 3.0 kg, respectively; p ¼ 0.02) over 24 weeks. Furthermore, it appears that weight gain with insulin therapy can be further minimized depending on the stage of the treatment pathway at which insulin is started, such as initial therapy in combination with metformin, or later in the treatment pathway when a combination of maximally tolerated doses of sulphonylurea plus metformin provides inadequate glycaemic control.
Ordinarily, insulin therapy would be expected to increase weight by 2.5 kg for every 1% reduction in HbA 1c . Indeed, in a study by Mäkimattila et al. [30], which assessed the impact of insulin therapy alone or insulin plus metformin therapy on weight gain in T2DM patients, improvements in HbA 1c were similar in both groups, but the insulin plus metformin group required 47% less insulin and experienced less weight gain than the insulin-alone group (3.8 vs. 7.5 kg, respectively; p < 0.05). Thus, unless contraindicated, metformin should continue when insulin is initiated in T2DM.

Insulin Analogues, Human Insulin or Premixed Insulin?
Moreover, the choice of insulin may limit the weight gain observed. Indeed, the LANMET study showed that when adding insulin glargine or NPH insulin to metformin, weight gain was lower with insulin glargine (þ2.6 vs. 3.5 kg, respectively; table 1) [35]. Similarly, the LAP-TOP (LANTUS þ Amaryl þ metformin vs. Premixed insulin in T2DM patients after failing Oral treatment Pathways) [39] and INITIATE (INITiation of Insulin to reach A1c TargEt) [40] studies demonstrated that the use of once-daily insulin glargine is associated with less weight gain than either twice-daily human insulin (þ1.4 vs. þ2.1 kg, respectively; p ¼ 0.0805) or twice-daily biphasic insulin aspart (þ3.5 vs. þ5.4 kg, respectively) (table 1), when added to existing oral agents. These findings are consistent with those reported in the 4-T (Treating to Targets in Type 2 diabetes) study, in which weight gain was lower with basal insulin vs. either biphasic or prandial insulin therapy [41]. However, in these studies, although the study populations comprised a majority of obese people (mean BMI was ;30 kg/m 2 ), it was not reported whether there were differential effects in normal weight, overweight or obese patients. Insulin glargine may offer benefits to these people, with lower risk of hypoglycaemia and potentially a decreased need for snacking or other preventative measures. Some studies have ascertained the effects of insulin glargine therapy in obese patients with T2DM.

Weight Management with Insulin Glargine in Randomized, Controlled Trials and Observational Studies
In randomized, controlled trials comparing insulin glargine with NPH insulin, when added to existing oral antidiabetic drugs (OADs), weight gain was seen with both insulins but was broadly similar across the trials (table 1). In contrast, in the two studies that compared insulin glargine therapy with premixed insulin, insulin glargine was associated with less weight gain, which was significant in one trial and of borderline significance in the other (table 1).
In the AT.LANTUS (A Trial comparing LANTUS Ò Algorithms to achieve Normal blood glucose Targets in patients with Uncontrolled blood Sugar) study [42], which compared a physician-managed algorithm (Algorithm 1; mean BMI at baseline: 29.0 AE 4.7 kg/m 2 ) with a patientmanaged algorithm (Algorithm 2; mean BMI at baseline: 29.0 AE 4.7 kg/m 2 ) for the initiation of insulin glargine, baseline to end-point increases in body weight were relatively modest in both algorithms (Algorithm 1: 79.8 AE 15.8 to 80.8 AE 16.0 kg; Algorithm 2: 79.8 AE 16.2 to 81.1 AE 16.5 kg) and similar to that expected for the magnitude of HbA 1c reduction (À0.9 and À1.1%, respectively) achieved [30].
Meanwhile, in an observational study of everyday clinical practice, which evaluated the switch from premixed insulin to insulin glargine plus OADs in 5045 patients (mean diabetes duration: 8.7 years) for 12 weeks, mean weight change was À1.5 AE 3.2 kg, while HbA 1c decreased by 1.1% and fasting blood glucose (FBG) decreased by 2.0 mmol/l (36 mg/dl) [43].
A recent, 32-month, open-label, uncontrolled, multicentre, observational study (which had previously demonstrated improvements in HbA 1c at 3, 9 and 20 months) assessed the efficacy and safety of initiating insulin glargine in addition to existing OADs [45,46]. In this study, the greatest reductions in HbA 1c (-1.8 AE 1.8%) were in patients who were obese (!30 kg/m 2 ) at the start of observation. Furthermore, some weight loss was also seen in these patients (À4.4 AE 10.7 kg) ( figure 3).
These results are promising for obese patients, suggesting that improvements in glycaemic control and weight loss can be achieved and, importantly, maintained over 32 months of treatment with insulin glargine.
In the majority of trials of insulin glargine, many of the patients are obese, as is common in T2DM. Several of these studies, including AT.LANTUS [42], GOT (Glycaemia Optimization Trial) [47] and GOAL A1C (Glycemic Optimization with Algorithms and Labs At po1nt of Care) [48], involved in excess of 3000 patients each. The size of these studies will provide an opportunity to evaluate through retrospective subanalyses the efficacy of insulin glargine therapy across the cut-points for obesity.

Discussion
Good blood glucose control reduces the risk of long-term diabetes complications. However, tight metabolic control can be particularly difficult in obese patients. Hypoglycaemia resulting from insulin therapy is a common fear Fig. 3 Change in body weight over 32 months in patients treated with insulin glargine plus oral glucose-lowering agents in everyday clinical practice according to BMI at the start of observation [45]. BMI, body mass index. *p < 0.0001. for many patients with T2DM, and it can be a possible influence on heath providers' and patients' treatment policies alike. This fear can be a major barrier to achieving good glycaemic control.
Insulin glargine provides at least equivalent glycaemic control when compared with NPH insulin regimes, and hypoglycaemic episodes are less common, as reported in a meta-regression of six studies of insulin glargine vs. NPH insulin [10]. This provides the opportunity for patients with diabetes to titrate to optimal doses and thereby increase the potential for further glycaemic control.
Weight gain is a commonly cited barrier to insulin therapy [7][8][9], and results of many studies indicate that weight gain is largely unavoidable, which may be a result of improved tissue glucose uptake, reduced glycosuria, snacking and reduced resting energy expenditure. However, studies in clinical practice suggest that weight gain can be minimized with insulin therapy [43,46], and this does not appear to compromise the improvements in glycaemic control seen in these studies.
A number of large, randomized, controlled trials have been performed to evaluate the efficacy and safety of insulin glargine in T2DM. In particular, the AT.LANTUS [42], GOT [47] and GOAL A1C [48] studies each recruited a large number of patients across a range of BMI values. Subanalyses of these studies, by stratifying the patients according to BMI, are in progress and should provide informative comparisons between normal weight and obese patients.
Such data would allow for a greater understanding of the impact of insulin therapy not only on glycaemic control and risk of hypoglycaemia but also on weight management in these patients. Indeed, this seems essential owing to differences in the pathophysiology of T2DM in lean and obese patients as well as optimum insulin titration regimens and dosing algorithms owing to the differential insulin secretory capability and insulin resistance in these patients [5]. Although both insulin glargine and insulin detemir are associated with less weight gain than other insulin preparations, particularly premixed insulin or regular human/NPH insulin, this does not preclude the continuation of lifestyle factors, such as diet and exercise, to help prevent excessive weight gain.
Finally, while weight gain may be an undesired factor and excessive adiposity is a risk factor for cardiovascular disease, the improvements in glycaemic control would be expected to outweigh these risks, as indicated by the United Kingdom Prospective Diabetes Study [49] and the Kumamoto [50] study. These studies demonstrated that the improvements in glycaemic control were associated with significantly reduced risk for cardiovascular end-points despite weight gain.
In summary, insulin therapy with insulin glargine is associated with clinically important improvements in glycaemic control. Retrospective analysis of earlier trials and new studies would help elucidate the impact of insulin therapy in normal weight to obese patients with T2DM. Such data are of interest to help the clinician (and patient) determine the optimal treatment algorithms.