Avondiabetes.nhs.uk

Clinical practice
Although not currently
covered by the licence for
either of the thiazolidinediones
available in the UK, triple oral
hypoglycaemic therapy including
glitazones is recommended by a number of
practitioners. Dr Ian Lawrence examines the latest
evidence and reports on his own practice in this
contentious area of diabetes care
Dr Ian Lawrence is Consultant Physician in
For many years, the conventional therapeutic Diabetes at the Leicester Royal Infirmary. He has
management of Type 2 diabetes in the United been a member of the University Hospitals of
Leicester Therapeutic Advisory Service since 1999,
together with one of the sulphonylureas, such as Chair of the Leicestershire and Rutland Local
glibenclamide, gliclazide and glimepiride. Once Diabetes Services Advisory Group since March
these agents are at maximum tolerated dose, then 2002, and Co-ordinating Editor of the Continuing
the next step has been to introduce insulin if Education Supplement to Diabetic Medicine since
glycaemic control remains poor or deteriorates to November 2002. He has sat on advisory boards
become poor (the definition of poor glycaemic for both GSK and Takeda and has received both
control given in England’s Diabetes NSF is an HbA speaker honoraria and support to attend
conferences from both companies.
The use of a third oral hypoglycaemic agent (OHA) has not been common practice, althoughthis option was available from the early 1990’s with Licence indications and NICE
acarbose. Problems with flatulence associated with The initial licence indication of the two glitazones acarbose and the drug’s disappointing performance was in dual therapy with either metformin (in in the United Kingdom Prospective Diabetes Study overweight and obese patients) or a sulphonylurea (UKPDS) have limited take-up of this form of triple (if metformin is not tolerated or contraindicated). hypoglycaemic therapy (THT) to a small number of More recently both glitazones have received a monotherapy licence, but neither have a The launch in 2000 of the thiazolidinediones, THT licence. However, in November 2004, GSK rosiglitazone and pioglitazone, renewed the possi- bility of THT, this time using a thiazolidinedione licensing authorities to the extension of the alongside both a sulphonylurea and metformin.
rosiglitazone licence to incorporate THT.
Indeed, other possibilities include quadruple Nevertheless, thus far, prescribing THT in Type 2 hypoglycaemic therapy in patients already on diabetes has been an out-of-licence usage of the triple therapy with acarbose, or THT incorporating glitazones, although not contraindicated, as is a prandial glucose regulator (repaglinide or the case with the glitazones and insulin in the nateglinide), metformin and a glitazone. UK due to the risk of fluid retention.
Diabetes Update Winter 2004
The National Institute for Clinical Excellence Author’s key points
(NICE) looked at both glitazones individually shortly after their launch, then in the blood • Patients on dual combination therapy with a glucose management guidance of 2002 and finally sulphonylurea and metformin, in the context of both glitazones together in 2003,1 the latter
poor glycaemic control, should not have one of guidance preceding the agents’ receipt of a the agents stopped and replaced by a glitazone, monotherapy licence. The preference for NICE as glycaemic control will initially deteriorate.
has been consistently for the combination of • Patients with progressive renal impairment metformin and a generic sulphonylurea, although should have a staged substitution (where this position has posed NICE some difficulties and, possible) of metformin with a glitazone, which in draft guidance at least, has led on occasions to involves a three to six month period of THT. • THT has been used successfully to improve An example of the difficulty posed by a steadfast glycaemic control in asymptomatic diabetic adherence to a position that excludes the option of THT is in patients with progressive renal impairment, when metformin can be electively discontinued and replaced by a glitazone (with • However, patients with osmotic symptoms, the proviso that the patient does not have fluid fatigue and weight loss should be commenced retention associated with the renal impairment). However, the glitazones are not particularly effective at improving glycaemic control rapidly, • When THT is employed, the improvement in and usually take three to six months to have glycaemic control is gradual, and a reduction in maximum effect. Thus, in the context of the sulphonylurea dose may be required after a patient who needs to discontinue metformin on account of renal impairment, the most appropriate • THT improves glycaemic control over six clinical management is an overlap period of months in placebo-controlled RCT’s with both both metformin and a glitazone, as well as a sulphonylurea (ie temporary THT for a period of • THT provides similar improvement in glycaemic control over six months compared to commonly To discontinue metformin immediately, and not used insulin initiation regimens, but with less overlap with a glitazone, puts the patient at risk hypoglycaemia. However, other side effects such of deteriorating glycaemic control, which is as peripheral oedema are increased, and some potentially preventable. Indeed, some of the poor patients will need to switch to insulin.
experiences with the glitazones shortly after their • THT is most likely to be effective at producing launch related to inappropriate substitutions, replacing either metformin or the sulphonylurea patients with HbA1c less than 8.0 per cent on with a glitazone, there being a predictable decline in glycaemic control and, invariably, insulin beingintroduced shortly afterwards.
Triple Oral Hypoglycaemic Therapy
pioglitazone in THT who have acceptable glycaemic However, many patients without progressive renal control (HbA1c < 7.5 per cent) after four years. impairment have also received THT, and this is not When initiating one of the glitazones in THT, for a temporary period of three to six months. patients are counselled that their glycaemic Such clinical decisions have been determined by a control is likely to improve, although this is not combination of both patient and clinician choice, guaranteed. Clearly if a patient is insulin deficient working outside the current licence indications of through loss of beta cell function, then insulin therapy alone will remedy this. However, such The most common scenarios informing these patients often have associated polydipsia, polyuria, clinical decisions have been either the implications nocturia and fatigue (and, more than occasionally, for employment of starting insulin (usually the weight loss), and THT has nothing to offer in such potential loss of a job) or a refusal to start insulin circumstances. I counsel patients that only insulin therapy (whether this be due to needle phobia or will remedy their symptoms, and this treatment not). On occasion, THT has ended up being a needs to be commenced as soon as possible.
staging post en route to commencing insulin, but However, the relatively asymptomatic patient many patients have experienced a clinically with Type 2 diabetes on metformin and a sulphony- significant improvement in glycaemic control and lurea with poor glycaemic control is quite a this is often a sustained improvement. In my own different scenario. Conventional practice remains clinic, there are patients on either rosiglitazone or the introduction of insulin, but THT is another Winter 2004 Diabetes Update
Clinical practice
therapeutic option, particularly in the context mean baseline HbA1c 8.7 per cent. After six months, of potential loss of employment or extreme the mean HbA1c fell by 0.58 per cent on rosiglita- reluctance to start insulin. When the glitazone is zone 4mg daily and 1.09 per cent on rosiglitazone commenced, glycaemic improvement is a gradual 8mg daily.2 Weight gains were 3.1kg and 5.1kg
process over months, and home blood glucose respectively, while the responder rate (an HbA1c monitoring is helpful in enabling the patient to fall of greater than 0.7 per cent) was 63 per cent.
gauge this improvement. Mild hypoglycaemia is a Similarly, a 12-week RCT in 283 Chinese patients possibility, and the sulphonylurea dose needs to be with diabees demonstrated improved glycaemic control with pioglitazone 30mg daily in THT My practice is to review patients after three and against dual combination therapy with placebo.3
six months, and then 12 months, with blood Another recent study, this time of 217 patients, investigations being repeated pre-visit. These compared rosiglitazone in THT against insulin include HbA1c , renal and liver chemistry, lipid glargine added to sulphonylurea and metformin profile and muscle enzymes – the latter being (the latter using a forced titration to target fasting required as the vast majority of patients are also plasma glucose ≤ 5.5mmol/l).4 Rosiglitazone was
receiving a statin. If a clinically significant, commenced at 4mg daily and an increase to 8mg sustained improvement in glycaemic control has daily was allowed if the FPG was > 5.5mmol/l. The been achieved, then THT is continued. If using improvements in HbA1c were 1.5 per cent with rosiglitazone, I commence at a dose of 4mg once rosiglitazone and 1.7 per cent with insulin glargine daily and increase to 4mg twice daily after three over 24 weeks (the baseline HbA1c being 8.7 per months if the HbA1c remains greater than 7.5 per cent and 8.8 per cent respectively). Weight gain cent. Meanwhile, with pioglitazone, I have usually was 3.0kg in the rosiglitazone group and 1.4kg in initiated at 30mg once daily, and then continued the insulin glargine group, while the glargine- at this dose (45mg once daily is now an option treated patients had more confirmed symptomatic and I suspect will increasingly be the dose used nocturnal hypoglycaemia with BG < 3.9mmol/l.
There were greater adverse reactions in the I also check carefully for mild hypoglycaemia, rosiglitazone group, while 12 per cent developed particularly if the HbA1c is less than 7.0 per cent, and have a low threshold for reducing the When this study was presented at the European sulphonylurea dose. It is critical that the patient is Association for the Study of Diabetes (EASD) engaged in the process, and if blood glucose meeting in September 2004, many of the UK readings are running less than 4.0mmol/l, then clinicians attending felt that the insulin glargine they are encouraged to prompt a reduction in results were as anticipated, whereas rosiglitazone sulphonylurea dose. This often occurs several in THT did better than anticipated. In particular, months after initiation – and in between clinic the similar improvement in HbA1c was noteworthy visits – and their usual contact point would be the when the insulin glargine group had the advan- diabetes specialist nurse or practice nurse.
tage of a forced titration regimen. Not surprisingly, Patients on THT, like any other patient on a the insulin glargine group had a better response in glitazone, may experience glitazone-related FPG (-3.6mmol/l vs -2.6mmol/l). Similarly, insulin side-effects, but these are rare. Most patients do glargine produced greater reductions in HbA1c gain weight (typically 2-3kg) through the first when the baseline HbA1c was ≥ 9.5 per cent. few months, and mild fluid retention is the most A smaller RCT of 62 patients with baseline HbA1c common side-effect. However, it is rare to > 8.0 per cent compared pioglitazone in THT discontinue glitazones on account of fluid against bedtime NPH insulin.5 Again there were
retention, the most usual reason for discontinua- similar improvements in HbA1c (1.9 per cent and tion being poor glycaemic control and the need 2.3 per cent reductions respectively), but hypogly- caemia was less frequent in the pioglitazone group (37 per cent vs 68 per cent). There were The evidence base for THT
similar effects upon weight, but the pioglitazone There is a growing evidence base to support THT group had an increase in high-density lipoprotein using glitazones, both in the form of randomised controlled trials (RCT’s) and real-life clinical THT has also been compared to a twice daily practice. The latter throws out surprises in that 30/70 insulin mixture and metformin over 24 weeks circumstances arise when THT is employed through in 188 patients, again with baseline HbA1c > 8.0 per sheer lack of choice (eg a complete refusal to start cent.6 There were comparable improvements in
insulin) but result in a much greater improvement both HbA1c and FBG, although only about one in glycaemic control than anticipated.
third of either group achieved an HbA1c < 7.0 per One recent placebo-controlled RCT with rosiglita- cent. About 10 per cent of the THT group needed zone in THT was undertaken in 837 patients with a to switch to insulin, and overall 16.3 per cent were Diabetes Update Winter 2004
Clinical practice
not receiving THT at the end of the study. On the 2005. Back in the ‘here and now’, though, THT with other hand, this means that 83.7 per cent were still a glitazone as a third agent is a useful therapeutic option in the asymptomatic patient with sub-optimal glycaemic control, particularly if there is a THT with sulphonylurea as the
reluctance to start insulin. Some patients have now third agent
achieved tight glycaemic control on THT for up to Thus far this article has assumed that the glitazone is added as the third agent in THT. On the basis of the UKPDS, metformin is the first-line OHA in Further information
overweight and obese patients with Type 2 For Diabetes UK’s own position statement on diabetes. However, many clinicians are now adding a glitazones, visit the information centre pages of glitazone as the second-line OHA after metformin, and this has the advantage of achieving or Glitazones are the first clinical topic covered in a maintaining tight glycaemic control without new series of position papers from the Association hypoglycaemia compared to sulphonylureas.
of British Clinical Diabeteologists. See Practical Furthermore, there are improvements in multiple Diabetes International (2004) 21 (7): 1-3. cardiovascular parameters, which are being investigated by large long-term studies looking atcardiovascular outcomes.
If these studies demonstrate cardiovascular References
benefits, then it is likely that the glitazones will 1. NICE Technology Appraisal 63 (August 2003).
become the conventional second-line agent after Guidance on the use of glitazones for the metformin in Type 2 diabetes. However, the issues regarding THT will become even more pertinent, as it is likely that most clinicians will wish to 2. Jones N, et al. Rosiglitazone in combination
introduce a sulphonylurea at this stage, rather than with glibenclamide plus metformin is effective go directly to insulin. This is yet another reason and well tolerated in Type 2 diabetic patients.
why it would be helpful for the pharmaceutical Diabetologia (2001); 44 (suppl 1): A235.
companies involved to obtain a THT licence in 3. Pan C, et al. The efficacy and safety of
pioglitazone hydrochloride in combination with A recent study compared glimepiride as the sulphonylureas and metformin in the treatment third agent in THT against placebo in 168 patients of Type 2 diabetes: a 12-week randomised multi-centre placebo-controlled parallel study metformin and glitazone therapy over six months.
(article in Chinese). Zhonghua Nei Ke Za Zhi Not surprisingly, glimepiride resulted in a significant improvement in HbA1c (- 1.31 per cent vs – 0.33 per cent), and the final mean HbA 4. Rosenstock J, et al. Triple therapy in Type 2
the glimepiride THT group was an impressive 6.83 diabetes: benefits of insulin glargine over rosigli- per cent. Furthermore, 62.2 per cent achieved a sulphonulurea plus metformin in insulin naïve was one severe hypoglycaemic episode in the patients. Diabetologia (2004); 47 (suppl 1): A57.
glimepiride group.7
5. Aljabri K, Kozak SE, Thompson DM. Addition
of pioglitazone or bedtime insulin to maximal
Conclusions
doses of sulphonylurea and metformin in Type 2 THT remains a controversial issue, partly because diabetes patients with poor glucose control: a of the limitations of the current licence and partly prospective, randomised trial. Am J Med (2004); because of guidance thus far by NICE. Nevertheless, it has been demonstrated to be effective in both 6. Schwartz S, et al. Insulin 70/30 mix plus
placebo-controlled and active comparator trials metformin versus triple oral therapy in the over six months. Indeed, these results have treatment of Type 2 diabetes after failure of two surprised many clinicians as these studies were oral drugs: efficacy, safety and cost analysis.
using glitazones later in the course of Type 2 Diabetes Care (2003); 26: 2238-2243.
diabetes, when they are less likely to be effective.
THT itself may well evolve to incorporate a 7. Roberts VL, et al. Efficacy and safety of
sulphonylurea as the third agent after metformin glimepiride in patients with Type 2 diabetes and a glitazone, but this will only become uncontrolled on metformin / thiazolidinedione routine practice after long-term trials with hard combination therapy. Diabetes (2004); 53 (suppl end-points have demonstrated beneficial results.
The first such data is likely to be available in late Diabetes Update Winter 2004

Source: http://www.avondiabetes.nhs.uk/FAQ/A%20Third%20Way.pdf