www.upmc-biosecurity.org www.upmc-cbn.org Wednesday, 4-29-09 | 11:00 AM FDA Emergency Use Authorizations (EUAs): What Are They and What Do By Brooke Courtney, JD, MPH What Do the Swine Flu EUAs Mean for the Current Outbreak? On April 27, 2009, the U.S. Food and Drug Administration (FDA) issued Emergency Use Authorizations (EUAs) for the swine flu outbreak in response to req
Diabetes politics………………………… 4 Philosophy and aims of care.…
Assessment of control
Treatment of type 2 diabetes
Preliminary observations………………… Overview of treatment…………………… Thiazoledinediones………………………. Prandial glucose regulators………………. Insulin and tablet combinations…………. Alternatives to insulin in poorly controlled patients 16 Concordance in type 2 diabetes………….
Treatment of cardiovascular risk factors
Education and lifestyle
Insulin-treated patients - practical notes
Types of insulin.……
Patient glucose monitoring
Home blood glucose monitoring…………
Urine testing……………………………. Notes on community clinics
Shared care philosophy………………….
The new diagnostic criteria. .
This book was originally written in June 1990 to provide a guide to the management of diabetic patients in the community, especially those with type 2 diabetes, who comprise the majority of patients. Since then it has been updated on three occasions, in close collaboration with many friends and colleagues from Ealing Hospital and elsewhere, to form a practical guide to diabetes care for colleagues working in West London. These are exciting times in diabetes care ! Remarkable advances have taken place since the last edition in 1999, necessitating extensive revision of this book. There are several new hypoglycaemic treatments - both tablets and insulin - providing new choices in treatment, although we have little comparative data to support the different options, so our guidelines need to be flexible and not too dogmatic. There are some new drugs for lowering lipids and blood pressure, and crucial new landmark studies defining their central role in treatment. Cardiovascular disease is by far the most important threat to our patients, and we now know that we can reduce this by up to 50% with intensive reduction of blood pressure and lipids. We can thus prolong the lives of thousands of local patients living locally by simple tablet treatments. This is both exciting and daunting – we have a huge potential to benefit patients, but there is a mountain of work out there, not least in persuading our patients to take their tablets ! The single most important factor in diabetes care is enthusiasm. A good diabetes service will soon produce its own momentum, and be immensely rewarding to everyone concerned - both patients and staff. However, it needs an abundance of energy, commitment, and a genuine concern for the welfare of people with diabetes. There are many other published guides to diabetes care, but they differ on various important aspects of management. This is because there is often no hard scientific information on which to base treatment - diabetes care remains as much an art as a science ! We have therefore produced this revised document to reflect our current local policies. It is intended to be a wholly practical guide, and we hope that it will provide answers to most problems which arise in diabetes care within the community - and within hospital clinics as well. Please ensure that it is readily available for reference. We hope that you find it useful and informative. Hugh Mather Consultant Physician DIABETES POLITICS
Diabetes care has been heavily ‘politicized’ in recent years, by 3 important initiatives, each designed
to improve standards of care.
Firstly, the National Service Framework (NSF) has given a whole new impetus, encouraging us
to review all aspects of our work and to establish local networks to improve standards. It is an
important and welcome initiative, although it comes with little extra funding. The most specific
targets relate to the construction of practice registers by *** and the introduction of digital retinal
screening by ****.
Secondly, the National Institute for Clinical Excellence (NICE) has produced detailed
guidelines on diabetes care, recommending many treatment targets. Unfortunately, they are rather
idiosyncratic. In the case of glycaemic control, they are over-optimistic and unrealistic, and in the
case of lipids and blood pressure they seem insufficiently rigorous. Most diabetologists regard them
with scepticism despite their ‘official’ status, but they are useful guidelines to therapy rather than rigid
Thirdly, the new GMS contract will provide a real incentive to primary care colleagues to improve
control and reduce cardiovascular risk, although the effort in collecting all the necessary data will be
considerable. Extra payments will be made for achieving clinical quality indicators, and of the total
of 550 points, 99 (18%) are for diabetes, with payments of about £2500 per GP in 2004/5 and
£4000 in 2005/6. The top indicators are concerned with control of glycaemic, blood pressure and
lipids, and are as follows.
55% of diabetic patients - last BP ≤ 145/85 17 points
50% - last HbA1c ≤7.4%
60% - last total cholesterol ≤5.0 mmol/l 90% - smokers offered advice or referral 90% - retinal screening in last 15 months 5 points etc. etc. Thus the total points are awarded for control of glucose, BP and cholesterol are 27,17 and 6 respectively! Points will still be awarded if GPs achieve a lower percentage cover, subject to a minimum of 25% cover. Three points each are awarded for several other targets, including achieving 90% targets in the recording of BMI, smoking status (in smokers), HbA1c, peripheral pulses, neuropathy testing, serum creatinine recorded, blood pressure, total cholesterol – all within the last 15 months. The disproportionate attention to glucose control, with 2 threshold levels at 7.4 and 10.0 %, and the large number of points they attract, may have considerable practical consequences. The lower threshold can be achieved by more vigorous early treatment of type 2 diabetes with combination therapy, and by screening more assiduously to detect ‘milder’ cases and thus boost the numbers of patients with lower HbA1c values – the best argument for screening advanced so far ! Patients above the higher threshold certainly justify a trial of insulin, although this may not improve their control. The target for total cholesterol conflicts with the NICE guidelines, which state that the decision re primary treatment should depend on overall cardiovascular risk rather than on the cholesterol value alone – the GMS target will undoubtedly encourage more liberal prescription of statins. PHILOSOPHY AND AIMS OF CARE IN TYPE 2 DIABETES
The main threat to patients with type 2 diabetes is from macro-vascular disease. The prevalence of ischaemic heart disease is an alarming 2-3 times higher than in non-diabetic subjects, and peripheral vascular disease and strokes are also increased. Up to 80% of patients will die from circulatory diseases. This is by far the most important consequence of type 2 diabetes. Patients are also susceptible to micro-vascular complications, namely retinopathy, nephropathy and neuropathy, but perhaps to a lesser extent than in type 1 diabetes. The Diabetes Control and Complications Trial (DCCT) and the United Kingdom Prospective Diabetes Study (UKPDS) have shown that tight glycaemic control decreases micro-vascular complications in both types of diabetes. However, there is still no rock-solid evidence that improving glucose control will decrease either mortality or morbidity from ischaemic heart disease. By contrast, the benefits of treating hypertension, hyperlipidaemia and other risk factors are truly amazing, with studies showing a reduction of 40-50% in cardiovascular events. Thus diabetes care is not synonymous with glucose control - it requires a much broader approach, especially in type 2 patients - with special emphasis on reducing cardiovascular risk above all else. The aims of care of type 2 diabetes can be summarised as follows: 1. To control blood glucose levels as tightly as possible, so that the patient has no symptoms due to hyperglycaemia or hypoglycaemia, and the risk of microvascular complications is reduced. 2. To screen for, and treat when necessary, diabetic complications, and risk factors for vascular disease, with special emphasis on treating hypertension, giving aspirin and/or lipid-lowering therapy, stopping patients smoking, and screening for retinopathy, early foot problems and nephropathy. 3. To educate patients to cope practically and psychologically with the diabetic life - we aim to produce happy and informed patients ! ASSESSMENT OF CONTROL
Whenever a patient is seen, we have to decide whether glycaemic control is satisfactory. It is useful
to consider how these decisions are made. Five factors are usually taken into account - 1) the
HbA1c concentration, if a recent value is available, 2) home blood or urine test results, 3) the
presence or absence of symptoms, 4) the clinic blood and urine glucose, and 5) serial weights.
1. HbA1c estimations
These give an indication of average blood glucose concentrations prevailing about 4-6 weeks before
the test is performed. They are currently the ‘gold standard’ of glycaemic control, although they
reflect baseline glucose values more than post-prandial rises. They help to identify patients who
starve themselves to obtain a good glucose result in the clinic, but whose control is otherwise less
than ideal. Instant HbA1c readings are not yet available in most clinics, but patients are usually
encouraged to attend for a blood test a week before the clinic, to provide a current reading. By
NICE guidelines, HbA1C readings should be obtained twice yearly, even in stable well-controlled
patients. *** Hb A1c readings can be misleading in patients with haemoglobinopathies, as occurs in
10% of Afro-Caribbeans, and occasionally it can be useful to measure fructosamine, which is an
alternative measure of medium-term glucose control, although in practice the test is now rarely
2. Home blood glucose or urine test results
Always look through the patient's test results, and discuss them with the patient, to give appropriate
feedback. Patients using home blood glucose monitoring should aim to keep their 2 hour
postprandial results below 10 mmol/l, but should avoid readings below 5 mmol/l, especially if taking
sulphonylureas, because of the important risk of dangerous hypoglycaemia. In practice these targets
are difficult to achieve in many patients. Patients performing urine tests should aim for negative
results, because then they will avoid all symptoms of hyperglycaemia (nocturia, polyuria, thirst and
weight loss). These generally occur only when urine tests are persistently 2% - patients with tests of
up to 1% are usually symptom-free. Many poorly controlled older patients with high blood levels
have no symptoms because they have a high renal threshold and thus have little or no glycosuria.
3. Presence of hyperglycaemic or hypoglycaemic symptoms.
The most useful indicator of poor control is nocturia. If patients have nocturia, ask them to test their
urine when they get up in the night, and also see whether they have glycosuria in the clinic. If this is
2%, their nocturia probably relates to poor control. However, if nocturnal tests are negative, it
implies that diabetes is not responsible for the nocturia, and it is more likely to be due to other
causes, notably prostatism, renal impairment or diuretic therapy. Other symptoms of poor control
are thirst, undue tiredness or lethargy.
It is important to enquire for symptoms of hypoglycaemia in patients on sulphonylureas or insulin.
Ask whether they feel unduly hungry, shaky or 'sweaty', particularly before lunch or the evening
meal. Another important symptom is morning headache, which often follows nocturnal
hypoglycaemia, and is a useful clinical indicator of over-control. Patients on insulin who drive need
to be asked very carefully about their hypos - whether they have warning symptoms, whether their
hypos have ever been disabling and whether they check their blood glucose before driving.
Fortunately, severe hypos are relatively uncommon in type 2 patients on insulin, presumably because
of their insulin resistance.
4. Clinic blood and urine glucose concentrations.
The fasting blood glucose (FBG) is a reasonably good guide to control in patients who are not
receiving insulin, because it is fairly constant from day to day. It is of much less value in insulin
treated patients. Thus clinics are ideally held in the morning, and patients not on insulin should be
asked to come fasting and to omit their morning dose of hypoglycaemic drugs – although they must
continue to take their other tablets, particularly their anti-hypertensive medication. Random (post-
prandial) glucose values are less useful, although by asking the time of the last meal and assessing
whether this was a light snack or a heavy meal, one can hazard a rough guess at the fasting value.
Fasting samples are less essential if a current HbA1C reading is available. The presence or
absence of glycosuria is useful in interpreting the significance of any polyuria or nocturia.
Serial weight readings are very helpful in assessing control, and adjusting therapy. They usually
remain remarkably constant, from year to year, even when diabetic control is only moderate. A
sudden marked change in weight always requires an adequate explanation.
A decrease in weight usually indicates successful dieting. It might suggest poor control, but this is
unlikely unless blood glucose and HbA1c values are very high and the patient is markedly
symptomatic. If weight suddenly falls, for no obvious reason, this may suggest an unrelated cause,
particularly malignant disease or thyrotoxicosis, and needs investigation.
An increase in weight may indicate that patients are not adhering to their diet. It may also show that
they are being overtreated with sulphonylureas or insulin, especially if their blood glucose and
HbA1c values are normal or only slightly elevated, and if they have hypoglycaemic symptoms.
Another common cause is congestive cardiac failure or nephrotic syndrome- look for ankle oedema
THE TREATMENT OF TYPE 2 DIABETES
Evidence from outcome studies – which is the best treatment ?
The only good evidence comes from the UKPDS, using the older treatments, and this showed firstly
that insulin and sulphonylureas produced similar effects on both control and complications (ie insulin
didn’t do any better than sulphonylureas) and secondly that metformin monotherapy showed
significant cardiovascular benefit in obese patients, although this was counter-balanced by an
unexpected adverse impact when added to sulphonylureas in poorly controlled patients.
Nevertheless, metformin is now the initial drug of choice for most patients, and has overtaken
sulphonylureas in this respect. Our choice of drugs has recently been expanded by the introduction
of the thiazoledinediones (rosiglitazone, pioglitazone) and the prandial glucose regulators
(repaglinide and nateglinide). They have promising theoretical advantages over older treatments,
but there are no long-term outcome studies, so we don’t yet know whether they are better or not.
There are several large studies using the newer drugs in progress, and they may alter our
management when the results are available. Meanwhile, we have to adopt a flexible ‘best guess’
approach. We simply don’t know which treatment is best !
Targets for glycaemic control
We aim to improve glucose levels as much as possible, both to abolish symptoms and to decrease
susceptibility to complications. But what is an acceptable degree of control ? We need to be
realistic ! The NICE guidelines specify a target HbA1C of 6.5-7.5%, but this is simply unachievable
in many patients, particularly in those who are overweight and have been diabetic for many years.
Our current treatments simply aren’t good enough ! The GMS contract provides incentives to
optimise control, with liberal quality payments depending on the proportions of patients at or below
thresholds of either 7.4% or 10.0%. Yet the UKPDS showed that control deteriorates inexorably
with time, regardless of type of treatment – and insulin treatment doesn’t prevent this deterioration.
It is an inherent feature of type 2 diabetes – and not necessarily due to poor concordance or
inadequate treatment. The average HbA1c in virtually all hospital clinics is around 8.5%. So it is
important not to become over-awed by published 'targets' for glucose control. Its importance may
be overemphasised, since the link with macrovascular complications is still unproven, and even the
reduction in microvascular endpoints in the UKPDS with tighter control was small and unimpressive.
More aggressive early treatment
There is a new vogue to treat patients more aggressively in the early stages of diabetes, in the hope
of preventing the deterioration in control found in UKPDS. Thus, some colleagues recommend
giving patients 2 drugs, or switching them to insulin, soon after diagnosis. This fits in well with the
targets set down by NICE and the GMS contract, and seems a sensible option, at least as far as
increasing the dose of tablets, and progressing from one type of oral therapy to two. The difficulty
comes in deciding when a trial of insulin is indicated in individual patients.
Tighter control versus increasing body weight
The influence of treatment on body weight is an important practical consideration, since most
patients are overweight. All treatments (even Metformin) tend to increase weight, because of the
reduction in glycosuria associated with improved control. This is especially so with insulin, but also
occurs with sulphonylureas and TZDs (to an equivalent degree), and (less so) with metformin or
acarbose. No-one knows how to treat the typical patient with a HbA1C of 9% and a BMI of 35
on maximum doses of metformin and a sulphonylurea ! Treatment with insulin will lead to an
increase in weight, may not improve control, may worsen cardiovascular risk factors, and there is
little evidence that it will extend the duration or quality of life. Often a compromise has to be
struck, between improving glycaemic control (usually with insulin), and a further increase in weight,
and a worsening of other adverse cardiovascular risk factors. The individual patient needs to be
involved in these difficult treatment decisions.
OVERVIEW OF TREATMENT – A STEP-WISE APPROACH…
The traditional treatment of type 2 diabetes is along a step-wise approach, and can be summarized
in note form as follows:
1. Diet alone – provided the patient is not too symptomatic – fasting blood glucose below 16
mmol/l – and scope from dietary history for improvement. Duration debatable – 3 months
traditionally, in practice 1 month probably sufficient.
2. Tablet monotherapy – if HbA1C > 7.0% - required for most patients. Metformin now the
drug of choice for both overweight and normal weight patients, although a sulphonylurea or prandial
glucose regulator might be used in the latter. Avoid in renal impairment or severe
gastroenterological problems. Initial dose 500 mg bd with food – increase gradually to 1 gram bd.
3. Two types of tablets - if HbA1C still above 7.5% despite maximum doses of single drug, add a
second drug – either a sulphonylurea (gliclazide 40 mg mane) (if on metformin) or metformin (if on a
sulphonylurea). A TZD (rosiglitazone 4 mg or pioglitazone) might be used instead of either -
theoretical advantages but more expensive, not supported by NICE and contraindicated in heart
failure. Build up the dose of second drug to maximum.
4. Next step – the big problem ! If HbA1C still ‘unacceptable’ (difficult to define – currently
HbA1C >8*%) consider trial of insulin – depending on presence of symptoms, degree of obesity,
age, ability to cope and willingness to start insulin etc. Bedtime long-acting insulin usually added to
However if patient unwilling, or unable, or unlikely to benefit (eg BMI >35) consider alternatives – triple therapy – metformin & sulphonylurea & TZD , or addition of acarbose, or trial of orlistat, or accept current control & concentrate on cardiovascular risk profile ! The diet required for type 2 patients is simply the same 'prudent' or healthy diet that we should all be eating, whether diabetic or not ! The main aims are to reduce fat and sugar consumption, to increase the intake of unrefined carbohydrate (wholemeal bread and cereals, rice, pasta, potatoes) and high-fibre foods (fruits and vegetables such as beans and lentils). Total calorie intake should be reduced if the patient is overweight. The most important single measure is the reduction in total and saturated fat intake, because this may help to prevent coronary artery disease, in addition to its effect on calorie intake. These principles can and must be taught by all of us, whether doctors, nurses or dietitians, and require constant reinforcement at each clinic visit. Most type 2 patients are overweight, and it is important to stress that even a modest reduction in weight may produce a marked beneficial effect on glucose control. All dietary advice must be practical, and should discuss specific foods rather than nutrients - e.g. potatoes not carbohydrate. It is worth quoting the basic advice given in one of our introductory leaflets for type 2 patients: "Diet is the most important part of your treatment. Being on a diet does NOT mean starving yourself ! It just means choosing healthy food, which we should all eat, whether diabetic or not. Follow these simple rules - they are very important : 1. Eat less fat - this is the single most important rule - to control weight and reduce heart disease. Do not fry food - grill or boil instead. Use a low fat spread instead of margarine or butter. Use skimmed or semi-skimmed milk, not full fat. Use leaner meats, and check food labels for large amounts of hidden fat. 2. Eat some starchy food with each meal – pasta, rice, bread, chapattis, potatoes or cereals. 3. Reduce sugar and sugar-containing foods, eg cakes, biscuits, sugary drinks. Use sweeteners, such as Canderal or Sweetex instead of sugar. 4. Eat more fruit and vegetables. Aim for at least 5 portions a day ! 5. Reduce the amount of salt in your cooking, and don't add salt at the table– this may raise blood pressure. 6. Avoid diabetic foods - these are fattening and expensive - but sugar-free or low-cal drinks are useful. 7. Drink alcohol in moderation. Never drink on an empty stomach. Have at least 2 alcohol-free days/week If you are overweight, losing weight - by reducing your fat, sugar and alcohol intake and taking regular exercise - will help your diabetes control, and make you feel much better. Even a little weight loss will make a big difference ! Buy some bathroom scales, weigh yourself regularly and record the results. It is important to be realistic about the limitations of dietary treatment. The problems of achieving long-term concordance are obvious, and relatively few type 2 patients remain controlled on diet alone. Most of us are all too aware, from personal experience, of the problems in adhering to any diet for any length of time ! Treatment of obese patients is especially difficult, and needs time and sympathy. Setting achievable goals is extremely important. Encourage patients to set their own targets, and record these for reference at their next visit. Some manage to lose weight initially, but the long term results are usually disappointing. If patients cannot lose weight, it is cruel to bully them incessantly ! Encourage them to ensure that their weight does not increase still further. Ask them to buy some bathroom scales, and to weigh themselves twice weekly. Patients with particular difficulties can be referred to the hospital clinic, or directly to the hospital or community dietitians. METFORMIN
Metformin is now the drug of choice for most patients who are inadequately controlled on diet alone, particularly those who are obese, since it may help weight reduction, or at least avoid any further weight increase. Its popularity is increasing because it is cheap, effective, safer than the sulphonylureas and may confer extra cardiovascular benefit. Curiously, it never causes hypoglycaemia - although it is a hypoglycaemic agent! It is absolutely contra-indicated in patients with severe renal impairment (plasma creatinine >140 µmol/l) because of the risk of lactic acidosis. Its use in the many patients with intermediate plasma creatinine values - between 120-140 µmol/l - is controversial, but the risks of lactic acidosis have been overemphasized, and most colleagues are now less concerned now about its use in this situation. It should also be avoided in patients with severe cardiac failure and hepatic failure, but it is probably safe in patients with less serious cardiac and liver problems – again the risks have been exaggerated. It is contraindicated in patients with significant colonic problems, such as inflammatory bowel disease, because of its well-known propensity to cause diarrhoea. Metformin (Glucophage) is available in tablets of 500 or 850 mg. The latter tablets are quite large – veritable ‘gob-stoppers’ - and may cause swallowing problems in elderly patients. Hence, most clinicians use 500mg tablets. The cost of 500 mg bd is £21/year. The important practical points are to start on a low dose (500 mg bd), to take the tablets with or after meals, and to build up the dose gradually, to a maximum of 1 gm bd. This will reduce the incidence of gastrointestinal side effects – mainly diarrhoea but sometimes dyspepsia as well - which otherwise affect about 25% of patients. If they occur, try a reduction in dosage, rather than stopping the drug altogether. With these precautions, most patients tolerate the drug well. Further increases in total daily dosage beyond 2 grams per day have no additional benefit. Its effect on glycaemic control is roughly equivalent to that of sulphonylureas, usually reducing HbA1C levels by 1-2% in type 2 patients. The full impact is delayed for a week or more, so if a very rapid effect is needed, as in some symptomatic patients, sulphonylureas are a more suitable choice. Metformin should be omitted for 2 days after X-ray procedures using intravenous contrast media because of the remote risk of renal failure and lactic acidosis. SULPHONYLUREAS
These drugs are still a reasonable choice as first-line treatments for patients who are not overweight,
if diet alone fails to achieve adequate control. They are very effective in rapidly relieving symptoms
due to hyperglycaemia (with a quicker treatment response than metformin), and any previous
doubts about their long-term safety have been dispelled by the results of the UKPDS, which
showed that they are as beneficial to control and complications as insulin treatment. However they
generally lead to weight gain (less so than insulin, but equivalent to TZDs) so obesity is a relative
Always start with small doses (gliclazide 40 mg, glimepride 1 mg, or glipizide 2.5 mg). Some
patients are very sensitive to these doses. Patients can be advised to increase the dose if home
glucose readings remain higher than the target range or urine tests remain persistently positive after a
week. Their dose-response curves are remarkably flat, compared with metformin which is more
linear. Thus doubling the dose of a sulphonylurea (say from gliclazide 80 bd to 160 bd) may have
little or no additional effect, whereas doubling the dose of metformin will have more obvious impact.
Hypoglycaemia is the main side-effect, and can be severe or even fatal. Minor degrees of
hypoglycaemia are very common, and severe episodes not uncommon. Always enquire for
hypoglycaemic symptoms, especially feeling very hungry and slightly shaky and 'sweaty' before
lunch. If symptoms are present, especially if the weight is increasing and if blood glucose and Hb
A1c values are relatively low, the dose of sulphonylurea must be reduced. Many patients are
overtreated. It is unfair to ask obese patients to lose weight, and give tablets which make them
hungry ! They can be dangerous especially when patients become ill, and stop eating but continue to
take their drugs. Elderly patients with renal impairment are most at risk, and deaths still occur every
year - please take care. If patients become severely hypoglycaemic, they must be admitted to
hospital - not sent home from Accident and Emergency department - and given an IV glucose
infusion for at least 24 hours, because the tendency for the hypoglycaemia to recur extends over
Gliclazide is a reasonable first choice sulphonylurea. It is safe and effective, and is especially useful
in the elderly, and in those with renal impairment because it is safer in these circumstances than
glibenclamide, and less likely to cause hypoglycaemia. The initial dose is 40 mg mane (1/2 tablet),
and this can be increased to 80 mg, then to 80mg bd, and then increased to a maximum of 160 mg
twice daily. However, as stated above, the dose response curve is very flat, and the benefit from
increasing the dose from 80 mg bd to 160 mg bd is negligible in practice. Check cost of generic
gliclazide*** It can also be given in a once daily form as modified release (Diamicron MR), and this
may be a useful aid to improving concordance. Diamicron MR 30 mg mane is equivalent to
gliclazide 40mg bd, and costs £52/year. The dose can be increased to a maximum of 120mg mane
(4 tablets mane – equivalent to gliclazide 160 mg bd, and costing £208/year.
Glimepride also has the advantage of once-daily dosage. Its starting dose is 1 mg mane, and the
maximum dose is 6 mg. Glimepride (Amaryl) 1mg costs £57/year. Tablets of 1mg, 2mg, 3mg and
4mg are available.
Glibenclamide is possibly the most potent agent, but is also the most dangerous and often causes
mild hypoglycaemia before lunch, because of its shorter effective half-life. It is generally less
preferable than the other sulphonylureas. Patients already on the drug can usually continue on the
drug, but elderly patients (aged over 70 years), or those with any renal impairment (creatinine >120
µmol/l) should be changed to a safer agent, such as gliclazide.
The thiazolidenediones (TZDs) (rosiglitazone and pioglitazone) reduce insulin resistance, acting particularly on muscle and fat, and lower HbA1C in type 2 patients by about 1%, similar to (or slightly less than) sulphonylureas and metformin. They also improve other cardiovascular risk factors, with a small but useful BP lowering and a decrease in microalbuminuria. Thus it is hoped that they will decrease cardiovascular risk, in addition to improving glycaemic control. There are good theoretical reasons for this, but we await the outcome studies currently in progress to see whether this happens in practice. Their full effect takes six weeks - and thus the benefit on HbA1C (reflecting control about 6 weeks previously) is not seen for about three months. It is important to emphasize this delayed onset of action, otherwise patients may assume that the drugs are useless and stop them after a week or so! The main side effect of TZDs is fluid retention, and they are contraindicated in patients with overt or
incipient heart failure. Body weight increases, partly due to fluid retention and partly due to fat
deposition, to a similar degree to that seen with sulphonylureas, so marked obesity is also a relative
contra-indication. They are safe in renal impairment unless end-stage (plasma creatinine >
****µmol/l. Our worries about hepatotoxicity, following the withdrawal of triglitazone, have
proved unfounded, and although the guidelines state that liver function tests should be checked
regularly, this may be unnecessary.
TZDs are indicated firstly in patients on maximum doses of a sulphonylurea who need a further oral
agent, but cannot tolerate metformin, either because gut side-effects or renal impairment. Secondly,
they can be added to maximum doses of metformin, instead of adding a sulphonylurea, in patients
who require a second drug. Many colleagues prefer this approach on theoretical grounds,
although it is against the NICE guidelines and is more expensive. Thirdly, they are licensed for use
as mono therapy in patients who cannot tolerate metformin because of renal impairment or gut side
effect. This usage is not supported by NICE, but many colleagues feel that this is the most logical use of TZDs. They are not generally used in combination with insulin because of concerns of precipitating heart failure. There is no point trying to improve control in patients on metformin and a sulphonylurea by stopping either metformin or a sulphonylurea and replacing it with a TZD, because the latter is no more powerful and takes six weeks to have its impact. The two available TZDs are rosiglitazone (Avandia) and Pioglitazone (Actos), and their profiles are very similar. There are some minor benefits in the impact on lipids with pioglitazone but these are probably of little clinical significance. The once daily dosage is an important aid to concordance. Rosiglitazone is given as 4mg mane (cost £312/year) and can be increased to 8mg (£650/year) but there is relatively little extra effect with the higher *(and very expensive)** dose. Pioglitazone is given as 15mg mane (£312/year) and can be increased to 45mg (£444/year). A combination of metformin and rosiglitazone – Avandamet – has recently been introduced, as an aid to patient concordance. This comes in 2 strengths – 500 mg metformin combined with either 1mg or 2 mg rosiglitazone, and the starting dose is 1 tablet of the weaker strength twice daily (£358/year). ACARBOSE
This drug works by reducing the breakdown of complex carbohydrates in the gut, and has a useful effect in reducing post-prandial rises in blood glucose. It is safe and does not induce any weight gain, but the effect on glycaemic control is relatively slight effect (usually lowering HbA1c by about 0.5%) and its gastro-intestinal side-effects are often severe. The latter can be reduced by starting with a small single daily dose (25mg or half a tablet), taken with the first mouthful of a meal, increasing to 50 mg daily and then gradually (by 50 mg per week) to 50 mg t.d.s. Acarbose has become less popular in recent years, but it may still have a role in poorly controlled patients who are unwilling to start insulin therapy. If used with sulphonylureas, it can cause hypoglycaemia, and in this situation treatment with glucose rather than sucrose (sugar) is needed because Acarbose inhibits the breakdown of sugar to glucose in the gut. Acarbose (50mg tds) costs about £116/year. PRANDIAL GLUCOSE REGULATORS
These drugs comprise repaglinide (Novonorm) and nateglinide (Starlix), and act like quick-acting sulphonylureas, causing a brisk release of insulin. They are taken with each meal, and their rapid action means that the post meal glucose “spikes” are well controlled and the short duration renders them less likely to cause hypoglycaemia or weight gain than sulphonylureas. However, the HbA1C concentration (which generally reflects base line rather than postprandial glucose levels) is no better. Theoretically, control of glucose spikes might prevent long term complications but this remains unproven, in the absence of outcome studies. Repaglinide may be useful as monotherapy in T2D patients who are recently diagnosed and have an erratic meal pattern, perhaps with normal fasting blood glucose but elevated postprandial levels. Both drugs may be added to metformin to improve control. Their use requires a motivated patient to ensure that the drug is taken with each meal – most patients have concordance problems with once-daily drugs, let alone thrice daily ! They should not be used with a sulphonylurea since they act on the same receptor on the beta cells, and will have no extra effect. Repaglinide (Novonorm) 1mg before each meal costs £149/year, and nateglinide (‘Starlix’) 120 mg before each meal costs £293/year. Nateglinide is not licensed for use as monotherapy. The prandial glucose regulators occupy only a small niche in treatment currently, although this might alter if studies confirm the importance of control of glucose spikes in preventing complications. INDICATIONS FOR INSULIN TREATMENT
This a notoriously difficult 'grey area'. As stated earlier, no-one really knows how to treat poorly controlled type 2 patients, and there is no really solid evidence to suggest that insulin therapy will extend either the life-span or quality of life of asymptomatic type 2 patients. Many patients fare little better when switched to insulin, and simply put on more weight. Yet some colleagues are switching many more type 2 patients to insulin earlier in the natural history of the disease, perhaps reinforced by NICE and GMS targets. It is difficult to give precise criteria for a trial of insulin, but currently we rarely switch type 2 patients to insulin unless the HbA1C level is over 8.0%. Above this level, the decision is influenced by several factors. These include: (1) the presence of symptoms particularly nocturia, especially if associated with prostatic problems, polyuria and polydypsia. However, many poorly controlled patients have few or no hyperglycaemic symptoms. (2) the patient's weight markedly influences the decision. The case for insulin is much stronger in patients who are not overweight, especially if actually losing weight, because they will be less resistant to insulin, and more likely to benefit from therapy. However, if a patient is very obese, insulin therapy may not help. Control may not be improved, and weight will almost certainly increase further. (3) the ability to cope with insulin injections, which is influenced by age, family back-up, language ability and other factors, is vitally important. (4) the presence of complications such as neuropathy or retinopathy, which reinforce the need for tighter control. (5) the height of the HbA1c concentration. Patients over 10% justify a trial of insulin. Those with HbA1C levels between 8 and 10% will require consideration of all the above factors in order to make a decision. In many instances, we offer patients insulin for a trial period of 1-3 months, and then leave the choice of treatment largely to them. Some patients feel much improved on insulin, and continue indefinitely, whereas others beg to restart tablets! INSULIN AND TABLET COMBINATIONS
Until recently, patients were treated with either tablets or insulin (plus diet) – rarely both together. Now, combinations of insulin and tablets are increasingly used, particularly with use of bedtime intermediate or long-acting insulin in combination with day-time tablets, for poorly-controlled type 2 patients. A dose of 10-20 units of isophane, lente insulin or insulin glargine at bedtime controls glucose levels overnight, and the tablets maintain control during the day. The advantages of this regime, compared with switching to insulin entirely, are that smaller and fewer doses of insulin are needed, less weight gain occurs and the risk of nocturnal hypoglycaemia is slight. The once-daily injection also seems more acceptable to patients who may be reluctant to switch to twice-daily insulin. The dose of bedtime insulin can be adjusted depending on the fasting blood glucose on the following morning, and often needs to be increased. This regime is sometimes called ‘BIDS’ - Bed-time Insulin Daytime Sulphonylurea, although a combination of insulin with metformin may be preferable, especially in obese patients. It is probably unnecessary to continue both sulphonylureas and metformin in this situation. TZDs should not be used with insulin. ALTERNATIVES TO INSULIN IN POORLY CONTROLLED PATIENTS
An increasingly popular but ‘unofficial’ role is as ‘triple therapy’, added to the regimen of patients
who are poorly controlled on both metformin and a sulphonylurea but who adamantly refuse insulin
therapy, because of fear or lack of motivation. Although this is against both the drug’s license and
NICE, there are no good reasons why it should not be used in this way, and the prescription of
‘triple therapy’ is now widespread.
Some patients are terrified of insulin treatment, and refuse point blank to commence injections, even
with supportive counselling. In this situation it seems perverse and unjustified to ‘twist arms’ of
patients, especially since the evidence base for insulin treatment may be rather flimsy. Now of
course there is an alternative, in the use of ‘triple therapy’ – the addition of a TZD to both
sulphonylurea and metformin. This is strictly ‘off license’ and against NICE guidelines, but it is a
logical drug combination, carries no known safety risks, and avoids the psychological trauma of
insulin injections. This treatment strategy is now commonly used by most diabetologists, although
long term efficacy remains to be established. At least it provides another therapeutic option, in a
difficult clinical situation.
Another option is to try Orlistat (Xenical). This is a lipase inhibitor which interferes with fat
absorption producing malabsorption with liquid oily stools, which can lead to urgency and faecal
incontinence, particularly after a fatty meal. They induce weight loss primarily by acting as a
‘behaviour therapy’ to persuade patients not to eat sinful fatty meals! It is a safe therapy and is
licensed for diabetic patients whose BMI is >28. However, according to the NICE guidelines,
patients must lose 2.5 kg with diet and exercise in the month prior to starting orlistat, to show that
they really can adhere to a low fat diet. After starting the drug, they should lose 5% of their weight
in the first 12 weeks (when most weight loss takes place) and 10% in the first six months. If these
targets are not met, the drug should be withdrawn. Patients on orlistat needs supervision and dietary
support and this is usually provided in primary care.
Its place in long term diabetes treatment is unclear. It is currently only licensed for two years use,
and is usually not given for more than a year, but most patients regain their former weight when the
drug is stopped. It can certainly improve glycaemic control, reducing HbA1C by about 0.5% in
T2D, but whether a short course of orlistat will reduce vascular risk accruing over several decades is
unclear, and there are no outcome studies in diabetes. Nevertheless, treatment of grossly obese
T2D patients is notoriously difficult, and orlistat at least provides an alternative, safe treatment.
Perhaps we don’t use it sufficiently and colleagues in primary care might be encouraged to use it
Sibutramine is the other anti-obesity drug treatment. It is centrally acting, avoids the gastro-
intestinal side effects of Orlistat, and seems equally good in achieving weight loss and improving
control. Unfortunately, may can elevate blood pressure, and is contra indicated in ischaemic heart
disease and uncontrolled hypertension. This is a severe limitation, and probably prohibits its use in
diabetes, until more evidence is available.
CONCORDANCE IN TYPE 2 DIABETES
This is the most underrated problem in diabetes care. Many patients are currently receiving 6-10 different drugs – perhaps 2 for diabetes, 2-4 for hypertension, a statin and aspirin, possibly with other tablets for heart failure, obstructive airways disease or hypothyroidism. We all know how difficult it is to complete even short courses of antibiotics ourselves ! It requires exceptional motivation to take several tablets per day for years on end, for prognostic rather than symptomatic benefit. The DARTS study, from Dundee, showed that only one third of patients bothered to collect their diabetic tablets from pharmacies, and the proportion was even less among patients taking twice daily therapy, or on multiple drugs. We have turned a blind eye to this issue for years ! There is no single solution to concordance problems, but several measures might help. Firstly, once daily drugs should be used whenever possible. Thus the use of once daily sulphonylureas (either gliclazide MR or glimepride) may improve concordance, and a once daily metformin preparation (available in the USA) is sorely needed in Europe as well. Secondly, the use of drug combinations should be encouraged, and our purist objections to these should be set aside. Thirdly, and most importantly, we must educate our patients about the importance of their tablets in preventing unpleasant and life-threatening complications, to ensure their involvement and concordance. We need to reinforce these messages at every visit. Otherwise, our treatment regimes may perhaps be little more than expensive charades ! TREATMENT OF CARDIOVASCULAR RISK FACTORS
Measurement – some practical points
Measurement of blood pressure is as important as that of blood glucose, and should be performed
with similar care. Important practical points are firstly to the record the BP to within 2 mm Hg (not
to the nearest 10 mm Hg!), and secondly to use an appropriate size of cuff - if the arm
circumference exceeds 30 cm, in obese patients, a large cuff must be used. The
sphygmomanometer pressure must fall slowly, not rapidly. High readings should be confirmed after
resting for 5-10 minutes, and treatment should only be instituted after hypertension has been
demonstrated on at least 2, and preferably 3, separate occasions, at intervals of 1-2 weeks.
Standards of BP measurement are generally unsatisfactory - please take care and time, because
important treatment decisions are based on their results. In elderly patients, particularly those on
several drugs, it is necessary to test for postural hypotension by measuring the lying and standing BP,
waiting a full minute ***before taking the latter. 24 hour ambulatory BP monitoring may be useful
in patients with suspected ‘white coat hypertension’.
Target Blood pressure
The importance of BP control in diabetes care is absolutely crucial – certainly as important, if not
more important, than glucose control ! Hypertension needs to be treated more vigorously in
diabetes, because patients are more susceptible to target organ damage at any given BP level.
Several studies have shown dramatic benefit from modest reductions in BP. Lowering BP from
about 155/90 to 145/85 produces a 30% - 40% reduction in cardiovascular risk, regardless of the
drug used. The target BP in diabetic people should be 140/80. Paradoxically, the NICE guidelines
don’t recommend drug therapy unless the BP is 160/100 or higher, when overall cardiovascular risk
is <15%, but most colleagues would aim for 140/80 in everyone. The GMS contract awards points
for ensuring that 55% of diabetic patients have a BP ≤145/85, as a compromise value. Even more
aggressive lowering, down to 135/75, is recommended in patients with MA or proteinuria. The
target levels may be difficult to achieve in practice, and 3-4 or more drugs may be necessary. We
used to be less aggressive in treating hypertension in the elderly, but studies have shown impressive
benefits at virtually any age. However, control is difficult in elderly patients with systolic
hypertension - it may be impossible to lower systolic BP to 140 mm Hg without reducing diastolic
BP below 60mm Hg and this should be avoided – the coronary arteries fill in diastole !
General Measures and hypertension
Some non-pharmacological measures can help treatment. A reduction in salt intake, by avoiding it
either in cooking or on the table, produces a modest but useful decrease of about 5 mm Hg
(systolic) and 3 mm Hg (diastolic). Excess alcohol raises blood pressure and reduction of heavy
alcohol intake is important, as is weight loss in obese patients. These simple measures are useful
and important, and don’t receive the attention they deserve.
It probably does not matter unduly, because all main classes seem equally effective in reducing both
BP and (more importantly) cardiovascular risk. ACE inhibitors or A2 blockers are the drugs of
choice in patients with microalbuminuria or overt proteinuria, because of their additional
renoprotection benefit. However, the massive ALLHAT study showed that the benefits from
chlorthalidone (a thiazide diuretic), lisinopril and amlodipine were virtually identical – in fact
chlorthalidone was marginally superior to lisinopril. In practice, many patients require 2 or more
drugs, so our basic strategy is to use small doses of several drugs, in an attempt to increase
effectiveness and reduce side effects. Patients below 55 tend to have ‘high renin’ hypertension and
may respond better to ACE inhibitors or beta blockers, whereas older patients tend to have more
volume dependent, low renin hypertension and respond better to calcium channel blockers or
diuretics. This gives rise to the ABCD rule, which states that A(ce inhibitors) or B(eta blockers)
work better with either C(alcium channel blockers) or D(iuretics), than with another similar drug. It
is necessary to wait about 2-4 weeks to see the full effect of a drug, before giving additional
treatment – treatment shouldn’t be stepped up too quickly.
Thiazides are definitely drugs of choice in diabetes, following ALLHAT and other studies. No
more than 2.5mg bendrofluazide should be given – this gives maximum effect on BP and minimizes
effects on glucose and lipids – and on the drug budget, costing £* /year ! A small proportion may
develop erectile dysfunction and will need other therapy. Thiazides are contra-indicated in gout,
and should not usually be given to patients already on loop diuretics, such as frusemide, because of
the synergistic diuretic effect. They are relatively ineffective in renal impairment, when hypertension
is more volume dependent and frusemide is a better choice.
ACE inhibitors are drugs of choice in patients with proteinuria or MA, and in those who have heart
failure or a history of myocardial infarction. They are relatively ineffective when used alone in Afro-
Caribbeans, although they still work in these patients in combination with other drugs. They are
markedly harmful to the foetus, and thus are contra-indicated in premenopausal women unless they
have reliable contraception. There is a slight but important risk of severe renal impairment in
patients with renovascular problems. Thus urea and electrolytes must be checked after 7-10 days
therapy. A modest rise in plasma creatinine is commonly observed, but an elevation of 30% or
more requires the drug to be stopped. About 15% of patients develop a troublesome cough
with ACE inhibitors, and can be switched to an A2 blocker. All the once-daily preparations -
lisinopril, ramipril, perindopril etc - are probably equally suitable. Lisinopril (‘Zestril’) 5mg costs
£102/year, ramipril (‘Tritace’) 2.5mg costs £107/year, perindopril (‘Coversyl’) 2mg £139/year.
The angiotensin II antagonists – or ‘A2 blockers’, such as irbesartan, losartan and valsartan are
gaining popularity because of their freedom from the ‘ACE’ cough, and because of their
renoprotective effect in type 2 patients with MA and proteinuria, although the evidence for benefit in
patients with CCF or IHD is not as strong as with ACE inhibitors. They may also cause renal
impairment in the presence of reno-vascular insufficiency, and thus urea and electrolytes need to be
checked a week after commencing therapy. They can usefully be combined with ACE inhibitors, to
produce extra BP lowering and reduction of MA. Irbesartan (‘Aprovel’) 150mg costs £214/year,
losartan 50mg costs £224/year and valsartan (‘Diovan’) 80mg costs £205/year,
Beta-blockers (given as atenolol) achieved similarly impressive results to ACE inhibitors in the
UKPDS, and are also drugs of choice. They must be avoided in patients with asthma or obstructive
airways disease, but concerns about the risk of hypoglycaemia are rarely a problem in practice.
They are especially important in the secondary prevention of myocardial infarction. They can cause
slight lethargy, cold extremities or erectile dysfunction, and thus may be less preferable as
monotherapy to the newer agents. Atenolol (Tenormin) 50 mg costs £74/year.
Calcium antagonists (or calcium channel blockers) are effective anti-hypertensive agents, and
the ALLHAT study showed that amlodipine produced equivalent benefit on vascular events to
lisinopril and chlorthalidone. Short acting dihydropyridine drugs such as nifedipine should no longer
be used. The predisposition to ankle oedema is often a limiting factor in the use of these drugs, but
their safety in asthma and heart failure may be valuable. Diltiazem and verapamil are less suitable
than the dihydopyridine drugs (amlodipine, felodipine etc) because of their tendency to cause heart
failure and bradycardia, and lack of evidence of long-term benefit. Amlodipine (‘Istin’) 5mg (cost
£170/year) or felodipine ((‘Plendil’) 5mg (cost £116/year) are suitable drugs.
Doxasozin has fallen from favour because it was withdrawn from the ALLHAT study because of a
slight predisposition to heart failure. However, the significance of this result is controversial, and
most colleagues continue to use it, although incipient or overt heart failure are contraindications. It
is effective, relatively free of side-effects, and useful when other drugs are contraindicated, such as in
peripheral vascular disease. It is also useful in patients with prostatic outflow obstruction, since it
benefits this as well. It can cause ankle oedema, in common with most vasodilator drugs. The
initial dose is 1 mg daily, increasing after 1-2 weeks to 2 mg daily. The dose needs to be titrated
carefully upwards, over the first few weeks, to a maximum of 8 mg daily. The cost of 1 mg
(‘Cardura’) is £138/year, and a useful long-acting form (Cardura XL) is available at 4mg (cost
Concordance with BP therapy
Concordance is a huge problem, but can be improved by patient empowerment. Patients should
know their BP reading, their target BP, and should be taught to take responsibility for their BP
control. Never tell a patient that their BP is ‘not too bad’ or ‘only up a bit’ – give them the figures!
Emphasise to all patients the need for life-long therapy, and the importance of BP control in
avoiding heart-attacks and strokes. Encourage patients to buy their own BP meter, preferably one
which measures at the elbow rather than the wrist. Home readings are usually lower than those at
clinics by about 12 mm Hg systolic / 7 mm diastolic, and this difference needs to be taken into
account when assessing the significance of home readings. All the key studies on BP lowering have used clinic readings so these should still form the basis of our targets and treatment. HYPERLIPIDAEMIA
The introduction of statins to diabetes care is perhaps the most important advance since the
discovery of insulin, and their impact may be as large as that of antibiotics in the 1940’s. They
reduce cardiovascular disease – the main killer - by 30-40%, and this benefit is over and above that
achieved by other treatments, notably BP lowering and aspirin. Furthermore, they have a
remarkably good safety record. The only limitation to their use in virtually all middle-aged and
elderly diabetic people is their high cost. However, the patent for simvastatin has now expired and
that of pravastatin will expire in August 2004, so the cost of the older statins should fall rapidly in
the near future.
So who should receive statins for primary prevention?
This is currently very confusing ! The landmark Heart Protection Study concluded that effectively all
groups of diabetic patients between 40 and 80 with a total cholesterol >3.5 mmol/l would benefit
from statin therapy (simvastatin 40 mg in the study), and the EHHHA expert lipid group supported
this policy. If statins were as cheap as aspirin, there is little doubt that all diabetic patients over 40
(except those with liver disease) would receive them. However, they are currently very expensive
and NICE have recommend statins only if the total cholesterol is ≥5.0 mmol/l or more (or
triglyceride is ≥2.3 mmol/l) and the patient’s combined cardiovascular risk score over 10 years is
15% or more. This policy is flawed because the scores are derived from the Framingham study,
which had relatively few diabetes people, and which did not allow for the effects of
microalbuminuria, family history and ethnicity. Being Asian, for example, probably justifies a 50%
increase in risk score. In addition, it is illogical to define a cut-off point in total cholesterol – one
component of cardiovascular risk - rather than considering overall combined risk. By contrast, the
GMS contract awards points for ensuring that 60% of all patients have a total cholesterol ≤5.0
mmol/l – regardless of their cardiovascular risk scores.
Nevertheless, we are obliged to adhere in principle to these NICE guidelines. This may deny statins
to two at-risk groups. These are, firstly, patients whose total cholesterol is <5 mmol/l, but whose
cardiovascular risk is 15% or more – some colleagues would give statins regardless. The second
group comprise patients whose cholesterol is ≥5.0 mmol/l, but whose risk score is under 15% -
here the case for statins is less robust here, but it might be reasonable to take a lower threshold of
10% in Asians patients because of their increased risk, especially since the GMS contract wishes us
to do so ! The use of older, potentially cheaper statins might perhaps compensate for a more
liberal prescription policy. There is little evidence to support the use of statins in primary prevention
in patients aged over 80, although the absolute risk (and thus the potential benefit from therapy) may
be higher, and many colleagues prescribe them regardless.
The benefits from statins are probably class effects, common to all, although this is not strictly
proven. The key studies have used simvastatin and pravastatin, and these drugs have the best
evidence of both efficacy and safety. Furthermore they may become cheaper with the expiry of
their patents. Atorvastatin is slightly more powerful than simvastatin but the evidence of outcome
benefit is currently less good (although more studies are imminent) and its patent will not expire until
2009, so ultimately it may prove more expensive. Another statin, cerivastatin, was withdrawn
because of an unexpectedly high incidence of rhabdomyolysis, so the safety profile of newer statins
cannot be taken for granted. The newly introduced rosuvastatin is the most powerful of all and may
be valuable in patients with resistant hyperlipidaemia but as yet there are no long term outcome
studies and less safety data. Pravastatin is preferable if there is a risk of drug interactions
(particularly in renal transplant patients) and may be less likely to cause myopathy, particularly in
combination with fibrates. Thus a reasonable plan might be to use either simvastatin or pravastatin
as a first line drug and switch to atorvastatin, and then finally to rosuvastatin if the lipid lowering
effect is insufficient. Another option is to add in ezetimibe (see below).
*** put in data from Stellar***
Which dose of statin?
The dose response curve to statin therapy is relatively flat, with little further lipid lowering from
increasing doses. Colleagues with a ‘purist’ streak would advocate the doses used in the landmark
studies, so 40mg of simvastatin, as used in the Heart Protection Study, is a popular choice, because
of its know efficacy and safety. Although a fixed dose was used, in this study, with no ‘fine tuning’
to achieve a target level, most colleagues continue to adjust the dosage, starting with simvastatin 10
mg nocte and increasing as necessary. Further studies are in progress to see whether extra benefit
can be obtained by more aggressive lipid lowering. Most types of statins need to be taken at night,
since this is the time of maximum cholesterol synthesis but this may lead to problems with
concordance. The need for night-time dosage may be less important with atorvastatin and
The current costs of statins are uncannily similar, with 10mg of proprietary simvastatin, atorvastatin
and rosuvastatin all costing an identical £235/year. The cost of simvastatin 20mg, 40mg or 80 mg
daily is identical to that of rosuvastatin 20mg and 40mg or atorvastatin 20, 40 and 80 mg – all the
different statins and dosages retail at about £387/year. This bizarre pricing structure at least allows
us to use higher doses without worrying about its effect on our drug budgets !
Problems with liver function tests and CPK levels
Many diabetic patients have minor elevations of liver enzymes, usually due to fatty infiltration.
Although statins are stated to be hepatotoxic, severe hepatic necrosis is excessively rare, and this
risk is usually far outweighed by the potential benefit in reducing vascular disease. Thus, we have a
liberal approach to giving statins even when hepatocellular enzymes are elevated, although it is
difficult to give precise figures. Nevertheless, statins are best avoided in patients with overt liver
disease, such as alcoholic cirrhosis. Serial LFTs may not be routinely necessary. Myalgia was
reported by over 30% of patients in the Heart Protection Study, but the prevalence was similar in
both placebo and statin groups, so patients complaining of this symptom may be reading the
package insert too assiduously. Severe rhabdomyolysis is very rare and usually occurs either in
renal failure or in combination therapy with fibrates, It is worth checking the creatine phosphokinase
(CPK) level if patients complain of myalgia, and statins should be stopped if there is more than a 5-
fold rise. It is worth remembering that Afro Caribbeans have a markedly raised normal range for
***check when need to stop statins temporarily when on cipro etc ***
This new drug inhibits cholesterol absorption from the gut, and acts as a useful adjunct to statins if
these drugs don’t lower lipid levels sufficiently. It might also be used as monotherapy in patients
who can’t tolerate statins. It is contraindicated in established liver disease, but otherwise seems
safe and useful, although as yet there is little evidence of long-term efficacy in reducing vascular
disease. Ezetimibe (‘Ezetrol’) is given in a dose of 10 mg daily, cost £343/year.
Fibrates are well suited for use in diabetes, because they tackle the classic lipid abnormalities of
elevated triglycerides and lowered HDL. However, trial data so far is limited, and confined largely
to secondary rather than primary prevention, compared with the huge body of evidence on the
benefits of statins. The significance of mildly raised triglyceride levels, and the benefits of lowering
them, are still unclear. The data suggest that fibrates are indeed beneficial, but unless patients have
a grossly raised triglyceride (say over 5 mmol/l) and relatively normal cholesterol, statins remain the
drugs of choice, especially since they also lower triglyceride levels. An obvious option is to add
fibrates to statins, but there are concerns re the risk of rhabdomyolysis, following the fatalities from
combined use of cerevastatin and gemfibrozil. These dangers may be less with fenofibrate and with
pravastatin but there are residual concerns amongst many clinicians, and there is no good evidence
of extra prognostic benefit from the use of both agents. Nevertheless, combination therapy might be
tried in high risk patients, for example in secondary prevention if triglycerides are still elevated on
statins. Fibrates should be avoided in patients with renal impairment or a history of gallstones.
Bezafibrate (Bezalip Mono) 400mg daily (cost £113/year) is a suitable drug for initial therapy, as is
fenofibrate (Lipantil) 200mg daily (cost £234/year), which may be more potent. Omega fatty acids
(Maxepa 5 capsules daily (£250/year)(** check – 5 caps daily or 10 caps daily!*) or Omacor
(£362/year) might also might also have a role in patients with grossly elevated triglycerides – say
over 10 mmol/l.
The case for aspirin therapy in diabetic patients with ischaemic heart disease is overwhelming, and must be given to all such patients unless there is a clear contra-indication such as a peptic ulcer or allergy to the drug. The case for aspirin in primary prevention is less robust, but it seems prudent to give it to all diabetic patients who are middle-aged and have any other evidence of complications or risk factors, such as proteinuria, retinopathy or hyperlipidaemia. South Asians diabetic patients in middle-age should probably receive aspirin therapy even in the absence of other risk factors, in view of their increased vascular risk. Since over 70% of type 2 patients have vascular deaths, and since aspirin has such a beneficial effect on myocardial infarction and acute coronary syndromes, this is a vitally important part of diabetes care. The debate on dosage - 75, 150 or 300 mg - is unresolved, but there is now a move towards lower doses, such as 75mg, as advocated in the NICE guidelines. If patients are “allergic to aspirin” or cannot tolerate its gastric side effects, clopidogrel might be given instead although this is much more expensive. There is circumstantial evidence that the effects of aspirin are nullified by concurrent ibuprofen therapy so this should be avoided if at all possible. Aspirin should be avoided if patients have systolic hypertension – say systolic BP>150 – because of the increased risk of cerebral haemorrhage may outweigh the reduction in ischaemic heart disease. The importance of stopping smoking can hardly be exaggerated. Diabetic subjects have such an alarming increase in ischaemic heart disease that it is sheer madness to superimpose another major risk factor voluntarily. Tell patients that we prefer them to stop smoking, even if it means putting on weight and poorer diabetic control ! Use any tactics you like - there are no holds barred ! This is the most important part of the management of any type 2 patient who smokes. With suitable threats and coercion, it is possible to persuade some patients to stop, although many still persist despite our best endeavours. Nicotine 'patches' or gum may have a limited role, but depend above all on adequate patient motivation, and their involvement in support groups. They are now available on prescription, but are contra-indicated in patients with established ischaemic heart disease ******Felicity / Ann Hamerton to check !!! Bupylpropion* (‘Zyban’) is now less popular after being incriminated in fatal epileptic seizures. There are local stop-smoking clinics in Ealing – phone 020 8321 2321 and an NHS smoking helpline – 0800 169 0169. Quality payments are earned, in the GMS contract, by offering advice to all smokers every 15 months. DIABETIC COMPLICATIONS
The term microalbuminuria (MA) refers to minor degrees of proteinuria which are elevated above
the normal (very low) levels, but cannot be detected by routine urinalysis. It is a good early marker
for nephropathy in type 1, and is also a surprisingly good marker for vascular disease in type 2 – ie
‘leaky kidneys’ are associated with a ‘leaky endothelium’, leading to atheroma. MA can be
measured with a stick test, or by laboratory analysis of a urine sample for albumin and creatinine,
preferably on an early morning sample. If the albumin/creatinine ratio (ACR) is elevated (strictly, on
2 out of 3 occasions), this constitutes MA, provided that a urinary tract infection has been excluded
by the absence of leucocytes and nitrates on urinalysis. The upper limit of normal for ACR is 2.5 for
men and 3.5 for women. By NICE guidelines, all type 2 patients should be tested for
microalbuminuria every year.
The detection of MA is important because it indicates the need for treatment with either ACE
inhibitors (in type 1 or type 2 patients) or A2 blockers (in type 2 patients) – even in patients who are
normotensive. Several studies have shown that this preserves renal function, delaying the
progression to renal failure, in addition to the effect on blood pressure. The best evidence in type 2
patients is with irbesartan (300 mg) or losartan (50mg), where their use prevents 2-4 deaths, or
patients developing renal failure, in every 100 patients treated. ACE inhibitors (rather than A2
blockers) are the first-line treatment in type 1 patients with MA. The combined use of ACE
inhibitors and A2 blockers together gives extra benefit both in renal protection and blood pressure
lowering. However, care is needed with these drugs in premenopausal women to ensure that there
is no risk of pregnancy, because they are significantly teratogenic. As renovascular disease may co-
exist with the nephropathy, especially in patients with peripheral vascular disease, it is essential to
check urea and electrolytes within a week of commencing ACE inhibitors or A2 blockers (and
strictly within a week of changes in drug dosage as well). Some rise in plasma creatinine is
commonly observed but provided this under 30% no further action is needed. If it rises more than
this, the drug should be stopped. Fortunately, this only happens rarely, but it is nevertheless vital to
monitor renal function carefully.
If overt proteinuria is detected on routine urinalysis, check for leucocytes and nitrates or send off an
MSU to exclude infection. If the MSU shows sterile pyuria, send off EMUs, to exclude renal TB,
especially in Asian patients, although, in practice, this is very rare. If proteinuria persists, and is not
due to infection, this strongly suggests early nephropathy, and indicates that the patient has an
increased risk of vascular disease. This requires aggressive management, with vigorous treatment of
any associated hypertension or lipid abnormalities. The target blood pressure in patients with either
MA or proteinuria is 135/75. Tight glycaemic control will also help to preserve renal function.
Nephropathy usually develops concurrently with retinopathy, and so if the fundi are normal it is much
less likely that the patient has significant diabetic nephropathy, although they may have other serious
renal disease. Asian patients are more prone to nephropathy than are Europeans.
Patients with raised plasma creatinine
Please refer any patient with plasma creatinine > 130 µmol/l to the hospital for assessment. It is
important to obtain a renal ultrasound examination if the creatinine value is over 150 mmol/l,
particularly in men to exclude obstructive uropathy from prostatic enlargement. Often such patients
have other medical problems, particularly CCF treated with powerful diuretic therapy and ACE
inhibitors. If the proteinuria is heavy (+++ on urinalysis), obtain a 24 hour urine collection to quantify
it, and check the plasma albumin. A full-blown nephrotic syndrome is relatively uncommon. The
risk of end-stage renal failure due to nephropathy is relatively small compared with the risk of
serious vascular disease, and will hopefully be further reduced by strict control of hypertension.
Patients approaching end-stage renal failure (with a creatinine >150 µmol/l, according to NICE
guidelines) should be managed jointly with a nephrologist, although many clinics have raised this
threshold to 200 because of the logistic pressures. Poor glycaemic control at this later stage has
little influence on the rate of decline in residual renal function. Insulin requirements often fall quite
markedly because insulin is partly metabolised by the kidney, and some patients may even be able to
stop their insulin or glicazide. Decreased insulin requirements are often an indicator of declining
renal function. Metformin and glibenclamide must never be used when the creatinine rises above
150 µmol/l, because of the real risk of lactic acidosis or profound hypoglycaemia. The preferred
oral treatment is gliclazide, but rosiglitazone or pioglitazone can also be used unless the patient is in
end-stage renal failure, with a * .
Screening for retinopathy is crucially important, and it is everyone’s responsibility to ensure that it happens! All patients attending hospital clinics can reasonably be assumed to receive eye screening there. Arrangements for patients managed entirely within the community will vary depending local circumstances. Some have set up schemes with trained optometrists, others use local ophthalmology services, and yet others use a mobile camera unit. The main priority is to ensure that some system is in place to cover all patients ! A system of district-wide digital retinal screening will be set up in **** to comply with the NSF *** details…. The benefits of early detection and treatment with laser are so impressive that blindness due to retinopathy is now largely preventable. Approximately 25% of all diabetic patients have retinopathy of some degree, and in about 10% this is severe enough to require laser therapy. Approximately 15% of type 2 patients may have retinopathy at diagnosis. Many retinopathy patients have already been diagnosed, are attending an eye unit and obviously do not require screening. The purpose of screening is to identify the small but vitally important proportion of patients - say 10% - who have undiagnosed retinopathy. Thus most fundi you examine will be entirely normal. Examination of fundi is remarkably easy under the right conditions. Many doctors are deterred from this when doing housejobs, looking at undilated eyes in bright wards, with poor ophthalmoscopes ! The important practical points are: 1) to dilate up the eyes with 1% Tropicamide drops. 3) to ensure that there are new or recharged batteries in the ophthalmoscope. Use 1% Tropicamide drops - not 0.5%. These may take 10 minutes to have their full effect, especially with dark irises. The risk of inducing acute glaucoma is negligible, but the drops should not be given to patients with known glaucoma, who will be attending an eye clinic in any case. All patients with type 1 diabetes must have their fundi viewed every year. By NICE guidelines, all type 2 patients also need yearly retinal screening, although although those with no retinopathy only need screening every 2 years. Some patients may object to the drops if they have to drive home. The drops do not affect distant vision, but lead to increased 'glare', which may make driving difficult. They wear off after 1-2 hours or more. When looking at fundi, you only need to decide whether they seem entirely normal or not. The main abnormalities are, of course, haemorrhages and exudates, which means any red or yellow/white lesions. Proliferative retinopathy ('new vessel formation') is uncommon in type 2, and is usually superimposed on obvious haemorrhages and exudates. So you should search carefully for any red or white/yellow blobs on the retina! The earliest changes are microaneurysms, which are tiny red spots, and small whitish exudates. Look particularly around the disc, and towards the macula, which is on the temporal (lateral) side. This is the commonest position for sight-threatening retinopathy. If you have any doubts whatsoever about whether the fundus is normal, please refer the patient to the hospital diabetic clinic. There are some normal, or at least non-diabetic, appearances which can cause confusion. Degenerative changes in the elderly often produce yellow lesions which look rather like hard exudates, but are not as clearly defined, and are not associated with haemorrhages. Many patients referred to hospital may not have retinopathy, but it is better to have a low threshold for referral. You don't need to be able to distinguish the different sorts of diabetic retinopathy - either the eye is totally normal, or it is possibly abnormal, in which case the patient needs referral to a hospital clinic. Those who have definite retinopathy will be referred on to a specialist eye unit. Some patients may develop macular oedema, with marked loss of vision and relatively little - or even no - evidence of retinopathy seen on fundoscopy. This is a potential diagnostic trap, and is suggested by an unexplained decrease in visual acuity which cannot be corrected by use of a pinhole. Thus it is essential to measure visual acuity in all patients before instilling eye drops. If this is impaired to 6/9 or less, the patient should be re-tested, using glasses or a pinhole, which will correct for refractive problems. If visual acuity remains impaired, and cannot be readily explained, for example by lens opacities, urgent referral to an ophthalmologist is necessary to exclude macular oedema. Some patients who are markedly hyperglycaemic have temporary blurring of vision, especially when newly diagnosed, due to changes in the lens shape produced by large blood glucose fluctuations. This is temporary, and is unrelated either to lens opacities or retinopathy. Nevertheless it can cause patients much anxiety, and it is important to explain to them that it has no long-term significance, and to advise them not to go to the optician for new glasses until their treatment is stabilised. CATARACTS
Cataracts (or lens opacities) are a major cause of poor vision in elderly diabetic patients. The best guide to their severity is the visual acuity (VA) - the correlation between the appearances on ophthalmoscopy and their impact on vision is remarkably poor. Until recently, ophthalmologists were unwilling to remove cataracts unless the VA in the better eye was reduced to 6/l8 or less. Now, with increasing skill in lens implantation, patients with less severe cataracts are being offered surgery. Unfortunately, cataract operations are generally less successful in diabetic patients, particularly if there is also significant retinopathy. It is important to establish how much inconvenience is caused by the lens opacities. Some elderly patients may not be unduly troubled by poor vision, and may wish to avoid the trauma of an operation. Conversely, younger patients may be severely distressed by a visual acuity of say 6/12. If in doubt, the patient should be referred for an eye opinion. Lens opacities often make fundi difficult to visualize, even when they don't need surgery. If this is a problem please refer the patient to the hospital clinic for an eye assessment. NEUROPATHY
Patients with neuropathy usually complain of paraesthesia - ‘burning feet’ - and numbness particularly in the lower limbs. Sometimes patients have symptoms despite preserved reflexes, although usually they are absent. Neurofilaments are a useful aid in assessing the severity of peripheral neuropathy, and identifying feet which are at risk of neuropathic ulceration, and are generally more useful than eliciting ankle reflexes. The 10 mg neurofilaments should be placed on the plantar surface of ** toes *** The treatment of severe neuropathic symptoms is unsatisfactory. They can sometimes be controlled with paracetamol and the importance of taking this drug regularly should be emphasised. If it is insufficient, try imipramine or amitriptyline, starting at low doses, such as 10 mg nocte (when symptoms are usually most troublesome) and increasing as necessary. These are sometimes effective, and their beneficial effect on neuropathic symptoms does not relate simply to its anti-depressant action. However, their use is often limited by daytime somnolence, and they are relatively contra-indicated in the presence of heart disease. Unfortunately, there is little evidence for similar benefit from SSRIs. If symptoms persist, the next drug to be tried is gabapentin, but at that stage they should be referred to a neurologist for assessment. Gabapentin has been proven to benefit neuropathy, and is relatively free of side-effects, but its dosage schedule is rather unusual – the dose needs to be increased rapidly over ** daysto 900mg and then *** Capsaicin cream (0.075%) applied to the affected areas is beneficial in some cases, and patients often find the cream more acceptable than additional tablets. Capsaicin is derived from peppers, and the cream initially causes warmth and burning, but with additional use it desensitises the skin, producing useful symptomatic benefit. Fortunately, painful neuropathy of this severity is relatively uncommon, and it tends to improve gradually. There is evidence that strict glycaemic control may help some patients with severe neuropathy particularly when it presents with weakness. Patients sometimes have a superimposed alcoholic neuropathy, and it is important to enquire for a high alcohol intake, which may exacerbate the neuropathy. If this seems a possibility, the patient should be given B vitamins, such as Thiamine compound forte tabs 1 t.d.s. which contain extra thiamine. The usual vitamin B preparations such as Multivite have too little thiamine. B vitamins are of no value in diabetic neuropathy. FOOT PROBLEMS
The treatment of early foot problems, and their prevention by education, are crucially important aspects of care. All patients should be taught to look after their feet. They should be advised to wash their feet daily, on how to cut their nails, to wear appropriate shoes and socks, and to avoid 'corn plasters'. Educational leaflets on foot care must be given to patients with potential foot problems. Many foot ulcers and amputations could be avoided by good patient education. Elderly patients need to be told to avoid self-chiropody, and instead should regularly see a State Registered Podiatrist. They are entitled to free priority treatment at Health Centres. Examination of feet in the clinic is important to document whether pulses and/or reflexes are absent, indicating that the feet are at risk. The importance of not walking 'barefoot' and of having correctly fitting shoes must be emphasised. If a patient develops an infected foot lesion, particularly if the skin is 'broken', early vigorous treatment with antibiotics is mandatory. Such patients should be referred urgently to the hospital clinic, for podiatry and careful follow-up. Please do not hesitate to refer any patient with these early but potentially serious foot problems. PERIPHERAL VASCULAR DISEASE
Patients with intermittent claudication should be strongly advised to stop smoking if they do so! Beta-blockers are contra-indicated in claudication, and ACE inhibitors should be used with caution because of the risk of concurrent renovascular disease. Alpha-blockers such as doxazosin are useful as anti-hypertensive agents in this situation. If the patient’s claudication distance is several hundred yards nothing more need be done. No drugs are beneficial, and symptoms often improve spontaneously. Low-dose aspirin should be given. If symptoms become severe, they need referral to a vascular surgical unit for Doppler studies initially, often leading to angiography. Surgical intervention is often unsuccessful in diabetic patients because their vascular disease tends to be more peripheral, but some patients may benefit from angioplasty or surgical revascularisation procedure, such as a femoro-popliteal bypass. ERECTILE DYSFUNCTION
It is important to enquire for this common problem. The cause is usually multifactorial, and is often compounded by psychogenic factors, especially when morning erections are preserved, or when it is intermittent. If peripheral reflexes are well preserved it is unlikely that the patient has autonomic neuropathy. Try to stop diuretics and betablockers, and advise the patient to reduce excessive alcohol intake. Endocrine causes are rare (<5%), and usually produce loss of libido rather than erectile dysfunction, but it is still worth estimating serum testosterone and prolactin. As everyone knows, the treatment has been transformed by sildenafil (‘Viagra’) which is taken about 1 hour before intercourse, and produces an erection in about 50% of diabetic patients in response to sexual stimulation, with little effect on the vasculature elsewhere, although some patients develop headache or dyspepsia, and rarely experience disturbances of colour vision. Patients with cardiac problems usually tolerate the treatment very well, but nitrate therapy is an absolute contraindication – this is extremely important ! The usual dose is 50mg, but this can be increased to 100 mg if necessary, and the cost is about £5 per tablet. They are allowed on prescription for diabetic patients at a rate of one per week. Newer related drugs – all PDE5 inhibitors - have now appeared, namely vardenafil (Levitra) which may work slightly faster and be marginally more effective, and tadalafil (Cialis) which lasts for 12-36 hours – perhaps more suitable for a romantic weekend ! All are strictly contraindicated in patients on nitrates. Patients who don’t respond to sildenafil or related drugs should be referred to an erectile dysfunction clinic. There are a range of more invasive and less aesthetic treatment options. The transurethral ‘MUSE’ (medicated urethral system for erection), in which alprostadil is administered directly into the urethra by a plastic cannula, works surprisingly well, with virtually no systemic side-effects and much better acceptability than some previous treatments, although about 10% experience local trauma from the urethral cannula. Another option is injection therapy. Many patients find this to be satisfactory, although it may have side-effects, particularly priapism and local pain, and surprisingly few patients persist with its use in the long-term. Various vacuum devices are now available. These have no side-effects, but patients have to buy them at a cost of about £300. The choice of therapy lies with patients and their partners. Patients who have an obvious psychogenic component may also benefit from referral to a Clinical Psychologist or a Psycho-sexual clinic. EDUCATION AND LIFESTYLE
Patient education is one of the most important aspects of diabetic care. It should be a continuing process, and involve all the medical, nursing, dietetic and chiropody staff. It is obviously important that everyone delivers the same basic messages. There are many booklets which provide detailed information for interested patients. Build up stocks of those which you find useful. Keep all your educational sheets and leaflets together in your clinic. Give them out personally, and ask your patients to read them carefully - this 'personal touch' lessens the chances of the leaflets ending prematurely in the waste-paper basket! There are good opportunities for group education sessions in mini-clinics. These approaches have potential, although in practice they are quite difficult to initiate and sustain. Try to form an educational group, with a few patients 'facilitated' by the diabetes or practice nurse. EXERCISE
Exercise has many potential benefits for type 2 patients. It may improve diabetic control, help weight reduction and decrease ischaemic heart disease. A reasonable target would be three sessions of a suitable exercise for 15-30 minutes per week. The aim is to have a pulse rate of 110/min at the end of the exercise. This should be modified in patients with known ischaemic heart disease. Alcohol abuse is a major health problem, especially in some sub-groups of Asian patients, and it is illogical to concentrate on prevention of long-term diabetic problems, whilst ignoring an obvious alcohol problem. Reducing alcohol consumption will improve diabetic control by decreasing calorie consumption. Pils lager is no more suitable for type 2 patients than ordinary beer, because it has just as many calories. Patients with severe alcohol problems are more prone to develop hypoglycaemia, and recovery is impaired, so tight control may be dangerous in this situation. However, a moderate alcohol intake is associated with reduced cardiovascular risk, at least in non-diabetic subjects. Type 2 patients who are taking oral hypoglycaemic drugs must inform the DVLA (Swansea) and their insurance company that they have diabetes. INSULIN-TREATED PATIENTS - PRACTICAL POINTS
Most patients on insulin attend a hospital clinic, and most of their practical problems will be sorted out there. The following notes cover some common and important topics which may arise when they attend a community clinic. Many patients will need to obtain supplies of a wide and bewildering range of items connected with their insulin treatment - syringes, pens, needles, lancets, needle clippers, strips etc. These are also summarised in the next few paragraphs. TYPES OF INSULIN
Insulin regimes overview
The range of different insulin regimes follow a few basic patterns, detailed below.
1. Twice daily insulin mixtures (Mixtard, Novomix 30 or Humalog Mix 25 insulin). This is the
basic pattern used for the majority of type I and some type II patients. They receive a fixed mixture
of both quick and intermediate-acting insulin, usually in a ratio of 30:70, and usually take two thirds
of the daily dose in the morning and a third in the evening. This works reasonably well but there is
no extra insulin to cover the post-lunch rise in blood glucose, and the evening intermediate insulin
may not last throughout the night.
2. Basal bolus regimes, with 4 injections a day, in which a long-acting insulin (ultratard or
glargine insulin) (usually injected at bedtime) provides a baseline of insulin, and short-acting insulin
injections are taken to cover each meal. This provides more flexibility and is the treatment of choice
for motivated type 1 patients and some type 2 patients, although glucose control in practice may be
no better than those on twice daily injections, and the regime requires 4 rather than 2 injections daily.
We previously used ultratard as the long acting insulin and actrapid insulin with each meal, but now
our standard regime is glargine insulin at bedtime, and Novorapid insulin with each meal.
3. Addition of bedtime insulin to tablets, for type 2 patients. This is a safe treatment for patients
who are poorly controlled despite full doses of tablets. Approximately 10 units of intermediate or
long acting insulin (Isophane, Ultratard insulin or glargine) are given at night, to reduce nocturia and
to lower early morning blood glucose. This regime serves as an introduction to insulin treatment,
with little risk of hypos and better acceptability than with two injections per day. The evening insulin
dose can be gradually increased, depending on the fasting blood glucose on the following morning.
Metformin is usually continued with bed time insulin, but sulphonylureas may be stopped.
4. Once daily insulin (without tablets) may rarely be given to elderly or infirm type 2 patients in
whom tight control may not be so crucial, and the main priority is the avoidance of hypoglycaemia.
Once daily monotard, mixtard or glargine insulin might be used for this.
New Insulin analogues
1. Rapidly acting insulin analogues - insulin aspart (Novorapid) and insulin lispro (Humalog) - act extremely rapidly and can be taken before, during or even after a meal. Their peak of action is 5-50*** minutes compared to 30-60 minutes of Actrapid insulin. This produces better cover of the postprandial rise in glucose, and fewer hypos, especially at night. However, it produces little or no improvement in the observed Hb A1C level, although this generally reflects basal rather than post
prandial blood glucose levels. Nevertheless, patients prefer these newer insulins, partly because it is
more convenient to take insulin immediately before a meal, or even after a meal, rather than half and
hour beforehand, and partly because they are delivered in better pen devices!
The rapid insulins are also used in Novomix 30 (Novorapid with Insulatard insulin), which is similar
to Mixtard insulin but which can be taken immediately before breakfast and supper, rather than 30
minutes beforehand. Humalog Mix 25 is the analagous produce from Lilley.
2. Insulin glargine (Lantus insulin) is a long acting insulin analogue which has a smoother profile than
other long acting insulins such as Ultratard. Thus, when given at bedtime it is less liable to cause
nocturnal hypos and it also produces less weight gain. Nevertheless, the absolute reduction in hypos
is relatively slight, and the insulin is expensive. Thus there is no need to switch all patients requiring
long acting insulin to this new agent. By NICE guidelines they are not recommended in type 2
patients unless the patient needs a carer, has recurrent symptomatic hypos or would otherwise
require two injections of a basal insulin per day. This last exception covers most type 2 patients in
Problems with glucose control
It is worth mentioning the inherent limitations to all insulin regimes. Whereas endogenous insulin is
secreted by the pancreas into the portal vein to act directly on the liver, exogenous insulin is injected
into subcutaneous tissues of diabetic patients as an unphysiological bolus, and leaches out into the
systemic (rather than the portal) circulation in a variable manner. Not surprisingly, this is an
imprecise art and blood glucose values often fluctuate widely in type 1 patients, in ways which
cannot easily be explained by food, exercise and insulin patterns. Many patients have unrealistic
expectations of achieving tight control. Insulin treatment is a crude art and not for the obsessional!
All patients should adjust their insulin dosage and base these changes on trends in glucose values,
rather than on individual values.
Human insulin versus animal insulins
The 'scare' that human insulin may cause hypos without any warning symptoms (and even unexplained deaths), has now largely subsided. A few patients had problems, and have remained on pork insulin again, but it has largely been a false alarm. Nevertheless, patients wishing to switch back to pork (or beef) insulins should be permitted to do so. There are a range of pork and beef insulins available. INSULIN PENS AND SYRINGES
Insulin pens – disposable (prefilled) or cartridge
Insulin pens are an important practical advance, because they are more convenient to use, and to
carry around, than syringes and insulin bottles, and virtually all patients now use a pen, although
patients should still be taught how to use conventional syringes and bottles in an emergency. There
are two types of pen - the disposable (or pre-filled) pens, which are discarded after all the insulin in
the pen is used, and cartridge pens, which take replacement insulin cartridges and are thus re-usable.
Disposable pens are slightly easier to teach to patients and more convenient, but are slightly more
expensive (see below). The major insulin manufacturers produce a range of both disposable and
cartridge pens, to take their respective insulins. As a rough guide, the annual cost of insulin
cartridges for a patient using 30 units of Mixtard 30 insulin daily is about £163, whereas the annual
cost of this insulin in disposable pens is £193. The equivalent costs for Novo Mix 30 are £196
(cartridges) or £216 (disposable pens). Insulin glargine is currently the most expensive insulin,
costing £287/year for 30 units per day, but in this case the cost of cartridges (for the OptiPen Pro 1)
and prefilled pens (Optiset pens) are identical.
Disposable (prefilled) pens
The new insulin analogues have been marketed in slightly better disposable pens, which are easier to
use and to teach to patients. The disposable pen for the Novo Nordisk insulin analogues (eg Novo
rapid, Novo Mix 30 etc) is the ‘Flexpen’, that of the Lilley analogues (eg Humalog, Humalog Mix
25) is simply called the ‘Humalog pen’ and that of insulin glargine, made by Aventis, is the ‘Optiset’
pen. The older Novo Nordisk insulins, such as Mixtard insulin, are contained in Novolet
disposable pens, and the Lilley range in the Humaject pen. Novo Nordisk also manufacture an
unusual disposable pen termed the ‘Innolet’ which looks more like an egg-timer than a pen, and is
useful both for older or arthritic patients, and for anxious patients.
The cartridge pen for older Novo Nordisk insulin, such as Mixtard insulin, is the Novopen 3, and
this takes Mixtard 30 penfills (or Actrapid or Insulatard penfills). The equivalent Lilley pen, for
Humulin M3, S or I insulins, is the Humapen Ergo. The cartridge pens now all take 3 ml (300 units)
insulin cartridges – the smaller 1.5ml cartridges are now obsolete.
Needles are required for both cartridge and disposable pens – they are prescribed separately, and
are not an integral part of the syringe. They are available in different needle lengths, and most
patients use 8mm lengths. Markedly obese patients may use 12mm needles, and children or very
thin adults may use 6 mm length. They are available on prescription. The cost to the NHS is
£8/100 needles or about £30/year for once daily injections etc. They are interchangeable and can
be used with all pens (disposable or cartridge pens) except for the Optipen Pro 1 pen, a cartridge
pen from Aventis, which specifically require Penfine needles, produced by Disetronic. Safe
disposal of used needles is obviously important. They should be placed in a safe container, such as
Journal of Antimicrobial Chemotherapy Advance Access published February 14, 2008 Journal of Antimicrobial Chemotherapydoi:10.1093/jac/dkn029Impact of CYP2B6 983T>C polymorphism on non-nucleoside reversetranscriptase inhibitor plasma concentrations in HIV-infected patientsChristoph Wyen1, Heidy Hendra1, Martin Vogel2, Christian Hoffmann3, Heribert Knechten4,Norbert H. Brockmeyer5, Johanne