Management of diabesity in primary care: individualisation of care

Abstract

‘The melting is rapid, the death speedy. Moreover, life is disgusting and painful; thirst unquenchable.’ Aretaeus the Cappadocian. Copyright © 2012 John Wiley & Sons. The word ‘diabesity’ has recently been coined, to describe the coexistence of the ‘twin epidemics’ of diabetes and obesity, but the expression is being used wrongly. Ethan Sims and his co-workers invented the term in the 1970s during studies of ‘experimental human obesity’1 whereby healthy prisoners were deliberately overfed, gaining weight to an average BMI of 28kg/m2. A BMI of 28 is overweight, not obese. Sims demonstrated reversible rises in fasting glucose and a deterioration of glucose tolerance as a consequence: not diabetes. Furthermore, the purpose of the studies were to explore the interactions between genetics and the environment, rather than relying on the over-simplistic energy balance equation; why did some prisoners require 10 000 extra calories a day to gain the same weight that others did at far lesser degrees of excess. Rather than merely re-stating the coexistence of diabetes and obesity, Sims' paper was more subtle, observing the mutual causes of impaired glycaemic control and weight gain, but also, crucially, the reversibility of both conditions on changing lifestyle habits. One inescapable conclusion is that diabetes cannot properly be said to be well managed without the effort to induce weight loss in overweight or obese patients, although many current glucose-lowering therapies have the opposite effect. Stevens and colleagues2 showed that around 90% of individuals who develop type 2 diabetes have a BMI >23.0kg/m2. Thanks to modern science, the fog is starting to clear in the management of type 2 diabetes, which is remarkable since not long ago it was not realised that any fog existed. After 1922, when 14-year-old Leonard Thompson was comatose and near death before becoming the first patient to be treated with insulin,3 surviving until 27, the priority for diabetes management was glucose control alone. Following the success of UKPDS in 1998, various large studies – ACCORD, ADVANCE and VADT – were designed to confirm and build upon UKPDS data, by treating more intensively, in more high-risk patients, but the results were unexpectedly negative precisely because disparate patient groups were being treated in inappropriate ways. Subtle individualisation of care for each patient, therefore, is now the priority, and primary care has assumed the mantle for such management in all but the most complex cases. In 1921, Frederick Banting, along with many other scientists whom Banting was at pains to acknowledge, altered the burden of diabetes from an inevitable death sentence to a treatable disease by successfully extracting useful insulin, for use by the most specialist medical centres. Now, at last, the burden of managing diabetes in the context of global cardiometabolic risk rests with primary care, where patients can be closely followed up, on a daily basis, if necessary, even at home, by a skilled and motivated multidisciplinary team. This has happened because of the medical advances made since Banting's era, and because of modern changes in commissioning, yet to hit primary care with their full force, which have underpinned the devolution of diabetes management into the community. However, the community must be ready to assume the role of managing diabesity, the metabolic syndrome and the broader ramifications. Since Banting's seminal work, the rate of increase in knowledge as judged by the volume of research and publications has accelerated, so that every week a new theory or paper is published. Each contemporary major paper throws new light upon the subject, introducing new subtleties with which primary care must be acquainted, or else patient care will suffer. Among the major studies, UKPDS4 highlighted long-term benefits of early intensive treatment; VADT5 sided against treating intensively in elderly, high-risk individuals with long duration of diabetes; ACCORD6 provided a stark warning against treating too intensively; STENO-27 promoted management of blood pressure and other risk factors in order to reduce macrovascular risk; and so on, so that each patient's individual needs are catered for. So now, in the ‘modern era’, different populations are treated differently, depending on age, degree of risk, duration of diabetes, different drug regimens, and of course individuals' weight and body morphology. The management of overweight and obese patients suffering from diabetes is a distinct issue, which merits individual consideration for individual patients, and highlights specific concerns. Metabolic syndrome was described by Gerald Reaven in 1988, as a means of co-defining obesity, diabetes, hypertension and dyslipidaemia under one clinically meaningful umbrella. The concept of ‘diabesity’ might be considered to represent ‘Metabolic Syndrome Lite’; a valuable reminder to non-specialists to engage actively with weight control alongside glycaemic control in overweight or diabetic patients, but not a reason to ignore blood pressure, lipids and other factors. There is a closeness in pathophysiology between diabetes and obesity, ensuring that they appear together on mutual algorithms, but metabolic syndrome goes further, linking abdominal obesity with the other major cardio-metabolic parameters including blood pressure and lipids, which invariably coincide with obesity and diabetes. Reaven, the Emeritus Professor of Medicine at Stanford, in his American Diabetes Association lecture in 1988 in honour of Frederick Banting8 described the method by which adipose tissue, skeletal muscle and liver become resistant to the effects of insulin, the subsequent hyperglycaemic drive and compensatory hyperinsulinaemia. Eventually, amidst increasing demands for insulin, beta-cell failure occurs, blood glucose rises unchallenged and type 2 diabetes develops. Other organs of the body, such as the ovaries, kidneys and brain, react badly to raised insulin levels, described by Reaven9 as ‘innocent bystanders’ of the hyperinsulinaemic state. The International Diabetes Federation (IDF) states:10 ‘With the metabolic syndrome driving the twin global epidemics of type 2 diabetes and CVD there is an overwhelming moral, medical and economic imperative to identify those individuals with metabolic syndrome early, so that lifestyle interventions and treatment may prevent the development of diabetes and/or cardiovascular disease.’ According to the IDF definition,11 obesity is the one single characteristic which must be present for an individual to be categorised as having the syndrome, removing any doubt about the clinical relevance of weight control. Arguably, even metabolic syndrome is too exclusive a definition, failing, as it does, to recognise conditions such as polycystic ovary syndrome (PCOS), sleep apnoea and depression as part of the bigger clinical picture. The concept of diabesity is even more narrow and simplistic in only linking weight and glycaemia. In observing the broad range of obesity-related comorbidities, highlighted by the metabolic syndrome, there is a strong argument, especially for primary care, on initial engagement of an obese individual, to be alert to conditions such as sleep apnoea – closely linked with hypertension, stroke and traffic accidents – which is only ever diagnosed in 15% of sufferers, with a pick-up rate of only 1.5% in any one appointment. There is an important clinical argument which ensures that anyone with any facet of the metabolic syndrome, or any other obesity-related condition, be broadly screened for all the others, not just diabetes. Much literature relating to diabesity, including most trials, overlook the fact that real-life management of diabetes and obesity does not involve giving obese volunteers the latest new pill or operation as part of a study; instead, it occurs in a primary care setting where anonymous fat people are encountered on a daily basis, some as patients, often with a condition unrelated to their weight, some not even as patients, merely the accompanying spouse, parent or offspring. Arguably, the most challenging part of the entire management process of obesity and diabetes is the initial conversion; the means by which anonymous fat people become engaged, aware and informed about their condition and associated risks. The identification of a high-risk individual is best done visually: an expanded waist is easy to identify, as soon as a person enters the room. Waist circumference, BMI and more complex body fat analyses can give extra information, but a person who looks as though they have a weight problem, has a weight problem, regardless of BMI. The most difficult but arguably most important step in the weight management programme is engagement – which can occur in the remaining one minute of an unrelated consultation but which can just as easily lead to alienation if done clumsily. The simple question ‘how's your general health?’ can introduce the appropriate conversation. (‘How's your general health?’ in a patient attending for an unrelated reason; ‘how's your general health?’ for an accompanying person.) This leads to one of the most important aspects of diabesity management: screening – height, weight, blood pressure, fasting glucose, HbA1c, lipids. The message from UKPDS is clear: early intensive management of diabetes improves outcomes, with a long-term so-called legacy effect, but can only be done with early diagnosis by screening. Studies such as ADDITION demonstrate that screening of high-risk patients in primary care can identify raised diabetes and cardiovascular risk factors, so that early intensive management can occur. There are arguably four groups of obese individuals (Table 1). Obese individuals who do not yet suffer overt illness: weight loss is beneficial, and accompanied by an improvement in global risk and specifically the risk of diabetes if successfully maintained long term Obese individuals already suffering from type 2 diabetes, polycystic ovary syndrome or sleep apnoea etc who suffer those conditions because of obesity, for whom weight reduction (or at the very least avoiding weight gain) is the essential cornerstone of treatment Those for whom the increasingly important concept of the ‘obesity paradox’ suggests that weight loss is inappropriate (e.g. some patients with heart failure, post percutaneous coronary artery intervention or renal failure), and that a degree of excess waist circumference seems to promote more favourable outcomes In another group of primary care patients, although weight loss would be beneficial, it cannot be achieved or maintained, but as long as identification, engagement and risk reduction – blood pressure, lipids etc – have occurred, they can still be considered successful in obesity management. Success should not simply be judged by kilograms lost Clearly, nutritional and physical activity interventions are cornerstones in the management of excess weight and diabetes, and also have the effect of improving blood pressure and lipids. Physical activity guidelines have tended to be somewhat simplistic and one-dimensional, rather than individually targeted. A person who is housebound will get more benefit from performing 100 steps a day than someone who increases from 9900 steps a day to the officially endorsed 10 000 per day. Suffice it to say, the more physical activity the better to improve health and insulin resistance, as well as reducing cardiovascular and cancer risk. Nutritional advice is more complex. Historically, in the 17th and 18th centuries, there were two schools of thought in dietetic management of obesity-related diabetes. The Scottish physician, John Rollo,12 considered that as sugar was being passed in the urine, the obvious solution was to decrease carbohydrate intake in order to redress the balance. Others such as Thomas Willis,13 however, had felt that the solution to glycosuria, and to regaining carbohydrate balance, was to replace the sugar lost in urine by adding sugar to the diet; a notion which seems comical today, although many dietetic guidelines promote a diet based on a carbohydrate-rich foundation, such as the Eatwell plate. The confusion about whether a low carb or a low fat diet is more favourable in diabetes may have emanated from Ancel Keys' Seven Countries Study,14-16 used as the basis for the COMA report upon which generations of guidelines have been based. Keys categorised certain foods as fat, or carbohydrate based. A slice of cake or a bowl of ice-cream, according to Keys, were defined as ‘saturated fat’ despite the immense sugar and carbohydrate they contained. A reliance on such flawed advice has led to carbohydrate-rich diets being promoted, although more enlightened trials such as the A TO Z Weight Loss Study17 and the Israeli Diet Study18 clearly show that low-carbohydrate approaches induce greater weight loss as well as improvements in blood pressure and cardiovascular risk factors, and are superior in the short- and long-term management of diabesity. Increasingly, glucose-lowering pharmacotherapy is being introduced early in the management algorithm. As metformin is generally considered as the first candidate as a glucose-lowering agent, assuming renal function is adequate, then weight is not initially an issue, metformin being weight-neutral. Otherwise, iatrogenic weight gain is a drawback of many conventional agents such as sulphonylureas (SUs), thiazolidinediones and insulin. Twenty-eight percent of the group treated intensively in ACCORD – in which increased mortality was noted – gained >10kg. In UKPDS 33, patients randomised to insulin or SUs had weight gains of 4.0kg and 1.7–2.6kg, respectively, compared with the conventional group receiving dietary advice alone. Of the traditionally widely-used drugs, only metformin is devoid of adverse effects on body weight and was the only drug shown by UKPDS to reduce atherosclerotic events and mortality. Acarbose has no adverse effect on weight, but is limited in use by other side effects. The advent of new classes of glucose-lowering agents such as the glucagon-like peptide-1 (GLP-1) mimetics, exenatide and liraglutide, dipeptidyl peptidase (DPP)-4 inhibitors – the ‘gliptins’ – and soon to launch sodium-glucose transport protein (SGLT)-2 inhibitors, has broadened the range of clinical options available for the drug management of diabetes. These agents offer advantages over some of the older agents in terms of avoidance of weight gain; their arrival demands a renewed critical appraisal of traditional drugs and offers new options for people with diabetes. Progressive weight loss has been demonstrated with exenatide: 1.6–3.6kg in 26–30-week studies19-23 and, among completers, 5.3kg after three years of treatment.24 Liraglutide is associated with clinically meaningful weight loss after 26 weeks, of 2.6–2.9kg for a 1.2mg dose, and 2.8–3.4kg for the 1.8mg dose,25,26 and is currently undergoing trials at a higher doses in anticipation of gaining a licence for obesity with or without the presence of diabetes. DPP-4 inhibitors are considered weight neutral,27,28 although a mean weight loss of 0.96kg has been reported with sitagliptin in a head-to-head study with liraglutide.26 Clearly, the weight profile of these more recently studied agents is superior in the management of diabesity than older traditional agents. NICE has provided exemplary guidance on glucose-lowering agents, allowing more than ever for appropriate individualisation of care, particularly useful for overweight and obese patients with ‘diabesity’; the guidance approves DPP-4 inhibitors ahead of an SU as second-line therapy to first-line metformin when blood glucose control remains or becomes inadequate if the person is at significant risk of hypoglycaemia or its consequences, or does not tolerate an SU or an SU is contraindicated. They are also promoted in preference to thiazolidinediones if further weight gain would cause or exacerbate significant problems associated with a high body weight. GLP-mimetics are deemed appropriate in patients with a BMI >35 and problems associated with high body weight, or for patients with a BMI <35 for whom losing weight would help other weight-related health problems, or taking insulin would greatly affect their ability to work. In other words, as long as there is a good reason, then therapy can be tailored to the individual patient. As the problem with diabesity worsens, the range of pharmacological weight loss options is diminishing. Sibutramine and rimonabant – highly effective drugs in terms of both weight and HbA1c reduction – have been controversially withdrawn, sibutramine being particularly badly mishandled by the licensing authorities in the author's opinion. Its withdrawal was based on information from the SCOUT trial,29 in which patients with cardiovascular disease (CVD) for whom the drug would have been contraindicated in clinical practice, were not only given the drug, but forced to stay on it for five years, even if they did not respond, and therefore would have been taken off the medication in normal practice. Subsequent, post hoc data from SCOUT30 have shown – remarkably – that modest weight loss with the drug is actually associated with reduction in subsequent cardiovascular mortality for the following four to five years in those with pre-existing CVD, if used according to its clinical licence. A study of over 15 000 people taking sibutramine in usual clinical practice in New Zealand31 revealed no increase, and a probable decrease in mortality on the drug. Rimonabant was shown to cause depressive mood changes in some individuals, but had beneficial effects in others, SERENADE32 revealing a reduction in HbA1c of 1.9%, albeit from a high baseline, alongside weight loss. The only remaining prescribable agent is orlistat, a pancreatic lipase inhibitor, which allows 30% of dietary fat to pass unabsorbed in faeces. Some research demonstrates reductions of >10%33 at one year, but most clinicians experience more modest benefit. XENDOS,34 however, demonstrated a 37% reduction in cumulative incidence of type 2 diabetes; a much stronger recommendation for its use. Qsymia, a combination of topiramate and phentermine, has been the subject of a NICE scoping exercise, based on trials including CONQUER,35 and SEQUEL,36 which demonstrate 10% weight loss sustained at two years, accompanied by improvements in metabolic risk factors. Lorcaserin is a serotonin receptor agonist, chemically similar to the withdrawn fenfluramine, apparently without the associated risks, and supported by the BLOOM37 and BLOSSOM38 studies which show more modest weight loss than topiramate/phentermine but a better side effect profile. Contrave combines naltrexone (currently for alcohol and opioid addiction) and bupropion (an antidepressant and smoking cessation aid) but is some way off approval. Empatic, by the same company, in an earlier stage of development, is bupropion with the anti-epileptic zonisamide. An intriguing approach towards obesity management within primary care is provided by the Edmonton staging model, described by Arya Sharma.39 According to the model, the severity of obesity is not simply assessed by a person's weight, waist or BMI, but by the degree of illness or dysfunction their level of obesity affords. Thus, someone with a very high BMI scores a low grade on the Edmonton Scale if they are mobile and free from comorbidities, whereas a person with a BMI of 30, who already suffers from diabetes, sleep apnoea or PCOS achieves a high score and is a candidate for more intensive primary care treatment. Furthermore, the Edmonton model allows for palliative management of obesity, in a person so obese that lifestyle therapy, drug management and even surgery are destined to fail. A novel role for primary care is the assessment, management and follow up of bariatric surgery patients. Many individuals are discharged after gastric bypass, their diabetes likely to resolve within days, yet still on insulin, without the primary care team even being aware that the patient has been admitted for surgery, and without even the most elementary guidelines on reducing and stopping insulin. Many bariatric units have excellent communications and guidelines, but this should become normal practice. It must also be remembered that although an individual post-bariatric surgery has a reduced cardiovascular risk profile compared to their former state, they still maintain a higher risk than their counterparts who have never been morbidly obese. Therefore, reduction and cessation of antihypertensive, cholesterol-lowering drugs and other medications should be performed slowly and closely monitored. Tertiary care centres have dwindling capacity for preoperative and follow-up work, and the burden is likely to fall upon primary care, which has the skills and motivation to assume the workload. The Roux-en-Y gastric bypass, in particular, has the potential to resolve diabetes, by an as yet unidentified mechanism. There is a perception that the bypass is the sole preserve of end-stage obesity, those with a BMI off the scale, whereas bariatric surgery may be best utilised in patients such as young, obese, insulin-resistant diabetic individuals, who are likely to suffer expensive microvascular and macrovascular complications, and for whom risks might more effectively be reduced. Cost analyses, such as the Picot Review,40 the National Bariatric Surgery Register,41 and the Office of Health Economics report ‘Shedding the Pounds’,42 have revealed that bariatric surgery is not only clinically effective but cost effective within a year. There is a risk that the new commissioning environment might endanger bariatric surgery, due to clinical commissioning group short-termism. It is the duty of primary care professionals to identify and insist on onward referral of patients deemed suitable by NICE, and therefore supported by the NHS Constitution, toward bariatric surgery. Gastric bypass surgery induces resolution of diabetes in around 86% of cases, more so in younger more newly-diagnosed patients. Gastric band has been shown to induce diabetes resolution in 73% of drug naïve newly-diagnosed diabetic patients. Obese individuals have a host of other comorbidities to contend with in addition to type 2 diabetes. Not everyone who is obese will develop type 2 diabetes. Not everyone with a high risk of type 2 diabetes is necessarily obese. Advisor and occasional speaker for MSD, BMS, BI, Novo Nordisk, and Sanofi. References are available online at www.practicaldiabetes.com.