Obesity and its management

Introduction
Obesity has medical consequences that are costly to society and the individual. The problem has been recognised by the World Health Organization (W.H.O.) which has recently published a technical report on obesity.¹ We all need to understand more about obesity and how to manage obese individuals. In Australia, the need for prevention has been realised and a National Health and Medical Research Council working party has published a report and plan, 'Acting on Australia's weight'.

While there is a belief that the majority of those in treatment programs regain their lost weight within 3 years, this does not reflect the aims, approach and results of recent management. It is not necessary to reach 'ideal weight', rather, current practice emphasises the loss of a reasonable amount of weight (6-10 kg), reduction in risk and emphasis of other benefits to the individual (e.g. effort tolerance, improvement in arthritis, fewer medications). Weight can be lost, but work needs to be done to maintain that loss.

There is prejudice against the obese individual ('will-power', 'gluttony and sloth' are often mentioned), against the obese individual in society and against treatment because of failure (either on the part of the health professional or the patient).

Risk associated with obesity (Table 1)
There is an increased risk of mortality, at any age, due to obesity alone. In addition, obesity causes, or is associated with, increased cardiovascular risk. There is hyperinsulinaemia, dyslipidaemia, hypertension, impaired glucose tolerance and diabetes, to name a few. Obesity is associated with sleep apnoea and its risks of sudden death and hypertension. These risks are reduced by weight loss.

Assessment of obesity
Adiposity, not weight, causes problems. While it is possible to measure the amount of fat an individual has, the techniques are too difficult or costly for routine practice. Instead, the body mass index (BMI, kg/m²) is used as an indirect measure of adiposity. There are some errors at the extremes of life, in the very fit and muscular or in some ethnic groups (e.g. some Pacific Islanders where BMI overestimates fatness and risk), but it is a useful measure (Table 2). Children can be assessed with BMI for age charts with those greater than the 85th percentile being `at risk' and those greater than the 95th percentile being obese.

As well as absolute adiposity, the distribution is important. Abdominal obesity conveys greater risk. Currently, males with a waist measurement >100 cm are at increased risk, while for women, waist measurement should be <95 cm. (It is likely that these recommendations will change to values of 94 cm in men and 88 cm in women.) These waist measurements are associated with increased visceral fat (>130 cm²) and it is this active depot that is implicated in many of the metabolic complications of obesity.

Prevalence of obesity
In Australia, 18% of adults are obese.² This has almost doubled since 1989. Over 62% of adult Australian males and 47% of adult females are overweight or obese. This is a worldwide trend.¹ It is interesting to compare the dramatic increase in obesity in the U.K. since 1985 with the less marked increase in the Netherlands, despite similar diets and living standards. One possible explanation for this discrepancy may be the greater use of bicycles in the Netherlands.

Aetiology
Fat is the major energy store of the body. An increase in fat stores can only occur with an increase in energy intake or a decrease in energy expenditure, but a number of factors or processes modify this simple equation.

Genes
There is a genetic component to obesity. The weight of adults adopted out as children is related to that of their natural parents and not to that of their adoptive parents. Twin studies also show that there is a genetic effect. When overfed (or underfed) by 1000 kcal per day (1 kcalorie = 4.18 kjoules), pairs of twins gained (or lost) very similar amounts. However, the weight gain (or loss) between different pairs of twins was very different. In a 100-day study, the weight gain on an extra 1000 kcal ranged from 3-14 kg.³

The direct genetic effect on obesity is estimated to be 25-40% and consideration of indirect genetic factors raises the genetic effect to some 70% of the cause of obesity. Over 24 genes associated with obesity have been identified. Leptin, the product of the lept gene in adipocytes, is a satiety factor in rodents. Mice unable to produce leptin or with abnormal receptors are obese. In those with defective production, giving leptin results in weight loss. However, in human obesity, leptin levels are elevated⁴, consistent with an increased fat mass. In general, the elevated leptin levels fall with weight loss and it is possible that this promotes a drive to eat and consequent weight regain.

Metabolism and diet
The obese do not have a reduced metabolic rate, except if dietary intake has been markedly reduced. It is the way dietary components are dealt with which is important. Those components which are oxidised do not promote fat storage e.g. there is the alcohol paradox: those with high intakes tending to be thin. The level of fat intake may be important as fat is stored preferentially and only oxidised to make up an energy deficit. Under normal living conditions, humans do not synthesise fat, but store fat eaten in the diet. The importance of fat intake is confirmed in epidemiological studies. Regions with a higher fat intake have more prevalent obesity. Fat intake may have been of major importance in the increase of obesity following the Second World War. However, recent studies suggest that fat intake is being reduced in Western cultures⁵, but the prevalence of obesity continues to increase. Over recent years, a decline in energy expenditure, through reductions in the activity of daily living, is seen as a major causative factor.

Society
Food is abundant and available, incidental activity is lower and there are many societal impediments to eating appropriately for one's age or being as active as necessary. There need to be major changes in the macro environment and in attitude before obesity can be prevented.

Medical factors
While hormonal changes, hypothyroidism, hypercortisolaemia and acromegaly, are often blamed, in reality they do not cause much obesity and produce relatively minor weight changes. The menopause is associated with an increase in abdominal adiposity in women. Drugs can cause weight gain and these include steroids, and some psychotropic and antidepressant drugs. Beta blockers may slow weight loss. Conditions which restrict mobility or activity are also associated with weight gain.

Consequences and cost of obesity (Table 1)¹
The direct health costs of obesity in Australia are estimated to be $830 million. Australians spend in excess of $500 million on commercial weight control measures. The indirect costs of obesity have yet to be assessed, but it will be necessary to do so in order to measure the cost-effectiveness of any intervention.

Management of obesity
While management is difficult, it is possible and can be effective so long as appropriate goals are set. Given regular follow-up, an 18 kg weight loss can be maintained for over 10 years. Treatment of obesity has to be long term because of genetic predisposition and the current environment (abundant, available food and wide choice and a tendency for much less activity in daily living). There should be an initial intensive treatment phase (weeks to months) and then continuing follow-up and treatment (years). There is a natural tendency for weight regain over years. This should be expected and intervention and treatment recommenced when necessary.

Who should be treated?
All those who are obese should be treated. As the degree of obesity increases, so the intensity of treatment and the modalities employed should increase. The presence or absence of risk factors should modify the treatment approach. Those at greater risk should have more intensive treatment. This may mean that those who are overweight (particularly men with abdominal adiposity) may require an intensive program.

Goals of treatment
Goals should be set before treatment begins. There should be no emphasis on reaching the 'ideal weight'. Short-term goals for 6-10 kg loss should be set. These are achievable and do produce changes in risk factors. There are many other goals which could and should be set, including better mobility, lower risk factors, control of metabolic disease (non-insulin dependent diabetes mellitus, dyslipidaemia, hypertension), less medication and increased well-being. Of course, there are often individual goals that a patient desires or needs.

Lifestyle program
Any obesity treatment must modify the combination of diet, physical activity, cognitions and behaviour. This approach may be all that is required for some individuals (e.g. overweight males). The emphasis on individual components of the program may be varied for the patient's needs.

Diet
Given the wide choice of foods available, it is easy to have a high energy intake. Whilst calorie-controlled diets may produce more rapid weight loss, from past experience it is obvious that there is poor adherence in the short to medium term. This may be because of physiological or psychological mechanisms.

At the moment, the recommended eating program is a low fat, ad libitum diet in which only the fat intake is specified. This leaves some sense of control in the hands of the patients, there is less problem with hunger and it appears to be both acceptable and effective in the medium term. The weight loss with such a diet is 1.0-1.5 kg a month and a weight loss of 10-15 kg in a year is an excellent result. This low-fat eating plan appears to be a good diet for the maintenance or follow-up phase of treatment. The fat intake can be determined on an individual basis. My preference is for a daily 40 g limit for men and a daily 30 g limit for women.

Activity
Increasing activity is an important part of the program. As well as increasing the energy deficit, it maintains muscle mass and promotes well being. Weight loss produced by increased exercise is not great, but may be of the order of 4-5 kg over 3 months. The type of exercise suggested should be low to medium-intensity aerobic activity.

Walking is the best example. Only a few minutes walking may be possible at first, but both the duration and intensity may be increased as the program proceeds and the patient loses weight. If patients have difficulty because of arthritis, hydrotherapy (activity in water) may be of great benefit and help them get started. It is not necessary to walk for an extended period of time.

The current emphasis is on an increase in the activities of daily living and variety in the activities undertaken. Where possible, one should choose an active alternative such as walking up stairs, walking to work and getting up to change television channels. This increased activity sums into increased daily energy expenditure. It is not necessary to attain cardiovascular fitness to lose weight. In general, men respond better than women to an exercise program, and this may be due in part to a familiarity with exercise or training.

Cognitions and behaviour
It is essential to identify those cues, social and psychological, and the habits that may induce inappropriate or overeating or under-activity. This requires time and effort. Habit recognition, the keeping of food and activity logs, cognitive restructuring, stress management and counselling are all part of this continuing approach. It is the changing of behaviour, sometimes lifelong habits, that will permit weight loss to be maintained.

While medical practitioners can supervise all the treatment, it may be necessary or essential to involve other health professionals, especially dietitians, in aspects of treatment. A team approach, with a set program backing up the multidisciplinary approach, may be the most effective way.

Pharmacotherapy
The drugs available for the treatment of obesity are not ideal, so the patients need to be selected carefully. Drugs are considered for those in whom hunger or overeating is a problem or those whose weight poses substantial risks.

Dexfenfluramine is an effective drug. It has been withdrawn following reports that use of this drug, or more particularly fenfluramine, especially if used in combination with phentermine, appeared to be associated with valvular heart disease.⁶ With its withdrawal, obesity treatment has taken a step backwards.

There was a tendency in the U.S.A. for the combined use of the appetite suppressants, fenfluramine and phentermine. At first, this was in low dose, but later, larger doses were used. This combination was used for some time and was seen to be very effective treatment. When dexfenfluramine became available, it was also used in combination therapy. In 1997, there were reports of cardiac valve thickening, even necessitating valve replacement in some women who had used combination therapy for weight loss.⁶ Most of the subjects were female, and the prevalence was higher in those who had been treated for >6 months. Dexfenfluramine and fenfluramine were voluntarily withdrawn. For patients who had been treated with these drugs, an echocardiogram was suggested if they were symptomatic.⁷

There have been few reported problems in Australia. There is no reported pulmonary hypertension and the Adverse Drug Reactions Advisory Committee has only recently reported 3 cases of valvular thickening. As the number of cases reported in the U.S.A. is greater than elsewhere, there are possible questions about the way these drugs were used in the U.S.A. In Australia, because of our prescribing limits and the fact that combination therapy is uncommon, there appears to be less of a problem.

What is available?
Older appetite suppressants are still available. These are amphetamine derivatives with adverse effects of excessive stimulation, tachycardia, hypertension and possible habituation. In a recent study, we found that phentermine is as effective as dexfenfluramine in producing weight loss and that adverse effects are few during a 3-month course. After ceasing therapy, patients treated with phentermine regained less weight over the next 6 months. So, phentermine in the dose of 15 mg twice daily for up to 3 months is an effective treatment in those who are experiencing hunger, problems with controlling eating or difficulty initiating weight loss. Appetite suppressants may still be used in those experiencing difficulty losing weight in a supervised program.

Future drugs
Two new compounds are being studied. They have different modes of action from the current drugs.

Sibutramine has appetite suppressant, serotonergic and thermogenic effects (noradrenergic). Its adverse effects include a slightly raised resting pulse and hypertension.

Orlistat is a pancreatic lipase inhibitor. It binds to lipases in the bowel and prevents fat breakdown and absorption. The adverse effects are substantially gastrointestinal and orlistat must be prescribed together with a low-fat eating program. The weight loss is similar to that with other drugs for obesity (an extra 30-50% of weight loss), but it does have the potential for long-term use. Weight loss has been maintained during two years of therapy. Orlistat also reduces LDL cholesterol and, through weight loss, has other beneficial effects on some aspects of the Metabolic Syndrome. This includes hypertension, non-insulin dependent diabetes mellitus, hypertriglyceridaemia, low HDL cholesterol as well as abdominal adiposity.

While it was hoped that obese humans would be deficient in leptin, its production from adipocytes is increased and serum levels are elevated in obesity. Obese humans are therefore either resistant to the satiety effects of leptin, or have learned to over-ride them. It is therefore unlikely that leptin will be a general treatment for obesity. However, it may have a role in weight maintenance after weight loss. With weight (and adipose tissue) loss, serum leptin levels decline. This fall in leptin may produce increased appetite, increased eating and intake and thus promote weight regain. Treatment with leptin, or an analogue, may prevent this. Leptin also has other actions. It is a signal of adequacy of fat stores and may be important in puberty and fertility. It may have peripheral actions as well as the central actions on appetite. For example, it improves glucose homeostasis in rodents. These additional actions of leptin are still being studied.

Very low energy (calorie) diets (VLCDs)
These are an effective form of therapy in those with class II or class III obesity. They produce rapid early weight loss. These diets, which contain between 400 and 800 calories per day, are largely protein-based with essential fatty acids, vitamins and minerals, but very little carbohydrate. Patients treated with these diets must be monitored carefully and there also must be a program of nutrition education and support for the time when these diets are ceased. Without lifestyle and habit change, the weight loss is not maintained. While there were problems with the early formulations, modern formulations enable patients to continue their normal daily activities. They do not need to be hospitalised. The method of administration varies. Some programs are a 6-12 week course of these meal replacements; in Sweden, they use two weeks' therapy every 3 months and achieve satisfactory weight loss. These diets may be used in patients with diabetes and with weight loss there is better glycaemic control. The diets may be repeated if and when weight is regained.

Surgery
This is the most effective long-term treatment for those with class III obesity. The interim analysis of the Swedish Obese Study, which has enrolled 10 000 patients with a 15-year follow-up, has determined that obesity surgery is the most effective therapy to date. Patients need to be carefully selected both on medical and psychological grounds. Those who cannot control their food intake are best suited to this approach. Operations include gastric bypass, banded gastroplasty and the more modern laparoscopic banding. Results of older studies show that 30% of patients will achieve a weight loss >30% of their excess weight, some 30% will have complications which include vomiting, stomal narrowing, pulmonary emboli and major postoperative complications. More recent studies, such as the Swedish Obese Study, are able to achieve a greater percentage of individuals with satisfactory weight loss. Surgery should be performed by an experienced surgeon and there must be a follow-up program. Local liposuction has no part to play in the treatment of generalised obesity. It may have a place after individuals have lost weight, in the removal of unsightly local collections of adipose tissue.

Conclusion
Obesity is a major public health problem throughout the world. It is caused by a combination of genetic factors, inappropriate eating and reduced activity. It can and should be treated. The mainstay of treatment remains a low fat diet and increased physical activity including, where appropriate, pharmacotherapy. Behaviour modification helps both initial weight loss and maintenance. Whilst dexfenfluramine has been withdrawn, other appetite suppressants may be of use as adjunctive therapy in a treatment program and there are newer and different agents being assessed. For the morbidly obese (class III obesity), treatment with VLCDs or surgery may be necessary. The aim of treatment is not to achieve ideal weight, but is rather to reduce risk factors by a smaller, achievable loss. Regular and continuing follow-up has been shown to be essential in maintaining weight loss. Prevention should be the ultimate goal and there is a plan for Australia to approach this in a co-ordinated way.

Dr Caterson has received funding from, or been involved with studies sponsored by, Servier Laboratories, 3M Pharmaceuticals and Roche Products.