OBESITY, AN OLD PROBLEM, A MODERN CRISIS

By now, practically everyone in the developed world is aware that there is an obesity problem if not a crisis. It has no socioeconomic boundaries since it is common to the rich and poor, inhabitants of developed and underdeveloped countries, and even seen in aboriginal populations. Obesity is generally regarded as the gateway to chronic and degenerative diseases, and those who worry about healthcare economics and the funding of public and private systems wring their hands in despair at the simple projections of future financial disaster common to most models.

The history of obesity, especially in prehistoric times, is unclear. There is evidence from so-called Venus figurines, the earliest apparently 35,000 years old, which showed clearly obese women. Figurines dated 12,000 years old also confirm the observation, but it is not known if these represented a stylised notion or how a significant fraction of women looked at the time. This was still the hunter-gatherer era and well before the advent of agriculture and the cultivation and grinding of grain. In fact, there are carvings going back beyond 100,000 years that also suggest obesity was present. In addition, there are both wall paintings and mummified remains dating to 1500 BC indicating that an Egyptian queen was clearly obese. A thousand years later Hippocrates described obesity and by the early 1800s, William Wadd, the Surgeon Extraordinary to the King of England had published a book on obesity. He also described the post-mortem examination of an obese person and found extensive fat deposits that are also seen today. Later in that century, a gentlemen named Banting (no relation to the discoverer of insulin) became famous for his pamphlet on a low-calorie low-carbohydrate diet he had found successful after consultation with the famous physician William Harvey.1

Much more recently, the rapid increase in prevalence of obesity has become of general concern. One piece of evidence comes from comparing National Health and Nutrition Examination Surveys (NHANES) which attempted to capture a vast number of health-related characteristic of the general U.S. population. By comparing the 1970 and 2000 surveys, it is found that women 20-29 years of age in the early 70s had a mean body mass index (BMI) of 23 which is mid normal. In the period of 1999-2002 the survey found that women 50-59, the age then of the earlier cohort, had a mean BMI of 29 (at the threshold of obesity), representing a weight gain of approximately 35 pounds in an average of 28 years.2 According to models discussed below, the energy intake would have to be about 370 calories more at the older age to maintain the new BMI as an equilibrium weight. This can occur gradually with small increases in energy intake over that required to maintain a stable weight at the younger age.2,3 Since control of such a small increment is impossible to implement on a daily basis, the only alternative would be to cut back on energy intake and/or increase activity by addressing weight increases when they becomes apparent. That this was not and is not being done is indicated by the above data and the prediction by the Organization for Economic Cooperation and Development that by 2020 three-fourths of Americans will be obese or overweight.

There is no lack of theories as to the cause of obesity, and certainly no shortage of schools of thought regarding the best diet to combat it. Book stores have shelf after shelf of diet books based on an amazing variety of philosophies. Gary Taubes has published his magnum opus titled Good Calories, Bad Calories, which goes on for 600 pages and has a level of sophistication equal to any good academic work. Furthermore, the academic nutritional literature is overloaded with studies on diets from many viewpoints and there is no shortage of what appears to be junk science. If one puts "obesity" as a title item in PubMed, this marvellous medical literature search engine brings up over 3000 papers in the past 12 months. The diet books keep coming out, dozens of studies report each month, the media eagerly cover new events in the nutritional world, generally without the proper perspective, and individuals of all ages just get fatter and fatter. At one extreme, there are some who no longer fit in the standard airplane seat, and some who can only be taken from home to the hospital by cutting a hole in a wall and using a crane and then a truck. Of course there is the other extreme described by the term anorexic, accompanied by anthromorphic features which fashion models currently exhibit in a somewhat mild form and some adolescents in an extreme form. But that is a different topic. There are, however, also large numbers of individuals who are of normal weight, but for some if not many of these, weight is probably creeping up and fat starting to accumulate around the waistline.

What we eat and how much we eat and then what happens is a highly complex subject containing a grand mixture of psychology, physiology, biochemistry, genetics, lifestyle, the nutritional environment and culture within the home and to some extent the neighbourhood and country of residence, and the human response to highly sophisticated food marketing. All this makes for a difficult subject to research and probably an impossible subject to cover adequately in medical school. The most relevant studies require carefully controlled human subjects studied over long periods, studies which in general are too expensive and offer difficulties in recruiting. Judging by the hundreds of published diet studies, it appears difficult if not impossible to impose an intervention for longer than about six months, many studies have shorter interventions, and the net result is a final weight loss that represents a small and perhaps insignificant percentage of the total weight the subjects should want to lose. Even the best studies are complicated by considerable human variability.

In the 21st century it is quite impossible to conduct studies lasting a year or two on large numbers of subjects typical of the general overweight or obese population who are confined to an institution, fed nothing but experimental diets prepared in-house, and for which physical exercise data is available. Out in the real world, people enrolled in diet studies tend to cheat, adherence is poor, the drop-out rate notoriously high, and it is not uncommon for target diet macronutrient distributions to be missed. Studies are plagued by inaccurate recall and a host of other biases and confounders. There is no general agreement regarding the macronutrient distributions in diets described by terms such as low-carbohydrate or low-fat. Health effects of diets have traditionally been judged partly by changes induced in blood lipid levels, and even small and probably clinically irrelevant elevations of LDL were viewed with alarm. Diet studies that focus on one issue or item such as fat, protein, or some micronutrient are of questionable relevance since the components of a diet are eaten together and to are some extent synergistic.

The complexity of the problem of potential causes of obesity is illustrated by an incomplete list of suggested factors, either in general or associated with its prevalence or the rate of increase.4

• Genetic factors
• Thyroid dysfunction
• Gut bacteria problems
• Energy dense foods
• Urban environment influencing exercise
• Increased portion sizes
• Inexpensive energy dense food sources
• Increased use of high fructose corn syrup
• Junk food vending machines
• Decreased physical education (PE) in schools
• Mother's age at individuals birth
• Selective mating of individuals genetically predisposed to obesity
• Decline in hours of sleep
• Environmental chemicals that are endocrine disrupters
• Pharmaceuticals
• Ambient temperature
• Intrauterine effects
• Sedentary lifestyle, TV video games, heavy computer use
• Psychological influences on eating behaviour
• Excessive sugar consumption
• Excessive refined carbohydrate consumption
• Aggressive marketing of energy dense junk food
• Metabolic dysfunction influencing basal metabolic rate and thermogenesis
• Natural defences against starvation or chronic food shortages
• Positive energy imbalance

There is considerable overlap of items in this list. Some represent potential modifiable factors. Many lack an evidence base of much significance. The problem with this list is that quantitative translation and weighting to produce meaningful guidance that is tailored to any given individual is difficult. Thus there is the need to zero in on the items of the greatest importance.

Those who prefer a simple picture merely state that all that matters is the balance of calories in and calories out and that obesity arises from the inability to control intake which is frequently attributed to weak willpower or a character defect, and the sedentary nature of modern life. Unfortunately, this leaves a lot unexplained and is a serious oversimplification of human nutritional biochemistry and the underlying physiology and control mechanisms. A major aspect ignored by this simple picture is the ability of the body to compensate and produce the changes in metabolism and energy usage precipitated by weight loss which work against the dieter. There is also a significant limit as to what most will tolerate in terms of caloric deficit, especially if the intervention is long-term. Exercise is also not an option with some. Therefore there are significant barriers to applying the approach based on energy balance. However, there is probably no sensible and viable alternative.

An argument for excess food consumption being a root cause of obesity is found in a recent study which was able to account for the increase in population weight in the U.S. over 30 years based on an analysis of national food consumption. Swinburn et al estimate that the food energy intake in the 1970s was 2398 calories/day whereas in the 2000s it was 2895 cal/day. According to their mathematical models of energy balance and weight gain, this is more than enough to account for the obesity epidemic.5 Hall et al have used a similar model to examine the same question and find that government food consumption data from which they calculate the change in population energy intake appears to be in excess of that required to explain the epidemic. They suggest that the official figures underestimate food waste.6 The model used by Hall and coworkers will be discussed below. It is important to note that weight loss or gain is not linear and metabolic adjustments related to weight loss or gain impact energy utilization and intervene to alter rates of weight change.

In this review we will and see what we can find that is convincing and useful. The two major issues are preventing obesity and losing weight after becoming overweight or obese.

ENERGY (CALORIC) BALANCE

The fundamental principle here is the conservation of energy, also called the First Law of Thermodynamics. But the application is not simple. Food has a basic caloric content which can be measured by complete combustion to water, carbon dioxide and the end products from the nitrogen and sulphur content. This is done in a calorimeter, usually termed a bomb calorimeter because the reaction occurs at high oxygen pressure in a heavy steel container about the size of a large yogurt container. But this is not what happens to ingested food. The fat, protein and carbohydrate are not completely converted into molecules used by cells to generate energy. The conventional rules of thumb which takes into account the various inefficiencies comes up with protein and carbohydrate contributing 4 calories per gram and fat contributing 9 calories per gram (actually kilocalories, but we will use the nutritional simplification of terminology). Refined calculations identify poorly metabolized food components and use lower numbers. The point is that these are only average values and they are universally applied to estimate the energy intake of diets.

The energy from food intake is used for a large number of functions and processes. Basic to the argument of the conservation of energenergy school is that if there is energy left over it is stored as mass, generally fat, and increases weight. If there is a deficit, then one has a decline in weight. But the energy utilization processes are complex and subject to large individual variation.

The total daily energy expenditure while at rest is called the basal metabolic rate (BMR). This is the energy at rest used to keep the system functioning. The BMR actually accounts for more than 50% of the daily energy expenditure. Some consider it an attractive target for pharmaceutical intervention.

Finally, there is a rule of thumb used in nutritional analysis that an energy intake deficit of 500 cal per day is equivalent to one pound weight loss per week. This is of course a rough generalization, but is amazing how often one can use this number to rather accurately predict the initial weight loss in a study given the calorie deficit and the duration. This translates into 500 calories per day yielding 52 lbs per year. But a 500 calories deficit out of say 2200 is far from starvation, and yet it predicts 156 pounds in three years. In other words, this rule of thumb implies that the fixed calorie restriction will produce a constant rate of decrease in weight over long periods. This is not true since metabolic adjustments intervene due to the loss of both fat mass and fat-free mass and this energy deficit no longer produces the indicated weight loss. Studies suggest that the adjustments start within a few months.

The bottom line appears to be that if one can account for the utilizable energy that gets into the system and all the energy expenditure processes, then one is bound to find that energy is conserved. The problem is with the details.

With the advent of the Atkins diet and the subsequent explosion of related low-carbohydrate diets (LCD), interest gradually developed in conducting controlled studies to evaluate this diet and its officially sanctioned and promoted competitor, the low-fat diet (LFD). There have now been an amazing number of studies and space does not permit a detailed review. But a general consensus seems to be developing.7 It is consistently observed that LCDs result in more initial weight loss than LFDs. Many explanations have been advanced, but the only one that does not appear to be refuted is that the LCD reduces spontaneous energy intake frequently without the dieter being aware this is happening. When comparison studies carefully equalize energy intake and take into account physical activity, the weight loss differences become small or negligible, but this is very difficult to do in studies with free-living populations who are monitored by interviews. Studies of this question are also confused by LCDs that are not really low-carbohydrate. But the evidence at present shifts the emphasis back to calories rather than the distribution of macronutrients as the main concern in weight loss. But when one is considering weight gain, there is still the viscous circle of high intake of high glycemic carbohydrates stimulating insulin secretion and fat storage and ultimately increasing insulin resistance which makes matters worse.

The LCD appears to result in an improved lipid profile, especially with regard to triglycerides (dramatic decrease) and HDL (significant increase), and an improvement in indicators of healthy glucose metabolism. The very low carbohydrate diet has been the traditional (old fashion) approach to dietary control of blood sugar levels in diabetics and prediabetics, both short-term and long-term (HbA1c), although this is not recognized or promoted in modern times by professional organizations, probably because of the belief that the increased fat content is dangerous and the low carbohydrate aspect is viewed as unacceptable to many patients. Pills are the modern answer with insulin injections used to "cover" high carbohydrate intake.

It has become increasingly more difficult to justify opposition to the elevated fat content of the LCD. As Mozzaffarian and Ludwig point out in a recent commentary, the proportion of total energy from fat appears largely unrelated to the risk of cancer, cardiovascular disease, diabetes or obesity and saturated fat, targeted by nearly all governmental agencies and professional organizations, has little relation to heart disease provided one is dealing with most prevailing dietary patterns.8 A meta-analysis just published confirms this view.9 No significant evidence was found for the view that dietary saturated fat is associated with increased risk of coronary heart disease or cardiovascular disease and a study from Japan has recently demonstrated saturated fat was inversely associated with both hemorrhage and ischemic stroke subtypes.10 Many in mainstream medicine or nutritional science still do not agree with this view and the food industry is still capitalizing big-time on the notion that fat is bad.

THE KETOGENIC DIET

The extreme form of the low-carbohydrate diet is the ketogenic diet. Through a very low intake of carbohydrates, and a high intake of fat, the metabolism shifts to using fat as the principal energy source. The true ketogenic diet is generally associated with the treatment of epilepsy and was developed in the 1920s for pediatric application. The induction and low-carbohydrate phases of the Atkins diet resemble a ketogenic diet, but there is no limitation on energy intake aside from that inadvertently instituted by the patient. Atkins describes the indication phase in terms of lipolysis, the burning of fat and this process is naturally accompanied by the production of ketones, thus ketosis. This is not to be confused with diabetic ketoacidosis which occurs in insulin-deficient individuals with out of control blood sugar levels. In view of the studies that find similar weight loss from either low-carbohydrate or low-fat diets if there is identical energy intake and physical exercise, there would appear to be no need to push carbohydrates down to the point where ketosis can be observed, for example, through the presence of ketones in the urine. In the Atkins diet, the induction phase carbohydrate intake is about 20 grams, which can produce mild ketosis in some. By comparison, a low-carbohydrate diet with 20% or 30% of energy derived from carbohydrates would, for an energy reduced diet of say 1500 calories, involve 75 to 112 g per day which would not be a ketogenic diet and exceeds the carbohydrate intake during most of the Atkins protocol. Aside from bad breath and constipation, Atkins found no evidence of side effects that merited concern, and there have been a number of controlled studies of this diet which also do not report serious side effects.

Therapeutic ketogenic diets should only be undertaken under the supervision of a physician experienced in this approach. There are potentially serious side effects, the most common of which can be identified by various blood and urine tests.11

THE HALL-JORDAN MODEL

If it is indeed true that the two limiting types of diet, low carbohydrate and low fat, produce approximately the same weight loss and problems with sustaining the loss, and that differences are probably due to absence of equivalence in energy intake, compliance or physical activity, then the focus shifts simply to the caloric intake, body composition and its relation to metabolism, and energy expenditures through metabolism, exercise and daily physical activity. It is important to recognise that what happens during calorie restriction and/or increased exercise is not a steady weight loss but weight changes which steadily diminish, and if the calorie deficit is maintained, a new constant weight will be reached which represents a metabolic adjustment, mostly due to the decrease in fat mass. The failure of most studies to achieve this new equilibrium appears to be a manifestation of the short term of the intervention relative to the follow-up followed by an inadequate or flawed protocol for weight maintenance. In the model to be discussed, the final new steady weight is the principal issue, not how long it takes to get there.

A model recently published by Hall and Jordan in the American Journal of Clinical Nutrition addresses these issues in what appears to be a highly rigorous fashion and the end result is a set of online calculators which enable patients and/or their physicians to estimate the caloric deficit required to reach a new but sustainable weight.3,12 The input concepts and parameters are simple enough, but the mathematics which we will not to worry about is a bit complicated.3 This model is much more complex than the 500 calories a day equals a pound a week or the tables easily found on the internet giving the miles one must walk or run to for a given caloric expenditure.

The Hall-Jordan model takes into account the initial weight, the energy input from food, the thermic effect associated with the energy input, the fat mass and fat-free mass as variables during the weight loss, and the influence of physical activity, its dependence on body weight and its changes introduced by the weight loss protocol. The various weighting factors are derived from a number of relevant studies and the final model was validated by comparison with controlled diet studies where the required information was available. The agreement seems impressive. For the correlation of measured vs. predicted steady-state weight after intervention the coefficient was 0.83. It was 0.91 for the measured vs. the predicted change in energy intake required to achieve the observed change in body weight to achieve the new steady state. For studies on humans, these are remarkably good correlations. The authors provide a table as an example. Full text of this paper is in the public domain at the journal website.13

The online calculators based on the Hall-Jordan model are also of interest. Of the four with links on the cited web page,12 the one easiest to use is the one that only requires input of gender, age, height, initial weight and the change in energy intake contemplated (a negative sign before the number is required). This particular calculator also requires input changes in physical activity. This last factor is introduced with a parameter called PAL and the input form has a box providing guidance for this number according to the level of activity running from sedentary to very active. A number of results are then calculated, the most interesting of course being the new steady-state weight expected. Note that these calculators use the proper unit kcal whereas in this discussion we have used the term calorie as equivalent, in keeping with lay nutritional parlance.

The Hall-Jordan model appears to represent considerable progress with a very serious problem which is that if one looks at the hundreds of diet studies in an attempt to gain insight and guidance as to what to do to lose weight in general, the only result will be confusion. There is a grand mixture of junk science, studies with unrealistic protocols (total daily energy intake of only 500 calories, i.e. a deficit of 1000 to 2500 calories per day), macronutrient distributions that are all over the map and frequently identified with vague terms, etc. A real zoo indeed. In contrast, the Hall-Jordan model relies on only "gold standard" studies of energy expenditure, detailed input from a number of studies of the physiology of nutrition, and the model takes into account the metabolic changes that accompany the loss of fat mass and the relationship between weight at any time during the diet and physical activity energy expenditure. The integration of all these concepts into a plan for weight loss requires a mathematical model which in would be very difficult to apply without the aid of a computer. Readers familiar with mathematics can view the equations in the cited paper.

Consider the following example based on this model. An obese man (BMI = 35) weighs 230 pounds (105 kg) and wishes to have a BMI indicating not overweight. To achieve a BMI of 24 requires the loss of 72 pounds or 33 kg. According the model, this will require a decrease in energy intake of about 600 calories per day in order to eventually reach the new equilibrium weight of 158 pounds. It is assumed that there is no change in physical activity. This model does not predict the time required. By increasing the calorie deficit the weight loss should be more rapid and then as the target is approach, the calorie deficit can be reduced and allowed to approach the new equilibrium value of 600, but the implication of the model is that one is stuck with this new energy intake if it is desired to maintain the new BMI.

Hall and coworkers have also examined the predicted time course of weight loss based on extensions of their steady-state model discussed above. For example, if a 100-kg individual were to be subjected to approximately a 500 cal energy deficit, two different models both predicted the approach to the new steady weight would be about 50% in one year, 85% in two years and 92% in 3 years. The new equilibrium weight was about 74 kg. A discussion of the details of the models is beyond the scope of this review.14,15 The researchers point out that the slow approach to a new steady state for humans has been found in a number of studies. In the example cited, extrapolation of the initial drop over the first few months yields a one-year weight that approaches the 24 kg that the 500 calories = 1 pound rule of thumb would predict at one year, but there is continuous compensation and it takes three years rather than one to reach a weight loss of about 24 kg or 53 pounds. These numbers are all approximate and are taken from graphs. As pointed out above, diet studies rarely keep the initial protocol beyond 6 months.

Hall and Jordan comment that since reaching a new desired weight that requires a substantial weight loss may require several years, consideration of an accelerated weight loss phase followed by a weight maintenance phase may be appropriate. In their model they provides guidance as to the dietary and activity changes necessary for weight maintenance, something for which no other clinical tool is currently available.3 This allows the estimate of the extent to which a caloric deficit in use exceeds that required to maintain the weight at any stage in the decline.

There are a number of online calculators that determine the energy burned by exercise. For example, one calculator found that if one weighed 230 lbs, walking 45 min at a moderate rate of 3 miles per hour yields 272 calories burned or about 1900 calories per week. Using 3500 calories burned as the equivalent of one pound, then this activity provides about an additional half pound compared to no exercise. While this may not seem like much, if one is on a diet with a reduced energy intake of 500 calories, then the initial rate of weight loss will be about one pound without exercise, but 1.5 pounds with exercise. It becomes significant. Results from these online calculators should be considered only approximate since the weight loss associated with exercise varies with weight.

The cited paper for the Hall-Jordan model does not discuss potential reasons for failure. The studies used to obtain weighting factors and other parameters presumably did not include individuals where there were unusual difficulties in achieving weight loss. The factors that might cause failure merit a brief discussion.

In his classic book Dr. Atkins' New Diet Revolution, a chapter is devoted to what he terms metabolic resistance. Over the years in his practice, Atkins treated a large number of patients who had great difficulty losing significant weight independent of the program they used, including his. He lists the major factors that can cause metabolic resistance.

• Excessive insulin resistance (low insulin sensitivity)
• Use of certain prescription drugs or hormones
• Hypothyroidism which may not be evident from blood tests

Prescription drugs most likely to interfere with weight loss include estrogens such as in hormone replacement therapy and birth control pills, antidepressants and antipsychotics, insulin and insulin stimulating drugs, steroids, diuretics and beta-blockers, antihistamines, and protease inhibitors (HIV and hepatitis medications).4

An under-active thyroid, generally called hypothyroidism, may be responsible for more obesity and resistance to weight loss programs than is generally recognized. In simple terms, an under-active thyroid slows metabolism and makes one more resistant to weight loss. Anecdotal reports suggest that even diets restricted to less than 1000 calories a day combined with walking several miles a day can fail to reverse weight gain when one has an under-active thyroid. For hypothyroidism, the first challenge is diagnosis since the standard blood tests, according to Atkins, do not catch all cases. Dr. David Brownstein, author of Overcoming Thyroid Disorders, strongly agrees and regards hypothyroidism as seriously under diagnosed. The conventional approach involves the thyroid stimulating hormone assay (TSH). An elevated level is an indication that the pituitary gland is sensing a low thyroid hormone levels. The standard reference range used is that TSH under 4.5 mIU/L is normal. But as Brownstein points out, there are many physicians and organizations that believe this upper limit is too high. He proposes a threshold for suspecting hypothyroidism at 2.0 mIU/L but other suggest 3.0 mIU/L. In Brownstein's practice, he finds the range of TSH of 0.3 to 2.0 mIU/L as indicating optimal thyroid function. In diagnosing hypothyroidism, he uses thyroid status blood tests, medical history, basal body temperature and a general physical exam. The blood test looks for elevated TSH and low levels of the thyroid hormone T3. Individuals can have normal TSH and T4 (the other thyroid hormone) and yet be have an under active thyroid because of poor conversion of T4 to T3. His protocol for measuring basal body temperature involves an armpit temperatur e measurement upon awakening but before getting out of bed. The measurement should last 10 min and be repeated for 5 consecutive days. For premenopausal women the temperature study should be started on the second day of menstruation. For everyone else, when to start makes no difference. A temperature below 98.2 degrees F (36.8 degrees C) may indicate hypothyroidism. The lower it is, the stronger the indication and degree of hypothyroidism.

Brownstein lists a number of factors that can inhibit T4 to T3 conversion. It is interesting that medications mentioned above that are associated with weight gain such as beta blockers, birth control pills, estrogen, and steroids appear in this list. He also mentions obesity which suggests a positive feedback component to the thyroid-obesity connection.

Clearly, individuals having trouble with the diet/exercise approach to weight loss need to worry about hypothyroidism and also about false reassurances provided by the normal laboratory ranges from blood tests and the conventional wisdom. Readers are referred to Dr. David Brownstein's book for a complete discussion of the diagnosis and treatment of hypothyroidism. Based on his clinical experience, the favored approach for treatment involves using dried (desiccated) glandular thyroid products which he prefers over the commonly used T4 derivatives. In his book, Atkins describes experimenting with escalating doses of the thyroid product to test the hypothesis that hypothyroidism is a cause of difficulty in weight loss.

NATURAL NON-DIET APPROACHES

The widespread difficulty many overweight and obese individuals have losing weight has made finding a prescription drug a high priority in the pharmaceutical industry. The drug discovery philosophy follows the standard paradigm of targeting some process involved in some aspect of the problem and designing an inhibitor or enhancer. Pharmaceuticals appear to have side-effect problems. Approaches using natural products employ the same paradigm. The most high-profile ones are as follows:

• White kidney bean extract (Phaseolus vulgaris). This extract contains an inhibitor of alpha-amylase, a digestive enzyme responsible for the breakdown of complex carbohydrates such as starches into simple sugars which can then be absorbed in the small intestine. It is marketed as a dietary supplement Blockal and the trade term of the enzyme inhibitor is Phase 2 or Phaseloamin 2250. There appear to be only two randomized studies. In one, treated subjects lost on average 6.5 pounds whereas the placebo group lost 1.45 pounds over 30 days.16 In the second study, a white bean extract prepared for the investigators or a placebo was given for 30 days to a group involved in a multicomponent weight-loss program. The treatment enhanced the weight loss from 4.7lbs to 6 lbs.17 Toxicity studies of Blockal involving high doses in rats suggest this extract is safe to use (on rats).

• Glucomannan. This is a gel producing, viscous water-soluble polysaccharide that comes from a plant root (konjac root). Glucomannan appears to reduce appetite because of its powerful bulk forming properties in the stomach. It decreases the rate of absorption of nutrients and thus decreases the glucose and insulin response to meals. There is also some evidence that it acts through promoting satiety and fecal energy loss.18 The reduction in the increase in blood glucose after eating can be considerable, e.g. by one-half.19 These hydrophilic polysaccharides are of great interest to the pharmaceutical industry with hundreds of papers published each year. Glucomannan, which can not be patented, is widely used in Asian countries and in the US for food thickening, improving bread texture, as a dietary fiber, etc. Toxicity studies are limited but it is a traditional food additive in China and Japan.20 Keithley and Swanson review twelve clinical trials. At doses of 2-4 g/day glucomannan was well tolerated and resulted in significant weight loss in both obese and overweight individuals.18 Glucomannan has also been observed to benefit constipated adults.21

• The seed extract from the West African plant Irvingia gabonensis. This extract from the seeds of a mango-like fruit also goes by the identifier IGOB131. It is thought to inhibit fat cell formation and to increase adiponectin levels and decrease leptin levels. The former is secreted from fat cells. Its increase may help induce weight loss. Leptin helps control food intake and energy expenditure.22 Egras et al23 recently reviewed studies of this extract for inducing weight loss. Two randomized controlled studies were identified. In one, after 12 weeks of treatment there was a 12% decrease in body weight and a 10% decrease in body fat compared to the placebo group. In a one-month study there was a 5.3% weight decrease compared to a 1.32% decrease in the placebo group.

• Green tea extracts. It has recently been demonstrated that green tea extracts taken orally can increase the resting energy expenditure and thermogenesis (production of body heat linked to the oxidation of body fat). This has led to studies of its potential for weight loss.24 The active ingredients are thought to be catechins including the extensively studied epigallocatechin-3-O-gallate (EGCG). A recent study by Di Pierro et al24 is of particular interest. One hundred individuals on a low-calorie diet were divided into two groups, one receiving a coated tablet containing highly bioavailable green tea extract, the other serving as a control. After 3 months, the treatment group lost 30 lbs. vs. 11 lbs. for the controls. Significant benefits in the treatment group compared to the controls were also seen in waistline, BMI, and cortisol levels. HDL increased by over 21% in the treatment group compared to 10% in the controls. Leptin, which was not measured in the control group, was dramatically decreased in the treatment group. These results were achieved with a green tea preparation (GreenSelect Phytosome) with greatly enhanced bioavailability. No adverse side effects were observed to be associated with this product, which Di Pierro et al claim is consistent with unpublished animal toxicity data. The importance of bioavailability is clear from a very recent systematic review and meta-analysis of green tea catechins and weight loss where only modest effects were found.25 Either green tea itself or poorly absorbed extracts were used in all the studies in the analysis. Thus there appears to be only one study indicating strong benefit simply because there is only one study that achieved high enough absorbed levels of EGCC and other catechins.

This approach attempts to replace or enhance energy intake restrictions with a somewhat artificial intervention, and if the protocol is terminated, then to maintain the new and perhaps more easily achieved weight it is necessary to find the new equilibrium energy intake for a given level of physical activity. It can be argued that it is not a good idea to directly interfere with digestive or metabolic processes by consuming chemicals, which while natural, are not necessarily present in our natural diets, at least not at the concentrations required for promoting weight loss, and it does not appear to make sense to do this as a life-long approach to weight control.

MEDICAL APPROACHES

The pharmaceutical industry is well aware of the problems associated with loosing weight through diet and exercise. The solution is dictated by the present drug development paradigm discussed several times in the Newsletter, i.e. use chemicals to block or inhibit, the same way the natural extracts work. This approach has been nothing but a disaster. Most recently, sibutramine (Meridas, originally used as an antidepressant) has been withdrawn due to unacceptable side effects. Among other things, it increases blood pressure and heart rate and also increases serious cardiovascular events. Most readers will remember the withdrawal of fen-phen in the 1990s. Recently full clinical trials of rimonabant, briefly marketed in Europe but not in the U.S. were discontinued. Another new drug as well as a combination of two drugs were turned down by the FDA because of concerns over a number of side effects. Those involved in "treating" the obese despair over a nearly empty toolbox. One can argue that the experience with diet drugs is just one more example of the flawed paradigm that requires meddling with biochemical pathways and processes using synthetic chemicals foreign to humans, meddling that clearly yields unintended results frequently unrelated to the target disorder and far removed for the site of the problem. Over the years there have been countless examples.

If diet drugs are required to realistically address the fact that most overweight and obese individuals can not accomplish and maintain significant weight loss by exercise and diet, then the pharmaceutical therapy presumably would be for life. Otherwise it would be a component in a yo-yo scheme. But long-term safety can not be tested in humans prior to regulatory approval, so there are unknown and potentially terrible risks with this approach. Some feel that the obesity crisis is such a serious public health threat that diet drugs should be allowed on the market for formal post-release safety trials which they point out may offer the only hope for discovering serious side effects. Industry support of such studies would be a risky business, given the history of forced drug withdrawal or restriction due to side effects that turned up even in lax and informal post-approval surveillance. Lifelong use is the dream of the industry but may be a nightmare for the patient.

Then there is the brute-force approach, bariatric surgery, which limits the stomach capacity.26 This definitely leads to highly significant weight loss and favorable changes in a number of risk markers, but most would consider it as a last resort. Once one has reached a desired weight and has the procedure reversed (if possible), then the energy intake to maintain the new weight becomes a big issue. The type of bariatric surgery called laparoscolaparoscopic Roux-en-Y gastric bypass has proved quite successful in producing dramatic weight loss with almost all patients achieving greater than 45% excess weight loss with mean changes typically of 40-50 kg (88-110 lbs).

Finally, it is being seriously suggested that a technique used successfully in Parkinson's disease called deep brain stimulation has potential in treating obesity.27 Appetite and satiety centers in the brain have now been well documented and offer the possibility of electrical manipulation. Electrodes are inserted into the appropriate spots in the brain and connected to an implanted device that is remotely controlled. While there is considerable experience with this technique for treating Parkinson's, there are no studies regarding efficacy for obesity.

CONCLUSIONS

The general conclusion seems to be that energy intake and energy expenditure, either internal or external, are the controlling factors on which weight control and the reduction in obesity must rely. For weight loss, the calorie restricted low-carbohydrate diet has merit over the calorie restricted low-fat diet for reasons associated with triglycerides, HDL cholesterol and insulin resistance, to which should be added enhanced satiety which makes calorie reduction easier. Physical activity must be factored in but dietary calorie intake has the potential for a much grater impact simply because exercise equivalent to a 500 or 1000 calorie daily deficit is not possible for many individuals.

The apparent equivalence of low-carbohydrate and low-fat diets does not invalidate the view that diets high in high-glycemic foods can have a deleterious effect on weight control through the impact on insulin levels and ultimately insulin sensitivity and thus insulin resistance. This involves the weight gain aspect of this subject directly, but insulin resistance can impact the success of weight loss interventions.

The guidelines developed by Hall and Jordan apply to the average individual and will require modification for some individuals. As with most chronic disorders, obesity is best prevented, not treated, and when the BMI enters into the overweight range of 25 or greater, this is the time to take action. Thus an important component of remaining healthy at all ages is to make weekly or monthly use of a scale and to act when weight creeps up by reducing energy intake and/or increasing physical activity until the increase is halted and the BMI returns to "baseline", which should be the lifelong target. Individuals for whom this does not work, or those who can not lose weight no matter what protocol they follow, then it is urgent to investigate other solutions. Hypothyroidism should be high on the list of suspects.

The emphasis on calories should not distract the dieter from concerns over what constitutes a healthy diet and it would appear important to include at least a typical multivitamin in the protocol. But there is considerable evidence that men and postmenopausal women should avoid iron containing multivitamins unless they are anemic, and natural folate is much to be preferred to folic acid. Attention also needs to be paid to the long-chain omega-3 content of the diet and supplement regime, as well as the vitamin C, magnesium, coenzyme Q-10 and vitamin D intake since desirable amounts are not generally provided by the typical multivitamin. Healthy diets and healthy diet patterns have been discussed repeatedly in this Newsletter.

Finally, the underlying assumption in this review is that obesity is bad. This ignores the well known paradox of the so-called healthy obese. They have high BMI but normal blood lipids, normal blood glucose metabolism which translates into normal fasting bloods glucose and insulin levels and a low HbA1c, and normal blood pressure. While they are lucky, there remain the majority who are no so fortunate. The evidence that obesity increases the risk of chronic degenerative and cardiovascular diseases is overwhelming. Obesity is associated with diabetes, and diabetes in turn is associated with not only elevated cardiovascular risks but vascular problems in general. The end result can have a much larger impact on the quality of life than a non-fatal heart attack as is illustrated by the prevalence of diabetic vascular problems which lead to visual problems as well as circulation problems ultimately requiring the amputation of toes, feet and even legs.