Obesity in the human has been widely studied by such authorities as Newburgh and Rony. It is generally accepted that fat in excess will be laid down only if food intake exceeds energy output. The treatment of obesity has generally followed this premise. Diets deficient in calories have been prescribed so that caloric intake does not exceed energy output. Weight loss should automatically follow when the instructions are faithfully followed. Indeed formulae have been devised to predict the loss in weight on a measured low caloric intake of a candidate of known height and weight. These low caloric diets are made up so as to be deficient in fat and carbohydrate and with protein at approximately 1 g. per kg. of body weight. It has been shown that weight loss can be achieved in this manner. The diet is followed and the desired results are obtained. Unfortunately, it is difficult for most patients seen in clinical practice to follow a low caloric diet. The literaturre is replete with instances of diet failure on such a regimen. The difficulty is in part due to inability to control appetite. Anorectic agents such as amphetamine, phenmetrazine hydrochloride and bulk substitutites have been utilized as a means of controlling appetite. These are of some value in the clinical mangement of an obese patient. Methods other than those of controlling appetite have also been applied. These include the administration of thyroid extract, the effect of regular exercise and psychotherapy administered both individually and in groups. However, in spite of all these methods, the long-term management of obesity presents many disappointments. Patients often lose weight only to regain it after a short interval. In many, weight loss is never achieved. 

Means other than the aforementioned have long been sought in the control of appetite. Appetite and satiety, i.e. the satisfaction of appetite, are complex problems. The latter, satiety, is dependent upon many variables. One of the chief factors is the production of body heat by the specific dynamic action of ingested food. Protein has much the highest index in this regard, while fat has the lowest." Rise in skin temperature and a resulting feeling of warmth are intimately correlated with the feeling of satiety. In fact, it has been suggested that the obese are slower in showing this rise, hence their desire for more food. Another theory relating to satiety is that of the arteriovenous (A-V) glucose difference and its regulation of glucoreceptors in the brain stem. Mayer feels that the glucoreceptors are the controlling centres of appetite and satiety. It is stated by others that satiety depends only on the body's caloric needs and the subsequent voluntary supply of calories. A most attractive hypothesis, well documented by physiological study, is that which proposes that satiety is experienced because the stomach is full. Nervous impulses are sent out to the brain when the stomach is filling or full and the sensation of satiety results. Satiety may then be related to many factors of the diet. If the bulk of food, its protein and its fat intake are increased, on the basis of some of these theories satiety may then more readily follow. Bulk, increased intake of fat and protein, and thus satiety, are not possible with the usual low caloric diet. 

As a diet for achieving satiety while effecting weight loss, the low carbohydrate diet of Pennington shows some promise. This diet allows as much bulk as desired. It is high in both fat and protein. Such a diet meets many of the requirements for achievement of satiety. It provides plenty of protein to be used for heat production by the body. Calories are supplied by the high fat intake and filling of the stomach is achieved by the usual bulky nature of the diet. Pennington claims that his patients have lost weight on this diet with a caloric intake of 3000 calories. Another consideration is that of a fat-mobilizing hormone which has been reported to be present in the urine of patients on this type of diet. Urine extracts from such fasting patients have been shiown to produce weight loss when injected into mice. Unfortunately, verification of this work has not as yet been reported by others. One may anticipate that with such a diet hunger may be avoided, appetite satisfied and a measurable weight loss achieved. The diet is not easy to follow. Its most important requirement is the strict necessity of restricting carbohydrate intake to less than 50 g. per day. One may consume as much fat and protein as desired to produce satiety. Of course diets high in fat and protein, and therefore meats, are somewhat expensive. These may be out of the reach of some economic classes. Other ethnic groups long accustomed to high carbohydrate intakes, such as Italians and Chinese, may find such a diet highly unpalatable. However, most well-motivated patients are prepared to follow such a diet. 

METHODS 

Forty-eight obese individuals were selected. These were patients attending a private practice, an industrial medical centre, or the outpatient clinic of a hospital. All expressed a desire to lose weight. A copy of the diet was given to each patient. The diet was made up to allow protein and fat ad libitunm. However, the carbohydrate component was carefully restricted to less than 50 g. per day. The object of the diet was to provide as much bulk as desired but at the same time to limit sharply the carbohydrate intake. These basic points were outlined to each patient. There were no other diet restrictions. Copies of the diet were mimeographed along with suggested menus for each meal. The patients were instructed regarding the approximate values of the usual daily foodstuffs. The high protein and high fat content foods were selected as being most useful for this type of diet. The whole regimen was reviewed with the patient after the diet had been followed for some weeks, so as to correct any misunderstanding that might have arisen. The patients at the outset found the concept of an ad libittum diet difficult to understand. However, with time they realized that the guiding principle of the extremely low carbohydrate intake (less than 50 g. daily) had first to be strictly maintained. They could then realize satiety by taking as much fat and protein as required. The patients' weights varied from 140 lb. in a young woman whose height was 58 in. to 274 lb. in an elderly woman 70 in. in height. The patients ranged in age from 16 to 62 years. They all fulfilled the true definition of obesity, being 20% more than the ideal (provided by the Metropolitan Life Insurance tables) weight for their height. Their weights were checked at monthly intervals for three months to one year. A small group, eight patients in all, were followed up for a two-year period. One patient was studied while in hospital and balance studies are available in this case (Fig. 1). The patients served as their own controls, since all had been on a low caloric diet without measurable success. At least half had used anorectic agents, seven patients had taken bulk substitutes, and eight had participated in group psychotherapy for a period of eight months. None of them showed a sustained loss of weight. 

RESULTS 

Forty-eight patients were seen initially. Of these, eight rapidly loist interest and did not wish to carry on with the diet after the first month. All of these patients complained, nonetheless, of the monotony of the diet, its constipating effect, the absence of taste and its failure to satisfy their desire for sweets. Of the remaining 40 patients, 12 felt that they were following instructions faithfully but did not lose weight. The remaining 28 patients achieved satisfactory weight loss during the period of at least six months in which the diet was followed. This loss varied from 10 to 40 lb., averaging 11/ lb. per week. Nine of the 28 patients who lost weight were able to reduce their weight to ideal chart indices.19 The others, although showing considerable weight losses during the period of study, did not reach this desired level. Results in the single case under balance study are shown in Fig. 1. The balance study was carried out on a woman (E.C.) who initially weighed 83 kg. (182 lb.). She was allowed a free diet, for the first seven days. It will be noted that the caloric intake was approximately 2800 calories and that little change in weight occurred. There was a substantial fall in weight from day 7 to day 15 when a low caloric diet of 1000 calories was taken. The high protein and fat diet of Pennington with only 50 g. of carbohydrate was followed for the final period from day 15 to day 24. The caloric value during this period was in the neighbourhood of 2000. There was a weight loss of at least 1 kg. (2.21 lb.) and, interesting to observe, a negative nitrogen balance and a positive sodium balance. The patients who did achieve weight loss showed a substantial fall as. illustrated by a representative case (Fig. 2). All patients, including those who dropped out of the study, expressed similar opinions regarding the diet. They agreed that it was monotonous and constipating. Many missed sweets and the oral satisfaction derived from sweets. However, none of the patients experienced hunger. since they were free to eat protein and fat at will. Hunger had been a factor to most of them on lowr caloric diets and they were quite familiar with this form of nagging discomfort. The new diet was preferred by them, if only for this reason. The eight patients followed up for two years maintained their weight loss w\hile following the diet. DIscussIoN The treatment of the obese patient has followxed a stereotyped pattern for the past 20 years. Prescribing a simple low caloric diet and sympathetic handling of the patient, the usual metlhod, had not been a rewarding form of clinical treatment. Usually, the patient was disturbed by a continual g,naving sense of hunger.0 Attempts to overcome the feeling of hlunger by the use of anorectic drugs and bulk substitutes have been found of value for limited periods.4 5 The use of food high in protein and fat in order to overcome hunger does not at first glance appear to be a likely treatment for obesity. However, such a diet, high in protein and fat but low in carbohydrate, has been suggested by Pennington, who has reported that weight loss can be achieved by such means.16 17 Pennington also has submitted the following theory in an attempt to show that fat and carbohydrate are metabolized in a different manner by obese as compared to normal subjects. A partial block in carbohydrate metabolism at the pyruvic acid level is postulated. Pyruvic acid becomes converted to fat. Glucose intake is increased in an attempt to overcome the block. Obesity results because of the increased intake and consequent fat deposition. By inhibiting glucose intake in the obese, Pennington feels that energy will be derived not from glucose but from fat (ketogenesis). Weight loss in the obese on such a diet is achieved through fat breakdown. The evidence for this theory is hardly complete. Our results do show that satisfactory weight loss may be accomplished by a full caloric, low carbohydrate diet. The patients ingested protein and fat as desired. Careful attention was paid to keeping carbohydrate intake to a minimum. It has been suggested that the diet was unpalatable and the caloric intake was unconsciouslv restricted for this reason, although the builk may have appeared to be sufficient. Another criticism might be that even if the total bulk and caloric intake were ingested, complete absorption may not have taken place. The answer to these points may be discussed in the light of the vork of Kekwick and Pawan,20 who have shown that obese patients will lose weight with diets of 1000 calories. Surprisingly, the rate of weight loss was increased when the composition of the diet was altered from the usual low caloric one to one predominantly made up of fat or protein. They also showed that more liberal diets, of approximately 2000 calories, sufficient to maintain an even weight in obese patients, would result in weight loss if this same caloric intake was altered to a high fat or high protein content of similar caloric value. Balance studies performed during the period showed that complete absorption occurred during the period of high fat or protein ingestion. They suggested that some aspects of metabolism are different in the obese as compared to the normal and that alteration in composition of food may alter energy output in the obese. Our results are compatible with these findings. The same studies have been extended by Pilkington,21 whose group has shown that obese patients on 1000 calorie diets consisting mainly of fat or protein, for long periods of time, lost weight at a constant rate. They found that after an initial rapid weight loss a steady state was achieved if the diet was continued for a sufficiently long period, usually months. The weight loss paralleled that seen in the usual isocaloric 1000 calorie diet consisting mainly of carbohydrate. One mtust bear in mind that these vere 1000 and not 2000 calorie diets. The detailed study on the single patient described in this report shows that weight loss occurred on a full caloric intake, consisting of high fat and protein and low carbohydrate content. The sodium balance was positive and the nitrogen balance negative during the periods of free diet and low caloric diet. However, while on the Pennington diet the sodium balance was negative and the nitrogen balance was positive. Although one is tempted to attempt it, it is not possible to interpret these findings decisively. Shifts in mineral balances are commonly observed phenomena in the obese when the caloric intake is manipulated. The patients who were successful in losing weight all did so on a liberal diet which prevented hunger and provided for satiety. All the other methods of weight reduction mentioned earlier have been utilized by the author in the past. The diet discussed was found to be the most satisfactory of all these methods in our hands. Weight reduction occurred dramatically with a rapid fall early and then proceeding slowly but surely. Only nine of the 28 were able to reach ideal weight indices.'9 The others did not do so well but did achieve significant weight losses. It is our feeling that the usual low isocaloric diet would be necessary to bring these remaining 18 subjects down to ideal weight indices, but this is not an established fact. As stated, the patients found this method of losing weight superior to others. They did not suffer from hunger, felt satisfied most 'of the time and were free to reach for food at any time. They found this to be of immeasurable comfort and thus they were able to lose weight to a greater degree and for a longer period of time than they had heretofore realized. The results reported indicated that a greater number of patients will follow such a diet for a long period with satisfactory achievement levels. 

SUMMARY Twenty-eight of 48 patients succeeded in losing weight on a liberal caloric diet high in protein and fat and low in carbohydrate, as proposed by Pennington. The results are discussed in the light of recent metabolic studies in obesity by Kekwick and Pawa.