Answers are grounded in this paper and cite their sources.
Most clinicians have met the patient with overweight, who repeatedly and convincingly states that he or she eats very little, still cannot lose weight or maintain weight loss. In the scientific literature there is an ongoing controversy whether such patients manage with an increased metabolic efficiency or whether the true energy intake has been underestimated. Many researchers argue that it is impossible to estimate the energy intake in free living individuals over any longer period of time with necessary precision. An argument that this might be true stems from the studies by Acheson et al., who monitored the energy intake of cooperative, motivated, normal weight members of an Antarctic expedition [1]. Such experimental conditions certainly eliminated the need to control food being eaten outside the research unit. When the 24 h energy intake, recorded objectively, was related to the spontaneously reported intake, the mean underestimate was found to be 34%. In a pilot study of moderately obese women at our Obesity Unit, Karolinska Hospital, Stockholm, we obtained 24-h dietary histories and found a mean intake of 2302 kcal/day [2]. This is a reasonably high figure, which could indeed have been accepted as an adequate daily energy intake in this patient category. But when the knowledge to elicit such dietary information was subsequently applied to an identical group of obese women, we found a daily intake of 2805 kcal/day, an increase of 22%. This suggests that detailed interviews by skilled dieticians are necessary, if 24-h recalls are to be relied upon at all, and indicates that in some obese individuals, high daily energy intakes may certainly be identified. Recently, interest has been focused on ‘weight cycling’ as a phenomenon that may lead to an increased metabolic efficiency and thus at least in part explain the conflicting results reported. There is as yet no established definition of the term weight cycling. In epidemiological work, weight cycling can be described as a mathematically defined variation in body weight around a certain set point. In clinical studies, individuals have been operationally defined as weight cyclers by identifying those who have repeatedly lost and regained a certain amount of body weight. In animal experiments weight cycling has been achieved by repeated semistarvation and refeeding. In the United States, a Weight Cycling Project has recently been initiated, headed by a group of scientists with varying backgrounds from psychiatrists to enzymologists. This multi-disciplinary project approaches the problem in several parallel ways, from the animal experiment to the epidemiological analysis. An interesting pattern seems to emerge and a possible causal relationship between weight cycling and reduced energy needs has been suggested. In animal experiments, rats have been treated with periods of dietary restrictions and compared to a control group. In one such study it was demonstrated that with each weight reduction cycle, designed to imitate the slimming procedure in humans, weight loss was slower and weight regain faster [3]. Lean Zucker rats have been found to show an increased food efficiency during ad-libitum feeding after four cycles of food restriction and refeeding [4]. In other studies, these findings have not been obtained in fat Zucker rats, adult lean Wistar rats or obese ovariectomized rats [4–6]. Clinical studies of obese humans, who received an energy restricted diet in a metabolic ward, have on the other hand suggested that weight loss was less during the second dietary period compared to the first [7]. The potential importance of variations in body weight on resting metabolic rate has recently been studied in a group of individuals, typically known to vary considerably in weight. High-school wrestlers may undergo variations in body weight of a mean of 3 kg. They generally reduce their weight within 10 days prior to the sport events to improve their chances in the competitions. Wrestlers may undertake such weight reductions several times during one season. Brownell's group recently studied the resting metabolic rate in a sample of 14 such wrestlers and 13 controls [8]. The wrestlers who were cyclers and the non-cyclers did not differ in age, weight, height, body surface area, lean body mass or percentage body fat. However, in cyclers a significantly lower metabolic rate per kg of lean body mass was found, and the resting energy expenditure was 14% lower in these cyclers than in the controls. These data are suggestive and provoking but, as the authors are anxious to point out, do not prove cause and effect. It could be argued that those wrestlers who have lower energy requirements have to take more drastic measures to cut their weight in order to be competitive. The mechanism by which weight cycling may result in reduced metabolic rate could be the following: although in wrestlers much of the weight lost is water, the food restriction used to cut the weight will result in a breakdown of lean body mass. When refeeding occurs it is not certain that the energy excess will be stored in lean body mass again, but may in part appear as fat deposition. The resting metabolic rate is closely correlated to lean body mass, but not to adipose tissue. Repeated cycles of lean body mass breakdown with an ensuing shift from protein to fat deposition may explain the subsequent reduction in resting metabolic rate. Although this hypothesis is attractive, the data of Brownell et al., as they themselves indicate, do not fully support such a concept, since body composition was found to be similar in both groups. But it is possible that although the total amount of fat was similar in cyclers and in non-cyclers, the distribution may have been altered. This question was not addressed in the wrestler study but seems to be an important problem to resolve in forthcoming work. Several studies, in particular by the Gothenburg group, have stressed the importance of fat distribution for the metabolic consequences [9]. It is possible that fat which has been lost in one region will be regained in another, with ensuing metabolic consequences in spite of an unchanged total fat amount. A shift of fat from peripheral locations towards the trunk might have such effects. It is obvious that more information is needed in order to clarify the role of weight cycling for the development of body weight in man. Since it is difficult to predict who is going to develop obesity, prospective studies are difficult to design, and studies of ex-obese individuals cannot distinguish cause and effect of weight changes. While we are awaiting further information, data from both Framingham and Gothenburg about the effects of weight cycling on coronary heart disease cause some concern [10, 11]. Both these preliminary studies suggest that weight variation is a risk factor for mortality from coronary heart disease for both men and women. These data suggest that individuals who have undergone repeated weight cycling by means of so-called ‘yo-yo’-dieting may be at greater risk than those who maintain a more stable body weight, even if that body weight is higher. It could be speculated that an accelerated atherogenic process, which is not completely reversed when subsequent weight loss occurs, could increase the likelihood of coronary heart disease in intermittent dieters [10, 11]. If these speculations were found to be true, the findings of the risks involved in ‘yo-yo’-dieting will have important implications, both for all those who wish to lose weight, and for those who are to help them in their efforts. Weight loss must be taken seriously, both by the individuals who consider it and by those — medics and paramedics — who recommend it. Weight reduction programmes should never be initiated when the likelihood of success is small. Long-term approaches with therapist involvement, probably over years, may become necessary. A moderate but sustained weight loss is worth more than a great weight loss which is rapidly regained. Weight reduction goals should be realistic and developed in professional ways. And finally — if the ‘yo-yo’-effects so well-known from all spectacular fraud diet programmes turn out to have truly negative effects on the cardiovascular system and not only lead to frustration and feelings of low self-esteem in desperate obese individuals, this makes it even more compelling to prevent such fraud activities, whenever possible.
No takes yet. Share an insight, caveat, or question.
Stephan Rössner (1989) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: