In this narrative review we sought to show the effects of WC on body composition in patients with a wide range of age and body weight. Moreover, we would evaluate if WC may represent a risk factor for sarcopenia or sarcopenic obesity. Both voluntary and involuntary weight loss followed by weight regain in the elderly were included.

A bibliographic preliminary search was performed through the main medical search engines, using the following main keywords: weight cycling or yo-yo diet and weight loss, weight regain, body composition, FM, body fat percentage (FM%), FFM or lean body mass (LBM) as reported by the authors. This search was followed by a manual, in-depth study of articles and their respective references. We decided only to include paper in English, published between 1989–2024, describing studies that used validated body composition methods (i.e. excluding those using skinfold thickness measurements), with data about FM and/or FFM. We did not included studies conducted on athletes, since they are not generalizable to the whole population and that with weight loss induced through the use of drugs or bariatric surgery.

After selecting articles based on their relevance to the research questions, 26 papers were finally included in the critical synthesis and revision of the literature (Tables 1 and 2) [31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56].

A large number of papers analyzed body composition changes in weight cyclers and had conflicting results. Differences in study characteristics may, in part, explain the discrepant results of the literature.

Tables 1 and 2 summarize the main results of the articles selected through database searches and present them by publication date [31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56]. Studies with voluntary [31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53] and involuntary [54,55,56] weight loss and regain are separately presented in Tables 1 and 2 respectively. The main characteristics of the study sample, WC definition, study design and type of intervention as well as methods of body composition used, were summarized in the two tables to facilitate the comparison between different studies and the interpretation of the results.

Most of the papers that present data on the effects of WC on body composition were conducted in in patients with overweight and obesity [31, 33,34,35,36,37,38, 40,41,42, 44,45,46,47,48,49,50,51,52,53], involved in weight control programs, and only a few in normal weight subjects [32, 39, 43].

The majority are represented by intervention studies, where subjects with overweight and obesity, undergoing different nutritional treatments, associated or not with physical exercise, were characterized by body composition changes at the end of the intervention and after weight regain with very variable lengths of follow-up, from a few months to many years (Table 1). These studies generally provide data about body composition changes after a single weight loss cycle [35,36,37,38, 40, 45,46,47,48,49, 51] (Table 1).

In a small group of 12 young women affected by obesity, Wadden et al. [36] did not find any significant increase in either FM% or FM when body weight returned to baseline 2–3 years after a weight loss program with very low-calorie diet (VLCD). Moreover, in the same study, WC showed no association with FFM changes, which led the authors to conclude that their results did not support the clinical hypothesis that WC may adversely affect body composition [36]. Similarly, in a wider study sample of 44 premenopausal women with obesity, Hammer et al. [38], reached the same conclusion, confirming no effect on total body FM after failure to maintain the weight loss at 6 and 12 months. The stability of FM and FFM, upon regaining of body weight after a nutritional program, was further confirmed in a sample of healthy young subjects of both sexes with excess weight [46], as well as in different racial groups [40].

Obesity studies often assess women only, presented small sample size with a potential lack of statistical power and do not take into account potential confounders as physical activity [36, 38,