During the developmental period, body mass index (BMI), which is used as an index of obesity, shows an interesting growth trajectory. In general, BMI rapidly increases during the first year of life, then subsequently decreases and reaches a nadir at around 6 years of age. Thereafter, BMI increases again throughout childhood, and this second rise is referred to as the adiposity rebound. Adiposity rebound was first reported in 1984 by Rolland-Cachera et al. [1] They found a relationship between the age at adiposity rebound and final adiposity (at age 16 years), and showed that an early rebound (before age 5.5 years) is followed by a higher adiposity level than a later rebound (after age 7 years) [1]. Since then, numerous studies have suggested early adiposity rebound as a predictive marker of obesity in later childhood, adolescence, and adulthood [2-9].

In 2010, approximately 6.7% of preschool-aged children worldwide were overweight or obese, and this prevalence is predicted to increase to 9.1% by 2020 [10]. According to the Korean National Health and Nutrition Examination Survey VII performed in 2016, 8.6% and 9.6% of children and adolescents aged 2-18 years were overweight and obese, respectively [11]. In line with these findings, the timing of adiposity rebound is expected to be different in the 21st century compared with that in the past. In this article, the trends and novel insights in the timing of adiposity rebound with conventional concepts are reviewed.

The exact timing of adiposity rebound is difficult to judge. The most traditional and best method is to trace an individual’s adiposity plot through visual inspection [1,12]. However, most previous studies were population studies or cross-sectional surveys, and only a few of them measured data with sufficient frequency to allow identifying the age of adiposity rebound. Moreover, owing to the small body size of children, even minimal errors in weight or height measurement may considerably influence the determination of BMI. To compensate for these practical weaknesses, various statistical approaches have been proposed [5,6,8,12,13].

In addition, depending on the characteristics of subjects, follow-up durations, and measurement intervals, some studies defined the timing of adiposity rebound as the age at which the lowest BMI occurred before the second BMI rise [5] or at which a specific BMI increase occurred during the period of adiposity rebound [14,15]. Apart from BMI, another study considered the timing of adiposity rebound based on weight and height separately, which implies the velocity rather than the nadir of BMI [16].

Many risk factors related to early adiposity rebound have been studied; however, for some of them, the evidence is not yet conclusive.

Because BMI reflects both FM and lean mass (LM), to confirm that the renewed increase of BMI is the true increase of adiposity, changes in body composition were studied. A small Dunedin study in girls that used dual-energy X-ray absorptiometry showed that early adiposity rebound was associated with increases in FM rather than LM, as expected [47]. However, several studies suggested that the direction of change in BMI and the LM index during the period of adiposity rebound was identical [48], whereas an increase in the FM index lagged behind 2-3 years, particularly for boys [48,49]. On the basis of the evidence on changes in body composition, adiposity rebound coincided with a cessation of FM decline and an increase in LM [48,50]. Therefore, the timing of adiposity rebound was the indicator of individual development rather than increased body fat [48], and a more accurate term for adiposity rebound would be “BMI rebound.” [28] In line with this, being tall at age 3 years, rather than having a high BMI at age 3 years, was shown to be associated with an earlier rebound [51].

Williams and Goulding [16] found different rates of physical development between the early, average, and late rebound groups, with the late-rebound group reaching maturity later. Several other studies have also noted that a more rapid linear growth in infancy or tallness in early childhood is associated with early pubertal maturation or later obesity [52-54]. German et al. [55] studied the onset and tempo of puberty in relation to adiposity rebound, and showed that the early AR group had earlier pubarche and menarche in girls and a more rapid progression of puberty in boys than the later adiposity rebound group. Therefore, early adiposity rebound indicates faster growth, more advanced development, and earlier puberty. In this context, the finding that the shifting trends in adiposity rebound reflect the acceleration of growth in the 21st century is not surprising.

Although the timing of adiposity rebound has complex meanings, an important point is that BMI rebound has a impact on adverse health outcomes. Therefore, targeting modifiable factors in very early life to delay the timing of adiposity rebound is important. Nowadays, 63.3% of 0- to 1-year-old children and 90.6% of 1- to 5-year-old children spend time in some form of childcare center based on a childcare survey in 2013 by the Korea Institute of Child Care and Education [56]. Therefore, childcare settings should be involved in intervention efforts to prevent obesity. Additionally, health check-up programs for infants and children should be emphasized to screen for those at a high risk for early adiposity rebound.