Brief Communication: Menarche is Related to Fat Distribution William D. Lassek* and Steven J.C. Gaulin Department of Anthropology, University of California at Santa Barbara, Santa Barbara, CA 93106-3210 KEY WORDS body fat; hip circumference; waist-hip ratio; leptin ABSTRACT The energy demands of pregnancy and lactation together with the accumulation of stored fat in human females during development suggest that a criti- cal level of fat may be required for menarche; but multi- variate analyses have supported the alternative view that skeletal growth is the main factor. However, signifi- cant differences between upper- and lower-body (gluteo- femoral) fat suggest that fat distribution may be more relevant than total fat. Using cross-sectional data from the third National Health and Nutrition Examination Survey (NHANES III) for females aged 10–14, we show that menarche is more closely related to fat distribution than to skeletal maturity. Unit increases in hip circum- ference are associated with 24% higher odds of menarche while increases in waist circumference and triceps skinfold lower the odds by 7 and 9%, respectively. Those with menarche despite low levels of total body fat have relatively more fat stored in gluteofemoral depots than those without menarche or those with menarche and greater total amounts of fat. In young women with com- pleted growth, age at menarche is negatively related to hip and thigh circumference and positively related to waist circumference, stature, and biiliac breadth; and blood leptin levels are much more strongly related to gluteofemoral than upper-body fat, suggesting that leptin may convey information about fat distribution to the hypothalamus during puberty. Fat distribution may be relevant because gluteofemoral fat may provide neu- rodevelopmentally important fatty acid reserves. Am J Phys Anthropol 133:1147–1151, 2007. V V C 2007 Wiley-Liss, Inc. Menarche is an important landmark in a woman’s reproductive career; and, to the degree that selection molds the life-history of a species, one would expect sex- ual maturation to be linked to the acquisition of resour- ces necessary for successful reproduction. The proximate cause of menarche is an increase in the frequency of the gonadotropin releasing hormone (GnRH) pulse generator in the hypothalamus, but the age at menarche varies widely and is delayed in populations with poor nutrition (Thomas et al., 2001; Gluckman and Hanson, 2006). Until recently it was generally accepted that the timing of menarche is related to skeletal growth, which comes about 1 year after the peak in height velocity (Simmons and Greulich, 1943; Elizondo, 1992). An alternative view is that menarche depends on a critical amount of stored fat, since the 16 kg of fat typi- cally stored during childhood and puberty can provide additional energy during pregnancy and lactation (Frisch and Revelle, 1970; Frisch and McArthur, 1974; Frisch et al., 1973; Frisch, 1976, 1994). The hormone leptin, produced by fat cells, provides a pathway to com- municate the size of fat stores to the GnRH secreting neurons in the hypothalamus via leptin receptors in KiSS-1 neurons (Smith et al., 2006). Leptin is required for puberty (Chehab et al., 1996, 1997; Clement et al., 1998; Ozato et al., 1999; Farooqi et al., 2002), and age at menarche in young women is inversely related to leptin levels (Matkovic et al., 1997), with a significant 28% increase in leptin levels during the 6 months preceding menarche (Blogowska et al., 2005). Despite the appeal of this hypothesis, studies of men- arche have generally failed to support the critical-fat theory. Menarche can occur despite low fat levels with little evidence of a threshold (Johnston et al., 1971; Bille- wicz et al., 1976; Trussell, 1978; Garn and LaVelle, 1983); and multivariate analyses have shown that height and biiliac breadth are much more important than meas- ures of total fat or body weight in predicting the age of menarche (van’t Hof and Roede, 1977; Ellison, 1982; Stark et al., 1989; Elizondo, 1992). However, another possibility is that menarche may be related to fat distribution rather than total fat, and in particular to the relative amount of lower-body (gluteo- femoral) vs. upper-body fat. Female waist-hip ratio (WHR) declines during childhood from 1.03 at 4 months of age to 0.78 at the time of menarche (Fredriks et al., 2005), and there is a steep increase in hip circumference just before menarche (Forbes, 1992). Young German women in higher quartiles for self-reported hip, thigh, and leg circumferences had higher odds of menarche in cross-sectional bivariate analyses (Merzenich et al., 1993). There is also evidence that gluteofemoral fat produces more leptin than upper-body fat. Subcutaneous gluteal fat contains more leptin mRNA than abdominal fat (Papaspyrou-Rao et al., 1997), and multivariate analyses indicate that hip circumference is a significant positive predictor of blood leptin levels while waist circumference is not (Bennett et al., 1997; Ho et al., 1999; Sudi et al., 2000). For example, in the study by Bennett et al. (1997), hip circumference explained 36% of the variance in blood leptin levels, total fat explained an additional 2%, and waist circumference was not related. There is *Correspondence to: William Lassek, Department of Anthropology, University of California at Santa Barbara, Santa Barbara, CA 93106-3210, USA. E-mail: will.lassek@gmail.com Received 5 June 2006; accepted 26 March 2007 DOI 10.1002/ajpa.20644 Published online 6 June 2007 in Wiley InterScience (www.interscience.wiley.com). V V C 2007 WILEY-LISS, INC. AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 133:1147–1151 (2007)