ORIGINAL

Niger J Paed 2014; 41 (2): 120 –124

Nwizu SE

Age and gender-related fat mass

index and fat- free mass index

patterns among adolescents in

Surulere LGA, Lagos

Njokanma OF

Okoromah CAN

David AN

DOI:http://dx.doi.org/10.4314/njp.v41i2,8

Accepted: 18th December 2013

Abstract: Background: Percent

body fat, fat mass and fat mass

index are measures of body fat

while fat free mass and fat free

mass index measure lean mass.

These measures of body composi-

tion vary with age and sex.

Aim: To show the age-related pat-

terns of measures of fat and lean

mass in adolescent Nigerians in

Lagos.

4.5±3.2kg: p < 0.001 respectively).

FMI for girls ranged from 3.2 to

4.5kg/m² peaking at 16 years. At

all ages, girls had higher mean

FMI than boys. The mean FMI for

males fell from 2.6 kg/m² at 10

years to a trough of 1.5 kg/m² at 16

years before a slight rise to 1.9 kg/

m² at 18 years. FFM in boys in-

creased consistently with age,

overtaking that of girls at 12 years

with the gap widening up to 18

years. Measures of body fat were

much lower in study subjects than

reported from western countries

even where lean mass was compa-

rable.

Nwizu SE⁽

)

Department of Paediatrics

Premier Specialist Medical Centre

Ogalade Close Victoria Island

Lagos, Nigeria

Email: ejaycosben@yahoo.com

Tel: +234 802 306 8450

Njokanma OF

Department of Paediatrics,

Lagos State University College of

Medicine, Ikeja, Lagos, Nigeria.

Methods: Percent body fat was

measured in school pupils aged

1

3

0years to 18years (377 boys and

76 girls) using Tanita ® body fat

Okoromah CAN

Lagos University Teaching Hospital,

Idi-Araba, Lagos, Nigeria.

monitor (BF 666). Fat mass (FM)

was derived from percentage body

fat and body mass and fat free

mass (FFM) was obtained by sub-

tracting fat mass from body mass.

Fat mass index (FMI) and fat free

mass index (FFMI) were derived

by dividing FM and FFM, respec-

tively by the square of the height.

Results: Overall percent body fat

and FM were significantly higher

in girls (18.9±7.5% Vs 9.5±4.5%:

David AN

Conclusion: Adolescent females

have higher body fat indices while

males have higher lean mass indi-

ces. Indices of body fat in the cur-

rent study are much lower than

reported for western counterparts.

Clinical Science Division

Nigerian Institute for Medical Research,

Lagos, Nigeria.

Keywords: Adolescents, Fat mass,

Fat mass index, Fat free mass and

Fat free mass index

p

< 0.001 and 9.7±6.1kg Vs

Introduction

as an index of body fat to the extent that it does not dis-

tinguish between excessive weight resulting from fat

mass and that resulting from fat free mass.

Some fat is necessary for overall health to protect inter-

nal organs, provide energy and regulate hormones that

perform various functions in the body. For example,

1

Degree of fatness may be expressed as percent body fat

(% of whole body mass) or as fat mass (kg). It has how-

ever been demonstrated that as much as 45% of the vari-

ance in FFM and 2% of the variance in FM are ex-

cholesterol metabolized in the adrenal gland leads to

corticosteroids and androgens. On the other hand, exces-

sive body fat (obesity) is associated with long term com-

plications of heart disease, diabetes and other chronic

health problems. It is important to periodically survey

populations at risk in ord₂er to determine the need for

appropriate interventions. Adolescence is one of the

four known sensitive periods in life for the development

of obesity, the other t₃hree being intrauterine life, infancy

and ages 5 – 7 years.

5

plained by stature. There is therefore, a need to correct

body fat measures for the effect of stature. This is

achieved by dividing fat mass by square of the height to

get fat mass index (FMI). It is analogous to derivation of

BMI by dividing body mass by square of the height.

Similarly, fat–free mass index (FFMI) is derived by

dividing fat–free mass by the square of the height. Thus

the sum₅ of FMI and FFMI will give the body mass index

(BMI).

Often, body mass index (BMI) is used as proxy for as-

sessing excessive accumulation of body fat, partly be-

cause of ease of measurement. However, it is deficient

4

BMI is commonly used as a proxy for body fatness. This

1

21

however, is not strictly accurate because BMI is made

up of both fat and non-fat components. Expressing body

mass in its component parts of fat mass and fat-free

mass is more informative as percent body fat is a more

specific index of obesity. Further, correction of body fat

mass and fat -free mass for height to give fat mass index

and fat free mass index is thought to improve reliability

a percentage body fat of 20% has a fat mass of: 20/100 x

40 = 8kg

The fat free mass therefore would be 32kg (40kg less

8kg).

The indices FMI and FFMIwere derived by dividing FM

and FFM, respectively by the square of the height.

The data were analyzed using Microsoft Excel program

supplemented by Megastat statistical package. Measures

of statistical location (mean, standard deviation and

range) were derived. Comparison of continuous vari-

ables was achieved using the Student t-test. In all cases,

probability (p) values less than 0.05 were accepted as

statistically significant.

5

of the measurements as indicators of body fatness. Few

previous studies have reported percent body fat in Nige-

rian children but the concept of FMI and FFMI has not

previously been applied to healthy Nigerian adolescents.

Thus, ranges of normal have not been described. The

current report aims to show the pattern of percent body

fat, FMI and FFMI in adolescent Nigerian secondary

school pupils in Lagos.

Results

Subjects and Methods

Seven hundred and fifty three pupils (377 boys and 376

girls) aged 10 years to 18 years were studied. One third

of the pupils (251) were recruited from private secon-

dary schools while twice that number, (502) were from

public schools.

The study was cross- sectional, school based and con-

ducted in Suru-Lere Local Government Area (LGA) of

Lagos State from September through November 2006.

The study population consisted of adolescent male and

female pupils aged between 10years and 18years. The

LGA has a total of 30 public secondary schools and 14

registered private secondary schools, each with six

classes. Nine schools (six public and three private) were

selected at random to represent at least 20% of the num-

ber of schools in the local government area. The choice

reflected the 2-to-1 ratio of public to private schools.

The private schools were all coeducational; three of the

public schools were exclusively for boys and three for

girls.

Figure 1 shows the mean percentage body fat distribu-

tion of pupils according to age and gender. There was a

slight age-related increase in girls with a peak at 16

years (20.8%). The graph for boys declined steadily

from 10 years until 17 years before a slight rise at 18

years. The mean overall value in girls was about twice

that of boys (18.9 ± 7.51 Vs 9.5 ± 4.48, p < 0.001).

Fig 1: Mean percentage body fat of boys and girls

In every participating school, fourteen pupils were ran-

domly selected from each class. Selecting equal num-

bers of subjects from each class ensured that the 2-to-1

ratio of public schools and private schools was main-

tained.For the coeducational schools, separate lists were

used for selecting boys and girls. At the first visit, health

talks were given and copies of a self-designed question-

naire were distributed. The questionnaire was designed

to record socio-demographic features of the child and

parents as well as presence or absence of chronic ill-

nesses in the subjects.

Table 1 shows mean fat mass and fat free mass accord-

ing to age and gender. In boys FFM increased steadily

as age increased. There was no consistent trend with FM

but the figures varied within a narrow range.

Measurements of height, weight and percent body fat

using Tanita ® body fat monitor (BF 666) were carried

out.

Height was measured to the nearest centimeter without

shoes, with the heels and back against a height meter

looking straight ahead. The Tanita body fat monitor/

scaleBF-681/BF-682, was used to measure the bioelec-

trical impedance analysis and the body weight. The

monitor displays the body weight in kg to the nearest

Thus growth in boys involved more of FFM than FM.

On the other hand, both FM and FFM increased with

age in girls reaching peaks at 16 years and 17 years re-

spectively. Comparison between both gender groups

showed that across all age groups except at 10 and 18

years, girls had significantly higher mean fat mass than

boys (p < 0.001). With respect to FFM, girls had higher

values at 10 and 11 years: the values were about equal at

12 and 13 years but boys had significantly higher values

from 14 years on.

1

00gms. In addition, the equipment derives the body fat

content from impedance recorded and displays it as a

percentage of body weight on a digital dial.

The fat mass index profile in figure 2 shows higher val-

ues for girls compared to boys. The graph for girls

showed no consistent pattern but the highest mean value

Fat mass was derived from percentage body fat and

body weight. For instance, a subject weighing 40kg with

1

22

was observed at 16 years corresponding to the observa-

tion with respect to percent body fat and fat mass. For

boys, the highest mean value of FMI was seen at age 10

with a steady decline to a trough between 13 and 17

years followed by a slight rise at 18 years of age but

never getting to the value of the 10 year olds.

On the contrary, fat free mass increased progressively

with age in both sexes with the boys gaining faster

(Figure 3). Boys caught up with girls at 12 years and the

divergence between both graphs widened until 18 years.

Analysis of findings with respect to school type in table

2 showed that in each age group, private school boys

had significantly more percent body fat, fat mass and fat

free mass than their public school counterparts (p at least

Table 1: Mean fat mass (kg) and mean fat free mass (kg)

according to age and gender

0

.03). The same pattern was observed among girls ex-

Boys

Girls

cept in the 16 to 18-year-old group where there was no

significant difference with respect to FM and FFM (p =

Age

n

Mean ± SD

n

Mean ± SD

t

p

0

.16 and 0.96 respectively).

1

0

1

2

3

4

5

6

7

26

FM

5.6 ± 5.40

32

7.5 ± 4.99

1.36

1.11

0.18

0.27

Next, FMI and FFMI were derived by correcting FM

and FFM respectively for stature (Figure 4). The graphs

show that with respect to FMI, mean values were consis-

tently higher in private school pupils of each age group.

On the other hand, mean FFMI values were almost iden-

tical between private and public school boys and girls in

each age group.

FFM

28.7 ± 5.87

30.4 ± 5.29

51

43

52

62

41

26

FM

4.2 ± 2.59

35

62

58

67

63

25

23

9.8 ± 6.82

4.66

4.95

< 0.001

FFM

29.9 ± 3.45

34.6 ± 4.82

FM

3.9 ± 1.86

8.7 ± 6.64

5.40

0.28

< 0.001

0.78

FFM

35.1 ± 6.68

35.5 ± 5.44

Table 2: Percent body fat, fat mass and fat mass index accord-

ing to school type

Private schools

Public schools

t

p

FM

4.4 ± 2.27

10.0 ± 7.66

38.2 ± 5.29

5.28

1.37

< 0.001

0.17

Age

n

Mean ± SD

n

Mean ±

FFM

39.8 ± 7.02

group yr

SD

Boys

FM

4.1 ± 2.31

9.3 ± 3.46

39.2± 4.10

10.13

2.91

< 0.001

0.005

1

0 - 12

3 - 15

6 - 18

PBF

FM

54

13.2 ± 5.9

5.5 ± 4.2

66

10.3 ± 3.7

3.6 ± 1.9

3.3

< 0.01

FFM

42.5 ± 7.94

FM

4.9 ± 4.41

47.8± 7.87

10.6 ± 5.72

41.0 ± 5.36

6.32

5.69

< 0.001

FFM

PBF

33.0 ± 6.6

10.0 ± 5.0

30.3 ± 5.2

8.2 ± 3.3

2.5

2.8

0.01

56

15

120

66

FFM

FM

5.5 ± 4.5

3.9 ± 2.2

3.2

< 0.01

FM

4.5 ± 3.60

11.5 ± 6.18

41.0 ± 4.51

5.16

7.75

< 0.001

FFM

PBF

46.5 ± 8.5

9.5 ± 5.9

42.2 ± 7.9

7.3 ± 2.5

3.3

2.3

< 0.01

0.03

FFM

52.4 ± 7.39

FM

6.8 ± 6.2

4.2 ± 2.0

2.9

2.3

< 0.01

0.03

FM

4.6 ± 3.28

52.7 ±

11.3 ± 6.69

44.2 ± 3.67

4.39

3.92

< 0.001

FFM

57.0 ± 10.3

51.8 ± 7.3

Girls

FFM

10 - 12

PBF

FM

21.6 ± 8.4

10.9 ± 7.2

16.9 ± 2.7

6.1 ± 3.7

4.1

4.6

< 0.01

1

0.30

70

50

59

FFM

PBF

35.7 ± 5.3

23.2 ± 7.5

31.9 ± 5.2

18.3 ± 2.7

4.1

6.6

< 0.01

1

8

14

FM

5.6 ± 2.40

11

8.6 ± 4.64

1.97

6.15

0.07

13 - 15

16 - 18

138

FFM

54.1 ± 5.76

43.1 ± 3.06

< 0.001

FM

13.5 ± 7.0

8.7 ± 4.7

5.4

< 0.01

FFM

PBF

41.9 ± 4.7

23.4 ± 7.9

38.6 ± 4.9

20.0 ± 3.0

4.1

2.1

< 0.01

0.04

To-

tal

377

FM

4.5 ± 3.23

41.7 ±

376

9.7 ± 6.06

14.7

5.68

< 0.001

6

53

FFM

38.1 ± 6.01

FM

14.2 ± 9.0

42.7 ± 6.0

10.5 ± 5.7

42.6 ± 4.0

1.4

0.1

0.16

0.96

1

0.89

FFM

Fig 2: Fat mass index of male and female subjects

Fig 3: Fat–free mass index of male and female subjects

1

23

Fig 4: Fat mass index and fat free mass index in private and

public pupils

parts. The same applied to girls except in late adoles-

cence where the differences were not statistically signifi-

cant. This exception is most probably the result of the

rather small number of subjects recruited at that age. To

the extent that school type reflects socioeconomic status,

it may be concluded that higher socioeconomic status is

associated with increased body fat.

Comparison with studies from Western countries shows

that the mean values of FM we observed in boys and

girls (5.6kg and 12.1kg) remain far below those reported

8

by the New York study. Also, the PBF of our private

school pupils (11.3% for boys and 22.3% for girls) were

9

below the figures observed in the New Zealand study.

In comparison with Italians, the private school boys and

girls in the current study had₀ appreciably less body fat

1

but comparable lean mass. Thus, while there is a

greater tendency towards fatness by private school

pupils, the levels are not yet comparable to Western

countries.

Discussion

The age-related graphs of PBF, FM and FMI in boys

were similar in shape showing a fairly consistent decline

in trend lines between ages 10 and 14/15 before a gentle

rise at 18 years. The trajectory is very similar to that

established for b₂lack and non-black boys of comparable

age in Texas. It is consistent with physiological

changes of adolescence which make body fat indices in

males fall from early to mid-adolescence. The rise at 18

years most probably represents a physiologic increase to

establish adult levels.

At all ages, girls had higher mean percent body fat, fat

mass and fat mass index than boys. This finding is con-

sistent with tre₆_,n₇ds observed in earlier studies both

8

within Nigeria

and elsewhere . A study of Nigerian

6

adolescents at Ile-Ife , reported a higher percentage fat

mass among girls. Also, a study of Fulani population of

Northern Nigeria showed higher fat levels in females

compared to males. This was also the pattern in New

York which show₈ed higher levels of fat mass in girls

compared to boys.

7

The observation was quite different in girls. Percent

body fat, fat mass and fat mass index increased with age,

reaching a peak at 16 years. Again, this is in agreement

with findings among non-black girls in the Texas study

in which the trajectory was still upwards in 15-year-

In terms of absolute measurements, the mean fat mass of

females in the current study (9.7kg) was higher than

6

.6kg observed at Ile-Ife , Nigeria. For the males how-

2

ever, our value of 4.5kg was comparable to the figure of

.7kg reported at Ife. On the contrary, the fat mass fig-

ures herein reported are much lower than 11kg to 12kg

Males) and 16kg to 22kg (Females) reported in a study

in New York comprising four different ethnic groups

olds. .The finding is also explainable on phy₁s₃iologic fat

4

deposition that accompanies puberty in girls.

With respect to measures of lean mass (FFM and FFMI),

a different pattern was observed. The figures for both

boys and girls increased progressively with age. At 10

years, girls had higher values than boys but with increas-

ing age, there was a faster increase in the mean values

for boys with a “catch-up” at 12 years, followed by a

persistently widening gap between the graphs in older

adolescents. These findings are con₇sistent w₁₂ith those of

(

8

Whites, Blacks, Hispanics and Asians). Also, our male

(

and female subjects had about half and two thirds of the

PBF of figures reported in New Zealand (17.2% and

9

2

8.2% resp₀ectively). Indeed, a group of Italian adoles-

1

cent boys of comparable age weighed 17kg more than

the subjects of the current study (63.7kg Vs 46.7kg).

However, there was practically no difference in lean

mass (42.6kg Vs 42.2kg) as 16.6kg of that difference

was accounted for by fat mass. Similarly, although Ital-

ian girls weighed 17.1kg more (64.9kgVs 47.8kg), their

counterparts in the present study had 17.6kg less fat

researchers in Northern Nigeria, Japan

York.

and New

8

,4, 9

The observed differences could be explained

by the fact that at puberty, sex hormones induce a pro-

nounced sexual dimorphism: males gain proportionately

more muscle and lean tissue compared to fat, and fe-

males deposit fat as a natural part of the ontogeny of

their sexual and reproductive phy₁₃siology. Thus boys

have more lean mass than body fat.

(

3

27.3kg Vs 9.7kg) and 0.5kg more lean mass (37.6kg Vs

8.1kg). Thus our subjects had appreciably less fat but

about the same fat free mass as their western counter-

parts. The reasons for the marked differences are most

likely dietary. Western diets are far more refined and of

higher caloric density than African foods.

Conclusion

Analysis of our findings with respect to school type in-

dicated that boys in private schools had more percent

body fat, FM and FFM than their public school counter-

The age-related patterns of measures of body fat and

lean mass of boys and girls in the current study followed

1

24

expected trends. The absolute figures however showed

that fat indices were much lower than in advanced west-

ern societies. The obvious interpretation is that obesity

is much less of a problem in the group of adolescents

studied than in Western societies, corro₁₄b_,₁o₅_,r₁a₆t_,₁i₇ng the low

signals. However, concerted effort must be made to en-

sure that dietary and sedentary lifestyles which predis-

pose to obesity are controlled before it becomes a public

health problem in Nigeria.

prevalence reported by earlier workers

. The spe-

Conflict of interest: None

Funding: None

cific periods at which percent body fat and FMI were

elevated represent physiologic changes rather than alert

References

7

.

Glew RH, Conn CA, Bhanji R,

12. Sampei MA, Novo NF, JulianoY

and Sigulem DM. Anthropometry

and body composition in ethnic

Japanese and Caucasian adolescent

boys. Pediatr. Int. 2008; 50:679-

686

13. McCarthy HD, Cole TJ, Fry T,

Jebb SA and Prentice AM. Pediat-

ric Highlight. Body fat reference

curves for children. Int J of Obesity

2006; 30: 598 – 602

14. Akesode EA, Ajibode HA. Preva-

lence of Obesity among Nigerian

School Children. Soc Sci Med

1983;17:107- 111

15. Ben- Bassey UP, Oduwole AO,

Ogundipe OO. Prevalence of Over-

weight and Obesity in Eti-Osa

LGA, Lagos, Nigeria. Obestet

Rev.2007;8(6):475-9

16. AnsaVO,OdigweCO,Anah MU.

Profile of Body Mass Index and

Obesity in Nigerian Children and

adolescents. Niger J Med 2001;78-

80

17. Ene-Obong H, Ibeanu V, Onuoha

N, Ejekwu A. Prevalence of over-

weight, obesity and thinness

among urban school aged children

and adolescents in Southern Nige-

ria. Food Nutr Bull. 2012;33

(4):242-50

1

2

.

Scott JR. What is body composi-

tion? About.com Health’s disease

and condition.

Eissa MA, Dai S, Mihalopoulos

NL, Ray S, Harrist RB, Labarthe

DR. Trajectories of Fat mass In-

dex, Fat Free mass Index and

Waist Circumference in Children.

Project Heartbeat. Am J Prev Med

Calderon P, Barnes C and Vander-

Jagt DJ. Survey of the Growth

Characteristics and Body Compo-

sition of Fulani Children in a Rural

Hamlet in Northern Nigeria. J

Trop Pediatr 2003; 49:313-322.

8. Freedman DS, Wang J, Maynard

LM et al. Relation of BMI to fat

free mass among children and

adolescents. Int J Obestet 2005;

29: 1 – 8

9. Taylor RW, Jones IE, Williams

SM, and Goulding A. Body fat

percentages measured by dual

energy X-ray absorpsiometry cor-

responding to recently recom-

mended body mass index cutoffs

for overweight and obesity in chil-

dren and adolescents aged 3 – 18

yr. Am J Clin Nutr 2002; 76: 1416

– 21

10. ZuguoMei, Laurence M Grummer-

Strawn, Angelo Pietrobelli et al.

Validity of body mass index com-

pared with other body composition

screening indexes for the assess-

ment of body fatness in children

and adolescents. Am J ClinNutr

2002; 75:978-85.

11. Bray GA, Delany JP, Harsha DW,

Volaufova J, Champagne CM.

Body composition of African

American and White Children: A

2

009;37(IS):S34-S39

3

4

.

Monyeki KD, Van Lenthe FJ and

Steyn NP. Obesity: does it occur

in African Children in a rural com-

munity in South Africa? Int J Epi-

demiol 1999; 28:287-292.

Komiya S, Masuda T, Nakao T

and Teramoto K.The relationships

between stature, fat-free mass

index, and fat mass index at before

and after BMI-rebound in chil-

dren. J Hlth Sci 2004; 26:31-39.

Nakao T and Komiya S. Reference

Norms for a Fat-free Mass Index

and Fat Mass Index in the Japa-

nese Child Population. J Physiol

Anthropol Appl Human Sci

5

.

2

003;22 (6):293-298 .

6

.

Owa JA and Adejuyigbe O. Fat

Mass, Fat Mass Percentage, Body

Mass Index and Mid–Upper Arm

Circumference in a Healthy Popu-

lation of Nigerian Children. J Trop

Paediatr 1997; 43: 13 -19

2

-Year Follow-up of the BAROC

Study. Obes Res 2001; 9:605-621.