ABSTRACT Mid upper arm circumference (MUAC) is recognized as an effective means of screening for poor nutritional status in adults. The efficacy of calf circumference (CC) as a screening tool, however, is not well reported. We studied 2000 healthy women (age range: 15–49 years) in Mashed, Islamic Republic of Iran, to assess anthropometric measurements at reproductive age and the usefulness of MUAC and CC for estimating body mass index (BMI) and screening of women at nutritional risk. We found a strong correlation between MUAC and CC with weight and BMI, suggesting that MUAC and CC can be used to estimate BMI and detect nutritional disorders.

RESUME Le périmètre brachial à mi-hauteur est reconnu comme moyen de dépistage efficace d’un mauvais état nutritionnel chez l’adulte. Toutefois, l’efficacité du périmètre du mollet comme outil de dépistage n’est pas bien documentée. Nous avons étudié 2000 femmes en bonne santé (fourchette d’âge : 15-49 ans) à Mashed (République islamique d’Iran) afin d’évaluer les mesures anthropométriques pendant la période reproductive et l’utilité du périmètre du mollet et du périmètre brachial à mi-hauteur pour estimer l’indice de Quételet et dépister les femmes qui présentent un risque nutritionnel. Nous avons trouvé une forte corrélation entre le périmètre brachial à mi-hauteur et le périmètre du mollet avec le poids et l’indice de Quételet, ce qui semble indiquer que ces deux mesures peuvent être utilisées pour estimer l’indice de Quételet et dépister les troubles nutritionnels.

Women are particularly vulnerable to developing nutritional problems. The many daily pressures they experience can lead to nutritional stress and poor maternal nutritional status. Anthropometric indicators are useful tools for screening women at nutritional risk, monitoring nutritional status and predicting unfavorable infant outcomes related to pregnancy [1]. In adults, body mass index (BMI), or weight (kg)/height squared (m2), is used to define underweight or overweight [2].

Mid upper arm circumference (MUAC) has recently emerged in the literature as a potential screening tool for poor nutritional status in adults. MUAC has been analysed in adults and cut-off points for chronic energy deficiency have been established using a range of datasets from developing countries [2].

The typical distribution of weight gain during pregnancy is to the conceptus and maternal organs such as the breasts and uterus [3]. Thus, in societies in which weight gain and MUAC changes in pregnancy are low, MUAC may be an accurate indicator of pre-pregnancy weight, especially in the early months of pregnancy. Measuring MUAC has several advantages. The measurement can be taken quickly and at little cost. It requires neither sophisticated equipment nor anything but the most basic literacy level to carry out [1].

Although the usefulness of the MUAC measurement is being increasingly acknowledged, the same cannot be said of calf circumference (CC). Therefore, to assess the relationship of anthropometric measures of MUAC and CC to BMI and the usefulness of the two indices in estimating BMI in woman of reproductive age, we examined both among women in the city of Mashed in 1997–1998.

Mashed is the capital of the Islamic Republic of Iran’s largest province, Khorasan, in the north-east of the country. It is also the country’s second largest city, with a population of approximately 1.9 million people (1996 census), one-quarter of whom are women of reproductive age (i.e. aged 15–49 years) with a literacy rate of 92.5%.

Using clustered, multi-stage random sampling, 2000 women from this age group were selected for a cross-sectional study. All were in good health. None were pregnant, lactating or users of hormonal contraceptives or other drugs. Sportswomen were excluded. All had experienced menarche and none had reached menopause. The women were interviewed and assessed in their homes, at their workplaces or at their educational institutions between the hours of 09.00 and 11.00.

Using non-stretchable tapes, MUAC and CC were measured to the nearest millimetre. Left MUAC was measured at the mid-point between the acromion process of the scapula and the olecranon process of the ulna, with the arm and forearm hanging loosely by the side [4]. Triceps skin fold (TSF) was measured with skin fold caliper at the mid-point of the arm [4]. Left leg CC was measured by passing the tape around the point of maximum circumference. Two measurements of CC were taken: the subject standing (CC standing); and the subject seated, with thigh flexed at 90º to the trunk and calf flexed at 90º to the thigh (CC sitting). The mean of three consecutive measurements was recorded to the nearest millimetre. All subjects were weighed wearing minimal clothing. Body weight was measured to within 100 g using scales, which were checked with standard weights each day. Height was measured with the subject standing at ground level without footwear against a wall. BMI was calculated as weight (kg) divided by height squared (m2). The cut-off points for overweight and underweight were 19.8 and 26, respectively [5]. A trained midwife took all measurements.

Data were analysed using the Pearson correlation coefficient. Data to assess the utility of MUAC and CC as an instrument for estimating BMI were analysed using multiple regression analysis. Sensitivity, specificity and positive and negative predictive values of MUAC and CC in screening underweight and overweight were calculated.

Of the 2000 women enrolled in the study, 22% were < 20 years of age; 37.8% were 20–29 years of age; 29.3% were 30–39 years of age; and 10.9% were > 40 years of age. The most common occupation of subjects was non-paid domestic duties in the family home (65.1%), followed by full-time study (23.7%) and paid employment outside the family home (11.2%). Educational levels attained were: primary school (24.4%), middle school (17.3%), high school (39.3%), tertiary (14.5%) and illiterate (4.5%).

BMI values had a mean value of 6.2 BMI units and tended to be within normal ranges. Of all women in the study, 21.1% were underweight (i.e. BMI < 19.8 kg/m2). All indices were lower in younger women, with 41.1% of those aged 15–19 years and 26.2% of those aged 20–24 years being underweight. Conversely, the prevalence of obesity (36.2%) and overweight (24.5%) were greatest in women aged 40–44 years. The mean and SD of anthropometric indices are given in Table 1. BMI correlated strongly with MUAC and CC, with CC standing correlating more strongly to BMI than CC sitting.

Anthropometric measurements were significantly correlated with each other and with age, raising the possibility that some measurements could replace others in a regression model (Table 2). Entering age, MUAC and BMI as dependent, stepwise regression analysis revealed that MUAC, but not age, independently correlated with BMI (Table 3).

The relationship between BMI and MUAC was formulated:

The relationship of BMI to MUAC and CC was formulated:

In our study, 88.8% of women with BMI > 29 kg/m2 had a MUAC > 30.5 cm, with only 10.2% of these women not identified by this means. On the other hand, 70.9% of the mothers with MUAC > 30.5 cm were overweight and 98.3% of women with MUAC > 30.5 cm were not overweight (Table 6). The sensitivity, specificity, PPV and NPV for detecting overweight for women with a CC of 36 cm was 83.1%, 87.2%, 49.3% and 97.2%, respectively.

The use of BMI for assessing undernutrition in adults is now being applied worldwide [6]. There is also a need to assess whether other, simpler measures that could be used in situations such as emergencies, when only semi-skilled monitors are available, or even in the home when a person wants to be certain her weight is within or near the normal range.

Mean weight and height in our sample were lower than the United States of America’s National Center for Health Statistics average [4], and greater than those reported in 1994 by James et al. for adults in eight African and Asian countries [6].

Our study demonstrated lower anthropometric lean BMIs in younger women, especially those < 25 years of age and higher indices in women aged > 40 years, indicating the need for nutritional education for women as they approach the end stages of their reproductive period. Similar findings for women 40–50 years of age were reported in a Brazilian study [2].

In the present study, MUAC and CC were significantly correlated with weight, BMI and percent of ideal body weight. Similarly, the James et al. study also demonstrated an excellent linear relationship between MUAC and BMI [6]. A study in Nigeria [7] found similar results.

MUAC and CC can be used to estimate BMI. The following formulae are suggested:

BMI = 0.1036 MUAC (mm) – 3.9, or

BMI = 0.07745 MUAC (mm) +

0.03795 CC (mm) – 9.7578

This study was supported by a grant from the Vice-Chancellor for Research Affairs at Mashed University of Medical Sciences. We appreciate the contribution of our colleagues Dr S. Parsai, Dr M. Al-Reza Hdjzadeh, M. Kolivand and S. Reza Mazloom. We also appreciate the contribution of women who participated in this study.