Assessment of the Bone Density of Nomadic Fulani Herdsmen in Northern Nigeria Using Calcaneal Ultrasonography

Abstract

Context

Osteoporosis is characterized by decreased bone density and increased bone fragility. Genetics, diet, and physical activity are established determinants of bone density. The seminomadic Fulani of northern Nigeria trek long distances on foot daily to graze and water their animals, and have access to calcium-rich dairy products.

Objective

We sought to determine whether the high level of physical activity and presumed calcium-rich diet of the Fulani would promote a higher bone density, compared with their relatively inactive counterparts in the general population.

Design

Cross-sectional study.

Setting

Three Fulani settlements (Toro, Tilden Fulani, and Magaman Gumau) within 5 to 15 kilometers of Jos Metropolis on the Jos Plateau in northern Nigeria.

Patients

We assessed the calcaneal characteristics of a consecutive sample of 51 active seminomadic Fulani men using the Lunar Achilles+ ultrasonometer.

Main outcome measures

Calcaneal broadband ultrasound attenuation (BUA), speed of sound (SOS), stiffness index (SI), and SI T-scores.

Results

The mean age of the herdsmen was 26 ± 9 years (range, 16 to 49), and the mean BMI was 19.9 ± 2.3 kg/m². The mean BUA was 124 ± 13 dB/MHz (95% CI, 120 to 128), the mean SOS was 1572 ± 33 m/s (95% CI, 1563 to 1581), the mean SI was 102 ± 17 (95% CI, 97 to 107], and the mean SI T-score was −0.74 ± 0.97 (95% CI, −0.47 to 1.01]. The mean SI was 1 T-score unit below that of an age-matched cohort in the general population.

Conclusion

Fulani herdsmen have a theoretically increased risk for bone fracture in a background of low BMI and potentially high calcium intake.

Introduction

Bone mass steadily increases from childhood through adolescence and peaks in the third decade of life.[1] Attaining and maintaining peak bone mass in adulthood depends on genetic, diet, and lifestyle factors[2–4] and is correlated with the lifetime risk for osteoporosis and fracture.[5–7] Adequate calcium intake and physical activity in childhood and adolescence have been linked with subsequent attainment of higher peak bone density.

The Fulani are a distinct nomadic African ethnic group that inhabits a hot, arid area that spans the region between Mauritania and Senegal in the west to Chad and Sudan in the east. Half of the estimated 15 million Fulani inhabit northern Nigeria where they represent one of the major ethnic groups.

The culture and economy of the seminomadic Fulani of northern Nigeria are centered on cattle. As such, Fulani males of the region habitually trek long distances on foot daily in search of water and pasture for their animals and have access to milk and other dairy products (yogurt, butter oil, cheese). Thus, the intense physical activity of these nomads and the presumably high calcium content of their diet should confer relatively high bone mineral density (BMD) and low risk for bone fracture, but this hypothesis has not been tested.

We have shown in previous reports that the Fulani nomads of Bauchi and Plateau States in northern Nigeria commonly have a body mass index (BMI) in the region of 20 kg/m².[8] It is well known that bone density has a positive correlation with BMI.[9] We sought to determine whether the extensive physical activity and dairy product-based diet of Fulani nomads would offset their risk for low bone density because of their relatively low BMI. To our knowledge, the literature contains no reports of bone density of Fulani nomads.

To answer this question, we used ultrasonography to measure the characteristics of the calcaneus of 51 very active Fulani herdsmen in the age range of 16 to 45 years who trek on foot daily to graze their cattle in the rural areas of northern Nigeria.

Dual-energy X-ray absorptiometry (DXA) is regarded as the gold standard for assessment of bone mineral density, a surrogate for bone quality.[10] Since DXA was unavailable, we used calcaneal ultrasonography to estimate the bone density of the Fulani herdsmen. Measurement of the stiffness index (SI) by using calcaneal ultrasonography has been shown to be a reliable surrogate of bone mineral density at the femoral head and spine determined by using DXA.[10] Calcaneal ultrasonography has been demonstrated to be as useful for predicting fracture risk as DXA.[11–13]

This paper reports the results of our analysis of the bones of Fulani nomads in northern Nigeria.

Materials and Methods

Study Area

Jos is the capitol of Plateau State in northern Nigeria. It lies between latitude 09° 52′N and longitude 008° 54′E at an elevation of 1285 m above sea level.[14] The high altitude favors an abundance of rain and a long wet season, both of which are suitable for cattle rearing. The seminomadic Fulani inhabit several hamlets located at the foot of the Jos Plateau and within 5 to 15 kilometers of Jos metropolis. The Jos University Teaching Hospital (JUTH) is a federal tertiary care facility located within Jos metropolis.

Study Sample

All active Fulani herdsmen aged 16 to 50 years were eligible for recruitment into the study. Exclusion criteria were recent bone injury (preceding 12 weeks); cigarette smoking; and intake of bone-altering medications, such as steroids. Approval for the study was obtained from the JUTH Ethical Committee and the Human Research Review Committee of the University of New Mexico. Written informed consent was obtained from all participants.

Study Protocol

This study was done in July 2007. Participants were recruited from the Fulani settlements of Toro, Tilden Fulani, and Magaman Gumau. Consent for recruitment was first obtained from the community leaders who then assigned an English-speaking herdsman to assist in recruiting participants. Interviews were conducted in the dominant local language (Hausa) by a member of the research team (EPL). The purpose of the study, as well as risks and benefits, were explained to all participants. Consecutive, eligible, consenting Fulani herdsmen were then transported in small groups to JUTH where anthropometric and calcaneal ultrasonography measurements were conducted.

Anthropometric Assessment

Height was measured without shoes and headgear, using a portable stadiometer calibrated to the nearest 0.1 cm. Participants were weighed daily wearing light clothing and no shoes, to the nearest 0.5 kg, using a scale calibrated with known weight daily (Metro, 2001 Model, USA, sensitivity, 0.5 kg). Body measurements were conducted according to the National Health and Nutrition Examination Survey III guidelines.[15] The BMI was defined as the quotient of the weight in kilograms and the square of the height in meters, expressed as kg/m².

Calcaneal Ultrasonography

Ultrasonography measurements of the calcaneus were obtained using the Achilles+ (Lunar Corporation, Madison, WI, USA) according to the manufacturer's instructions. Since there is no consensus on whether the dominant or nondominant foot should be used to measure BMD or to predict fracture risk,[16] we performed ultrasonographic measurements on the right foot, which was the dominant foot in most participants. Each participant was seated with his right foot resting in the heel bath of the instrument. The ultrasonography instrument then automatically introduces surfactant-containing warm water into the heel bath, and measures the broadband ultrasound attenuation (BUA) and speed-of-sound (SOS) transmission. This technique has been fully described in a previous report.[17] BUA is defined as the slope of the regression line derived from the ratio of the signal amplitude of the calcaneus to that of water (reference) at each ultrasound frequency (dB/MHz). SOS refers to the speed of the sound wave traveling through the calcaneus, measured in m/s. The stiffness index (SI) was calculated for each participant using the following equation:

$$\text{SI}=\left(0.67\times \text{BUA}\right)+\left(0.28\times \text{SOS}\right)-420$$

[18]

The SI is a single clinical measure whose calculation has a 50% contribution from BUA and a 50% contribution from SOS, with a lower precision error than either BUA or SOS alone. Calibration of the instrument was monitored by using a phantom heel provided by the manufacturer. The coefficients of variation for BUA and SOS measurements were 9% and 2%, respectively, indicating good precision.

The T-score for SI was calculated as follows:

$$\text{T-score}=(\text{participant SI}-\text{mean SI of the reference group})/\left(\text{standard deviation of the reference group}\right)$$

We used the mean SI of 115 ± 17 for healthy Nigerian men aged 20 to 29 years reported by VanderJagt and colleagues[17] as the mean SI of the reference group.

Statistical Methods

Data were analyzed with Epi Info 3.4.1 (CDC, Atlanta, Georgia) and Microsoft Office Excel 2003 (Microsoft Corporation, Redmond, Washington). Results are expressed as mean and standard deviation except where specified. Linear regression was used to assess correlations between continuous variables. We examined the independent association of SI with age, weight, height, and BMI in a multiple linear regression model.

Results

A total of 51 eligible herdsmen gave consent and were recruited into the study. None of the herdsmen were excluded. The anthropometric characteristics of the participants by decade of age are summarized in Table 1.

Table 1.

Anthropometric Characteristics (Mean ± SD) of 51 Fulani Men by Decade of Age

Age Interval (years)	n	Age (years)	Weight (kg)	Height (cm)	BMI* (kg/m2)
16–19	11	18 ± 1.0	51.4 ± 6.7	166.9 ± 7.6	18.3 ± 1.2
20–29	25	23 ± 2.0	56.9 ± 8.2	168.3 ± 6.8	19.9 ± 2.1
30–39	11	34 ± 3.0	60.9 ± 5.4	168.7 ± 7.5	21.5 ± 2.1
40–49	4	47 ± 3	60.8 ± 5.4	178.8 ± 7.6	18.8 ± 3.9

^*BMI = Body mass index

The mean age of the herdsmen was 26 ± 9 years (range, 16 to 49). Mean weight was 56.9 ± 8.6 kg, mean height was 168.9 ± 8 cm, and mean BMI was 19.9 ± 2.3 kg/m² (95% CI, 19.3 to 20.5) (Table 2). Thirteen (26%) of the participants were underweight[19] (defined as BMI < 18.5), and none were obese. Weight (r = .3; P = .03) and height (r = .3; P = .04) but not BMI (r = .2; P = .2) were positively correlated with age in this cohort.

Table 2.

Descriptive Statistics for 51 Fulani Men^*

Characteristic	Mean ± SD (min, max)	95% Confidence Interval
Age (years)	26 ± 9 (16, 49)	23 to 29
Weight (kg)	56.9 ± 8.6 (38.3, 8.0)	54.5 to 59.3
Height (cm)	168.9 ± 7.6 (153.7, 187)	166.7 to 171.1
BMI (kg/m2)	19.9 ± 2.3 (13.9, 27.4)	19.3 to 20.5
BUA (dB/MHz)	124 ± 13 (97, 187)	120 to 128
SOS (m/s)	1572 ± 33 (1495, 1644)	1563 to 1581
SI	102 ± 17 (63, 136)	97 to 107
T-score	−0.74 ± 0.97 (−3.0, 1.24)	−0.47 to 1.01

^*BMI = body mass index, BUA = broadband ultrasound attenuation; SI = stiffness index; SOS = speed of sound

The mean BUA of the herdsmen was 124 ± 13 dB/MHz (95% CI, 120 to 128) (Table 2). BUA increased with age, peaked in the fourth decade, and declined afterward (Table 3). Similarly, the mean SOS was 1572 ± 33 m/s (95% CI, 1563 to 1581) (Table 2). SOS increased with age, peaked in the fourth decade, and sharply declined thereafter (Table 3).

Table 3.

Ultrasonography Parameters by Decade of Age for 51 Fulani Men^*

Age interval (years)	n	BUA (dB/MHz)	SOS (m/s) Mean ± SD	SI Mean ± SD	T-Score Mean ± SD	T-Score Nigerian Reference
16–19	11	111 ± 8	1549 ± 25	87 ± 12	−1.6 ± 0.7	−0.94 (−1.52,0.53)
20–29	25	125 ± 10	1582 ± 30	106 ± 13	−0.5 ± 0.8	No data
30–39	11	136 ± 11	1585 ± 30	114 ± 14	−0.04 ± 0.8	−0.35 (−3.4, 1.53)
40–49	4	118 ± 17	1536 ± 33	89 ± 21	−1.5 ± 1.2	−0.73 (−3.47, 1.94)

^*BUA = broadband ultrasound attenuation, SI = stiffness index, SOS = speed of sound

The mean SI of herdsmen was 102 ± 17 (95% CI, 97 to 107), and the mean SI T-score was −0.74 ± 0.97 (95% CI, −0.47 to 1.01) (Table 2). Both participant SI and T-score increased with age, peaked in the fourth decade, and sharply declined afterward (Table 3).

BMI was positively correlated with SI (r = .4; P = .003). The equation that best described the relationship between BMI and SI was SI = 2.8047 BMI + 46.695 (Figure 1). Likewise, weight was positively correlated with SI (r = 0.3; P = 0.01). The equation that best described the relationship between weight and SI was SI = 0.6726 Weight + 64.14 (Figure 2). Neither age (r = .1; P = .3) nor height (r = .0; P = .3) correlated with SI. However, the correlations between both BMI and weight with SI were no longer significant after age was controlled for in a multivariate model (P = .7).

Figure 1.

Correlation of body mass index with stiffness index of Fulani men.

Figure 2.

Correlation of weight with stiffness index of Fulani men.

Discussion

We have described our observations regarding bone density in a cross-section of nomadic Fulani herdsmen in northern Nigeria. It is widely assumed that the diet of Fulani pastoralists is rich in calcium and related minerals, since their economy is centered on cattle from which they derive a variety of dairy products. In addition, the Fulani are known to trek long distances on foot daily in search of pasture and water for their animals. However, despite the high level of physical activity and unlimited access to milk and other dairy products, such as yogurt and cheese, we found the mean calcaneal SI of the participants to be 1 T-score unit below that reported for an age-matched urban cohort of 321 Nigerian men in the general population.[17] In a recent report, the mean dietary calcium intake for Nigerian men was 551 mg/day,[20] which is well below the 1000 mg/day recommended for adults in the US.[21]

The low calcaneal SI of the Fulani herdsmen raises several questions. Are the Fulani taking advantage of the availability of milk and other dairy products for their nourishment? Or are most of their calcium-rich dairy products being sold for income to meet other household needs? In our view, the low SI of the Fulani men may be attributed to a combination of low BMI (23% of herdsmen were underweight) and inadequate calcium intake. However, if the Fulani pastoralists had optimal calcium intake and were not underweight, then other, as-yet-unidentified factors, such as decreased bioavailability of calcium despite adequate dietary intake, may account for our findings. The diets of most sub-Saharan African populations are cereal-based and contain chelators, such as oxalates, that decrease calcium bioavailability.[22] Furthermore, genetics may be largely responsible for our findings since genetic factors account for between 70% to 80% of an individual's potential peak bone mass.[23]

The variation of bone mass with age is thoroughly documented: It increases steadily through childhood and adolescence and peaks between the ages of 17 and 30 years.[24, 25] Furthermore age-related bone loss begins after the third decade of life and progresses at a slower rate in individuals who have attained a higher peak bone mass, have optimum calcium intake, and engage in physical activity.[26, 27] In contrast with other populations in previous reports, we observed that the SI of the Fulani men peaked in the fourth decade, and found no significant correlation between SI and age. That Fulani nomads attain peak SI later than other populations in previous reports may be a chance finding or a true phenomenon. Our failure to demonstrate a correlation between SI and age may be attributed to the small sample size, as reflected in the wide confidence intervals for some of the estimates. It may also result from the narrow age range of the participants.

Both height and weight are known to increase mechanical loading of bones, and Sowers and colleagues[28] have suggested that more fat may increase the production of hormones that slow bone loss. In agreement with previous reports, we observed a significant correlation between SI and both weight[29] and BMI, but not height as cited by Messenger and colleagues.[30] Interestingly, the correlations between SI and both weight and BMI were no longer significant after controlling for age in a multivariate model, probably due to small-study bias.

In terms of significance, our data demonstrate that both a high level of physical activity and availability of milk and other dairy products may not offset the risk for low bone density in Fulani men. Our findings suggest the presence of a heightened theoretical risk for fractures in Fulani nomads, a risk that appears to be compounded by low BMI.

Our study has several limitations. First, estimation of daily dietary calcium intake by the Fulani herdsmen would have been illuminating, as it would have allowed for assessment of the correlation between calcium intake and SI. Second, serum calcium assays and estimation of serum markers of bone turnover, such as bone N-telopeptides and bone-specific alkaline phosphatase, would have provided useful information regarding the dynamics of bone metabolism in the herdsmen. Third, given the mobile nature of the Fulani men, our inability to recruit more participants might have introduced a bias due to the small sample.

Future studies should include direct or indirect assessment of calcium intake by the herdsmen as well as assays of serum calcium and markers of bone turnover in Fulani men. In addition, an assessment of the incidence of fractures among Fulani herdsmen will be enlightening. Furthermore, the contribution of genetic factors to bone density of Nigerian men warrants investigation.

We conclude that the Fulani herdsmen have a theoretically increased risk for bone fracture in a background of low BMI and potentially high calcium intake. Our findings may be supported by future studies of fracture rates in Fulani herdsmen. However, the wide confidence intervals for some of the estimates limit the conclusions that can be drawn.

Funding Information

This study was supported by a Fulbright Scholar award to RHG from the US State Department. All authors state that they have no competing financial relationships.