QUS characteristics in Normal Population: a Mini Review and our experience

Abstract

Objectives

Quantitative ultrasound (QUS) is a bone densitometry method that is less expensive and more portable than DXA. It is also noninvasive. QUS parameters include speed of sound (SOS), broad band ultrasound attenuation (BUA), and stiffness index (SI). This study defined normal values of QUS parameters in Iranian men and women.

Methods

QUS of heels measured in 258 Iranian men and women, aged 20–76 y/o. They were participants of Iranian Multicenter Osteoporosis study (IMOS), selected by randomized sampling. QUS device was an Achilles+ (GE-Lunar) device.

Results

Percentiles of SI (2.5%, 50%, and 97.5%) determined. We found a good agreement between the Iranian reference values and western reference (used by device) value in defining normal and osteoporotic people (κ = 0.875).

Conclusion

Results from this study suggest that QUS of the heel may be a good method for diagnosis of low bone mass in different regions.

Introduction

Osteoporosis is a common progressive metabolic bone disease resulting in low bone mineral density (BMD) and affects bone microarchitecture. In postmenopausal women it is characterized by reduction of BMD ≥ 2.5 SD below the average density in young adults. This disease is a major public health concern as it links to osteoporosis fracture. Osteoporotic fracture is the most common cause of morbidity and mortality of this disease and more (about 90%) occurs in vertebrate and femur [1–3]. Dual energy x-ray absorptiometry (DXA) is the gold standard for osteoporosis diagnosis and evaluating the fracture risk by measuring BMD at the region of interests (ROIs). However this 2-dimentional method is not widely available and portable, is expensive and uses radiation. On the other hand DXA is not a good method for assessing other characteristics of bone like microstructure. It results fully appreciation of other methods such as quantitative ultrasound (QUS) that have the ability to assess quality of bone [4, 5] .QUS measures speed of sound (SOS) [6]. The QUS machine used in this study, Achilles+, also measures broadband ultrasound attenuation (BUA) and calculates the stiffness index (SI).

However there is not a good correlation between QUS and DXA in the diagnosis of osteoporosis. An Iranian study showed the kappa values as 0.29–0.35 in calcaneus. Other researchers found it to be 0.2–0.8 [7–11]. In phalanx area kappa value shown to be 0.317 [12].

In a meta-analysis, the authors concluded that the determined sensitivity and specificity of QUS are too low to diagnose of osteoporosis according to DXA-determined osteoporosis threshold (T-score= -2.5) [13]. This is why many researchers investigating other cut-off point for osteoporosis diagnosis using QUS [7, 14].

However, many studies have found that QUS can differentiate between patients with and without fragility fracture [15], and also predict the pathologic fractures [3, 16, 17]. May it means, DXA based BMD and BMD by QUS, are associated independently with the risk of fracture [18].

Broadband ultrasound attenuation depends on skeleton characteristics such as diffraction, scattering and absorption and trabecular orientation [19]. It seems SOS relates to elasticity of cancellous bone [18]. Bone elasticity relates to collagen cross-links, size and composition of mineral crystals and anisotropy [20]. Microarchitecture or geometrical characteristics and volumetric BMD are also named as properties that QUS measures [21]. Even in some studies, QUS in comparison with DXA, showed better correlation with fertility parameters and vitamin deficiency [22–24]. In clinical studies, a significant relationship between the QUS parameters and age and gender is found [25, 26].

Before this method can be used for screening or diagnosis, the normal range of its parameters should be established for any specific population due to different genetic and environmental factors. On the other hand in special groups such as institutionalized patients, QUS may be first choice because BMD measurement by DXA needs much effort from patients and health care staffs. It is another reason for use of normal reference range of any population that these patients belong to [13].

Normal range of quantitative ultrasound parameters in the different populations are established [27]. In this study we assess and characterized of the QUS parameters and the normal range in the Iranian population.

Methods:

Subjects

Normal subjects aged between 20 and 76 years entered this study. Female to male ratio was 151/107. Mean age of men and women was 43.28 ± 15.59 and 42.67 ± 12.12, respectively. They were enrolled participants of the Iranian Multicenter Osteoporosis study [IMOS). Number of post-menopause women was 42. To confirm inclusion of healthy participants to the study, subjects with the following diseases or conditions were excluded: known history or evidence of rheumatoid arthritis, thyroid, parathyroid or adrenal disease, hepatic or renal failure, metabolic bone disease, type I diabetes mellitus, sterility, oligomenorrhea, malignancy, malabsorption, immobility for more than one week, pregnancy, lactation, smoking more than 10 cigarettes per day, alcoholism, and medications influencing bone metabolism [28].

Bone mineral density measurement

Quantitative ultrasonography of heel of 258 participants performed by one ultrasonography instrument (Achilles; Lunar Corporation). According to the manufacturer’s instruction, there is no preference for the right or left foot. The CV of Achilles + is 1.7%. For quality assurance, the Achilles + is supplied with two quality assurance phantoms: a water reference insert and a calibration standard. A single technician performed all QUS measurements. Achilles measures broadband ultrasound attenuation [BUA), speed of sound (SOS), and calculates stiffness index (SI). SI calculates by the machine using this equation: SI = 0.67 X BUA + 0.28 X SOS—420. Higher SI signifies better bone health. In an Iranian study T-score of − 1, found to be a suitable cutoff level for osteoporosis in Iranians, so we used this cut-off for low BMD diagnosis in our study [7].

Statistical analyses

Continuous parameters are reported as means, standard deviations, and ranges. Categorical parameters are reported as simple percentages.

One-way ANOVA test was used to define heterogeneity and homogeneity for the QUS parameters between the age groups (10 year groups) in both sexes. Prior to this, transformation methods were used for establishing the normal distributions of the QUS parameters (an assumption needed for the use of ANOVA). Age groups that had homogenous results were combined, and only heterogenic age groups were reported. SI percentiles for both sexes were established at 2.5%, 50%, and 97.5%.

The diagnostic agreement in identifying individuals as low BMD was assessed with the use of κ scores for T scores of QUS calculated with Iranian reference range and one that calculated with reference ranges of device.

SPSS software (SPSS Inc, Chicago, IL) was used to perform statistical analyses.

Results

Table1 shows the characteristics of the subjects participating in the study. There were three heterogeneous groups of QUS parameter in the women studied: ages 20–49, 50–59, and 60–69. The relationship between age and sex with QUS parameters (as is shown in Table2) was significant. The effect of years after menopause was not significant on changes of QUS parameters. QUS parameter in men were homogenous in all of the men. Tables3 and 4 show the mean QUS parameters for the different age groups and gender. There was no significant difference in the mean SI for men and women aged between 20 and 40 years of age when compared with reference data for the QUS device. 2.5, 50, 97.5 percentiles of SI (stiffness index) for both sexes established (Figs.1 and 2).

Table 1.

Characteristics of participants:

Parameters	Numbers
Men/women	108/148
Age of women ( range)	42.37+- 12.12 ( 20–72)
Age of men ( range)	43.28+- 15.59 ( 20–76)
Age of menopause ( range)	47.54 ± 5.90 (30–57)
Weight of women ( range)	68.87 ± 12.34 (40–105)
Weight of men ( range)	74.35 ± 12.10 (53–114)
height of women ( range)	157.25 ± 6.2 (140–173)
height of men ( range)	171.16 ± 7.08 (157–193)
BMI of women ( range)	27.90 ± 5.08 (18.73–43.17)
BMI of men ( range)	25.46 ± 4.33 (16.77–39.91)

Table 2.

Beta-coefficients (standardized) of parameters

Beta- Coefficients Parameters	Age	Sex
SOS	0.406	-
BUA	0.214	0.313
SI	0.149	0.424

Table 3.

Mean ± SD of QUS parameters in different age groups of women

	QUS Parameters Mean (CI 95%)
Age Groups	SOS	BUA	SI
20–49	1558.9 (1497.4-1620.4)	126.6 (100.1-148.5)	100.9 (74.6-127.2)
50–59	1530.2 (1466.9-1593.5)	116.9 (95.6-134.8)	86.1 (60.4-111.8)
60–69	1514.1 (1469.6-1558.6)	107.7 (68.2-136.1)	74.6 (45.6-103.6)

Table 4.

Mean ± SD of QUS parameters in 20–70 age groups of men

	QUS Parameters Mean (CI 95%)
Age Groups	SOS	BUA	SI
20–69	1552 (1472.6-1640.5)	128.3 (100.3-151.2)	100 (62.8-137.1)

Fig. 1.

Percentiles of Stiffness Index (SI) in normal Iranian women

Fig. 2.

Percentiles of Stiffness Index (SI) in normal Iranian men

(Drown by SPSS software)

There was a good agreement between the diagnosis of osteoporosis in participants, when 20–40 year old groups from different countries used as reference groups (κ = 0.875).

Discussion

This study provides population-based reference data for the stiffness index of the heel in Iranian men and women. To the best of our knowledge, this is the first study that evaluated normal range of QUS in Iran.

We also found a good agreement between the diagnoses of low BMD in participants, when 20–40 year old groups from different countries used as reference groups.

In our study, BUA and stiffness index significantly related to age and sex. This was similar to other studies in Spain and Poland [25, 26]. It comes clear when we use this system. It means when we work with QUS (Achilles+), changing the data of height and weight, don’t changes the results of BMD diagnosis. It may be a defect when it compares with DXA.

The effect of years after menopause was not significant on QUS parameters. Menopause is an important factor for bone loss, so finding why there was no such significant correlation, needs more investigation, though it is shown also in another study [29]. The participants of this study were 201 postmenopausal Malaysian women aged 45–71 years.

For plotting normal range of our people, we calculated the means of QUS parameter. In Italy, Gonnelli and colleagues project on 268 male without pathologic fracture, showed average of SI as 87.2 [30]. In Malaysia 2017, Malaysian men had mean of stiff index as high as 91.9 [31].

Hammad in Saudi Arabia aimed to find normal data for calcaneus QUS in normal women. Their results showed mean of SI as 93.2 [32]. In young Swedish women, in a study published in 2016, mean of stiffness index found to be 99 [33].

SI of our people is higher than above populations. However when we wanted to find the agreement between the diagnosis of low BMD by two reference data, we needed to have the two normal ranges. After defining this in our population, the kappa method was used for defining the agreement between low BMD diagnosis when two different normal ranges used. We found a good agreement between the normal range of Iranian people and Germans and Americans that used as reference data in this device. This is important because when we used DXA as a method of bone mineral densitometry, Iranian females had a higher BMD than Japanese, Canadian, Hong Kong and Lebanese females and lower than Americans in spinal region [34], and using two reference data for diagnosis of low BMD by DXA, caused different diagnosis in patients [35]. In the latter study, when BMD results of adult beta thalassemia major patients (as cases of secondary osteoporosis) compared with reference range of different population (Iranians and Americans), different prevalence of low BMD reported. This difference may suggest a preference for using QUS as the primary method for assessing the bone disorders. This good agreement may relate to the ability of QUS in assessing qualitative factors that are more similar between different populations [4, 5]. Relatively low sample size and cross-sectional method of this study are from limitations of this study.

Conclusion

QUS of the heel, as a low cost and non-invasive method, may is a good method for diagnosis bone diseases in different regions. Before establishment of this suggestion, larger and multidimensional studies are needed.

Conflicts of interest/Competing interests:

Minoo Najafi, Maryam Najafi, Mitra Mahdavi-Mazdeh, Sima Maziar, Abbas Ali Keshtkar, Mojtaba Sedaghat, Fatemeh Zare-Bidaki, Bagher Larijani, Zohreh Hamidi declare that there is no conflict of interest/competing interests to report.

Authors contribution:

Study design: Mojtaba Sedaghat, Bagher Larijani, Zohreh Hamidi. Study conduct: Zohreh Hamidi. Data collection: Zohreh Hamidi, Ali Reza Khalilifard Data analysis: Abbas-Ali Keshtkar, Mojtaba Sedaghat Data interpretation: Abbas-Ali Keshtkar, Mojtaba Sedaghat, Zohreh Hamidi. Drafting manuscript: Zohreh Hamidi, Minoo Najafi, Maryam Najafi, Revising manuscript content: Abbas-Ali Keshtkar, Mojtaba Sedaghat, Bagher Larijani, Zohreh Hamidi Approving final version of manuscript: Zohreh Hamidi

Funding:

Funding Provided by Endocrinology and Metabolism Research Center of Tehran University of Medical Sciences.

Declarations

Consent to participate:

Available.