Osteoporosis Self Assessment Tool for Asian ( OSTA ) Index in Comparison to Quantitative Ultrasound of the Calcaneal in Predicting Low Bone Density

Introduction: Osteoporosis represent a worldwide public health problem, frequently resulting in fractures and leading to psychological problem, social consequences, functional limitation and poor quality of life. So it is important to identify those people who have high risk of osteoporosis, in order to reduce the incidence of osteoporotic fractures. The Osteoporosis Self-Assessment Tool for Asians (OSTA) index is a simple tool based on age and body weight. Calcaneal quantitative ultrasound (QUS) is another simple and low-cost instrument used to prescreen osteoporotic subjects. The aim of this study was to correlate between these two screening methods and to validate usefulness of Osteoporosis self assessment tool for Asian ( OSTA) in comparison with QUS of the calcaneum for the prediction of low bone density in Nepalese women. Methods: This was a two-year descriptive observational study comprising 100 subjects performed at TU Teaching Hospital, Kathmandu, Nepal, from 2007 January to 2009 January who completed semi structured questionnaire with subsequent measurement of Quantitative Ultrasound (QUS) of the calcaneum. The sensitivity, speciÞ city, and diagnostic accuracy of OSTA index was validated with the QUS T-score. Results: When the risk category was deÞ ned as OSTA index -1, and low QUS value as t-score -2.5, the sensitivity, speciÞ city and diagnostic accuracy of the index were 93.3%, 56.5%, and 62% respectively, and the area under the curve was 0.7651. When the low QUS value was taken as t score -1.0, the sensitivity, speciÞ city and diagnostic accuracy was 85.2%, 89.1% and 87% respectively. Conclusions: The OSTA index, a simple and free risk assessment tool, can be used to estimate the prevalence of low QUS values in Nepalese women and may help to increase awareness and prevention of low bone mineral density.


INTRODUCTION
Osteoporosis represent a worldwide public health problem, affecting 45% of women who are Þ fty years of age or older resulting in life time risk of 40% for the fractures of hip, vertebrae, and distal forearm 1 .It is important to identify people who have high risk of osteoporosis, in order to reduce the incidence of osteoporotic fractures.
Dual Energy X-ray Absorptiometry (DEXA) is gold standard to Bone Mineral Density (BMD), however it is not readily available in Nepal and also is a costly method for screening purpose.Osteoporosis Self assessment Tool for Asian (OSTA) is a clinical decisions making risk index originally developed for the use in post menopausal Asian population 2 .It is an inexpensive, simple tool based on age and body weight.Purpose of OSTA index is not to diagnose osteoporosis or low BMD but to identify women who are more likely to have low BMD who could then undergo BMD

Original Article
Osteoporosis Self Assessment Tool for Asian (OSTA) Index in Comparison to Quantitative Ultrasound of the Calcaneal in Predicting Low Bone Density.measurement for a deÞ nitive assessment.
Calcaneal quantitative ultrasound (QUS) is another simple low cost, instrument use to screen osteoporotic subjects 3 .
The aim of this study was to correlate between these two screening methods and to validate usefulness of Osteoporosis self assessment tool for Asian ( OSTA) in comparison with QUS of the calcaneum for the prediction of low bone density in Nepalese women.

METHODS
It was a descriptive observational study conducted at Department of Orthopedics, TUTH from 2007 January to 2009 January.Any women age 40 or above with no previous diagnosis of osteoporosis with at least 1 additive risk for osteoporosis were included in this study.Previous diagnosis of osteoporosis and secondary osteoporosis were excluded from the study.Ethical clearance was taken from the ethical clearance committee.Informed consent from all the eligible patients was taken after full explanation about the nature of the study.

Semi Structured Interview Schedule
All participants completed a structured questionnaire.Information on demographic proÞ le, age, age of menarche, age of menopause were obtained.Risk factors were stratiÞ ed into non modiÞ able and modiÞ able.Non modiÞ able risk factors were deÞ ned by; age 65 years, history of low energy fracture after the age of 40, history of Osteoporotic fracture in Þ rst degree relative and surgical menopause.ModiÞ able risk factors were deÞ ned by; smoking (current or past), weight of 127 lb or less, early menopause [age 45 yrs], prolonged premenopausal amenorrhea [ 1 yr]), Late menarche (age 15 years or more), low calcium intake (lifelong), excessive alcohol intake ( 2 drinks/ day), Excessive caffeine intake (2 to 4 cups/day) and inadequate physical activity. 4,5dentary lifestyle was deÞ ned as the occupation or activity that the subject engaged in most frequently in her life and accordingly subject was categorized as sedentary or heavy worker

Anthropometry measurement
Weight of the subject was measured in Kilogram with the standard measuring tool while subjects were standing, wearing light clothing and no shoes.

Bone mass assessment
Bone mass was assessed by broad band ultrasound attenuation using a QUS device.This device is small and portable, with a gel-coupled (dry) system that can measure BUA and speed of sound at the calcaneus.For all subjects, QUS was performed at the right calcaneus.
The T-score for each subject was calculated by using the peak BUA value for a deÞ ned population of young adults ( ASIAN)

Statistical analysis
Here data are presented as percentage and as mean (standard deviation).Sensitivity, speciÞ city, Diagnostic accuracy was calculated.Receiver operating curve analysis was performed and area under curve was calculated.To assess the internal validity of the index sensitivity was deÞ ned as the proportion of the subject with low T score correctly classiÞ ed by the risk index N OA J J uly -D e c e m be r 2 01 3| Vol 3| Iss ue 2 (true positive) and speciÞ city was deÞ ned as the proportion with normal T score correctly identiÞ ed by the risk index (true negative) .ROC curve provided a graphical representation of the overall accuracy of a test by plotting sensitivity against (1-speciÞ city) for all thresholds, while AUC quantiÞ ed the accuracy of the test.All statistical analysis were performed by using SPSS software for Windows, version 11.5 (SPSS, Inc., Chicago, Illinois).

RESULTS
Mean age of the subject was 58. 14  with the mean index of -0.860.
When the T-Score were calculated, 46 subjects were identiÞ ed as Normal with mean T Score of 0.4089, 39 subjects were having Osteopenia with the mean T score of -1.7764 and 12 subjects were identiÞ ed as Osteoporosis with the mean T-Score of -2.8792.Only 3 were identiÞ ed as severe osteoporosis with the mean T score of -3.1067.
When cut off point for the T score is taken as -1.0, 95% of the subjects in High risk , 86% of the subjects in intermediate risk and 16% of the subjects in low risk group were identiÞ ed as having low bone density (Table 1).There were 14 true positive , 37 false positive, 1 false negative and 48 true negative cases.When the QUS T score cutoff value was taken as -2.5 the OSTA index has Sensitivity of 93.3%, SpeciÞ city 56.5% and Diagnostic accuracy was 62%( When QUS T score cutoff value was taken as -1.0 , there were 46 true positive , 5 false positive , 8 false negative and 41 true negative cases and their Sensitivity was 85.2%, SpeciÞ city was 89.1% and Diagnostic accuracy was 87% (Table 3).
T score cutoff value of -1.0 has better speciÞ city and diagnostic accuracy than cutoff value of -2.5(Table 4).The sensitivity and speciÞ city of the OSTA index in relation to T score obtained by QUS calcaneum were plotted as receiver-operating characteristic (ROC) curves.The areas under the curves (AUC) were calculated and the OSTA index provided AUC of 0.7651(Figure 1).The results show that OSTA index is capable of selecting patients with low bone density as measured by QUS calcaneum.

DISCUSSION
Twenty one (21%) of the subject fall into OSTA class A i.e. high risk sub group with mean index of -5.590, a total of 30 (30%) of the subject fall into OSTA class B i.e. intermediate risk group with the mean index of -2.393 and 49 (49%) of the subject fall into OSTA class C i.e. low risk subgroup with the mean index of -0.860.

F. Richy, et al performed a validation and comparative study of OST in Caucasian in Belgium between 1996
to 1999 among 4035 post menopausal women and they had High risk group comprising 11%, intermediate risk group 47 % and low risk group 42%. 6Our study is comparable to their results in respect to the OSTA research group classiÞ cation of the subject included in our study.
T-score quantiÞ es the differences between the patients BMD and the mean value for young adults from the reference group. 6,7,8When the T-Score were calculated 46 (46%) of the subject were identiÞ ed as a Normal group with mean T Score of 0.4089, 39(39%) of the subjects were identiÞ ed as having Osteopenia with the mean T score of -1.7764 and 12 (12%) of the subjects were identiÞ ed as having Osteoporosis with the mean T-Score of -2.8792.Only 3 (3%)were identiÞ ed as having severe osteoporosis with the mean T score of -3.1067.Anand et al from India in 2000 performed QUS calcaneum of 1713 subjects during a nine month period from sep 98 to may 99.Using the WHO standard guideline they found that 48.9 % of the subject were found to have bone mineral density within normal limit, 39.9% were found to have osteopenia and 11% were found to have osteoporosis. 9Vu Thi Thu Hien, et al from Vietnam performed population based cross sectional survey at Hanoi city in 2003 and determined the crude prevalence of osteoporosis to be 15.4% when the QUS T-score cut off point was taken as < -1.8. 10 Our study has comparable results other studies despite of relatively small sample size.
We validated the OSTA index with the two different cutoff T score value obtained by the QUS calcaneum.The abnormal value taken for OSTA index is -1 as classiÞ ed by the OSTA research group.When the cutoff T score value obtained by the QUS calcaneum is taken as -2.5 as classiÞ ed by WHO as a Osteoporosis, Sensitivity , SpeciÞ city and diagnostic accuracy of OSTA Index was 93.3%, 56.5%, and 62% respectively.Similarly when the cutoff T score value obtained by the QUS calcaneum is taken as -1.0 as classiÞ ed by WHO as a osteopenia, Sensitivity , SpeciÞ city and diagnostic accuracy of OSTA Index was 85.2%, 89.1%, and 87% respectively.
The sensitivity and speciÞ city of the OSTA index in relation to T score obtained by QUS calcaneum were plotted as receiver-operating characteristic (ROC) curves.The areas under the curves (AUC) was ).The prevalence of osteoporosis ranged from 2% among women classiÞ ed as low risk (OSTA > -1) to 64% among those classiÞ ed as high risk (OSTA < -4). 12ris et al, in 2001 validated in Caucasian women using original SCORE population with1102 post menopausal women aged 45 or more with sensitivity of 88% and speciÞ city of 52%. 1 In a population-based sample of postmenopausal Japanese women 14 , the OST had a sensitivity of 90% and speciÞ city of 45%.This tool was similarly validated in Philippine 15 Studies have reported ranges varying from 24-95% and 56-93% respectively for sensitivity and speciÞ city depending upon the various cutoff. 16Our result is similar to the other validational studies to exclude persons with low risk for osteoporosis.However there is a difference in the percentage of women identiÞ ed as Osteoporosis in OSTA high risk group, which is less in our study than others.Similarly signiÞ cant proportion of women in intermediate risk group had osteoporosis.This may be due to the fact that we had enrolled women age 40 years or above irrespective of menstrual status where as these studies mostly included post menopausal or older age group.
OSTA index value of = or < -1 should be taken as a cutoff point to screen women so that maximum subjects with low bone density would be correctly identiÞ ed and subjected to further evaluation.When cut off point for the T score is taken as = or < -1.0, 95% of the subjects in High risk , 86% of the subjects in intermediate risk and 16% of the subjects in low risk group were identiÞ ed as having low bone density.
Though the OSTA risk level classiÞ cation had poor correlation with the classiÞ cation with the WHO category based on T score, OSTA index cutoff value of -1.0 was equally good on identifying women with the low QUS value of -1.0.

CONCLUSION
OSTA index can be used as a Þ rst line screening tools in the clinic where detailed evaluation of osteoporosis is neither practical nor beneÞ cial.OSTA cutoff value of = or <-1.0 is the most accurate index in our study.

Figure 1 .
Figure 1.QUS device and Measurement of bone density

Figure 1 .
Figure 1.Receiver operating curve analysis years with minimum age of 40 to maximum age of 84 years.Similarly mean weight of the subject was 53.91 kg with minimum weight of 28 kg to maximum weight of 92 kg.Mean age of menopause was 46.60 years with the minimum age of 38 years to maximum age of 57 years.The Osteoporosis Self Assessment Tool for the Asian was classiÞ ed according to the original OSTA research group classiÞ cation.A total of 21 subjects fell into OSTA class A with mean index of -5.590; 30 subject fell into OSTA class B with the mean index of -2.393 and 49 subject fell into OSTA class C (low risk subgroup)