Objectives: To determine fracture rate, fractural risk factor, result of predicting the risk of femoral fracture by FRAX and GARVAN model. Subjects and method: We performed studying on 206 women over 40 years old including 176 in the community and 30 had fracture who lived in the Rachgia city, Kiengiang province.
Trang 1INVESTIGATION OF FRACTURE RATE, FRACTURAL RISK FACTOR DUE TO OSTEOPOROSIS AND PREDICTING
FACTURE RISK BY FRAX AND GARVAN MODELS
Hoang Ngoc Tang*; Pham Thanh Binh*; Doan Van De*
SUMMARY
Objectives: To determine fracture rate, fractural risk factor, result of predicting the risk of femoral fracture by FRAX and GARVAN model Subjects and method: We performed studying
on 206 women over 40 years old including 176 in the community and 30 had fracture who lived
in the Rachgia city, Kiengiang province On each women, we performed clinical examination, found the fracture risk factors due to osteoporosis and measured bone mineral density by DEXA method on Osteocore Station Mobile machine, then we used FRAX and GARVAN model to
determine level of neck femoral facture risk and other fracture risks Result: In fracture risk
factors, proportion of high age was the highest (91.7%), BMI < 18.5% was the lowest (2.4%) 2.4% of patients didn’t have risk factors fracture Patients have more than 2 factors accounted for high rate Osteoporosis and ages are significant to predict neck of femoral fracture in FRAX and GARVAN model BMI is valid for predicting neck of femur in FRAX model but isn’t valid for
predicting in GARVAN model Conclusion: In the fracture risk factors, age accounted for the
highest proportion and BMI account for the lowest proportion A number of patients has 4 risk factors accounted for the highest proportion FRAX and GARVAN model has valid for prognosing the risk of femoral neck fracture after 10 years
* Keywords: Osteoporosis; FRAX and GARVAN model
INTRODUCTION
Osteoporosis is a dysmetabolic of bones
causing losing intact of bone quantity,
decreasing bone density and leading to
fracture which is last result of osteoporosis
The rate of patients with facture caused
by osteoporosis is increasing In the United
States, there are over 300.000 patients
with femoral neck fracture caused by
osteoporosis and about 5.2 millions patients
who have fracture is over 45 years old in
the beginning of 21th century [1] In the
fracture types, vertebral fracture, femoral
neck fracture, wristbone neck fracture (Colles and Smiths fracture) are considered as the special types of fracture due to osteoporosis In fact, studies have shown that fracture of elderly patients related with low bone density state [1] Therefore, most
of fracture types relating to elderly patients can be considered as a fracture due to osteoporosis In a epidemiological research
on 10,750 women in 7.6 years, Robbins
et al (2007) showed that there were numerous fractural risk factors such as: age, weight, height, race, physical activity,
* 103 Military Hospital
Corresponding author: Pham Thanh Binh (thanhbinh4121d@gmail.com)
Date received: 20/11/2017 Date accepted: 24/01/2018
Trang 2fracture history, smoking, history of using
glucocorticoid, diabetes mellitus [6] In
clinical practice, not only bone density is
risk factor but also many others related to
a fracture Measurement of bone density
by dual-energy X-ray absorptiometry has
many limitations Therefore, there must
be a much better system to predict
fracture And FRAX and GARVAN model
showed, they based on risk factors of
patient specially bone density to evaluate
fracture risk of patients We performed
this study with the aim: To detect fracture
osteoporosis and predict fracture risk
following FRAX and GARVAN model
SUBJECTS AND METHODS
1 Subject
206 women in our study were randomly
selected and lived in Rachgia city,
Kiengiang province, satisfied the condition
* Sample size:
Applying formula for determining
sample size of a proportion of population:
2
2 2 1
1 d
p p Z n
−
=
−α
- n: is necessary sample size for
cross-sectional study
- p: is osteoporosis proportion in the
population chosen following a result of
domestic study (15.4% following Vu Thi
Thu Hien)
- d: is estimated error, with d = 0.05
then Z2(1 – α/2) = 1.962
Sample size has to be higher than 196
subjects who respond condition
* Selected criteria:
+ Women over 40 years old living at
Rachgia city, Kiengiang province
+ Agree to take part in this research
* Excluded criteria: women having
Cushing syndrome, hyperparathyroidism, hyperthyroidism, chronic liver disease, malabsorption syndrome, getting ovariotomy, using replaced hormone, metastasis cancer, multiple myeloma, motionless over 3 months, chronic bone-joint disease, pregnant , didn’t agree to take part in this study
2 Method
- Descriptive cross-sectional study
- Patients were asked about disease history, taken clinical examination to discover fracture risk factors due to osteoporosis
- Measure bone mineral density
- Use FRAX and GARVAN model to estimate fracture risk To classify fracture risk of neck of femur after 10 years following FRAX and GARVAN model: ≥ 3% high risk, < 3% low risk
- Data was processed by SPSS software, compared 2 proportions, compared
2 average values
RESULTS AND DISCUSSION
Osteoporosis has numerous risk factors, fracture also has numerous risk factors, but the risk factors of fracture and osteoporosis are different In this study,
we investigated proportion of fracture risk factors of 206 subjects
* Risk factors of fractures:
Age ≥ 60: 189 patients (91.7%); BMI
< 18.5: 5 patients (2.4%); no procreation:
21 patients (10.2%); menstruation after
15 years: 151 patients (73.3%); menopause before 53 years: 159 patients (77.2%);
after menopause > 10 years: 165 patients (80.1%); osteoporosis: 23 patients (11.2%);
Trang 3fall history: 38 patients (18.4%); family has
member get fracture: 22 patients (10.7%)
Group of risk factors with low proportion
included: BMI < 18.5, no procreation,
osteoporosis, fall history, family has
member get fracture
* Distribution of patients following
amount of risk factors:
No risk factor: 5 patients (2.4%); 1 factor:
10 patients (4.9%); 2 factors: 27 patients
(3.1%); 3 factors: 56 patients (27.2%);
4 factors: 64 patients (31.1%)
- There was only 2.4% of subject
having no factor
- Subject had 4 risk factors took the highest proportion 31.1%, next was 3 risk factors 27.2%, from 5 risk factors was 21.3% In comparison with other authors’s study, we have seen that our result was different because the feature of sample selection There are many studies which selected subject over 20 years, therefore, proportion of risk factors was low Our subject is menopause women whose the lowest age was 48, and the highest was
85, therefore proportion of fracture risk factors was high and subjects appear many risk factors were also high
Table 1: Relation between fracture and osteoporosis
OR, p
OR = 9.8 (3.43 - 28.71)
p < 0.05
- Fracture had relation with osteoporosis
Proportion of fracture subjects in osteoporosis
group was significantly higher than
non-osteoporosis group OR = 9.8, p < 0.05
- Osteoporosis makes increasing fracture
risk, specialy neck of femur and vertebrae
Low bone density is one of the most
important factors of osteoporosis which
cause fracture Bone mineral density of
adult is detected by both peak bone
density at adulthood and lose-bone state
in the next years Osteoporosis-induced
fracture is caused by consequence of
trauma, can be slight or severe and
influence their quality and strength of
bones Although there is a close relation between mineral density with fracture risk, but there are also others factors play the important roles in the development of the bone’s strength and are absolute risk induce fracture Decreasing of mineral density and strength of bone is cause of fracture due to osteoporosis
- Currently, the general trend in the assessment of fracture risk is based on multi-factors, rather than just relying on bone density factors or a history of previous fractures There are two most popular models to estimate the risk of fracture: Fracture Risk Calculator Model
Trang 4GARVAN (Garvan) of the GARVAN
Institute of Medical Research (Australia)
and FRAX model of World Health
Organization Both models have not been
widely applied in Vietnam In this study,
we applied the calculation model and
Garvan Frax assessment predicted a
high risk of fracture in Vietnam World Health Organization recommends high-risk prediction model FRAX fracture ≥ 3% broken femur and ≥ 20% for the whole body fracture (and need treatment) In our study, determining the value of high-risk prognostic FRAX > 3% and GARVAN > 3%
Table 2: Predicting 10 years risk of femoral neck fracture by age group
Age group
(94.1)
1 (5.9)
16 (94.1)
1 (5.9)
(98.4)
2 (1.6)
60 (46.9)
68 (53.1)
(80.3)
12 (19.7)
8 (13.1)
53 (86.9)
- Both the FRAX and GARVAN models
have the predicted values with the risk of
femoral neck fracture The higher the age
predicted, the higher fracture risk
- The FRAX model predicts age from
60 to 69: high risk 1.6%, low risk 98.4%;
age over 70, high risk 19.7%, low risk
80.3%; the difference was statistically
significant with p < 0.05
- The GARVAN model showed the
high risk of 60 - 69 age (53.1%), age over
70 was 86.9%, the difference was
statistically significant with p < 0.05
- Both the FRAX and GARVAN models
predict the group with age under 60 with
high risk (5.9%) and low risk (94.1%)
- The higher the age is, the higher the
risk of fractures is For every 5 years of
age, the risk of fracture increased 1.8
times [7] According to Kung (2007), in
10 years post-menopausal women: 10 years old increased risk of fractures predicted increase in 2.2 times
- In our study, the value of prognosis for FRAX risk fractures accounted for 5.9%, 1.6% and 19.7%, respectively aged 60, 69 - 60 < age and ≥ 70 years
of age, prognosis value is higher-risk age GARVAN < 60 was 5.9%, from 60 -
69 age was 53.1% and from 70 years of 86.9% Therefore, the prognosis looks GARVAN close to reality higher than the clinical prediction of FRAX, this also conforms with the review of a number of authors [2, 3, 5] So to identify high risk fracture of an individual, the doctor needs to refer to both models rather than just a model
Trang 5Table 3: Predicting 10 years risk of femoral neck fracture by BMI group
BMI group
(60.0)
2 (40.0)
4 (80.0)
1 (20.0)
(88.4)
7 (11.6)
52 (86.7)
8 (13.3)
(95.7)
6 (4.3)
132 (93.6)
9 (6.4)
- In FRAX model, BMI can prognose
the risk factor of femoral neck fracture,
but it has no value in GARVAN model
- High risk of femoral neck fracture with
BMI <18.5 was 40%, BMI > 23 was 4.3%,
and this significant difference with p < 0.05
in FRAX model
- By the time, in GARVAN, this risk
factor was 20% and 6.4%, there was no
significant difference with p > 0.05
According to Kanis (2008), low BMI
can prognosis the risk of femoral neck
fracture and this value will increase if
clinical factors are added Women at
65 years old with BMI < 20 kg/m2 had risk
of femoral neck fracture 1.3%, which increased 1.7% to 3.2% per 1 clinical factor added, with 6 risk factors, the risk
of femoral neck fracture was 30% [4]
In this research, high risk of femoral neck fracture in under weigh women was 40.0% at FRAX predictive value (p < 0.05) and at GARVAN model was 20%
So low BMI - one of 12 predictive factors
of FRAX model - is one of the valuable factor in predicting the risk of fracture
Table 4: Predict 10 years risk of femoral neck fracture due to history fracture of family
History fracture of family
(96.2)
5 (2.8)
84 (45.6)
100 (54.4)
(54.5)
10 (45.5)
0 (0.0)
22 (10.,0)
History fracture of family has predictive
value in both models
According to FRAX model, predictive
value of femoral neck fracture in family
which had history fracture was 45.5%, while the rate was 2.8% in group of family without history factor, there was no significant difference with p < 0.05
Trang 6In GARVAN model, predictive value of
femoral neck fracture in family has history
fracture was 100% and in the non-history
fracture family was 54.4%, difference has
no meaning
History fracture of family is one of 12
predictive factors in FRAX model But the
high risk of femoral fracture that FRAX
model can predict is very low (0%) Study
by Billington (2016) has shown that
history fracture of family has significant
effect on predicting the risk of fracture in
elderly women of FRAX model If we
exclude this factor out of FRAX model,
predictive value will decrease 1.5 - 4.3%
per women [2]
CONCLUSION
In the fracture risk factors, age related
factor accounted for the highest
proportions and BMI account for lowest
proportion
A number of patients with 4 risk factor
that accounts for the highest proportion
FRAX and GARVAN models have valid
for prognosing the risk of femoral neck
fracture after 10 years
REFERENCES
1 Nguyen Van Tuan, Nguyen Dinh Nguyen
Osteoporosis: Cause, diagnosis, therapy, prevention Medical Publishing Hourse 2007,
pp.120-143
2 Billington E.O, Gamble G.D, Reid I.R
Reasons for discrepancies in hip fracture risk
calculators Maturitas 2016, 85, pp.11-18
3 Bolland M.J et al Evaluation of the
FRAX and GARVAN fracture risk calculators in older women J Bone Miner Res 2011, 26 (2), pp.420-427
4 Kanis J.A et al FRAX and the
assessment of fracture probability in men and women from the UK Osteoporos Int 2008, 19 (4), pp.385-397
5 Pluskiewicz W et al High fracture
probability predicts fractures in a 4-year follow-up in women from the RAC-OST-POL study Osteoporos Int 2015, 26 (12), pp.2811-2820
6 Robbins J et al Factors associated with
5-year risk of hip fracture in postmenopausal women Jama 2007, 298 (20), pp.2389-2398
7 Taylor B.C et al Long-term prediction of
incident hip fracture risk in elderly white women: study of osteoporotic fractures J Am Geriatr Soc 2004, 52 (9), pp.1479-1486