Osteoporosis research using peripheral quantitative computarized tomography (pQCT) of the tibia

The Rotterdam Study is a single-center prospective population-based study of
determinants of chronic disabling diseases in the elderly and has been approved
by the medical ethics committee of Erasmus MC. The design and rationale of the
study has been described earlier1. The original cohort (RS-I) includes 7,983
respondents. As of April 2000, an extension cohort of 3,011 individuals (RS-II)
was included; and since March 2007, a younger cohort, including 3,932
individuals age >45 years are being assessed under the same protocols.
Historically, DXA measurements have been obtained using GE-Lunar densitometers
(DPX-L, Prodigy and iDXA). Incident fractures are collected from computerized
records of the general practitioners (GPs) followed from baseline currently
until January 1, 2007 (mean follow-up 12.2 years). In addition, X-rays scans of
the thoracolumbar region have been collected and scored for vertebral fracture.

Osteoporosis has been operationally defined using bone mineral density (BMD).
Nevertheless, the BMD measurement alone is not optimal for detecting
individuals at high risk of fracture. Even though fracture risk is very high
when BMD is low (high specificity), risk is not negligible when BMD is normal
(low sensitivity). Bone strength needs to be significantly weakened before
density criteria are met to diagnose osteoporosis. By implementing a
high-resolution imaging technique one can expect to identify parameters of bone
structure and quality that are subject to be detected earlier and can help in
the prediction of osteoporosis and fracture. Most important this can be done in
a timely manner and within a period when the consequences of the disease are
still reversible.

One of these techniques is computerized tomography (CT). CT scanning allows
higher spatial resolution and improved delineation of bone architecture
overcoming many limitations of the DXA measurements. Yet, volumetric CT has
limitations including cost, accessibility and radiation exposure. The
peripheral quantitative computerized tomography (pQCT) has become the method of
choice in large scale projects, being fast, requiring no external shielding for
radiation protection (radiation exposure < 0.001 mSV) and being highly
reproducible.Trabecular architecture and cortical thinning are readily
appreciable in the osteoporotic subject, and are reflected on the strength
indexes measured by pQCT. Cortical bone parameters measured with pQCT include
volumetric BMD, periosteal circumference and cortical thickness assessments.
Trabecular volume, spacing, connectivity and number can only be assessed with
MRI or high-resolution pQCT. Nevertheless, trabecular vBMD by pQCT is strongly
correlated with total to trabecular bone volume (BV/TV) ratio and can be used
as a reliable surrogate.

The current proposal is to acquire a STRATEC XCT2000 device to make pQCT
measurements in all (n~13,000) Rotterdam study participants. CT scanning allows
higher spatial resolution and improved delineation of bone architecture
overcoming many limitations of the DXA measurements. Yet, volumetric CT has
limitations including cost, accessibility and radiation exposure. pQCT has
become the method of choice in large scale projects, being fast, requiring no
external shielding for radiation protection (radiation exposure < 0.001 mSV)
and being highly reproducible. Trabecular architecture and cortical thinning
are readily appreciable in the osteoporotic subject, and are reflected on the
strength indexes measured by pQCT. Cortical bone parameters measured with pQCT
include volumetric BMD, periosteal circumference and cortical thickness
assessments. Trabecular volume, spacing, connectivity and number can only be
assessed with MRI or high-resolution pQCT. Nevertheless, trabecular vBMD by
pQCT is strongly correlated with total to trabecular bone volume (BV/TV) ratio
and can be used as a reliable surrogate.

The current proposal envisions using an STRATEC XCT 2000 device to make pQCT
measurements in ~12,000 Rotterdam study participants. This is attempted
considering that further research is needed to understand the underlying
biology, improve risk prediction and develop novel disease interventions
(treatments) for osteoporosis disease. The implementation of this pQCT imaging
technology in the Rotterdam Study will bring the field one step further in this
direction.

Note: the use of the exact same device has been approved by the ERASMUS MC
METC (MEC-2012-165 amendment 2 NL40020.078.12 v06) to measure all children
from the Generation R study (n ~6000 children).

Primary Objective
This proposal concerns the inclusion of a peripheral quantitative computed
tomography (pQCT) device in the Rotterdam Study to perform osteoporosis
etiological research. Since 1990, the Rotterdam Study has performed research in
the general population, regarding diverse conditions and disease including
osteoporosis. Most of this research has been done using a dual-energy X-Ray
Absorptiometry (DXA) device, measuring bone mineral density to assess the
presence of osteoporosis and risk of fracture. One of the disadvantages of DXA
is that it is a two dimensional assessment, which does not characterize
appropriately the three-dimensional structure of bone. Bone parameters like
trabecular and cortical structures and material distribution (geometry) cannot
be studied using DXA. It is for this reason that we want to expand the
osteoporosis research program by means of including a peripheral quantitative
computerized tomography (pQCT) device. PQCT is a 3-D technique, which allows
the study of skeletal and muscle tissue at the tibia. Including this
complementary measure in the Rotterdam Study will allow studying additional
determinants of fracture in the elderly general population.

Secondary objectives
Given the multidisciplinary origin of the Rotterdam study we will be able to
use the pQCT measurements to address additional research questions surrounding
the use of additional assessments already present in the study, like medication
use, comorbidity and detailed state of the art genetic measurements among
others. This approach will allow identifying new approximations for the
diagnosis and treatment of osteoporosis. In particular, we will study the
biomechanics determinants of osteoporosis and fracture susceptibility in
individuals from the general population aged 45 years and over. Ultimately we
will pursue the construction of a risk profile that can complement the existent
risk factor by inclusion of genetic and imaging data of the pQCT.

The Rotterdam Study is a long-standing prospective population-based study in
15.000 aged 45 years and over living in the Rotterdam suburb Ommoord. The
Rotterdam Study is an initiative from the Department of Epidemiology of the
Erasmus University Medical Center in Rotterdam. The Rotterdam Study is
nationally and internationally recognized for its contribution to the
understanding of the determinants of disease and disability of hundreds of
entities, including those related to locomotor disease. Description of the
rationale of the study with a succinct listing of the scientific publications
is to be found in a recent publication in the European Journal of Epidemiology
(Hofman et al. 2014). All scientific publications (more than 1000) are listed
in the website of the Department of Epidemiology of Erasmus MC
(http://www.epib.nl). The medical ethics committee of Erasmus MC has approved
the Rotterdam Study and all participants have provided written informed
consent.

As with any device employing radiation sources care must be taken to avoid
unnecessary exposure. The device is approved for use by the Erasmus MC
Radiation Protection Unit (Stralingsbeschermingseenheid - SBE). This provisions
are clearly described for the operator in the protocol of operations of the
pQCT device (see Annex) and for the Rotterdam Study participants in the
information folder (see Annex).

The XCT 2000 radiation dose is measured using thermoluminescent dosimetry in
order to simulate the x-ray attenuation and scatter of a human forearm. An
array of TLD chips is used so that the actual distribution of the dose may be
determined. The skin dose is 90 µSv for the CT scan and 35 µSv for the scout
scan. According to W. Kalender (Osteoporosis Int (1992) 2: 83-97) the effective
dose can be calculated: Multiplication of the skin dose of 90 µSv with a
relative amount of 2% of the irradiated marrow from the total bone marrow
yields the organ dose of 1.8 µSv. Multiply this value with the weighting factor
for bone marrow recommended by the ICRP of 0.12. The resulting effective dose
is 0.22 µS. For the operator the dose is negligible. During a scan procedure,
the total leakage plus scatter radiation is less than 10 µSv/hr at the scanner
aperture.

The Stratec device has been used since 1998 in numerous studies including
hundreds of thousands of individuals placing no question about the safety of
its operation.