Diagnostic performance of dynamic four-dimensional computed tomography (4D CT) for analyzing wrist ligament injuries - a reproducibility study of the image acquisition protocol.

The wrist is a complex array of joints. It involves articulation of eight small
carpal bones, two forearm bones and five hand bones, with intricate inter- and
intra-osseous kinematics. The wrist facilitates multidirectional movement in a
synergistic manner that allow for the unique range of mobility of the hand and
forearm. Intrinsic (interosseous) and extrinsic (capsular) ligaments are the
primary passive stabilizers of the wrist. The muscles of the hand and forearm
are secondary, dynamic stabilizers.
Of all the injuries, hand and wrist injuries have the highest social and
economic impact; annual health costs and loss of labor productivity of
approximately 540 million Euros1. Part of the wrist injuries concerns ligament
lesions which, untreated, will progress to carpal instability and finally
osteoarthritis2-5. Out of the annual 25.000 wrist injuries, approximately 10%
will lead to instability2-5, for which reconstruction and salvage operations
are needed, costing each ¤3.700 including
physiotherapy; a total of 9.25M¤. After surgery, patients get 3-6 months
revalidation and are unable to work, costing ¤410 per day7; a total of ¤55.350
per person or 138M¤ in total. If patients are diagnosed earlier, less surgeries
will be necessary, drastically decreasing the societal costs and patient
discomfort.

Patients with an early diagnosis of wrist ligament injuries may benefit from
less-invasive treatment options, have a better prognosis, and experience less
residual disability 13. However, (pre)dynamic carpal instabilities demonstrate
abnormalities only during motion. That is why the diagnosis of these wrist
disorders currently relies largely on clinical findings, as patients with
(pre)dynamic instability often have normal appearances on static conventional
imaging studies, including plain radiographs, magnetic resonance imaging (MRI)
and ultrasound 14-17. Real-time fluoroscopy is currently the only imaging
modality that can be used to detect dynamic abnormalities in these patients18.
However, fluoroscopy suffers from several drawbacks: images are limited to
2-dimensions, bone superimposition is constantly present, and the sensitivity
of the examination is highly operator dependent. This makes it impossible to
reproducibly and objectively quantify any abnormalities, if present.

In cases of inconclusive radiographic evaluation, MRI or CT arthrography can be
used to assess ligament structural integrity but not ligament sufficiency,
which can be particularly problematic in patients with partial ligament tears.
The sensitivity and specificity of these static imaging techniques are
relatively low compared to arthroscopy and varies from 60-80% and 70-90%,
respectively.

Because of its high sensitivity and specificity, wrist arthroscopy is still the
gold standard for diagnosing intra-articular pathology and ligament tears of
the wrist19,20. Arthroscopy however, is an invasive and relatively expensive
procedure with a complication rate of 2%. Another drawback of arthroscopy is
the fact that the function of the ligaments can still not be dynamically
assessed for three reasons: 1) the wrist and soft tissue are distended during
the procedure to open the wrist joints for introduction of the scope and
instruments preventing movement and 2) the patient is unable to move the wrist
because of the anesthesia and 3) movements of the wrist is not possible because
of the rigidity of the scope that has been inserted into the wrist.

A promising imaging method that could contribute to this field is 4D CT
scanning. Using recently developed, continuous acquisition of 3D CT images
during movement, it may be capable of assessing patients with suspected
instability with higher diagnostic performance by facilitating kinematic
assessment of wrist motions. The superior 3-dimensional spatial resolution
allows this technique to accurately detect even slight positional changes of
the bones, in contrast to 2D fluoroscopy21. While conventional static 3D
imaging methods can provide valuable information about carpal bone anatomy and
alignment, dynamic imaging has the advantage of assessing the carpus, distal
radius and ulna, and joint spaces throughout the range of movements and
capturing their complex interplay. This will provide a way to better
understanding of normal wrist kinematics and improve diagnosis of subtle carpal
instabilities and decrease the need for arthroscopy. Moreover, if surgical
intervention is deemed appropriate, the source of the instability might be more
precisely identified. Detailed knowledge of the nature of instability will
allow clinicians to offer more specific and appropriate surgery for each
patient. Following from this, the effect of surgical interventions on wrist
kinematics can be more thoroughly investigated, helping to guide and shape
further clinical treatment.

An important drawback of 4D CT scan is the radiation exposure, like in
fluoroscopy. However, others have shown the possibility to get accurate 4D
images of the moving wrist with a mean effective radiation dose of 0.15 mSv,
which compares favorably with a normal background radiation for any individual
(2 mSv per year).

A few studies have been published on wrist kinematics using 4D CT scans. These
studies have demonstrated its potential for the evaluation of carpal
instability, particularly in cases of inconclusive initial clinical assessment.
However, the protocols used in these studies are not uniform or standardized
and hardly any study have been performed to analyze the effect of the protocol
used in the motion analysis outcome. Moreover, only one study performed a
test-retest to determine the reproducibility of their protocol. However, this
protocol uses a specific hand guiding device, which is not publicly available
and will therefore not contribute to a standardized and uniform 4D CT wrist
image acquisition protocol28.

The study wants to investigate the intra-patient test-retest reliability to
show the robustness of the acquisition protocol to contribute to the
standardization of a 4D CT wrist image acquisition protocol, with the use of a
forearm fixation device instead of a hand guiding device.

Single institution, a diagnostic clinical study in the Radboud University
Medical Center. We will include 20 patients with chronic unilateral wrist pain
suspect ligament injury. Currently, in the standard clinical practice, a
conventional X-ray of both wrists with a posterior- anterior (with and without
clenched fist position) and lateral view are made to evaluate the injured wrist
and compare it with the asymptomatic one. The following radiographic signs are
evaluated by the radiologist: scapholunate distance, carpal angles
(scapholunate, radiolunate, radioscaphoid, and lunocapitate), presence of
dorsal or ventral intercalated segment instability (DISI or VISI resp.) and a
scaphoid ring sign. Further investigation consists of a MRI, CT scan and
arthroscopy.

For this study, we will test the reproducibility of the scanning protocol.
Therefore, the patients will do the dynamic part of the protocol twice.
Participants first undergo a static bilateral 3D CT scan in neutral wrist
position for usage of reference. The field of view of this scan includes all
carpal bones and the two forearm bones. The scan technique is a Toshiba
delivered application and CE certified.

Subsequently, the forearms are placed in a supporting frame which minimizes
lower-arm motion during 4DCT image acquisition. Prior to image acquisition,
participants will undergo a training session on how to move their wrist
according to the imaging protocol.

A bilateral dynamic 4DCT scan is made while actively moving the wrists
according to a protocoled cycle of movements. To reduce radiation exposure, the
z-axis coverage is reduced to 8 cm in the 4DCT scan. Videos of the wrist
movements are shown to the participant during image acquisition, which will
help the participants to perform the movements at a constant pace. This will
provide images with highest quality and will provide the source data
considering carpal bone movements and function of ligaments during active wrist
motions. Following the initial dynamic CT scan, patients will walk one lap
around the room before repositioning their arms in the supporting frames. The
second bilateral 4DCT scan is performed with the same movements as during the
first dynamic CT scan.
The total effective dose of this study is 0.07 mSv. This burden is a manageable
risk compared to the natural background radiation in the Netherlands (2mSv).

The dynamic 4D CT scan of the wrist has an estimated total effective radiation
dose of approximately
0.03 mSv. Since the patients will perform the scanning protocol twice, a total
effective dose of this study is 0.07 mSv. This will not burden a trivial risk
compared to the natural background radiation in the Netherlands (2mSv).
Therefore, there is no additional risk associated with participation.
The goal of the study project is to investigate the imaging acquisition
protocol and contribute to a standard or uniform scanning protocol for wrist
injuries. Therefore, the comparison of the left and right wrist will provide an
understanding on the validity of comparing the injured wrist with the uninjured
wrists in clinical practice. The test-retest reliability will help to provide
normal ranges which will help to correctly diagnose patients and help to
prevent false-positive or false-negative indication for surgical treatment. So,
by investigating the scanning protocol, we contribute to the optimization and
standardization of the image acquisition and the results of this study (and the
studies performed before) will benefit future patients with wrist injuries, as
it could replace the costly and invasive procedure of arthroscopy with a more
accessible and less intrusive alternative.