| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Radiology, Mayo Clinic Arizona, 13400 E Shea Blvd., Scottsdale,
AZ 85259.
2 Present address: East Valley Diagnostic Imaging, Mesa, AZ.
3 Department of Orthopedics, Mayo Clinic Arizona, Scottsdale, AZ.
Received May 16, 2008;
accepted after revision July 29, 2008.
Address correspondence to P. T. Liu.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. We performed a retrospective review of 42 consecutive MDCT examinations from 29 patients. All patients had previously undergone arthrodesis of their ankle or subtalar joints and had persistent or recurrent hindfoot or ankle pain. Two musculoskeletal radiologists examined in consensus sagittal 2-mm-thick reformatted slices, measuring on each image the length of the joint surface and the length of the fused portion of the joint space. The sum of the lengths of the fused segments on all slices was then divided by the sum of the lengths of the joint surfaces to calculate the fusion ratio. For the standard of reference, the medical records were reviewed and operative reports, diagnostic injections, and physical examinations were used to classify the joints as stable or unstable.
RESULTS. Twelve clinically unstable joints had fusion ratios of 0–32.8%, whereas 30 clinically stable joints had fusion ratios of 33.2–100%. Using receiver operating characteristic analysis, we selected the cutoff level that maximized Youden's index (the sum of sensitivity and specificity). Using a 33% fusion ratio as the lower limit cutoff for joint stability, the sensitivity was 100%; specificity, 100%; and accuracy, 100%.
CONCLUSION. After arthrodesis of the ankle or subtalar joint, MDCT scans can be used to determine whether that joint is likely to be stable if > 33% of the joint has visible bone fusion on sagittal MDCT images.
Keywords: arthrodesis • fusion • MDCT • nonunion
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Nonunion or incomplete union is suspected clinically when pain or tenderness persists at the arthrodesis site beyond the expected time for normal healing, usually 6 months. In patients with such a history, it is important to have a reliable imaging test that can accurately differentiate between stable and unstable arthrodeses. Finding an unstable incomplete union leads to additional treatment with prolonged immobilization, application of bone growth stimulators, or revision surgery. Alternatively, if the joint is found to be fused and stable, the clinician would search for an alternative source of the pain. MDCT is often ordered at our institution to evaluate these patients because of its cross-sectional capabilities and excellent depiction of bone detail. Unfortunately, it has not yet been determined what degree of bone fusion visible across a postoperative joint on MDCT correlates with a stable fused joint.
In the literature, only a few publications have described the use of MDCT to evaluate fracture or healing after arthrodesis [4–9]. In the largest of these studies, Ohashi et al. [9] reviewed 114 MDCT examinations performed for suspected orthopedic hardware complications. Their sensitivity for complications using MDCT was 74% and their specificity, 95%. Six cases of fracture nonunion and four cases of nonfusion after arthrodesis were successfully detected on CT and confirmed at surgery, but the authors did not address the imaging criteria they used to make these diagnoses [9]. In the other articles, the patient samples were small and MDCT was considered the gold standard for itself.
To our knowledge, no study has correlated the percentage of bone fusion on CT after arthrodesis with a degree of clinical stability. Jones et al. [3] did develop a CT grading system for evaluating the effect of ultrasound stimulation on bone fusion, but they had no standard of reference. The authors arbitrarily defined a CT finding of nonunion as 0–33% fusion on CT, partial union as 34–66% fusion, and complete union as 67–100% fusion [3].
The purpose of this study was to correlate the degree of bone fusion on CT with clinical stability in patients who had undergone ankle or subtalar joint arthrodeses but who had recurrent hindfoot or ankle pain.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Patient Sample
A retrospective query of our institutional radiology report database was
performed to identify CT examinations of subtalar or ankle arthrodeses during
the period from January 1, 2000, to Sept ember 1, 2006. The examinations were
all per formed for persistent or recurrent pain in the ankle or hindfoot after
arthrodesis. Medical records were examined to identify patients who had
sufficient follow-up to be classified clinically as having unstable or stable
arthrodeses.
Our study sample consisted of a total of 42 scanned joints in 29 patients. Nine patients underwent attempted arthrodesis on both the subtalar and ankle joints, three patients had repeat examinations after revision surgery, and one patient had a repeat examination after a period of conservative therapy. Patient records were reviewed to deter mine the cause of arthrodesis nonunion or delayed union.
Clinical Criteria
Physical examination notes from foot and ankle orthopedic surgeons were
reviewed retrospectively. Patients were classified as having clinically stable
unions at their sites of attempted arthrodesis if they were found to lack
motion at the joint on physical examination and one of the following occurred:
solid joint fusion was observed during repeat surgery; fluoroscopically guided
injection of 3–5 mL of 0.5% bupivacaine of the operated joint did not
relieve pain, indicating that the source of pain was elsewhere;
fluoroscopically guided injection of 3–5 mL of 0.5% bupivacaine into
another non communicating joint relieved pain; surgery at another joint or
bone relieved pain; pain was neurogenic and improved with gabapentin; pain on
physical examination was isolated to a joint other than the operated
joint.
The criteria for clinically unstable union included any one of the following: pain and motion at the operated joint on physical examination, surgical revision confirmed motion at joint, fluoroscopically guided injection of 3–5 mL of 0.5% bupivacaine into the operated joint relieved the patients' pain, and surgical revision of arthrodesis relieved pain.
CT Technique
CT studies were performed on 4-, 16-, or 64-MDCT scanners (Sensation,
Siemens Medical Solutions). The imaging parameters included detector
collimation of 0.5–0.6 mm, 200–400 mAs, 120–140 kVp, bone
algorithm, and field of view of 15–20 cm. Axial source images for
reformatting were created with 0.5- to 0.6-mm slice thickness and 0.3- to
0.4-mm overlap. Sagittal reformatted images were created with 2-mm slice
thickness.
Image Review
Soft-copy sagittal images were retrospectively reviewed on a PACS
workstation by two musculoskeletal radiologists working in consensus who were
blinded to patient data. The readers measured the degree of bone fusion across
the operated joints using a grading system developed by Jones et al.
[3]. Using 2-mm-thick sagittal
multiplanar reformatted images through the operated joint, the length of the
joint surface and the length of the successfully fused portion of that joint
were measured on each slice that passed through the joint
(Fig. 1). Fusion of a joint
segment was defined as trabeculation or calcific density crossing the former
joint space.
|
The lengths of the fused segments and the lengths of each joint on each slice were recorded, and the fusion ratio score was calculated using the formula: fusion ratio = 100 x (sum of lengths of fused segments on all slices / sum of lengths of joint surface). When more than one operated joint was assessed in a patient, the images from each joint were separated in random order and scored separately.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
The time interval from arthrodesis to CT examination ranged from 4 to 716 months (mean, 77 months; median, 34 months). Five joints had loosening or fracture of fixation screws and all were unstable at surgery. No patients with stable joints had failed hardware. There were no cases with infected joints.
Using receiver operating characteristic analysis, we selected the cutoff level that maximized Youden's index (the sum of sensitivity and specificity). A cutoff level of 33% yielded 100% sensitivity (12/12) and 100% specificity (30/30) in this patient sample, with a positive predictive value of 100% and negative predictive value of 100%.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Coincidentally, 33% fusion was also the threshold arbitrarily chosen by Jones et al. [3] to divide nonunion from partial union in their CT grading system; complete union was defined as a joint fused over more than 67% of its length. Those authors created their grading system as a way to evaluate the success of ultrasound stimulation for arthrodesis nonunions, even though their CT grading system was never validated.
Assessment of arthrodesis fusion ratios on CT can be easily performed in patients with persistent or recurrent pain after arthrodesis of the subtalar or tibiotalar joint, or in postoperative follow-up in patients in whom the clinical findings and radiographic findings are discordant. In our study, measurement of joint lengths and fusion lengths, combined with calculation of fusion ratios, took approximately 5 minutes per case. Measurement would not be necessary in cases with obviously complete fusion or complete nonfusion.
Although radiographs are less sensitive than CT for findings of nonunion after arthrodesis, they remain a first-line imaging test because of their low cost and ease of access. The finding of persistent radiolucency crossing the entire joint space is still likely a highly specific sign of arthrodesis nonunion. Loosening, bending, or fracture of fixation screws, plates, and nails is also a specific radiographic sign for nonunion and persistent motion in a joint after surgery.
Limitations of this study include its retrospective design and limited number of patients. A larger study would be helpful to confirm our results and would also likely reveal some false-positive and false-negative results using the 33% fusion ratio cutoff. Although no false-positive and no false-negative results occurred using the 33% cutoff ratio, only a slim gap existed between the largest fusion ratio found in an unstable joint (32.8%) and the lowest fusion ratio found in a stable joint (33.2%).
The study design instilled a potential selection bias because no asymptomatic patients were included in our study. Probably at least a small number of asymptomatic patients would have a low percentage of bone fusion after attempted arthrodesis. If such patients had been included in this study and were found to have stable joints on physical examination, they would have caused some crossover of fusion ratio results between the stable and unstable cases, decreasing the accuracy of CT using the cutoff ratio and potentially altering the ideal fusion ratio cutoff level.
An additional limitation was the lack of surgical proof of all clinically stable joints. However, such surgeries were thought to be clinically unnecessary if an alternate source of pain was proven by another means, and would therefore not have been ethically appropriate.
Because the surgeons were aware of the CT results and were able to view the CT images in the symptomatic patients before surgery, their observations at surgery could have been biased by the imaging findings, potentially increasing the number of true-positive results.
Our use of fluoroscopically guided anesthetic blocks to diagnose a nearby nonoperated joint as a source of pain could also be seen as a potential limitation because we did not exclude the possibility of a delayed communication with the operated joint that was not initially visible.
In conclusion, in patients with recurrent symptoms after hindfoot or ankle arthrodesis, measurement of the percentage of joint fusion can differentiate stable unions from unstable unions when a 33% fusion ratio threshold is used. This interpretation technique may help to identify patients who could benefit from prolonged immobilization or additional surgical treatment. Further prospective study is needed to test this threshold in a larger population.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
