| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Diagnostic Radiology, Mayo Clinic Scottsdale, 13400 E Shea
Blvd., Scottsdale, AZ 85259.
2 Department of Orthopedic Surgery, Foot and Ankle Division, Mayo Clinic
Scottsdale, Scottsdale, AZ 85259.
3 East Valley Foot and Ankle, Ste. 114, 215 S Power Rd., Mesa, AZ 85206.
Received April 16, 2003;
accepted after revision June 2, 2003.
Address correspondence to P. T. Liu
(liu.patrick{at}mayo.edu).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Three musculoskeletal radiologists graded the visibility of these three signs on standing lateral radiographs of 32 feet that had been proven to have coalitions of the middle facets of the subtalar joint and 62 feet that had normal subtalar joints. The reviewers were unaware of the presence or absence of subtalar coalitions in these feet.
RESULTS. The absent middle facet sign had a sensitivity, specificity, and accuracy of 75%, 98%, and 90%, respectively, for the diagnosis of subtalar joint coalition in this patient population, whereas these values were 56%, 100%, and 85% for the C sign and 53%, 90%, and 78% for the talar beak sign, respectively. A finding of either a positive absent middle facet sign or a positive C sign resulted in a sensitivity, specificity, and accuracy of 84%, 98%, and 94%, respectively.
CONCLUSION. In this study population, the absent middle facet sign was more sensitive than and nearly as specific as the talar beak sign and C sign for diagnosing subtalar coalition on standing lateral foot radiographs. The highest accuracy was obtained when a finding of either a completely absent middle facet or a complete C sign was considered as a positive indicator of a subtalar coalition.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The subtalar joint is formed by anterior, middle, and posterior facet joints of the talus and calcaneus (Fig. 1A, 1B), although the anterior and middle facets are often continuous and difficult to distinguish. Of these joints, the middle subtalar joint is the most commonly involved segment in talocalcaneal coalition [4].
|
|
The planes of the middle and posterior calcaneal facets that form the subtalar joint lie parallel to one another, inclined approximately 45° anteriorly relative to the long axis of the calcaneus [5]. Relative to the sole of the foot, these facets are angled down anteriorly between 25° and 40° in 90% of cases [6]. A slight overlap of the middle and posterior subtalar joints is also present on lateral radiographs. The lateral process of the talus provides a clearly visible marker on the lateral radiograph for the anterior limit of the posterior facet (Fig. 2A, 2B). When a lateral foot radiograph is obtained with the patient standing with the central ray directed horizontally, the middle and posterior subtalar joint spaces will appear radiolucent if their mediolateral axes are maintained in the normal horizontal orientation (Fig. 3).
|
|
|
The appearance of subtalar coalition on unenhanced radiographs varies and can be subtle. Because standard radiographic views are thought to not reliably show subtalar coalitions, the Harris-Beath view and lateral tomograms have been advocated for better direct visualization of these coalitions [5–7]. However, the referring physician or radiologist must first suspect the presence of a coalition before these views will be requested. More recently, CT and MRI have been used for definitive imaging in patients suspected of having subtalar coalitions [8–13].
Harris [6] noted that "absence of a clear picture of the joint between the sustentaculum tali and the neck of the talus" occurs in feet with a subtalar joint coalition. We have also observed that the middle facets of the subtalar joint are clearly visible as thin sclerotic lines surrounding a radiolucent joint space on standing lateral radiographs of normal feet but are obscured in feet with subtalar joint or hindfoot alignment abnormalities ("absent middle facet" sign). This radiographic finding has not been previously evaluated, to the best of our knowledge. We therefore designed this study to determine the performance of the absent middle facet sign on standing lateral foot radiographs for the diagnosis of subtalar coalition.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We performed a search of patient records dating back to 1981 at our medical center and of radiology records for diagnoses of subtalar or talocalcaneal coalition. This search identified a study group of 21 patients with 32 feet that had coalitions of the subtalar joint middle facet confirmed by surgery (25 feet), CT (six feet), or MRI (one foot). The patients ranged in age from 11 to 67 years (mean, 26.6 years; median, 20 years).
To select the control group, one of two orthopedic foot and ankle surgeons or a podiatrist evaluated the subtalar joint motion of 100 consecutive patients seen in an outpatient foot and ankle clinic. In all patients, subtalar joint motion was assessed and graded as good, moderate, or poor. Inclusion criteria for the control cohort were a lack of pain from the subtalar region, good or moderate subtalar joint motion, the absence of any prior hindfoot or ankle surgery, and the availability of standing lateral radiographs. Sixty-two feet fulfilled these criteria and comprised the control group. The patients in the control group ranged in age from 15 to 83 years (mean, 60 years; median, 64 years).
The lateral radiographs were obtained with the patients standing on an elevated platform and the X-ray beam parallel to the floor and centered on the base of the first metatarsal. The tube-to-film distance was 10 ft (3.05 m).
Image Evaluation
The standing lateral radiographs of the 32 feet in the study group and the
62 feet in the control group were numerically coded and details about patient
demographics were obscured. These radiographs were then intermixed randomly
for review by a panel of two fellowship-trained musculoskeletal radiologists
and one musculoskeletal radiology fellow working in consensus. These three
reviewers were unaware of patient information.
Images were subjectively scored for the presence of the absent middle facet sign using a scale of 1–3 (1 = good visualization of the radiolucent middle facet joint space with clearly visible subchondral bone cortex of the talar and calcaneal facets, 2 = partial visualization of the radiolucent joint space and subchondral cortex, 3 = no visualization of radiolucent joint space or subchondral cortex). The reviewers were instructed to look for the middle facet joint space at a location anterior and parallel to the joint space of the posterior subtalar joint, the anterior extent of which is marked by the lateral process of the talus.
The images were also simultaneously scored with a 3-point scale for the presence of the "talar beak" sign (1 = normal head of talus, 2 = talar head vertical osteophyte at talonavicular joint, 3 = anteriorly directed talar head osteophyte) and the C sign (1 = normal talus, 2 = partial C-shaped radiopaque arc from the talar dome to the sustentaculum tali, 3 = complete C-shaped radiopaque arc from the talar dome to the sustentaculum tali).
In a preliminary training session, the reviewers were shown a series of sample cases that exhibited the different grades of visibility of the three signs.
Sensitivity, specificity, accuracy, and 95% confidence intervals were calculated from the tabulated data using scores of 3 as positive and 1 or 2 as negative and also using scores of 2 and 3 as positive and 1 as negative.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
When scores of 2 or 3 (incomplete or complete) for the C sign were considered positive findings of coalition, the C sign had a sensitivity, specificity, and accuracy of 84%, 94%, and 90%, respectively. When only a score of 3 (complete C sign) was considered positive, these values were 56%, 100%, and 85%, respectively.
When talar beak sign scores of 2 or 3 (superior or anteriorly directed excrescence) were considered positive findings of coalition, this sign had a sensitivity, specificity, and accuracy of 53%, 90%, and 78%, respectively. When only a score of 3 was considered positive, these values were 19%, 100%, and 72%, respectively.
For 24 of the 32 feet with subtalar joint coalition, reviewers assigned a score of 3 for the absent middle facet sign. Examination of the data shows that three of the eight coalitions not detected with a score of 3 for the absent middle facet sign had a score of 3 for the C sign. Therefore, when we consider a composite criterion—a score of 3 for either the absent middle facet sign or the C sign—as positive for coalition, we found the highest accuracy (94%) of all the measured criteria.
Figures 4A and 5A show lateral radiographs of two different patients that received true-positive scores for the absent middle facet, talar beak, and C signs. Figures 4B and 5B show the corresponding CT scans and MRIs of these patients that confirm the coalitions of the middle facet joints. Figures 6 and 7 are lateral radiographs from patients with surgically proven middle subtalar joint coalitions; these radiographs had false-negative scores for the talar beak and C signs, and these cases would not have been diagnosed correctly if only these signs were used. However, in both cases, the absent middle facet sign was positive.
|
|
|
|
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Some physicians think that the term "talar beak" should be reserved for describing an enthesophyte at the ankle joint capsule insertion on the dorsum of the talar neck and should not be used to describe the articular spurring seen with subtalar coalitions [15]. In our review of the English-language literature on subtalar coalitions, however, we found that most publications use the term "talar beak" to refer to the articular margin spurring seen in subtalar coalitions [5, 6, 14, 16, 17].
The C sign refers to the finding of a continuous radiopaque C-shaped arch on a lateral radiograph, created by the combined shadows of the talar dome and the fused posterior and middle facets of the subtalar joint (Fig. 4A). This sign can also be seen with fusion of the subtalar joint middle facets combined with an adjoining prominent bony ridge posterior to the sustentaculum tali. Initially, the C sign was found to have very high sensitivity (87–98%) and specificity (93–98%) [14, 17]. However, a later study found the C sign to be only 40% sensitive and 50% specific for subtalar coalition. The authors of the latter study believe that the C sign was actually more sensitive and specific (50% and 100%, respectively) for flatfoot deformity than for subtalar coalition [16].
Although the talar beak and C signs are indirect signs of subtalar coalition, the finding of osseous density bridging the subtalar middle facets is a direct sign of coalition that can be seen on a standing lateral radiograph. This sign, which we have termed the "absent middle facet" sign, has been previously discussed [5, 6] but has never, to our knowledge, been scientifically evaluated as a sign of subtalar coalition.
We have shown that in this patient population the absent middle facet sign has higher sensitivity than and a specificity nearly equal to the C sign and the talar beak sign for the diagnosis of subtalar coalition (Fig. 8). By classifying the cases that received grade 3 (complete) scores for either the absent middle facet or C sign as positive for coalition, the highest accuracy of all the criteria tested was achieved. Our data indicate that the absent middle facet and C signs are complementary for the detection of subtalar coalition. Three of the eight coalitions found to have a false-negative score for the absent middle facet sign (score of 1 or 2) were correctly diagnosed by a true-positive score for the C sign (score of 3). Eight of the 14 coalitions that were found to have a false-negative score for the C sign (score of 1 or 2) received a true-positive score for the absent middle facet sign (score of 3). The positive scores for the absent middle facet and C signs may be related to different factors of the coalitions, such as the degree of involvement of the posterior facet or the amount of hypertrophy of the sustentaculum.
|
If we were instead to consider findings of both partial and complete absent middle facet signs (scores of 2 and 3) as positive for subtalar coalitions, then the sensitivity of the sign would increase to 94%, but specificity would decrease to 61%. This loss of specificity may be because of a high number of false-positive findings caused by the overlap of the middle facet by the anterior process of the calcaneus or by abnormal angulation of the middle facet from the hindfoot valgus. Positioning the foot in abduction or inversion has been shown to obscure the middle facet joint space as a result of overlap by the lateral process of the talus or the sustentaculum tali [18].
The absent middle facet sign will probably not be a reliable indicator of subtalar joint coalition when evaluating nonstanding foot radiographs. Although we have not quantified the frequency of false-positive absent middle facet signs on nonstanding foot radiographs, anecdotally we have seen numerous false-positive absent middle facet signs on nonstanding radiographs obtained in our emergency department. Figure 9A is one such nonstanding lateral radiograph that shows an absent middle facet sign in a patient who suffered minor foot trauma. A standing lateral radiograph of the same foot (Fig. 9B) obtained 3 months later shows a clearly visible middle facet. The corresponding CT scan (Fig. 9C) that was obtained the same day as Figure 9B shows a normal middle facet.
|
|
|
When C sign scores of 2 and 3 were considered positive, our results for the performance of the C sign correlate more closely with those obtained by Lateur et al. [17] and Sakellariou et al. [14] than with those obtained by Brown et al. [16]. Of possible significance is the observation that the study by Lateur et al. was the only one of these three studies that used only weight-bearing lateral radiographs. Whether the use of non–weight-bearing radiographs by Brown et al. was responsible for the low sensitivity and specificity for the C sign in that study is not known.
Unlike those prior studies, our study graded the C sign findings as "partial" or "complete" (scores of 2 and 3, respectively). We believe that partial sclerosis of the sustentaculum tali might be seen in degenerative arthritis of the subtalar joint or pes planovalgus deformities. This sclerosis could have potentially produced partial C signs and false-positive diagnoses of coalition if they had been considered as positive signs.
Our results for the performance of the talar beak sign, when both scores of 2 and 3 were considered positive findings of subtalar coalition (sensitivity, specificity, and accuracy = 53%, 90%, and 78%, respectively), are nearly identical to those obtained by Sakellariou et al. [14] (Table 2). Our results echo those researchers' observation that the talar beak sign has low sensitivity but high specificity for subtalar coalition. We could find no other study that evaluated the performance of this sign for the diagnosis of coalition. If we limit the diagnosis of subtalar coalition to only those cases with talar beak scores of 3 (anteriorly directed talar head osteophyte), the sensitivity of the sign drops to an undesirably low level of 19%. Researchers who have previously discussed the talar beak sign did not specify the direction of this osteophyte in their studies [5, 14, 16].
|
Although we could find no prior studies that evaluated the visibility of the middle facet of the subtalar joint, Sakellariou et al. [14] examined the radiographic finding of "subtalar joint degenerative joint disease," which they defined as only "narrowing of the posterior facet of the subtalar joint." This sign had a sensitivity and specificity of 63% and 99%, respectively. However, that study did not assess the visibility of the middle facets of the subtalar joint, which are more commonly involved in subtalar joint coalitions.
We recognize that our study has several limitations. The retrospective design of our study, using a patient population with an unnaturally high incidence of subtalar coalitions, might not permit true measurement of the sensitivity and specificity of these radiographic signs for subtalar coalition. A prospective study would be more accurate, but would be challenging because of the low natural incidence of subtalar coalition. In such a study, correlation by either cross-sectional imaging or physical examination in a large number of patients would be necessary to realize a significant sample of patients with actual subtalar joint coalitions.
In our study, observer bias was also present because the observing radiologists were aware that many patients had subtalar coalitions. The lower incidence of subtalar coalitions in the general population may result in a lower specificity for the absent middle facet sign if more false-positive interpretations were made.
The average age of our control subjects exceeded the average age of the patients with subtalar joint coalition, and this discrepancy may have introduced an additional source of error. This discrepancy is mainly because of the demographics of the patient population at our foot and ankle clinic and to the relatively early presentation of symptoms in most patients with subtalar joint coalition. The fact that four of our patients with subtalar joint coalition were older than 50 years, however, indicates that symptoms of coalition can be fairly mild in some patients and these patients may not present until they are more advanced in age.
In conclusion, we have shown that a normal middle subtalar joint is visible on standing lateral radiographs of most normal feet, and the presence of either an absent middle facet sign or C sign should raise the radiologist's suspicion of a subtalar coalition. Because the clinical presentation of subtalar joint coalition can be nonspecific—with symptoms ranging from chronic hindfoot pain to recurrent ankle sprains—the radiologist should be aware of these unenhanced radiography findings of subtalar coalition in order to alert the referring clinician to the possible diagnosis.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  D. C. Moe, J. J. Choi, and K. W. Davis Posterior Subtalar Facet Coalition with Calcaneal Stress Fracture Am. J. Roentgenol., January 1, 2006; 186(1): 259 - 264. [Full Text] [PDF]
|
|
|
|
 |
|
 G. E. Rassi, E. C. Riddle, and S. J. Kumar Arthrofibrosis Involving the Middle Facet of the Talocalcaneal Joint in Children and Adolescents J. Bone Joint Surg. Am., October 1, 2005; 87(10): 2227 - 2231. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
