Bulletin of the NYU Hospital for Joint Diseases 2008;66(1):14-2114

Nalaboff KM, Schweitzer ME. MRI of tarsal coalition: frequency, distribution, and innovative signs. Bull NYU Hosp Jt Dis. 2008;66(1):14-21.

Abstract

Objective: Tarsal coalition is uncommon, but not rare. Since it may present with chronic symptoms and since prior studies have shown that magnetic resonance (MR) has a high accu-racy for diagnosis, we evaluated the prevalence, frequency, location, and types of coalition, as well as secondary signs, as seen in a large clinical MR population. Subjects and Methods: Initially a “word search” program was used to find all diagnoses of “tarsal coalition” in our RIS database as part of 27,483 ankle MRIs performed over a ten-year period. We revaluated 101 of 169 identified cases for location (calca-neonavicular vs. talocalcaneal/subtalar) and type (osseous, cartilaginous, or fibrous, utilizing previously described MR criteria) of coalition. Additional imaging findings, including marrow edema and fractures, and associated secondary signs, such as talar “beaks,” the “anteater” (hyperplastic anterior calcaneal process) sign, and the innovative “reverse anteater” (hyperplastic navicular) and “drunken waiter” (hypo/dysplastic sustentaculum tali) signs, were reported as well. As a second part to the study, in order to more ac-curately determine prevalence in a population requiring an MRI of the ankle, two observers reviewed 667 consecutive ankle MRIs performed on 640 patients. Patients less than 12 years of age (N = 20, 3.0%) or with nondiagnostic exams (N = 73, 10.9%) were excluded. Note was made: 1. if a coalition was present, 2. if it was definite or possible, 3. of the location, 4. of the type, 5. of secondary signs, and 6. the

frequency of bilaterality. Results: In the initial retrospective study, coalitions were identified in 0.6% of exams performed, including 78 calcaneonavicular coalitions. Seven fractures were seen, all in the fibrous calcaneonavicular subgroup. We found only eight cases (10.3%) of the “anteater” sign, while identifying 19 cases (24.4%) of the “reverse” anteater sign in the calcaneonavicular group. Talar “beaks” were seen in 15 cases (19.2%) of the calcaneonaviuclar group and nine (39.1%) of the subtalar group. Sixteen (69.6%) of the subtalar cases demonstrated the “drunken waiter” sign, a dysplastic sustentaculum. We saw a dramatic difference in frequency of tarsal coalitions when we applied a prospec-tive approach; 11.5% of all patients (66/574) had a coali-tion, while 12.2% of studies (70/574) demonstrated tarsal coalitions: 25.7% (18/70) were subtalar, 71.4% (50/70) were calcaneonavicular, and 2.9% (2/70) had both. The calcaneonavicular coalitions were roughly evenly divided between the cartilaginous (56%) and fibrous (44%) subtypes, without any osseous coalitions being identified. The subtalar coalitions were also nearly evenly divided between osse-ous/partially osseous (33.3%), cartilaginous (27.8%), and fibrous (38.9%). Talar “beaks” were seen in 25 cases (50%) of the calcaneonavicular group and five (27.%) of the sub-talar group. We identified six cases (12%) of the “anteater” and nine cases (18%) of the “reverse anteater” signs in our calcaneonavicular group. Conclusion: Tarsal coalitions ap-pear to be more common than previously described. This is likely to be because fibrous and cartilaginous coalitions are infrequently recognized by plain radiography. The frequency of tarsal coalitions may be as high as 11%. We also found a disproportionately higher relative frequency of calcaneo-navicular coalitions, either because subtalar coalitions are more subtle on MRI or because calcaneonavicular coalitions can be overly diagnosed. Calcaneonavicular coalitions tend to be overwhelmingly nonosseous, while there is a more even distribution for subtalar coalitions. Newly described signs,

MRI of Tarsal CoalitionFrequency, Distribution, and Innovative Signs

Kenneth M. Nalaboff, M.D., and Mark E. Schweitzer, M.D.

Kenneth M. Nalaboff, M.D., was from the Department of Radiol-ogy, Thomas Jefferson University Hospital, Philadelphia, Pennsyl-vania, at submission. Mark E. Schweitzer, M.D., is Professor of Radiology and Orthopaedic Surgery, New York University School of Medicine, and Chairman, Division of Radiology, NYU Hospital for Joint Diseases, NYU Medical Center, New York, New York.Correspondence: Mark.Schweitzer, M.D., Department of Radiol-ogy, NYU Hospital for Joint Diseases, 301 East 17th Street, New York, New York 10003; mark.schweitzer@nyumc.org.

15Bulletin of the NYU Hospital for Joint Diseases 2008;66(1):14-21

such as the “reverse anteater” and “drunken waiter,” are commonly seen in calcaneonavicular and subtalar coali-tions, respectively.

Tarsal coalition is an abnormal fibrous, cartilaginous, or osseous fusion of one or more of the tarsal ossifi-cation centers. It is an uncommon condition, gener-

ally believed to affect less than 1% of the population. The condition was described as early as 250 years ago.1-4 Only during the early part of the last century were clinical find-ings noted, including tarsal or hindfoot pain and stiffness, decreased subtalar motion, valgus deformity, and non-heal-ing ankle sprains.5-7

The goal of this study was to determine the prevalence and relative frequency, by location and morphologic sub-type, of tarsal coalitions utilizing magnetic resonance (MR) criteria. We also evaluated previously described, as well as innovative, secondary MR signs. Our hypothesis was that, since MR is more sensitive than plain radiography or CT for the detection and description of tarsal coalition, MR evaluation would yield the true prevalence and distribution of location and morphologic type in a population requiring an ankle MRI and that MR would also better evaluate for the presence of secondary signs.

Subjects and MethodsA two-part study was performed. In the first part, we sought to determine the frequency in which two experienced mus-culoskeletal radiologists described tarsal coalitions. We initially searched reports on all ankle MRI examinations per-formed at our institution over a ten-year period that contained

the word “coalition” in the comments or impression. Out of the 27,483 cases, 169 cases of tarsal coalition had been described. Of these cases, we were able to retrospectively, systematically evaluate 101 of them, utilizing previously described MR criteria for coalition type and location. We also reviewed these cases for the frequency of the following secondary signs: blunted or flat anterior calcaneal process,

Table 1 Ankle MRI Pulse Sequences

Sagittal SE T1 (400-800/min.) STIR (> 2000/20-40/150)Axial FSE PD (> 2000/30-40) FSE T2 (2000-6000/60-70) with fat saturation with fat saturationCoronal FSE T2 (2000-6000/60-70)

Table 2 MR Signs of Tarsal Coalition

Osseous Cartilaginous Fibrous Calcaneonavicular Talocalcaneal

Primary MR criteria Continuity Loss of joint Areas of Loss of marrow fat Loss of marrow of marrow space, with intermediate between calcaneus fat between space between signal intensity or low signal and navicular talus and the suspect of cartilage or intensity calcaneus bones joint fluid bridging the Irregularity of between the adjacent bones the talocalcaneal suspect bones articulation

Secondary MR criteria Blunted anterior “Drunken calcaneal process, waiter” sign, “Anteater” sign, talar “beak,” “reverse anteater” marrow edema, sign, talar “beak,” fracture marrow edema, fracture

Figure 1 Sagittal T1-weighted MR image. Osseous subtalar coali-tion with a continuous marrow space (arrow).

Bulletin of the NYU Hospital for Joint Diseases 2008;66(1):14-2116

“anteater” sign, “reverse anteater” sign, talar beak, “drunken waiter” sign, marrow edema, and fractures. For the second part of study, 667 consecutive ankle MRIs, performed over a 15-week period, were reviewed by the same two musculoskeletal radiologists for the presence of tarsal coalition. Our ankle MRI protocol includes sagittal Spin Echo (SE) T1-weighted and STIR sequences, axial Fast Spin Echo (FSE) Proton Density (PD) and frequency selective fat-satu-rated FSE T2-weighted sequences, and a coronal fat-saturated FSE T2-weighted sequence (Table 1). We utilized previously published criteria for the diagnosis of tarsal coalition.8,9 The criterion for osseous coalition was demonstration of a continuous marrow space (Figs. 1 and 2). The criteria for fibrous and cartilaginous coalitions were the loss of the fat plane at the affected articulation, geographic proximity, and a vertically oriented anterior process. If the intervening tissue was with low signal bridging on both T1 and T2 or STIR-weighted sequences, it was considered a fibrous coalition (Fig. 3). If it was gray on the T2 or STIR-weighted sequence, with or without abnormal fluid, it was considered a cartilaginous coalition (Fig. 4).8

Secondary signs evaluated for calcaneonavicular coalitions were similar to those utilized in plain radiographic evaluation: blunted or flat anterior calcaneal process10 (Figs. 3 and 4), “anteater” sign (hyperplastic anterior calcaneal process with anterior elongation)10-12; “reverse anteater” sign (hyperplastic

navicular with posterior elongation) (Fig. 5); or a talar “beak,” proximal to the talonavicular joint (Fig. 6).13-15

The secondary signs for talocalcaneal coalitions were: hypo/dysplastic sustentaculum tali (“drunken waiter”

Figure 2 Axial proton density-weighted MR image. Osseous cal-caneonavicular coalition, with a continuous marrow space (dashed circle). Incidental note is made for peroneal tenosynovitis.

Figure 3 Sagittal STIR MR image. Low signal fibrous calcaneo-navicular coalition on STIR image (dashed circle). The anterior calcaneal process is blunted, a secondary sign of calcaneonavicular coalition. There is subchondral bone marrow edema in the calca-neus, navicular and cuboid, likely related to biomechanical stress associated with the coalition.

Figure 4 Sagittal STIR MR image. Fluid-like gray signal is seen at the calcaneonavicular coalition on this STIR image, consistent with a cartilaginous coalition (dashed circle). Notice, again, the blunted anterior calcaneal process, a secondary sign of coalition.

17Bulletin of the NYU Hospital for Joint Diseases 2008;66(1):14-21

sign)9 (Figs. 7 and 8) and talar “beaks.”13,14 The former sign was a subjective evaluation on coronal images, based on the cranial height of the sustenaculum as well as how well formed the inner curve appeared. The dysplastic sustenaculum tali occurring with a subtalar coalition

has been likened to a waiter having trouble carrying his tray (either an upturned or downturned hand and plate) while intoxicated, where the body of the calcaneus is considered the waiter and the sustentaculum tali is the tray. We also evaluated areas of bone marrow edema (Figs. 3 and 7) and fractures.

Table 3 Study #1, Distribution

Partially Osseous Osseous Cartilaginous Fibrous Total

Calcaneonavicular 0 2 33 43 78Talocalcaneal 6 3 7 7 23Total 6 5 40 50 101

Table 4 Study #1, Secondary Signs

“Reverse “Drunken Talar “Anteater” Anteater” Waiter” “beak” sign sign sign Fracture

Calcaneonavicular Osseous 0 0 0 0 0 Partially Osseous 0 0 0 0 0 Cartilaginous 6 2 4 0 0 Fibrous 9 6 15 0 7Total 15 8 19 0 7

Talocalcaneal Osseous 3 0 0 0 0 Partially Osseous 0 0 0 0 0 Cartilaginous 6 0 0 13 0 Fibrous 0 0 0 3 0Total 9 0 0 16 0

Figure 5 Sagittal T1-weighted MR image. Posterior elongation of the navicular, in the setting of a calcaneonavicular coalition, is termed the “reverse anteater” sign (arrow). Also notice the low signal on this T1-weighted image, which could represent either a fibrous or cartilaginous coalition.

Figure 6 Sagittal T1-weighted MR image. Large osseous excres-cence emanating from the dorsal aspect of the talus, just proximal to the talonavicular joint. This is a traction osteophyte, termed a talar “beak” (arrow).

Bulletin of the NYU Hospital for Joint Diseases 2008;66(1):14-2118

ResultsIn our initial retrospective study, a prevalence of 0.6% (169 coalitions out of 27,483 cases) was seen. Of the 101 cases we re-evaluated, 77.2% (78/101) were calcaneonavicular

in location. Of those, 55.1% (43/78) were fibrous, 42.3% (33/78) were cartilaginous, and 2.6% (2/78) were osseous. Regarding location, 22.8% (23/101) of cases were subtalar, with 30.4% (7/23) fibrous, 30.4% (7/23) cartilaginous, and 39.1% (9/23) osseous or partially osseous. Coalitions in other locations were not identified (Table 3). Only 10.3% (8/78) of the calcaneonavicular coalitions demonstrated the well-known “anteater” sign, while 24.4% (19/78) demonstrated the previously undescribed “reverse anteater” sign. Talar “beaks,” which were only called in the absence of osteoarthritis, in order to prevent confusion with

Figure 7 Coronal FSE T2-weighted MR image. “Drunken waiter” sign (arrows). The dysplastic sustenaculum tali occurring with a subtalar coalition has been likened to a waiter having trouble carry-ing his tray (either an upturned or downturned hand and plate) while intoxicated. Notice the bone marrow edema and cysts, indicative of biomechanical stress.

Table 6 Study #2, Secondary Signs

“Reverse “Drunken Bone Talar “Anteater” Anteater” Waiter” Marrow “beak” sign sign sign Fracture Edema

Calcaneonavicular 25 6 9 0 0 18“Coalition-like” 7 0 0 0 0 2Talocalcaneal 5 0 0 18 0 5“Coalition-like” 0 0 0 0 0 0

Table 5 Study #2, Distribution

Osseous/Partially “Coalition-like” Osseous Cartilaginous Fibrous (N = 40)

Calcaneonavicular 0 28 22 38 (N = 50)Talocalcaneal 6 5 7 2 (N = 18)Calcaneonavicular 2 (osseous 0 2 (fibrous 0and Talocalcaneal talocalcaneal calcaneonavicular (N = 2) portion) portion)

Figure 8 Coronal FSE T2-weighted MR image. Another ex-ample of the “drunken waiter” sign, with the “tray” pointing up (arrow).

19Bulletin of the NYU Hospital for Joint Diseases 2008;66(1):14-21

osteophytes, were seen in 19.2% (15/78) of cases. Seven fractures were seen, all in the fibrous subgroup. There were no fractures seen with the group of subtalar coalitions; 39.1% (9/23) demonstrated a talar “beak” and 69.6% (16/23) demonstrated the “drunken waiter” sign (Table 4). In the second prospective study, a higher prevalence of 11.5% (66/574) of patients and 12.2% (70/574) of studies (four patients had bilateral coalitions) were seen. Calcaneo-navicular coalitions accounted for 71.4% (50/70) of cases, and the subtalar location accounted for 25.7% (18/70). Two patients (2.9%) had both types of coalitions. In both instances, the subtalar component was osseous in nature and the calcaneonavicular was fibrous. Coalitions in other locations were not seen (Table 4). Of the calcaneonavicular coalitions identified, 56.0% (28/50) were cartilaginous and 44.0% (22/50) were fibrous. There were no osseous calcaneonavicular coalitions seen. One-half (25/50) of the cases demonstrated a talar “beak,” 12.0% (6/50) demonstrated the “anteater” sign, and 18.0% (9/50) demonstrated the “reverse anteater” sign. Within the group of subtalar coalitions, 33.3% (6/18) were osseous or partially osseous, 27.8% (5/18) were cartilaginous, and 38.9% (7/18) were fibrous in nature. All cases demonstrated the “drunken waiter” sign, and 27.8% (5/18) had talar “beaks.” No fractures were seen and bone marrow edema was present in 36.0% (18/50) of calcaneonavicular and 27.8% (5/18) of subtalar coalitions. No discreet pattern of marrow edema, other than involvement of the bones directly involved in the coalition, was discerned. There were only five patients with bilateral studies, and at least one ankle was involved with coalition. Eighty percent (4/5) were calcaneonavicular and 20% (1/5) were subtalar in location. We also made note of “coalition-like” appearances, where there was close apposition of the bones, but they did not meet MR criteria for true tarsal coalition. Overall, there was a 7.0% (40/574) prevalence of “coalition-like” appearances, some with evidence of biomechanical stress. Talar “beaks” were seen in 17.5% (7/40) of cases, and bone marrow edema was seen in 5.0% (2/40). All but two of those cases were calcaneonavicular in location (Table 5).

DiscussionTarsal coalitions in typical locations are generally felt to be a developmental anomaly.6,9,13,16 The precise cause is unclear but is thought to be due to a failure of mesenchymal segmentation. There is possible autosomal dominance with variable penetrance.17-20 Symptoms of coalitions include peroneal spastic flatfoot or rigid flatfoot attributed to peroneal spasm or adaptive peroneal shortening in response to heel valgus.9,10,13,14,21-23 Symptoms are attributed to progressive ossification of the coalition.10,22,24,25 Calcaneonavicular coalitions ossify earlier and become symptomatic around 8 to 12 years of age.10,22,25,26 Talocalcaneal coalitions ossify later and become symptomat-ic around 12 to 16 years of age.10,22,25 Conservative treatment

options range from arch supports for minor symptoms to a short-leg walking cast immobilization in neutral or a slight varus positioning and decreased activity and anti-inflamma-tory medications. Surgical treatment includes resection with arthrodesis or arthrodesis alone. The radiographic and cross-sectional criteria for the diagnosis of the various types of tarsal coalitions have been previously described.6,8-11,13-15,21,22,26 The first formal radio-logic analysis of tarsal coalition was described by Conway and Cowell in 1968.13 Several investigators suggest that MR is more sensitive to nonosseous coalitions then CT.8,27-29 This is the first study to utilize cross-sectional techniques to describe the frequency, location, and distribution of the different subtypes of coalitions. More recent radiologic literature has focused on the value of cross-sectional imaging, compared to plain radiography and the relative merits of computed tomography (CT) and MRI.8,9,21,29,30 Calcaneonavicular and talocalcaneal (subtalar) coalitions are accepted as the two most common types, but opinion varies regarding relative distribution and histologic subtype.16,22,23,31 Calcaneocuboid and talonavicular coalitions are far less frequent. Stormont and Peterson reported that the incidences of calcaneonavicular and subtalar coalitions were 53% and 37%, respectively, with talonavicular and calcaneocuboid coalitions occurring far less commonly.16 More recent work by Kulik and Clanton asserted that calca-neonavicular and subtalar coalitions occurred with roughly the same frequency.22 Studies by Harris as well as Mosier and Asher even claim that subtalar coalitions are the most common.23,32 To our knowledge, the only study that evaluated the relative frequencies of the various histologic subtypes, as well as the incidence of secondary signs, utilizing modern cross-sectional techniques was a cadaveric study using helical CT evaluation by Solomon and colleagues.33 They reported an approximately 3:1 ratio of calcaneonavicular to subtalar coalitions (only nonosseous coalitions were identi-fied). We also found that calcaneonavicular coalitions are far more common than subtalar coalitions (approximately a 3:1 ratio in both the retrospective and prospective portions of the study), and that calcaneonavicular coalitions tend to be overwhelmingly nonosseous, while there is a more even distribution of histologic subtype for subtalar coalitions. This variability in histologic type explains why radiographic series may misrepresent not only the frequency but also the relative histologic distribution. Most previous investigators have noted overall prevalence to be less than 1% of the general population,6,16,19,31,34 al-though the recent cadaveric study by Solomon and coworkers demonstrated an incidence of almost 13%.33 Their finding is similar to the frequency seen in the prospective portion of our study. Another recent investigation by Lysack and Fenton, which intended to evaluate radiographic variations in calcaneonavicular morphology, incidentally noted a general prevalence of calcaneonavicular coalition to be 5.6%.35 Our

Bulletin of the NYU Hospital for Joint Diseases 2008;66(1):14-2120

initial retrospective study demonstrated an overall prevalence of less than 1%, however, which is consistent with most of the previous reports. Our study, as well as most prior ones, is a “victim” of some degree of population bias, from the perspective of who was imaged and for what indications. We did not image the general population, but a select one: only those patients with ankle pain or other symptoms requiring an MRI. We cannot be sure how many of these tarsal coali-tions were actually the cause of their symptomatology and whether subtle fibrous and cartilaginous coalitions tended to be less symptomatic than osseous coalitions. In addition, we were unable to obtain surgical proof that the findings that the MRI studies demonstrated were actually tarsal coalitions. The prior study by Solomon and associates also made conclusions similar to ours, where large numbers of non-osseous coalitions are likely symptomatic, they are often missed with spiral CT examinations.33 A more recent study by Taniguchi and colleagues indicated that relying on the classic “C” radiographic sign to evaluate for subtalar coalitions will lead to numerous false negative diagnoses of nonosseous coalitions.36 Taking those studies as well as ours into consideration, we believe that prior to the advent of MRI, many, if not most, nonosseous coalitions may have been missed, even with CT. Although not called “true” tarsal coalitions, the “coali-tion-like” appearance was seen in an additional group of cases (7.0%). We are not certain what these represent, but present them only to add emphasis to the fact that coalitions are a spectrum of disorders. It is unclear if the these “coali-tion-like” appearances correlate at all to the morphologic types 2 and 3 described by Lysack and Fenton35 on medial oblique radiographs of the ankle, although these investiga-tors claimed that those morphologic types corresponded to cartilaginous and fibrous calcaneonavicular coalitions. Our cases showed many of the secondary MR signs, as in the more overt group, although at a lesser frequency. Prior studies have reported a bilateral incidence of tarsal coalition of between 50% and 80%, with more frequent bilaterality of the calcaneonavicular subtype.13,16,19,23,37 Al-though we observed an 80% bilateral occurrence in those patients with bilateral studies and at least one side positive for a coalition, our number of cases (5) was too small to draw any conclusions. We also demonstrated that previously undescribed secondary signs of coalition are fairly common. These anatomic changes associated with tarsal coalitions are pre-sumably due to the altered biomechanics that a relatively or completely immobile articulation would place upon the midfoot and hindfoot. Although the hyperplastic anterior calcaneal process, or the “anteater” sign, is a well-known sign for calcaneonavicular coalitions seen on lateral ankle radiographs11 and sagittal MR images, we saw a far more common occurrence of the hyperplastic navicular, or “re-verse anteater” sign, on sagittal imaging. Both findings are almost certainly due to similar biomechanical stresses, but it

is uncertain why the posterior aspect of the tarsal navicular may become hypertrophied more often than the anterior calcaneal process. Nearly all the talocalcaneal/subtalar coalitions dem-onstrated a hypo/dysplastic sustentaculum tali, where the upturned or downturned sustentaculum can be likened to a “drunken waiter” unable to hold his tray at a level position (Fig. 7). This finding has only been cursorily described in coronal cross-sectional imaging by previous investigators in association with calcaneal valgus,9,38 although it does contribute to the “C” sign on lateral ankle radiographs. There has been no published formal evaluation of the frequency of talar “beaks,” a well known sign of both calca-neonavicular and subtalar coalitions that represents a traction spur attempting to compensate for abnormal motion at the talonavicular joint,13,14,23 although there have been many descriptive papers on this topic. We observed a frequency of approximately 25% in subtalar coalitions in both of our studies and a range of 20% to 50% for calcaneonavicular coalitions. This is not surprising, considering how often this sign is identified anecdotally. We saw fractures involving only patients with fibrous calcaneonavicular patients and only in the first study. This information is nonspecific.

ConclusionsTarsal coalitions appear to be much more common than previously described, likely because fibrous and cartilagi-nous coalitions are infrequently recognized by plain radi-ography. This also suggests that many coalitions may be asymptomatic. We also found a disproportionately higher relative frequency of calcaneonavicular coalitions, either because subtalar coalitions are subtler on MRI or because calcaneonavicular coalitions can be overcalled.

Disclosure StatementNone of the authors have a financial or proprietary interest in the subject matter or materials discussed, including, but not limited to, employment, consultancies, stock ownership, honoraria, and paid expert testimony.

References1. Buffon GL. Compte de Histoire Naturelle, Generale et Par-

ticuliere. Paris: Pankoucke, 1769.2. Anderson PJ. The presence of an astraguloscaphoid bone in

man. J Anat Physiol. 1879;14:452-5.3. Cuveilhier J. Anatomie Pathologique du Corps Humain. Paris:

JB Balliere, 1829.4. Zuckerkandl E. Ueber einen Fall von Synostose zwischen

Talus und Calcaneus. Allg Wien Med Zeitung. 1877;22:293-4.

5. Slomann, HC. On coalition calcaneo-navicularis. J Orthop Surg. 1921;3:586-602.

6. Harris RI, Beath T. Etiology of peroneal spastic flatfoot. J Bone Joint Surg Br. 1948;30:624-34.

7. Badgley CE. Coalition of the calcaneus and navicular. Arch Surg. 1927;15:75-88.

21Bulletin of the NYU Hospital for Joint Diseases 2008;66(1):14-21

8. Wechsler RJ, Schweitzer M, Deely DM, et al. Tarsal coalition: depiction and characterization with CT and MR imaging. Radiology. 1994;193:447-52.

9. Newman JS, Newberg AH. Congenital tarsal coalition: mul-timodality evaluation with emphasis on CT and MR imaging. Radiographics. 2000;20:321-32.

10. Jayakumar S, Cowell HR. Rigid flatfoot. Clin Orthop Rel Res. 1977;(122):77-84.

11. Oestreich AE, Mize WA, Crawford AH, Morgan RC. The “anteater nose”: a direct sign of calcaneonavicular coalition on the lateral radiograph. J Pediatr Orthop. 1987;7:709-11.

12. Gonzalez P, Kumar JS. Calcaneonavicular coalition treated by resection and interposition of the extensor digitorum brevis muscle. J Bone Joint Surg Am. 1990;72:71-7.

13. Conway JJ, Cowell HR. Tarsal coalition: clinical sig-nificance and roentgenographic demonstration. Radiology. 1969;92:799-811.

14. Bohne WHO. Tarsal coalition. Curr Opin Pediatr. 2001;13:29-35.

15. Resnick D. Talar ridges, osteophytes and beaks: a radiologic commentary. Radiology. 1984;151:329-32.

16. Stormont DM, Peterson HA. The relative incidence of tarsal coalition. Clin Orthop Rel Res. 1983;(181):28-36.

17. Harris RI. Rigid valgus foot due to talocalcaneal bridge. J Bone Joint Surg Am. 1955;37:169-83.

18. Harris RI. Retrospect—peroneal spastic flat foot (rigid valgus foot). Follow up notes on articles previously published in the journal. J Bone Joint Surg Am.1965;47 (8):1657-67.

19. Leonard MA. The inheritance of tarsal coalition and its rela-tionship to spastic flat foot. J Bone Joint Surg Br. 1974;56:520-6.

20. Wray JB, Herndon CN. Hereditary transmission of congenital coalition of the calcaneus to the navicular. J Bone Joint Surg Am. 1963;45:365-72.

21. Hochman M, Reed MH. Features of calcaneonavicular co-alition on coronal computed tomography. Skeletal Radiol. 2000;29:409-12.

22. Kulik SA, ClantonTO. Tarsal coalition. Foot Ankle Int. 1996;17:286-96.

23. Mosier KM, Asher MA. Tarsal coalitions and peroneal spastic

flatfoot: A review. J Bone Joint Surg Am. 1984;66:976-84.24. Cowell HR, Elener V. Rigid painful flatfoot secondary to tarsal

coalition. Clin Orthop Rel Res.1983;(177):54-60.25. Jack EA. Bone anomalies of the tarsus in relation to “peroneal

spastic flat foot.” J Bone Joint Surg Br. 1954; 36:530-42.26. Vicent KA. Tarsal coalition and painful flatfoot. J Am Acad

Orthop Surg. 1998;6:274-81.27. Pachuda NM, Lasday SD, Jay RM. Tarsal coalition: etiology,

diagnosis and treatment. J Foot Surg 1990;29:474-88.28. Munk PL, Vellet AD, Levin MF, Helms CA. Current status of

magnetic resonance imaging of the ankle and hindfoot. Can Assoc Radiol J. 1992;43:19-30.

29. Masciocchi C, D’Archivio C, Barile A, et al. Talocalcaneal coalition: computed tomography and magnetic resonance imaging diagnosis. Eur J Radiol. 1992;15:22-5.

30. Emery KH, Bisset GS 3rd, Johnson ND, Nunan PJ. Tarsal coalition: a blinded comparison of MRI and CT. Pediatr Radiol. 1998 Aug;28(8):612-6.

31. Ehrlich MG, Elmer EB. Tarsal Coalition. In: Disorders of the Foot and Ankle, (2nd ed). Philadelphia: WB Saunders, 1991, pp. 921-940.

32. Harris BJ. Anomalous structures in the developing human foot. Anat Rec. 1955;121:399.

33. Solomon LB, Ruhli FJ, Taylor J, et al. A dissection and com-puter tomography study or tarsal coalitions in 100 cadaver feet. J Orthop Res. 2003;21(2):352-8.

34. Snyder RB, Lipscomb AB, Johnston RK. The relationship of tarsal coalition to ankle sprain in athletes. Am J Sports Med. 1981;9:313-7.

35. Lysack JT, Fenton PV. Variations in calcaneonavicular mor-phology demonstrated with radiography. Radiology. 2004 Feb;230(2):493-7.

36. Taniguchi A, Tanaka Y, Kadono K, et al. C sign for diagnosis of talocalcaneal coalition. Radiology. 2003 Aug;228(2):501-5.

37. Rankin EA, Baker GI. Rigid flatfoot in the young adult. Clin Orthop Rel Res. 1974;(104):244-8.

38. Lateur LM, Van Hoe LR, Van Ghillewe KV, et al. Subtalar coalition: diagnosis with the C-sign on lateral radiographs of the ankle. Radiology. 1994;193:847-51.