MRI of Metatarsal Head Subchondral Fractures in Patients with Forefoot Pain

Transcription

1 Musculoskeletal Imaging Original Research Torriani et al. MRI of Metatarsal Head Subchondral Fractures Musculoskeletal Imaging Original Research Martin Torriani 1 Bijoy J. Thomas Miriam A. Bredella Hugue Ouellette Torriani M, Thomas BJ, Bredella MA, Ouellette H Keywords: foot, Freiberg s infraction, metatarsal, MRI, stress fracture DOI: /AJR Received July 10, 2007; accepted after revision September 25, All authors: Musculoskeletal Radiology, Massachusetts General Hospital, 55 Fruit St., YAW 6E, Boston, MA Address correspondence to M. Torriani (mtorriani@hms.harvard.edu). AJR 2008; 190: X/08/ American Roentgen Ray Society MRI of Metatarsal Head Subchondral Fractures in Patients with Forefoot Pain OBJECTIVE. The purpose of our study was to determine the MRI features of metatarsal head subchondral fractures in symptomatic adults. MATERIALS AND METHODS. A retrospective review of foot MRI procedures was performed to detect cases of metatarsal head subchondral fractures over a 6-year period. MR images of selected cases were analyzed by two reviewers for the presence of subchondral fracture, marrow edema-like pattern, metatarsal head flattening, and subchondral sclerosis. Patients with a history of foot surgery, infection, or inflammatory arthritis were excluded. Assessment for coexisting osseous and soft-tissue abnormalities was also performed. RESULTS. Subchondral fractures of the metatarsal heads were seen in 14 patients. All patients were women. The metatarsal head most commonly affected was the second (71%, 10/14) and the dorsal third of the metatarsal articular surface was involved in 79% (11/14). MRI findings of subchondral fracture of the metatarsal head with severe marrow edema-like pattern were seen in 71% (10/14), suggesting early stage changes. Metatarsal head collapse with subchondral sclerosis and mild or absent marrow edema-like pattern were seen in 29% (4/14) indicating late-stage changes. Concurrent abnormalities included three patients (21%) with metatarsal shaft fractures and one patient (7%) with an interdigital neuroma. One subject was treated surgically. CONCLUSION. Subchondral fractures of the metatarsal heads can be detected on MR examinations of adults with forefoot pain. A subchondral fracture with associated marrow edema-like pattern is the most common presentation and likely reflects early stages of metatarsal head infraction. F ractures of metatarsal bones may involve the shaft or subchondral region of the metatarsal head. Shaft fractures most commonly involve the middle and distal shaft of the second and third metatarsals [1]. Subchondral fractures of metatarsal heads are less common and believed to occur secondary to overuse or in patients with underlying conditions predisposing to insufficiency fractures [1, 2]. Although occurring in distinct age groups, metatarsal head subchondral fractures and Freiberg s infraction share several similarities in radiographic appearance and pathophysiology [1, 3, 4]. In addition, symptoms from these entities can be confused with those produced by interdigital (Morton s) neuroma, injuries to the second plantar plate (second ray syndrome), or crystal-induced arthropathy [1]. Prior reports on metatarsal head subchondral fractures mainly discuss radiographic features, with description of MRI findings in one case [1, 2]. To our knowledge, our study is the first to specifically evaluate the MRI features of metatarsal head subchondral fractures in a series of cases also describing concurrent injuries of metatarsals and periarticular soft tissues. In this article, we report our experience with MRI of metatarsal head subchondral fractures in a retrospective analysis of MRI studies of the foot performed at our institution. Materials and Methods This study was approved by the institutional review board, with exemption status for individual informed consent. A retrospective search was performed using Boolean operators (Folio, Open Market) [5] in reports generated for 1,887 consecutive cases of foot MRI studies from adults (more than 18 years old) obtained at our institution from January 2001 to December Selection of cases with metatarsal head subchondral fractures was performed by searching the body and impression of reports for combinations of terms including metatarsal, head, edema, 570 AJR:190, March 2008

2 MRI of Metatarsal Head Subchondral Fractures subchondral, stress, fracture, and Freiberg s. Patients with a history of diabetic foot, osteomyelitis, and inflammatory arthritis were excluded. Patients with a history of foot surgery (bunionectomies, toe amputations, metatarsal resections) were excluded to avoid confounding factors regarding metatarsal head morphology. Data on age, sex, history, and side of injury were collected. Treatment decisions were obtained from clinical records. MRI was performed on 1.5-T scanners, using routine imaging protocols that included sagittal T1-weighted (TR/TE, 700/19; number of excitations [NEX], 2; matrix, ; slice thickness, 4 mm; field of view, 16 cm), sagittal STIR (3,100/43; inversion time, 140 milliseconds; NEX, 2; matrix, ; slice thickness, 4 mm; field of view, 16 cm), coronal fat-suppressed T2- weighted (3,600/44; NEX, 2; matrix, ; slice thickness, 3 mm; field of view, 13 cm), and coronal proton density (2,500/13; NEX, 2; matrix, ; slice thickness, 3 mm; field of view, 13 cm). In select cases, axial T1-weighted (700/16; NEX, 2; matrix, ; slice thickness, 4 mm; field of view, 15 cm) and contrast-enhanced sagittal, coronal, and axial fat-suppressed T1- weighted pulse sequences (650/16; NEX, 2; matrix, ; slice thickness, 4 mm; field of view, 15 cm) after IV injection of 0.1 mmol/kg of gadopentetate dimeglumine (Magnevist, Bayer HealthCare) were performed. MRI interpretation of selected cases was performed by consensus of two experienced musculoskeletal radiologists. MRI criteria included presence of a subchondral fracture, marrow edemalike pattern, flattening of articular surface, and subchondral sclerosis. Subchondral fractures were characterized by an area of linear hypointense signal intensity on T1-weighted images, STIR or T2- weighted images, or both. Edema-like pattern was identified by the presence of diffuse hypointense signal intensity on T1-weighted images, hyperintense signal intensity on STIR or T2-weighted images, or hyperintense signal intensity on contrast-enhanced T1-weighted images. Flattening was identified by focal or diffuse loss of articular surface convexity on sagittal images. Subchondral sclerosis was characterized by ill-defined and irregular articular surface with nodular areas of mixed (predominantly hypointense) signal intensity. Sagittal MR images were used to determine the location and length of metatarsal head abnormalities. The location of subchondral fractures and metatarsal head flattening was evaluated using the articular surface as a reference, dividing it equally into dorsal, central, and plantar thirds. The length of subchondral fracture was obtained by measurements on sagittal MR images using a measuring tool in our PACS (Impax 4.0, Agfa). The length of subchondral sclerosis was determined using the same method. Measurements are presented followed by the standard error (SE) of the mean. Marrow edema-like pattern was evaluated semiquantitatively: absent (normal marrow signal intensity), mild (abnormal marrow signal intensity limited to subchondral region), moderate (abnormal marrow signal intensity involving 50% or less of metatarsal head), and severe (abnormal marrow signal intensity involving 50% or more of metatarsal head and adjacent soft tissues). Metatarsal shaft fractures were characterized by hypointense cortical disruption on T1-weighted images with surrounding hyperintense bone marrow edema on STIR or T2-weighted images; tendinous and ligamentous injuries were identified by the presence of hyperintense signal intensity on STIR or T2-weighted images surrounding the metatarsophalangeal (MTP) joint capsule or the flexor or extensor tendons; and MTP joint effusion and bursitis were determined by the presence of fluidlike hyperintense signal intensity on STIR or T2-weighted images within joint spaces and between metatarsal heads, respectively. MRI findings of periarticular soft tissues such as the plantar plate, joint capsule, and flexor and extensor tendons were classified as normal, sprain (laxity without discontinuity), partial (partial or interstitial discontinuity), or full-thickness (complete discontinuity) tears. In cases in which foot radiographs were available, interpretation was performed by consensus of two experienced musculoskeletal radiologists. The radiographic criteria for identification of metatarsal head subchondral fractures included presence of a zone of subchondral sclerosis or flattening of the metatarsal head. The presence of degenerative joint disease (DJD) of MTP joints was evaluated on MRI and radiography when available. DJD was characterized by evidence of marginal osteophytosis and joint space narrowing, classified as absent, mild (presence of marginal osteophytes, joint space narrowing less than 50%), moderate (presence of marginal osteophytes, joint space narrowing more than 50%), and severe (prominent osteophytes, bone-on-bone contact of articular surfaces). Using available radiographs, we measured the hallux valgus angle (HVA), defined as the angle formed by the intersection of longitudinal axes of the diaphyses of the first metatarsal and proximal phalanx, adopting 15 as the upper limit of normal for this measurement [6]. Results A total of 18 cases fulfilled at least one search criteria. Four patients were excluded: two women had involvement of the second Fig year-old woman with forefoot pain over third and fourth metatarsal heads. Sagittal T1- weighted MR image through third metatarsal shows subchondral low signal intensity in subchondral region consistent with subchondral fracture (arrowheads) without flattening. Radiographs (not shown) were unremarkable in this patient. metatarsal with coexisting active osteomyelitis of the first toe, of which one had undergone a partial amputation; one woman had involvement of the second metatarsal head with prior bunionectomy; and one man with psoriatic arthritis had edema of the second and third metatarsal heads with severe inflammatory changes of the MTP joints. The remaining 14 cases formed our study cohort. Two cases were from one patient with concurrent subchondral fractures of two metatarsal heads (total of 13 patients). The mean age of the patients was 52.8 years (age range, years) at the time of MRI. All patients were women and only one had a history of acute trauma immediately preceding foot symptoms. In eight patients, the mean duration of symptoms (pain and swelling in the metatarsal head region) was 12 weeks (range, 1 35 weeks). Information regarding the duration of symptoms was not available in the remaining six patients, of whom one had history of treatment for giant cell arteritis with prednisone for 14 months and another had undergone ankle surgery with the use of a weight-bearing boot for 6 weeks preceding forefoot pain. The metatarsal head most frequently involved was the second (71%, 10/14), followed AJR:190, March

3 Torriani et al. Fig year-old woman with forefoot pain for 6 weeks. Sagittal T1-weighted MR image through second metatarsal shows subchondral fracture involving all thirds of articular surface (arrow) without flattening. by the fourth (14%, 2/14) and third (14%, 2/14). One subject had subchondral fractures of the third and fourth metatarsal heads of the same foot. The right foot was affected more frequently (57%, 8/14). Radiographs were available in 86% (12/14) of our cases. There were two clusters of MRI findings. The first group was composed of 10 cases (71%, 10/14) that exhibited subchondral fractures with associated severe marrow edema-like pattern. These cases showed involvement of dorsal and central (n = 3), dorsal (n = 2), dorsal and plantar (n = 1), plantar (n = 3), and all thirds (n = 1) of the metatarsal head articular surface (Figs. 1 and 2). The average length of subchondral fractures was 9.5 ± 1.4 mm. MR evidence of metatarsal head flattening was seen in 40% (4/10) (Fig. 3). There was no evidence of subchondral sclerosis or cystic or degenerative changes. All cases had MTP joint effusion. Radiographs were available in nine cases, of which three showed metatarsal head flattening. The second group included four cases (29%, 4/14) that presented subchondral sclerosis with metatarsal head flattening and mild or absent marrow edema-like pattern. The dorsal and central (n = 3) and dorsal (n = 1) thirds of the metatarsal head articular surface were involved. The average length of subchondral sclerosis measured 8.7 ± 0.5 mm. Edema-like pattern was mild in one and A B Fig year-old woman with forefoot pain after wearing walking boot for 6 weeks after ankle surgery. A and B, Sagittal T1-weighted (A) and STIR (B) MR images through second metatarsal show metatarsal head flattening (arrowheads, A) and subchondral fracture (arrows) involving dorsal and central thirds of articular surface. There is severe soft-tissue and bone marrow edema-like pattern. Skin marker is present in dorsum of foot. absent in three cases. Cystic changes in the subchondral region without surrounding marrow edema were noted in two cases (Fig. 4). Mild degenerative changes of the MTP joints were present in three cases. No patients in this group had MTP joint effusion. Radiographs were available in three cases and showed metatarsal head flattening. In a combined analysis of all cases, metatarsal head flattening was present in 57% (8/14). The dorsal third of the metatarsal head was affected by flattening or subchondral A Fig year-old woman with forefoot pain referred for evaluation of possible stress fracture. A and B, Axial T1-weighted (A) and sagittal STIR (B) MR images through second metatarsal show flattening, subchondral sclerosis, and rounded area of cystic change without surrounding marrow edema-like pattern (arrow). B 572 AJR:190, March 2008

4 MRI of Metatarsal Head Subchondral Fractures fracture in isolation or combination with other thirds in 79% (11/14) of cases. Simultaneous involvement of dorsal and central thirds (43%, 6/14) was most common followed by the dorsal (21%, 3/14), dorsal and plantar (7%, 1/14), plantar (21%, 3/14), and all thirds (7%, 1/14) of the metatarsal head articular surface. Metatarsal shaft fractures were present in 21% (3/14) and were the most common coexisting abnormality. In one case, serial radiographs during 17 months initially showed a nondisplaced fracture of the second metatarsal shaft with progressive flattening and a subchondral fracture of the second metatarsal head at MRI. In another case, flattening of the second metatarsal head coexisted with an acute stress fracture at the base of the same metatarsal (Fig. 5). A third case was of an obese 62-year-old woman with equinus deformity of the right foot, fourth metatarsal neck fracture, and third metatarsal head subchondral fracture at MRI. Another subject had a Morton s neuroma between the third and fourth metatarsal heads. Only non-weight-bearing anteroposterior views of the foot were available in our series. The mean HVA was 20 (range, ), and in 10 cases the HVA was larger than 15. No abnormalities of the plantar plate or tendons were identified. Intermetatarsal bursitis was not seen in our series. None of our cases were evaluated on CT. One patient was treated operatively because of persistent pain for 7 months that did not resolve with use of a weight-bearing boot. Partial improvement of pain was obtained after surgical débridement. Three patients were treated with immobilization, yielding relief of symptoms. Medical records for the remaining 10 patients did not reveal details regarding conservative or surgical management. A B Fig year-old woman with forefoot pain for 1 week. A and B, Sagittal T1-weighted (A) and STIR (B) MR images through second metatarsal show metatarsal head flattening with subchondral sclerosis and mild edema-like pattern (arrowheads). Acute stress fracture through base of same metatarsal (arrows) with severe soft-tissue and bone marrow edema-like pattern are noted. Discussion The location and radiographic appearance of metatarsal head subchondral fractures are similar to those seen in Freiberg s infraction. The term infraction refers to an incomplete fracture of bone without displacement of fragments [7]. Freiberg s infraction is defined as an osteochondrosis of the metatarsal heads, particularly the second metatarsal, with a lesion that presents the radiologic, anatomic, clinical, and evolutionary characteristics of subchondral cancellous bone necrosis [8]. Although Freiberg s infraction affects adolescents (12 to 18 years) and metatarsal head subchondral fractures are seen in adults [1], both entities likely share a cause that combines mechanical stress, subchondral fracture, vascular injury, and subsequent osteonecrosis [1, 8 10]. Because of this multifactorial nature, we refer to such metatarsal head fractures as subchondral (on the basis of anatomic location) rather than stress fractures (on the basis of cause) [1, 2]. Furthermore, we consider our case series as representative of metatarsal head subchondral fractures because all patients were adults (more than 18 years old). Prior studies indicate the second metatarsal is most commonly involved [2, 9, 10]. The second metatarsal is usually the longest and is subjected to the greatest reactive ground forces during ambulation, which may increase its susceptibility to repetitive trauma [4, 9]. Prior studies of patients with Freiberg s infraction have shown the longest metatarsal was affected in 85% [4] and 42% of cases [10]. This preferential involvement is further supported by evidence that the second and third rays are less mobile and have more difficulty in dispersing the applied forces during weight bearing [8]. During the toe-off phase of ambulation, the base of the proximal phalanges creates compressive and shearing forces as the phalanges ride along the dorsal articular surfaces of the metatarsal heads. This type of action predisposes to shearing injuries at the interface between cartilage and subchondral bone, which is the weakest point of force dissipation [8]. In a prior case report in which resection of a collapsed metatarsal head was performed in a symptomatic adult, a shearing injury at the interface between mineralized and nonmineralized cartilage was identified at histopathology [8]. This mechanism may be accentuated in women because of the use of high-heeled shoes that maintain the phalanges in a dorsiflexed position, which also contributes to retrograde pressure on the metatarsal heads [9]. Studies of the vascularity of the metatarsal heads indicate two main arterial sources: the dorsal metatarsal arteries, which arise from the dorsalis pedis artery, and the plantar metatarsal arteries, which are branches of the posterior tibial artery [11]. These two vessels anastomose forming an arterial network around the metatarsal heads with nutrient arteries traversing the metaphyseal cortex to supply the subchondral bone [11]. This vascularity may be affected by surgical procedures such as metatarsal head osteotomies, in which extensive capsular stripping may result in damage to the medial and lateral head vessels [11]. Trauma, metatarsal shaft fractures, or vasculopathy may also represent potential causes for disruption of the tenuous blood supply to the metatarsal heads. AJR:190, March

5 Torriani et al. Therefore, it is possible that Freiberg s infraction and metatarsal head subchondral fractures occurring in adults have the same pathogenesis. The primary lesion may initiate as a subchondral fissure most commonly involving the dorsal aspect of the metatarsal head. This fissure may lead to disruption of the epiphyseal vascular supply, evolving to ischemic bone necrosis with subsequent repair or collapse [8 10]. Five stages of anatomic changes have been described, with progression or consolidation at any stage [12]. No sequelae are seen when consolidation takes place at an early stage, whereas flattening or arthrosis may be seen when consolidation occurs at later stages [10]. The results of our study are concordant with several of these observations. All patients in our series were women and the second metatarsal head was affected in the majority of cases. There was a predilection for involvement of the dorsal and central portions of the metatarsal heads, best characterized on sagittal T1-weighted or STIR images. Such involvement was manifested by either the presence of a subchondral fracture or flattening at this level. On the other hand, we also noted involvement of other areas, such as the plantar and central thirds. These findings suggest compressive and shearing forces may also occur in such areas, indicating variability in biomechanical characteristics of patients. In our study, two clusters of MRI findings were noted. One group of patients had subchondral fractures with severe bone marrow edema-like pattern that extended into the periarticular soft tissues, suggesting an early stage phenomenon. Another group exhibited metatarsal head flattening with subchondral sclerosis and mild or absent marrow edema-like pattern, pointing toward a late-stage process. In the study by Chowchuen and Resnick [1], MRI was available in one case of metatarsal head flattening without significant marrow edema-like pattern surrounding a focal area of hypointense signal intensity on T1-weighted images and hyperintense signal intensity on STIR images. In our series, this pattern was present in two cases. Because of the lack of marrow and soft-tissue edema, we believe such a pattern likely reflects a late-stage process, with focal hyperintense areas in the subchondral region representing granulation tissue [8] or early cystic changes from superimposed DJD. The two groups observed in our study apparently correlate with the five-stage classification of Freiberg s disease by Smillie [12]: early stage MRI findings may correspond to stages I through III, which represent an initial fissure fracture that progresses to flattening, and late-stage MRI findings possibly relate to stages IV and V in which additional flattening, deformity, and arthrosis are seen. Taken together, our observations suggest MRI is capable of differentiating early from late stages of metatarsal head infraction. Furthermore, our results also highlight the value of forefoot MRI for detection of acute subchondral fractures, which may be the only finding in early stage cases and is unlikely to be visualized on radiographs [12]. MRI was also sensitive to the presence of joint effusion, metatarsal shaft stress fractures, and Morton s neuroma. Coexisting abnormalities seen in our series included metatarsal shaft stress fractures (n = 3) and one case of interdigital neuroma. However, because of the lack of statistical power, it is not possible to determine a relationship with metatarsal head subchondral fractures. Prior reports suggest infraction of a metatarsal head may result from altered weight bearing because of unrelated processes that place abnormal stress on the metatarsal heads [3]. Binek et al. [3] described two patients who evolved with metatarsal head flattening after a surgical procedure to the foot (64-year-old woman) and placement of a short leg cast (8-year-old boy). In our study, patients with history of foot surgery (bunionectomies, toe amputations, metatarsal resections) were not included to avoid confounding factors regarding metatarsal head morphology. Of note, one of our patients presented with metatarsal head subchondral fracture after wearing a walking boot for 6 weeks recovering from ankle surgery (tibiotalar débridement and lateral ligament reconstruction). Another subject showed late-stage changes to a metatarsal head with concurrent acute stress fracture of the same metatarsal, suggesting the head abnormality may have contributed to altered biomechanics. On the basis of nonweight-bearing anteroposterior radiographs, the mean HVA in our series was larger than 15. Although such methodology may underestimate hallux valgus severity, the potential association of this entity with increased load on the metatarsal heads may represent an area for future investigation. A pitfall in the diagnosis of metatarsal head subchondral fracture or Freiberg s infraction is the presence of a normal anatomic variant characterized by a flat configuration of the second metatarsal head with widening of the MTP joint space [13]. Jensen and de Carvalho [13] described this finding in 9.7% of healthy volunteers as bilateral, symmetric, and equally distributed between male and female subjects. In patients with acute symptoms and borderline radiographic findings, the presence of subchondral fracture and edema-like pattern on MRI may be helpful to exclude the possibility of anatomic variation. Chronic DJD may also lead to confusion because of hypertrophic and subchondral changes that may affect metatarsal head morphology. However, involvement of both sides of the MTP joint and joint space narrowing may provide further indication of a degenerative phenomenon. In addition, symptoms from subchondral fractures of the metatarsal heads may be confused with injuries to the second plantar plate (second ray syndrome) or crystal-induced arthropathy [1]. The treatment of metatarsal head subchondral fractures and Freiberg s infraction includes modification of activities, orthotics, and surgical intervention [14]. Surgery is considered when conservative measures do not relieve symptoms and consists of joint débridement, drilling, metatarsal osteotomies, joint arthroplasty, elevation of the depressed articular fragment, and bone grafting. Arthroscopic management has recently been described as an alternative technique for removal of loose bodies, arthroscopic débridement, and osteochondral transplantation [14]. In our study, only one subject was treated surgically, undergoing joint débridement with partial improvement of symptoms. Our study has a few limitations. The case selection methodology used in our study did not allow evaluation of the true prevalence of metatarsal head subchondral fractures in the general population. There is a possibility that metatarsal head abnormalities were missed or misinterpreted in cases not selected by our search technique, and some cases of subchondral fracture may have been described using terminology distinct from our method. Follow-up clinical history and imaging was not available for most patients. This could be useful for diagnostic confirmation, especially in cases that were imaged during early stage disease (subchondral fracture without collapse). No histologic data were available in our patients to confirm the precise cause of imaging findings. However, the imaging findings are comparable to those of subchondral injury in other anatomic areas and correlate to histologic abnormalities described in previous reports [8, 10]. Finally, it was not possible 574 AJR:190, March 2008

6 MRI of Metatarsal Head Subchondral Fractures to determine if the initial injuries in cases with late-stage MRI findings occurred during adulthood. In summary, our study shows that subchondral fractures of the metatarsal heads can be detected using MRI of the foot. Subchondral fracture and marrow edema-like pattern likely represent the early stages of the disease, whereas metatarsal head collapse and subchondral sclerosis suggest a late-stage process. These findings may be useful to increase the specificity of MR interpretation in the context of adults with forefoot pain undergoing MR examination. References 1. Chowchuen P, Resnick D. Stress fractures of the metatarsal heads. Skeletal Radiol 1998; 27: Lechevalier D, Fournier B, Leleu T, Crozes P, Magnin J, Eulry F. Stress fractures of the heads of FOR YOUR INFORMATION the metatarsals: a new cause of metatarsal pain. Rev Rhum Engl Ed 1995; 62: Binek R, Levinsohn EM, Bersani F, Rubenstein H. Freiberg disease complicating unrelated trauma. Orthopedics 1988; 11: Stanley D, Betts RP, Rowley DI, Smith TW. Assessment of etiologic factors in the development of Freiberg s disease. J Foot Surg 1990; 29: Thomas BJ, Ouellette H, Halpern EF, Rosenthal DI. Automated computer-assisted categorization of radiology reports. AJR 2005; 184: Easley ME, Trnka HJ. Current concepts review: hallux valgus. Part 1. Pathomechanics, clinical assessment, and nonoperative management. Foot Ankle Int 2007; 28: Thomas CL, ed. Taber s cyclopedic medical dictionary, 17th ed. Philadelphia, PA: F. A. Davis Company, 1993: Young MC, Fornasier VL, Cameron HU. Osteochondral disruption of the second metatarsal: a variant of Freiberg s infraction? Foot Ankle 1987; 8: Nguyen V, Keh R, Daehler R. Freiberg s disease in diabetes mellitus. Skeletal Radiol 1991; 20: Gauthier G, Elbaz R. Freiberg s infraction: a subchondral bone fatigue fracture a new surgical treatment. Clin Orthop Relat Res 1979: Petersen WJ, Lankes JM, Paulsen F, Hassenpflug J. The arterial supply of the lesser metatarsal heads: a vascular injection study in human cadavers. Foot Ankle Int 2002; 23: Smillie IS. Treatment of Freiberg s infraction. Proc R Soc Med 1967; 60: Jensen EL, de Carvalho A. A normal variant simulating Freiberg s disease. Acta Radiol 1987; 28: Lui TH. Arthroscopic interpositional arthroplasty for Freiberg s disease. Knee Surg Sports Traumatol Arthrosc 2007; 15: Mark your calendar for the following ARRS annual meetings: April 13 18, 2008 Marriott Wardman Park Hotel, Washington, DC April 26 May 1, 2009 John B. Hynes Veterans Memorial Convention Center, Boston, MA May 2 7, 2010 Grand Hyatt San Diego, San Diego, CA May 1 6, 2011 Hyatt Regency Chicago, Chicago, IL April 29 May 4, 2012 Vancouver Convention Center, Vancouver, BC, Canada AJR:190, March