Nogah Haramati

The diagnostic role of gadolinium enhanced MRI in distinguishing between acute medullary bone infarct and osteomyelitis

The objective of the study was to evaluate the diagnostic utility of contrast enhanced magnetic resonance imaging (MRI) for distinguishing between acute medullary bone infarct and osteomyelitis. There were 11 patients (age 6-34 years) presented to our institution between December 1994 and February 1998 with a clinical differential diagnosis of acute bone infarct versus osteomyelitis and inconclusive radiographs were imaged using MRI. All but one received i.v. gadolinium. Nine of the patients had homozygous Sickle Cell disease (SCD) and two had Systemic Lupus Erythematosus (SLE), the latter requiring chronic methylprednisolone. Osteomyelitis was confirmed either by biopsy alone or by the combination of Gallium(67) scan in conjunction with positive blood cultures and clinical resolution following antibiotics. Infarcts without osteomyelitis were confirmed either by biopsy or resolution of symptoms without antibiotic therapy. All patients had at least six months clinical follow-up. The results found that seven of nine patients with SCD had acute infarct only. One patient with SCD had osteomyelitis only. Three patients (two SLE and one SCD) had both acute-on-chronic infarcts and superimposed osteomyelitis, one with an adjacent soft tissue abscess. Accurate distinction between infarct and osteomyelitis was impossible for one patient with SLE who did not receive contrast. All other cases were correctly diagnosed prospectively based on distinct patterns of MRI contrast enhancement. In all adult patients, acute infarcts demonstrated thin, linear rim enhancement on MRI while osteomyelitis revealed more geographic and irregular marrow enhancement. Two of four cases of osteomyelitis also demonstrated subtle cortical defects with abnormal signal traversing marrow and soft tissue. The single pediatric patient demonstrated elongated, serpiginous central medullary enhancement with periostitis. We concluded that the pattern of MR contrast enhancement may allow accurate distinction between acute infarct and osteomyelitis, or recognition of osteomyelitis superimposed on bone infarction.

CT scans through metal scanning technique versus hardware composition.

OBJECTIVE Streak artifact on CT scans of metal containing areas has been a long standing problem. Although several artifact reducing methods have been used to improve image quality, most have been limited by requiring specialized equipment or lengthy complex calculations that are not automated. Others have shown that increasing the beam energy results in increased thickness of metal that may be imaged by CT without severe image degradation. We have studied the image quality of bone surrounding metal both with titanium and cobalt-chrome prostheses using various scanning techniques. METHODS In a double blind fashion, 28 radiology residents and attendings were surveyed as to the best technique for imaging bone detail surrounding metal. A series of images was arranged of an implanted titanium prosthesis, a cobalt-chrome prosthesis and a pelvis repaired with stainless steel pelvic reconstruction plates. Scans were performed using three techniques: 120 kVp, 170 mA, 2 s, 360 degrees rotation, 140 kVp, 140 mA, 3 s, 360 degrees rotation, 140 kVp, 140 mA, 4 s, 420 degrees rotation. RESULTS Titanium was superior to cobalt-chrome (p < .0001 Wilcoxon Signed Rank Test). No advantage was noted for higher kVp or increased scan arc of 420 degrees compared to the standard 360 degrees. CONCLUSION Titanium allows improved bone detail surround the metal than CT cobalt-chrome. We have found no advantage to using either high kVp or a 420 degrees scan arc to improve the image quality of bone surrounded by metal.

Magnetic resonance imaging of occult fractures of the proximal femur

The evaluation of the painful hip in the elderly osteoporotic patient with normal plain radiographs can be difficult. We studied 15 osteopenic patients with normal plain radiographs and suspected hip fractures with magnetic resonance (MR) imaging and found MR to be an excellent aid in detecting occult fractures. A clear fracture was seen in 10 of the 15 patients, who then underwent surgical repair based on the MR study. The remaining patients had no MR-demonstrable fracture and were successfully treated nonoperatively. Some believe that a negative bone scan in this population of patients should be repeated within 3 days prior to a definitive “no fracture” decision being made. Unfortunately, bone scanning lacks spatial resolution, and increased osteoblastic activity may be caused by other pathologic processes besides fracture. Two of the 15 patients had MR-demonstrated bone infarcts near the fracture. One patient also had femoral head osteonecrosis on the side of the fracture. One patient with metastatic prostatic carcinoma had a hip fracture and one patient with metastatic breast carcinoma had no fracture. Not only is MR imaging an excellent technique for delineating occult fractures, but due to its spatial resolution, associated bone disorders adjacent to fractures can be detected in most instances. From a cost perspective, rapid diagnosis and early treatment of an occult femoral fracture is advisable. A reduced hospital stay pending diagnosis and the early institution of definitive therapy also decrease the chance that a simple non-displaced fracture will displace and require more complex management with resultant increased morbidity and cost. We propose that, especially in elderly, osteopenic patients with normal plain radiographs and a high index of suspicion for hip fracture, MR can serve as the sole additional imaging study in most instances.

Slipped capital femoral epiphysis: a physeal lesion diagnosed by MRI, with radiographic and CT correlation.

Abstract Objective. To define and compare early lesions associated with slipped capital femoral epiphysis (SCFE) on magnetic resonance imaging (MRI), computed tomography (CT) and radiography. Design and patients. Thirteen patients with 15 symptomatic hips due to SCFE underwent radiography and MRI; CT was performed in 12 patients. SCFE was graded on radiographs, head/neck angles and qualitative changes were evaluated on CT, and morphologic/signal abnormalities were determined on MRI. Results. Physeal widening, apparent on T1-weighted MRI, was evident in every case of SCFE, including one presumed “pre-slip.” T2-weighted images demonstrated synovitis and marrow edema but obscured physeal abnormalities. CT head/neck angles ranged from 4–57° for symptomatic to 0–14° for asymptomatic hips. Physeal and metaphyseal changes were variably identified on both radiographs and CT in all cases of SCFE, but not in the pre-slip. Conclusion. MRI clearly delineates physeal changes of both pre-slip and SCFE, and demonstrates very early changes at a time when radiographs and CT may appear normal.

MRI features of confirmed "pre-slip" capital femoral epiphysis: a report of two cases

Abstract.We describe the morphologic and signal changes detected about the proximal femoral growth plate in two patients with hip pain preceding the progression to slipped capital femoral epiphysis using magnetic resonance imaging.

MRI of the Stener lesion.

AbstractObjective. To assess the utility of MR in detecting surgically induced Stener lesions (displaced thumb ulnar collateral ligaments) in cadaveric models. Design. Six cadaver thumbs had ulnar collateral ligament (UCL) tears created surgically. MR examinations (2D STIR and 3D GRASS) were performed identically on all specimens both before displacement (non-Stener) and after displacement (Stener lesion) of the UCL. The MR images were then randomly numbered. Each image was evaluated separately in blinded fashion by four musculoskeletal radiologists for the presence or absence of a Stener lesion. Each radiologist reinterpreted the images after an interval of several days. The interpretation was based on previously published criteria for Stener lesion diagnosis by MR. Results. The sensitivity of GRASS ranged from 0.17 to 0.67 with the most experienced reader scoring the lowest. The specificity of GRASS ranged from 0.33 to 1.0 (most experienced reader 0.67, 0.83). STIR had a sensitivity of 0.00–0.17 and a specificity of 0.53–0.83. The κ values for inter- and intraobserver agreement were measured. The intraobserver κ for GRASS was 0.27–0.75 (most experienced reader 0.75). Conclusions. 2D imaging is probably inadequate for the evaluation of Stener lesions. The most likely reason is that the STIR slice thickness of 3 mm limits resolution of small UCLs. The poor sensitivity and specificity of GRASS as well as poor interobserver agreement suggest that MR may not be sufficiently accurate for Stener lesion evaluation.

Calcification of the ligamentum arteriosum in adults: CT features

OBJECTIVE The present study describes the frequency and pattern of ligamentum arteriosum calcification seen on chest CT in adults. MATERIALS AND METHODS We retrospectively reviewed 402 sequential unenhanced chest CT studies for ligamentum arteriosum calcification, atherosclerotic cardiac or aortic calcification, and granulomatous calcification. The pattern of calcification was characterized as curvilinear, punctate, or clumped. RESULTS Mean patient age was 60 (+/- 18, range 18-97) years with 214 women (53%) and 188 men (47%). Of these patients 194 (48%) had calcification in the ligamentum arteriosum including 26 (6%) with calcification in the ligamentum arteriosum alone, 108 (27%) with atherosclerotic calcification, 11 (3%) with granulomatous calcification, and 49 (12%) with both. A total of 100 patients (25%) had no calcifications. In the study population 169 patients (42%) had atherosclerotic calcification, 32 (8%) had granulomatous calcification, and 75 (19%) had both. The patterns of calcification for the 26 patients with calcification of the ligamentum arteriosum alone were curvilinear (7 patients, 27%), punctate (17 patients, 65%), and clumped (2 patients, 8%). The pattern of ligamentum arteriosum calcification for the 108 patients with only atherosclerotic and ligamentum arteriosum calcification was curvilinear (28 patients, 26%), punctate (32 patients, 30%), and clumped (48 patients, 44%). The patients without any calcifications and the patients with ligamentum arteriosum calcification formed one group (based on their similarity in age, mean 47 years) with a prevalence of ligamentum arteriosum calcification of 21%. The patients with atherosclerotic and ligamentum arteriosum calcification formed a second group (mean age 71 years) with a prevalence of ligamentum arteriosum calcification of 65%. CONCLUSION On unenhanced chest CT calcification of the ligamentum arteriosum is a common finding in adults and increases in prevalence with increasing age and atherosclerosis.

Abnormal geometry of the distal radioulnar joint: MR findings

Previously developed criteria of normal distal radioulnar joint (DRUJ) axial geometry were applied to routine magnetic resonance (MR) images of 50 wrists. All wrists lacked clinically evident DRUJ instability. An attempt to apply the three geometric criteria to seven of the cases was not possible, since the prescribed landmarks were not visible. The remaining 43 cases were retrospectively divided into a symptomatic group (25 wrists), with clinical abnormalities referable to the ulnar side of the wrist, and an asymptomatic group (18 wrists), with no such abnormalities. Abnormal DRUJ geometry was found in 12 symptomatic and 2 asymptomatic wrists. This difference between the groups is significant (p=0.02), suggesting that many symptomatic wrists exhibit different DRUJ geometry than is found in most asymptomatic wrists.

Proximal wrist imaging

Ligamentous injuries to the triangular fibrocartilage are often difficult to evaluate. The plain film examination along with the physical examination serves as the basis for the initial assessment of most wrist injuries. Ligamentous injuries of the wrist often present at the subacute or chronic state with the physical examination and plain films being insufficient to make a definitive diagnosis in many cases and the plain film radiographic exam being of little additional value. This paper reviews the triangular fibrocartilage complex including the underlying anatomy and biomechanics, along with its radiographic evaluation by modern imaging techniques.

Pulmonary pseudonodules on computed tomography: a common pitfall caused by degenerative arthritis.

This study assessed the prevalence of pseudonodules of the lung on computed tomography (CT) related to degenerative arthritis at the distal first rib and sternum. A total of 250 serial chest CT scans were retrospectively reviewed by a chest and a musculoskeletal radiologist. A pseudonodule was defined as a round or oval opacity identified on lung window surrounded by aerated lung and superimposable on an osseous structure on the immediate cephalad slice. The pseudonodules were classified as originating from one of the following three locations: first costochondral junction, first costosternal junction, or sternoclavicular junction. CT slice thickness was 8 mm or 10 mm. For each side, the presence of a pseudonodule and its classification was noted. The presence or absence of degenerative arthritis was also noted for each side. The median patient age of the 111 men and 139 women was 62 years (range, 2–97). Degenerative arthritis was noted bilaterally in 94 (38%) and unilaterally in two. Pseudonodules were present in 29 patients (12%). Eighteen (62%) were men with a median age of 73 years (range, 50–84), and 11 (38%) were women with a median age of 67 years (range, 52–85). All of the patients with pseudonodules had bilateral degenerative arthritis. Twenty-seven pseudonodules were caused by degenerative arthritis at the first costochondral junction, four at the first costosternal junction, and one at the sternoclavicular junction. Fifteen pseudonodules were right sided and 17 were left sided. Three patients had bilateral pseudonodules. Pulmonary pseudonodules caused by degenerative arthritis, most commonly at the first costochondral junction, were present in 12% of our series. This common finding occasionally may be misinterpreted as a true pulmonary nodule.