Imaging the Postoperative Elbow.

Evaluation of postoperative images of any joint can be a daunting task, and the elbow is no exception. Patients may be imaged with a complication of the repair, or the postoperative changes may be incidentally observed as the patient is imaged for other reasons. We divide the postoperative elbow into soft tissue procedures (covering ligament and tendon repairs, as well as compartmental release and nerve transposition), joint-related procedures (osteochondral lesion treatment, ostectomy, and joint replacement), and bone procedures (fracture fixation). We summarize the procedures and their indications, show normal imaging appearances, and finally cover common complications.

Lateral femoral condyle insufficiency fractures: imaging findings, demographics, and analysis of outcomes.

OBJECTIVE: To describe imaging characteristics and demographics of lateral femoral condyle insufficiency fractures (LFCIFs) and identify characteristics associated with progression to epiphyseal collapse. MATERIALS AND METHODS: A retrospective review of 105 consecutive patients with LFCIF was performed (mean age 58.1 years) after excluding post-traumatic and pathological fractures. Lesion size and location, presence of bone marrow edema-like signal, soft tissue edema, chondrosis grade, and meniscus pathology were documented. Demographics were recorded from the electronic patient record. Follow-up MRI and/or radiographs were evaluated for healing/stability or progression to epiphyseal collapse. Bone mineral density was assessed from dual-energy x-ray absorptiometry (DEXA) scans and/or radiographs. RESULTS: Fifty-six female and 49 male subjects were included. Female subjects were older at presentation (60.5 versus 56.3 years, p = .02). A total of 61.7% of the subjects with available DEXA and/or radiographs had osteopenia/osteoporosis. The central weight-bearing (61%) and outer condyle (54.3%) were most involved. High-grade chondrosis was present in ≥ 1 compartment in 70.5% including 42% in the lateral compartment. A total of 67.6% had ≥ 1 meniscus tear with similar frequency of medial and lateral tears (47.6% versus 41%). Bone marrow edema-like signal was present in all cases; soft tissue edema was present in 83.8%. Fifty-three subjects had available follow-up MRI (n = 24) and/or radiographs (n = 29). Increased age, fracture dimensions, presence of medial meniscus tears, and high-grade patellofemoral chondrosis were associated with progression (p ≤ .05). CONCLUSION: LFCIFs are associated with meniscus tears, high-grade chondrosis, and osteopenia/osteoporosis with more global knee pathology present when compared with medial femoral condyle insufficiency fracture. Increased age, medial meniscus tears, fracture dimensions, and high-grade patellofemoral chondrosis were associated with progression.

CT-guided Cervical Bone Biopsy in 43 Patients: Diagnostic Yield and Safety at Two Large Tertiary Care Hospitals.

RATIONALE AND OBJECTIVES: The cervical spine is a high-risk area for percutaneous biopsy compared to the thoracic and lumbar regions. Biopsy of the cervical spine is less commonly undertaken, and previously published series on diagnostic yield and safety of cervical spine biopsy have been limited to 12 patients or less. The purpose of our study is to further define the diagnostic yield of computed tomography (CT)-guided biopsy for bony lesions identified in the cervical spine, by combining data from two large tertiary care referral centers. METHODS: A retrospective review of an imaging database was performed to identify all percutaneous CT-guided biopsies of the cervical spine performed at two tertiary care hospitals from 2010 to 2015. Core biopsies were obtained whenever possible and supplemented with fine-needle aspiration in some cases. Histopathologic results of the biopsy were recorded, as were changes in subsequent management, need for repeat biopsy, and complications. RESULTS: Forty-three patients underwent CT-guided biopsy of the cervical spine. Sufficient tissue for histopathologic analysis was obtained in 41 out of 43 cases, for a yield of 95%. One case was false-negative and one was deemed insufficient by the pathologist for diagnostic purposes; in both of these cases, only a fine-needle aspiration was obtained. There were no immediate or delayed complications. CONCLUSIONS: Percutaneous biopsy of the cervical spine is a safe and high-yield method of obtaining a tissue diagnosis when performed under image guidance with CT.

Prevalence and pattern of gluteus medius and minimus tendon pathology and muscle atrophy in older individuals using MRI.

PURPOSE: To evaluate gluteus medius and minimus tendon pathology and muscle atrophy in older individuals using MRI. METHODS: A retrospective MRI study of 185 individuals was performed. The inclusion criterion was age ≥50. Exclusion criteria were hip surgery, fracture, infection, tumor, or inadequate image quality. Greater trochanteric bursitis was graded none, mild, moderate, or severe. Gluteus medius, gluteus minimus, and iliopsoas tendinopathy was graded normal, tendinosis, low-grade partial tear, high-grade partial tear, or full thickness tear. Gluteus medius, gluteus minimus, tensor fascia lata, and iliopsoas muscle atrophy was scored using a standard scale. Insertion site of tendinopathy and location of muscle atrophy were assessed. Descriptive and statistical analysis was performed. RESULTS: There was increasing greater trochanteric bursitis and gluteus medius and minimus tendinopathy and atrophy with advancing age with moderate to strong positive associations (p < 0.0001) for age and tendinopathy, age and atrophy, bursitis and tendinopathy, and tendinopathy and atrophy for the gluteus medius and minimus. There is a weak positive association (p < 0.0001) for age and tensor fascia lata atrophy, and no statistically significant association between age and tendinopathy or between age and atrophy for the iliopsoas. Fisher's exact tests were statistically significant (p < 0.0001) for insertion site of tendon pathology and location of muscle atrophy for the gluteus medius. CONCLUSIONS: Gluteus medius and minimus tendon pathology and muscle atrophy increase with advancing age with progression of tendinosis to low-grade tendon tears to high-grade tendon tears. There is an associated progression in atrophy of these muscles, which may be important in fall-related hip fractures.

Shoulder, hip, and knee arthrography needle placement using fluoroscopic guidance: practice patterns of musculoskeletal radiologists in North America.

OBJECTIVE: The aim of this study was to evaluate the range of techniques used by radiologists performing shoulder, hip, and knee arthrography using fluoroscopic guidance. MATERIALS AND METHODS: Questionnaires on shoulder, hip, and knee arthrography were distributed to radiologists at a national radiology meeting. We enquired regarding years of experience, preferred approaches, needle gauge, gadolinium dilution, and volume injected. For each approach, the radiologist was asked their starting and end needle position based on a numbered and lettered grid superimposed on a radiograph. RESULTS: Sixty-eight questionnaires were returned. Sixty-eight radiologists performed shoulder and hip arthrography, and 65 performed knee arthrograms. Mean experience was 13.5 and 12.8 years, respectively. For magnetic resonance arthrography, a gadolinium dilution of 1/200 was used by 69-71%. For shoulder arthrography, an anterior approach was preferred by 65/68 (96%). The most common site of needle end position, for anterior and posterior approaches, was immediately lateral to the humeral cortex. A 22-gauge needle was used by 46/66 (70%). Mean injected volume was 12.7 ml (5-30). For hip arthrography, an anterior approach was preferred by 51/68 (75%). The most common site of needle end position, for anterior and lateral approaches, was along the lateral femoral head/neck junction. A 22-gauge needle was used by 53/68 (78%). Mean injected volume was 11.5 ml (5-20). For knee arthrography, a lateral approach was preferred by 41/64 (64%). The most common site of needle end position, for lateral and medial approaches, was mid-patellofemoral joint level. A 22-gauge needle was used by 36/65 (56%). Mean injected volume was 28.2 ml (5-60). CONCLUSION: Arthrographic approaches for the shoulder, hip, and knee vary among radiologists over a wide range of experience levels.

Musculoskeletal biopsy: utilization and provider distribution in the United States Medicare population.

RATIONALE AND OBJECTIVES: Physicians from many specialties perform musculoskeletal biopsy. Using the Medicare database, we sought to determine which specialties represent the physicians who are performing the majority. MATERIALS AND METHODS: Using the CMS physician supplier procedure summary master file for 1996-2003, we extracted all claims for biopsy procedure codes (including marrow aspiration, muscle biopsy, percutaneous bone biopsy, and open surgical biopsy) categorized by provider specialty, and we analyzed procedure volumes. RESULTS: Since 1996, the rate of utilization of percutaneous bone biopsy has remained stable. In 2003, marrow aspiration was most commonly performed by hematology/oncology (80,038, 57%), followed by medical oncology (23,428, 17%); radiologists performed 755 (0.5%). Muscle biopsies were predominantly performed by radiologists (4,761, 40%), followed by neurosurgery (591, 5%). Percutaneous bone biopsy was mostly performed by radiologists (14,830, 53%), but orthopedic surgeons, neurosurgeons, and hematology/oncology specialists performed a large minority (6,879, 2,296, and 1,048 respectively; in aggregate, 37%). From 1996 to 2003, radiologists performed 71% more muscle biopsies (2,788 to 4,761) and 60% more percutaneous bone biopsies (9,259 to 14,830). Although most specialties are performing fewer percutaneous bone biopsies (e.g., oncologists: 7,217 to 1,048, -85%), orthopedic surgeons are performing 247% more (1,983 to 6,879) and neurosurgeons are performing 2,343% more (94 to 2,296). CONCLUSION: Excluding marrow aspiration, radiologists perform the majority of percutaneous bone biopsies, and the volume is increasing in the U.S. Medicare population. The overall volume has remained relatively stable from 1996 to 2003; although medical specialties are performing fewer, the volume performed by surgeons is increasing rapidly.

MRI in the diagnosis of cartilage injury in the wrist.

OBJECTIVE: Our purpose was to evaluate the accuracy of MRI in identifying articular cartilage abnormalities in the distal radius, scaphoid, lunate, and triquetrum of patients with wrist pain. MATERIALS AND METHODS: Eighty-six MRI examinations of the wrist in 85 patients (41 indirect MR arthrograms and 45 unenhanced [nonarthrographic] MR images) were evaluated. The study population consisted of 47 male (54.7%) and 38 female (45.3%) patients with an average age of 37.5 years (range, 7-62 years). Three experienced musculoskeletal radiologists who were unaware of surgical findings retrospectively evaluated the MRI examinations for cartilage abnormalities in the distal radius, scaphoid, lunate, and triquetrum. All patients underwent arthroscopy of the radiocarpal joint with inspection of the articular surfaces of the distal radius, scaphoid, lunate, and triquetrum. The articular cartilage was evaluated on the basis of the 5-point scale of the Outerbridge classification system. RESULTS: When at least two of the three radiologists had concordant interpretations, sensitivity for abnormalities in the distal radius was 27%; the scaphoid, 31%; the lunate, 41%; and the triquetrum, 18%. Specificity for the distal radius was 91%; the scaphoid, 90%; the lunate, 75%; and the triquetrum, 93%. Weighted kappa values among the three observers showed only fair agreement (0.279-0.360). High-grade more extensive cartilage lesions were no more accurately identified than low-grade lesions. Indirect MR arthrograms were not statistically more sensitive, specific, or accurate than unenhanced studies. No bone was more frequently or less frequently graded correctly or incorrectly with statistical significance. The variables of sex, age, and the presence of multiple bones with lesions did not affect accuracy. CONCLUSION: Our findings suggest that MRI of the wrist with the techniques described is not adequately sensitive or accurate for diagnosing cartilage defects in the distal radius, scaphoid, lunate, or triquetrum.

Internal derangement of the wrist: indirect MR arthrography versus unenhanced MR imaging.

PURPOSE: To compare indirect magnetic resonance (MR) arthrography with unenhanced MR imaging of the wrist for evaluation of the central disk of the triangular fibrocartilage complex (TFCC) and the scapholunate and lunotriquetral interosseous ligaments. MATERIALS AND METHODS: Eighty-six wrists were evaluated at MR imaging (41 indirect MR arthrography and 45 unenhanced MR imaging examinations). Three musculoskeletal radiologists independently evaluated the central disk of the TFCC and scapholunate and lunotriquetral ligaments and compared the results with those of wrist arthroscopy. Sensitivity and specificity were calculated for each of the readers, and the means were obtained. Sensitivities and specificities were compared with the Student t test. RESULTS: Thirty-three tears of the central disk of the TFCC and 13 scapholunate and 18 lunotriquetral ligament tears were identified at arthroscopy. Sensitivities and specificities were 54%-73% and 83%-91%, respectively, in the evaluation of the central disk of the TFCC, with no significant difference between indirect MR arthrography (P =.666) and unenhanced MR imaging (P =.559). Sensitivities and specificities in the evaluation of the scapholunate ligament were 38%-69% and 75%-99%, respectively, with a significant improvement in sensitivity at indirect MR arthrography (P =.017) and no significant difference in specificity (P =.876). Sensitivities in the evaluation of the lunotriquetral ligament were poor, 0%-22%, though the specificities were 88%-99%, with no significant difference between indirect MR arthrography and unenhanced MR imaging (P =.592 and P =.354, respectively, for sensitivity and specificity. CONCLUSION: Indirect MR arthrography significantly improves sensitivity in the evaluation of the scapholunate ligament when compared with unenhanced MR imaging of the wrist but does not significantly improve the ability to evaluate the central disk of the TFCC or the lunotriquetral ligament.

Limitations of MR imaging in the diagnosis of peripheral tears of the triangular fibrocartilage of the wrist.

OBJECTIVE: The treatment of peripheral tears of the triangular fibrocartilage complex is radically different from the more typical central, degenerative tears. To our knowledge, no reports in the imaging literature specifically evaluate tears of the ulnar attachment of the triangular fibrocartilage complex. We evaluated the accuracy of MR imaging in these patients. MATERIALS AND METHODS: Eighty-six MR imaging examinations of the wrist (41 indirect MR arthrograms and 45 unenhanced MR images) were evaluated: 20 wrists with surgically confirmed peripheral triangular fibrocartilage complex tears and 66 wrists with surgically documented normal ulnar attachment. These cases were evaluated by three experienced musculoskeletal radiologists, who were unaware of the surgical findings, to assess the presence of peripheral triangular fibrocartilage complex tears or fluid signal at the ulnar attachment of the triangular fibrocartilage complex. RESULTS: The sensitivity for evaluation of the peripheral triangular fibrocartilage complex tear was 17%, with a specificity of 79% and an accuracy of 64%. High signal intensity at the ulnar insertion of the triangular fibrocartilage complex as a marker for tear showed a sensitivity of 42%, a specificity of 63%, and an accuracy of 55%. Weighted kappa values revealed only fair agreement among the three observers. CONCLUSION: MR imaging does not adequately reveal the peripheral attachment of the triangular fibrocartilage complex.