Utility of 3D reformatted images in the diagnosis of sternoclavicular joint injury

INTRODUCTION

The purpose of this study is to evaluate the ability of musculoskeletal radiologists to diagnose sternoclavicular joint injuries in the standard CT imaging planes compared to 3D volume rendered images to define the most accurate plane to improve prospective diagnosis.

MATERIALS AND METHODS

A retrospective query of our institutional database was performed. Twenty-six patients with a diagnosis of sternoclavicular joint injury, who had been evaluated with CT and treated by orthopedic surgery, and 30 control patients who did not have a sternoclavicular joint injury were included for analysis. Two blinded radiologists with specialty training in musculoskeletal radiology independently reviewed axial, coronal, sagittal, and 3D reformatted CT images and documented whether injury was present or not present.

RESULTS

Accuracy was good for both radiologists on all views. It was lowest on the sagittal view for both readers. Accuracy was highest for the 3D view. When comparing the accuracy of the four views for each radiologist, there was a significant difference for Radiologist A, whose 3D images were more accurate compared to the axial and sagittal views. There was no significant difference for Radiologist B. There was good inter-reader agreement, which was highest on the 3D images.

CONCLUSION

3D volume renderings of the sternoclavicular joints have the potential to improve radiologist accuracy for detection of sternoclavicular joint injury/dislocation in the setting of chest well trauma, which could decrease instances of missed or delayed diagnosis.

Clinical value of weight-bearing CT and radiographs for detecting patellofemoral cartilage visualized by MRI in the MOST study

OBJECTIVE

The patellofemoral joint is frequently affected by osteoarthritis (PFOA) and is incompletely imaged on radiographs (XR). Weight-Bearing CT (WBCT) could offer advantages for visualization. This study determined the sensitivity, specificity, and accuracy of axial WBCT and lateral XR for detection of PFOA features in comparison with cartilage damage on MRI.

DESIGN

A convenience sample of 60 right knees from the MOST cohort were analyzed. WBCT and XR were read for OARSI JSN score and MRI for MOAKS cartilage score by two experienced musculoskeletal radiologists blinded to participant. Using MOAKS scoring on MRI (referent standard), the sensitivity, specificity and accuracy of patellofemoral OARSI JSN scores based on WBCT and XR were compared.

RESULTS

The mean ± SD age and BMI for the participants included (66.7% women) were 67.6 ± 9.8 years and 30.0 ± 5.3 kg/m² respectively. WBCT demonstrated significantly greater sensitivity (0.85-0.97 on WBCT vs 0.47-0.57 on XR) and accuracy (0.85-0.92 on WBCT vs 0.48-0.57 on XR) for all parameters except lateral full-thickness cartilage loss (McNemar's test p-values all <0.001). There was moderate-to-strong and low-to-moderate agreement between PFOA findings on WBCT and XR, respectively, and semi-quantitative scores of PF cartilage on MRI. Inter-rater reliability for XR JSN [weighted kappa = 0.83 (0.64, 1.0)], WBCT JSN [kappa = 0.60 (0.48, 0.72)] and MRI MOAKS-CM [kappa = 0.70 (0.61, 0.79)] readings were good.

CONCLUSION

WBCT demonstrates significantly greater sensitivity and accuracy than radiographs for identification of PFOA. Given the same Relative Radiation Level as XR and improved visualization, WBCT holds promise to improve understanding of the weight-bearing patellofemoral joint.

Artificial Intelligence Algorithm Improves Radiologist Performance in Skeletal Age Assessment: A Prospective Multicenter Randomized Controlled Trial

Background Previous studies suggest that use of artificial intelligence (AI) algorithms as diagnostic aids may improve the quality of skeletal age assessment, though these studies lack evidence from clinical practice. Purpose To compare the accuracy and interpretation time of skeletal age assessment on hand radiograph examinations with and without the use of an AI algorithm as a diagnostic aid. Materials and Methods In this prospective randomized controlled trial, the accuracy of skeletal age assessment on hand radiograph examinations was performed with (n = 792) and without (n = 739) the AI algorithm as a diagnostic aid. For examinations with the AI algorithm, the radiologist was shown the AI interpretation as part of their routine clinical work and was permitted to accept or modify it. Hand radiographs were interpreted by 93 radiologists from six centers. The primary efficacy outcome was the mean absolute difference between the skeletal age dictated into the radiologists' signed report and the average interpretation of a panel of four radiologists not using a diagnostic aid. The secondary outcome was the interpretation time. A linear mixed-effects regression model with random center- and radiologist-level effects was used to compare the two experimental groups. Results Overall mean absolute difference was lower when radiologists used the AI algorithm compared with when they did not (5.36 months vs 5.95 months; P = .04). The proportions at which the absolute difference exceeded 12 months (9.3% vs 13.0%, P = .02) and 24 months (0.5% vs 1.8%, P = .02) were lower with the AI algorithm than without it. Median radiologist interpretation time was lower with the AI algorithm than without it (102 seconds vs 142 seconds, P = .001). Conclusion Use of an artificial intelligence algorithm improved skeletal age assessment accuracy and reduced interpretation times for radiologists, although differences were observed between centers. Clinical trial registration no. NCT03530098 © RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Rubin in this issue.

Cervical CT-Dependent Diagnosis of Crowned Dens Syndrome in Calcium Pyrophosphate Dihydrate Crystal Deposition Disease

PURPOSE

The purpose of this study is to assess the presence of crowned dens syndrome in patients with calcium pyrophosphate disease. We report 34 patients with crowned dens syndrome in one of the largest series from a single tertiary medical center in North America.

METHODS

A retrospective chart review was conducted at the University of Kansas Medical Center from November 1, 2005-November 1, 2017. A total of 191 patients with calcium pyrophosphate disease were identified. The available cervical computed tomography scans were analyzed by a musculoskeletal radiologist for the presence of periodontoid calcifications and erosions.

RESULTS

Of the 191 patients with calcium pyrophosphate disease, 57 had cervical computed tomography scans; 34 of them (34/57, 59.64%) had periodontoid calcifications. Only 12/34 patients were formally evaluated and diagnosed by rheumatologists with crowned dens syndrome. Twenty-two of 34 were either not seen by a rheumatologist or were not diagnosed with crowned dens syndrome. The median age was 78.5 years, with 73.52% over 70 years old; 24/34 (70.58%) were female; 17/34 patients (50%) were symptomatic; 28/34 (82.35%) had additional sites of chondrocalcinosis on available radiographs; 8 (28.57%) had 3 or more sites of chondrocalcinosis in typical calcium pyrophosphate disease locations. Six patients did not have any radiographs.

CONCLUSION

Crowned dens syndrome is an under-recognized entity that should be considered in elderly patients with neck pain in the setting of calcium pyrophosphate disease. Our data demonstrated a high percentage (about 60%) of patients with calcium pyrophosphate disease who had cervical computed tomography findings consistent with crowned dens syndrome. This underscores the importance of performing cervical computed tomography when evaluating patients with neck pain and calcium pyrophosphate disease.

Association between imaging findings and microbiological findings for image‑guided biopsies for spine infections

BACKGROUND

In the evaluation of spondylodiscitis, a number of factors are thought to contribute to the positive‑predictive‑value of spine biopsy including biopsy technique and equipment, number of sample obtained, timing of antibiotics, imaging findings and lab values. The purpose of this study was to examine which technical, magnetic resonance imaging (MRI), laboratory and clinical findings are most frequently associated with positive cultures or histopathology among patients receiving spine biopsies with a focus on the association of imaging findings and positive cultures/histopathology.

METHODS

Following International Review Board approval, we retrospectively reviewed a consecutive series of spine biopsies performed at our institution over a 28-month period in patients who received spine biopsies for radiographically and clinically suspicious spondylodiscitis. All patients underwent MRI prior to biopsy. Patient charts were reviewed for the following data: erythrocyte sedimentation rate (ESR), C‑reactive protein (CRP), and white blood cell level at the time of biopsy, prior back surgeries, timing of last dose of antibiotics prior to biopsy, and length of antibiotic therapy prior to biopsy. We also reviewed procedure notes from the biopsy and collected the following data: technique (transpedicular versus parapedicular), equipment used, type of imaging guidance (CT versus fluoroscopy) and number of samples obtained. MRI studies were evaluated by two radiologists and rated on an ordinal 1-4 scale based on increasing suspicion for spondylodiscitis (1 lowest, 4 highest). All categorical variables were compared using chi‑squared tests. All continuous variables were compared using Student's t‑tests.

RESULTS

Seventy‑seven patients are included in this study. Overall, 62% (48/77) were positive for infection on pathological and/or microbiological studies. There was a significant association between radiological index of suspicion and biopsy positivity as 0% (0/3) with index of 1, 31% (4/13) with an index of 2, 59% (10/17) with an index of 3 and 83% (30/36) with an index of 4 had positive pathology (P=0.001). Biopsy approach, type of imaging guidance, CBC, and number of passes were not associated with biopsy positivity. Elevated CRP was associated with biopsy positivity (P=0.002) while elevated ESR was not (P=0.12). On multivariate analysis adjusting for the MRI scale, ESR and CRP, increasing degree of suspicion on MRI was independently associated with biopsy positivity (P<0.01) while CRP and ESR were not independently associated with biopsy positivity.

CONCLUSIONS

In this study of 77 patients receiving spinal biopsy for diagnosis of spondylodiscitis, a high index of suspicion based on MRI imaging was strongly associated with positive biopsy samples on culture and/or histopathology. In addition, imaging findings were independently associated with positive biopsy findings while elevated CRP and ESR were not. These findings suggest that MRI findings should be used to guide practitioners in their pursuit of a causative organism among patients with suspected discitis‑osteomyelitis.

Neuromuscular choristoma: characteristic magnetic resonance imaging findings and association with post-biopsy fibromatosis

OBJECTIVE

To describe imaging characteristics of neuromuscular choristomas (NMC) and to differentiate them from fibrolipomatous hamartomas (FLH).

MATERIALS AND METHODS

Clinical and imaging characteristics of six patients with biopsy-proven NMC and six patients with FLH were reviewed by musculoskeletal, a pediatric, and two in-training radiologists with a literature review to define typical magnetic resonance imaging features by consensus. Five radiology trainees blinded to cases and naive to the diagnosis of NMC and a musculoskeletal-trained radiologist rated each lesion as having more than or less than 50% intralesional fat, as well as an overall impression using axial T1 images. Sensitivity, specificity, accuracy, and interobserver agreement kappa were determined.

RESULTS

Typical features of NMC include smoothly tapering, fusiform enlargement of the sciatic nerve or brachial plexus elements with T1 and T2 signal characteristics closely following those of muscle. Longitudinal bands of intervening low T1 and T2 signal were often present and likely corresponded to fibrous tissue by pathology. Four of five patients with long-term follow-up (80%) developed aggressive fibromatosis after percutaneous or surgical biopsy. Nerve fascicle thickening often resulted in a "coaxial cable" appearance similar to classic FLH, however, using a cutoff of <50% intralesional fat allowed for differentiation with 100% sensitivity by all reviewers and 100% specificity when all imaging features were utilized for impressions. Agreement was excellent with all differentiating methods (kappa 0.861-1.0).

CONCLUSIONS

NMC can be confidently differentiated from FLH and malignancies using characteristic imaging and clinical features. When a diagnosis is made, biopsy should be avoided given frequent complication by aggressive fibromatosis.