Accuracy and Consistency of 3D Magnetic Resonance Imaging Is Comparable to 3D Computed Tomography in Assessing Glenohumeral Instability: A Systematic Review

PURPOSE

The purpose of this study is to compare the accuracy of 3D magnetic resonance imaging (MRI) with that of 3D CT in evaluating glenoid bone loss (GBL).

METHODS

This review aligned with PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. PubMed, the Cochrane Library, Embase, and Web of Science were obtained from data inception to Aug 28, 2023. The search term "glenoid bone loss" was extracted and analyzed via stringent inclusion and exclusion criteria. The Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) combined with the Quality Assessment of Diagnostic Accuracy Studies-Comparative (QUADAS-C) to assess the heterogeneity of included studies.

RESULTS

A total of 1589 related studies were retrieved, and 10 studies were finally included, of which a total of 143 shoulders were evaluated. The index test in QUADAS-C was low risk in 9 studies. 3D MRI measurements of GBL were primarily best-fit circles (n = 9). In both clinical and cadaveric studies, the mean percentages of GBL measured by 3D MRI were 0.38% to 2.19% and 0.25% to 6.1% when compared with 3D CT and standard reference values, respectively. ICC agreement greater than 0.9 between GBL percentages measured by 3D CT and 3D MRI. 3D MRI can also accurately measure glenoid width, glenoid height, humeral head width and height, etc. 3D MRI reconstruction time was similar to that of 3D CT, which was mainly 10-15 min.

CONCLUSIONS

In both clinical and cadaveric studies, compared to 3D CT, 3D MRI is accurate and consistent in assessing glenohumeral bone, especially in measuring GBL, and the reconstruction time of 3D MRI is similar to 3D CT.

Cinematic rendering of non-traumatic thoracic aorta emergencies: a new look at an old problem

Non-traumatic thoracic aorta emergencies are acute conditions associated with substantial morbidity and mortality. In the emergency setting, timely detection of aortic injury through radiological imaging is crucial for prompt treatment planning and favorable patient outcomes. 3D cinematic rendering (CR), a novel rendering algorithm for computed tomography (CT) image processing, allows for life-like visualization of spatial details and contours of highly complex anatomic structures such as the thoracic aorta and its vessels, generating a photorealistic view that not just adds to diagnostic confidence, but is especially useful for non-radiologists, including surgeons and emergency medicine physicians. In this pictorial review, we demonstrate the utility of CR in the setting of non-traumatic thoracic aorta emergencies through 10 cases that were processed at a standalone 3D CR station at the time of presentation, including its role in diagnosis and management.

The True-Blue technique: Enhanced chondral assessment of the chest wall

There is a growing understanding and identification of costal cartilage injuries, however, diagnosis of these injuries remains difficult. We present a novel radiodensity based coloring technique, termed the True-Blue technique, to manipulate 3D CT imaging and more accurately diagnose costochondral injuries.

Atlantoaxial Instability with Persistent Second Intersegmental Artery

Understanding the anatomy of the vertebral artery is essential while manipulating the craniovertebral joint during surgery. Its anomalous course in congenital atlantoaxial dislocation makes it more vulnerable to injury. Preoperative dedicated computed tomography (CT) angiography helps identify the artery's position and plan for surgical procedure. A 13-year-boy presented with neck pain and spastic quadriparesis for 1 year. Radiological imaging of the craniovertebral junction revealed atlantoaxial instability with basilar invagination. His CT angiography of neck and brain vessels revealed an anomalous course of the vertebral artery due to a persistent second intersegment artery. He underwent posterior atlantoaxial fixation after mobilization of the vertebral artery. His clinical condition significantly improved after surgery. We report a case of management of an atlanto axial dislocation with persistent second intersegment artery and describe the role of vertebral artery mobilization during surgery.

Post-operative Baseplate Radiographic Evaluation Using Routine pre-Operative CT

Background

There is limited data evaluating post-operative component position and fixation in reverse shoulder arthroplasty (RSA). Therefore, the purpose of this study was to evaluate baseplate position and fixation using routine pre-operative CT and post-operative radiographs.

Methods

A retrospective analysis of a series consecutive patient who underwent primary RSA was performed. Pre-operative and post-operative glenoid retroversion and inclination were measured using radiographs aligned with projection silhouettes of 3D scapula models in Mimics software. Baseplate retroversion and inclination were measured followed by evaluating for the presence of radiolucent lines (RLLs).

Results

Twenty-four patients met inclusion criteria. The average age was 73.4 ± 10.7 years (range, 45-89 years). Radiographic follow-up was 3.4 ± 1.3 years. Post-operative glenoid baseplate retroversion was 2 ± 10 degrees (range, 30 to -9). Post-operative glenoid baseplate inclination was 3.8 ± 9.1 (range, -13 to 19). Five (21%) RSAs had baseplate retroversion >10 degrees. Follow-up radiographs revealed no RLLs around the baseplate, central post, or peripheral screws in any patient.

Conclusions

Pre-operative CT imaging enabled evaluation of baseplate component placement and fixation on post-operative radiographs. Baseplate version was within 10 degrees of neutral in 79% (19/24) of patients. No RLLs or loss of fixation were found in any cases.

Level of Evidence

Level IV: Diagnostic Study.

Evaluation of extensive inflammatory conditions of the bowel using three-dimensional CT cinematic rendering: focus on inflammatory bowel disease

Inflammatory conditions that affect long segments of bowel and/or the mesentery and mesenteric vasculature are a common cause of emergency department visits and evaluation by cross-sectional imaging. Inflammatory bowel disease, specifically Crohn disease and ulcerative colitis, can be unsuspected at presentation and may only be eventually diagnosed based on initial imaging findings. Traditional 2D axial reconstructions and multi-planar reformations can be limited in their ability to globally assess the extent of disease. 3D methods such as volume rendering (VR) are often used as adjunctive means of visualizing the pathology in such patients. Recently, a novel technique known as cinematic rendering (CR) has emerged, utilizing advanced lighting models and ray tracing to simulate photon interactions with tissues, resulting in realistic shadows and enhanced surface detail compared to VR. Generating CR images from select presets takes an experienced radiologist approximately 5 min, meaning that the technique can be incorporated into meaningful emergency department workflows. Given the apparent advantages of CR, we highlight its application in a series of cases in which patients had inflammatory conditions that affected long segments of bowel and/or involved the mesentery, particularly those patients with inflammatory bowel disease, but also including patients with mesenteric venous thrombosis and lymphedema. Those conditions included inflammatory bowel disease, mesenteric venous thrombosis, and bowel lymphedema. We present examples of those conditions in this pictorial essay and describe the potential of CR to visualize key findings. As CR exhibits possible advantages, further studies are warranted to support its broader clinical adoption and assess its efficacy in diagnosing and guiding managing of inflammatory conditions in emergency settings.

Current Concepts in the Measurement of Glenohumeral Bone Loss

PURPOSE

The extent of glenohumeral bone loss seen in anterior shoulder dislocations plays a major role in guiding surgical management of these patients. The need for accurate and reliable preoperative assessment of bone loss on imaging studies is therefore of paramount importance to orthopedic surgeons. This article will focus on the tools that are available to clinicians for quantifying glenoid bone loss with a focus on emerging trends and research in order to describe current practices.

RECENT FINDINGS

Recent evidence supports the use of 3D CT as the most optimal method for quantifying bone loss on the glenoid and humerus. New trends in the use of 3D and ZTE MRI represent exciting alternatives to CT imaging, although they are not widely used and require further investigation. Contemporary thinking surrounding the glenoid track concept and the symbiotic relationship between glenoid and humeral bone loss on shoulder stability has transformed our understanding of these lesions and has inspired a new focus of study for radiologists and orthopedist alike. Although a number of different advanced imaging modalities are utilized to detect and quantify glenohumeral bone loss in practice, the current literature supports 3D CT imaging to provide the most reliable and accurate assessments. The emergence of the glenoid track concept for glenoid and humeral head bone loss has inspired a new area of study for researchers that presents exciting opportunities for the development of a deeper understanding of glenohumeral instability in the future. Ultimately, however, the heterogeneity of literature, which speaks to the diverse practices that exist across the world, limits any firm conclusions from being drawn.

Distal Clavicle Autograft Versus Traditional and Congruent Arc Latarjet Procedures: A Comparison of Surface Area and Glenoid Apposition With 3-Dimensional Computed Tomography and 3-Dimensional Magnetic Resonance Imaging

BACKGROUND

Limited studies have compared graft-glenoid apposition and glenoid augmentation area between the Latarjet procedure and distal clavicle graft in glenohumeral stabilization. Additionally, preoperative planning is typically performed using computed tomography (CT), and few studies have used 3-dimensional (3D) magnetic resonance imaging (MRI) reformations to assess graft dimensions.

PURPOSE

The purpose of this study was 2-fold: (1) to compare bony apposition, glenoid augmentation, and graft width among coracoid and distal clavicle bony augmentation techniques and (2) to determine the viability of 3D MRI to assess bone graft dimensions.

STUDY DESIGN

Controlled laboratory study.

METHODS

A total of 24 patients with recurrent glenohumeral instability and bone loss were included in this study. 3D CT and 3D MRI reformations were utilized to measure pertinent dimensions for 5 orientations of coracoid and distal clavicle autografts: (1) standard Latarjet procedure (SLJ), (2) congruent arc Latarjet procedure (CLJ), (3) distal clavicle attached by its posterior surface (DCP), (4) distal clavicle attached by its inferior surface (DCI), and (5) distal clavicle attached by its resected end (DCR). Glenoid augmentation was defined as the graft surface area contributing to the glenoid. Bone-on-bone apposition was defined as the graft-glenoid contact area for bone healing potential, and graft width was pertinent for fixation technique. Glenoid bone loss ranged from 0% to 34%. Paired t tests were used to compare graft sizes between patients and compare 3D CT versus 3D MRI measurements.

RESULTS

The CLJ had the largest graft surface area (mean, 318.41 ± 74.44 mm²), while the SLJ displayed the most bone-on-bone apposition (mean, 318.41 ± 74.44 mm²). The DCI had the largest graft width (mean, 20.62 ± 3.93 mm). Paired t tests revealed no significant differences between the Latarjet techniques, whereas distal clavicle grafts varied significantly with orientation. All 3D CT and 3D MRI measurements were within 1 mm of each other, and only 2 demonstrated a statistically significant difference (coracoid width: 13.03 vs 13.98 mm, respectively [P = .010]; distal clavicle thickness: 9.69 vs 10.77 mm, respectively [P = .002]). 3D CT and 3D MRI measurements demonstrated a strong positive correlation (r > 0.6 and P < .001 for all dimensions).

CONCLUSION

Glenoid augmentation, bony apposition, and graft width varied with coracoid or distal clavicle graft type and orientation. Differences between 3D CT and 3D MRI were small and likely not clinically significant.

CLINICAL RELEVANCE

3D MRI is a viable method for preoperative planning and graft selection in glenoid bone loss.

What are the resection accuracy and guide-fitting errors associated with 3D-printed, patient-specific resection guides for bone tumour resections?

AIMS

This study aimed to analyze the accuracy and errors associated with 3D-printed, patient-specific resection guides (3DP-PSRGs) used for bone tumour resection.

METHODS

We retrospectively reviewed 29 bone tumour resections that used 3DP-PSRGs based on 3D CT and 3D MRI. We evaluated the resection amount errors and resection margin errors relative to the preoperative plans. Guide-fitting errors and guide distortion were evaluated intraoperatively and one month postoperatively, respectively. We categorized each of these error types into three grades (grade 1, < 1 mm; grade 2, 1 to 3 mm; and grade 3, > 3 mm) to evaluate the overall accuracy.

RESULTS

The maximum resection amount error was 2 mm. Out of 29 resection amount errors, 15 (51.7%) were grade 1 errors and 14 (38.3%) were grade 2 errors. Complex resections were associated with higher-grade resection amount errors (p < 0.001). The actual resection margins correlated significantly with the planned margins; however, there were some discrepancies. The maximum guide-fitting error was 3 mm. There were 22 (75.9%), five (17.2%), and two (6.9%) grade 1, 2, and 3 guide-fitting errors, respectively. There was no significant association between complex resection and fitting error grades. The guide distortion after one month in all patients was rated as grade 1.

CONCLUSION

In terms of the accurate resection amount according to the preoperative planning, 3DP-PSRGs can be a viable option for bone tumour resection. However, 3DP-PSRG use may be associated with resection margin length discrepancies relative to the planned margins. Such discrepancies should be considered when determining surgical margins. Therefore, a thorough evaluation of the preoperative imaging and surgical planning is still required, even if 3DP-PSRGs are to be used.Cite this article: Bone Joint J 2023;105-B(2):190-197.

Fracture through tracking pin sites following a robotic-assisted total knee arthroplasty

The use of computer-navigated, robotic-assisted total knee arthroplasty (TKA) is expanding as the result of widespread efforts to improve the precision of implant placement and to restore mechanical, anatomic, and kinematic alignment. The procedure requires placement of femoral and tibial pins to mount the robotic tracking system. Fractures through tracking pin sites following robotic-assisted TKA are rare; only 30 cases have been reported to our knowledge. It is probable that this complication will become more frequent as the use of robotic-assisted TKAs continues to grow. We report the case of a 67-year-old female who suffered a fracture of the femur through two pin sites 3 months after a robotic-assisted TKA. We believe that this report is the first case in the radiologic literature. Our case demonstrates some of the difficulties of recognizing this complication and raises questions about appropriate post-operative imaging.

Plain Radiographs Underestimate Varus Deformity of the Tibial Plafond

Understanding plain radiograph in association with 3-dimensional (3D) morphology of the ankle is essential for treatment about varus ankle osteoarthritis (OA). The aims of this study were to investigate whether the alignment of the tibial plafond as determined on plain radiograph reflected the alignment of the tibial plafond on computed tomography (CT) in varus ankle OA and whether the alignment of the tibial plafond changed as the OA progressed. The 3D CT and plain radiographs from 101 ankles with varus ankle OA were analyzed and compared with 40 ankles in control group. The tibial plafond was assessed in the coronal and sagittal planes using 3D CT. The medial angle between the vertical line and the tibial plafond was measured on 3 different coronal plane CT images which was anterior, middle and posterior area of the tibial plafond. The medial distal tibial angle on plain radiograph reflected the posterior area of the tibial plafond on CT. The amount of varus angulation on CT was larger in anterior and middle area of the tibial plafond than the posterior area. There was a difference in the degree of varus of the tibial plafond between control group and OA patients; however, there was no difference among patients in different stages of varus ankle OA. Weightbearing plain radiographs underestimate the varus deformity in anterior and middle area of the tibial plafond and there is no significant difference in deformity of the tibial plafond among patients in different stages of varus ankle OA.

Role of 3D reconstructed images in primary pulmonary vein stenosis

Primary pulmonary vein stenosis (PPVS) is a rare congenital cardiac anomaly. The surgical management of PPVS requires anatomical delineation of pulmonary venous drainage along with exact localization of the stenosis. Since echocardiography cannot visualize the whole length of pulmonary veins (PV), computed tomography (CT) pulmonary venography becomes necessary. Three-dimensional (3D) reconstruction helps in planning PV augmentation and for prognostication. Here, we present 3D reconstructed CT pulmonary venography images of a 3-month-old child who presented to us with PPVS.

Three-dimensional computed tomography-guided excision of an intrathoracic giant thymoma with elongated thymic vessels

We report the case of an intrathoracic giant thymoma with elongated thymic vessels, which was successfully resected under three-dimensional computed tomography guidance. A large, left-sided intrathoracic mass was incidentally found in a 41-year-old woman during a routine work-up for uterine cancer. Six vessels were noted arising from the tumor, five of which were connected to the anterosuperior mediastinum. The vasculature suggested that the tumor originated from the thymus and grew into the left pleural cavity, which pulled and elongated the associated vessels. Preoperative computed tomography imaging demonstrated that these vessels were located behind the tumor, which increased the risk for catastrophic intraoperative bleeding. We created a detailed surgical plan using our preoperative computed tomography data and successfully excised the tumor using intraoperative three-dimensional computed tomography guidance. Histopathological examination revealed a type AB thymoma without capsular invasion. This case highlighted the role of preoperative planning and intraoperative imaging in resecting an intrathoracic giant thymoma safely. In the video, we demonstrate how we performed the procedure under three-dimensional navigation.

How Important is the Tunnel Position in Outcomes Post-ACL Reconstruction: A 3D CT-Based Study

BACKGROUND

Drilling the femoral and tibial tunnels at their anatomical locations are critical for good outcomes and involve seeing the footprints well. We intended to compare two techniques of drilling the tunnels and the patient-reported outcomes and knee stability of patients undergoing single bundle ACL reconstruction using 3D CT to evaluate if the tunnels were anatomical or not.

MATERIALS AND METHODS

Sixty single bundle ACL reconstructions were analyzed, 30 each with Technique A and B. Pre-operative and after a minimum 27 month follow-up Lysholm, IKDC, Tegner score, hop test, and Lachman test were noted. 3D CT was done to classify femoral tunnels positions as being well placed, slightly or grossly misplaced and tibial tunnels as optimal or suboptimal and compared.

RESULTS

Sixty ACL reconstructions had full follow-up with a mean follow-up of 34 months. There was no significant difference between tunnel positions between the two techniques. Well-placed femoral tunnel had better Lysholm score (62.2 ± 16.2 v/s 48.5 ± 17.2, p 0.002) and IKDC score (62.5 ± 14.3 v/s 52.7 ± 15.1, p 0.012).). Those who had their surgeries within 3 months of their injury had better hop test (4.4 ± 0.9 v/s 3.9 ± 1, p 0.034) and IKDC scores (62.5 ± 15.8 v/s 33.2 ± 13.8, p 0.026) as compared to those that had surgery done after 3 months.

CONCLUSION

Tibial tunnel positions were optimal in most cases and did not differ between the two techniques. Well-placed femoral tunnels and surgeries done within 3 months of the injury produced best results.

Injury pattern simulation and mapping of complex tibial plateau fractures that involve the posterior plateau with three-dimensional computed tomography

Background

Tibial plateau fractures involving the posterior plateau (TPFIPs) are complex intra-articular fractures that are difficult to stabilize. Understanding the characteristics of these fractures together with the injury pattern is beneficial for surgeons to choose an optimal treatment strategy. However, the complicated morphology and injury patterns of TPFIPs are poorly characterized. The purpose of this retrospective study was to investigate the injury patterns and fracture characteristics of complex TPFs by applying three-dimensional (3D) simulation and fracture mapping methods.

Methods

In total, 171 TPFIPs were retrospectively reviewed, and the injury pattern was simulated and analyzed by applying a 3D method with Mimics software, which allowed matching of the fractured articular surfaces of the tibial plateau to the femoral condyle surface. The major articular fracture lines were mapped and then superimposed on a template. The tibial motion angle after fracture injury pattern simulation and the major fracture line angle were quantitatively analyzed, while the injury patterns and fracture characteristics were qualitatively analyzed.

Results

Four main injury patterns with distinctive fracture characteristics were observed in this study. In total, 72 TPFs exhibited extension as the pattern of injury with a split posterolateral fragment, and 61 fractures exhibited the flexion-internal rotation injury pattern; compression was the main feature of posterolateral fractures. Furthermore, 21 fractures exhibited the flexion-external rotation injury pattern, with a small posteromedial fragment, and 17 fractures exhibited the flexion-neutral injury pattern, with both parts of the posterior plateau fracture and anterior dislocation being observable. The major articular fracture line angles were significantly different between the four main injury patterns (85.92°, 46.79°, 148.26°, and 16.21°, median values, P<0.05). Two injury patterns, namely, flexion-internal rotation and flexion-external rotation, exhibited rotation in the axial plane (24.13°±8.33°, −15.13°±5.14°, P<0.05).

Conclusions

In this study, a method involving a simulated injury pattern was developed and combined with evaluations of fracture characteristics, including two-dimensional (2D) and 3D analyses, to comprehensively describe both the morphologies and injury patterns of TPFIPs.

High-resolution 3D abdominal segmentation with random patch network fusion

Deep learning for three dimensional (3D) abdominal organ segmentation on high-resolution computed tomography (CT) is a challenging topic, in part due to the limited memory provide by graphics processing units (GPU) and large number of parameters and in 3D fully convolutional networks (FCN). Two prevalent strategies, lower resolution with wider field of view and higher resolution with limited field of view, have been explored but have been presented with varying degrees of success. In this paper, we propose a novel patch-based network with random spatial initialization and statistical fusion on overlapping regions of interest (ROIs). We evaluate the proposed approach using three datasets consisting of 260 subjects with varying numbers of manual labels. Compared with the canonical "coarse-to-fine" baseline methods, the proposed method increases the performance on multi-organ segmentation from 0.799 to 0.856 in terms of mean DSC score (p-value < 0.01 with paired t-test). The effect of different numbers of patches is evaluated by increasing the depth of coverage (expected number of patches evaluated per voxel). In addition, our method outperforms other state-of-the-art methods in abdominal organ segmentation. In conclusion, the approach provides a memory-conservative framework to enable 3D segmentation on high-resolution CT. The approach is compatible with many base network structures, without substantially increasing the complexity during inference. Given a CT scan with at high resolution, a low-res section (left panel) is trained with multi-channel segmentation. The low-res part contains down-sampling and normalization in order to preserve the complete spatial information. Interpolation and random patch sampling (mid panel) is employed to collect patches. The high-dimensional probability maps are acquired (right panel) from integration of all patches on field of views.

3D mapping of scaphoid fractures and comminution

OBJECTIVE

Acute and subacute scaphoid fractures were assessed using 3D computer tomography (CT). The aims were to describe fracture morphology, to map fractures onto a 3D scaphoid model and to correlate this to scaphoid anatomy.

MATERIALS AND METHODS

A retrospective, multicentre database search was performed to identify CT studies of acute and subacute scaphoid fractures. CT scans of scaphoid fractures less than 6 weeks from time of injury were included in this retrospective, multicentre study. CTs were segmented and converted into three-dimensional models. Following virtual fracture reduction, fractures were mapped onto a three-dimensional scaphoid model.

RESULTS

Seventy-five CT scans were included. The median delay from injury to CT was 29 days. Most studies were in male patients (89%). Most fractures were comminuted (52%) or displaced (64%). A total of 73% of displaced fractures had concomitant comminution. Waist fractures had higher rates of comminution and displacement when compared with all other fractures. Comminution was located along the dorsal ridge and the volar scaphoid waist.

CONCLUSION

Our study is the first to describe acute fracture morphology using 3D CT and to correlate comminution and displacement to fracture types. The dorsal ridge and volar waist need prudent assessment, especially in waist fractures.

Does 3D CT reconstruction help the surgeon to preoperatively assess calcaneal fractures?

BACKGROUND

With the advent of 3D volume rendered CT scans, more information is potentially available to aid the surgeon in complex calcaneal fractures. The primary aim was to determine if there is a difference in inter-observer and intra-observer reliability of 3D CT reconstructions compared to 2D CTs of calcaneus fractures based on classification, identification of specific fracture characteristics and proposed treatment. It is hypothesized that the correlation will be greater between experienced surgeons and trainees when using 3D CT. As a secondary aim, the authors wished to investigate the surgeons' satisfaction with 3D CT.

PATIENTS AND METHODS

There were six raters, split into 2 groups: high and low surgical experience. Both 2D and 3D scans (10 patients) were reviewed by each rater by filling out a validated questionnaire. This sitting was over a period of six months and all scans were sent separately and randomly by our research coordinator. This process was repeated after a four-week break.

RESULTS

The overall evaluation and satisfaction of CT scans of calcaneal fractures was improved by the additional use of volume rendered 3D images. Inexperienced surgeons benefited more than experienced surgeons in terms of reliabilities with fracture configuration. 3D CTs were more likely to change the surgeons' operative strategy. The consistency of deciding on the fracture classification and satisfaction was more uniform between all the raters when 3D CT was used.

CONCLUSION

Inter-observer and intra-observer reliability of 3D CT reconstructions compared to 2D CTs of calcaneus fractures was greater based on classification, identification of specific fracture characteristics and proposed treatment. Inexperienced surgeons were more consistent when interpreting the scans using 3D CTs (improved inter-observer reliability) and were always more satisfied using the 3D CTs. As there is no extra cost or radiation exposure, we propose that 3D CTs may be valuable with preoperative teaching and planning.

Impact of two- and three-dimensional computed tomography use on intraobserver and interobserver reliabilities of pilon fracture classification and treatment recommendation

BACKGROUND

Two-dimensional (2D) and three-dimensional (3D) computed tomography (CT) have been increasingly used in various intra-articular fractures including pilon fracture. However, no study has investigated intraobserver and interobserver reliabilities of pilon fracture classification using 3D CT images.

HYPOTHESIS

(1) Intraobserver and interobserver agreements of fracture classifications and treatment recommendations will improve by using 2D CT images compared to using plain radiographs only; (2) agreements will improve by adding 3D CT images compared to adding 2D CT images; and (3) agreements of orthopedic residents rather than specialists will be influenced more by imaging modality.

MATERIALS AND METHODS

Ten orthopedic specialists and 10 residents completed a survey to classify the fractures according to the Rüedi-Allgöwer and AO/OTA classifications and to select treatment options using 25 pilon fracture images. The survey was conducted using plain radiographs, with 2D and 3D CT images introduced 3 and 6weeks later, respectively. Kappa coefficients were calculated to determine reliabilities.

RESULTS

Intraobserver reliabilities for fracture classifications in specialists significantly improved by using 2D images compared to using plain radiographs only. Addition of 3D CT did not significantly improve intraobserver reliabilities compared to those with 2D CT. Use of 2D CT images significantly improved overall interobserver agreement of both classifications, with the improvement being greater for residents. Use of 3D CT images did not improve the interobserver reliability of both classifications. Overall interobserver reliabilities for treatment recommendations did not significantly differ according to the imaging modality. However, interobserver agreement among residents significantly improved from slight agreement using radiographs only to fair agreement using 2D CT images.

DISCUSSION

Intraobserver and interobserver reliabilities of pilon fracture classification and treatment recommendations did not improve between using 3D and 2D CT. Using 2D CT images improved the intraobserver and interobserver reliabilities of the fracture classifications in specialists and the interobserver reliabilities of the fracture classifications and the treatment recommendations in residents.

LEVEL OF EVIDENCE

IV, case control study.

3D-MRI versus 3D-CT in the evaluation of osseous anatomy in femoroacetabular impingement using Dixon 3D FLASH sequence

OBJECTIVE

To determine if hip 3D-MR imaging can be used to accurately demonstrate femoral and acetabular morphology in the evaluation of patients with femoroacetabular impingement.

MATERIALS AND METHODS

We performed a retrospective review at our institution of 17 consecutive patients (19 hips) with suspected femoroacetabular impingement who had both 3D-CT and 3D-MRI performed of the same hip. Two fellowship-trained musculoskeletal radiologists reviewed the imaging for the presence and location of cam deformity, anterior-inferior iliac spine variant, lateral center-edge angle, and neck-shaft angle. Findings on 3D-CT were considered the reference standard. The amount of radiation that was spared following introduction of 3D-MRI was also assessed.

RESULTS

All 17 patients suspected of FAI had evidence for cam deformity on 3D-CT. There was 100% agreement for diagnosis (19 out of 19) and location (19 out of 19) of cam deformity when comparing 3D-MRI with 3D-CT. There were 3 type I and 16 type II anterior-inferior iliac spine variants on 3D-CT imaging with 89.5% (17 out of 19) agreement for the anterior-inferior iliac spine characterization between 3D-MRI and 3D-CT. There was 64.7% agreement when comparing the neck-shaft angle (11 out of 17) and LCEA (11 out of 17) measurements. The use of 3D-MRI spared each patient an average radiation effective dose of 3.09 mSV for a total reduction of 479 mSV over a 4-year period.

CONCLUSION

3D-MR imaging can be used to accurately diagnose and quantify the typical osseous pathological condition in femoroacetabular impingement and has the potential to eliminate the need for 3D-CT imaging and its associated radiation exposure, and the cost for this predominantly young group of patients.

The Anterior Ethmoidal "Genu": A Newly Appreciated Anatomic Landmark for Endoscopic Sinus Surgery

Human sinonasal anatomy varies widely between patients, challenging surgeons operating in the sinuses. Ethmoid sinus anatomy is so variable it has been referred to as a labyrinth. Accordingly, reliable, consistent anatomic landmarks aid surgeons operating in this region. The goal of this investigation was to explore our observations and hypothesis that the ethmoidal bulla and the uncinate process are not entirely separate structures but rather attach, and the attachment could potentially provide a landmark for surgeons performing ethmoid and frontal recess surgery. Ethmoid sinus anatomy was studied in 57 sinonasal complexes through a variety of methods including gross anatomic dissection, endoscopic dissection and 3D CT stereoscopic imaging. The uncinate process and ethmoidal bulla were noted to fuse at the superior aspect of the hiatus semilunaris in 57/57 cases, forming a genu-like feature in the anterior ethmoid. This consistent anatomic feature related closely to the frontal sinus drainage pathway, which drained medial to it in 44/57 (77%) cases. The anterior ethmoidal "genu" appears to be an excellent anatomic feature that surgeons can use during ethmoid and frontal recess surgery. High resolution 3D stereoscopic CT scan is capable of demonstrating sinonasal anatomy in a detailed fashion previously only achieved by cadaveric dissection. This technology can potentially allow for a virtual dissection of a patient's anatomy prior to surgery and could improve minimally invasive procedures and reduce complications. Clin. Anat. 32:534-540, 2019. © 2019 Wiley Periodicals, Inc.

New Imaging, Diagnostic, and Assessment Techniques in Orthopedic Trauma

This article examines new imaging, diagnostic, and assessment techniques that may affect the care of patients with orthopedic trauma and/or infection. Three-dimensional imaging has assisted in fracture assessment preoperatively, whereas improvement in C-arm technology has allowed real-time evaluation of implant placement and periarticular reduction before leaving the operating room. Advances in imaging techniques have allowed earlier and more accurate diagnosis of nonunion and infection. Innovations in bacteriologic testing have improved the sensitivity and specificity of perioperative and peri-implant infections. It is critical that surgeons remain up to date on the options available for optimal patient care.

Anterior Inferior Iliac Spine Morphology: Comparison of Symptomatic Hips With Femoroacetabular Impingement and Asymptomatic Hips

OBJECTIVE

The objective of our study was to compare anterior inferior iliac spine (AIIS) morphology in symptomatic hips with femoroacetabular impingement (FAI) and in asymptomatic hips, determine the prevalence of impingement morphology in patients with a radiographic "crossover" sign, and identify potential risk factors for having impingement morphology.

MATERIALS AND METHODS

For this retrospective study, we identified consecutive symptomatic hips with FAI (n = 54) and asymptomatic hips (n = 35) in patients who underwent CT from 2015 to 2017. Two radiologists blindly and independently evaluated 3D CT images of each hip and graded the AIIS morphology according to the Hetsroni classification scheme. The prevalence of AIIS morphology types was calculated. Associations of AIIS morphology types with symptoms and the crossover sign were evaluated with a chi-square test. A multivariable logistic regression determined risk factors for abnormal AIIS morphology (i.e., type 2 or 3).

RESULTS

There was no difference in the prevalence of AIIS morphology types for symptomatic hips with FAI versus asymptomatic hips (p = 0.44) or for hips with a positive versus those with a negative crossover sign (p = 0.21). There was moderate interobserver agreement (κ = 0.44) and good-to-excellent intraobserver agreement (κ = 0.67 and 0.90) for grading AIIS morphology. Age, sex, femoral version, acetabular version, alpha angle, lateral center edge angle, and the crossover sign were not significant risk factors for abnormal AIIS morphology in patients with FAI (p = 0.11-0.79).

CONCLUSION

There is no difference in AIIS morphology between symptomatic hips with FAI versus asymptomatic hips or between hips with and those without the radiographic crossover sign. Age, sex, and other FAI parameters are not risk factors for developing AIIS impingement morphology.

Cinematic rendering of small bowel pathology: preliminary observations from this novel 3D CT visualization method

3D visualization methods for volumetric CT data have played an important role in diagnostic imaging of the small bowel, a structure which intrinsically crosses numerous slices in any 2D imaging plane. Recently, a new approach to 3D CT image creation has become available-cinematic rendering (CR). CR differs from other 3D methods in making use of a global lighting model that produces high surface detail and realistic shadowing effects that lead to 3D visualizations with photorealistic quality. Although the utility of these images for improving diagnostic accuracy has not yet been established, our group's early experience in regions of complex anatomy and pathology has been encouraging. In this pictorial review, we review the established role of 3D CT in many of the most common small bowel pathologies, provide examples of those pathologies visualized with CR, and suggest future directions for researchers to pursue.

Customized implants for acetabular Paprosky III defects may be positioned with high accuracy in revision hip arthroplasty

PURPOSE

In revision hip arthroplasty, custom-made implants are one option in patients with acetabular Paprosky III defects.

METHODS

In a retrospective analysis, we identified 11 patients undergoing cup revision using a custom-made implant. The accuracy of the intended position of the implant was assessed on post-operative 3D CT and compared to the pre-operative 3D planning in terms of inclination, anteversion, and centre of rotation. In addition, the accuracy of post-operative plain radiographs for measuring implant position was evaluated in relation to the 3D CT standard.

RESULTS

We found a mean deviation between the planned and the final position of the custom-made acetabular implant on 3D CT of 3.6° ± 2.8° for inclination and of - 1.2° ± 7.0° for anteversion, respectively. Restoration of center of rotation succeeded with an accuracy of 0.3 mm ± 3.9 mm in the mediolateral (x) direction, - 1.1 mm ± 3.8 mm in the anteroposterior (y) direction, and 0.4 mm ± 3.2 mm in the craniocaudal (z) direction. The accuracy of the post-operative plain radiographs in measuring the position of the custom-made implant in relation to 3D CT was 1.1° ± 1.7° for implant inclination, - 2.6° ± 1.3° for anteversion and 1.3 mm ± 3.5 mm in the x-direction, and - 0.9 mm ± 3.8 mm in the z-direction for centre of rotation.

CONCLUSION

Custom-made acetabular implants can be positioned with good accuracy in Paprosky III defects according to the pre-operative planning. Plain radiographs are adequate for assessing implant position in routine follow-up.

Anatomical differences in the bony structure of L5 and L4: A possible classification according to the lateral tilt of the pedicles

The aim of this study is to underline the necessity of a better knowledge of pedicles anatomy in order to improve surgical treatment of spine disorders such us low back pain, spinal fractures and scholiosis. A classification of pedicles lateral tilt which could help surgeons before the application of screws during transpedicular fixation is reported. Anatomical differences in the orientiation of the pedicles of L5 and L4 have been found. For each patient that met the inclusion criteria underwent: Radiography of the lumbo-sacral region, CT examination, MRI acquisition. Patients were divided into three categories thanks to 3D direct volume rendering of CT scan. Subjects belonged to W-Type, V-Type and U-type depending on their morphometric features. The subdivision was further implemented with measurements of the distance between pedicles and adjacent nervous structures. Concerning L5, W-Type (WT) exhibited a lateral tilt of L5 larger than 36°, V-Type exhibited a lateral tilt of L5 from 30° to 36°, U-type exhibited a lateral tilt of L5 smaller than 30°. Concerning L4, WT exhibited a lateral tilt of 28.4°, VT exhibited a lateral tilt of of 25.1, UT exhibited a lateral tilt of 22.2°; we assume that the degree of lateralization of L4 depends on the one of L5. The way the screw is applied during surgical treatment is clinically relevant, thus our classification may be very useful in order to decrease surgical risk and improve conditions of patients after surgical treatment.

Radiographic Location Does Not Ensure a Precise Anatomic Location of the Femoral Fixation Site in Medial Patellofemoral Ligament Reconstruction

Background:

A frequently used method to determine the anatomic femoral fixation point in the operating room during medial patellofemoral ligament (MPFL) reconstruction is the radiographic method. However, the ability of this radiological method to establish an anatomic femoral attachment point might not be as accurate as expected.

Purpose:

(1) To evaluate the accuracy of the radiological method to locate the anatomic femoral fixation point in MPFL reconstruction surgery and (2) to determine the factors influencing the predictability of this method to obtain this objective.

Study Design:

Cohort study (diagnosis); Level of evidence, 2.

Methods:

A total of 100 consecutive 3-dimensional computed tomography (3D CT) knee examinations were performed at 0° of extension in 87 patients treated for chronic lateral patellar instability. For each knee, 2 virtual 7 mm–diameter femoral tunnels were created: 1 using the adductor tubercle as a landmark (anatomic tunnel) and the other according to the radiological method described by Schöttle et al (radiographic tunnel). We measured the percentage of overlap between both tunnels. Moreover, of the 100 included knees, 10 were randomly selected for a variability study.

Results:

Considering an overlap area greater than 50% as reasonable, the radiographic method achieved this in only 38 of the 100 knees. Intrarater and interrater reliability were excellent. There was a trend for female patients with severe trochlear dysplasia to have less overlap. This model accounted for 64.2% of the initial variability in the data.

Conclusion:

An exact anatomic femoral tunnel placement could not be achieved with the radiographic method. Radiography provided only an approximation and should not be the sole basis for the femoral attachment location. Moreover, in female patients with severe trochlear dysplasia, the radiographic method was less accurate in determining the anatomic femoral fixation point, although differences were not statistically significant.

Fetal and placental anatomy visualized with cinematic rendering from volumetric CT data

Avoiding unnecessary radiation exposure in children, including fetuses or embryos, is of paramount importance. However, emergent clinical situations will arise that necessitate the use of ionizing radiation-based modalities, such as computed tomography (CT), in this patient population. In such circumstances, the use of advanced visualization methods may provide optimum diagnostic utility. We present the case of a pregnant patient with Loeys-Dietz syndrome who was evaluated with CT angiography to rule out an acute aortic syndrome. The CT data from the fetus and placenta were reconstructed using the new cinematic rendering technique that allows for photorealistic display. The potential advantages of cinematic rendering relative to traditional volume rendering are discussed.

Does the position of the scapula in relation to the glenopolar angle change the preferred treatment of extra-articular fractures?

OBJECTIVE

To analyse the glenopolar angle (GPA) at different inclinations of the scapula using 3D CT, to test the hypothesis that the result could change the indication from conservative treatment to surgery.

MATERIALS AND METHODS

Analysis of 30 3D CT images of patients' scapulae, measuring the GPA. The GPA was measured with scapulae at 0^° and at 20^° and 30^° of internal and external rotation. Angles were compared by age, sex and examiner for the different angles of rotation of the scapulae.

RESULTS

The GPA of scapulae in rotation tended to be smaller than the GPA without rotation, and the larger the degree of rotation, the more the angle was underestimated. Additionally, for the same degree of rotation, internal rotation was associated with greater underestimation of the GPA than external rotation. Two different examiners achieved an excellent level of agreement between angle measurements. The GPA with the scapula at 0^° was significantly higher among elderly patients. The variation in GPA with the scapula in rotation (20^° and 30^°) in relation to the GPA without rotation was significantly greater for female patients.

CONCLUSIONS

As the rotation of the scapula was displaced from the scapula in the coronal position (GPA 0^°), both in internal rotation and in external rotation, the GPA reduced. Therefore, rotational displacement may lead to an error in GPA measurement, resulting in incorrect indication of treatment. It is recommended that whenever possible, GPA measurements should be taken in neutral rotation, with the scapula in a neutral position at 0^°.

Optimal trajectory and insertion accuracy of sacral alar iliac screws

Objective

The aim of this study was to analyse the optimal trajectories for sacral alar iliac screws (SAISs) in a Japanese patient population and the clinical assessment of insertion accuracies.

Methods

The ideal trajectories of SAISs, starting from 2 mm medial to the apex of the lateral sacral crest on the midline between S1 and S2 dorsal foramina, were measured in 80 consecutive spinal disease patients (40 males and 40 females; average age: 67.4 ± 8.1 years) using three-dimensional computed tomographic image software. Following these anatomic analyses, accuracies of 32 inserted SAISs in consecutive patients, who underwent long spinal posterior fusion, were investigated clinically.

Results

Lateral angulations of optimal SAIS trajectories in males (left: 37.9; right: 37.7) were significantly larger than those than in females (left: 32.8; right: 32.4). Caudal SAIS angulations for females (left: 33.4; right: 33.9) were significantly larger than those in males (left: 27.5; right: 28.0). The 32 SAISs (100 mm long and 9 mm in diameter) assessed clinically were accurately inserted on optimal trajectories.

Conclusion

The optimal trajectories of SAISs in a Japanese patient population are more lateral in males and more caudal in females. This study examines the clinical safety and accuracy of SAIS insertion on these optimal trajectories.

Practicality of exchanging transparent 3D CT for radiography for pelvic fractures

OBJECTIVE

We assess the utility of transparent 3D reconstructed CT images for evaluation of traumatic pelvic bony injuries compared to traditional radiographs.

METHODS

Radiographs and 3D reconstructed CT were anonymized and randomized before review by 4 board certified physicians using a standardized questionnaire and compared to a gold-standard axial CT by a fifth board certified physician.

RESULTS

49 patients were included. We found significant agreement (K=[0.5-0.92], p<0.001) and comparable accuracy (K=[0.36-0.38], p<0.02) and ghost images of radiographs and transparent 3D reconstructed CT without a difference in confidence (p=0.38).

CONCLUSION

Transparent 3D reconstructed CT images may be sufficient for pelvic trauma injury without the use of radiographs.

Three-Dimensional Reconstruction of Intracardiac Anatomy Using CTA and Surgical Planning for Double Outlet Right Ventricle: Early Experience at a Tertiary Care Congenital Heart Center

BACKGROUND

Although transthoracic echocardiography (TTE) routinely establishes the diagnosis of double outlet right ventricle (DORV), it can be suboptimal for depicting exact ventricular septal defect (VSD) position, especially with respect to the outflow tracts. Advanced imaging with computed tomography angiography (CTA) can help visualize structures and relationships not easily seen by echo. Using computer-aided design, we have the ability to create three-dimensional (3D) models of the intracardiac anatomy, which can be helpful for better depicting the overall anatomy to assist surgical planning.

METHODS

Patients with a diagnosis of DORV were retrospectively reviewed at our institution from October 2013 to April 2015. Patients who preoperatively underwent both TTE and CTA with 3D reconstruction of the intracardiac anatomy were included. Computed tomography angiography findings with 3D intracardiac model creation were compared to the surgical findings.

RESULTS

Twenty-five patients underwent surgical repair of DORV during the study period. Five patients had CTA with 3D reconstruction, in addition to the standard TTE images, and were included in the study. In all five cases, CTA with 3D reconstruction of the intracardiac anatomy accurately depicted the VSD position relative to important adjacent structures, including the outflow tracts.

CONCLUSION

Three-dimensional reconstruction of the intracardiac anatomy using CTA data can provide accurate data for presurgical planning of DORV repair and has the potential for being especially useful in patients for whom intracardiac anatomy and VSD position cannot be well seen by TTE. A larger prospective analysis is warranted to help validate this approach.

Fracture morphology of AO/OTA 31-A trochanteric fractures: A 3D CT study with an emphasis on coronal fragments

AIMS

This study was designed to assess the incidence and morphology of coronal plane fragments in AO/OTA 31-A trochanteric fractures.

PATIENTS AND METHODS

156 cases of AO/OTA 31-A trochanteric fractures were retrospectively evaluated. Lateral radiographs were analyzed for the presence of coronal plane fragments followed by analysis of 3D CT reconstructions in these fractures. The incidence of coronal fragments identified on the lateral radiograph and 3D CT reconstructions were both calculated. Coronal fragment morphology was described based upon the origin and exit points of fracture lines and the number of fragments.

RESULTS AND CONCLUSION

On plain radiographs, a coronal plane fracture was identified in 59 cases, an incidence of 37.8% (59/156). In comparison, 3D CT reconstructions identified coronal plane fractures in 138 cases for an incidence of 88.4% (138/156). 3D CT reconstructions identified coronal fracture fragments in 81.9% (50/61) of AO/OTA 31-A1 cases, 94.5% (69/73) of 31-A2 cases, and 86.3% (19/22) of 31-A3 cases. Incidence of coronal fractures identified on plain radiographs of 3 AO/OTA 31-A1,A2,A3 groups was lower when compared to the incidence of coronal fractures identified on 3D CT. Of the 138 cases that had coronal plane fracture, 82 cases (59.4%) had a single coronal fragment (GT fragment 35 cases, GLT fragment 19 cases, GLPC fragment 28 cases). The remaining 56 cases (40.5%) had two coronal fragments. There is a high incidence of coronal fragments in intertrochanteric femur fractures when analyzed with 3D CT reconstructions. Our study suggests that these coronal fragments are difficult to identify on plain radiographs. Knowledge of the incidence and morphology of coronal fragments helps to avoid potential intraoperative pitfalls.

Tunnel widening in single- versus double-bundle anterior cruciate ligament reconstructed knees

PURPOSE

The consequence of tunnel widening after ACL reconstructions is foremost of importance in case of revision surgery. Tunnel expansion leads to bone loss close to the joint, and additional surgery with bone grafting prior to revision surgery might be necessary. The purpose of the study was to measure widening of the tunnels in single-bundle (SB) and double-bundle (DB) ACL reconstructed knees during the first year after surgery, detected by a novel, semi-automated 3D CT imaging modality. Our hypothesis was that there would be a difference between the initial tunnel size and the size measured one year post-operatively due to the tunnel widening process. Further, the purpose was to evaluate whether there were any differences in the amount of tunnel widening between the two surgical techniques.

METHODS

Twenty patients who underwent DB ACL reconstruction, and 22 patients who underwent SB ACL reconstruction, performed a CT scan of the bony tunnels, during their first days after surgery and one year post-operatively. The CT scans were transformed into 3D CT reconstructions, and the tunnels were measured with the "best-fit cylinder" method, measurements at the level of tunnel aperture and 10.0 mm from the joint line.

RESULTS

All tunnels in the DB and SB ACL reconstructed knees exhibited widening during the first year after the operation (p < 0.001). The SB femoral tunnels showed more widening compared to the DB femoral AM tunnels (1.4 ± 0.9 vs. 0.5 ± 0.6 mm) (p < 0.001), and the SB tibial tunnels widened more compared to the DB tibial PL tunnels (1.0 ± 1.0 vs. 0.5 ± 0.6) (p < 0.043).

CONCLUSION

All tunnels widened during the first year after the ACL reconstruction with a larger amount of widening in the SB tunnels compared to the DB femoral AM tunnels and the DB tibial PL tunnels. This is the first study to detect tunnel widening in DB reconstructed knees through a semi-automated 3D CT imaging modality.

LEVEL OF EVIDENCE

Prospective cohort study, Level III.

Is radiograph glenopolar angle accurate for extraarticular scapular neck fractures?

BACKGROUND

Plain radiographs still play a role in management of extraarticular scapular neck fractures. Glenopolar angle (GPA) is one of the radiograph measurements that is used to determine the necessity for surgery. Our aim was to establish reliability of GPA on plain radiograph in patients with extraarticular scapular neck fractures.

METHODS

We performed a multicentre retrospective study including all patients with extraarticular scapular neck fractures with available imaging between 2006 and 2012. We excluded intra-articular glenoid fractures, scapular blade fractures, acromion fractures, and scapular spine fractures. We compared GPA on plain radiograph with three dimensional computed tomography (3D CT) measurement, as well as contribution of radiograph rotational error, glenoid inclination, and medial shortening of glenoid fragment towards GPA measurement.

RESULTS

One hundred patients met the inclusion criteria. The mean difference between the GPA measurements on radiographs and 3D CT was 6.1±0.85° (95% confidence interval) as an absolute value. In terms of contribution to GPA values, GPA changed by one degree with ten degrees of radiograph rotational error, three degrees of glenoid inclination, and three millimetres of glenoid fragment medial shortening.

CONCLUSION

Plain radiograph can provide a clinician with a reasonable estimation of the GPA. Glenoid inclination has a greater influence on GPA compared to medial shortening.

Automatic abdominal multi-organ segmentation using deep convolutional neural network and time-implicit level sets

PURPOSE

Multi-organ segmentation from CT images is an essential step for computer-aided diagnosis and surgery planning. However, manual delineation of the organs by radiologists is tedious, time-consuming and poorly reproducible. Therefore, we propose a fully automatic method for the segmentation of multiple organs from three-dimensional abdominal CT images.

METHODS

The proposed method employs deep fully convolutional neural networks (CNNs) for organ detection and segmentation, which is further refined by a time-implicit multi-phase evolution method. Firstly, a 3D CNN is trained to automatically localize and delineate the organs of interest with a probability prediction map. The learned probability map provides both subject-specific spatial priors and initialization for subsequent fine segmentation. Then, for the refinement of the multi-organ segmentation, image intensity models, probability priors as well as a disjoint region constraint are incorporated into an unified energy functional. Finally, a novel time-implicit multi-phase level-set algorithm is utilized to efficiently optimize the proposed energy functional model.

RESULTS

Our method has been evaluated on 140 abdominal CT scans for the segmentation of four organs (liver, spleen and both kidneys). With respect to the ground truth, average Dice overlap ratios for the liver, spleen and both kidneys are 96.0, 94.2 and 95.4%, respectively, and average symmetric surface distance is less than 1.3 mm for all the segmented organs. The computation time for a CT volume is 125 s in average. The achieved accuracy compares well to state-of-the-art methods with much higher efficiency.

CONCLUSION

A fully automatic method for multi-organ segmentation from abdominal CT images was developed and evaluated. The results demonstrated its potential in clinical usage with high effectiveness, robustness and efficiency.

Aortic dissection albeit negative d-dimers and zero clinical probability - Another facet of medicine's deadly chameleon

Aortic dissection, the rupture of the aorta's intimal and medial layers, leading to the formation of a false lumen, is a relatively common disease with high mortality. So far, while not addressing penetrating aortic ulcer or intramural hematoma, current guidelines take a negative value for d-dimers for a sufficient method to rule out aortic dissection in patients with a low clinical probability, as calculated by the ADD risk score. We present two cases of patients with acute aortic dissection, albeit presenting with a low clinical probability for acute aortic dissection and negative values for d-dimers. <Learning objective: One must entertain a high level of suspicion for acute aortic syndrome as even laboratory results within the normal range of d-dimers connected with a low clinical probability can be misleading. According to the patient's symptoms, a battery of different diagnostic tools should play in concert, including imaging studies, repeat clinical evaluation, and repeat laboratory tests, such as d-dimers, before dismissing a diagnosis as severe as aortic dissection.>.

Real-Time 3D CT Image Guidance for Transjugular Intrahepatic Portosystemic Shunt Creation Using Preoperative CT: A Prospective Feasibility Study of 20 Patients

OBJECTIVE

The purpose of this study is to prospectively evaluate the feasibility and efficacy of real-time 3D CT image guidance during transjugular intrahepatic portosystemic shunt (TIPS) creation.

SUBJECTS AND METHODS

Between October 2013 and December 2013, a total of 20 patients were prospectively enrolled in the present study. Previously acquired portal venous phase CT datasets and intraoperative CT datasets were registered on a dedicated workstation. We accomplished semiautomatic registration for the datasets of 11 of 20 patients (55%), and we performed manual registration for the datasets of the remaining nine patients. The selected volume of interest of the CT image showing the portal vein vasculature was overlaid onto the fluoroscopic display to provide real-time 3D CT image guidance during the procedure.

RESULTS

For all 20 patients, TIPS procedures were successfully performed by the same operator. The mean (± SD) number of needle passes required for portal vein entry was 1.8 ± 1.1 passes (range, 1-5 passes). The mean duration of radiographic fluoroscopy was 3.5 ± 1.1 minutes for portal vein entry and 11.4 ± 2.1 minutes for the whole procedure. The mean radiation dose used for the whole TIPS procedure was 295.5 ± 66.6 Gy · cm². No major technical complications were observed.

CONCLUSION

Real-time 3D guidance with the use of preoperative CT is feasible, safe, and effective for assisting in the creation of TIPS. This approach may result in a shorter procedural time and less radiation exposure. However, future studies are required to compare this method with other mapping techniques.

Optimal area of lateral mass mini-screws implanted in plated cervical laminoplasty: a radiography anatomy study

PURPOSE

Lateral mass mini-screws used in plated cervical laminoplasty might penetrate into facet joints. The objective is to observe this complication incidence and to identify the optimal areas for 5- and 7-mm-long mini-screws to implant on lateral mass.

METHODS

47 patients who underwent plated cervical laminoplasty were included. The optimal area for mini-screws implanting was set according to pre-operative 3D CT reconstruction data. Then, each posterior-lateral mass surface was divided into three regions: 7-mm region, 5-mm region, and dangerous area. The mini-screw implanted region was recorded. Post-operative CT images were used to identify whether the mini-screws penetrated into facet joints.

RESULTS

235 mini-plates and 470 lateral mass mini-screws were used in the study. 117 (24.9%) mini-screws penetrated 88 (37.4%) facet joints. The 5-mm-long mini-screw optimal area occupied the upper 72, 65, 65, 64, and 65 % area of the posterior-lateral mass surface for C3-7, while the 7-mm-long mini-screw optimal area encompassed the upper 54, 39, 40, 33, and 32 %. Only 7-mm-long mini-screws were used to fix the plate to the lateral mass. 4 of 240 mini-screws in 7-mm region, 67 of the 179 mini-screws in 5-mm region, and 46 of the 51 mini-screws in dangerous region penetrated into the facet joint. The differences in the rate of facet joint penetration related to region were statistically significant (P < 0.001).

CONCLUSIONS

The facet joint destruction by mini-screws was not a rare complication in plated cervical laminoplasty. The optimal areas we proposed may help guide the mini-screw implantation positions.

Transportal femoral drilling creates more horizontal ACL graft orientation compared to transtibial drilling: A 3D CT imaging study

BACKGROUND

The principle of anatomic anterior cruciate ligament (ACL) reconstruction is to create a femoral and tibial tunnel that resembles the insertion of the native ACL. Anatomic reconstruction leads to a more horizontal graft orientation that provides more rotational stability. The aim of this study is to investigate the best method to achieve anatomical reconstruction of femoral insertion of the ACL and thus, a more horizontal orientation of the ACL. We compared tunnel position and orientation between transportal femoral drilling technique and transtibial technique.

METHODS

Thirty-two patients were included. Post-operative CT scans were obtained and femur, tibia and ACL tunnels were reconstructed. The position and orientation of tibial and femoral tunnels were quantified using the quadrant method, and femoral tunnel length, ellipticity and posterior wall breakage were assessed. We also investigated clinical outcome.

RESULTS

Analyses show that transportal drilled femoral tunnels were situated significantly lower than transtibial drilled tunnels (p<0.0001), resulting in a significantly more horizontal oriented ACL in the transportal group in coronal (p<0.0001) and sagittal plane (p=0.01). No differences were observed in depth of femoral tunnel position (p=0.44). Femoral tunnel length was shorter in the transportal group (p=0.01) with a more ellipsoidal femoral aperture (p=0.01). There were no differences between both groups in tibial position. There were no differences in clinical outcome measure between the transportal and transtibial groups.

CONCLUSION

This study indicates that transportal drilling of the femoral tunnel leads to a more horizontal graft orientation of the ACL, without differences in clinical outcome.

Is volumetric 3-dimensional computed tomography useful to predict histological tumour invasiveness? Analysis of 211 lesions of cT1N0M0 lung adenocarcinoma

OBJECTIVES

The purpose of this study was to use Hounsfield unit (HU) thresholds of computed tomography (CT) images to predict pathological lymph node metastasis and tumour invasiveness of cT1N0M0 lung adenocarcinoma on 3D evaluations.

METHODS

Preoperative CT images of 211 lesions of surgically resected cT1N0M0 lung adenocarcinoma were retrospectively examined. The tumour size was calculated in 1D, 2D and 3D views. Tumours with -300 HU and over were defined as 'solid tumours', and those between -800 and -301 HU were defined as 'ground glass opacity tumours'. Tumours with -800 HU and over were assumed to be the whole tumour entity. The proportion of 'solid tumour' within the whole tumour entity was also calculated as the 'solid tumour ratio'. These were compared with pathological information.

RESULTS

Solid tumour size and ratio were positively correlated with microscopic invasion to pleura, vessels and lymphatics in all dimensional evaluations. Pathological lymph node metastases were also well predicted by solid tumour size and ratio in all dimensional evaluations. The P-values for the receiver operating characteristic (ROC) curves of 1D, 1D ×2, 2D and 3D evaluations were: solid tumour size P = 0.013, 0.014 and 0.032; and solid tumour ratio 0.016, 0.0032 and <0.0001. In comparisons of 1D, 2D and 3D evaluations, 'solid tumour size' of the area under the curve (AUC) of ROC to detect pathological lymph node metastases was not significant. However, strikingly, the 3D solid tumour ratio was found to be significantly more accurate for the prediction of pathological lymph node metastases than the 1D and 2D solid tumour ratios on ROC evaluation (AUC: 1D 0.736, 2D 0.803 and 3D 0.882; P-values for the AUC comparisons were P = 0.013 for 3D versus 1D and P = 0.022 for 3D versus 2D). The correlations of subtypes of adenocarcinoma and the 3D solid tumour ratio were also investigated. Subtypes of adenocarcinoma were well correlated with the 3D solid tumour ratio.

CONCLUSIONS

Preoperative 3D CT using threshold values of -800 and -300 HU was useful for predicting pathological lymph node metastases and tumour invasiveness of cT1N0M0 lung adenocarcinoma.

The effect of feedback from post-operative 3D CT on placement of femoral tunnels in single-bundle anatomic ACL reconstruction

PURPOSE

To evaluate the effect of feedback from post-operative 3D CT in the learning process of placing the femoral graft tunnel anatomically using the anteromedial (AM)-portal technique in single-bundle anterior cruciate ligament (ACL) reconstruction.

METHODS

An experienced knee surgeon converting from transtibial to AM-portal technique was offered post-operative feedback on tunnel placement. Three groups of patients were included: transtibial drilling, (AM1) anteromedial drilling without feedback and (AM2) anteromedial drilling with post-operative CT feedback. Intra-articular landmarks were used as the only guidance for tunnel placement. Tunnel position was compared to an ideal anatomical ACL position using the Bernard and Hertel grid and visual feedback was given on tunnel placements. The effect of feedback was measured as the distance from the anatomical centre, and spread of tunnel placements on post-operative CT performed feedback was initiated.

RESULTS

When comparing the femoral tunnel placement to an ideal anatomical centre, there was an improvement in the mean tunnel position after (A) changing from a transtibial to an anatomical technique and a further improvement after (B) initializing the radiological feedback. There was a great variation of femoral tunnel localizations when initially only using intra-articular landmarks as guidance for tunnel placement--this variation, however, converged towards the anatomical centre throughout the feedback period and the AM2 group had a femoral tunnel closer (P = 0.001) to the anatomical centre than the AM1 group.

CONCLUSIONS

Post-operative 3D CT is effective in the learning process of placing femoral tunnels anatomically by giving post-operative feedback on tunnel placement. Bony landmarks and ACL remnants were found unreliable as the only guidance for femoral tunnel placement in the AM-portal technique-therefore, the use of an aid is recommended to reduce unwanted tunnel variations in a learning phase.

LEVEL OF EVIDENCE

Cohort Study, Level III.

Measurement of the radiographic anatomy of the small and ring metacarpals using computerized tomographic scans

BACKGROUND

To date, only plain radiographic definitions of normal anatomical parameters have been described. Our study aims to describe normal anatomic measurements of small and ring metacarpals using a novel digital reconstruction technique based on raw CT image data. We hypothesize that current plain radiographic data incorrectly describes normal metacarpal anatomy in the lateral plane.

METHODS

Thirty-five scans of the small and 30 scans of the ring metacarpals form the basis for this study. Using a custom digital 3D image reformatting software, CT sections were reconstructed in the plane of each studied metacarpal. The 3D images were converted to sagittal and coronal weighted projections to represent lateral and posteroanterior (PA) 2D images that are equivalent to "perfect orthogonal" radiographs. Using a customized image measurement program, shaft lengths, shaft-bending angle (SBA), and capital-axis angle (CAA) were measured.

RESULTS

Our results show that CAA averaged 14 and 12° in the ring and small metacarpals, respectively. Apex dorsal SBA averaged 12 and 10° in the ring and small metacarpals, respectively. On the PA images, the shafts are nearly straight. In contrast to prior reported values, we found the CAA to be less acute and the metacarpal curvature less pronounced on the lateral projection. We also demonstrated that much of the metacarpal apex dorsal bend is in the shaft itself.

CONCLUSION

Normal anatomic parameters of metacarpals are based primarily on radiographic data, and as such are limited due to bony overlap in the lateral plane, as well as imperfect radiographic projections that are known to distort anatomical relationships. This novel method of image reconstruction eliminates metacarpal overlap and defines precise anatomical reference for metacarpals.

Three-dimensional CT might be a potential evaluation modality in correction of asymmetrical masseter muscle hypertrophy by botulinum toxin injection

For correction of this asymmetrical hypertrophy, botulinum toxin type A (BTxA) injection is one of convenient treatment modalities. Unfortunately, physical examination of masseter muscle is not enough to estimate the exact volume of muscle hypertrophy difference. Two Koreans, male and female, of bilateral masseter hypertrophy with asymmetricity were evaluated. BTxA (NABOTA(®), Daewoong, Co. Ltd., Seoul, Korea) was injected at master muscle site with total 50 U (25 U at each side) and volume change was evaluated with three-dimensional (3D) CT image analysis. Maximum reduction of masseter hypertrophy was recognized at 2-month follow-up and reduced muscle size started to restore after 3 months. Mean reduction of masseter muscle volume was 36% compared with baseline. More hypertrophied side of masseter muscle presented 42% of volume reduction at 2-month follow-up but less hypertrophied side of masseter muscle showed 30% of volume shrinkage. In conclusion, 3D CT image analysis might be the exact evaluation tool for correction of asymmetrical masseter hypertrophy by botulinum toxin injection.

Pleural invasion by peripheral lung cancer: prediction with three-dimensional CT

RATIONALE AND OBJECTIVES

To evaluate the clinical utility of three-dimensional (3D) computed tomography (CT) for predicting pleural invasion by peripheral lung cancer.

MATERIALS AND METHODS

CT findings (tumor size, vertical diameter, length and area of the interface between tumor and the pleura, ratios of length and area [Rarea] of interface between tumor and the pleura to tumor size, angle between the tumor and adjacent pleura, presence or absence of pleural thickening, and originally developed 3D pleural patterns) in 201 consecutive patients with lung cancer of ≤3 cm in contact with pleural surface were correlated with pathologic findings. Logistic modeling was used for determining the significant factors for prediction of pleural invasion, and receiver operating characteristic (ROC) curves were used for investigating diagnostic capability of significant factors, resulting in a recommendation to the optimal criteria for predicting pleural invasion and to the optimal threshold for differentiating parietal from visceral invasion.

RESULTS

Sixty-one (30%) of the 201 patients had pathologically verified pleural invasion. Logistic modeling revealed that the 3D pleural pattern was the only significant factor (P < .001; relative risk of 7.34). Among every combination of the 3D patterns, skirt-like pattern showed the highest accuracy of 77% for predicting pleural invasion. In differentiating parietal from visceral pleural invasion, ROC analysis revealed that Rarea was optimal for differentiating parietal from visceral pleural invasion, and the highest accuracy of 77% was obtained with a cut-off value of 13.4 for this criterion.

CONCLUSIONS

Computer-aided 3D CT analysis of the pleura was useful for predicting pleural invasion.

X-ray strain tensor imaging: FEM simulation and experiments with a micro-CT

In tissue elasticity imaging, measuring the strain tensor components is necessary to solve the inverse problem. However, it is impractical to measure all the tensor components in ultrasound or MRI elastography because of their anisotropic spatial resolution. The objective of this study is to compute 3D strain tensor maps from the 3D CT images of a tissue-mimicking phantom. We took 3D micro-CT images of the phantom twice with applying two different mechanical compressions to it. Applying the 3D image correlation technique to the CT images under different compression, we computed 3D displacement vectors and strain tensors at every pixel. To evaluate the accuracy of the strain tensor maps, we made a 3D FEM model of the phantom, and we computed strain tensor maps through FEM simulation. Experimentally obtained strain tensor maps showed similar patterns to the FEM-simulated ones in visual inspection. The correlation between the strain tensor maps obtained from the experiment and the FEM simulation ranges from 0.03 to 0.93. Even though the strain tensor maps suffer from high level noise, we expect the x-ray strain tensor imaging may find some biomedical applications such as malignant tissue characterization and stress analysis inside the tissues.

Location of the femoral tunnel aperture in single-bundle anterior cruciate ligament reconstruction: comparison of the transtibial, anteromedial portal, and outside-in techniques

BACKGROUND

Previous 3-dimensional computed tomography (3D CT) studies of knees after anterior cruciate ligament (ACL) reconstruction have compared femoral tunnel positions obtained using the transtibial and anteromedial drilling techniques. This study used postoperative in vivo 3D CT analysis to compare the locations of the femoral tunnel aperture among 3 drilling techniques used in ACL reconstruction: transtibial, anteromedial portal, and outside-in.

HYPOTHESIS

The use of the transtibial drilling technique might result in a less anatomically accurate femoral tunnel placement than the anteromedial portal and outside-in techniques.

STUDY DESIGN

Cohort study; Level of evidence, 3.

METHODS

Immediate postoperative in vivo 3D CT was used to assess the location of the femoral tunnel aperture in 153 patients who underwent single-bundle ACL reconstruction using the transtibial (n = 42), anteromedial portal (n = 73), or outside-in (n = 38) techniques. Femoral tunnel positions were measured by an anatomic coordinate axis method in the low-to-high and deep-to-shallow directions of the distal femur at 90° of knee flexion.

RESULTS

The low-to-high femoral tunnel positions were significantly higher in the transtibial group than in the anteromedial portal (P < .001) and outside-in (P < .001) groups. There were no differences among the 3 groups in the deep-to-shallow femoral tunnel positions (P = .773).

CONCLUSION

The transtibial technique of anatomic reconstruction resulted in more highly positioned femoral tunnels in the low-to-high direction than did the anteromedial portal and outside-in techniques. However, no significant differences in the femoral tunnel location were observed in the deep-to-shallow direction.

Coverage of the anterior cruciate ligament femoral footprint using 3 different approaches in single-bundle reconstruction: a cadaveric study analyzed by 3-dimensional computed tomography

BACKGROUND

Performing a single-bundle anterior cruciate ligament (ACL) reconstruction within the femoral footprint is important to obtain a functional graft and a stable knee.

HYPOTHESIS

There will be a significant difference in the ability of 3 ACL reconstruction techniques to reach and cover the native femoral footprint.

STUDY DESIGN

Controlled laboratory study.

METHODS

The percentage of the ACL footprint covered by the femoral tunnel was compared after 3 different techniques to target the footprint: transtibial (TT), inside-out/anteromedial (IO), and outside-in/transfemoral (OI). Fourteen cadaveric knee specimens with a mean age of 67.5 years were used. For each knee, the TT technique utilized a 7.5-mm offset guide, the IO technique was performed through an accessory anteromedial portal, and the OI technique was carried out through the femur from the external wall of the lateral condyle. Entry points in the footprint were spotted with markers, and orientations (sagittal and frontal) of each drill guide were noted. The distal femurs were sawed and scanned, and 3-dimensional image reconstructions were analyzed. The virtual drilled area (reamer diameter, 8 mm) depending on the entry point and the sagittal/frontal orientation of the drill guide was calculated and reported for each of the 3 techniques. The distance from the tunnel center to the ACL center, percentage of the femoral tunnel within the ACL footprint, and percentage of the ACL footprint covered by the tunnel were calculated and statistically compared (analysis of variance and t test).

RESULTS

The average distance to the native femoral footprint center was 6.8 ± 2.68 mm for the TT, 2.84 ± 1.26 mm for the IO, and 2.56 ± 1.39 mm for the OI techniques. Average percentages of the femoral tunnel within the ACL footprint were 32%, 76%, and 78%, and average percentages of the ACL footprint covered by the tunnel were 35%, 54%, and 47%, for the TT, IO, and OI techniques, respectively. No significant difference was observed between the IO and OI techniques (P = .11). The TT approach gave less satisfactory coverage on all testing criteria.

CONCLUSION

The IO and OI techniques allowed for creation of a tunnel closest to the ACL femoral footprint center. Despite this fact and even if the average percentage of the drilled area included in the femoral footprint was close to 80% for these 2 techniques, the average percentage of the ACL footprint covered by the tunnels was <55% for all 3 techniques. Coverage of the ACL footprint depended on the entry point, orientation, and diameter of the drilling but also on the size of the footprint.

CLINICAL RELEVANCE

To improve the coverage of the native femoral footprint with a single-bundle graft, in addition to the entry point it may also be necessary to consider the orientation of the drilling to increase the dimensions of the area while respecting the anatomic constraints of the femoral bone and graft geometry.

A comparison study of different facial soft tissue analysis methods

OBJECTIVES

The purpose of this study was to evaluate several different facial soft tissue measurement methods.

MATERIALS AND METHODS

After marking 15 landmarks in the facial area of 12 mannequin heads of different sizes and shapes, facial soft tissue measurements were performed by the following 5 methods: Direct anthropometry, Digitizer, 3D CT, 3D scanner, and DI3D system. With these measurement methods, 10 measurement values representing the facial width, height, and depth were determined twice with a one week interval by one examiner. These data were analyzed with the SPSS program.

RESULTS

The position created based on multi-dimensional scaling showed that direct anthropometry, 3D CT, digitizer, 3D scanner demonstrated relatively similar values, while the DI3D system showed slightly different values. All 5 methods demonstrated good accuracy and had a high coefficient of reliability (>0.92) and a low technical error (<0.9 mm). The measured value of the distance between the right and left medial canthus obtained by using the DI3D system was statistically significantly different from that obtained by using the digital caliper, digitizer and laser scanner (p < 0.05), but the other measured values were not significantly different. On evaluating the reproducibility of measurement methods, two measurement values (Ls-Li, G-Pg) obtained by using direct anthropometry, one measurement value (N'-Prn) obtained by using the digitizer, and four measurement values (EnRt-EnLt, AlaRt-AlaLt, ChRt-ChLt, Sn-Pg) obtained by using the DI3D system, were statistically significantly different. However, the mean measurement error in every measurement method was low (<0.7 mm). All measurement values obtained by using the 3D CT and 3D scanner did not show any statistically significant difference.

CONCLUSION

The results of this study show that all 3D facial soft tissue analysis methods demonstrate favorable accuracy and reproducibility, and hence they can be used in clinical practice and research studies.

Actual concepts in scaphocephaly : (an experience of 98 cases)

Craniosynostoses are recognized as a group of birth defects that impair the skull structures by early closure of one or more sutures, causing an abnormal cranial shape. Among the "simple" craniosynostoses, (a single closed suture) the most common is scaphocephaly. The 3D CT scan is the most relevant and rapid diagnostic test. The authors present the personal experience of 98 scaphocephaly cases diagnosed and surgically treated in the Neurosurgical Department of "Bagdasar-Arseni" Emergency Hospital during a period of 10 years (2000 - 2009). The procedure of choice was the Stein & Schut (1977) extensive craniotomy that removes the early closed suture. There were no post-operatory death cases and no abnormally closed sutures. The routine use of the craniotome facilitates the lateral osteotomy that allows a normal brain growth and a normal symmetrical skull shape development. The authors advocate for early surgery during the first 6 months of life.