Interdisciplinary consensus of virtual monochromatic dual-energy CT images: is there discrepancy in preferred photon energy between surgeons and radiologists for the assessment of non-unions?

AIM: To investigate possible differences between surgeons and radiologists in selecting optimal photon energy settings from a set of virtual monochromatic dual-energy computed tomography (CT) images for the assessment of bone union in patients with a suspected non-union of the appendicular skeleton. MATERIALS AND METHODS: Fifty patients suspected of having bone non-union after operative fracture treatment with a variety of fixation implants were included. Patients were scanned on a dual-source CT machine using 150/100-kVp. Monochromatic images were extracted at 70, 90, 110, 130, 150, and 190 keV. Images were reviewed by 159 orthopaedic trauma surgeons and 12 musculoskeletal radiologists in order to select the best and worst energy setting to assess bone union. Furthermore, a confidence score (1-4) was given in selecting the best and worst setting to assess bone union. RESULTS: Monochromatic 190 keV images were selected most frequently as the optimal energy in titanium (34.8%), stainless steel (40%), and combined implants of stainless steel and titanium (40.5%). Confidence scores and average optimal energies were higher and average worst energies were lower for radiologists compared to surgeons in all hardware (p<0.05). Differences in optimal energy were not statistically significant for different alloys or type of fixation implant in both groups. CONCLUSIONS: In both observer groups, 190 keV images were selected most frequently as the optimal energy to assess bone union in patients with a suspected non-union of the appendicular skeleton with hardware in situ. On average, musculoskeletal radiologists selected higher optimal and lower worst energy settings and were more confident in selecting both energy settings than orthopaedic trauma surgeons.

Three-dimensional topography of scapular nutrient foramina.

PURPOSE: The aim of this study is to describe the number and location of the nutrient foramina in human scapulae which can minimize blood loss during surgery. METHODS: 30 cadaveric scapulae were macerated to denude the skeletal tissue. The nutrient foramina of 0.51 mm and larger were identified and labeled by adhering glass beads. CT scans of these scapulae were segmented resulting in a surface model of each scapula and the location of the labeled nutrient foramina. All scapulae were scaled to the same size projecting the nutrient foramina onto one representative scapular model. RESULTS: Average number of nutrient foramina per scapula was 5.3 (0-10). The most common location was in the supraspinous fossa (29.7%). On the costal surface of the scapula, most nutrient foramina were found directly inferior to the suprascapular notch. On the posterior surface, the nutrient foramina were identified under the spine of the scapula in a somewhat similar fashion as those on the costal surface. Nutrient foramina were least present in the peri-glenoid area. CONCLUSION: Ninety percent of scapulae have more than one nutrient foramen. They are located in specific areas, on both the posterior and costal surface.

Correction to: Three­dimensional topography of scapular nutrient foramina.

Correction to: Surgical and Radiologic Anatomy.

The "coracoid tunnel view": a simulation study for finding the optimal screw trajectory in coracoid base fracture fixation.

PURPOSE: Coracoid fractures represent approximately 3-13% of all scapular fractures. Open reduction and internal fixation can be indicated for a coracoid base fracture. This procedure is challenging due to the nature of visualization of the coracoid with fluoroscopy. The aim of this study was to develop a fluoroscopic imaging protocol, which helps surgeons in finding the optimal insertion point and screw orientation for fixations of coracoid base fractures, and to assess its feasibility in a simulation study. METHODS: A novel imaging protocol was defined for screw fixation of coracoid base fractures under fluoroscopic guidance. The method is based on finding the optimal view for screw insertion perpendicular to the viewing plane. In a fluoroscopy simulation environment, eight orthopaedic surgeons were invited to place a screw down the coracoid stalk through the coracoid base and into the neck of 14 cadaveric scapulae using anatomical landmarks. The surgeons placed screws before and after they received an e-learning of the optimal view. Results of the two sessions were compared and inter-rater reliability was calculated. RESULTS: Screw placement was correct in 33 out of 56 (58.9%) before, and increased to 50 out of 56 (89.3%) after the coracoid tunnel view was explained to the surgeons, which was a significant improvement (p < 0.001). CONCLUSIONS: Our newly developed fluoroscopic view based on simple landmarks is a useful addendum in the orthopaedic surgeon's tool box to fixate fractures of the coracoid base.

Locking plate constructs in subtrochanteric fixation: a biomechanical comparison of LCP screws and AO-nuts.

OBJECTIVES: Various studies have reported the use of the 95-degree condylar blade plate in the treatment of a subtrochanteric fracture or non-union. However, the holding power of standard screws in the metaphyseal and diaphyseal area is often diminished due to osteopenia. The alternative in this area is the use of locking plates, Schühlis or AO-nuts. With the latter two, non-locking screws in the blade plate can be converted to a fixed angle fixation. The objective of this study was to compare the stiffness and strength of the AO-nut augmented 95-degree condylar blade plate construct with that of a locking plate construct. In addition, a clinical series of eight patients treated with the AO-nut augmented 95-degree condylar blade plate construct is presented. METHODS: Single screw-plate constructs of a 5.0 mm locking screw/locking compression plate (LCP) and a 4.5 mm non-locking screw/4.5 mm dynamic compression plate (DCP), converted to a fixed-angle screw construct using AO-nuts, were tested by cantilever bending. During loading, force and displacement were recorded, from which the bending stiffness (N/mm) and the yield strength (N) were determined. Secondarily, all patients that underwent surgical treatment for subtrochanteric fracture, malunion or non-union by the senior author using this technique, underwent chart review. RESULTS: The stiffness of the locking screws was about four times higher compared to the AO-nut augmented construct. The yield strength was 2.3 times higher for the locking screw construct. In none of the eight patients treated with the fixed-angle blade plate, failure of the AO-nut augmented construct occurred. CONCLUSIONS: Although the stiffness and strength of the AO-nut augmented construct is less than of the locking screw, excellent clinical outcomes can be achieved utilizing this construct.

Improved diagnostic confidence in evaluating bone non-union using virtual monochromatic dual-energy CT.

PURPOSE: The aim of this study was to determine whether virtual monochromatic dual-energy CT imaging improves the evaluation of suspected non-union of the appendicular skeleton treated with titanium or stainless steel intramedullary nails and plates. METHOD: Forty-one patients with a clinical suspected non-union with hardware in place were included and scanned on a dual-source CT-scanner using 100/Sn150kVp. Images including titanium hardware were extracted at 130 keV. Images including stainless steel hardware were extracted at 150 keV. Monochromatic 70 keV images served as reference. Non-union confirmed during revision surgery was used as gold standard. A musculoskeletal radiologist and orthopedic trauma surgeon evaluated images on image quality, degree and location of consolidation, non-union type and diagnostic confidence. RESULTS: Likert scores with respect to image quality improved from 0.88 to 1.83 (p < 0.001) in high (130 and 150) keV images. High keV images reduced the number of false negative non-unions based on consolidation grade with 5% (p = 0.283). Agreement between observers regarding location of consolidation and non-union type did not improve in 130 and 150 keV images. Diagnostic confidence improved from 1.43 to 2.37 in high keV images compared to 70 keV images (p < 0.001). Overall diagnostic confidence was higher in intramedullary nails than plates (p < 0.05). CONCLUSIONS: Use of virtual monochromatic 130 and 150 keV dual-energy CT compared to 70 keV images improves the evaluation of suspected non-union of the appendicular skeleton treated with titanium or stainless steel intramedullary nails and plates.